WebKit-https.git
12 months agoUse enum classes and OptionSets for PaintPhase and PaintBehavior
achristensen@apple.com [Mon, 6 Aug 2018 20:41:05 +0000 (20:41 +0000)]
Use enum classes and OptionSets for PaintPhase and PaintBehavior
https://bugs.webkit.org/show_bug.cgi?id=188323

Reviewed by Simon Fraser.

Source/WebCore:

No change in behaviour.

* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlTextTrackContainerElement::createTextTrackRepresentationImage):
* page/FrameSnapshotting.cpp:
(WebCore::snapshotFrameRectWithClip):
* page/FrameView.cpp:
(WebCore::FrameView::reset):
(WebCore::FrameView::willPaintContents):
(WebCore::FrameView::paintContents):
(WebCore::FrameView::setPaintBehavior):
(WebCore::FrameView::paintBehavior const):
(WebCore::FrameView::paintContentsForSnapshot):
* page/FrameView.h:
* rendering/InlineElementBox.cpp:
(WebCore::InlineElementBox::paint):
* rendering/InlineFlowBox.cpp:
(WebCore::InlineFlowBox::paint):
(WebCore::InlineFlowBox::paintBoxDecorations):
(WebCore::InlineFlowBox::paintMask):
* rendering/InlineTextBox.cpp:
(WebCore::InlineTextBox::paint):
* rendering/PaintInfo.h:
(WebCore::PaintInfo::PaintInfo):
(WebCore::PaintInfo::forceBlackText const):
(WebCore::PaintInfo::forceWhiteText const):
(WebCore::PaintInfo::skipRootBackground const):
(WebCore::PaintInfo::paintRootBackgroundOnly const):
* rendering/PaintPhase.h:
(): Deleted.
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::paint):
(WebCore::RenderBlock::paintContents):
(WebCore::RenderBlock::paintObject):
(WebCore::RenderBlock::paintSelection):
* rendering/RenderBlockFlow.cpp:
(WebCore::RenderBlockFlow::paintFloats):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::paintMask):
(WebCore::RenderBox::paintClippingMask):
(WebCore::RenderBox::paintMaskImages):
(WebCore::RenderBox::pushContentsClip):
(WebCore::RenderBox::popContentsClip):
* rendering/RenderBox.h:
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::decodingModeForImageDraw const):
(WebCore::RenderBoxModelObject::paintMaskForTextFillBox):
(WebCore::RenderBoxModelObject::fixedBackgroundPaintsInLocalCoordinates const):
* rendering/RenderDetailsMarker.cpp:
(WebCore::RenderDetailsMarker::paint):
* rendering/RenderElement.cpp:
(WebCore::RenderElement::paintAsInlineBlock):
(WebCore::RenderElement::selectionColor const):
* rendering/RenderEmbeddedObject.cpp:
(WebCore::RenderEmbeddedObject::paint):
(WebCore::RenderEmbeddedObject::paintReplaced):
* rendering/RenderFileUploadControl.cpp:
(WebCore::RenderFileUploadControl::paintObject):
* rendering/RenderFrameSet.cpp:
(WebCore::RenderFrameSet::paint):
* rendering/RenderHTMLCanvas.cpp:
(WebCore::RenderHTMLCanvas::paintReplaced):
* rendering/RenderImage.cpp:
(WebCore::RenderImage::paintReplaced):
(WebCore::RenderImage::paint):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::renderableTransform const):
(WebCore::transparencyClipBox):
(WebCore::expandClipRectForDescendantsAndReflection):
(WebCore::paintingExtent):
(WebCore::RenderLayer::paint):
(WebCore::RenderLayer::paintOverlayScrollbars):
(WebCore::RenderLayer::paintLayer):
(WebCore::RenderLayer::paintLayerContents):
(WebCore::RenderLayer::paintBackgroundForFragments):
(WebCore::RenderLayer::paintForegroundForFragments):
(WebCore::RenderLayer::paintForegroundForFragmentsWithPhase):
(WebCore::RenderLayer::paintOutlineForFragments):
(WebCore::RenderLayer::paintMaskForFragments):
(WebCore::RenderLayer::paintChildClippingMaskForFragments):
(WebCore::RenderLayer::calculateClipRects const):
* rendering/RenderLayer.h:
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::paintIntoLayer):
(WebCore::RenderLayerBacking::paintContents):
* rendering/RenderLayerBacking.h:
* rendering/RenderLineBoxList.cpp:
(WebCore::RenderLineBoxList::paint const):
* rendering/RenderListBox.cpp:
(WebCore::RenderListBox::paintObject):
* rendering/RenderListMarker.cpp:
(WebCore::RenderListMarker::paint):
* rendering/RenderReplaced.cpp:
(WebCore::RenderReplaced::paint):
(WebCore::RenderReplaced::shouldPaint):
* rendering/RenderReplica.cpp:
(WebCore::RenderReplica::paint):
* rendering/RenderScrollbarPart.cpp:
(WebCore::RenderScrollbarPart::paintIntoRect):
* rendering/RenderSnapshottedPlugIn.cpp:
(WebCore::RenderSnapshottedPlugIn::paint):
* rendering/RenderTable.cpp:
(WebCore::RenderTable::paintObject):
(WebCore::RenderTable::paintMask):
(WebCore::RenderTable::overflowClipRect):
* rendering/RenderTable.h:
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::paint):
(WebCore::RenderTableCell::paintCollapsedBorders):
(WebCore::RenderTableCell::paintMask):
* rendering/RenderTableRow.cpp:
(WebCore::RenderTableRow::paintOutlineForRowIfNeeded):
(WebCore::RenderTableRow::paint):
* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::paint):
(WebCore::RenderTableSection::paintCell):
(WebCore::RenderTableSection::paintObject):
* rendering/RenderThemeMac.mm:
(WebCore::RenderThemeMac::paintSnapshottedPluginOverlay):
(WebCore::RenderThemeMac::paintImageControlsButton):
* rendering/RenderVideo.cpp:
(WebCore::RenderVideo::paintReplaced):
* rendering/RenderWidget.cpp:
(WebCore::RenderWidget::paintContents):
(WebCore::RenderWidget::paint):
* rendering/RootInlineBox.cpp:
(WebCore::RootInlineBox::paintEllipsisBox const):
* rendering/SimpleLineLayoutFunctions.cpp:
(WebCore::SimpleLineLayout::paintFlow):
* rendering/mathml/MathOperator.cpp:
(WebCore::MathOperator::paint):
* rendering/mathml/RenderMathMLBlock.cpp:
(WebCore::RenderMathMLBlock::paint):
* rendering/mathml/RenderMathMLFraction.cpp:
(WebCore::RenderMathMLFraction::paint):
* rendering/mathml/RenderMathMLMenclose.cpp:
(WebCore::RenderMathMLMenclose::paint):
* rendering/mathml/RenderMathMLToken.cpp:
(WebCore::RenderMathMLToken::paint):
* rendering/svg/RenderSVGContainer.cpp:
(WebCore::RenderSVGContainer::paint):
* rendering/svg/RenderSVGForeignObject.cpp:
(WebCore::RenderSVGForeignObject::paint):
* rendering/svg/RenderSVGImage.cpp:
(WebCore::RenderSVGImage::paint):
* rendering/svg/RenderSVGResource.cpp:
(WebCore::requestPaintingResource):
* rendering/svg/RenderSVGResourceClipper.cpp:
(WebCore::RenderSVGResourceClipper::drawContentIntoMaskImage):
* rendering/svg/RenderSVGResourceSolidColor.cpp:
(WebCore::RenderSVGResourceSolidColor::applyResource):
* rendering/svg/RenderSVGRoot.cpp:
(WebCore::RenderSVGRoot::paintReplaced):
* rendering/svg/RenderSVGShape.cpp:
(WebCore::RenderSVGShape::paint):
* rendering/svg/RenderSVGText.cpp:
(WebCore::RenderSVGText::paint):
* rendering/svg/SVGInlineFlowBox.cpp:
(WebCore::SVGInlineFlowBox::paintSelectionBackground):
(WebCore::SVGInlineFlowBox::paint):
* rendering/svg/SVGInlineTextBox.cpp:
(WebCore::SVGInlineTextBox::paintSelectionBackground):
(WebCore::SVGInlineTextBox::paint):
* rendering/svg/SVGRenderingContext.cpp:
(WebCore::isRenderingMaskImage):
(WebCore::SVGRenderingContext::renderSubtreeToImageBuffer):
* rendering/svg/SVGRootInlineBox.cpp:
(WebCore::SVGRootInlineBox::paint):

Source/WebKit:

* WebProcess/InjectedBundle/DOM/InjectedBundleNodeHandle.cpp:
(WebKit::imageForRect):
* WebProcess/InjectedBundle/DOM/InjectedBundleRangeHandle.cpp:
(WebKit::InjectedBundleRangeHandle::renderedImage):
* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::shouldCreateTransientPaintingSnapshot const):

Source/WebKitLegacy/mac:

* WebCoreSupport/WebContextMenuClient.mm:
(WebContextMenuClient::imageForCurrentSharingServicePickerItem):
* WebView/WebFrame.mm:
(-[WebFrame _paintBehaviorForDestinationContext:]):
(-[WebFrame _drawRect:contentsOnly:]):
* WebView/WebHTMLView.mm:
(imageFromRect):
(selectionImage):

Source/WTF:

* wtf/OptionSet.h:
(WTF::OptionSet::OptionSet):
Allow a zero-valued enum value in the OptionSet constructor.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234619 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agofast/custom-elements/custom-element-registry-wrapper-should-stay-alive.html always...
rniwa@webkit.org [Mon, 6 Aug 2018 20:35:53 +0000 (20:35 +0000)]
fast/custom-elements/custom-element-registry-wrapper-should-stay-alive.html always timeouts on debug bots
https://bugs.webkit.org/show_bug.cgi?id=188353

Add a timeout expectation to the test while I verify the fix for the test.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234618 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, rolling out r234552.
ryanhaddad@apple.com [Mon, 6 Aug 2018 20:17:04 +0000 (20:17 +0000)]
Unreviewed, rolling out r234552.

Introduced 2 layout test failures on High Sierra.

Reverted changeset:

"Check with SafeBrowsing during navigation in WKWebView"
https://bugs.webkit.org/show_bug.cgi?id=188133
https://trac.webkit.org/changeset/234552

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234615 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Caret disappears after resigning and becoming first responder if active focus...
wenson_hsieh@apple.com [Mon, 6 Aug 2018 19:31:06 +0000 (19:31 +0000)]
[iOS] Caret disappears after resigning and becoming first responder if active focus state is retained
https://bugs.webkit.org/show_bug.cgi?id=188322
<rdar://problem/42455270>

Reviewed by Tim Horton.

Fixes an API test added in r234600, which had incorrect test expectations — namely, the width of the caret rect
was not in content view coordinates, unlike the x, y, and height.

* TestWebKitAPI/Tests/ios/KeyboardInputTestsIOS.mm:
(rounded):
(-[TestWKWebView waitForCaretViewFrameToBecome:]):
(-[TestWKWebView waitForSelectionViewRectsToBecome:]):
(TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234614 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, rolling out r234569.
ryanhaddad@apple.com [Mon, 6 Aug 2018 18:46:59 +0000 (18:46 +0000)]
Unreviewed, rolling out r234569.

Breaks internal builds.

Reverted changeset:

"We should cache the compiled sandbox profile in a data vault"
https://bugs.webkit.org/show_bug.cgi?id=184991
https://trac.webkit.org/changeset/234569

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234613 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed test gardening for mac-wk1.
ryanhaddad@apple.com [Mon, 6 Aug 2018 18:37:24 +0000 (18:37 +0000)]
Unreviewed test gardening for mac-wk1.

* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234612 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoFix IPC::Connection leak in StorageManager
cdumez@apple.com [Mon, 6 Aug 2018 18:31:36 +0000 (18:31 +0000)]
Fix IPC::Connection leak in StorageManager
https://bugs.webkit.org/show_bug.cgi?id=188321
<rdar://problem/42748485>

Reviewed by Alex Christensen.

When a StorageMap is destroyed on WebContent process side, StorageManager::destroyStorageMap()
gets called via IPC with a (IPC::Connection, StorageMapID) pair. Normally, it removes this
pair from m_storageAreasByConnection. However, if this is a *transient* StorageMap (sessionStorage),
then we keep the pair in the map and we merely remove the StorageMapID as a listener from the
StorageArea. We do this so that:
1. The StorageArea stays alive so that it can be reused later on for the same security origin, on
   the same IPC::Connection (logic for this is in StorageManager::createTransientLocalStorageMap()
2. Removing the StorageMapID as a listener from the StorageArea is important because
   StorageArea::m_eventListeners holds a strong reference to the IPC::Connection in a std::pair
   with the StorageMapID (HashSet<std::pair<RefPtr<IPC::Connection>, uint64_t>> m_eventListeners).

As mentioned in 1 above, in StorageManager::createTransientLocalStorageMap(), there is logic to
check if there is already an existing StorageArea for the given IPC::Connection that is transient
and is for the same security origin. In this case, we could avoid constructing a new StorageArea
and we would:
1. Add a new entry to m_storageAreasByConnection with the key (connection, newStorageMapID), using
   same same StorageArea as value.
2. Remove the previous (connection, oldStorageMapID) key from m_storageAreasByConnection.

Step 2 here is wrong and is updated in this patch. It is only safe to remove the previous
(connection, oldStorageMapID) if this oldStorageMapID no longer exists (i.e. destroyStorageMap()
was already called for it). This patch thus adds a check before removing (connection, oldStorageMapID)
from the HashMap to make sure that the oldStorageMapID is no longer a listener of the StorageArea).

This would cause leaks in the following case:
1. We construct a StorageArea for (connection1, storageMapId1)
2. We ask for a StorageArea for (connection1, storageMapId2) and decide to reuse the existing StorageArea
   since it has the same SecurityOrigin.
3. As a result of step2, we would remove (connection1, storageMapId1) from m_storageAreasByConnection
   and add (connection1, storageMapId2), even though there is still a StorageMap with storageMapId1
   on WebContent process side.
4. Later on, we would try to call destroyStorageMap(connection1, storageMap1), it would fail to find
   it in m_storageAreasByConnection and return early. It would therefore fail to remove storageMapId1
   as a listener of the StorageArea which still exists.
-> This would leak the IPC::Connection that there would be a std::pair<RefPtr<IPC::Connection>, StorageMapID>
   with value (connection1, storageMap1) which would get leaked and it would ref the IPC::Connection.

This code should really be refactored to be less leak prone but I have kept the patch minimal for now
to facilitate cherry-picking.

Note that this would reproduce very easily on sina.com.cn, when clicking bold links at the top, which
opens new tabs to different pages in the same WebContent process. When closing all Safari windows, the
IPC::Connection for this WebContent process would stay alive.

* UIProcess/WebStorage/StorageManager.cpp:
(WebKit::StorageManager::StorageArea::hasListener const):
(WebKit::StorageManager::createTransientLocalStorageMap):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234611 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoMake BlendMode an enum class
commit-queue@webkit.org [Mon, 6 Aug 2018 18:19:43 +0000 (18:19 +0000)]
Make BlendMode an enum class
https://bugs.webkit.org/show_bug.cgi?id=188325

Patch by Alex Christensen <achristensen@webkit.org> on 2018-08-06
Reviewed by Darin Adler.

Source/WebCore:

No change in behavior.

* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
(WebCore::CSSPrimitiveValue::operator BlendMode const):
* html/HTMLImageElement.cpp:
(WebCore::HTMLImageElement::parseAttribute):
* html/canvas/CanvasRenderingContext2DBase.cpp:
(WebCore::CanvasRenderingContext2DBase::State::State):
(WebCore::CanvasRenderingContext2DBase::setGlobalCompositeOperation):
(WebCore::CanvasRenderingContext2DBase::drawImageFromRect):
* platform/graphics/BitmapImage.h:
* platform/graphics/GraphicsContext.h:
(WebCore::ImagePaintingOptions::ImagePaintingOptions):
* platform/graphics/GraphicsContextImpl.h:
* platform/graphics/GraphicsLayer.cpp:
(WebCore::GraphicsLayer::dumpProperties const):
* platform/graphics/GraphicsLayer.h:
* platform/graphics/GraphicsTypes.cpp:
(WebCore::parseBlendMode):
(WebCore::parseCompositeAndBlendOperator):
(WebCore::compositeOperatorName):
(WebCore::blendModeName):
(WebCore::operator<<):
* platform/graphics/GraphicsTypes.h:
* platform/graphics/Image.cpp:
(WebCore::Image::drawTiled):
* platform/graphics/Image.h:
* platform/graphics/ImageBuffer.h:
* platform/graphics/ca/cocoa/PlatformCAFiltersCocoa.mm:
(PlatformCAFilters::setBlendingFiltersOnLayer):
* platform/graphics/cg/GraphicsContextCG.cpp:
(WebCore::selectCGBlendMode):
* platform/graphics/cpu/arm/filters/FEBlendNEON.h:
(WebCore::FEBlend::platformApplyNEON):
* platform/graphics/displaylists/DisplayListItems.h:
* platform/graphics/displaylists/DisplayListRecorder.h:
* platform/graphics/filters/FEBlend.cpp:
(WebCore::FEBlend::externalRepresentation const):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::paintFillLayers):
* rendering/RenderImage.cpp:
(WebCore::RenderImage::paintIntoRect):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::RenderLayer):
(WebCore::RenderLayer::updateBlendMode):
(WebCore::RenderLayer::beginTransparencyLayers):
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateBlendMode):
* rendering/style/FillLayer.cpp:
(WebCore::FillLayer::FillLayer):
(WebCore::FillLayer::hasOpaqueImage const):
* rendering/style/FillLayer.h:
(WebCore::FillLayer::initialFillBlendMode):
* rendering/style/RenderStyle.h:
(WebCore::RenderStyle::setBlendMode):
(WebCore::RenderStyle::hasBlendMode const):
(WebCore::RenderStyle::blendMode const):
(WebCore::RenderStyle::initialBlendMode):
* rendering/style/StyleRareNonInheritedData.cpp:
(WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
* rendering/svg/SVGRenderingContext.cpp:
(WebCore::SVGRenderingContext::prepareToRenderSVGContent):
* svg/SVGAnimatedEnumeration.cpp:
(WebCore::enumerationValueForTargetAttribute):
* svg/SVGFEBlendElement.cpp:
(WebCore::SVGFEBlendElement::SVGFEBlendElement):
(WebCore::SVGFEBlendElement::parseAttribute):
* svg/SVGFEBlendElement.h:
(WebCore::SVGPropertyTraits<BlendMode>::highestEnumValue):
(WebCore::SVGPropertyTraits<BlendMode>::toString):
* svg/graphics/SVGImage.cpp:
(WebCore::SVGImage::nativeImageForCurrentFrame):
(WebCore::SVGImage::nativeImage):
(WebCore::SVGImage::drawPatternForContainer):
(WebCore::SVGImage::draw):

Source/WebKit:

* Shared/RemoteLayerTree/RemoteLayerTreeTransaction.mm:
(WebKit::RemoteLayerTreeTransaction::LayerProperties::LayerProperties):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234610 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Indeterminate checkboxes appear unchecked
commit-queue@webkit.org [Mon, 6 Aug 2018 18:06:25 +0000 (18:06 +0000)]
[iOS] Indeterminate checkboxes appear unchecked
https://bugs.webkit.org/show_bug.cgi?id=160484
<rdar://problem/31143185>

Patch by Aditya Keerthi <akeerthi@apple.com> on 2018-08-06
Reviewed by Wenson Hsieh.

Source/WebCore:

Add support for rendering indeterminate checkboxes on iOS.

We now paint the unchecked appearance only if the checkbox is not checked and is
not indeterminate. Otherwise, we draw a checkmark if the checkbox is in the
checked state and a dash if the checkbox is in the indeterminate state. Both of
these states share the same dark background.

Test: fast/forms/ios/render-indeterminate-checkbox.html

* css/html.css:
(input[type="checkbox"]:indeterminate):
(input[type="checkbox"]:indeterminate:disabled):
* rendering/RenderThemeIOS.mm:
(WebCore::drawJoinedLines):
(WebCore::RenderThemeIOS::paintCheckboxDecorations):

LayoutTests:

Added reftest to ensure that the indeterminate checkbox does not appear the same
as an unchecked checkbox.

* fast/forms/ios/render-indeterminate-checkbox-expected-mismatch.html: Added.
* fast/forms/ios/render-indeterminate-checkbox.html: Added.
* platform/ios/fast/forms/indeterminate-expected.txt: Rebaseline.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234609 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoHTML parser should execute custom element reactions for setting attributes immediatel...
rniwa@webkit.org [Mon, 6 Aug 2018 17:45:41 +0000 (17:45 +0000)]
HTML parser should execute custom element reactions for setting attributes immediately after creating a custom element
https://bugs.webkit.org/show_bug.cgi?id=188336

Reviewed by Frédéric Wang.

LayoutTests/imported/w3c:

Rebaseline the test now that the relevant test case is passing.

* web-platform-tests/custom-elements/parser/parser-sets-attributes-and-children-expected.txt:

Source/WebCore:

Push and pop an element queue from the custom element reactions stack when constructing a custom element:
https://html.spec.whatwg.org/multipage/parsing.html#create-an-element-for-the-token

To do this, we instantiate CustomElementReactionStack in HTMLDocumentParser::runScriptsForPausedTreeBuilder
where we synchronously construct a custom element. We don't have to worry about whether *will execute script*
is set or not since the presence of an element queue should not be observable in the case where we're constructing
a fallback element (since it would not enqueue any new custom element reaction).

Tests: imported/w3c/web-platform-tests/custom-elements/parser/parser-sets-attributes-and-children.html

* html/parser/HTMLDocumentParser.cpp:
(WebCore::HTMLDocumentParser::runScriptsForPausedTreeBuilder): Instantiate CustomElementReactionStack. Note that we
don't insert the custom element into the parser until we finish processing the custom element reactions.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234608 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Win][DLLLauncher] Remove --highDPI switch
Hironori.Fujii@sony.com [Mon, 6 Aug 2018 17:43:42 +0000 (17:43 +0000)]
[Win][DLLLauncher] Remove --highDPI switch
https://bugs.webkit.org/show_bug.cgi?id=188255

Reviewed by Yusuke Suzuki.

Bug 149417 has added `--highDPI` switch to make MiniBrowser DPI
aware. But, Bug 158733 has made MiniBrowser always DPI aware. It
became useless since the change. Remove the switch.

This change fixes the issue all dump render tree tests are failing
in WinCairo port if high DPI display is connected.

* MiniBrowser/win/Common.cpp:
(parseCommandLine): Removed `--highDPI` switch.
* win/DLLLauncher/DLLLauncherMain.cpp:
(wWinMain): Do not call SetProcessDPIAware.
(shouldUseHighDPI): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234607 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, fix the build by ignoring some deprecation warnings.
ryanhaddad@apple.com [Mon, 6 Aug 2018 17:25:13 +0000 (17:25 +0000)]
Unreviewed, fix the build by ignoring some deprecation warnings.

* wtf/MD5.cpp:
(WTF::MD5::MD5):
(WTF::MD5::addBytes):
(WTF::MD5::checksum):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234606 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoReturn extracted key ids as an optional
cturner@igalia.com [Mon, 6 Aug 2018 16:47:21 +0000 (16:47 +0000)]
Return extracted key ids as an optional
https://bugs.webkit.org/show_bug.cgi?id=188303

Reviewed by Darin Adler.

An empty list of extracted key ids was being considered a failure
case before this patch. In the PSSH boxes from the CENC standard,
it's not uncommon for the box to be version 0, meaning it has no
embedded key ids, so the case when there's an empty list should
not be treated as an error. Given this, the interface should be
more general and allow for a sentinel value indicating a parsing
error rather than an absence of key ids.

Covered by existing tests.

* Modules/encryptedmedia/InitDataRegistry.cpp:
(WebCore::extractKeyIDsKeyids): Change return type to be wrapped
in an optional, and make parsing errors return a nullopt rather
than an empty vector.
(WebCore::sanitizeKeyids): Use the new optional interface, return
a null RefPtr in the case of a parsing error, this method may now
return an empty vector.
(WebCore::extractKeyIDsCenc): Not implemented, so return an error
value rather than an empty vector.
(WebCore::extractKeyIDsWebM): Ditto.
(WebCore::InitDataRegistry::extractKeyIDs): Ditto.
* Modules/encryptedmedia/InitDataRegistry.h: Update the interface
to use an optional return type.
* platform/graphics/avfoundation/CDMFairPlayStreaming.cpp:
(WebCore::CDMPrivateFairPlayStreaming::extractKeyIDsSinf): Update
to use the new interface.
(WebCore::CDMPrivateFairPlayStreaming::extractKeyIDsSkd): Ditto.
* platform/graphics/avfoundation/CDMFairPlayStreaming.h: Ditto.
* platform/graphics/avfoundation/objc/CDMInstanceFairPlayStreamingAVFObjC.mm:
(WebCore::CDMInstanceFairPlayStreamingAVFObjC::keyIDs): Convert
the optional value into a vector, since it is assumed you can not
have an empty vector of key ids in the init datas FPS supports.
* testing/MockCDMFactory.cpp:
(WebCore::MockCDMInstance::requestLicense): Only return an error
if there really was a parsing error, rather than the case of there
being zero key ids in the init data payload.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234603 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agostorage/indexeddb/modern/opendatabase-after-storage-crash.html is causing tests in...
tsavell@apple.com [Mon, 6 Aug 2018 16:13:19 +0000 (16:13 +0000)]
storage/indexeddb/modern/opendatabase-after-storage-crash.html is causing tests in directory svg/ to timeout at random
https://bugs.webkit.org/show_bug.cgi?id=187931

Unreviewed test gardening.

* platform/wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234602 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Tests that try to interact with the QuickType bar time out on iOS 11+
wenson_hsieh@apple.com [Mon, 6 Aug 2018 15:01:47 +0000 (15:01 +0000)]
[iOS] Tests that try to interact with the QuickType bar time out on iOS 11+
https://bugs.webkit.org/show_bug.cgi?id=188335
<rdar://problem/32542437> and <rdar://problem/32542433>

Reviewed by Tim Horton.

Tools:

Remove a UIScriptController helper method that attempted to wait for QuickType buttons to appear, and then send
touch events to the buttons. These were only used for the two tests which this patch refactors. See the
LayoutTests ChangeLog for more detail.

* DumpRenderTree/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::selectTextCandidateAtIndex): Deleted.
* TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
* TestRunnerShared/UIScriptContext/UIScriptController.cpp:
(WTR::UIScriptController::selectTextCandidateAtIndex): Deleted.
(WTR::UIScriptController::waitForTextPredictionsViewAndSelectCandidateAtIndex): Deleted.
* TestRunnerShared/UIScriptContext/UIScriptController.h:
* TestRunnerShared/spi/UIKitTestSPI.h:

Remove an internal class declaration that is now unnecessary.

* WebKitTestRunner/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::selectTextCandidateAtIndex): Deleted.
(WTR::UIScriptController::waitForTextPredictionsViewAndSelectCandidateAtIndex): Deleted.

LayoutTests:

Refactors a couple of tests that currently rely on the QuickType bar being shown when the hardware keyboard is
presented in the iOS simulator, so that it can insert text suggestions by tapping QuickType bar buttons; in the
first place, it was never guaranteed that this bar would show up, or even have more than 1 text suggestion.
Instead, we use the existing UIScriptController::applyAutocorrection method to trigger text replacement in the
same way tapping a button on the QuickType bar would replace the current word with the suggested word.

* fast/events/ios/before-input-events-prevent-candidate-insertion-expected.txt:
* fast/events/ios/before-input-events-prevent-candidate-insertion.html:
* fast/events/ios/input-events-insert-replacement-text-expected.txt:
* fast/events/ios/input-events-insert-replacement-text.html:
* platform/ios/TestExpectations:

Removes a [Timeout] expectation.

* resources/ui-helper.js:
(window.UIHelper.typeCharacter):

Add a new UIHelper method to type a character using the keyboard. Sends hardware keyboard events on the WebKit2
port of iOS, and uses EventSender elsewhere.

(window.UIHelper.applyAutocorrection):

Add a new UIHelper method that wraps UIScriptController::applyAutocorrection. Only supported in WebKit2
currently.

(window.UIHelper):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234601 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Caret disappears after resigning and becoming first responder if active focus...
wenson_hsieh@apple.com [Mon, 6 Aug 2018 14:14:42 +0000 (14:14 +0000)]
[iOS] Caret disappears after resigning and becoming first responder if active focus state is retained
https://bugs.webkit.org/show_bug.cgi?id=188322
<rdar://problem/42455270>

Reviewed by Tim Horton.

Source/WebKit:

Prior to r230745, when a user selects a word in non-editable web content without a prior selection, we would
always try to activate the text interaction assistant, creating a selection view (a UITextSelectionView). After
the long press is recognized, this text selection view is configured for "highlight mode", which is a special
mode for presenting selection UI where the grabber handles at the start and end of the selection are suppressed.
UIKit then prepares to show the selection by asking WKContentView for the number of selection rects; if this
number is zero, the UITextSelectionView is removed from the superview, and state that keeps track of whether the
selection view is in "highlight mode" is reset.

In the case where there's no prior selection, our cached EditorState in the UI process will not be up to date
yet when the gesture is recognized. This means that when UIKit asks us for the number of selection rects, we'll
return 0, which causes any state tracking "highlight mode" for the selection to be reset, subsequently resulting
in selection handles showing up before the user has ended the initial loupe gesture.

r230745 addressed this bug by removing logic to activate the text selection when becoming first responder,
instead deferring until the next `-_selectionChanged` call with post-layout editor state data to activate the
selection. While this does ensure that selection handles don't erroneously appear, it also means that clients
that call -becomeFirstResponder to show selection UI and the keyboard in a web view while an element is already
focused will not have an active selection assistant (i.e. the selection view will still be hidden). One way this
happens is when Safari uses `-_retainActiveFocusedState` in combination with `-resignFirstResponder` and
`-becomeFirstResponder` to temporarily switch focus away from the web view when the URL bar is tapped.

To fix both the inactive selection after `-becomeFirstResponder` as well as the selection handles showing up
when performing a loupe gesture, we simply make the check in `-becomeFirstResponderForWebView` more nuanced.
Instead of always activating the selection or never activating the selection, only activate the selection if the
current editor state has information about a selection to avoid causing the selection view to be immediately
removed and "highlight mode" to be reset when selecting a word via loupe gesture for the first time.

Tests:  KeyboardInputTests.CaretSelectionRectAfterRestoringFirstResponder
        KeyboardInputTests.RangedSelectionRectAfterRestoringFirstResponder
        editing/selection/ios/selection-handles-after-touch-end.html

* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView becomeFirstResponderForWebView]):
(-[WKContentView canShowNonEmptySelectionView]):

Tools:

Adds plumbing in UIScriptController to grab the start and end selection handle rects for use in the new layout
test. Also adds new API tests to verify that when a web view resigns first responder, both caret and range
selection views are hidden, and when first responder status is restored, both caret and range selection views
are made visible again.

* DumpRenderTree/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::selectionStartGrabberViewRect const):
(WTR::UIScriptController::selectionEndGrabberViewRect const):
* TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
* TestRunnerShared/UIScriptContext/UIScriptController.cpp:
(WTR::UIScriptController::selectionStartGrabberViewRect const):
(WTR::UIScriptController::selectionEndGrabberViewRect const):
* TestRunnerShared/UIScriptContext/UIScriptController.h:
* TestWebKitAPI/Tests/ios/KeyboardInputTestsIOS.mm:
(-[TestWKWebView waitForCaretViewFrameToBecome:]):
(-[TestWKWebView waitForSelectionViewRectsToBecome:]):
(webViewWithAutofocusedInput):

Pull out some common logic for creating a web view that allows programmatic focus to present the keyboard, and
immediately loading a web page with an autofocusing text field.

(TestWebKitAPI::TEST):
* TestWebKitAPI/cocoa/TestWKWebView.h:
* TestWebKitAPI/cocoa/TestWKWebView.mm:
(-[TestWKWebView caretViewRectInContentCoordinates]):
(-[TestWKWebView selectionViewRectsInContentCoordinates]):
* WebKitTestRunner/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::selectionStartGrabberViewRect const):
(WTR::UIScriptController::selectionEndGrabberViewRect const):

LayoutTests:

Adds a new layout test to verify that (1) selection handles are not shown when selecting a word by long
pressing prior to ending the touch, and (2) selection handles are shown after ending the touch.

* editing/selection/ios/selection-handles-after-touch-end-expected.txt: Added.
* editing/selection/ios/selection-handles-after-touch-end.html: Added.
* platform/win/TestExpectations:

Skip iOS selection tests on Windows.

* resources/ui-helper.js:

Introduces new hooks in UIHelper to grab the frames of the start and end selection handle views.

(window.UIHelper.getSelectionStartGrabberViewRect.return.new.Promise.):
(window.UIHelper.getSelectionStartGrabberViewRect.return.new.Promise):
(window.UIHelper.getSelectionStartGrabberViewRect):
(window.UIHelper.getSelectionEndGrabberViewRect.return.new.Promise.):
(window.UIHelper.getSelectionEndGrabberViewRect.return.new.Promise):
(window.UIHelper.getSelectionEndGrabberViewRect):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234600 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Layout tests that send HID events cause WebKitTestRunner to crash on recent...
wenson_hsieh@apple.com [Mon, 6 Aug 2018 13:59:04 +0000 (13:59 +0000)]
[iOS] Layout tests that send HID events cause WebKitTestRunner to crash on recent SDKs
https://bugs.webkit.org/show_bug.cgi?id=188334
<rdar://problem/40630074>

Reviewed by Tim Horton.

To mark the end of previously dispatched IOHID events, HIDEventGenerator currently sends a vendor-defined event
and stores the completion callback ID for the previously dispatched events as vendor-defined data. When this
vendor-defined marker event is handled by the application, we then read the callback ID back from the event, map
it to a completion block, and invoke the completion block to signal that the previous HID event has been
processed.

This callback ID is an unsigned, so we tell IOKit that we need `sizeof(unsigned)` (4 bytes) to store it. On
shipping software, IOKit clamps this to a minimum of 8 bytes, i.e. `sizeof(CFIndex)`. When we later call
IOHIDEventGetIntegerValue to read the value of our vendor-defined data as a CFIndex, we get our expected
callback ID because the buffer was clamped to 8 bytes.

However, on recent iOS SDKs that contain the fix for <rdar://problem/20082284>, IOKit no longer clamps the size
of the vendor-defined data buffer to 8 bytes. This means that when we try to use IOHIDEventGetIntegerValue to
read our callback ID back, we end up getting a CFIndex where the lower 4 bytes are the callback ID we wrote, and
the upper 4 bytes are garbage. In the case where any of these upper 4 bytes are non-zero, we fail to map the
callback ID to a completion handler, and so we never finish dispatching the HID event, causing an exception to
be thrown.

To fix this, we adjust callback ID to be a CFIndex, which matches IOHIDEventGetIntegerValue's return type.

* WebKitTestRunner/ios/HIDEventGenerator.mm:
(+[HIDEventGenerator nextEventCallbackID]):
(-[HIDEventGenerator _sendMarkerHIDEventWithCompletionBlock:]):

Also refactor a bit of `-_sendMarkerHIDEventWithCompletionBlock:` by using auto and move semantics.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234599 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WPE] New webgl 2.0 failures.
csaavedra@igalia.com [Mon, 6 Aug 2018 12:59:41 +0000 (12:59 +0000)]
[WPE] New webgl 2.0 failures.

Unreviewed gardening.
* platform/wpe/TestExpectations: Two webgl 2.0 tests are failing,
mark them.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234598 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[GTK] Skip new webgl 2.0 tests. Unreviewed gardening.
csaavedra@igalia.com [Mon, 6 Aug 2018 12:45:38 +0000 (12:45 +0000)]
[GTK] Skip new webgl 2.0 tests. Unreviewed gardening.

* platform/gtk/TestExpectations: Expanded webgl 2.0 skipped
test to include all, since there were new failures.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234597 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoMake two-arguments versions of scrollBy/scrollTo depend on the one-argument versions
commit-queue@webkit.org [Mon, 6 Aug 2018 09:56:08 +0000 (09:56 +0000)]
Make two-arguments versions of scrollBy/scrollTo depend on the one-argument versions
https://bugs.webkit.org/show_bug.cgi?id=188300

Patch by Frederic Wang <fwang@igalia.com> on 2018-08-06
Reviewed by Darin Adler.

This patch refactors a bit the scrollBy/scrollTo code, so that the two-arguments versions
share the same code path as the more generic one-argument versions. In particular, this
helps to implement the ScrollBehavior option (bug 188043) since the one-argument versions
will require to distinguish between smooth and instant scrolling. The logic to normalize
non finite left/right values or to use a fallback when they are absent is also factored out
into ScrollToOptions.

References:
https://drafts.csswg.org/cssom-view/#dom-element-scroll
https://drafts.csswg.org/cssom-view/#dom-element-scrollby
https://drafts.csswg.org/cssom-view/#dom-window-scroll
https://drafts.csswg.org/cssom-view/#dom-window-scrollby

No new tests, behavior is unchanged.

* dom/Element.cpp:
(WebCore::Element::scrollBy): Make two-parameter version depends on one-parameter version
and rewrite the normalize / fallback logic.
(WebCore::Element::scrollTo): Rewrite the normalize / fallback logic.
(WebCore::normalizeNonFiniteValue): Deleted. The logic is moved to ScrollToOptions.
* page/DOMWindow.cpp:
(WebCore::DOMWindow::scrollBy const): Make two-parameter version depends on one-parameter
version and rewrite the normalize / fallback logic.
(WebCore::DOMWindow::scrollTo const): Make two-parameter version depends on one-parameter
version and rewrite the normalize / fallback logic.
* page/ScrollToOptions.h: Add <cmath> to use std::isfinite
(WebCore::ScrollToOptions::normalizeNonFiniteCoordinatesOrFallBackTo): New function to
normalize left/right values or fallback to the specified value if it is missing.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234596 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed follow-up to r234594.
zandobersek@gmail.com [Mon, 6 Aug 2018 09:24:36 +0000 (09:24 +0000)]
Unreviewed follow-up to r234594.

Add missing constructor and destructor definitions for the bunch of new
classes that were added in that revision.

* platform/graphics/nicosia/NicosiaPlatformLayer.cpp:
(Nicosia::ContentLayer::ContentLayer):
(Nicosia::BackingStore::BackingStore):
(Nicosia::ImageBacking::ImageBacking):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234595 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Nicosia] Add additional layer state classes, use impl-based approach to make them...
zandobersek@gmail.com [Mon, 6 Aug 2018 08:48:01 +0000 (08:48 +0000)]
[Nicosia] Add additional layer state classes, use impl-based approach to make them extendable
https://bugs.webkit.org/show_bug.cgi?id=188341

Reviewed by Carlos Garcia Campos.

Add the ContentLayer, BackingStore and ImageBacking classes in the
Nicosia namespace. State objects of CompositionLayer instances keep
references to objects of these classes depending on the content that's
presented in the associated GraphicsLayer.

ContentLayer derives from the PlatformLayer class. It's used for layers
that display things like WebGL and HTML5 canvas and media content in a
platform-specific way. In case of TextureMapper the hardware-accelerated
content is piped into that rendering pipeline.

BackingStore is meant to represent the painted contents of a layer. The
equivalent current functionality is using a tiled backing store that
has its contents copied into the CoordinatedBackingStore instance.

ImageBacking is used for a layer whose content is a simple Image object.
Image's pixel data is rasterized and again managed through
CoordinatedBackingStore for rendering.

All these classes, along with the CompositionLayer class, should now be
constructed with a factory function that returns an object that derives
the class-specific Impl interface. This will allow for simpler
implementation of different approaches in parallel. The TextureMapper
variants will be the first ones, replicating the current behavior as it
is implemented across classes in the CoordinatedGraphics and
TextureMapper layers.

* platform/graphics/nicosia/NicosiaPlatformLayer.cpp:
(Nicosia::CompositionLayer::CompositionLayer):
* platform/graphics/nicosia/NicosiaPlatformLayer.h:
(Nicosia::PlatformLayer::isContentLayer const):
(Nicosia::CompositionLayer::Impl::isTextureMapperImpl const):
(Nicosia::CompositionLayer::create):
(Nicosia::CompositionLayer::impl const):
(Nicosia::ContentLayer::Impl::isTextureMapperImpl const):
(Nicosia::ContentLayer::create):
(Nicosia::ContentLayer::impl const):
(Nicosia::BackingStore::Impl::isTextureMapperImpl const):
(Nicosia::BackingStore::create):
(Nicosia::BackingStore::impl const):
(Nicosia::ImageBacking::Impl::isTextureMapperImpl const):
(Nicosia::ImageBacking::create):
(Nicosia::ImageBacking::impl const):
* platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:
(WebCore::CoordinatedGraphicsLayer::CoordinatedGraphicsLayer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234594 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Nicosia] Add Nicosia::Scene
zandobersek@gmail.com [Mon, 6 Aug 2018 08:46:30 +0000 (08:46 +0000)]
[Nicosia] Add Nicosia::Scene
https://bugs.webkit.org/show_bug.cgi?id=188340

Reviewed by Carlos Garcia Campos.

Source/WebCore:

Add the Nicosia::Scene class that controls the scene's state and enables
thread-safe updates to that state.

The State struct itself holds a set of all the layers in the scene, and
separates out the root layer specifically. An ID counter is also present
there, allowing consumers of this state to easily check for any updates.

A reference to the Nicosia::Scene object now replaces the
HashSet-and-root-layer pair in the CoordinatedGraphicsState struct.

* platform/TextureMapper.cmake:
* platform/graphics/nicosia/NicosiaScene.cpp: Added.
* platform/graphics/nicosia/NicosiaScene.h: Added.
(Nicosia::Scene::create):
(Nicosia::Scene::accessState):
* platform/graphics/texmap/coordinated/CoordinatedGraphicsState.h:

Source/WebKit:

CompositingCoordinator spawns a Nicosia::Scene object that it shares
with the CoordinatedGraphicsSceneState instance. All the
Nicosia::CompositionLayer objects indirectly managed by
CompositingCoordinator are now stored in a local Nicosia::Scene::State
member object. Upon each flush that requires frame synchronization the
Nicosia::Scene object is updated in a thread-safe manner, increasing
the scene ID value as well as copying the local HashSet and root layer
values into the shared Scene object, allowing for the consumer (which
currently is the related CoordinatedGraphicsScene instance) to update
its output accordingly.

* WebProcess/WebPage/CoordinatedGraphics/CompositingCoordinator.cpp:
(WebKit::CompositingCoordinator::CompositingCoordinator):
(WebKit::CompositingCoordinator::flushPendingLayerChanges):
(WebKit::CompositingCoordinator::initializeRootCompositingLayerIfNeeded):
(WebKit::CompositingCoordinator::createGraphicsLayer):
(WebKit::CompositingCoordinator::detachLayer):
(WebKit::CompositingCoordinator::attachLayer):
* WebProcess/WebPage/CoordinatedGraphics/CompositingCoordinator.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234593 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[GTK] Buttons are drawn too large, text not centered
commit-queue@webkit.org [Mon, 6 Aug 2018 08:25:34 +0000 (08:25 +0000)]
[GTK] Buttons are drawn too large, text not centered
https://bugs.webkit.org/show_bug.cgi?id=185854

Patch by Carlos Eduardo Ramalho <cadubentzen@gmail.com> on 2018-08-06
Reviewed by Carlos Garcia Campos.

Some pages require small buttons and the GTK theme was
drawing only down to a minimum size.

No new tests required. Current layout tests and ManualTests/gtk/theme.html
already cover it.

* platform/gtk/RenderThemeGadget.cpp:
(WebCore::RenderThemeGadget::create):
(WebCore::RenderThemeButtonGadget::RenderThemeButtonGadget): Added
(WebCore::RenderThemeButtonGadget::minimumSize const):
Allow sizes smaller than the minimum size.
* platform/gtk/RenderThemeGadget.h:
* platform/gtk/RenderThemeWidget.cpp:
(WebCore::RenderThemeButton::RenderThemeButton):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234592 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoASSERTION !data.customElementReactionQueue() when creating custom element inside...
commit-queue@webkit.org [Mon, 6 Aug 2018 08:25:29 +0000 (08:25 +0000)]
ASSERTION !data.customElementReactionQueue() when creating custom element inside an SVG document
https://bugs.webkit.org/show_bug.cgi?id=187802

The assertion has been removed in bug 188327. This patch updates the test expectations and
add some non-regression test to verify different cases of XML custom element construction.

Patch by Frederic Wang <fwang@igalia.com> on 2018-08-06
Reviewed by Ryosuke Niwa.

* TestExpectations: This test no longer crashes but times out because of bug 187800.
* fast/custom-elements/xml-parsing-create-and-element-for-a-token-expected.txt: Added.
* fast/custom-elements/xml-parsing-create-and-element-for-a-token.svg: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234591 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoImplement self.queueMicrotask in Workers
utatane.tea@gmail.com [Mon, 6 Aug 2018 07:59:13 +0000 (07:59 +0000)]
Implement self.queueMicrotask in Workers
https://bugs.webkit.org/show_bug.cgi?id=188247

Reviewed by Ryosuke Niwa.

Source/WebCore:

This patch adds support for self.queueMicrotask to workers.

Tests: http/wpt/workers/queue-microtask.any.html
       http/wpt/workers/queue-microtask.any.worker.html

* bindings/js/JSWorkerGlobalScopeCustom.cpp:
(WebCore::JSWorkerGlobalScope::queueMicrotask):
* page/DOMWindow.idl:
Move queueMicrotask declaration to WindowOrWorkerGlobalScope.idl.
* page/WindowOrWorkerGlobalScope.idl:

LayoutTests:

* http/wpt/workers/queue-microtask.any-expected.txt: Added.
* http/wpt/workers/queue-microtask.any.html: Added.
* http/wpt/workers/queue-microtask.any.js: Added.
(promise_test):
(promise_test.):
(queueMicrotask.promise_test):
(test):
* http/wpt/workers/queue-microtask.any.worker-expected.txt: Added.
* http/wpt/workers/queue-microtask.any.worker.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234588 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[GTK][WPE] Fetch tests assert in SubresourceLoader::didReceiveResponse()
commit-queue@webkit.org [Mon, 6 Aug 2018 06:43:51 +0000 (06:43 +0000)]
[GTK][WPE] Fetch tests assert in SubresourceLoader::didReceiveResponse()
https://bugs.webkit.org/show_bug.cgi?id=188163

Patch by Rob Buis <rbuis@igalia.com> on 2018-08-05
Reviewed by Frédéric Wang.

Source/WebCore:

Fetch in manual redirect mode uses didReceiveResponse instead of
willSendRequestInternal, so update the ASSERT.

* loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::didReceiveResponse):

LayoutTests:

Unskip since this will not crash in Debug anymore.

* platform/wpe/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234587 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoAdd support for microtasks in workers
utatane.tea@gmail.com [Mon, 6 Aug 2018 04:37:10 +0000 (04:37 +0000)]
Add support for microtasks in workers
https://bugs.webkit.org/show_bug.cgi?id=188246

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

This test is wrong[1,2]. It starts failing since our worker starts using correct microtask queues.

[1]: https://github.com/web-platform-tests/wpt/pull/8936
[2]: https://github.com/w3c/ServiceWorker/issues/1213#issuecomment-342640579

* web-platform-tests/service-workers/service-worker/extendable-event-async-waituntil.https-expected.txt:

Source/WebCore:

This patch adds the microtask mechanism to workers. To adopt the existing microtask mechanism from the main thread,
we extend JSMainThreadExecState for non-main-threads. We rename it to JSExecState, and store stacked
ExecState* data in thread local storage in ThreadGlobalData instead of a static variable s_mainThreadState.

We add MicrotaskQueue to WorkerGlobalScope since each worker has its own thread and it should have its
own microtask queue.

* Modules/encryptedmedia/legacy/LegacyCDMSessionClearKey.cpp:
* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSBindingsAllInOne.cpp:
* bindings/js/JSCallbackData.cpp:
(WebCore::JSCallbackData::invokeCallback):
* bindings/js/JSCustomElementInterface.cpp:
(WebCore::constructCustomElementSynchronously):
(WebCore::JSCustomElementInterface::upgradeElement):
(WebCore::JSCustomElementInterface::invokeCallback):
* bindings/js/JSCustomXPathNSResolver.cpp:
(WebCore::JSCustomXPathNSResolver::lookupNamespaceURI):
* bindings/js/JSDOMGlobalObjectTask.cpp:
* bindings/js/JSDOMWindowBase.cpp:
(WebCore::JSDOMWindowBase::queueTaskToEventLoop):
Queue a microtask to MicrotaskQueue instead of posting a macrotask.

(WebCore::JSDOMWindowMicrotaskCallback::create): Deleted.
(WebCore::JSDOMWindowMicrotaskCallback::call): Deleted.
(WebCore::JSDOMWindowMicrotaskCallback::JSDOMWindowMicrotaskCallback): Deleted.
(): Deleted.
Extract JSDOMWindowMicrotaskCallback as JSMicrotaskCallback and create a new file for it.

* bindings/js/JSErrorHandler.cpp:
(WebCore::JSErrorHandler::handleEvent):
* bindings/js/JSEventListener.cpp:
(WebCore::JSEventListener::handleEvent):
* bindings/js/JSExecState.cpp: Renamed from Source/WebCore/bindings/js/JSMainThreadExecState.cpp.
(WebCore::JSExecState::didLeaveScriptContext):
If we are in a main thread, we consume main thread microtask queue. If we are in worker thread,
we consume a microtask queue per worker.

(WebCore::functionCallHandlerFromAnyThread):
(WebCore::evaluateHandlerFromAnyThread):
* bindings/js/JSExecState.h: Renamed from Source/WebCore/bindings/js/JSMainThreadExecState.h.
(WebCore::JSExecState::currentState):
(WebCore::JSExecState::call):
(WebCore::JSExecState::evaluate):
(WebCore::JSExecState::profiledCall):
(WebCore::JSExecState::profiledEvaluate):
(WebCore::JSExecState::runTask):
(WebCore::JSExecState::loadModule):
(WebCore::JSExecState::linkAndEvaluateModule):
(WebCore::JSExecState::JSExecState):
(WebCore::JSExecState::~JSExecState):
(WebCore::JSExecState::setCurrentState):
Store and load ExecState in thread local storage, ThreadGlobalData. This allows us to use it for workers.

(WebCore::JSMainThreadNullState::JSMainThreadNullState):
(WebCore::JSMainThreadNullState::~JSMainThreadNullState):
We keep this name "JSMainThreadNullState" since CustomElementReactionStack should be stick to the main thread.
And this class is only used in the main thread.

* bindings/js/JSExecStateInstrumentation.h: Renamed from Source/WebCore/bindings/js/JSMainThreadExecStateInstrumentation.h.
(WebCore::JSExecState::instrumentFunctionInternal):
(WebCore::JSExecState::instrumentFunctionCall):
(WebCore::JSExecState::instrumentFunctionConstruct):
* bindings/js/JSMicrotaskCallback.h: Copied from Source/WebKitLegacy/mac/DOM/DOMHTMLBaseElement.mm.
(WebCore::JSMicrotaskCallback::create):
(WebCore::JSMicrotaskCallback::call):
(WebCore::JSMicrotaskCallback::JSMicrotaskCallback):
* bindings/js/JSWorkerGlobalScopeBase.cpp:
(WebCore::JSWorkerGlobalScopeBase::queueTaskToEventLoop):
* bindings/js/ScheduledAction.cpp:
(WebCore::ScheduledAction::executeFunctionInContext):
* bindings/js/ScriptController.cpp:
(WebCore::ScriptController::evaluateInWorld):
(WebCore::ScriptController::loadModuleScriptInWorld):
(WebCore::ScriptController::linkAndEvaluateModuleScriptInWorld):
(WebCore::ScriptController::canAccessFromCurrentOrigin):
* bindings/js/WorkerScriptController.cpp:
(WebCore::WorkerScriptController::evaluate):
* bridge/objc/WebScriptObject.mm:
(-[WebScriptObject callWebScriptMethod:withArguments:]):
(-[WebScriptObject evaluateWebScript:]):
* html/canvas/WebGLRenderingContextBase.cpp:
(WebCore::WebGLRenderingContextBase::compileShader):
(WebCore::WebGLRenderingContextBase::printToConsole):
* inspector/InspectorCanvas.cpp:
(WebCore::InspectorCanvas::buildObjectForCanvas):
(WebCore::InspectorCanvas::buildAction):
* inspector/InspectorController.cpp:
* inspector/InspectorFrontendHost.cpp:
* inspector/TimelineRecordFactory.cpp:
(WebCore::TimelineRecordFactory::createGenericRecord):
* inspector/WorkerInspectorController.cpp:
* inspector/agents/InspectorCanvasAgent.cpp:
* inspector/agents/InspectorNetworkAgent.cpp:
(WebCore::InspectorNetworkAgent::buildInitiatorObject):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::postMessage):
* page/PageConsoleClient.cpp:
(WebCore::PageConsoleClient::addMessage):
* page/csp/ContentSecurityPolicy.cpp:
(WebCore::ContentSecurityPolicy::reportViolation const):
* platform/ThreadGlobalData.h:
(WebCore::ThreadGlobalData::ThreadGlobalData::currentState const):
(WebCore::ThreadGlobalData::ThreadGlobalData::setCurrentState):
* workers/WorkerGlobalScope.cpp:
(WebCore::WorkerGlobalScope::WorkerGlobalScope):
(WebCore::WorkerGlobalScope::removeMicrotaskQueue):
* workers/WorkerGlobalScope.h:
(WebCore::WorkerGlobalScope::microtaskQueue const):
* workers/WorkerThread.cpp:
(WebCore::WorkerThread::stop):
* workers/service/ExtendableEvent.cpp:
(WebCore::ExtendableEvent::addExtendLifetimePromise):
When dispatching an "install" event from service worker, we first create an event,
dispatch it, and set a handler which is called when a pending promise count becomes zero.
However, the old code checked pending promise count in a queued microtask. It worked
previously because microtask is actually a macrotask in the service worker. So this check
is done after a handler is set. But this patch introduces real microtask, and this check
happens before a handler is set because dispatching an event can exhaust microtask queue.
According to the spec, this check should not be done in microtask[1]. We make this checking
part as a macrotask. We note that algorithm noted as FIXMEs should be done in this
microtask while the checking part should not be done.

[1]: https://w3c.github.io/ServiceWorker/#installation-algorithm

Source/WebKit:

Rename JSMainThreadExecState.h to JSExecState.h.

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMAttr.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMBlob.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMCDATASection.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMCSSRule.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMCSSRuleList.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMCSSStyleDeclaration.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMCSSStyleSheet.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMCSSValue.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMCharacterData.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMClientRect.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMClientRectList.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMComment.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDOMImplementation.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDOMSelection.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDOMTokenList.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDOMWindow.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDeprecated.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDocumentFragment.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDocumentGtk.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDocumentType.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMElementGtk.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMEvent.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMFile.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMFileList.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLAnchorElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLAppletElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLAreaElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLBRElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLBaseElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLBodyElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLButtonElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLCanvasElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLCollection.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLDListElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLDirectoryElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLDivElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLDocument.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLEmbedElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLFieldSetElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLFontElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLFormElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLFrameElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLFrameSetElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLHRElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLHeadElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLHeadingElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLHtmlElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLIFrameElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLImageElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLInputElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLLIElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLLabelElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLLegendElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLLinkElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLMapElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLMarqueeElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLMenuElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLMetaElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLModElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLOListElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLObjectElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLOptGroupElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLOptionElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLOptionsCollection.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLParagraphElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLParamElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLPreElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLQuoteElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLScriptElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLSelectElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLStyleElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLTableCaptionElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLTableCellElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLTableColElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLTableElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLTableRowElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLTableSectionElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLTextAreaElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLTitleElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMHTMLUListElement.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMKeyboardEvent.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMMediaList.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMMouseEvent.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMNamedNodeMap.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMNodeGtk.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMNodeIterator.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMNodeList.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMProcessingInstruction.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMRange.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMStyleSheet.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMStyleSheetList.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMText.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMTreeWalker.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMUIEvent.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMWheelEvent.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMXPathExpression.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMXPathNSResolver.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMXPathResult.cpp:

Source/WebKitLegacy/mac:

Rename JSMainThreadExecState.h to JSExecState.h.

* DOM/DOMAttr.mm:
* DOM/DOMBlob.mm:
* DOM/DOMCDATASection.mm:
* DOM/DOMCSSCharsetRule.mm:
* DOM/DOMCSSFontFaceRule.mm:
* DOM/DOMCSSImportRule.mm:
* DOM/DOMCSSMediaRule.mm:
* DOM/DOMCSSPageRule.mm:
* DOM/DOMCSSPrimitiveValue.mm:
* DOM/DOMCSSRule.mm:
* DOM/DOMCSSRuleList.mm:
* DOM/DOMCSSStyleDeclaration.mm:
* DOM/DOMCSSStyleRule.mm:
* DOM/DOMCSSStyleSheet.mm:
* DOM/DOMCSSUnknownRule.mm:
* DOM/DOMCSSValue.mm:
* DOM/DOMCSSValueList.mm:
* DOM/DOMCharacterData.mm:
* DOM/DOMComment.mm:
* DOM/DOMCounter.mm:
* DOM/DOMDocument.mm:
* DOM/DOMDocumentFragment.mm:
* DOM/DOMDocumentType.mm:
* DOM/DOMElement.mm:
* DOM/DOMEvent.mm:
* DOM/DOMFile.mm:
* DOM/DOMFileList.mm:
* DOM/DOMHTML.mm:
* DOM/DOMHTMLAnchorElement.mm:
* DOM/DOMHTMLAppletElement.mm:
* DOM/DOMHTMLAreaElement.mm:
* DOM/DOMHTMLBRElement.mm:
* DOM/DOMHTMLBaseElement.mm:
* DOM/DOMHTMLBaseFontElement.mm:
* DOM/DOMHTMLBodyElement.mm:
* DOM/DOMHTMLButtonElement.mm:
* DOM/DOMHTMLCanvasElement.mm:
* DOM/DOMHTMLCollection.mm:
* DOM/DOMHTMLDListElement.mm:
* DOM/DOMHTMLDirectoryElement.mm:
* DOM/DOMHTMLDivElement.mm:
* DOM/DOMHTMLDocument.mm:
* DOM/DOMHTMLElement.mm:
* DOM/DOMHTMLEmbedElement.mm:
* DOM/DOMHTMLFieldSetElement.mm:
* DOM/DOMHTMLFontElement.mm:
* DOM/DOMHTMLFormElement.mm:
* DOM/DOMHTMLFrameElement.mm:
* DOM/DOMHTMLFrameSetElement.mm:
* DOM/DOMHTMLHRElement.mm:
* DOM/DOMHTMLHeadElement.mm:
* DOM/DOMHTMLHeadingElement.mm:
* DOM/DOMHTMLHtmlElement.mm:
* DOM/DOMHTMLIFrameElement.mm:
* DOM/DOMHTMLImageElement.mm:
* DOM/DOMHTMLInputElement.mm:
* DOM/DOMHTMLLIElement.mm:
* DOM/DOMHTMLLabelElement.mm:
* DOM/DOMHTMLLegendElement.mm:
* DOM/DOMHTMLLinkElement.mm:
* DOM/DOMHTMLMapElement.mm:
* DOM/DOMHTMLMarqueeElement.mm:
* DOM/DOMHTMLMediaElement.mm:
* DOM/DOMHTMLMenuElement.mm:
* DOM/DOMHTMLMetaElement.mm:
* DOM/DOMHTMLModElement.mm:
* DOM/DOMHTMLOListElement.mm:
* DOM/DOMHTMLObjectElement.mm:
* DOM/DOMHTMLOptGroupElement.mm:
* DOM/DOMHTMLOptionElement.mm:
* DOM/DOMHTMLOptionsCollection.mm:
* DOM/DOMHTMLParagraphElement.mm:
* DOM/DOMHTMLParamElement.mm:
* DOM/DOMHTMLPreElement.mm:
* DOM/DOMHTMLQuoteElement.mm:
* DOM/DOMHTMLScriptElement.mm:
* DOM/DOMHTMLSelectElement.mm:
* DOM/DOMHTMLStyleElement.mm:
* DOM/DOMHTMLTableCaptionElement.mm:
* DOM/DOMHTMLTableCellElement.mm:
* DOM/DOMHTMLTableColElement.mm:
* DOM/DOMHTMLTableElement.mm:
* DOM/DOMHTMLTableRowElement.mm:
* DOM/DOMHTMLTableSectionElement.mm:
* DOM/DOMHTMLTextAreaElement.mm:
* DOM/DOMHTMLTitleElement.mm:
* DOM/DOMHTMLUListElement.mm:
* DOM/DOMHTMLVideoElement.mm:
* DOM/DOMImplementation.mm:
* DOM/DOMKeyboardEvent.mm:
* DOM/DOMMediaError.mm:
* DOM/DOMMediaList.mm:
* DOM/DOMMouseEvent.mm:
* DOM/DOMMutationEvent.mm:
* DOM/DOMNamedNodeMap.mm:
* DOM/DOMNode.mm:
* DOM/DOMNodeIterator.mm:
* DOM/DOMNodeList.mm:
* DOM/DOMOverflowEvent.mm:
* DOM/DOMProcessingInstruction.mm:
* DOM/DOMProgressEvent.mm:
* DOM/DOMRGBColor.mm:
* DOM/DOMRange.mm:
* DOM/DOMRect.mm:
* DOM/DOMStyleSheet.mm:
* DOM/DOMStyleSheetList.mm:
* DOM/DOMText.mm:
* DOM/DOMTextEvent.mm:
* DOM/DOMTimeRanges.mm:
* DOM/DOMTokenList.mm:
* DOM/DOMTreeWalker.mm:
* DOM/DOMUIEvent.mm:
* DOM/DOMWheelEvent.mm:
* DOM/DOMXPathExpression.mm:
* DOM/DOMXPathResult.mm:
* DOM/ObjCEventListener.mm:

LayoutTests:

* http/wpt/workers/microtasks.any-expected.txt: Added.
* http/wpt/workers/microtasks.any.html: Added.
* http/wpt/workers/microtasks.any.js: Added.
(promise_test):
* http/wpt/workers/microtasks.any.worker-expected.txt: Added.
* http/wpt/workers/microtasks.any.worker.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234586 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoProperties set on window.customElements can disappear due to GC
rniwa@webkit.org [Mon, 6 Aug 2018 00:44:55 +0000 (00:44 +0000)]
Properties set on window.customElements can disappear due to GC
https://bugs.webkit.org/show_bug.cgi?id=172575

Lexicologically sort the IDL attributes per Darin's comment.

* dom/CustomElementRegistry.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234585 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][BFC] contentHeightForFormattingContextRoot uses the wrong coordinate system...
zalan@apple.com [Sun, 5 Aug 2018 23:19:57 +0000 (23:19 +0000)]
[LFC][BFC] contentHeightForFormattingContextRoot uses the wrong coordinate system to compute the height.
https://bugs.webkit.org/show_bug.cgi?id=188319

Reviewed by Antti Koivisto.

Display::Box::marginBox() is in the coordinate system of the box itself (8px top margin translates to y = -8px).
What we need instead to compute the content height is the top/bottom position in the containing block's coordinate system (rect and rectWithMargin).

* layout/FormattingContextGeometry.cpp:
(WebCore::Layout::contentHeightForFormattingContextRoot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234584 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[GStreamer] Remove unsound assertions in MediaPlayerPrivateGStreamerBase.
Ms2ger@igalia.com [Sun, 5 Aug 2018 10:10:10 +0000 (10:10 +0000)]
[GStreamer] Remove unsound assertions in MediaPlayerPrivateGStreamerBase.
https://bugs.webkit.org/show_bug.cgi?id=188162

Reviewed by Philippe Normand.

These assertions sometimes fail, and there's a runtime check right
after them anyway.

Tests: fast/canvas/webgl/texImage2D-video-flipY-false.html
       fast/canvas/webgl/texImage2D-video-flipY-true.html

* platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:
(WebCore::MediaPlayerPrivateGStreamerBase::copyVideoTextureToPlatformTexture):
(WebCore::MediaPlayerPrivateGStreamerBase::nativeImageForCurrentTime):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234583 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[MediaCapabilities] Platform integration
philn@webkit.org [Sun, 5 Aug 2018 09:26:00 +0000 (09:26 +0000)]
[MediaCapabilities] Platform integration
https://bugs.webkit.org/show_bug.cgi?id=187850

Reviewed by Eric Carlson.

Source/WebCore:

Mock platform support for the MediaCapabilities specification.

Test: media/mediacapabilities/mock-decodingInfo.html
      media/mediacapabilities/mock-encodingInfo.html

* Modules/mediacapabilities/MediaCapabilities.cpp:
(WebCore::MediaCapabilities::decodingInfo):
(WebCore::MediaCapabilities::encodingInfo):
* Modules/mediacapabilities/MediaCapabilitiesInfo.h:
(WebCore::MediaCapabilitiesInfo::create):
* PlatformGTK.cmake:
* PlatformMac.cmake:
* PlatformWPE.cmake:
* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
* platform/mediacapabilities/MediaEngineConfiguration.cpp: Added.
(WebCore::MediaEngineVideoConfiguration::MediaEngineVideoConfiguration):
(WebCore::MediaEngineAudioConfiguration::MediaEngineAudioConfiguration):
(WebCore::MediaEngineConfiguration::MediaEngineConfiguration):
* platform/mediacapabilities/MediaEngineConfiguration.h: Added.
(WebCore::MediaEngineVideoConfiguration::create):
(WebCore::MediaEngineVideoConfiguration::contentType const):
(WebCore::MediaEngineVideoConfiguration::size const):
(WebCore::MediaEngineVideoConfiguration::bitrate const):
(WebCore::MediaEngineVideoConfiguration::framerate const):
(WebCore::MediaEngineAudioConfiguration::create):
(WebCore::MediaEngineAudioConfiguration::contentType const):
(WebCore::MediaEngineAudioConfiguration::channels const):
(WebCore::MediaEngineAudioConfiguration::bitrate const):
(WebCore::MediaEngineAudioConfiguration::samplerate const):
(WebCore::MediaEngineConfiguration::audioConfiguration const):
(WebCore::MediaEngineConfiguration::videoConfiguration const):
* platform/mediacapabilities/MediaEngineConfigurationFactory.cpp: Added.
(WebCore::mockEnabled):
(WebCore::MediaEngineConfigurationFactory::createDecodingConfiguration):
(WebCore::MediaEngineConfigurationFactory::createEncodingConfiguration):
(WebCore::MediaEngineConfigurationFactory::enableMock):
(WebCore::MediaEngineConfigurationFactory::disableMock):
* platform/mediacapabilities/MediaEngineConfigurationFactory.h: Added.
* platform/mediacapabilities/MediaEngineDecodingConfiguration.h: Added.
(WebCore::MediaEngineDecodingConfiguration::MediaEngineDecodingConfiguration):
(WebCore::MediaEngineDecodingConfiguration::canDecodeMedia):
(WebCore::MediaEngineDecodingConfiguration::canSmoothlyDecodeMedia):
(WebCore::MediaEngineDecodingConfiguration::canPowerEfficientlyDecodeMedia):
(WebCore::MediaEngineDecodingConfiguration::decodingType const):
* platform/mediacapabilities/MediaEngineEncodingConfiguration.h: Added.
(WebCore::MediaEngineEncodingConfiguration::MediaEngineEncodingConfiguration):
(WebCore::MediaEngineEncodingConfiguration::canEncodeMedia):
(WebCore::MediaEngineEncodingConfiguration::canSmoothlyEncodeMedia):
(WebCore::MediaEngineEncodingConfiguration::canPowerEfficientlyEncodeMedia):
(WebCore::MediaEngineEncodingConfiguration::encodingType const):
* platform/mock/MediaEngineDecodingConfigurationMock.cpp: Added.
(WebCore::MediaEngineDecodingConfigurationMock::canDecodeMedia):
(WebCore::MediaEngineDecodingConfigurationMock::canSmoothlyDecodeMedia):
(WebCore::MediaEngineDecodingConfigurationMock::canPowerEfficientlyDecodeMedia):
* platform/mock/MediaEngineDecodingConfigurationMock.h: Added.
* platform/mock/MediaEngineEncodingConfigurationMock.cpp: Added.
(WebCore::MediaEngineEncodingConfigurationMock::canEncodeMedia):
(WebCore::MediaEngineEncodingConfigurationMock::canSmoothlyEncodeMedia):
(WebCore::MediaEngineEncodingConfigurationMock::canPowerEfficientlyEncodeMedia):
* platform/mock/MediaEngineEncodingConfigurationMock.h: Added.
* testing/Internals.cpp:
(WebCore::Internals::resetToConsistentState):
(WebCore::Internals::enableMockMediaCapabilities):
* testing/Internals.h:
* testing/Internals.idl:

LayoutTests:

Mock tests for the MediaCapabilities specification.

* media/mediacapabilities/mock-decodingInfo-expected.txt: Added.
* media/mediacapabilities/mock-decodingInfo.html: Added.
* media/mediacapabilities/mock-encodingInfo-expected.txt: Added.
* media/mediacapabilities/mock-encodingInfo.html: Added.
* media/video-test.js:
(checkMediaCapabilitiesInfo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234582 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoMake WebCore::Timer more space-efficient
simon.fraser@apple.com [Sat, 4 Aug 2018 22:57:13 +0000 (22:57 +0000)]
Make WebCore::Timer more space-efficient
https://bugs.webkit.org/show_bug.cgi?id=187455

Reviewed by Brent Fulgham.

Steal a bit from m_heapIndex for m_wasDeleted to eliminate padding in Timer.
This reduces the size of Document, which has 8 Timers, from 3360 to 3296 bytes (saving 64 bytes).

* platform/Timer.cpp:
(WebCore::TimerBase::TimerBase):
* platform/Timer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234581 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoREGRESSION (r208953): TemplateObjectDescriptor constructor calculates m_hash on use...
ddkilzer@apple.com [Sat, 4 Aug 2018 20:07:13 +0000 (20:07 +0000)]
REGRESSION (r208953): TemplateObjectDescriptor constructor calculates m_hash on use-after-move variable
<https://webkit.org/b/188331>

Reviewed by Yusuke Suzuki.

* runtime/TemplateObjectDescriptor.h:
(JSC::TemplateObjectDescriptor::TemplateObjectDescriptor):
Use `m_rawstrings` instead of `rawStrings` to calculate hash.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234580 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoSort the WebCore project file
simon.fraser@apple.com [Sat, 4 Aug 2018 17:38:13 +0000 (17:38 +0000)]
Sort the WebCore project file

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234579 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoProperties set on window.customElements can disappear due to GC
rniwa@webkit.org [Sat, 4 Aug 2018 09:02:44 +0000 (09:02 +0000)]
Properties set on window.customElements can disappear due to GC
https://bugs.webkit.org/show_bug.cgi?id=172575
<rdar://problem/32440668>

Reviewed by Saam Barati.

Source/WebCore:

Fixed the bug that JS wrapper of CustomElementsRegistry can erroneously get collected during GC
by keeping it alive as long as the global object is alive.

Test: fast/custom-elements/custom-element-registry-wrapper-should-stay-alive.html

* dom/CustomElementRegistry.cpp:
(WebCore::CustomElementRegistry::create):
(WebCore::CustomElementRegistry::CustomElementRegistry):
* dom/CustomElementRegistry.h:
(WebCore::CustomElementRegistry): Make this inherited from ContextDestructionObserver.
* dom/CustomElementRegistry.idl: Set GenerateIsReachable=ImplScriptExecutionContext in IDL. This will
make CustomElementRegistry reachable from the global object.
* page/DOMWindow.cpp:
(WebCore::DOMWindow::ensureCustomElementRegistry):

LayoutTests:

Added a regression test.

* fast/custom-elements/custom-element-registry-wrapper-should-stay-alive-expected.txt: Added.
* fast/custom-elements/custom-element-registry-wrapper-should-stay-alive.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234578 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoinnerHTML should not synchronously create a custom element
rniwa@webkit.org [Sat, 4 Aug 2018 06:35:37 +0000 (06:35 +0000)]
innerHTML should not synchronously create a custom element
https://bugs.webkit.org/show_bug.cgi?id=188327
<rdar://problem/42923114>

Reviewed by Daniel Bates.

LayoutTests/imported/w3c:

Rebaselined the test now that all test cases are passing.

* web-platform-tests/custom-elements/connected-callbacks-html-fragment-parsing-expected.txt:

Source/WebCore:

Fixed the bug that the fragment parsing algorithm was synchronously constructing a custom element instead of
enqueuing an element to upgrade.

The fragment parsing algorithm creates an element for a token with *will execute script* flag set to false:
https://html.spec.whatwg.org/multipage/parsing.html#create-an-element-for-the-token
which results in creating an element with synchronous custom elements flag *not* set:
https://dom.spec.whatwg.org/#concept-create-element

When synchronous custom elements flag is false, we're supposed to create an element and enqueue a custom element
upgrade reaction. createHTMLElementOrFindCustomElementInterface was missing this last logic.

Also fixed a bug that Element::enqueueToUpgrade would hit a debug assertion when a custom element which has been
enqueued to upgrade is enqueued to upgrade for the second time. In this case, we need to put the element into the
current element queue (https://html.spec.whatwg.org/multipage/custom-elements.html#current-element-queue) again.

While the specification simply enqueues another upgrade reaction and bails out immediately in the first step of
the upgrade, WebKit's implementation simply avoids this redundancy in the first place:
https://html.spec.whatwg.org/multipage/custom-elements.html#concept-upgrade-an-element

Existing tests such as imported/w3c/web-platform-tests/custom-elements/reactions/Document.html exercises this
code path after the fragment parsing algorithm fix.

Tests: imported/w3c/web-platform-tests/custom-elements/connected-callbacks-html-fragment-parsing.html

* dom/CustomElementReactionQueue.cpp:
(WebCore::CustomElementReactionQueueItem::type const): Added for an assertion.
(WebCore::CustomElementReactionQueue::enqueueElementUpgrade): Enqueue this element to the current element queue
by calling ensureCurrentQueue and avoid inserting a redundant upgrade reaction.
* dom/CustomElementReactionQueue.h:
* dom/Element.cpp:
(WebCore::Element::enqueueToUpgrade): Handle the case when a custom element is enqueued to upgrade for the second
time while it had been waiting in some element queue. In this case, the reaction queue for this element has
already been created and we simply need to put this element back into the current element queue (i.e. this element
now belongs to both element queues).
* html/parser/HTMLConstructionSite.cpp:
(WebCore::findCustomElementInterface): Extracted out of createHTMLElementOrFindCustomElementInterface.
(WebCore::HTMLConstructionSite::createHTMLElementOrFindCustomElementInterface): Fixed the bug that the HTML parser
was synchronously constructing a custom element even for the fragment parsing algorithm.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234577 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoGive the `jsc` shell the JIT entitlement
sbarati@apple.com [Sat, 4 Aug 2018 06:27:23 +0000 (06:27 +0000)]
Give the `jsc` shell the JIT entitlement
https://bugs.webkit.org/show_bug.cgi?id=188324
<rdar://problem/42885806>

Reviewed by Dan Bernstein.

This should help us in ensuring the system jsc is able to JIT.

* Configurations/JSC.xcconfig:
* JavaScriptCore.xcodeproj/project.pbxproj:
* allow-jit-macOS.entitlements: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234576 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoWe should cache the compiled sandbox profile in a data vault
commit-queue@webkit.org [Sat, 4 Aug 2018 01:09:37 +0000 (01:09 +0000)]
We should cache the compiled sandbox profile in a data vault
https://bugs.webkit.org/show_bug.cgi?id=184991

Patch by Ben Richards <benton_richards@apple.com> on 2018-08-03
Reviewed by Ryosuke Niwa.

Source/WebCore:

Added functionality to FileHandle so that it can lock a file while open.
Added a function to FileSystem to delete non empty directories.

* platform/FileHandle.cpp:
(WebCore::FileHandle::FileHandle):
(WebCore::FileHandle::open):
(WebCore::FileHandle::close):
* platform/FileHandle.h:
* platform/FileSystem.h:
* platform/cocoa/FileSystemCocoa.mm:
(WebCore::FileSystem::deleteNonEmptyDirectory):

Source/WebKit:

This patch changes a few things (note: data vaults and sandbox entitlements are only used in internal builds):
(1) Instead of compiling a sandbox every time a process is launched, processes now look for a cached sandbox
    in a process specific data vault on macOS platforms. (ChildProcessMac.mm)
(2) If a valid cached sandbox is not found, a process will create the data vault (or ensure that it exists),
    compile a sandbox, and cache it.
(3) In order to create process specific data vaults, each process now has their own <process name>-OSX-sandbox.entitlements
    file which contains an entitlement with a process specific "storage class" which ensures that each process
    can only ever access its own data vault. (See the article on confluence "Data Vaults and Restricted Files" for more info)
(4) The sandbox entitlements file for the Network, WebContent and Plugin services are loaded dynamically
    through Scripts/<process name>-process-entitlements.sh which is triggered in a new build phase for each service.
    The Storage process sandbox entitlements are loaded directly in Configurations/StorageService.xcconfig.
    The reason that the sandbox entitlements are applied dynamically is so that these sandbox entitlements
    are only applied when WK_USE_RESTRICTED_ENTITLEMENTS is YES. This means that open source builds will still work.

* Configurations/Network-OSX-sandbox.entitlements: Added.
* Configurations/Plugin-OSX-sandbox.entitlements: Added.
* Configurations/Storage-OSX-sandbox.entitlements: Added.
* Configurations/StorageService.xcconfig:
* Configurations/WebContent-OSX-sandbox.entitlements: Added.
* Configurations/WebKit.xcconfig:
* NetworkProcess/NetworkProcess.h:
* PluginProcess/PluginProcess.h:
* Scripts/process-network-entitlements.sh: Added.
* Scripts/process-plugin-entitlements.sh: Added.
* Scripts/process-webcontent-entitlements.sh:
* Shared/ChildProcess.h:
* Shared/EntryPointUtilities/mac/XPCService/XPCServiceEntryPoint.h:
(WebKit::XPCServiceInitializer):
* Shared/SandboxInitializationParameters.h:
(WebKit::SandboxInitializationParameters::setOverrideSandboxProfilePath):
(WebKit::SandboxInitializationParameters::overrideSandboxProfilePath const):
(WebKit::SandboxInitializationParameters::setSandboxProfile):
(WebKit::SandboxInitializationParameters::sandboxProfile const):
(): Deleted.
* Shared/mac/ChildProcessMac.mm:
(WebKit::SandboxProfileDeleter::operator()):
(WebKit::SandboxParametersDeleter::operator()):
(WebKit::SandboxInfo::SandboxInfo):
(WebKit::fileContents):
(WebKit::processStorageClass):
(WebKit::setAndSerializeSandboxParameters):
(WebKit::getUserCacheDirectory):
(WebKit::sandboxDataVaultParentDirectory):
(WebKit::sandboxDirectory):
(WebKit::sandboxFilePath):
(WebKit::ensureSandboxCacheDirectory):
(WebKit::writeSandboxDataToCacheFile):
(WebKit::compileAndCacheSandboxProfile):
(WebKit::tryApplyCachedSandbox):
(WebKit::webKit2Bundle):
(WebKit::sandboxProfilePath):
(WebKit::compileAndApplySandboxSlowCase):
(WebKit::applySandbox):
(WebKit::initializeSandboxParameters):
(WebKit::ChildProcess::initializeSandbox):
* Shared/mac/SandboxInitialiationParametersMac.mm:
(WebKit::SandboxInitializationParameters::SandboxInitializationParameters):
* StorageProcess/StorageProcess.h:
* WebKit.xcodeproj/project.pbxproj:
* WebProcess/WebProcess.h:

Source/WTF:

Added trace points for sandbox initialization and exposed functions needed for sandbox caching

* wtf/SystemTracing.h:
* wtf/spi/darwin/SandboxSPI.h:

Tools:

Added trace points for sandbox initialization

* Tracing/SystemTracePoints.plist:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234569 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoWebGL 2 conformance: vertex_arrays/vertex_array_object.html
commit-queue@webkit.org [Sat, 4 Aug 2018 00:11:49 +0000 (00:11 +0000)]
WebGL 2 conformance: vertex_arrays/vertex_array_object.html
https://bugs.webkit.org/show_bug.cgi?id=188291
<rdar://problem/42792709>

Source/WebCore:

Patch by Justin Fan <justin_fan@apple.com> on 2018-08-03
Reviewed by Dean Jackson.

Get https://www.khronos.org/registry/webgl/conformance-suites/2.0.0/conformance2/vertex_arrays/vertex-array-object.html
working on Mac OS and iOS, and check in the new test expectations.

Existing test: webgl/2.0.0/conformance2/vertex_arrays/vertex-array-object.html

* html/canvas/WebGL2RenderingContext.cpp:
(WebCore::WebGL2RenderingContext::initializeVertexArrayObjects):
(WebCore::WebGL2RenderingContext::deleteVertexArray):
* html/canvas/WebGLRenderingContextBase.cpp:
(WebCore::WebGLRenderingContextBase::deleteObject):
* html/canvas/WebGLVertexArrayObject.cpp:
(WebCore::WebGLVertexArrayObject::WebGLVertexArrayObject):
* platform/graphics/opengl/GraphicsContext3DOpenGLCommon.cpp:
(WebCore::GraphicsContext3D::checkVaryingsPacking const):
(WebCore::GraphicsContext3D::createVertexArray):
(WebCore::GraphicsContext3D::deleteVertexArray):
(WebCore::GraphicsContext3D::isVertexArray):
(WebCore::GraphicsContext3D::bindVertexArray):

LayoutTests:

Get https://www.khronos.org/registry/webgl/conformance-suites/2.0.0/conformance2/vertex_arrays/vertex-array-object.html
working on Mac OS and iOS, and check in the new test expectations.

Patch by Justin Fan <justin_fan@apple.com> on 2018-08-03
Reviewed by Dean Jackson.

* TestExpectations: No longer skipping vertex_arrays tests.
The following FAILed before this patch but are now expecting PASSes.
* webgl/2.0.0/conformance2/glsl3/array-as-return-value-expected.txt:
* webgl/2.0.0/conformance2/glsl3/array-assign-constructor-expected.txt:
* webgl/2.0.0/conformance2/glsl3/array-assign-expected.txt:
* webgl/2.0.0/conformance2/glsl3/array-complex-indexing-expected.txt:
* webgl/2.0.0/conformance2/glsl3/array-element-increment-expected.txt:
* webgl/2.0.0/conformance2/glsl3/array-equality-expected.txt:
* webgl/2.0.0/conformance2/glsl3/array-in-complex-expression-expected.txt:
* webgl/2.0.0/conformance2/glsl3/bool-type-cast-bug-uint-ivec-uvec-expected.txt:
* webgl/2.0.0/conformance2/glsl3/compare-structs-containing-arrays-expected.txt:
* webgl/2.0.0/conformance2/glsl3/const-array-init-expected.txt:
* webgl/2.0.0/conformance2/glsl3/frag-depth-expected.txt:
* webgl/2.0.0/conformance2/glsl3/loops-with-side-effects-expected.txt:
* webgl/2.0.0/conformance2/glsl3/no-attribute-vertex-shader-expected.txt:
* webgl/2.0.0/conformance2/glsl3/short-circuiting-in-loop-condition-expected.txt:
* webgl/2.0.0/conformance2/glsl3/tricky-loop-conditions-expected.txt:
* webgl/2.0.0/conformance2/glsl3/vector-dynamic-indexing-expected.txt:
* webgl/2.0.0/conformance2/glsl3/vector-dynamic-indexing-nv-driver-bug-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234566 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoSkip imported/w3c/web-platform-tests/FileAPI/reading-data-section/filereader_abort...
ryanhaddad@apple.com [Fri, 3 Aug 2018 23:59:33 +0000 (23:59 +0000)]
Skip imported/w3c/web-platform-tests/FileAPI/reading-data-section/filereader_abort.html.
https://bugs.webkit.org/show_bug.cgi?id=187269

Unreviewed test gardening.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234565 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRemove WebGPUObject
commit-queue@webkit.org [Fri, 3 Aug 2018 23:25:56 +0000 (23:25 +0000)]
Remove WebGPUObject
https://bugs.webkit.org/show_bug.cgi?id=188114

Patch by Sam Weinig <sam@webkit.org> on 2018-08-03
Reviewed by Darin Adler.

This removes the unecessary WebGPUObject base class and updates the classes
that were inheriting from it to instead inherit directly from RefCounted. This
shrinks the size of all the classes that were inheriting from it, as they no
longer store a RefPtr<WebGPURenderingContext> member, which is unused. Most
of the classes also shrink because they can be devirtualized.

This also updates most of the create functions and constructors for the classes
that were inheriting from WebGPUObject to no longer require the WebGPURenderingContext
to be passed in.
- For WebGPUBuffer, WebGPUCommandQueue, WebGPUComputeCommandEncoder, WebGPUComputePipelineState,
WebGPUDepthStencilState, WebGPUDrawable, WebGPUFunction, WebGPULibrary, WebGPURenderCommandEncoder,
WebGPURenderPassDepthAttachmentDescriptor, WebGPURenderPipelineColorAttachmentDescriptor,
WebGPURenderPipelineState, and WebGPUTexture the WebGPURenderingContext and any other additional
parameters needed for initialization were replaced with the underlying type (e.g. WebGPUBuffer
now gets passed a GPUBuffer).
- WebGPUCommandBuffer is passed a GPUCommandQueue, since it needs to construct the underlying
GPUCommandBuffer itself so it can pass in the completion handler.
- For WebGPUDepthStencilDescriptor, WebGPURenderPassAttachmentDescriptor,
WebGPURenderPassColorAttachmentDescriptor, WebGPURenderPassDescriptor and WebGPURenderPipelineDescriptor
the create and constructor functions are empty.

* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
Remove WebGPUObject.h/cpp

* html/canvas/WebGPUBuffer.cpp:
(WebCore::WebGPUBuffer::WebGPUBuffer):
* html/canvas/WebGPUBuffer.h:
* html/canvas/WebGPUBuffer.idl:
Switch from inheriting from WebGPUObject to directly inheriting from
RefCounted<WebGPUBuffer>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create
and constructor to take a GPUBuffer directly, rather than the context.

* html/canvas/WebGPUCommandBuffer.cpp:
(WebCore::WebGPUCommandBuffer::create):
(WebCore::WebGPUCommandBuffer::WebGPUCommandBuffer):
Update create and constructor to no longer require a context, as it would
now be unused, as it no longer inherits from WebGPUObject nor needs to hold
onto the context for any reason.

(WebCore::WebGPUCommandBuffer::createRenderCommandEncoderWithDescriptor):
Update to call the new WebGPURenderCommandEncoder::create, which no longer
requires a context.

(WebCore::WebGPUCommandBuffer::createComputeCommandEncoder):
Update to call the new WebGPUComputeCommandEncoder::create, which no longer
requires a context.

* html/canvas/WebGPUCommandBuffer.h:
* html/canvas/WebGPUCommandBuffer.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPUCommandBuffer>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create and constructor
to no longer take a context, and just take a GPUCommandQueue. We can't directly
pass a GPUCommandBuffer as WebGPUCommandBuffer needs to pass in a completion handler
on construction of the underlying GPUCommandBuffer.

* html/canvas/WebGPUCommandQueue.cpp:
(WebCore::WebGPUCommandQueue::WebGPUCommandQueue):
(WebCore::WebGPUCommandQueue::createCommandBuffer):
Update to call the new WebGPUCommandBuffer::create, which no longer
requires a context.

* html/canvas/WebGPUCommandQueue.h:
* html/canvas/WebGPUCommandQueue.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPUCommandQueue>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create and
constructor to take a GPUCommandQueue directly.

* html/canvas/WebGPUComputeCommandEncoder.cpp:
(WebCore::WebGPUComputeCommandEncoder::create):
(WebCore::WebGPUComputeCommandEncoder::WebGPUComputeCommandEncoder):
* html/canvas/WebGPUComputeCommandEncoder.h:
* html/canvas/WebGPUComputeCommandEncoder.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPUComputeCommandEncoder>. Since this de-virtualizes it,
also add ImplementationLacksVTable to the IDL file. Also update create and
constructor to take a GPUComputeCommandEncoder directly

* html/canvas/WebGPUComputePipelineState.cpp:
(WebCore::WebGPUComputePipelineState::WebGPUComputePipelineState):
* html/canvas/WebGPUComputePipelineState.h:
* html/canvas/WebGPUComputePipelineState.idl:
Switch from inheriting from WebGPUObject to directly inheriting from
RefCounted<WebGPUComputePipelineState>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create and
constructor to take a GPUComputePipelineState directly.

* html/canvas/WebGPUDepthStencilDescriptor.cpp:
Remove unneeded include of include "WebGPURenderingContext.h".

* html/canvas/WebGPUDepthStencilDescriptor.h:
* html/canvas/WebGPUDepthStencilDescriptor.idl:
Switch from inheriting from WebGPUObject to directly inheriting from
RefCounted<WebGPUDepthStencilDescriptor>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file.

* html/canvas/WebGPUDepthStencilState.cpp:
(WebCore::WebGPUDepthStencilState::WebGPUDepthStencilState):
* html/canvas/WebGPUDepthStencilState.h:
* html/canvas/WebGPUDepthStencilState.idl:
Switch from inheriting from WebGPUObject to directly inheriting from
RefCounted<WebGPUDepthStencilState>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create and
constructor to take a GPUDepthStencilState directly.

* html/canvas/WebGPUDrawable.cpp:
(WebCore::WebGPUDrawable::WebGPUDrawable):
* html/canvas/WebGPUDrawable.h:
* html/canvas/WebGPUDrawable.idl:
Switch from inheriting from WebGPUObject to directly inheriting from
RefCounted<WebGPUDrawable>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create and
constructor to take a GPUDrawable directly.

* html/canvas/WebGPUFunction.cpp:
(WebCore::WebGPUFunction::create):
(WebCore::WebGPUFunction::WebGPUFunction):
* html/canvas/WebGPUFunction.h:
* html/canvas/WebGPUFunction.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPUFunction>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create and
constructor to take a GPUFunction directly.

* html/canvas/WebGPULibrary.cpp:
(WebCore::WebGPULibrary::WebGPULibrary):
(WebCore::WebGPULibrary::functionWithName const):
Update for new WebGPUFunction::create, which no longer requires a
context.

* html/canvas/WebGPULibrary.h:
* html/canvas/WebGPULibrary.idl:
Switch from inheriting from WebGPUObject to directly inheriting from
RefCounted<WebGPULibrary>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create and
constructor to take a GPULibrary directly, rather than the context.
Unlike the other create functions that take their underlying type, this
one also continues to need the sourceCode String to be passed, as that
is not available from the underlying GPULibrary and must be stored
seperately.

* html/canvas/WebGPUObject.cpp: Removed.
* html/canvas/WebGPUObject.h: Removed.
Deleted.

* html/canvas/WebGPURenderCommandEncoder.cpp:
(WebCore::WebGPURenderCommandEncoder::create):
(WebCore::WebGPURenderCommandEncoder::WebGPURenderCommandEncoder):
* html/canvas/WebGPURenderCommandEncoder.h:
* html/canvas/WebGPURenderCommandEncoder.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPURenderCommandEncoder>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create and
constructor to take a GPURenderCommandEncoder directly.

* html/canvas/WebGPURenderPassAttachmentDescriptor.cpp:
(WebCore::WebGPURenderPassAttachmentDescriptor::WebGPURenderPassAttachmentDescriptor):
* html/canvas/WebGPURenderPassAttachmentDescriptor.h:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPURenderPassAttachmentDescriptor>. Also update create and
constructor to no longer take a context, as it is now not used. This remains virtual
as it is inherited from by WebGPURenderPassColorAttachmentDescriptor and WebGPURenderPassDepthAttachmentDescriptor.

* html/canvas/WebGPURenderPassColorAttachmentDescriptor.cpp:
(WebCore::WebGPURenderPassColorAttachmentDescriptor::create):
(WebCore::WebGPURenderPassColorAttachmentDescriptor::WebGPURenderPassColorAttachmentDescriptor):
* html/canvas/WebGPURenderPassColorAttachmentDescriptor.h:
Update create and constructor to no longer take a context, as it is now not used. Declare final
as this is not inherited by anything.

* html/canvas/WebGPURenderPassDepthAttachmentDescriptor.cpp:
(WebCore::WebGPURenderPassDepthAttachmentDescriptor::create):
(WebCore::WebGPURenderPassDepthAttachmentDescriptor::WebGPURenderPassDepthAttachmentDescriptor):
* html/canvas/WebGPURenderPassDepthAttachmentDescriptor.h:
Update create and constructor to no longer take a context, as it is now not used. Declare final
as this is not inherited by anything.

* html/canvas/WebGPURenderPassDescriptor.cpp:
(WebCore::WebGPURenderPassDescriptor::depthAttachment):
Update for new WebGPURenderPassDepthAttachmentDescriptor::create function which no longer requires
a context.

Semi-interesting note: This function was previously calling "*context()" to pass a WebGPURenderingContext&
to WebGPURenderPassDepthAttachmentDescriptor::create. However, "context()" was always returning null, as
the constructor of WebGPURenderPassDescriptor was not initializing the WebGPUObject with a WebGPURenderingContext.
Now, one might think that this would cause a null-deref, but, because it was being assigned into a
WebGPURenderingContext&, no dereference happens right away (it essentially just passes a pointer) and since
the function it is being passed to does not use it, it never actually gets dereferenced.

(WebCore::WebGPURenderPassDescriptor::colorAttachments):
Update for new WebGPURenderPassColorAttachmentDescriptor::create function which no longer requires
a context.

* html/canvas/WebGPURenderPassDescriptor.h:
* html/canvas/WebGPURenderPassDescriptor.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPURenderPassDescriptor>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file.

* html/canvas/WebGPURenderPipelineColorAttachmentDescriptor.cpp:
(WebCore::WebGPURenderPipelineColorAttachmentDescriptor::create):
(WebCore::WebGPURenderPipelineColorAttachmentDescriptor::WebGPURenderPipelineColorAttachmentDescriptor):
* html/canvas/WebGPURenderPipelineColorAttachmentDescriptor.h:
* html/canvas/WebGPURenderPipelineColorAttachmentDescriptor.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPURenderPipelineColorAttachmentDescriptor>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create and constructor to
no longer take a context, as it is now not used.

* html/canvas/WebGPURenderPipelineDescriptor.cpp:
(WebCore::WebGPURenderPipelineDescriptor::colorAttachments):
Update for new WebGPURenderPipelineColorAttachmentDescriptor::create which no longer requires
a context.

* html/canvas/WebGPURenderPipelineDescriptor.h:
* html/canvas/WebGPURenderPipelineDescriptor.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPURenderPipelineDescriptor>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file.

* html/canvas/WebGPURenderPipelineState.cpp:
(WebCore::WebGPURenderPipelineState::WebGPURenderPipelineState):
* html/canvas/WebGPURenderPipelineState.h:
* html/canvas/WebGPURenderPipelineState.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPURenderPipelineState>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file. Also update create and
constructor to take a GPURenderPipelineState directly.

* html/canvas/WebGPURenderingContext.cpp:
(WebCore::WebGPURenderingContext::createLibrary):
(WebCore::WebGPURenderingContext::createRenderPipelineState):
(WebCore::WebGPURenderingContext::createDepthStencilState):
(WebCore::WebGPURenderingContext::createComputePipelineState):
(WebCore::WebGPURenderingContext::createCommandQueue):
(WebCore::WebGPURenderingContext::nextDrawable):
(WebCore::WebGPURenderingContext::createBuffer):
(WebCore::WebGPURenderingContext::createTexture):
Update for new create functions which require the caller to pass
the underlying type.

* html/canvas/WebGPUTexture.cpp:
(WebCore::WebGPUTexture::createFromDrawableTexture):
Renamed to create.

(WebCore::WebGPUTexture::create): Deleted
All callers now pass a GPUTexture directly.

* html/canvas/WebGPUTexture.h:
* html/canvas/WebGPUTexture.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPUTexture>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file.

* html/canvas/WebGPUTextureDescriptor.h:
* html/canvas/WebGPUTextureDescriptor.idl:
Switch from inheriting from GPUObject to directly inheriting from
RefCounted<WebGPUTextureDescriptor>. Since this de-virtualizes it, also add
ImplementationLacksVTable to the IDL file.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234564 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoFix spelling of "overridden"
commit-queue@webkit.org [Fri, 3 Aug 2018 23:13:57 +0000 (23:13 +0000)]
Fix spelling of "overridden"
https://bugs.webkit.org/show_bug.cgi?id=188315

Patch by Alex Christensen <achristensen@webkit.org> on 2018-08-03
Reviewed by Darin Adler.

Source/JavaScriptCore:

* API/JSExport.h:
* inspector/InjectedScriptSource.js:

Source/ThirdParty/ANGLE:

* src/libANGLE/renderer/ProgramImpl.h:

Source/ThirdParty/libwebrtc:

* Source/webrtc/p2p/client/basicportallocator.h:

Source/WebCore:

* accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
(-[WebAccessibilityObjectWrapper accessibilityLabel]):
(-[WebAccessibilityObjectWrapper accessibilityValue]):
* contentextensions/ContentExtensionsBackend.h:
* editing/EditingStyle.cpp:
(WebCore::EditingStyle::removeStyleFromRulesAndContext):
* inspector/CommandLineAPIModuleSource.js:
* platform/ScrollView.h:
* platform/ScrollableArea.h:
* platform/mediasession/MediaSessionInterruptionProvider.h:
* rendering/RenderFragmentedFlow.cpp:
(WebCore::RenderFragmentedFlow::adjustedPositionRelativeToOffsetParent const):
* rendering/RenderLayer.h:
* rendering/RenderListBox.h:
* rendering/svg/RenderSVGResourceClipper.cpp:
(WebCore::RenderSVGResourceClipper::drawContentIntoMaskImage):
* svg/SVGElement.cpp:
(WebCore::SVGElement::localCoordinateSpaceTransform const):
* svg/SVGLengthContext.cpp:
(WebCore::SVGLengthContext::SVGLengthContext):
(WebCore::SVGLengthContext::convertValueToUserUnits const):
(WebCore::SVGLengthContext::determineViewport const):
* svg/SVGLengthContext.h:

Source/WebInspectorUI:

* UserInterface/Views/ContentBrowserTabContentView.js:
(WI.ContentBrowserTabContentView.prototype.treeElementForRepresentedObject):
* UserInterface/Views/GeneralTreeElement.js:
(WI.GeneralTreeElement.prototype.ondetach):
* UserInterface/Views/NavigationBar.js:
(WI.NavigationBar.prototype.get sizesToFit):
* UserInterface/Views/NavigationSidebarPanel.js:
(WI.NavigationSidebarPanel.prototype.shouldFilterPopulate):
* UserInterface/Views/TextEditor.js:
(WI.TextEditor.prototype.canBeFormatted):
* UserInterface/Views/TreeElement.js:
(WI.TreeElement.prototype.onpopulate):

Source/WebKit:

* Platform/IPC/mac/ConnectionMac.mm:
(IPC::Connection::receiveSourceEventHandler):
* Shared/WebPreferencesStore.cpp:
(WebKit::WebPreferencesStore::encode const):
(WebKit::WebPreferencesStore::decode):
(WebKit::valueForKey):
(WebKit::setValueForKey):
(WebKit::WebPreferencesStore::setStringValueForKey):
(WebKit::WebPreferencesStore::getStringValueForKey const):
(WebKit::WebPreferencesStore::setBoolValueForKey):
(WebKit::WebPreferencesStore::getBoolValueForKey const):
(WebKit::WebPreferencesStore::setUInt32ValueForKey):
(WebKit::WebPreferencesStore::getUInt32ValueForKey const):
(WebKit::WebPreferencesStore::setDoubleValueForKey):
(WebKit::WebPreferencesStore::getDoubleValueForKey const):
(WebKit::WebPreferencesStore::setOverrideDefaultsStringValueForKey):
(WebKit::WebPreferencesStore::setOverrideDefaultsBoolValueForKey):
(WebKit::WebPreferencesStore::setOverrideDefaultsUInt32ValueForKey):
(WebKit::WebPreferencesStore::setOverrideDefaultsDoubleValueForKey):
* Shared/WebPreferencesStore.h:
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView dragInteraction:previewForLiftingItem:session:]):
(-[WKContentView dragInteraction:previewForCancellingItem:withDefault:]):
* UIProcess/ios/WKScrollView.mm:
(-[WKScrollView _systemContentInset]):

Source/WebKitLegacy/mac:

* Misc/WebKitVersionChecks.mm:
(setWebKitLinkTimeVersion):
(WebKitLinkTimeVersion):
* Plugins/WebBaseNetscapePluginView.mm:
(-[WebBaseNetscapePluginView handleMouseMoved:]):
* WebCoreSupport/PopupMenuMac.mm:
(PopupMenuMac::populate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234563 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, rolling out r234517.
david_fenton@apple.com [Fri, 3 Aug 2018 23:01:35 +0000 (23:01 +0000)]
Unreviewed, rolling out r234517.

Caused API test failures on iOS

Reverted changeset:

"[Cocoa] setCookie API fails to set session cookies for
defaultDataStore if processPool created but not used"
https://bugs.webkit.org/show_bug.cgi?id=188209
https://trac.webkit.org/changeset/234517

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234561 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoAdd configuration for automatic process pre-warming
commit-queue@webkit.org [Fri, 3 Aug 2018 22:50:15 +0000 (22:50 +0000)]
Add configuration for automatic process pre-warming
https://bugs.webkit.org/show_bug.cgi?id=187108

Patch by Ben Richards <benton_richards@apple.com> on 2018-08-03
Reviewed by Ryosuke Niwa.

Source/WebKit:

Added configurations to allow setting the maximum number of processes that should be automatically prewarmed.

* UIProcess/API/APIProcessPoolConfiguration.cpp:
(API::ProcessPoolConfiguration::copy):
* UIProcess/API/APIProcessPoolConfiguration.h:
* UIProcess/API/C/WKContext.cpp:
(WKContextSetMaximumNumberOfPrewarmedProcesses):
* UIProcess/API/C/WKContextPrivate.h:
* UIProcess/API/Cocoa/WKProcessPool.mm:
(-[WKProcessPool _setMaximumNumberOfPrewarmedProcesses:]):
* UIProcess/API/Cocoa/WKProcessPoolPrivate.h:
* UIProcess/API/Cocoa/_WKProcessPoolConfiguration.h:
* UIProcess/API/Cocoa/_WKProcessPoolConfiguration.mm:
(-[_WKProcessPoolConfiguration setMaximumPrewarmedProcessCount:]):
(-[_WKProcessPoolConfiguration maximumPrewarmedProcessCount]):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::didFinishLoadForFrame): Moved call to notifyProcessPoolToPrewarm from didFirstVisuallyNonEmptyLayoutForFrame to here.
This is to try to ensure that frame loading and prewarming don't happen at the same time as this would be heavy for some devices.
(WebKit::WebPageProxy::didFirstVisuallyNonEmptyLayoutForFrame):
* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::setMaximumNumberOfProcesses): Condition changed so that calling setMaximumNumberOfProcesses after warmInitialProcess
doesn't result in a crash.
(WebKit::WebProcessPool::setMaximumNumberOfPrewarmedProcesses):
(WebKit::WebProcessPool::warmInitialProcess):
(WebKit::WebProcessPool::didReachGoodTimeToPrewarm):
* UIProcess/WebProcessPool.h:

Tools:

Added new test case for setting maximum prewarmed process count and updated test case for process swap on navigation to set the maximum prewarmed process count to be 1 where relevant.

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKitCocoa/InitialWarmedProcessUsed.mm:
(TEST):
* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
* TestWebKitAPI/Tests/WebKitCocoa/SetMaximumPrewarmedProcessCount.mm:
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234560 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[ews-build] Add build step to Check Patch Relevance
aakash_jain@apple.com [Fri, 3 Aug 2018 21:35:49 +0000 (21:35 +0000)]
[ews-build] Add build step to Check Patch Relevance
https://bugs.webkit.org/show_bug.cgi?id=188295

Reviewed by Lucas Forschler.

* BuildSlaveSupport/ews-build/steps.py:
(CheckPatchRelevance): Added step to check patch relevance.
(CheckPatchRelevance._patch_is_relevant): Checks if the patch is relevant.
(CheckPatchRelevance._get_patch): Retrieves the patch from buildbot.
(CheckPatchRelevance._addToLog): Add the log message.
(CheckPatchRelevance.start):
* BuildSlaveSupport/ews-build/factories.py: Added CheckPatchRelevance step appropriately.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234559 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LayoutTests][websocket] Return correct value even when cookie is not set.
Basuke.Suzuki@sony.com [Fri, 3 Aug 2018 21:10:44 +0000 (21:10 +0000)]
[LayoutTests][websocket] Return correct value even when cookie is not set.
https://bugs.webkit.org/show_bug.cgi?id=188317

Reviewed by Darin Adler.

> ws://127.0.0.1:8880/websocket/tests/hybi/echo-cookie_wsh.py
This web socket script throw exception when cookie is not set. That is ended up with
immediate disconnection and causes timeout for failure tests which makes debugging hard.

* http/tests/websocket/tests/hybi/echo-cookie_wsh.py:
(web_socket_transfer_data):
* platform/wincairo/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234558 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Curl] Test gardening
Basuke.Suzuki@sony.com [Fri, 3 Aug 2018 19:19:07 +0000 (19:19 +0000)]
[Curl] Test gardening
https://bugs.webkit.org/show_bug.cgi?id=188316

Unreviewed test gardening.

Skip some flaky tests.
Add comments why we skip for some platform dependent reason.

* platform/wincairo/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234557 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoFix invalid optional access in WebDriver Session::createTopLevelBrowsingContext.
Ms2ger@igalia.com [Fri, 3 Aug 2018 18:36:31 +0000 (18:36 +0000)]
Fix invalid optional access in WebDriver Session::createTopLevelBrowsingContext.
https://bugs.webkit.org/show_bug.cgi?id=188304

Reviewed by Brian Burg.

* Session.cpp:
(WebDriver::Session::createTopLevelBrowsingContext): remove incorrect value() call.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234555 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][BFC] Special case the document renderer when computing the height and margin
zalan@apple.com [Fri, 3 Aug 2018 18:00:23 +0000 (18:00 +0000)]
[LFC][BFC] Special case the document renderer when computing the height and margin
https://bugs.webkit.org/show_bug.cgi?id=188313

Reviewed by Antti Koivisto.

The document renderer is certainly special and its geometry is likely to be computed as part of "10.6.6 Complicated cases" (spec is unclear) but probably not by
making the it a formatting root. Let's just special case it at contentHeightForFormattingContextRoot(), since having the document renderer as a context root has unwanted side effect.

* layout/FormattingContextGeometry.cpp:
(WebCore::Layout::contentHeightForFormattingContextRoot):
* layout/layouttree/LayoutBox.cpp:
(WebCore::Layout::Box::establishesBlockFormattingContext const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234554 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][BFC] Layout out-of-flow descendants only on the formatting context root.
zalan@apple.com [Fri, 3 Aug 2018 17:57:35 +0000 (17:57 +0000)]
[LFC][BFC] Layout out-of-flow descendants only on the formatting context root.
https://bugs.webkit.org/show_bug.cgi?id=188312

Reviewed by Antti Koivisto.

Collect out-of-flow descendants on the formatting root container instead of the containing block and run layout on them on the root level. It's more efficient this way.

* layout/blockformatting/BlockFormattingContext.cpp:
(WebCore::Layout::BlockFormattingContext::layout const):
* layout/layouttree/LayoutTreeBuilder.cpp:
(WebCore::Layout::TreeBuilder::createSubTree):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234553 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoCheck with SafeBrowsing during navigation in WKWebView
achristensen@apple.com [Fri, 3 Aug 2018 17:56:29 +0000 (17:56 +0000)]
Check with SafeBrowsing during navigation in WKWebView
https://bugs.webkit.org/show_bug.cgi?id=188133

Reviewed by Chris Dumez.

This turns WebFramePolicyListenerProxy into an object that now listens for the results
of two processes happening in parallel: the API::NavigationClient's decidePolicyForNavigation{Action, Response}
(which it was already waiting for) and, on platforms that support it, the SafariSafeBrowsing framework's check.
The first result is stored as it waits for the second result unless the first result is the API::NavigationClient
saying to cancel or convert the navigation to a download, in which cases we don't care what the safe browsing
framework results are because we won't show the URL in the browser.

Nothing is done with the safe browsing results yet.

* UIProcess/Cocoa/SafeBrowsingResultCocoa.mm: Added.
(WebKit::SafeBrowsingResult::SafeBrowsingResult):
* UIProcess/Cocoa/WebPageProxyCocoa.mm:
(WebKit::WebPageProxy::beginSafeBrowsingCheck):
* UIProcess/SafeBrowsingResult.h: Added.
(WebKit::SafeBrowsingResult::provider const):
(WebKit::SafeBrowsingResult::isPhishing const):
(WebKit::SafeBrowsingResult::isMalware const):
(WebKit::SafeBrowsingResult::isUnwantedSoftware const):
(WebKit::SafeBrowsingResult::isKnownToBeUnsafe const):
* UIProcess/WebFramePolicyListenerProxy.cpp:
(WebKit::WebFramePolicyListenerProxy::WebFramePolicyListenerProxy):
(WebKit::WebFramePolicyListenerProxy::didReceiveSafeBrowsingResults):
(WebKit::WebFramePolicyListenerProxy::use):
(WebKit::WebFramePolicyListenerProxy::download):
(WebKit::WebFramePolicyListenerProxy::ignore):
* UIProcess/WebFramePolicyListenerProxy.h:
(WebKit::WebFramePolicyListenerProxy::create):
* UIProcess/WebFrameProxy.cpp:
(WebKit::WebFrameProxy::setUpPolicyListenerProxy):
* UIProcess/WebFrameProxy.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::decidePolicyForNavigationAction):
(WebKit::WebPageProxy::decidePolicyForNewWindowAction):
(WebKit::WebPageProxy::decidePolicyForResponse):
* UIProcess/WebPageProxy.h:
* WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234552 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LayoutTests][PHP] Fix old-style script which uses function name as a keyword.
Basuke.Suzuki@sony.com [Fri, 3 Aug 2018 17:51:21 +0000 (17:51 +0000)]
[LayoutTests][PHP] Fix old-style script which uses function name as a keyword.
https://bugs.webkit.org/show_bug.cgi?id=188289

Reviewed by Alex Christensen.

Function name should be passed as a string literal. In PHP, named function is
not the first class citizen.

* http/tests/cookies/resources/echo-cookies.php:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234551 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoContentFilterUnblockHandler's constructors access moved-from variables
aestes@apple.com [Fri, 3 Aug 2018 17:51:04 +0000 (17:51 +0000)]
ContentFilterUnblockHandler's constructors access moved-from variables
https://bugs.webkit.org/show_bug.cgi?id=188308
<rdar://problem/42882758>

Reviewed by Alex Christensen.

* platform/cocoa/ContentFilterUnblockHandlerCocoa.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234550 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LayoutTests] Suppress PHP Error message by removing an execution time limit.
Basuke.Suzuki@sony.com [Fri, 3 Aug 2018 17:48:47 +0000 (17:48 +0000)]
[LayoutTests] Suppress PHP Error message by removing an execution time limit.
https://bugs.webkit.org/show_bug.cgi?id=188288

Reviewed by Alex Christensen.

Remove execution timeout.

* http/tests/multipart/resources/multipart.php:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234549 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed gardening; r234534 changed the policy for what audio session is set when a
jer.noble@apple.com [Fri, 3 Aug 2018 17:06:28 +0000 (17:06 +0000)]
Unreviewed gardening; r234534 changed the policy for what audio session is set when a
media element is paused. Updating the test to reflect the new policy.

* TestWebKitAPI/Tests/WebKitLegacy/ios/AudioSessionCategoryIOS.mm:
(TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234547 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][Floating] Now that the document renderer belongs to "complicated cases", adjust...
zalan@apple.com [Fri, 3 Aug 2018 15:38:10 +0000 (15:38 +0000)]
[LFC][Floating] Now that the document renderer belongs to "complicated cases", adjust viewport stretching.
https://bugs.webkit.org/show_bug.cgi?id=188306

Reviewed by Antti Koivisto.

Source/WebCore:

And add passing cases for floats.

* layout/blockformatting/BlockFormattingContextGeometry.cpp:
(WebCore::Layout::initialContainingBlock):
(WebCore::Layout::isStretchedToInitialContainingBlock):
(WebCore::Layout::stretchHeightToInitialContainingBlock):
(WebCore::Layout::stretchWidthToInitialContainingBlock):
(WebCore::Layout::BlockFormattingContext::Geometry::inFlowNonReplacedHeightAndMargin):
(WebCore::Layout::BlockFormattingContext::Geometry::inFlowNonReplacedWidthAndMargin):
(WebCore::Layout::BlockFormattingContext::Geometry::inFlowHeightAndMargin):
(WebCore::Layout::isStretchedToViewport): Deleted.

Tools:

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234546 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, rolling out r234513.
ryanhaddad@apple.com [Fri, 3 Aug 2018 15:26:13 +0000 (15:26 +0000)]
Unreviewed, rolling out r234513.

14 API tests fail the assertions added in this change.

Reverted changeset:

"Check with SafeBrowsing during navigation in WKWebView"
https://bugs.webkit.org/show_bug.cgi?id=188133
https://trac.webkit.org/changeset/234513

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234545 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC] Do not check margin box while validating geometry.
zalan@apple.com [Fri, 3 Aug 2018 14:21:21 +0000 (14:21 +0000)]
[LFC] Do not check margin box while validating geometry.
https://bugs.webkit.org/show_bug.cgi?id=188297

Reviewed by Antti Koivisto.

The current RenderBox::marginBox implementation is not spec compliant. Ignore it for now.

* layout/Verification.cpp:
(WebCore::Layout::outputMismatchingBlockBoxInformationIfNeeded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234544 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WPE] WebDriver: add support for action commands
carlosgc@webkit.org [Fri, 3 Aug 2018 10:43:14 +0000 (10:43 +0000)]
[WPE] WebDriver: add support for action commands
https://bugs.webkit.org/show_bug.cgi?id=188301

Reviewed by Žan Doberšek.

WPE doesn't support action commands because the platform specific code for handling events is not implemented.

* SourcesWPE.txt: Add new file to compilation.
* UIProcess/API/glib/WebKitUIClient.cpp: Use the drawing area size as window size in WPE.
* UIProcess/API/wpe/PageClientImpl.cpp:
(WebKit::PageClientImpl::viewBackend): Return the WPE backend of the view.
* UIProcess/API/wpe/PageClientImpl.h:
* UIProcess/Automation/WebAutomationSession.cpp:
(WebKit::WebAutomationSession::performMouseInteraction):
(WebKit::WebAutomationSession::performKeyboardInteractions):
(WebKit::WebAutomationSession::performInteractionSequence):
(WebKit::WebAutomationSession::cancelInteractionSequence):
* UIProcess/Automation/wpe/WebAutomationSessionWPE.cpp: Added.
(WebKit::modifiersToEventState):
(WebKit::mouseButtonToWPEButton):
(WebKit::stateModifierForWPEButton):
(WebKit::doMouseEvent):
(WebKit::doMotionEvent):
(WebKit::WebAutomationSession::platformSimulateMouseInteraction):
(WebKit::doKeyStrokeEvent):
(WebKit::keyCodeForVirtualKey):
(WebKit::modifiersForKeyCode):
(WebKit::WebAutomationSession::platformSimulateKeyboardInteraction):
(WebKit::WebAutomationSession::platformSimulateKeySequence):
* UIProcess/WebPageProxy.h:
* UIProcess/wpe/WebPageProxyWPE.cpp:
(WebKit::WebPageProxy::viewBackend): Return the WPE backend.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234543 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WPE] Use the new key mapper API from WPEBackend
carlosgc@webkit.org [Fri, 3 Aug 2018 10:07:55 +0000 (10:07 +0000)]
[WPE] Use the new key mapper API from WPEBackend
https://bugs.webkit.org/show_bug.cgi?id=188216

Reviewed by Žan Doberšek.

* WebKitTestRunner/wpe/EventSenderProxyWPE.cpp:
(WTR::EventSenderProxy::keyDown): Use wpe_input_xkb_context_get_entries_for_key_code() to get the hardware key
code.
* wpe/backends/WindowViewBackend.cpp:
(WPEToolingBackends::WindowViewBackend::WindowViewBackend):
(WPEToolingBackends::WindowViewBackend::handleKeyEvent):
(WPEToolingBackends::WindowViewBackend::modifiers const):
* wpe/backends/WindowViewBackend.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234542 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WPE] Implement MouseEvent.buttons
carlosgc@webkit.org [Fri, 3 Aug 2018 09:38:18 +0000 (09:38 +0000)]
[WPE] Implement MouseEvent.buttons
https://bugs.webkit.org/show_bug.cgi?id=187998

Reviewed by Žan Doberšek.

Source/WebKit:

Pass buttons currently pressed to WebMouseEvent.

* Shared/wpe/WebEventFactory.cpp:
(WebKit::pressedMouseButtons): Helper to get the pressed buttons mask for the WPE modifiers.
(WebKit::WebEventFactory::createWebMouseEvent): Use pressedMouseButtons().
* UIProcess/API/wpe/PageClientImpl.cpp:
(WebKit::PageClientImpl::doneWithTouchEvent): Update the event modifiers.

Tools:

Pass modifiers to mouse events.

* WebKitTestRunner/EventSenderProxy.h:
* WebKitTestRunner/wpe/EventSenderProxyWPE.cpp:
(WTR::senderButtonToWPEButton):
(WTR::modifierForButton):
(WTR::EventSenderProxy::mouseDown):
(WTR::EventSenderProxy::mouseUp):
(WTR::EventSenderProxy::mouseMoveTo):
(WTR::EventSenderProxy::mouseScrollBy):
(WTR::wkEventModifiersToWPE):
(WTR::wpeKeySymForKeyRef):
(WTR::EventSenderProxy::keyDown):
(WTR::EventSenderProxy::prepareAndDispatchTouchEvent):
* wpe/backends/WindowViewBackend.cpp:
(WPEToolingBackends::WindowViewBackend::handleKeyEvent):
(WPEToolingBackends::WindowViewBackend::modifiers const):
* wpe/backends/WindowViewBackend.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234541 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WPE] Use WPE key symbols and new API instead of xkbcommon and the key mapper
carlosgc@webkit.org [Fri, 3 Aug 2018 09:16:18 +0000 (09:16 +0000)]
[WPE] Use WPE key symbols and new API instead of xkbcommon and the key mapper
https://bugs.webkit.org/show_bug.cgi?id=188093

Reviewed by Žan Doberšek.

Source/WebCore:

Add helper functions to get key code, hardware key code, key identifier and windows key code from a WPE key symbol.

* platform/PlatformKeyboardEvent.h:
* platform/wpe/PlatformKeyboardEventWPE.cpp:
(WebCore::PlatformKeyboardEvent::keyValueForWPEKeyCode):
(WebCore::PlatformKeyboardEvent::keyCodeForHardwareKeyCode):
(WebCore::PlatformKeyboardEvent::keyIdentifierForWPEKeyCode):
(WebCore::PlatformKeyboardEvent::windowsKeyCodeForWPEKeyCode):
(WebCore::PlatformKeyboardEvent::singleCharacterString):

Source/WebKit:

* Shared/WebEvent.h: Add WebKeyboardEvent constructor for WPE.
* Shared/WebKeyboardEvent.cpp:
(WebKit::WebKeyboardEvent::WebKeyboardEvent): WebKeyboardEvent constructor for WPE.
* Shared/wpe/WebEventFactory.cpp:
(WebKit::isWPEKeyCodeFromKeyPad): Helper tpo check if a key symbols is from key pad.
(WebKit::WebEventFactory::createWebKeyboardEvent): Use PlatformKeyboardEvent API to provide key code, hardware
key code, key identifier and windows key code to WebKeyboardEvent.
* UIProcess/API/wpe/WPEView.cpp:
(WKWPE::m_backend): Use WPE key symbols.

Tools:

* MiniBrowser/wpe/main.cpp: Use WPE key symbols.
* WebKitTestRunner/wpe/EventSenderProxyWPE.cpp:
(WTR::wpeKeySymForKeyRef): Use WPE key symbols and wpe_unicode_to_key_code().
(WTR::EventSenderProxy::keyDown): Do not set unicode char to the event since it was removed.
* flatpak/org.webkit.WPE.yaml: Bump WPEBackend and WPEBackend-fdo versions.
* wpe/backends/WindowViewBackend.cpp:
(WPEToolingBackends::WindowViewBackend::handleKeyEvent): Do not set unicode char to the event since it was removed.
* wpe/jhbuild.modules: Bump WPEBackend and WPEBackend-fdo versions.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234540 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRelease assert when throwing exceptions in custom element reactions
rniwa@webkit.org [Fri, 3 Aug 2018 08:16:48 +0000 (08:16 +0000)]
Release assert when throwing exceptions in custom element reactions
https://bugs.webkit.org/show_bug.cgi?id=187805
<rdar://problem/42432714>

Reviewed by Saam Barati.

LayoutTests/imported/w3c:

Generated the expected result.

* web-platform-tests/custom-elements/reactions/with-exceptions-expected.txt: Added.

Source/WebCore:

The release assertion was hit because we were not catching & re-throwing the exception thrown by DOM API
before trying to execute custom elements reactions in ~CustomElementReactionStack as specified here:
https://html.spec.whatwg.org/multipage/custom-elements.html#cereactions
Fixed the bug by capturing the exception and re-throwing the exception as specified.

Tests: imported/w3c/web-platform-tests/custom-elements/reactions/with-exceptions.html

* bindings/js/JSMainThreadExecState.h:
(WebCore::JSMainThreadNullState::JSMainThreadNullState): Use the previous JS state.
* bindings/scripts/CodeGeneratorJS.pm:
(GeneratePut): Pass in the exec state to CustomElementReactionStack.
(GeneratePutByIndex): Ditto.
(GenerateDefineOwnProperty): Ditto.
(GenerateDeletePropertyCommon): Ditto.
(GenerateAttributeSetterBodyDefinition): Ditto.
(GenerateOperationBodyDefinition): Ditto.
* bindings/scripts/test/JS/JSTestCEReactions.cpp:
(WebCore::setJSTestCEReactionsAttributeWithCEReactionsSetter):
(WebCore::setJSTestCEReactionsReflectAttributeWithCEReactionsSetter):
(WebCore::jsTestCEReactionsPrototypeFunctionMethodWithCEReactionsBody):
* bindings/scripts/test/JS/JSTestCEReactionsStringifier.cpp:
(WebCore::setJSTestCEReactionsStringifierValueSetter):
* dom/CustomElementReactionQueue.cpp:
(WebCore::CustomElementReactionQueue::ElementQueue::processQueue): Added. If there is a script running
in the stack (i.e. ExecState is not null), catch any exception before executing custom element reactions,
then re-throw the exception afterwards. ExecState is null when DOM API is invoked via Objective-C bindings
or when custom element reactions are executed in the backup queue (e.g. for editing operations).
(WebCore::CustomElementReactionStack::processQueue):
(WebCore::CustomElementReactionQueue::processBackupQueue):
* dom/CustomElementReactionQueue.h:
(WebCore::CustomElementReactionStack::CustomElementReactionStack):
(WebCore::CustomElementReactionStack::~CustomElementReactionStack):

LayoutTests:

Unskipped the previously crashing test.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234539 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Curl] Test gardening
Basuke.Suzuki@sony.com [Fri, 3 Aug 2018 05:24:07 +0000 (05:24 +0000)]
[Curl] Test gardening
https://bugs.webkit.org/show_bug.cgi?id=188293

Unreviewed test gardening.

Unskip some http tests.

* platform/wincairo/TestExpectations:
  - contentextensions
  - events
  - favicon.ico
  - fullscreen
  - globalhistory
  - gzip-content-encoding
  - history
  - local
  - messaging
  - mime
  - multipart
  - performance
  - permissionclient
  - referrer-policy
  - resourceLoadStatistics
* platform/wincairo/http/tests/local/file-url-sent-as-referer-expected.txt: Added.
* platform/wincairo/http/tests/multipart/invalid-image-data-expected.txt: Added.
* platform/wincairo/http/tests/multipart/invalid-image-data-standalone-expected.txt: Added.
* platform/wincairo/http/tests/uri/css-href-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234538 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][BFC] Apply the "10.6.6 Complicated cases" when computing height and margin...
zalan@apple.com [Fri, 3 Aug 2018 04:32:05 +0000 (04:32 +0000)]
[LFC][BFC] Apply the "10.6.6 Complicated cases" when computing height and margin for the document renderer
https://bugs.webkit.org/show_bug.cgi?id=188296

Reviewed by Simon Fraser.

The spec is not clear about what to do with the document renderer. It behaves as if it was a formatting context root when
it comes to computing the content height. Let's apply "10.6.6 Complicated cases" for now.

* layout/FormattingContext.h:
* layout/FormattingContextGeometry.cpp:
(WebCore::Layout::FormattingContext::Geometry::complicatedCases):
(WebCore::Layout::FormattingContext::Geometry::floatingHeightAndMargin):
(WebCore::Layout::FormattingContext::Geometry::floatingNonReplacedHeightAndMargin): Deleted.
* layout/blockformatting/BlockFormattingContextGeometry.cpp:
(WebCore::Layout::BlockFormattingContext::Geometry::inFlowNonReplacedHeightAndMargin):
(WebCore::Layout::BlockFormattingContext::Geometry::inFlowHeightAndMargin):
* layout/layouttree/LayoutBox.cpp:
(WebCore::Layout::Box::establishesBlockFormattingContext const):
* layout/layouttree/LayoutBox.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234537 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][Floating] Add FloatingState::bottom() to enable content height computation...
zalan@apple.com [Fri, 3 Aug 2018 02:32:24 +0000 (02:32 +0000)]
[LFC][Floating] Add FloatingState::bottom() to enable content height computation for formatting roots.
https://bugs.webkit.org/show_bug.cgi?id=188294

Reviewed by Simon Fraser.

"10.6.7 'Auto' heights for block formatting context roots
...
In addition, if the element has any floating descendants whose bottom margin edge is below the element's bottom content edge,
then the height is increased to include those edges. Only floats that participate in this block formatting context are taken into
account, e.g., floats inside absolutely positioned descendants or other floats are not..."

* layout/FloatingState.cpp:
(WebCore::Layout::FloatingState::bottom const):
* layout/FloatingState.h:
* layout/FormattingContextGeometry.cpp:
(WebCore::Layout::contentHeightForFormattingContextRoot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234536 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoControl center controls disappear when pausing, locking device.
jer.noble@apple.com [Fri, 3 Aug 2018 01:54:15 +0000 (01:54 +0000)]
Control center controls disappear when pausing, locking device.
https://bugs.webkit.org/show_bug.cgi?id=188282
<rdar://problem/42497809>

Reviewed by Eric Carlson.

Source/WebCore:

Test: platform/mac/media/audio-session-category-video-paused.html

MediaRemote will set any app whose audio session category changes from MediaPlayback to
Ambient as not eligable for being the "now playing" app. Previously, due to the ordering of
events, we never moved from MediaPlayback to Ambient when pausing <video>, even though that
was the intention. Now that that bug is fixed, it exposed this new issue with MediaRemote.

To solve the new issue, make our audio session category policy more explicit: once we are in
MediaPlayback category, we will remain so as long as the media element in queston stays
loaded, has played, and is never interrupted by the system.

Make MediaSessionManagerCocoa a proper subclass of PlatformMediaSessionManager, and make
MediaSessionManageriOS and -Mac subclasses of the -Cocoa class. Add a new
m_hasPlayedSinceLastInterruption ivar to PlatformMediaSession, set when the state() changes
to Playing, and cleared during an interruption. Check this flag when deciding what audio
sessino category to set in MediaSessionManagerCocoa.

* WebCore.xcodeproj/project.pbxproj:
* platform/audio/PlatformMediaSession.cpp:
(WebCore::PlatformMediaSession::setState):
* platform/audio/PlatformMediaSession.h:
(WebCore::PlatformMediaSession::hasPlayedSinceLastInterruption const):
(WebCore::PlatformMediaSession::clearHasPlayedSinceLastInterruption):
* platform/audio/PlatformMediaSessionManager.cpp:
(WebCore::PlatformMediaSessionManager::updateSessionState): Deleted.
* platform/audio/PlatformMediaSessionManager.h:
(WebCore::PlatformMediaSessionManager::audioHardwareListener):
(WebCore::PlatformMediaSessionManager::updateSessionState):
* platform/audio/cocoa/MediaSessionManagerCocoa.cpp:
(MediaSessionManagerCocoa::updateSessionState):
(MediaSessionManagerCocoa::beginInterruption):
(PlatformMediaSessionManager::updateSessionState): Deleted.
* platform/audio/cocoa/MediaSessionManagerCocoa.h: Added.
* platform/audio/ios/MediaSessionManagerIOS.h:
* platform/audio/ios/MediaSessionManagerIOS.mm:
(WebCore::MediaSessionManageriOS::MediaSessionManageriOS):
* platform/audio/mac/MediaSessionManagerMac.h:
* platform/audio/mac/MediaSessionManagerMac.mm:
(WebCore::MediaSessionManagerMac::MediaSessionManagerMac):

LayoutTests:

* platform/mac/media/audio-session-category-video-paused-expected.txt: Added.
* platform/mac/media/audio-session-category-video-paused.html: Added.
* platform/ios/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234534 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoPDFPlugin: Context menus in RTL are left-aligned
timothy_horton@apple.com [Fri, 3 Aug 2018 01:11:46 +0000 (01:11 +0000)]
PDFPlugin: Context menus in RTL are left-aligned
https://bugs.webkit.org/show_bug.cgi?id=188292
<rdar://problem/32293787>

Reviewed by Simon Fraser.

* WebProcess/Plugins/PDF/PDFLayerControllerSPI.h:
Add some SPI.

* WebProcess/Plugins/PDF/PDFPlugin.mm:
(WebKit::PDFPlugin::handleContextMenuEvent):
Translate UI layout direction back into the platform enum, and pass it to PDFKit.

* WebProcess/WebPage/WebPage.h:
(WebKit::WebPage::userInterfaceLayoutDirection const):
Add a getter for UI layout direction.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234530 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRegression(r234486): assertion hit in ~CallbackAggregator()
cdumez@apple.com [Fri, 3 Aug 2018 00:24:13 +0000 (00:24 +0000)]
Regression(r234486): assertion hit in ~CallbackAggregator()
https://bugs.webkit.org/show_bug.cgi?id=188283
<rdar://problem/42851342>

Reviewed by Alex Christensen.

[NSHTTPCookieStorage _saveCookies:] SPI may call its completion block on the background queue
so we need to make sure we dispatch back to the main thread before calling our completion
handler.

* NetworkProcess/cocoa/NetworkProcessCocoa.mm:
(WebKit::saveCookies):
(WebKit::NetworkProcess::platformSyncAllCookies):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234529 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoSource/JavaScriptCore:
sbarati@apple.com [Fri, 3 Aug 2018 00:14:11 +0000 (00:14 +0000)]
Source/JavaScriptCore:
Reading instructionPointer from PlatformRegisters may fail when using pointer profiling
https://bugs.webkit.org/show_bug.cgi?id=188271
<rdar://problem/42850884>

Reviewed by Michael Saboff.

This patch defends against the instructionPointer containing garbage bits.
See radar for details.

* runtime/MachineContext.h:
(JSC::MachineContext::instructionPointer):
* runtime/SamplingProfiler.cpp:
(JSC::SamplingProfiler::takeSample):
* runtime/VMTraps.cpp:
(JSC::SignalContext::SignalContext):
(JSC::SignalContext::tryCreate):
* tools/CodeProfiling.cpp:
(JSC::profilingTimer):
* tools/SigillCrashAnalyzer.cpp:
(JSC::SignalContext::SignalContext):
(JSC::SignalContext::tryCreate):
(JSC::SignalContext::dump):
(JSC::installCrashHandler):
* wasm/WasmFaultSignalHandler.cpp:
(JSC::Wasm::trapHandler):

Source/WTF:
Reading instructionPointer from PlatformRegisters may fail when using pointer tagging
https://bugs.webkit.org/show_bug.cgi?id=188271
<rdar://problem/42850884>

Reviewed by Michael Saboff.

* wtf/PtrTag.h:
(WTF::isTaggedWith):
(WTF::usesPointerTagging):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234528 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoAX: [iOS] add support to return the attributed string under the element
n_wang@apple.com [Thu, 2 Aug 2018 23:54:45 +0000 (23:54 +0000)]
AX: [iOS] add support to return the attributed string under the element
https://bugs.webkit.org/show_bug.cgi?id=188276
<rdar://problem/42872357>

Reviewed by Chris Fleizach.

Source/WebCore:

Provided a way on iOS to return the attributed string under the element for
better performance.

Test: accessibility/ios-simulator/attributed-string-for-element.html

* accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:
(-[WebAccessibilityObjectWrapper _stringFromStartMarker:toEndMarker:attributed:]):
(-[WebAccessibilityObjectWrapper _stringForRange:attributed:]):
(-[WebAccessibilityObjectWrapper attributedStringForElement]):

Tools:

* DumpRenderTree/AccessibilityUIElement.cpp:
(attributedStringForElementCallback):
(AccessibilityUIElement::getJSClass):
* DumpRenderTree/AccessibilityUIElement.h:
* DumpRenderTree/ios/AccessibilityUIElementIOS.mm:
(AccessibilityUIElement::attributedStringForElement):
* WebKitTestRunner/InjectedBundle/AccessibilityUIElement.cpp:
(WTR::AccessibilityUIElement::attributedStringForElement):
* WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h:
* WebKitTestRunner/InjectedBundle/Bindings/AccessibilityUIElement.idl:
* WebKitTestRunner/InjectedBundle/ios/AccessibilityUIElementIOS.mm:
(WTR::AccessibilityUIElement::attributedStringForElement):

LayoutTests:

* accessibility/ios-simulator/attributed-string-for-element-expected.txt: Added.
* accessibility/ios-simulator/attributed-string-for-element.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234526 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][Floating] Remove redundant LayoutContext member variable.
zalan@apple.com [Thu, 2 Aug 2018 23:46:10 +0000 (23:46 +0000)]
[LFC][Floating] Remove redundant LayoutContext member variable.
https://bugs.webkit.org/show_bug.cgi?id=188286

Reviewed by Simon Fraser.

* layout/FloatingContext.cpp:
(WebCore::Layout::begin):
(WebCore::Layout::end):
(WebCore::Layout::FloatingContext::floatingPosition const):
(WebCore::Layout::FloatingPair::FloatingPair):
(WebCore::Layout::Iterator::Iterator):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234524 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoREGRESSION (r234488): Layout test fast/events/ios/keyboard-scrolling-repeat.html...
timothy_horton@apple.com [Thu, 2 Aug 2018 23:35:08 +0000 (23:35 +0000)]
REGRESSION (r234488): Layout test fast/events/ios/keyboard-scrolling-repeat.html is timing out
<rdar://problem/42878480>

* fast/events/ios/keyboard-scrolling-repeat-expected.txt: Removed.
* fast/events/ios/keyboard-scrolling-repeat.html: Removed.
Remove a test that can't work anymore.
If we need to re-write it it will probably need to be an API test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234522 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoSkip API test DataInteractionTests.CustomActionSheetPopover on current SDKs.
ryanhaddad@apple.com [Thu, 2 Aug 2018 23:09:24 +0000 (23:09 +0000)]
Skip API test DataInteractionTests.CustomActionSheetPopover on current SDKs.
https://bugs.webkit.org/show_bug.cgi?id=175204

Unreviewed test gardening.

* TestWebKitAPI/Tests/ios/DataInteractionTests.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234521 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoMove queue processing logic from CustomElementReactionStack to CustomElementReactionQueue
rniwa@webkit.org [Thu, 2 Aug 2018 22:37:09 +0000 (22:37 +0000)]
Move queue processing logic from CustomElementReactionStack to CustomElementReactionQueue
https://bugs.webkit.org/show_bug.cgi?id=188277

Reviewed by Wenson Hsieh.

Moved ensureCurrentQueue, processBackupQueue, ElementQueue, ensureBackupQueue, and backupElementQueue
from CustomElementReactionStack to CustomElementReactionQueue to simplify CustomElementReactionStack.

No new tests since there is no behavior change.

* dom/CustomElementReactionQueue.cpp:
(WebCore::CustomElementReactionQueue::enqueueElementUpgrade):
(WebCore::CustomElementReactionQueue::enqueueConnectedCallbackIfNeeded):
(WebCore::CustomElementReactionQueue::enqueueDisconnectedCallbackIfNeeded):
(WebCore::CustomElementReactionQueue::enqueueAdoptedCallbackIfNeeded):
(WebCore::CustomElementReactionQueue::enqueueAttributeChangedCallbackIfNeeded):
(WebCore::CustomElementReactionQueue::ElementQueue::add): Moved from CustomElementReactionStack.
(WebCore::CustomElementReactionQueue::ElementQueue::invokeAll): Ditto.
(WebCore::CustomElementReactionQueue::ensureCurrentQueue): Ditto.
(WebCore::CustomElementReactionQueue::ensureBackupQueue): Ditto.
(WebCore::CustomElementReactionQueue::processBackupQueue): Ditto.
(WebCore::CustomElementReactionQueue::backupElementQueue): Ditto.
* dom/CustomElementReactionQueue.h:
(WebCore::CustomElementReactionStack::hasCurrentProcessingStack): Deleted. It was never called.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234520 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoFix some builds after r234516
achristensen@apple.com [Thu, 2 Aug 2018 22:31:48 +0000 (22:31 +0000)]
Fix some builds after r234516
​https://bugs.webkit.org/show_bug.cgi?id=188250

* Platform/IPC/ArgumentCoders.h:
This assertion was comparing integers of different signs.
Since we don't allow exception throwing in WebKit, valueless_by_exception won't be true,
so this assertion will always be true.  I'll just remove it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234519 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agohttpCookieStore.getAllCookies() does not always call completion block
sihui_liu@apple.com [Thu, 2 Aug 2018 22:20:55 +0000 (22:20 +0000)]
httpCookieStore.getAllCookies() does not always call completion block
https://bugs.webkit.org/show_bug.cgi?id=188242

Reviewed by Chris Dumez.

Take background assertion for network process to keep it alive during cookieStore API calls.

* UIProcess/GenericCallback.h:
* UIProcess/WebCookieManagerProxy.cpp:
(WebKit::WebCookieManagerProxy::getHostnamesWithCookies):
(WebKit::WebCookieManagerProxy::deleteCookie):
(WebKit::WebCookieManagerProxy::deleteAllCookiesModifiedSince):
(WebKit::WebCookieManagerProxy::setCookie):
(WebKit::WebCookieManagerProxy::setCookies):
(WebKit::WebCookieManagerProxy::getAllCookies):
(WebKit::WebCookieManagerProxy::getCookies):
(WebKit::WebCookieManagerProxy::setHTTPCookieAcceptPolicy):
(WebKit::WebCookieManagerProxy::getHTTPCookieAcceptPolicy):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234518 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Cocoa] setCookie API fails to set session cookies for defaultDataStore if processPoo...
sihui_liu@apple.com [Thu, 2 Aug 2018 22:18:36 +0000 (22:18 +0000)]
[Cocoa] setCookie API fails to set session cookies for defaultDataStore if processPool created but not used
https://bugs.webkit.org/show_bug.cgi?id=188209

Reviewed by Geoffrey Garen.

Source/WebKit:

If defaultDataStore is already created when launching network process, it means user
probably set cookies for default session via API, so we should update m_websiteDataStore to
make sure pendingCookies is passed to network process.

* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::ensureNetworkProcess):

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/WKHTTPCookieStore.mm:
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234517 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUse WTF::Variant for WebPreferencesStore::Value
achristensen@apple.com [Thu, 2 Aug 2018 21:48:21 +0000 (21:48 +0000)]
Use WTF::Variant for WebPreferencesStore::Value
https://bugs.webkit.org/show_bug.cgi?id=188250

Reviewed by Sam Weinig.

It was conceptually a variant. This just uses an actual WTF::Variant.
Since it's encoded and decoded, I wrote variadic template encoders and decoders
like we have for other types (Expected, std::tuple, KeyValuePair, etc.)

* Platform/IPC/ArgumentCoders.h:
(IPC::VariantCoder::encode):
(IPC::VariantCoder::decode):
Introduce ArgumentCoder<Variant>, which encodes which type the Variant currently
holds followed by the encoding of the current type.
* Platform/IPC/Decoder.cpp:
(IPC::Decoder::Decoder):
* Platform/IPC/Decoder.h:
* Platform/IPC/mac/ConnectionMac.mm:
(IPC::createMessageDecoder):
Use move semantics to prevent an unnecessary copy of the Vector<Attachment>
* Scripts/PreferencesTemplates/WebPreferencesStoreDefaultsMap.cpp.erb:
* Shared/WebPreferencesStore.cpp:
(WebKit::WebPreferencesStore::decode):
Use modern std::optional-based decoding supported by the new Variant decoding.
(WebKit::valueForKey):
Use WTF::get and WTF::holds_alternative instead of custom type checks.
(WebKit::WebPreferencesStore::Value::encode const): Deleted.
(WebKit::WebPreferencesStore::Value::decode): Deleted.
(WebKit::as<String>): Deleted.
(WebKit::as<bool>): Deleted.
(WebKit::as<uint32_t>): Deleted.
(WebKit::as<double>): Deleted.
* Shared/WebPreferencesStore.h:
(WebKit::WebPreferencesStore::Value::Value): Deleted.
(WebKit::WebPreferencesStore::Value::operator=): Deleted.
(WebKit::WebPreferencesStore::Value::~Value): Deleted.
(WebKit::WebPreferencesStore::Value::type const): Deleted.
(WebKit::WebPreferencesStore::Value::asString const): Deleted.
(WebKit::WebPreferencesStore::Value::asBool const): Deleted.
(WebKit::WebPreferencesStore::Value::asUInt32 const): Deleted.
(WebKit::WebPreferencesStore::Value::asDouble const): Deleted.
(WebKit::WebPreferencesStore::Value::destroy): Deleted.
(): Deleted.
Use WTF::Variant instead of a custom type/union pair.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234516 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoReformat editable selection tests to remove potential flakeyness due to autoscroll
megan_gardner@apple.com [Thu, 2 Aug 2018 21:26:43 +0000 (21:26 +0000)]
Reformat editable selection tests to remove potential flakeyness due to autoscroll
https://bugs.webkit.org/show_bug.cgi?id=188263
<rdar://problem/40565375>

Reviewed by Wenson Hsieh

Change the format of these editable selection tests to put the editable content at the top of the page so
that when selecting we do not run into the page trying to autoscroll while selecting, which makes the tests more
flakey, and more difficult to get accurate and consistent answers.

* fast/events/touch/ios/double-tap-on-editable-content-for-selection-then-drag-down-to-change-selected-text.html:
* fast/events/touch/ios/double-tap-on-editable-content-for-selection-then-drag-left-to-change-selected-text.html:
* fast/events/touch/ios/double-tap-on-editable-content-for-selection-then-drag-right-to-change-selected-text.html:
* fast/events/touch/ios/double-tap-on-editable-content-for-selection-then-drag-up-to-change-selected-text.html:
* fast/events/touch/ios/long-press-on-editable-content-then-drag-down-to-change-selected-text.html:
* fast/events/touch/ios/long-press-on-editable-content-then-drag-left-to-change-selected-text.html:
* fast/events/touch/ios/long-press-on-editable-content-then-drag-right-to-change-selected-text.html:
* fast/events/touch/ios/long-press-on-editable-content-then-drag-up-to-change-selected-text.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234515 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoAdd some Variant types to Forward.h
achristensen@apple.com [Thu, 2 Aug 2018 20:24:57 +0000 (20:24 +0000)]
Add some Variant types to Forward.h
https://bugs.webkit.org/show_bug.cgi?id=188250

* wtf/Forward.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoCheck with SafeBrowsing during navigation in WKWebView
achristensen@apple.com [Thu, 2 Aug 2018 20:13:05 +0000 (20:13 +0000)]
Check with SafeBrowsing during navigation in WKWebView
https://bugs.webkit.org/show_bug.cgi?id=188133

Reviewed by Chris Dumez.

This turns WebFramePolicyListenerProxy into an object that now listens for the results
of two processes happening in parallel: the API::NavigationClient's decidePolicyForNavigation{Action, Response}
(which it was already waiting for) and, on platforms that support it, the SafariSafeBrowsing framework's check.
The first result is stored as it waits for the second result unless the first result is the API::NavigationClient
saying to cancel or convert the navigation to a download, in which cases we don't care what the safe browsing
framework results are because we won't show the URL in the browser.

Nothing is done with the safe browsing results yet.

* UIProcess/Cocoa/SafeBrowsingResultCocoa.mm: Added.
(WebKit::SafeBrowsingResult::SafeBrowsingResult):
* UIProcess/Cocoa/WebPageProxyCocoa.mm:
(WebKit::WebPageProxy::beginSafeBrowsingCheck):
* UIProcess/SafeBrowsingResult.h: Added.
(WebKit::SafeBrowsingResult::provider const):
(WebKit::SafeBrowsingResult::isPhishing const):
(WebKit::SafeBrowsingResult::isMalware const):
(WebKit::SafeBrowsingResult::isUnwantedSoftware const):
(WebKit::SafeBrowsingResult::isKnownToBeUnsafe const):
* UIProcess/WebFramePolicyListenerProxy.cpp:
(WebKit::WebFramePolicyListenerProxy::WebFramePolicyListenerProxy):
(WebKit::WebFramePolicyListenerProxy::didReceiveSafeBrowsingResults):
(WebKit::WebFramePolicyListenerProxy::use):
(WebKit::WebFramePolicyListenerProxy::download):
(WebKit::WebFramePolicyListenerProxy::ignore):
* UIProcess/WebFramePolicyListenerProxy.h:
(WebKit::WebFramePolicyListenerProxy::create):
* UIProcess/WebFrameProxy.cpp:
(WebKit::WebFrameProxy::setUpPolicyListenerProxy):
* UIProcess/WebFrameProxy.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::decidePolicyForNavigationAction):
(WebKit::WebPageProxy::decidePolicyForNewWindowAction):
(WebKit::WebPageProxy::decidePolicyForResponse):
* UIProcess/WebPageProxy.h:
* WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234513 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoText selection color is hard to see in dark mode web views.
timothy@apple.com [Thu, 2 Aug 2018 20:12:40 +0000 (20:12 +0000)]
Text selection color is hard to see in dark mode web views.
https://bugs.webkit.org/show_bug.cgi?id=188260
rdar://problem/42721294

Reviewed by Simon Fraser.

Stop using blendWithWhite() to transform the AppKit selection color in dark mode.
Using an alpha of 80% gives good contrast, and still works good for selections over images.

* platform/graphics/Color.cpp:
(WebCore::Color::blendWithWhite const): Mark new colors as semantic if the original is.
(WebCore::Color::colorWithAlpha const): Ditto.
* rendering/RenderElement.cpp:
(WebCore::RenderElement::selectionBackgroundColor const): Use transformSelectionBackgroundColor.
* rendering/RenderTheme.cpp:
(WebCore::RenderTheme::activeSelectionBackgroundColor const): Use transformSelectionBackgroundColor.
(WebCore::RenderTheme::inactiveSelectionBackgroundColor const): Ditto.
(WebCore::RenderTheme::transformSelectionBackgroundColor const): Added. Just blend with white.
* rendering/RenderTheme.h:
* rendering/RenderThemeMac.h:
* rendering/RenderThemeMac.mm:
(WebCore::RenderThemeMac::transformSelectionBackgroundColor const): Added. Use an alpha with the color
in dark mode, otherwise fallback to RenderTheme.
(WebCore::RenderThemeMac::systemColor const): Use activeListBoxSelectionBackgroundColor()
and activeSelectionBackgroundColor() instead of caching the colors again. Update hardcoded color.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234512 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][Floating] Use displayBox.rectWithMargin().bottom instead of displayBox.bottom...
zalan@apple.com [Thu, 2 Aug 2018 20:07:43 +0000 (20:07 +0000)]
[LFC][Floating] Use displayBox.rectWithMargin().bottom instead of displayBox.bottom() to where applicable.
https://bugs.webkit.org/show_bug.cgi?id=188274

Reviewed by Antti Koivisto.

Float placement uses the margin box.

* layout/FloatingContext.cpp:
(WebCore::Layout::FloatingPair::bottom const):
(WebCore::Layout::Iterator::operator++):
(WebCore::Layout::Iterator::set):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234511 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][Floating] Right aligned float's horizontal candidate position is miscomputed.
zalan@apple.com [Thu, 2 Aug 2018 19:56:46 +0000 (19:56 +0000)]
[LFC][Floating] Right aligned float's horizontal candidate position is miscomputed.
https://bugs.webkit.org/show_bug.cgi?id=188273

Reviewed by Antti Koivisto.

rightAlignedBoxLeft is already shifted with the size of the margin box.

* layout/FloatingContext.cpp:
(WebCore::Layout::FloatingContext::alignWithFloatings const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234510 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC] Display:Box::rectWithMargin()'s width and height don't include margin left...
zalan@apple.com [Thu, 2 Aug 2018 19:47:57 +0000 (19:47 +0000)]
[LFC] Display:Box::rectWithMargin()'s width and height don't include margin left/top.
https://bugs.webkit.org/show_bug.cgi?id=188272

Reviewed by Antti Koivisto.

* layout/displaytree/DisplayBox.h:
(WebCore::Display::Box::rectWithMargin const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][Floating] Containing block of a float could push the candidate position beyond...
zalan@apple.com [Thu, 2 Aug 2018 19:40:37 +0000 (19:40 +0000)]
[LFC][Floating] Containing block of a float could push the candidate position beyond the current float.
https://bugs.webkit.org/show_bug.cgi?id=188264

Reviewed by Antti Koivisto.

<div id=first style="float: left; width: 100px; height: 10px;"></div>
<div id=second style="float: left; width: 10px; height: 10px;"></div>
<div style="width: 50px; height: 50px;>
    <div id=third style="float: left; width: 10px; height: 10px;"></div>
</div>
In the example above by the time the "third" float comes in, we are already at 110px horizontally. However the containing block of the "third" float forces
the candidate position at 0px. In such cases intersecting won't work since the candidate position is beyond the last(second) float (horizontally).
We could either keep going back to the previous floats on this vertical position and eventually intersect with "first" or just compare the horizontal
positions and ensure that the left/right float is not going to be placed to the left/right of the last float on the current vertical position.

* layout/FloatingContext.cpp:
(WebCore::Layout::FloatingPair::intersects const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234508 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoImplement customElements.upgrade()
rniwa@webkit.org [Thu, 2 Aug 2018 19:25:23 +0000 (19:25 +0000)]
Implement customElements.upgrade()
https://bugs.webkit.org/show_bug.cgi?id=183397

Reviewed by Frédéric Wang.

LayoutTests/imported/w3c:

Rebaseline the test now that we're passing.

* web-platform-tests/custom-elements/custom-element-registry/upgrade-expected.txt:

Source/WebCore:

Added the support to upgrade custom elements directly. Ordinarily, custom elements get upgraded as they are
inserted / connected into a document but some script libraries and authors want to be able to upgrade them before that.
Also see https://github.com/w3c/webcomponents/issues/710

Implemented the method as specified at:
https://html.spec.whatwg.org/multipage/custom-elements.html#dom-customelementregistry-upgrade

    When invoked, the upgrade(root) method must run these steps:
    1. Let candidates be a list of all of root's shadow-including inclusive descendant elements,
       in shadow-including tree order.
    2. For each candidate of candidates, try to upgrade candidate.

Tests: imported/w3c/web-platform-tests/custom-elements/custom-element-registry/upgrade.html

* dom/CustomElementReactionQueue.cpp:
(WebCore::CustomElementReactionQueue::enqueueElementUpgradeIfDefined): Removed the assertion that the upgraded element
is connected since the whole point of this API is to upgrade a disconnected element.
* dom/CustomElementRegistry.cpp:
(WebCore::upgradeElementsInShadowIncludingdescendants): Added.
(WebCore::CustomElementRegistry::upgrade): Added.
* dom/CustomElementRegistry.h: Forward declare DeferredPromise instead of unnecessarily including JSDOMPromiseDeferred.h.
* dom/CustomElementRegistry.idl:
* dom/Element.cpp:
(WebCore::Element::insertedIntoAncestor): Moved the assertion here.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoConvert calc-related enums to 8-bit enum classes
simon.fraser@apple.com [Thu, 2 Aug 2018 19:19:33 +0000 (19:19 +0000)]
Convert calc-related enums to 8-bit enum classes
https://bugs.webkit.org/show_bug.cgi?id=188215

Reviewed by Zalan Bujtas.

This shrinks CSSCalcOperation from 48 to 32 bytes.

* css/CSSCalculationValue.cpp:
(WebCore::unitCategory):
(WebCore::determineCategory):
(WebCore::resolvedTypeForMinOrMax):
(WebCore::isIntegerResult):
(WebCore::CSSCalcExpressionNodeParser::parseValue):
(WebCore::CSSCalcExpressionNodeParser::parseValueMultiplicativeExpression):
(WebCore::CSSCalcExpressionNodeParser::parseAdditiveValueExpression):
(WebCore::CSSCalcExpressionNodeParser::parseMinMaxExpression):
(WebCore::createBlendHalf):
(WebCore::createCSS):
* css/CSSCalculationValue.h:
* css/CSSPrimitiveValue.cpp:
(WebCore::CSSPrimitiveValue::primitiveType const):
* css/StyleBuilderConverter.h:
(WebCore::StyleBuilderConverter::convertTo100PercentMinusLength):
* css/parser/CSSPropertyParserHelpers.cpp:
(WebCore::CSSPropertyParserHelpers::CalcParser::consumeNumberRaw):
(WebCore::CSSPropertyParserHelpers::CalcParser::consumePositiveIntegerRaw):
(WebCore::CSSPropertyParserHelpers::consumeInteger):
(WebCore::CSSPropertyParserHelpers::consumePositiveIntegerRaw):
(WebCore::CSSPropertyParserHelpers::consumeNumberRaw):
(WebCore::CSSPropertyParserHelpers::consumeNumber):
(WebCore::CSSPropertyParserHelpers::consumeFontWeightNumber):
(WebCore::CSSPropertyParserHelpers::consumeLength):
(WebCore::CSSPropertyParserHelpers::consumePercent):
(WebCore::CSSPropertyParserHelpers::canConsumeCalcValue):
(WebCore::CSSPropertyParserHelpers::consumeLengthOrPercent):
(WebCore::CSSPropertyParserHelpers::consumeAngle):
(WebCore::CSSPropertyParserHelpers::consumeAngleOrPercent):
(WebCore::CSSPropertyParserHelpers::consumeTime):
* platform/CalculationValue.cpp:
(WebCore::CalcExpressionNumber::operator== const):
(WebCore::CalcExpressionOperation::evaluate const):
(WebCore::CalcExpressionOperation::operator== const):
(WebCore::CalcExpressionOperation::dump const):
(WebCore::CalcExpressionLength::operator== const):
(WebCore::CalcExpressionBlendLength::CalcExpressionBlendLength):
(WebCore::CalcExpressionBlendLength::operator== const):
(WebCore::operator<<):
* platform/CalculationValue.h:
(WebCore::CalcExpressionNumber::CalcExpressionNumber):
(WebCore::toCalcExpressionNumber):
(WebCore::CalcExpressionLength::CalcExpressionLength):
(WebCore::toCalcExpressionLength):
(WebCore::CalcExpressionOperation::CalcExpressionOperation):
(WebCore::toCalcExpressionOperation):
(WebCore::toCalcExpressionBlendLength):
* platform/Length.cpp:
(WebCore::convertTo100PercentMinusLength):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoFixed WebKit.AttrStyle timing out on build.webkit.org.
mitz@apple.com [Thu, 2 Aug 2018 19:06:42 +0000 (19:06 +0000)]
Fixed WebKit.AttrStyle timing out on build.webkit.org.

* TestWebKitAPI/Tests/WebKitCocoa/AttrStyle.html: Changed from a URL that requires network
  access to one that doesn’t.
* TestWebKitAPI/Tests/WebKitCocoa/AttrStyle.mm:
(TEST): Updated expected value.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234505 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Keyboard becomes unresponsive after pressing delete while pressing down on...
wenson_hsieh@apple.com [Thu, 2 Aug 2018 18:32:22 +0000 (18:32 +0000)]
[iOS] Keyboard becomes unresponsive after pressing delete while pressing down on a character key with accents
https://bugs.webkit.org/show_bug.cgi?id=188251
<rdar://problem/37842108>

Reviewed by Tim Horton.

Source/WebKit:

Fixes a bug in key event handling where invoking -handleKeyWebEvent:withCompletionHandler: from within the
completion callback of a previous call to -handleKeyWebEvent:withCompletionHandler: would cause the completion
callback to be cleared out prematurely. In some cases (as described in the title of this bug), UIKit exercises
this codepath and subsequently hangs due to their completion block never getting invoked by WebKit.

Test: KeyboardInputTests.CanHandleKeyEventInCompletionHandler

* UIProcess/ios/WKContentViewInteraction.h:
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView handleKeyWebEvent:withCompletionHandler:]):
(-[WKContentView _didHandleKeyEvent:eventWasHandled:]):

Tools:

Adds a new API test to verify that clients (in this case, UIKit) is allowed to invoke
-handleKeyWebEvent:withCompletionHandler: within the completion block of a prior invocation.

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/ios/KeyboardInputTestsIOS.mm: Added.
(TestWebKitAPI::TEST):
* TestWebKitAPI/Tests/ios/TestInputDelegate.h: Added.

Pull some logic used to force an input session to start out from an existing API test file
(WKWebViewAutofillTests) and into a separate helper class that is used by both the existing API tests and the
new keyboard input test.

* TestWebKitAPI/Tests/ios/WKWebViewAutofillTests.mm:
(-[AutofillTestView initWithFrame:]):
(TestWebKitAPI::TEST):
(-[TestInputDelegate init]): Deleted.
(-[TestInputDelegate _webView:focusShouldStartInputSession:]): Deleted.
* TestWebKitAPI/ios/UIKitSPI.h:

Add some UIKit SPI utilized by the new API test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234504 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoHandle zero-sized ISOMP4 boxes appropriately
cturner@igalia.com [Thu, 2 Aug 2018 17:43:40 +0000 (17:43 +0000)]
Handle zero-sized ISOMP4 boxes appropriately
https://bugs.webkit.org/show_bug.cgi?id=188256

Reviewed by Jer Noble.

According to ISO/IEC 14496-12:2012(E), when the Box classes' size
field is zero, the implied size of the box extends to the end of
the file. If this detail is not taken into account, CENC
sanitization can incorrectly report an invalid box size, since 0
!= the number of bytes in this box, specifically, the data layout
of Box is as follows,

aligned(8) class Box (unsigned int(32) boxtype,
      optional unsigned int(8)[16] extended_type) {
   unsigned int(32) size;
   unsigned int(32) type = boxtype;
   if (size==1) {
     unsigned int(64) largesize;
   } else if (size==0) { // This is the case now handled.
     // box extends to end of file
   }
   if (boxtype==‘uuid’) {
     unsigned int(8)[16] usertype = extended_type;
   }
}

Tested by imported/w3c/web-platform-tests/encrypted-media/clearkey-generate-request-disallowed-input.https.html

* platform/graphics/iso/ISOBox.cpp:
(WebCore::ISOBox::peekBox): Check if the parsed size is zero, and
if it is, the size is calculated as the total number of bytes in
the incoming DataView.
(WebCore::ISOBox::parse): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234503 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoOptionally expose Attr::style to JavaScript
mitz@apple.com [Thu, 2 Aug 2018 16:42:39 +0000 (16:42 +0000)]
Optionally expose Attr::style to JavaScript
https://bugs.webkit.org/show_bug.cgi?id=188226
<rdar://problem/42818113>

Reviewed by Darin Adler.

Source/WebCore:

Test: TestWebKitAPI/Tests/WebKitCocoa/AttrStyle.mm

* dom/Attr.cpp: Update comment.
* dom/Attr.idl: Define the style attribute, enabled at runtime by the AttrStyle feature.

* page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::setAttrStyleEnabled): Added this accessor.
(WebCore::RuntimeEnabledFeatures::attrStyleEnabled const): Ditto.

Source/WebKit:

* Shared/WebProcessCreationParameters.cpp:
(WebKit::WebProcessCreationParameters::encode const): Encode new attrStyleEnabled member.
(WebKit::WebProcessCreationParameters::decode): Decode new attrStyleEnabled member.
* Shared/WebProcessCreationParameters.h: Define new attrStyleEnabled member, initialized to
  false.

* UIProcess/API/APIProcessPoolConfiguration.cpp:
(API::ProcessPoolConfiguration::copy): Copy new m_attrStyleEnabled member.
* UIProcess/API/APIProcessPoolConfiguration.h: Define new m_attrStyleEnabled member.

* UIProcess/API/Cocoa/_WKProcessPoolConfiguration.h: Declare new attrStyleEnabled boolean
  property.
* UIProcess/API/Cocoa/_WKProcessPoolConfiguration.mm:
(-[_WKProcessPoolConfiguration attrStyleEnabled]): New accessor.
(-[_WKProcessPoolConfiguration setAttrStyleEnabled:]): Ditto.

* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::initializeNewWebProcess): Initialize parameters.attrStyleEnabled
  from the configuration.

* WebProcess/WebProcess.cpp:
(WebKit::WebProcess::initializeWebProcess): Set the attrStyleEnabled runtime feature based
  on the creation parameters.

Tools:

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKitCocoa/AttrStyle.html: Added.
* TestWebKitAPI/Tests/WebKitCocoa/AttrStyle.mm: Added.
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, rolling out r234489.
david_fenton@apple.com [Thu, 2 Aug 2018 15:45:09 +0000 (15:45 +0000)]
Unreviewed, rolling out r234489.

Caused 50+ crashes and 60+ API failures on iOS

Reverted changeset:

"[WTF] Rename String::format to String::deprecatedFormat"
https://bugs.webkit.org/show_bug.cgi?id=188191
https://trac.webkit.org/changeset/234489

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][Floating] Do not pass formatting root to FloatingContext
zalan@apple.com [Thu, 2 Aug 2018 15:01:48 +0000 (15:01 +0000)]
[LFC][Floating] Do not pass formatting root to FloatingContext
https://bugs.webkit.org/show_bug.cgi?id=188257

Reviewed by Antti Koivisto.

If we ever need it, we can get it from FloatingState.

* layout/FloatingContext.cpp:
(WebCore::Layout::FloatingContext::FloatingContext):
* layout/FloatingContext.h:
* layout/blockformatting/BlockFormattingContext.cpp:
(WebCore::Layout::BlockFormattingContext::layout const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234500 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[LFC][Floating] Convert all coordinates relative to the formatting context root.
zalan@apple.com [Thu, 2 Aug 2018 14:28:28 +0000 (14:28 +0000)]
[LFC][Floating] Convert all coordinates relative to the formatting context root.
https://bugs.webkit.org/show_bug.cgi?id=188249

Reviewed by Antti Koivisto.

The idea here is that all the float related computation should be happening in the same coordinate system.
FloatItem holds the transformed Display::Box for both the layout box and its containing block.

* layout/FloatingContext.cpp:
(WebCore::Layout::begin):
(WebCore::Layout::end):
(WebCore::Layout::FloatingContext::computePosition const):
(WebCore::Layout::FloatingContext::floatingPosition const):
(WebCore::Layout::FloatingContext::initialVerticalPosition const):
(WebCore::Layout::FloatingContext::alignWithContainingBlock const):
(WebCore::Layout::FloatingContext::alignWithFloatings const):
(WebCore::Layout::FloatingContext::toContainingBlock const):
(WebCore::Layout::FloatingPair::FloatingPair):
(WebCore::Layout::FloatingPair::left const):
(WebCore::Layout::FloatingPair::right const):
(WebCore::Layout::Iterator::Iterator):
(WebCore::Layout::previousFloatingIndex):
(WebCore::Layout::Iterator::operator++):
(WebCore::Layout::Iterator::set):
* layout/FloatingContext.h:
* layout/FloatingState.cpp:
(WebCore::Layout::FloatingState::FloatItem::FloatItem):
(WebCore::Layout::FloatingState::remove):
(WebCore::Layout::FloatingState::append):
* layout/FloatingState.h:
(WebCore::Layout::FloatingState::isEmpty const):
(WebCore::Layout::FloatingState::FloatItem::layoutBox const):
(WebCore::Layout::FloatingState::FloatItem::containingBlock const):
(WebCore::Layout::FloatingState::FloatItem::displayBox const):
(WebCore::Layout::FloatingState::FloatItem::containingBlockDisplayBox const):
(WebCore::Layout::FloatingState::floats const):
(WebCore::Layout::FloatingState::last const):
(WebCore::Layout::FloatingState::root const):
(WebCore::Layout::FloatingState::floatings const): Deleted.
* layout/displaytree/DisplayBox.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234499 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[GStreamer] Dynamically adjust blocksize
cturner@igalia.com [Thu, 2 Aug 2018 12:21:53 +0000 (12:21 +0000)]
[GStreamer] Dynamically adjust blocksize
https://bugs.webkit.org/show_bug.cgi?id=188194

Reviewed by Xabier Rodriguez-Calvar.

Update the blocksize depending on how much is obtained from a read
of the input stream. This avoids doing too many reads in small
chunks when larger amounts of data are available and also prevents
using a very large memory area to read a small chunk of data.

Adapted from https://bugzilla.gnome.org/show_bug.cgi?id=767833

No tests new tests since no change in functionality.

* platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:
(webkit_web_src_init):
(CachedResourceStreamingClient::checkUpdateBlocksize):
(CachedResourceStreamingClient::dataReceived):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234498 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[GStreamer] Stop pushing buffers when seeking status changes
cturner@igalia.com [Thu, 2 Aug 2018 09:23:42 +0000 (09:23 +0000)]
[GStreamer] Stop pushing buffers when seeking status changes
https://bugs.webkit.org/show_bug.cgi?id=188193

Reviewed by Xabier Rodriguez-Calvar.

After switching to splitting buffers into smaller block sizes in

    https://bugs.webkit.org/show_bug.cgi?id=182829

It was found that during the individual buffer pushes, the seeking
status could change behind our backs from another thread. When
this happens, buffers from incorrect offsets would find their way
into appsrc and eventually the demuxer itself, which would start
parsing from a random place and at best give a confusing error
message.

The solution here is break from pushing buffers when the seeking
status has been has changed. Flushes will clear out what we've
already delivered into the appsrc, then we must make sure to not
continue sending buffers in there after the flush.

No new tests since this is a timing bug.

* platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:
(CachedResourceStreamingClient::dataReceived):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234497 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, revert TransformationMatrix::operator== change
utatane.tea@gmail.com [Thu, 2 Aug 2018 04:24:11 +0000 (04:24 +0000)]
Unreviewed, revert TransformationMatrix::operator== change
https://bugs.webkit.org/show_bug.cgi?id=188197

This change partially revert cleaning up of TransformationMatrix since memcmp does not
follow double comparison algorithm. So semantics was accidentally changed.

* platform/graphics/transforms/TransformationMatrix.h:
(WebCore::TransformationMatrix::operator== const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234493 268f45cc-cd09-0410-ab3c-d52691b4dbfc