WebKit-https.git
10 months ago[Web Animations] Rebase some flaky tests
graouts@webkit.org [Fri, 26 Oct 2018 18:55:45 +0000 (18:55 +0000)]
[Web Animations] Rebase some flaky tests
https://bugs.webkit.org/show_bug.cgi?id=190969

Reviewed by Dean Jackson.

A few Web Animations tests that are flaky have outdated expectations. Rebasing them to have the output
that is most likely to be produced.

LayoutTests/imported/mozilla:

* css-animations/test_document-get-animations-expected.txt:

LayoutTests/imported/w3c:

* web-platform-tests/web-animations/animation-model/keyframe-effects/effect-value-context-expected.txt:
* web-platform-tests/web-animations/interfaces/Animatable/animate-expected.txt:
* web-platform-tests/web-animations/timing-model/animations/current-time-expected.txt:
* web-platform-tests/web-animations/timing-model/animations/updating-the-finished-state-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237475 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[Web Animations] Remove useless internals methods
graouts@webkit.org [Fri, 26 Oct 2018 18:53:32 +0000 (18:53 +0000)]
[Web Animations] Remove useless internals methods
https://bugs.webkit.org/show_bug.cgi?id=190968

Reviewed by Dean Jackson.

Source/WebCore:

We had a few internals methods added early on in the Web Animations implementation that are no longer
relevant now that the full API is implemented. We can safely remove them now.

* animation/AnimationTimeline.cpp:
(WebCore::AnimationTimeline::setCurrentTime): Deleted.
(WebCore::AnimationTimeline::description): Deleted.
* animation/AnimationTimeline.h:
(WebCore::AnimationTimeline::currentTime):
(WebCore::AnimationTimeline::pause): Deleted.
* animation/DocumentTimeline.cpp:
(WebCore::DocumentTimeline::currentTime):
(WebCore::DocumentTimeline::pause): Deleted.
* animation/DocumentTimeline.h:
* testing/Internals.cpp:
(WebCore::Internals::timelineDescription): Deleted.
(WebCore::Internals::pauseTimeline): Deleted.
(WebCore::Internals::setTimelineCurrentTime): Deleted.
* testing/Internals.h:
* testing/Internals.idl:

LayoutTests:

Remove a test that is no longer useful.

* webanimations/animation-creation-addition-expected.txt: Removed.
* webanimations/animation-creation-addition.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoFix missing edge cases with JSGlobalObjects having a bad time.
mark.lam@apple.com [Fri, 26 Oct 2018 18:30:13 +0000 (18:30 +0000)]
Fix missing edge cases with JSGlobalObjects having a bad time.
https://bugs.webkit.org/show_bug.cgi?id=189028
<rdar://problem/45204939>

Reviewed by Saam Barati.

JSTests:

* stress/regress-189028.js: Added.

Source/JavaScriptCore:

Consider the following scenario:

    let object O1 (of global G1) have an indexing type that is not SlowPut.
    let global G2 have a bad time.
    let object O2 (of global G2) be set as the prototype of O1.
    let object O3 (of global G2) have indexed accessors.

In the existing code, if we set O3 as O2's prototype, we'll have a bug where
O1 will not be made aware that that there are indexed accessors in its prototype
chain.

In this patch, we solve this issue by introducing a new invariant:

    A prototype chain is considered to possibly have indexed accessors if any
    object in the chain belongs to a global object that is having a bad time.

We apply this invariant as follows:

1. Enhance JSGlobalObject::haveABadTime() to also check if other global objects are
   affected by it having a bad time.  If so, it also ensures that those affected
   global objects have a bad time.

   The original code for JSGlobalObject::haveABadTime() uses a ObjectsWithBrokenIndexingFinder
   to find all objects affected by the global object having a bad time.  We enhance
   ObjectsWithBrokenIndexingFinder to also check for the possibility that any global
   objects may be affected by other global objects having a bad time i.e.

        let g1 = global1
        let g2 = global2
        let o1 = an object in g1
        let o2 = an object in g2

        let g1 have a bad time
        g2 is affected if
            o1 is in the prototype chain of o2,
            and o2 may be a prototype.

   If the ObjectsWithBrokenIndexingFinder does find the possibility of other global
   objects being affected, it will abort its heap scan and let haveABadTime() take
   a slow path to do a more complete multi global object scan.

   The slow path works as follows:

   1. Iterate the heap and record the graph of all global object dependencies.

      For each global object, record the list of other global objects that are
      affected by it.

   2. Compute a list of global objects that need to have a bad time using the
      current global object dependency graph.

   3. For each global object in the list of affected global objects, fire their
      HaveABadTime watchpoint and convert all their array structures to the
      SlowPut alternatives.

   4. Re-run ObjectsWithBrokenIndexingFinder to find all objects that are affected
      by any of the globals in the list from (2).

2. Enhance Structure::mayInterceptIndexedAccesses() to also return true if the
   structure's global object is having a bad time.

Note: there are 3 scenarios that we need to consider:

    let g1 = global1
    let g2 = global2
    let o1 = an object in g1
    let o2 = an object in g2

    Scenario 1: o2 is a prototype, and
                g1 has a bad time after o1 is inserted into the o2's prototype chain.

    Scenario 2: o2 is a prototype, and
                o1 is inserted into the o2's prototype chain after g1 has a bad time.

    Scenario 3: o2 is NOT a prototype, and
                o1 is inserted into the o2's prototype chain after g1 has a bad time.

    For scenario 1, when g1 has a bad time, we need to also make sure g2 has
    a bad time.  This is handled by enhancement 1 above.

    For scenario 2, when o1 is inserted into o2's prototype chain, we need to check
    if o1's global object has a bad time.  If so, then we need to make sure o2's
    global also has a bad time (because o2 is a prototype) and convert o2's
    storage type to SlowPut.  This is handled by enhancement 2 above in conjunction
    with JSObject::setPrototypeDirect().

    For scenario 3, when o1 is inserted into o2's prototype chain, we need to check
    if o1's global object has a bad time.  If so, then we only need to convert o2's
    storage type to SlowPut (because o2 is NOT a prototype).  This is handled by
    enhancement 2 above.

3. Also add $vm.isHavingABadTime(), $vm.createGlobalObject() to enable us to
   write some tests for this issue.

* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::fireWatchpointAndMakeAllArrayStructuresSlowPut):
(JSC::JSGlobalObject::haveABadTime):
* runtime/JSGlobalObject.h:
* runtime/JSObject.h:
(JSC::JSObject::mayInterceptIndexedAccesses): Deleted.
* runtime/JSObjectInlines.h:
(JSC::JSObject::mayInterceptIndexedAccesses):
* runtime/Structure.h:
* runtime/StructureInlines.h:
(JSC::Structure::mayInterceptIndexedAccesses const):
* tools/JSDollarVM.cpp:
(JSC::functionHaveABadTime):
(JSC::functionIsHavingABadTime):
(JSC::functionCreateGlobalObject):
(JSC::JSDollarVM::finishCreation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237469 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoWeb Inspector: Crash in http/tests/inspector/network/resource-response-source-memory...
commit-queue@webkit.org [Fri, 26 Oct 2018 18:16:54 +0000 (18:16 +0000)]
Web Inspector: Crash in http/tests/inspector/network/resource-response-source-memory-cache-revalidate-expired-only.html
https://bugs.webkit.org/show_bug.cgi?id=190955

Patch by Antoine Quint <graouts@apple.com> on 2018-10-26
Reviewed by Dean Jackson.

We can get in situations when running tests where runtime flags are not consistent throughout the time a test is run since
showing the Web Inspector can cause flags to be re-set after the initial test was loaded. As such, to avoid crashes due to
the ASSERT(!frame().animation().hasAnimations()) in FrameView::didDestroyRenderTree(), we now cancel animations upon teardown
no matter what the value of the runtime flag for the Web Animations CSS Integration on both the DocumentTimeline (if it exists)
and the CSSAnimationController.

* dom/Element.cpp:
(WebCore::Element::removedFromAncestor):
* dom/PseudoElement.cpp:
(WebCore::PseudoElement::clearHostElement):
* page/FrameView.cpp:
(WebCore::FrameView::didDestroyRenderTree):
* rendering/updating/RenderTreeUpdater.cpp:
(WebCore::RenderTreeUpdater::tearDownRenderers):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[PSON] Avoid tearing down the drawing area when suspending a WebPage due to process...
cdumez@apple.com [Fri, 26 Oct 2018 18:01:39 +0000 (18:01 +0000)]
[PSON] Avoid tearing down the drawing area when suspending a WebPage due to process-swap
https://bugs.webkit.org/show_bug.cgi?id=190879

Reviewed by Antti Koivisto.

Avoid tearing down the drawing area when suspending a WebPage due to process-swap. We really only need to reset
the drawing area upon resuming the WebPage. There is no strict need to destroy the drawing area on suspension
and this has caused various crashes because code usually assumes we always have a drawing area.

This patch also drops various drawing area null checks that were added to address PSON crashes.

* UIProcess/SuspendedPageProxy.cpp:
(WebKit::SuspendedPageProxy::tearDownDrawingAreaInWebProcess): Deleted.
* UIProcess/SuspendedPageProxy.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::suspendCurrentPageIfPossible):
(WebKit::WebPageProxy::receivedNavigationPolicyDecision):
(WebKit::WebPageProxy::didCompletePageTransition):
(WebKit::WebPageProxy::enterAcceleratedCompositingMode):
* UIProcess/WebPageProxy.h:
* WebProcess/WebCoreSupport/WebChromeClient.cpp:
(WebKit::WebChromeClient::invalidateContentsAndRootView):
(WebKit::WebChromeClient::invalidateContentsForSlowScroll):
(WebKit::WebChromeClient::contentsSizeChanged const):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::transitionToCommittedForNewPage):
* WebProcess/WebPage/RemoteLayerTree/RemoteScrollingCoordinator.mm:
(WebKit::RemoteScrollingCoordinator::scheduleTreeStateCommit):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::reinitializeWebPage):
(WebKit::WebPage::exitAcceleratedCompositingMode):
(WebKit::WebPage::setIsSuspended):
(WebKit::WebPage::tearDownDrawingAreaForSuspend): Deleted.
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUse dark appearance scrollbar when page background is dark or document supports dark...
timothy@apple.com [Fri, 26 Oct 2018 17:47:10 +0000 (17:47 +0000)]
Use dark appearance scrollbar when page background is dark or document supports dark mode.
https://bugs.webkit.org/show_bug.cgi?id=190937
rdar://problem/41225839

Reviewed by Beth Dakin.

Source/WebCore:

* page/FrameView.cpp:
(WebCore::FrameView::useDarkAppearance const): Added. Ask the document.
(WebCore::FrameView::paintScrollCorner): Set LocalDefaultSystemAppearance based
on the scrollbar overlay style too.
* page/FrameView.h:
* platform/ScrollableArea.h:
(WebCore::ScrollableArea::useDarkAppearance const): Added. Default to false.
* platform/mac/ScrollAnimatorMac.mm:
(-[WebScrollerImpDelegate effectiveAppearanceForScrollerImp:]): Added.
Ask the ScrollableArea if a dark appearance is desired.

Source/WebCore/PAL:

* pal/spi/mac/NSScrollerImpSPI.h: Added effectiveAppearanceForScrollerImp:.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[High Sierra/Sierra Release] Layout Test media/track/track-cue-css.html is a flaky...
pvollan@apple.com [Fri, 26 Oct 2018 17:44:38 +0000 (17:44 +0000)]
[High Sierra/Sierra Release] Layout Test media/track/track-cue-css.html is a flaky image failure
https://bugs.webkit.org/show_bug.cgi?id=190876

Unreviewed test fix.

This is a speculative flakiness fix.

* media/track/track-cue-css-expected.html:
* media/track/track-cue-css.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[iOS] Unreviewed test gardening.
zalan@apple.com [Fri, 26 Oct 2018 17:34:55 +0000 (17:34 +0000)]
[iOS] Unreviewed test gardening.

* platform/ios/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237464 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUse random() instead of begin() to limit cache sizes
antti@apple.com [Fri, 26 Oct 2018 17:07:49 +0000 (17:07 +0000)]
Use random() instead of begin() to limit cache sizes
https://bugs.webkit.org/show_bug.cgi?id=190957

Reviewed by Chris Dumez.

Source/WebCore:

We currently use cache.remove(cache.begin()) pattern to limit sized of various caches.
This is a bad pattern for tables that never rehash (because they have fixed maximum size) as most of the
keys get permanently stuck in the table.

* css/CSSValuePool.cpp:
(WebCore::CSSValuePool::createColorValue):
(WebCore::CSSValuePool::createFontFamilyValue):
(WebCore::CSSValuePool::createFontFaceValue):
* dom/InlineStyleSheetOwner.cpp:
(WebCore::InlineStyleSheetOwner::createSheet):
* dom/SelectorQuery.cpp:
* platform/graphics/FontCascade.cpp:
(WebCore::retrieveOrAddCachedFonts):
* platform/graphics/cocoa/FontCacheCoreText.cpp:
(WebCore::shouldAutoActivateFontIfNeeded):
* platform/mac/PublicSuffixMac.mm:
(WebCore::topPrivatelyControlledDomain):

Source/WebKit:

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237463 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoJSC xcconfig should set DEFINES_MODULE
keith_miller@apple.com [Fri, 26 Oct 2018 16:47:37 +0000 (16:47 +0000)]
JSC xcconfig should set DEFINES_MODULE
https://bugs.webkit.org/show_bug.cgi?id=190952

Reviewed by Mark Lam.

This should mean that the JavaScriptCore.framework will have a module map.

* Configurations/JavaScriptCore.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237462 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agohashSet.remove(hashSet.random()) doesn't build
antti@apple.com [Fri, 26 Oct 2018 16:27:23 +0000 (16:27 +0000)]
hashSet.remove(hashSet.random()) doesn't build
https://bugs.webkit.org/show_bug.cgi?id=190953

Reviewed by Chris Dumez.

Source/WTF:

* wtf/HashSet.h:

Remove non-const random(). HashSet only returns const iterators (it is immutable via iterator).

* wtf/HashTable.h:
(WTF::HashTable::random const):

Invoke const_iterator() by using static_cast<> instead of trying to do it directly.

Tools:

* TestWebKitAPI/Tests/WTF/HashSet.cpp:
(TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoModernize / Simplify IPC::Connection::sendOutgoingMessage()
cdumez@apple.com [Fri, 26 Oct 2018 15:34:18 +0000 (15:34 +0000)]
Modernize / Simplify IPC::Connection::sendOutgoingMessage()
https://bugs.webkit.org/show_bug.cgi?id=190931

Reviewed by Alex Christensen.

* Platform/IPC/mac/ConnectionMac.mm:
(IPC::Connection::sendOutgoingMessage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237460 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[GTK] Unreviewed test gardening.
ajuma@chromium.org [Fri, 26 Oct 2018 15:29:11 +0000 (15:29 +0000)]
[GTK] Unreviewed test gardening.
https://bugs.webkit.org/show_bug.cgi?id=190962

Add a baseline for the test added in r237449.

* platform/gtk/fast/repaint/text-selection-overflow-hidden-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoAdopt -setOverrideRouteSharingPolicy:routingContextUID: SPI
jer.noble@apple.com [Fri, 26 Oct 2018 15:04:44 +0000 (15:04 +0000)]
Adopt -setOverrideRouteSharingPolicy:routingContextUID: SPI
https://bugs.webkit.org/show_bug.cgi?id=190951
<rdar://problem/45213065>

Reviewed by Alex Christensen.

Source/WebCore:

Request the correct route policy and context from the VideoFullscreenModel.

* platform/cocoa/VideoFullscreenModel.h:
(WebCore::VideoFullscreenModel::requestRouteSharingPolicyAndContextUID):
* platform/cocoa/VideoFullscreenModelVideoElement.h:
* platform/cocoa/VideoFullscreenModelVideoElement.mm:
(WebCore::VideoFullscreenModelVideoElement::requestRouteSharingPolicyAndContextUID):
* platform/ios/VideoFullscreenInterfaceAVKit.h:
* platform/ios/VideoFullscreenInterfaceAVKit.mm:
(-[WebAVPlayerViewController setWebKitOverrideRouteSharingPolicy:routingContextUID:]):
(VideoFullscreenInterfaceAVKit::setVideoFullscreenModel):
(VideoFullscreenInterfaceAVKit::doSetup):

Source/WebCore/PAL:

* pal/spi/cocoa/AVKitSPI.h:

Source/WebKit:

Add an asyncronous reply request to VideoFullscreenManager.

Drive-by fix: messages.py generates "WTFMove(*arg1, *arg2)" instead of "WTFMove(*arg1), WTFMove(*arg2)"
in its reply handler generation.

* Platform/IPC/MessageSender.h:
(IPC::MessageSender::sendWithAsyncReply):
* Scripts/webkit/messages.py:
* UIProcess/Cocoa/VideoFullscreenManagerProxy.h:
* UIProcess/Cocoa/VideoFullscreenManagerProxy.mm:
(WebKit::VideoFullscreenModelContext::requestRouteSharingPolicyAndContextUID):
* WebProcess/cocoa/VideoFullscreenManager.h:
* WebProcess/cocoa/VideoFullscreenManager.messages.in:
* WebProcess/cocoa/VideoFullscreenManager.mm:
(WebKit::VideoFullscreenManager::requestRouteSharingPolicyAndContextUID):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237458 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[LFC][IFC] Add more inline information to outputLayoutTree
zalan@apple.com [Fri, 26 Oct 2018 14:45:51 +0000 (14:45 +0000)]
[LFC][IFC] Add more inline information to outputLayoutTree
https://bugs.webkit.org/show_bug.cgi?id=190945

Reviewed by Antti Koivisto.

* layout/layouttree/LayoutTreeBuilder.cpp:
(WebCore::Layout::outputInlineRuns):
(WebCore::Layout::outputLayoutBox):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[LFC][IFC] Layout floats as part of the inline content
zalan@apple.com [Fri, 26 Oct 2018 14:42:06 +0000 (14:42 +0000)]
[LFC][IFC] Layout floats as part of the inline content
https://bugs.webkit.org/show_bug.cgi?id=190942

Reviewed by Antti Koivisto.

Source/WebCore:

Add float handling to InlineFormattingContext::layoutInlineContent.
Note that floats don't actually generate inline runs.

Test: fast/block/basic/inline-content-with-floating-image.html

* layout/blockformatting/BlockFormattingContext.cpp:
(WebCore::Layout::BlockFormattingContext::layoutFormattingContextRoot const):
* layout/floats/FloatAvoider.cpp:
(WebCore::Layout::FloatAvoider::FloatAvoider): Check if any mapping is needed.
* layout/inlineformatting/InlineFormattingContext.cpp:
(WebCore::Layout::isTrimmableContent):
(WebCore::Layout::InlineFormattingContext::layoutInlineContent const):
(WebCore::Layout::trimLeadingRun): Deleted.
* layout/inlineformatting/InlineFormattingContext.h:
(WebCore::Layout::InlineFormattingContext::Line::isClosed const):
* layout/inlineformatting/Line.cpp:
(WebCore::Layout::InlineFormattingContext::Line::init):
(WebCore::Layout::InlineFormattingContext::Line::adjustLogicalLeft):
(WebCore::Layout::InlineFormattingContext::Line::adjustLogicalRight):
(WebCore::Layout::InlineFormattingContext::Line::appendContent):
(WebCore::Layout::InlineFormattingContext::Line::close):

Tools:

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

LayoutTests:

* fast/block/basic/inline-content-with-floating-image-expected.txt: Added.
* fast/block/basic/inline-content-with-floating-image.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[LFC][IFC] Adjust current line with float constraints.
zalan@apple.com [Fri, 26 Oct 2018 14:40:00 +0000 (14:40 +0000)]
[LFC][IFC] Adjust current line with float constraints.
https://bugs.webkit.org/show_bug.cgi?id=190940

Reviewed by Antti Koivisto.

* layout/inlineformatting/InlineFormattingContext.cpp:
(WebCore::Layout::InlineFormattingContext::initializeNewLine const):
(WebCore::Layout::InlineFormattingContext::layoutInlineContent const):
* layout/inlineformatting/InlineFormattingContext.h:
* layout/inlineformatting/InlineRun.h:
(WebCore::Layout::InlineRun::moveHorizontally):
* layout/inlineformatting/Line.cpp:
(WebCore::Layout::InlineFormattingContext::Line::adjustLogicalLeft):
(WebCore::Layout::InlineFormattingContext::Line::adjustLogicalRight):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237455 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[LFC][IFC] Compute float box size and position
zalan@apple.com [Fri, 26 Oct 2018 14:36:30 +0000 (14:36 +0000)]
[LFC][IFC] Compute float box size and position
https://bugs.webkit.org/show_bug.cgi?id=190938

Reviewed by Antti Koivisto.

* layout/inlineformatting/InlineFormattingContext.cpp:
(WebCore::Layout::InlineFormattingContext::computeWidthAndHeight const):
(WebCore::Layout::InlineFormattingContext::computeFloatPosition const):
* layout/inlineformatting/InlineFormattingContext.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[LFC][IFC] Add line logical top and bottom
zalan@apple.com [Fri, 26 Oct 2018 14:34:54 +0000 (14:34 +0000)]
[LFC][IFC] Add line logical top and bottom
https://bugs.webkit.org/show_bug.cgi?id=190934

Reviewed by Antti Koivisto.

* layout/inlineformatting/InlineFormattingContext.cpp:
(WebCore::Layout::InlineFormattingContext::initializeNewLine const):
(WebCore::Layout::InlineFormattingContext::layoutInlineContent const):
* layout/inlineformatting/InlineFormattingContext.h:
(WebCore::Layout::InlineFormattingContext::Line::isFirstLine const):
(WebCore::Layout::InlineFormattingContext::Line::logicalTop const):
(WebCore::Layout::InlineFormattingContext::Line::logicalBottom const):
* layout/inlineformatting/Line.cpp:
(WebCore::Layout::InlineFormattingContext::Line::init):
(WebCore::Layout::InlineFormattingContext::Line::contentLogicalRight):
(WebCore::Layout::InlineFormattingContext::Line::close):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237453 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[LFC][IFC] Find floating constraints for a given vertical position
zalan@apple.com [Fri, 26 Oct 2018 14:29:22 +0000 (14:29 +0000)]
[LFC][IFC] Find floating constraints for a given vertical position
https://bugs.webkit.org/show_bug.cgi?id=190928

Reviewed by Antti Koivisto.

https://www.w3.org/TR/CSS22/visuren.html#inline-formatting

"In general, the left edge of a line box touches the left edge of its containing block and the right edge touches the right edge of its containing block.
However, floating boxes may come between the containing block edge and the line box edge.
Thus, although line boxes in the same inline formatting context generally have the same width
(that of the containing block), they may vary in width if available horizontal space is reduced due to floats."

This patch adds support for retrieving left/right constraints for a given line (vertical position).

* layout/floats/FloatingState.cpp:
(WebCore::Layout::FloatingState::constraints const):
* layout/floats/FloatingState.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237452 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[LFC] Inline formatting context has higher priority than block
zalan@apple.com [Fri, 26 Oct 2018 14:25:14 +0000 (14:25 +0000)]
[LFC] Inline formatting context has higher priority than block
https://bugs.webkit.org/show_bug.cgi?id=190924

Reviewed by Antti Koivisto.

When an element establishes both inline and block formatting contexts, we need to pick one to
create (and we choose the content driven formatting type (inline)). See example below:

<div style="overflow: hidden">This text should be inside an inlines formatting context.</div>

* layout/LayoutContext.cpp:
(WebCore::Layout::LayoutContext::formattingContext const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237451 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[MSE][WTF][Media] Invalid MediaTime should be falsy
aboya@igalia.com [Fri, 26 Oct 2018 14:05:25 +0000 (14:05 +0000)]
[MSE][WTF][Media] Invalid MediaTime should be falsy
https://bugs.webkit.org/show_bug.cgi?id=190893

Reviewed by Jer Noble.

Source/WTF:

This patch modifies the definition of MediaTime so that invalid times
are evaluated to false in the context of a boolean expression.

* wtf/MediaTime.cpp:
(WTF::MediaTime::operator! const):
(WTF::MediaTime::operator bool const):

Tools:

This patch adds additional API tests for the conversions from
MediaTime to boolean.

* TestWebKitAPI/Tests/WTF/MediaTime.cpp:
(TestWebKitAPI::TEST):

LayoutTests:

A test is added that replicates a bug fixed by this patch.

* media/media-source/media-source-timestampoffset-then-zero-expected.txt: Added.
* media/media-source/media-source-timestampoffset-then-zero.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoREGRESSION (r237255): Text selection is broken in form fields
ajuma@chromium.org [Fri, 26 Oct 2018 13:16:03 +0000 (13:16 +0000)]
REGRESSION (r237255): Text selection is broken in form fields
https://bugs.webkit.org/show_bug.cgi?id=190899

Reviewed by Ryosuke Niwa.

Source/WebCore:

Fix missing negation when checking for a fully-clipped-out rect. This was causing
RenderObject::computeVisibleRectInContainer to incorrectly early-out.

Test: fast/repaint/text-selection-overflow-hidden.html

* rendering/RenderObject.cpp:
(WebCore::RenderObject::computeVisibleRectInContainer const):

LayoutTests:

Add layout test coverage.

* fast/repaint/text-selection-overflow-hidden-expected.txt: Added.
* fast/repaint/text-selection-overflow-hidden.html: Added.
* platform/win/fast/repaint/text-selection-overflow-hidden-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237449 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoSome internal projects include wtf headers and build with C++11
keith_miller@apple.com [Fri, 26 Oct 2018 08:25:43 +0000 (08:25 +0000)]
Some internal projects include wtf headers and build with C++11
https://bugs.webkit.org/show_bug.cgi?id=190791

Reviewed by Alexey Proskuryakov.

C++11 doesn't support constexpr functions that contain
statements. This patch removes getLSBSet set from builds before
C++14 to avoid this for now.

* wtf/MathExtras.h:
(getLSBSet):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237448 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[JSC] havingABadTimeWatchpoint is not required in Array#indexOf optimization
yusukesuzuki@slowstart.org [Fri, 26 Oct 2018 07:56:56 +0000 (07:56 +0000)]
[JSC] havingABadTimeWatchpoint is not required in Array#indexOf optimization
https://bugs.webkit.org/show_bug.cgi?id=190941

Reviewed by Saam Barati.

While "Rest" operation fast path requires havingABadTimeWatchpoint since it allocates
JSArray, Array#{indexOf,lastIndexOf} do not require it when we use the fast path for them.
This patch removes watching on havingABadTimeWatchpoint in Array#indexOf. The test causing
"havingABadTime" is already included in our test suites (e.g. array-indexof-have-a-bad-time.js).

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::handleIntrinsicCall):
* runtime/JSArrayInlines.h:
(JSC::JSArray::canDoFastIndexedAccess):
* runtime/JSGlobalObject.h:
* runtime/JSGlobalObjectInlines.h:
(JSC::JSGlobalObject::isArrayPrototypeIteratorProtocolFastAndNonObservable):
(JSC::JSGlobalObject::isArrayPrototypeIndexedAccessFastAndNonObservable): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237447 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[PSON] Navigating cross-site with locked history but unlocked back/forward list fails...
cdumez@apple.com [Fri, 26 Oct 2018 05:06:24 +0000 (05:06 +0000)]
[PSON] Navigating cross-site with locked history but unlocked back/forward list fails to create a new BackForwardListItem
https://bugs.webkit.org/show_bug.cgi?id=190915
<rdar://problem/45059069>

Reviewed by Geoffrey Garen.

Source/WebCore:

* history/PageCache.cpp:
(WebCore::canCacheFrame):
Make sure we do not put into PageCache a page whose main frame is showing the initial empty document.
We usually do not try to put those into PageCache because we do not have a HistoryItem to save the
PageCache entry on. However, when we process-swap on a navigation with the history locked, the new
process has a HistoryItem and is initially showing the initial empty document before continuing
the load from the previous process. Note that saving the initial empty document in PageCache would
lead to crashes later on previous the initial empty document's Window is taken and reused for the
next load.

* loader/FrameLoader.cpp:
(WebCore::FrameLoader::load):
Stop assuming that we're continuing a client-side redirect when lockHistory is Yes. It is
lockBackForwardList that is actually Yes when we're doing a client-side redirect.

* loader/PolicyChecker.cpp:
(WebCore::PolicyChecker::checkNavigationPolicy):
Stop using calling the completion handler with an invalid URL when the policy decision is 'Suspend' and
use 'about:blank' instead. Without this change, FrameLoader::continueLoadAfterNavigationPolicy() would
not load 'about:blank' when its AllowNavigationToInvalidURL parameter is No.

Tools:

Add API test coverage.

* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237446 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUnreviewed, partial rolling in r237254
yusukesuzuki@slowstart.org [Fri, 26 Oct 2018 04:51:59 +0000 (04:51 +0000)]
Unreviewed, partial rolling in r237254
https://bugs.webkit.org/show_bug.cgi?id=190340

We do not use the added function right now to investigate what is the reason of the regression.
If it causes the regression, it seems that Parser.cpp's inlining decision seems culprit.

* bytecode/UnlinkedFunctionExecutable.cpp:
(JSC::UnlinkedFunctionExecutable::fromGlobalCode):
* bytecode/UnlinkedFunctionExecutable.h:
* parser/Parser.cpp:
(JSC::Parser<LexerType>::parseInner):
(JSC::Parser<LexerType>::parseSingleFunction):
(JSC::Parser<LexerType>::parseFunctionInfo):
(JSC::Parser<LexerType>::parseFunctionDeclaration):
(JSC::Parser<LexerType>::parseAsyncFunctionDeclaration):
* parser/Parser.h:
(JSC::Parser<LexerType>::parse):
(JSC::parse):
(JSC::parseFunctionForFunctionConstructor):
* parser/ParserModes.h:
* parser/ParserTokens.h:
(JSC::JSTextPosition::JSTextPosition):
(JSC::JSTokenLocation::JSTokenLocation): Deleted.
* parser/SourceCodeKey.h:
(JSC::SourceCodeKey::SourceCodeKey):
(JSC::SourceCodeKey::operator== const):
* runtime/CodeCache.cpp:
(JSC::CodeCache::getUnlinkedGlobalCodeBlock):
(JSC::CodeCache::getUnlinkedGlobalFunctionExecutable):
* runtime/CodeCache.h:
* runtime/FunctionConstructor.cpp:
(JSC::constructFunctionSkippingEvalEnabledCheck):
* runtime/FunctionExecutable.cpp:
(JSC::FunctionExecutable::fromGlobalCode):
* runtime/FunctionExecutable.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237445 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoLayout Test http/tests/websocket/tests/hybi/handshake-ok-with-legacy-sec-websocket...
ryanhaddad@apple.com [Fri, 26 Oct 2018 00:59:04 +0000 (00:59 +0000)]
Layout Test http/tests/websocket/tests/hybi/handshake-ok-with-legacy-sec-websocket-response-headers.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=173041

Unreviewed test gardening.

* platform/ios/TestExpectations: Mark test as flaky.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237443 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoCheck in corresponding C code in JetStream 2
sbarati@apple.com [Fri, 26 Oct 2018 00:44:44 +0000 (00:44 +0000)]
Check in corresponding C code in JetStream 2
https://bugs.webkit.org/show_bug.cgi?id=190733

Rubber-stamped by Filip Pizlo.

* JetStream2/simple/float-mm.c: Added.
* JetStream2/wasm/HashSet.cpp: Added.
* JetStream2/wasm/gcc-loops.cpp: Added.
* JetStream2/wasm/quicksort.c: Added.
* JetStream2/wasm/TSF: Added.
* JetStream2/wasm/TSF/build.sh: Added.
* JetStream2/wasm/TSF/config.h: Added.
* JetStream2/wasm/TSF/gpc.h: Added.
* JetStream2/wasm/TSF/gpc_code_gen_util.c: Added.
* JetStream2/wasm/TSF/gpc_instruction.c: Added.
* JetStream2/wasm/TSF/gpc_instruction_dispatch.gen: Added.
* JetStream2/wasm/TSF/gpc_instruction_size.gen: Added.
* JetStream2/wasm/TSF/gpc_instruction_stack_effects.gen: Added.
* JetStream2/wasm/TSF/gpc_instruction_static_size.gen: Added.
* JetStream2/wasm/TSF/gpc_instruction_to_string.gen: Added.
* JetStream2/wasm/TSF/gpc_int_common.h: Added.
* JetStream2/wasm/TSF/gpc_intable.c: Added.
* JetStream2/wasm/TSF/gpc_internal.h: Added.
* JetStream2/wasm/TSF/gpc_interpreter.gen: Added.
* JetStream2/wasm/TSF/gpc_program.c: Added.
* JetStream2/wasm/TSF/gpc_proto.c: Added.
* JetStream2/wasm/TSF/gpc_stack_height.c: Added.
* JetStream2/wasm/TSF/gpc_threaded.c: Added.
* JetStream2/wasm/TSF/gpc_worklist.h: Added.
* JetStream2/wasm/TSF/tsf.h: Added.
* JetStream2/wasm/TSF/tsf_adaptive_reader.c: Added.
* JetStream2/wasm/TSF/tsf_asprintf.c: Added.
* JetStream2/wasm/TSF/tsf_atomics.h: Added.
* JetStream2/wasm/TSF/tsf_buf_reader.c: Added.
* JetStream2/wasm/TSF/tsf_buf_writer.c: Added.
* JetStream2/wasm/TSF/tsf_buffer.c: Added.
* JetStream2/wasm/TSF/tsf_build_defines.h: Added.
* JetStream2/wasm/TSF/tsf_config.h: Added.
* JetStream2/wasm/TSF/tsf_config_stub.h: Added.
* JetStream2/wasm/TSF/tsf_copier.c: Added.
* JetStream2/wasm/TSF/tsf_define_helpers.c: Added.
* JetStream2/wasm/TSF/tsf_define_helpers.h: Added.
* JetStream2/wasm/TSF/tsf_destructor.c: Added.
* JetStream2/wasm/TSF/tsf_error.c: Added.
* JetStream2/wasm/TSF/tsf_format.h: Added.
* JetStream2/wasm/TSF/tsf_fsdb.c: Added.
* JetStream2/wasm/TSF/tsf_fsdb_protocol.c: Added.
* JetStream2/wasm/TSF/tsf_fsdb_protocol.h: Added.
* JetStream2/wasm/TSF/tsf_generator.c: Added.
* JetStream2/wasm/TSF/tsf_gpc_code_gen.c: Added.
* JetStream2/wasm/TSF/tsf_indent.h: Added.
* JetStream2/wasm/TSF/tsf_internal.h: Added.
* JetStream2/wasm/TSF/tsf_internal_config.h: Added.
* JetStream2/wasm/TSF/tsf_internal_config_stub.h: Added.
* JetStream2/wasm/TSF/tsf_inttypes.h: Added.
* JetStream2/wasm/TSF/tsf_io.c: Added.
* JetStream2/wasm/TSF/tsf_io_utils.c: Added.
* JetStream2/wasm/TSF/tsf_ir.c: Added.
* JetStream2/wasm/TSF/tsf_ir.h: Added.
* JetStream2/wasm/TSF/tsf_ir_different.c: Added.
* JetStream2/wasm/TSF/tsf_ir_different.h: Added.
* JetStream2/wasm/TSF/tsf_ir_speed.c: Added.
* JetStream2/wasm/TSF/tsf_limits.c: Added.
* JetStream2/wasm/TSF/tsf_named_type.c: Added.
* JetStream2/wasm/TSF/tsf_native.c: Added.
* JetStream2/wasm/TSF/tsf_parser.c: Added.
* JetStream2/wasm/TSF/tsf_primitive.c: Added.
* JetStream2/wasm/TSF/tsf_ra_type_man.c: Added.
* JetStream2/wasm/TSF/tsf_reflect.c: Added.
* JetStream2/wasm/TSF/tsf_region.h: Added.
* JetStream2/wasm/TSF/tsf_serial_in_man.c: Added.
* JetStream2/wasm/TSF/tsf_serial_out_man.c: Added.
* JetStream2/wasm/TSF/tsf_serial_protocol.h: Added.
* JetStream2/wasm/TSF/tsf_sha1.c: Added.
* JetStream2/wasm/TSF/tsf_sha1.h: Added.
* JetStream2/wasm/TSF/tsf_sha1_writer.c: Added.
* JetStream2/wasm/TSF/tsf_sort.c: Added.
* JetStream2/wasm/TSF/tsf_st.c: Added.
* JetStream2/wasm/TSF/tsf_st.h: Added.
* JetStream2/wasm/TSF/tsf_st_typetable.c: Added.
* JetStream2/wasm/TSF/tsf_stream_file_input.c: Added.
* JetStream2/wasm/TSF/tsf_stream_file_output.c: Added.
* JetStream2/wasm/TSF/tsf_type.c: Added.
* JetStream2/wasm/TSF/tsf_type_in_map.c: Added.
* JetStream2/wasm/TSF/tsf_type_out_map.c: Added.
* JetStream2/wasm/TSF/tsf_type_table.c: Added.
* JetStream2/wasm/TSF/tsf_types.h: Added.
* JetStream2/wasm/TSF/tsf_util.h: Added.
* JetStream2/wasm/TSF/tsf_version.c: Added.
* JetStream2/wasm/TSF/tsf_zip_abstract.c: Added.
* JetStream2/wasm/TSF/tsf_zip_abstract.h: Added.
* JetStream2/wasm/TSF/tsf_zip_attr.c: Added.
* JetStream2/wasm/TSF/tsf_zip_reader.c: Added.
* JetStream2/wasm/TSF/tsf_zip_writer.c: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237442 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoFix build after r237431 for platforms that don't support FULLSCREEN_API
drousso@apple.com [Fri, 26 Oct 2018 00:39:36 +0000 (00:39 +0000)]
Fix build after r237431 for platforms that don't support FULLSCREEN_API
Need the bug URL (OOPS!).

Reviewed by Joseph Pecoraro.

No new tests. No change in behavior.

* inspector/agents/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::addEventListenersToNode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237441 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUnreviewed build fix for Visual Studio 2017
bfulgham@apple.com [Fri, 26 Oct 2018 00:21:00 +0000 (00:21 +0000)]
Unreviewed build fix for Visual Studio 2017

* API/tests/testapi.c:
(testMarkingConstraintsAndHeapFinalizers):
(main):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237440 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoWeb Inspector: Canvas Recording loading goes significantly slower when "Frame" tree...
drousso@apple.com [Thu, 25 Oct 2018 23:59:43 +0000 (23:59 +0000)]
Web Inspector: Canvas Recording loading goes significantly slower when "Frame" tree element is expanded
https://bugs.webkit.org/show_bug.cgi?id=190497

Reviewed by Joseph Pecoraro.

Source/WebInspectorUI:

Save the `WI.TreeElement` that are in the DOM, and compare that with any `WI.TreeElement`
that would be added to the DOM. If all of the following are true, don't edit the DOM:
 - no focused `WI.TreeElement` that isn't currently visible (we would scroll otherwise)
 - the set of previously visible `WI.TreeElement`s (e.g. not used for scroll padding) has at
   least one of the `WI.TreeElement`s that would be added from this update
    - this means that the user hasn't scrolled beyond the padding `WI.TreeElement`
 - there are no `WI.TreeElement`s that would be added from this update that were NOT added
   in a previous update
    - this covers the case that a `WI.TreeElement` is inserted in the visible area

* UserInterface/Base/Utilities.js:

* UserInterface/Views/TreeOutline.js:
(WI.TreeOutline):
(WI.TreeOutline.prototype.registerScrollVirtualizer):
(WI.TreeOutline.prototype.updateVirtualizedElements.walk):
(WI.TreeOutline.prototype.updateVirtualizedElements):

LayoutTests:

* inspector/unit-tests/set-utilities-expected.txt: Added.
* inspector/unit-tests/set-utilities.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237436 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoJSContext Inspector: Broken frontend, DOMAgent is used but does not exist
commit-queue@webkit.org [Thu, 25 Oct 2018 23:55:12 +0000 (23:55 +0000)]
JSContext Inspector: Broken frontend, DOMAgent is used but does not exist
https://bugs.webkit.org/show_bug.cgi?id=190922
<rdar://problem/45569827>

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2018-10-25
Reviewed by Devin Rousso.

* UserInterface/Controllers/DOMManager.js:
(WI.DOMManager.prototype.requestDocument):
Feature check for agents that may not exist in a JSContext.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237435 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoLayout Test compositing/iframes/display-none-subframe.html is a flaky failure.
realdawei@apple.com [Thu, 25 Oct 2018 23:42:26 +0000 (23:42 +0000)]
Layout Test compositing/iframes/display-none-subframe.html is a flaky failure.
https://bugs.webkit.org/show_bug.cgi?id=186406

Unreviewed.

Patch by Dawei Fenton <realdawei@apple.com> on 2018-10-25

* platform/mac-wk1/TestExpectations: Mark test flaky.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237433 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoWeb Inspector: add corner rounding to the network timing "blocks"
drousso@apple.com [Thu, 25 Oct 2018 23:09:59 +0000 (23:09 +0000)]
Web Inspector: add corner rounding to the network timing "blocks"
https://bugs.webkit.org/show_bug.cgi?id=190379

Reviewed by Timothy Hatcher.

* UserInterface/Views/NetworkTableContentView.css:
(.waterfall .block):
(body[dir=ltr] .waterfall .block): Added.
(body[dir=rtl] .waterfall .block): Added.
(.network-table .waterfall .block + .block): Added.
(.network-table .waterfall .block.filler + .block, .network-table .waterfall .block:not(.request, .response) + :matches(.request, .response)): Added.
(.network-table .waterfall .block:last-child): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237432 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoWeb Inspector: display fullscreen enter/exit events in Timelines and Network node...
drousso@apple.com [Thu, 25 Oct 2018 22:59:29 +0000 (22:59 +0000)]
Web Inspector: display fullscreen enter/exit events in Timelines and Network node waterfalls
https://bugs.webkit.org/show_bug.cgi?id=189874
<rdar://problem/44700000>

Reviewed by Joseph Pecoraro.

Source/JavaScriptCore:

* inspector/protocol/DOM.json:
Allow `data` to be passed to the frontend with `didFireEvent`.

Source/WebCore:

Updated existing test: http/tests/inspector/dom/didFireEvent.html

* inspector/agents/InspectorDOMAgent.h:
* inspector/agents/InspectorDOMAgent.cpp:
(WebCore::EventFiredCallback::handleEvent):
(WebCore::InspectorDOMAgent::didCreateFrontendAndBackend):
(WebCore::InspectorDOMAgent::addEventListenersToNode):
(WebCore::InspectorDOMAgent::discardBindings):
(WebCore::InspectorDOMAgent::eventDidResetAfterDispatch): Added.
Prevent the same event from being sent to the frontend more than once.

* dom/Event.cpp:
(WebCore::Event::resetAfterDispatch):

* dom/Document.cpp:
(WebCore::Document::Document):

* inspector/InspectorInstrumentation.h:
(WebCore::InspectorInstrumentation::eventDidResetAfterDispatch): Added.
* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::eventDidResetAfterDispatchImpl): Added.

Source/WebInspectorUI:

* Localizations/en.lproj/localizedStrings.js:

* UserInterface/Protocol/DOMObserver.js:
(WI.DOMObserver.prototype.didFireEvent):
* UserInterface/Controllers/DOMManager.js:
(WI.DOMManager.prototype.didFireEvent):
Allow `data` to be passed to the frontend with `didFireEvent`.

* UserInterface/Models/DOMNode.js:
(WI.DOMNode):
(WI.DOMNode.getFullscreenDOMEvents): Added.
(WI.DOMNode.prototype.didFireEvent):
(WI.DOMNode.prototype._handleDOMNodeDidFireEvent): Added.
(WI.DOMNode.prototype._addDOMEvent):
(WI.DOMNode.prototype._shouldListenForEventListeners): Added.
If an event is fired on an ancestor of this node, also record that event in this node's
`domEvents`, including the `originator` node.

* UserInterface/Views/NetworkTableContentView.js:
(WI.NetworkTableContentView.prototype._populateWaterfallGraph):
* UserInterface/Views/NetworkTableContentView.css:
(.network-table :not(.header) .cell.waterfall .waterfall-container > .dom-fullscreen): Added.

* UserInterface/Views/DOMEventsBreakdownView.js:
(WI.DOMEventsBreakdownView.prototype.initialLayout):
(WI.DOMEventsBreakdownView.prototype._populateTable):
* UserInterface/Views/DOMEventsBreakdownView.css:
(.dom-events-breakdown .graph > .area.fullscreen): Added.
(.dom-events-breakdown .inherited > .name, .dom-events-breakdown .inherited > .graph > .point): Added.
(.dom-events-breakdown:not(.has-inherited) .originator): Added.

LayoutTests:

* http/tests/inspector/dom/didFireEvent-expected.txt:
* http/tests/inspector/dom/didFireEvent.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237431 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoWeb Inspector: Network: more aggressively snap timing blocks together
drousso@apple.com [Thu, 25 Oct 2018 22:57:45 +0000 (22:57 +0000)]
Web Inspector: Network: more aggressively snap timing blocks together
https://bugs.webkit.org/show_bug.cgi?id=190439

Reviewed by Timothy Hatcher.

* UserInterface/Views/NetworkTableContentView.js:
(WI.NetworkTableContentView.prototype._populateWaterfallGraph.appendBlock):
(WI.NetworkTableContentView.prototype._populateWaterfallGraph):
If the time difference between the end of the previous block and the start of this block
would result in less than 2px of space, extend the next block back to the previous block.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237430 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoCleanup: inline constexpr is redundant as constexpr implies inline
ross.kirsling@sony.com [Thu, 25 Oct 2018 22:23:23 +0000 (22:23 +0000)]
Cleanup: inline constexpr is redundant as constexpr implies inline
https://bugs.webkit.org/show_bug.cgi?id=190819

Reviewed by Mark Lam.

Source/bmalloc:

* bmalloc/Algorithm.h:
(bmalloc::max):
(bmalloc::min):
(bmalloc::mask):
(bmalloc::test):
(bmalloc::isPowerOfTwo):
(bmalloc::roundDownToMultipleOf):
(bmalloc::sizeOf):
(bmalloc::bitCount):
(bmalloc::log2):
* bmalloc/Bits.h:
(bmalloc::bitsArrayLength):
* bmalloc/Sizes.h:
(bmalloc::Sizes::maskSizeClass):

Source/JavaScriptCore:

* bytecode/ArrayProfile.h:
(JSC::asArrayModes):
* runtime/IndexingType.h:
(JSC::isCopyOnWrite):
* runtime/MathCommon.h:
(JSC::maxSafeInteger):
(JSC::minSafeInteger):
* runtime/StackAlignment.h:
(JSC::stackAlignmentBytes):
(JSC::stackAlignmentRegisters):

Source/WTF:

* wtf/Bitmap.h:
(WTF::WordType>::Bitmap):
* wtf/LEBDecoder.h:
(WTF::LEBDecoder::maxByteLength):
* wtf/MathExtras.h:
(defaultMinimumForClamp):
(defaultMaximumForClamp):
(clampToAccepting64):
(isLessThan):
(isLessThanEqual):
(isGreaterThan):
(isGreaterThanEqual):
(WTF::roundUpToPowerOfTwo):
(WTF::maskForSize):
* wtf/Optional.h:
* wtf/PtrTag.h:
(WTF::tagCodePtr):
(WTF::untagCodePtr):
(WTF::retagCodePtr):
(WTF::removeCodePtrTag):
* wtf/StdLibExtras.h:
(WTF::roundUpToMultipleOf):
* wtf/Variant.h:
(WTF::operator==):
(WTF::operator!=):
(WTF::operator>=):
(WTF::operator<=):
(WTF::operator>):
(WTF::operator<):
* wtf/text/StringImpl.h:
(WTF::StringImplShape::StringImplShape):
(WTF::StringImpl::StaticStringImpl::StaticStringImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237429 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agohttps://bugs.webkit.org/show_bug.cgi?id=175597
ap@apple.com [Thu, 25 Oct 2018 20:04:50 +0000 (20:04 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=175597
XMLHttpRequest should not sniff content encoding

Actually, the discussion in Bugzilla claims that it should always have worked on
iOS. Not sure why the test was not enabled from the start.

* platform/ios/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237426 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUpdate test results for https://bugs.webkit.org/show_bug.cgi?id=175597
ap@apple.com [Thu, 25 Oct 2018 20:01:25 +0000 (20:01 +0000)]
Update test results for https://bugs.webkit.org/show_bug.cgi?id=175597
XMLHttpRequest should not sniff content encoding

This code doesn't work like planned on macOS High Sierra, and I'm not sure what's
up with iOS and other platforms. But I can to re-enable the test for Mojave.

* platform/mac/TestExpectations: As expected, this test is passing on macOS Mojave.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237425 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUnreviewed, silence a -Wreturn-type warning
mcatanzaro@igalia.com [Thu, 25 Oct 2018 19:55:12 +0000 (19:55 +0000)]
Unreviewed, silence a -Wreturn-type warning

When every return is supposed to be covered by a switch statement, a release assert or CRASH
is required by GCC.

* Modules/mediastream/libwebrtc/LibWebRTCCertificateGenerator.cpp:
(WebCore::LibWebRTCCertificateGenerator::keyParamsFromCertificateType):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237424 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUnreviewed, another try at fixing the GTK build with sandbox enabled
mcatanzaro@igalia.com [Thu, 25 Oct 2018 19:53:15 +0000 (19:53 +0000)]
Unreviewed, another try at fixing the GTK build with sandbox enabled

* UIProcess/Launcher/glib/BubblewrapLauncher.cpp:
(WebKit::memfd_create): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237423 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoWeb Inspector: Add a watchlist message rule to warn about feature checking new inspec...
commit-queue@webkit.org [Thu, 25 Oct 2018 19:51:42 +0000 (19:51 +0000)]
Web Inspector: Add a watchlist message rule to warn about feature checking new inspector protocol changes
https://bugs.webkit.org/show_bug.cgi?id=190896

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2018-10-25
Reviewed by Matt Baker.

* Scripts/webkitpy/common/config/watchlist:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237422 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[Web Animations] Turn Web Animations CSS Integration off by default
graouts@webkit.org [Thu, 25 Oct 2018 19:17:04 +0000 (19:17 +0000)]
[Web Animations] Turn Web Animations CSS Integration off by default
https://bugs.webkit.org/show_bug.cgi?id=190901

Reviewed by Dean Jackson.

Source/WebCore:

* page/RuntimeEnabledFeatures.h:

Source/WebKit:

* Shared/WebPreferences.yaml:

Source/WebKitLegacy/mac:

* WebView/WebPreferences.mm:
(+[WebPreferences initialize]):

Source/WebKitLegacy/win:

* WebPreferences.cpp:
(WebPreferences::initializeDefaultSettings):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237421 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[High Sierra/Sierra Release] Layout Test media/track/track-cue-css.html is a flaky...
pvollan@apple.com [Thu, 25 Oct 2018 17:44:38 +0000 (17:44 +0000)]
[High Sierra/Sierra Release] Layout Test media/track/track-cue-css.html is a flaky image failure
https://bugs.webkit.org/show_bug.cgi?id=190876

Reviewed by Alexey Proskuryakov.

This is a speculative fix; avoid an unnecessary video seek step in test.

* media/track/track-cue-css-expected.html:
* media/track/track-cue-css.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237420 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoHashMap should support selecting a random entry
ggaren@apple.com [Thu, 25 Oct 2018 17:41:35 +0000 (17:41 +0000)]
HashMap should support selecting a random entry
https://bugs.webkit.org/show_bug.cgi?id=190814

Reviewed by Antti Koivisto.

Source/WTF:

In some cases, remove(begin()) is not quite good enough as a random
eviction strategy. (See https://bugs.webkit.org/show_bug.cgi?id=190792.)
So, let's support real random eviction.

(And by "real" I mean "pseudo".)

* wtf/HashCountedSet.h:
* wtf/HashMap.h:
* wtf/HashSet.h:
* wtf/ListHashSet.h:
(WTF::ListHashSet::random):
(WTF::ListHashSet::random const):
* wtf/LoggingHashMap.h:
* wtf/LoggingHashSet.h: Plumb through the random() iterator.

* wtf/HashTable.h:
(WTF::HashTable::random):
(WTF::HashTable::random const): Implement the random() iterator.
makeIterator() already supports starting from any bucket, so this is
pretty easy.

In the subtle case where we select end(), we choose to wrap around and
return begin(). We expect that clients don't really think of the end()
bucket as being in the domain of the random search. Also, we don't want
to annoy clients who know their tables are non-empty with needless
checks for end().

* wtf/RandomNumber.cpp:
(WTF::weakRandomUint32):
* wtf/RandomNumber.h: Added a free function for weak random numbers so
that clients that want cheap random numbers aren't required to allocate
storage for a WeakRandom generator.

Tools:

Unit testing is fun and easy!

* TestWebKitAPI/Tests/WTF/HashMap.cpp:
(TestWebKitAPI::ZeroHash::hash):
(TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237419 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoSource/WebCore:
jond@apple.com [Thu, 25 Oct 2018 17:09:23 +0000 (17:09 +0000)]
Source/WebCore:
Changed "Under Development" status to use "In Development" instead
https://bugs.webkit.org/show_bug.cgi?id=187615

Reviewed by Joseph Pecoraro.

* features.json: Updated CSS Painting API Level 1 and CSS Properties and Values API Level 1.

Websites/webkit.org:
Added support for "deprecated" feature status
https://bugs.webkit.org/show_bug.cgi?id=187615

Reviewed by Joseph Pecoraro.

* wp-content/themes/webkit/status.php:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237418 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUnreviewed, try to fix GTK build with sandbox enabled
mcatanzaro@igalia.com [Thu, 25 Oct 2018 17:06:53 +0000 (17:06 +0000)]
Unreviewed, try to fix GTK build with sandbox enabled

* UIProcess/Launcher/glib/BubblewrapLauncher.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237417 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[CMake][Ninja][Win] midl.exe fails to create, open and write a temporary file in...
Hironori.Fujii@sony.com [Thu, 25 Oct 2018 16:33:33 +0000 (16:33 +0000)]
[CMake][Ninja][Win] midl.exe fails to create, open and write a temporary file in parallel build in Docker container
https://bugs.webkit.org/show_bug.cgi?id=187725

Reviewed by Michael Catanzaro.

* PlatformWin.cmake: Added USES_TERMINAL to serialize midl.exe
executions.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237416 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[WPE][GTK] Follow-up fixes to the certificate coders
csaavedra@igalia.com [Thu, 25 Oct 2018 15:55:35 +0000 (15:55 +0000)]
[WPE][GTK] Follow-up fixes to the certificate coders
https://bugs.webkit.org/show_bug.cgi?id=190900

Reviewed by Michael Catanzaro.

Cleanup the GRefPtr mess I created earlier. Also we now bail out
if any certificate in the chain has no data instead of returning a
cropped chain.

* Shared/soup/WebCoreArgumentCodersSoup.cpp:
(IPC::ArgumentCoder<CertificateInfo>::encode):
(IPC::ArgumentCoder<CertificateInfo>::decode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237415 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[GTK] Reenable the sandbox
mcatanzaro@igalia.com [Thu, 25 Oct 2018 14:53:09 +0000 (14:53 +0000)]
[GTK] Reenable the sandbox

Previously reviewed by, er, myself. This just reverts the sabotague as our bots should have
the required deps now. Hopefully.

* Source/cmake/OptionsGTK.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237414 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoREGRESSION (236779) scandinaviandesigns.com product pages auto redirect to product...
cdumez@apple.com [Thu, 25 Oct 2018 14:15:21 +0000 (14:15 +0000)]
REGRESSION (236779) scandinaviandesigns.com product pages auto redirect to product image
https://bugs.webkit.org/show_bug.cgi?id=190891
<rdar://problem/45296796>

Reviewed by Antti Koivisto.

Source/WebCore:

When a radio element gets clicked, we should only fire the 'input' and 'change' if the checked state
of the radio element has changed.

Test: fast/dom/HTMLInputElement/radio-element-fires-change-event-only-when-checked-state-changes.html

* html/RadioInputType.cpp:
(WebCore::RadioInputType::didDispatchClick):

LayoutTests:

Add layout test coverage.

* fast/dom/HTMLInputElement/radio-element-fires-change-event-only-when-checked-state-changes-expected.txt: Added.
* fast/dom/HTMLInputElement/radio-element-fires-change-event-only-when-checked-state-changes.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237413 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUnreviewed test gardening.
fred.wang@free.fr [Thu, 25 Oct 2018 14:03:30 +0000 (14:03 +0000)]
Unreviewed test gardening.

Patch by Frederic Wang <fwang@igalia.com> on 2018-10-25

* platform/ios-wk1/TestExpectations: Add failures for tests requiring findString.
* platform/ios-wk1/fast/scrolling/adjust-scroll-offset-on-zoom-expected.txt: Added. Adjust
expectation to handle whitespace difference.
* platform/ios-wk1/fast/scrolling/scrolling-tree-includes-frame-expected.txt: Added. The
scrolling tree is not displayed on WK1.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237412 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoAdd more tests for the Find UI
fred.wang@free.fr [Thu, 25 Oct 2018 12:46:28 +0000 (12:46 +0000)]
Add more tests for the Find UI
https://bugs.webkit.org/show_bug.cgi?id=190823

Patch by Frederic Wang <fwang@igalia.com> on 2018-10-25
Reviewed by Antonio Gomes.

We add more tests for iOS's find UI to cover cases fixed by r224284. Equivalent tests for
subframes currently fails when frame flattening is disabled, they will be handled in bug
184297.

* platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position-expected.png: Added.
* platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position-expected.txt: Added.
* platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position-limit-expected.png: Added.
* platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position-limit-expected.txt: Added.
* platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position-limit.html: Added.
Test with a match at the limit of the scrolling element.
* platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position.html: Added. Test
with a match at the middle of the scrolling element.
* platform/ios/fast/scrolling/find-text-in-overflow-node.html: Use names for parameters to
make it more explicit.
* platform/ios/fast/scrolling/find-text-in-subframe-expected.txt: Added.
* platform/ios/fast/scrolling/find-text-in-subframe.html: Added. Test with a subframe instead
of an overflow node.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237411 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[GTK][AC] Resizing the window doesn't always update the visible rect
zandobersek@gmail.com [Thu, 25 Oct 2018 11:09:40 +0000 (11:09 +0000)]
[GTK][AC] Resizing the window doesn't always update the visible rect
https://bugs.webkit.org/show_bug.cgi?id=189743

Reviewed by Michael Catanzaro.

Disassociate resizing that can be done on an AcceleratedSurface object
between "host resizes" and "client resizes".

The former is done from ThreadedCoordinatedLayerTreeHost directly, and
is currently used only for GTK on X11, where a new pixmap object is
created on each resize, which affects the context ID that is based on
that pixmap object's address.

The latter is done from the composition thread. It's used for GTK on
Wayland and WPE. In both cases, the underlying window object does not
change on each resize, but it's necessary to perform the actual resize
on the composition thread.

So far it hasn't been performed there, which ended up pushing sizes of
the WebKit view and the underlying window object out of sync during
resizing, leaving parts of window unrendered.

* Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.cpp:
(WebKit::ThreadedCompositor::renderLayerTree):
* Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.h:
* WebProcess/WebPage/AcceleratedSurface.cpp:
(WebKit::AcceleratedSurface::hostResize):
(WebKit::AcceleratedSurface::resize): Deleted.
* WebProcess/WebPage/AcceleratedSurface.h:
(WebKit::AcceleratedSurface::clientResize):
* WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.cpp:
(WebKit::ThreadedCoordinatedLayerTreeHost::deviceOrPageScaleFactorChanged):
(WebKit::ThreadedCoordinatedLayerTreeHost::sizeDidChange):
* WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.h:
* WebProcess/WebPage/gtk/AcceleratedSurfaceWayland.cpp:
(WebKit::AcceleratedSurfaceWayland::clientResize):
(WebKit::AcceleratedSurfaceWayland::resize): Deleted.
* WebProcess/WebPage/gtk/AcceleratedSurfaceWayland.h:
* WebProcess/WebPage/gtk/AcceleratedSurfaceX11.cpp:
(WebKit::AcceleratedSurfaceX11::hostResize):
(WebKit::AcceleratedSurfaceX11::resize): Deleted.
* WebProcess/WebPage/gtk/AcceleratedSurfaceX11.h:
* WebProcess/WebPage/wpe/AcceleratedSurfaceWPE.cpp:
(WebKit::AcceleratedSurfaceWPE::clientResize):
(WebKit::AcceleratedSurfaceWPE::resize): Deleted.
* WebProcess/WebPage/wpe/AcceleratedSurfaceWPE.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237410 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoInspectorCanvas is not getting cleared properly for OffscreenCanvas
commit-queue@webkit.org [Thu, 25 Oct 2018 08:34:54 +0000 (08:34 +0000)]
InspectorCanvas is not getting cleared properly for OffscreenCanvas
https://bugs.webkit.org/show_bug.cgi?id=190894
<rdar://problem/45498435>

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2018-10-25
Reviewed by Simon Fraser.

Covered by existing tests not crashing with guard malloc.

InspectorCanvasAgents tracks all CanvasRenderingContexts and needs to
remove its reference when the containing CanvasBase goes away. It does
this by registering as a notification observer, but if it can't map
from the CanvasBase back to the rendering context we were failing to
remove our reference. Enforce CanvasBase classes to notify observers
of destruction while they still have their CanvasRenderingContext.

* html/CanvasBase.cpp:
(WebCore::CanvasBase::~CanvasBase):
(WebCore::CanvasBase::notifyObserversCanvasDestroyed):
* html/CanvasBase.h:
Assert that subclasses notify observers of the canvas being destroyed,
since they will need to do this before m_context is cleared.

* html/CustomPaintCanvas.cpp:
(WebCore::CustomPaintCanvas::~CustomPaintCanvas):
* html/OffscreenCanvas.cpp:
(WebCore::OffscreenCanvas::~OffscreenCanvas):
Follow the new expected pattern of notifying observers before clearing
the context. HTMLCanvasElement already followed this pattern.

* inspector/agents/InspectorCanvasAgent.cpp:
(WebCore::InspectorCanvasAgent::canvasDestroyed):
Add an assertion that would catch this earlier.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237409 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[WPT] Update resource-timing WPT tests to use the right cross origin
jiewen_tan@apple.com [Thu, 25 Oct 2018 07:24:59 +0000 (07:24 +0000)]
[WPT] Update resource-timing WPT tests to use the right cross origin
https://bugs.webkit.org/show_bug.cgi?id=190554

Reviewed by Youenn Fablet.

LayoutTests/imported/w3c:

* web-platform-tests/resource-timing/resource_TAO_cross_origin_redirect_chain-expected.txt:
* web-platform-tests/resource-timing/resource_TAO_cross_origin_redirect_chain.html:
* web-platform-tests/resource-timing/resource_timing_cross_origin_redirect_chain-expected.txt:
* web-platform-tests/resource-timing/resource_timing_cross_origin_redirect_chain.html:

LayoutTests:

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237408 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoREGRESSION (r237331): DismissingActionSheetShouldNotDismissPresentingViewController...
timothy_horton@apple.com [Thu, 25 Oct 2018 06:12:50 +0000 (06:12 +0000)]
REGRESSION (r237331): DismissingActionSheetShouldNotDismissPresentingViewController times out
https://bugs.webkit.org/show_bug.cgi?id=190897

* TestWebKitAPI/Tests/ios/ActionSheetTests.mm:
(TestWebKitAPI::TEST):
Disable the test for now.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237407 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoClean up some obsolete macOS version guards
ap@apple.com [Thu, 25 Oct 2018 00:19:22 +0000 (00:19 +0000)]
Clean up some obsolete macOS version guards
https://bugs.webkit.org/show_bug.cgi?id=190887

Reviewed by Dan Bernstein.

Source/WebCore:

* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::createAVAssetForURL):
(WebCore::MediaPlayerPrivateAVFoundationObjC::maximumDurationToCacheMediaTime const):
(WebCore::MediaPlayerPrivateAVFoundationObjC::didPassCORSAccessCheck const):
(WebCore::MediaPlayerPrivateAVFoundationObjC::wouldTaintOrigin const):

Source/WebCore/PAL:

* pal/spi/mac/TUCallSPI.h:

Source/WebKit:

* PluginProcess/mac/com.apple.WebKit.plugin-common.sb.in:
* UIProcess/Cocoa/WebProcessPoolCocoa.mm:
(WebKit::WebProcessPool::platformInitializeWebProcess):

Source/WebKitLegacy/mac:

* Misc/WebNSPasteboardExtras.mm:
(+[NSPasteboard _web_dragTypesForURL]):
* WebCoreSupport/PopupMenuMac.mm:
(PopupMenuMac::show):
* WebCoreSupport/WebEditorClient.h:
* WebCoreSupport/WebEditorClient.mm:
(WebEditorClient::respondToChangedSelection):
* WebView/WebPDFView.mm:
(-[WebPDFView _clipViewForPDFDocumentView]):

Tools:

* TestWebKitAPI/cocoa/TestWKWebView.mm:
(__simulated_forceClickAssociatedEventsMask):
(-[TestWKWebViewHostWindow _mouseDownAtPoint:simulatePressure:clickCount:]):
* WebKitTestRunner/cocoa/TestControllerCocoa.mm:
(WTR::initializeWebViewConfiguration):
* WebKitTestRunner/mac/EventSenderProxy.mm:
(-[EventSenderSyntheticEvent initPressureEventAtLocation:globalLocation:stage:pressure:stageTransition:phase:time:eventNumber:window:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237405 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoAttachment filenames with RTL characters should format similar to Finder
timothy_horton@apple.com [Wed, 24 Oct 2018 22:51:03 +0000 (22:51 +0000)]
Attachment filenames with RTL characters should format similar to Finder
https://bugs.webkit.org/show_bug.cgi?id=190736
<rdar://problem/44735946>

Reviewed by Dan Bernstein.

Source/WebCore:

Test: fast/attachment/attachment-title-with-rtl.html

* html/HTMLAttachmentElement.cpp:
(WebCore::HTMLAttachmentElement::attachmentTitleForDisplay const):
* html/HTMLAttachmentElement.h:
Add attachmentTitleForDisplay(), which wraps the non-extension part of
attachmentTitle in BiDi isolates, matching Finder's behavior.

* rendering/RenderThemeIOS.mm:
(WebCore::RenderAttachmentInfo::buildWrappedLines):
(WebCore::RenderAttachmentInfo::RenderAttachmentInfo):
* rendering/RenderThemeMac.mm:
(WebCore::AttachmentLayout::layOutTitle):
Adopt attachmentTitleForDisplay, and ask CoreText to use a subrange of
the original string for the last line, instead of splitting the string
ourselves. This ensures that BiDi control characters are respected
even in the last line of the string.

LayoutTests:

* fast/attachment/attachment-title-with-rtl-expected.html: Added.
* fast/attachment/attachment-title-with-rtl.html: Added.
Add a test that directionality marks in the attachment's title are
isolated from the file extension.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237403 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoTurn on Conic Gradients
megan_gardner@apple.com [Wed, 24 Oct 2018 22:28:42 +0000 (22:28 +0000)]
Turn on Conic Gradients
https://bugs.webkit.org/show_bug.cgi?id=190810

Reviewed by Tim Horton.

Source/JavaScriptCore:

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

Added tests previously, only switching feature from experimental to always avaiable.

* Configurations/FeatureDefines.xcconfig:
* css/parser/CSSParserContext.cpp:
(WebCore::CSSParserContext::CSSParserContext):
(WebCore::operator==):
* css/parser/CSSParserContext.h:
(WebCore::CSSParserContextHash::hash):
* css/parser/CSSPropertyParserHelpers.cpp:
(WebCore::CSSPropertyParserHelpers::consumeConicGradient):
* page/Settings.yaml:

Source/WebCore/PAL:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit:

Remove conic gradients from experimental features and turn it on always.

* Configurations/FeatureDefines.xcconfig:
* Shared/WebPreferences.yaml:

Source/WebKitLegacy/mac:

* Configurations/FeatureDefines.xcconfig:

Source/WTF:

* wtf/FeatureDefines.h:

Tools:

* TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

LayoutTests:

No longer necessary to manually turn on conic gradients for testing.

* http/wpt/css/css-images-4/conic-gradient-parsing.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237402 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoWeb Inspector: popovers for function source code are malformed
drousso@apple.com [Wed, 24 Oct 2018 22:12:54 +0000 (22:12 +0000)]
Web Inspector: popovers for function source code are malformed
https://bugs.webkit.org/show_bug.cgi?id=190859

Reviewed by Joseph Pecoraro.

Previously, the `WI.Popover` would be drawn twice: once when the
`WI.CodeMirrorTokenTrackingController` determines that the user has hovered on a function
token, and once again when the source for that function is formatted (after being retrieved).
In the case that the formatter is able to return the prettified source within a frame (or
two), the `WI.Popover` is still in the middle of animating to its new size, meaning that the
changes made by the formatted `update` will be overridden on the next rAF (creates a flash).

* UserInterface/Views/Popover.js:
(WI.Popover.prototype):
(WI.Popover.prototype._animateFrame.drawBackground):
(WI.Popover.prototype._drawBackground):
Add a member variable to make sure that there is only ever one rAF firing at a time.
Drive-by: rework the background code to only use one canvas.
* UserInterface/Views/SourceCodeTextEditor.js:
(WI.SourceCodeTextEditor.prototype._showPopoverForFunction.didGetDetails):
Don't show the `WI.Popover` until the formatter has finished prettifying the function's
source code, as otherwise there is brief moment that the popover appears and is empty.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237401 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago2018-10-24 Michael Saboff <msaboff@apple.com>
msaboff@apple.com [Wed, 24 Oct 2018 21:35:35 +0000 (21:35 +0000)]
2018-10-24  Michael Saboff  <msaboff@apple.com>

Increase executable memory pool from 64MB to 128MB for ARM64
https://bugs.webkit.org/show_bug.cgi?id=190453

Unreviewed, rolling back in r237024.

The original change did impact ARES-6 performance by 4-8%.  That will
be investigated separately.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237400 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoAdd BPLATFORM(IOS_FAMILY)
ap@apple.com [Wed, 24 Oct 2018 21:01:27 +0000 (21:01 +0000)]
Add BPLATFORM(IOS_FAMILY)
https://bugs.webkit.org/show_bug.cgi?id=190878

Reviewed by Saam Barati.

* bmalloc/AvailableMemory.cpp:
(bmalloc::memorySizeAccordingToKernel):
(bmalloc::computeAvailableMemory):
* bmalloc/AvailableMemory.h:
(bmalloc::isUnderMemoryPressure):
* bmalloc/BPlatform.h:
* bmalloc/Gigacage.h:
* bmalloc/Logging.cpp:
(bmalloc::logVMFailure):
* bmalloc/VMAllocate.h:
(bmalloc::vmPageSizePhysical):
* bmalloc/bmalloc.h:
* bmalloc/darwin/MemoryStatusSPI.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237399 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoWeb Inspector: REGRESSION (r237232): changing resources via up/down blurs the TreeOut...
drousso@apple.com [Wed, 24 Oct 2018 20:19:39 +0000 (20:19 +0000)]
Web Inspector: REGRESSION (r237232): changing resources via up/down blurs the TreeOutline in Resources
https://bugs.webkit.org/show_bug.cgi?id=190862

Reviewed by Matt Baker.

* UserInterface/Views/TextEditor.js:
(WI.TextEditor.prototype.set selectedTextRange):
Don't `focus` unless the `activeElement` is not a WebInspector element (e.g. <body>),
meaning that the previously focused element has been removed from the DOM and no longer
needs to keep the focus.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237397 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoWeb Inspector: CSP request payload on medium.com is unreadable, should be pretty...
drousso@apple.com [Wed, 24 Oct 2018 19:27:27 +0000 (19:27 +0000)]
Web Inspector: CSP request payload on medium.com is unreadable, should be pretty-printed
https://bugs.webkit.org/show_bug.cgi?id=190354
<rdar://problem/45090894>

Reviewed by Joseph Pecoraro.

* UserInterface/Views/TextEditor.js:
(WI.TextEditor.prototype.set string.update):
(WI.TextEditor.prototype._attemptToDetermineMIMEType): Added.
If the content doesn't already have a MIME type, attempt to determine one by trying to
format it as "javascript" (e.g. request JSON that is simply missing a MIME type).

* UserInterface/Views/TextContentView.js:
(WI.TextContentView):
(WI.TextContentView.prototype._handleTextEditorMIMETypeChanged): Added.
* UserInterface/Views/TextResourceContentView.js:
(WI.TextResourceContentView):
(WI.TextResourceContentView.prototype._handleTextEditorMIMETypeChanged): Added.
* UserInterface/Views/ScriptContentView.js:
(WI.ScriptContentView):
(WI.ScriptContentView.prototype._handleTextEditorMIMETypeChanged): Added.
Enable the "Pretty Print" navigation button if the MIME type changes to something that is
able to be formatted.
Drive-by: reorder the creation of the "Pretty Print" button so that it exists if
`_attemptToDetermineMIMEType` finishes synchronously.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237396 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoOnly report resource timing to parent frame for the first iframe load
jiewen_tan@apple.com [Wed, 24 Oct 2018 18:49:03 +0000 (18:49 +0000)]
Only report resource timing to parent frame for the first iframe load
https://bugs.webkit.org/show_bug.cgi?id=190498
<rdar://problem/44347398>

Reviewed by Youenn Fablet.

Source/WebCore:

Only the first iframe navigation or the first iframe navigation after about:blank should be reported.
https://www.w3.org/TR/resource-timing-2/#resources-included-in-the-performanceresourcetiming-interface

Test: http/tests/misc/resource-timing-navigation-in-restored-iframe.html

* loader/FrameLoader.cpp:
(WebCore::FrameLoader::loadWithDocumentLoader):
* loader/FrameLoader.h:
(WebCore::FrameLoader::shouldReportResourceTimingToParentFrame):
(WebCore::FrameLoader::setShouldReportResourceTimingToParentFrame): Deleted.
* loader/ResourceTimingInformation.cpp:
(WebCore::ResourceTimingInformation::addResourceTiming):

LayoutTests:

The layout test is from Chromium change:
https://chromium-review.googlesource.com/c/chromium/src/+/1186215.

* http/tests/misc/resource-timing-navigation-in-restored-iframe-expected.txt: Added.
* http/tests/misc/resource-timing-navigation-in-restored-iframe.html: Added.
* http/tests/misc/resources/alert-then-back.html: Added.
* http/tests/misc/resources/navigate-on-message.html: Added.
* http/tests/misc/resources/post-message-to-parent.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237395 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoCure Windows Direct2D Backend of a nasty case of bitrot
bfulgham@apple.com [Wed, 24 Oct 2018 18:46:39 +0000 (18:46 +0000)]
Cure Windows Direct2D Backend of a nasty case of bitrot
https://bugs.webkit.org/show_bug.cgi?id=190875
<rdar://problem/45523268>

Reviewed by Zalan Bujtas.

The Direct2D backend has drifted out of date with the rest of WebKit.
This patch updates things so we can get a runnable build.

* platform/graphics/FontPlatformData.h:
* platform/graphics/ImageBuffer.cpp:
(WebCore::ImageBuffer::create):
* platform/graphics/ImageBuffer.h:
* platform/graphics/ImageDecoder.h:
* platform/graphics/ImageSource.cpp:
(WebCore::ImageSource::setTargetContext):
* platform/graphics/opentype/OpenTypeMathData.cpp:
(WebCore::OpenTypeMathData::OpenTypeMathData): Need non-default constructor to make Visual
Studio happy.
* platform/graphics/win/GradientDirect2D.cpp:
(WebCore::Gradient::generateGradient):
* platform/graphics/win/GraphicsContextDirect2D.cpp: Update for new display list implementation.
(WebCore::GraphicsContextPlatformPrivate::setAlpha):
(WebCore::GraphicsContext::savePlatformState):
(WebCore::GraphicsContext::restorePlatformState):
(WebCore::GraphicsContext::drawNativeImage):
(WebCore::GraphicsContext::drawPattern):
(WebCore::GraphicsContext::drawRect):
(WebCore::GraphicsContext::drawLine):
(WebCore::GraphicsContext::drawEllipse):
(WebCore::GraphicsContext::drawPath):
(WebCore::GraphicsContext::fillPath):
(WebCore::GraphicsContext::strokePath):
(WebCore::GraphicsContext::fillRect):
(WebCore::GraphicsContext::platformFillRoundedRect):
(WebCore::GraphicsContext::fillRectWithRoundedHole):
(WebCore::GraphicsContext::clip):
(WebCore::GraphicsContext::clipOut):
(WebCore::GraphicsContext::clipPath):
(WebCore::GraphicsContext::clipBounds const):
(WebCore::GraphicsContext::beginPlatformTransparencyLayer):
(WebCore::GraphicsContext::endPlatformTransparencyLayer):
(WebCore::GraphicsContext::setMiterLimit):
(WebCore::GraphicsContext::clearRect):
(WebCore::GraphicsContext::strokeRect):
(WebCore::GraphicsContext::setLineCap):
(WebCore::GraphicsContext::setLineDash):
(WebCore::GraphicsContext::setLineJoin):
(WebCore::GraphicsContext::scale):
(WebCore::GraphicsContext::rotate):
(WebCore::GraphicsContext::translate):
(WebCore::GraphicsContext::concatCTM):
(WebCore::GraphicsContext::setCTM):
(WebCore::GraphicsContext::getCTM const):
(WebCore::GraphicsContext::roundToDevicePixels):
(WebCore::GraphicsContext::drawLinesForText):
(WebCore::GraphicsContext::setURLForRect):
(WebCore::GraphicsContext::setIsCALayerContext):
(WebCore::GraphicsContext::isCALayerContext const):
(WebCore::GraphicsContext::setIsAcceleratedContext):
(WebCore::GraphicsContext::isAcceleratedContext const):
(WebCore::GraphicsContext::setPlatformShouldAntialias):
(WebCore::GraphicsContext::setPlatformShouldSmoothFonts):
(WebCore::GraphicsContext::setPlatformCompositeOperation):
(WebCore::GraphicsContext::platformFillEllipse):
(WebCore::GraphicsContext::platformStrokeEllipse):
* platform/graphics/win/GraphicsLayerDirect2D.cpp:
(WebCore::GraphicsLayer::create):
(): Deleted.
* platform/graphics/win/ImageBufferDataDirect2D.cpp:
(WebCore::ImageBufferData::putData):
* platform/graphics/win/ImageBufferDataDirect2D.h:
* platform/graphics/win/ImageBufferDirect2D.cpp:
(WebCore::ImageBuffer::createCompatibleBuffer):
(WebCore::ImageBuffer::ImageBuffer):
* platform/graphics/win/ImageDecoderDirect2D.cpp:
(WebCore::ImageDecoderDirect2D::bytesDecodedToDetermineProperties const):
(WebCore::ImageDecoderDirect2D::encodedDataStatus const):
(WebCore::ImageDecoderDirect2D::repetitionCount const):
(WebCore::ImageDecoderDirect2D::frameOrientationAtIndex const):
(WebCore::ImageDecoderDirect2D::frameDurationAtIndex const):
(WebCore::ImageDecoderDirect2D::frameAllowSubsamplingAtIndex const):
(WebCore::ImageDecoderDirect2D::frameHasAlphaAtIndex const):
(WebCore::ImageDecoderDirect2D::createFrameImageAtIndex):
(WebCore::ImageDecoderDirect2D::setData):
(WebCore::ImageDecoderDirect2D::bytesDecodedToDetermineProperties): Deleted.
(WebCore::ImageDecoderDirect2D::createFrameImageAtIndex const): Deleted.
* platform/graphics/win/ImageDecoderDirect2D.h:
* platform/graphics/win/ImageDirect2D.cpp:
(WebCore::BitmapImage::drawFrameMatchingSourceSize):
* platform/graphics/win/PatternDirect2D.cpp:
(WebCore::Pattern::createPlatformPattern const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237394 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoCleanup: MIMETypeRegistry functions
commit-queue@webkit.org [Wed, 24 Oct 2018 18:42:22 +0000 (18:42 +0000)]
Cleanup: MIMETypeRegistry functions
https://bugs.webkit.org/show_bug.cgi?id=190838

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2018-10-24
Reviewed by Simon Fraser.

Source/WebCore:

Modernize the initialization of the lists in MIMETypeRegistry. Make the
functions that return these list be self-contained. Use NeverDestroy<> to
allocate the local static variable. Use std::initializer_list() and
makeNeverDestroyed() to initialize NeverDestroy<> variables only once.

supportedImageResourceMIMETypes will be deleted and all the calls to it
will be replaced by supportedImageMIMETypes because they are identical.

* loader/archive/ArchiveFactory.cpp:
(WebCore::ArchiveFactory::registerKnownArchiveMIMETypes): This function
is called while initializing the supportedNonImageMIMETypes(). So it
should not have a direct call to it. Instead, supportedNonImageMIMETypes
is passed to it.
* loader/archive/ArchiveFactory.h:
* platform/MIMETypeRegistry.cpp:
(WebCore::MIMETypeRegistry::supportedImageMIMETypes):
(WebCore::supportedImageMIMETypesForEncoding):
(WebCore::supportedJavaScriptMIMETypes):
(WebCore::MIMETypeRegistry::supportedNonImageMIMETypes):
(WebCore::MIMETypeRegistry::supportedMediaMIMETypes):
(WebCore::pdfMIMETypes):
(WebCore::MIMETypeRegistry::unsupportedTextMIMETypes):
(WebCore::MIMETypeRegistry::isSupportedImageMIMEType):
(WebCore::MIMETypeRegistry::isSupportedImageMIMETypeForEncoding):
(WebCore::MIMETypeRegistry::isSupportedJavaScriptMIMEType):
(WebCore::MIMETypeRegistry::isSupportedNonImageMIMEType):
(WebCore::MIMETypeRegistry::isSupportedMediaMIMEType):
(WebCore::MIMETypeRegistry::isUnsupportedTextMIMEType):
(WebCore::MIMETypeRegistry::isPDFMIMEType):
(WebCore::MIMETypeRegistry::systemPreviewMIMETypes):
(WebCore::MIMETypeRegistry::isSystemPreviewMIMEType):
(WebCore::initializeSupportedImageMIMETypes): Deleted.
(WebCore::initializeSupportedImageMIMETypesForEncoding): Deleted.
(WebCore::initializeSupportedJavaScriptMIMETypes): Deleted.
(WebCore::initializePDFMIMETypes): Deleted.
(WebCore::initializeSupportedNonImageMimeTypes): Deleted.
(WebCore::initializeSupportedMediaMIMETypes): Deleted.
(WebCore::initializeUnsupportedTextMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::isSupportedImageResourceMIMEType): Deleted.
(WebCore::MIMETypeRegistry::getSupportedImageMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getSupportedImageResourceMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getSupportedNonImageMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getSupportedMediaMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getPDFMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getUnsupportedTextMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getSystemPreviewMIMETypes): Deleted.
* platform/MIMETypeRegistry.h:
* platform/graphics/Image.cpp:
(WebCore::Image::supportsType):
* platform/graphics/cg/ImageDecoderCG.cpp:
(WebCore::ImageDecoderCG::encodedDataStatus const):
* platform/graphics/cg/UTIRegistry.cpp:
(WebCore::supportedDefaultImageSourceTypes):
(WebCore::isSupportImageSourceType):
(WebCore::allowedImageUTIs): Deleted.
(WebCore::isAllowedImageUTI): Deleted.
* platform/graphics/cg/UTIRegistry.h:
* platform/mac/PasteboardMac.mm:
(WebCore::Pasteboard::write):

Source/WebKit:

* UIProcess/Cocoa/WKWebViewContentProviderRegistry.mm:
(-[WKWebViewContentProviderRegistry initWithConfiguration:]):

Source/WebKitLegacy/mac:

* Misc/WebNSPasteboardExtras.mm:
(-[NSPasteboard _web_writePromisedRTFDFromArchive:containsImage:]):
Replace isSupportedImageResourceMIMEType() by isSupportedImageMIMEType()
because they were identical.

* WebView/WebHTMLRepresentation.mm:
(+[WebHTMLRepresentation supportedMediaMIMETypes]):
(+[WebHTMLRepresentation supportedNonImageMIMETypes]):
(+[WebHTMLRepresentation supportedImageMIMETypes]):
(+[WebHTMLRepresentation unsupportedTextMIMETypes]):
* WebView/WebView.mm:
(+[WebView _unregisterViewClassAndRepresentationClassForMIMEType:]):
(+[WebView _registerViewClass:representationClass:forURLScheme:]):
(+[WebView registerViewClass:representationClass:forMIMEType:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237393 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[macOS] Layout Test legacy-animation-engine/animations/suspend-resume-animation.html...
ryanhaddad@apple.com [Wed, 24 Oct 2018 18:26:57 +0000 (18:26 +0000)]
[macOS] Layout Test legacy-animation-engine/animations/suspend-resume-animation.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=190883

Unreviewed test gardening.

* platform/mac/TestExpectations: Mark test as flaky.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237392 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoMarked webgl/1.0.2/conformance/glsl/misc/shader-with-reserved-words.html as a flaky...
jlewis3@apple.com [Wed, 24 Oct 2018 18:24:02 +0000 (18:24 +0000)]
Marked webgl/1.0.2/conformance/glsl/misc/shader-with-reserved-words.html as a flaky timout.
https://bugs.webkit.org/show_bug.cgi?id=170877

Unreviewed test gardening.

* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237391 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[WPE] Fix CMake rules in for TestWebKitAPIBase library building in developer mode
commit-queue@webkit.org [Wed, 24 Oct 2018 17:44:21 +0000 (17:44 +0000)]
[WPE] Fix CMake rules in for TestWebKitAPIBase library building in developer mode
https://bugs.webkit.org/show_bug.cgi?id=190328

Patch by Pablo Saavedra <psaavedra@igalia.com> on 2018-10-24
Reviewed by Michael Catanzaro.

* TestWebKitAPI/CMakeLists.txt:
* TestWebKitAPI/PlatformWPE.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237390 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[PSON] When navigating back and forth, 'about:blank' shows up in the back/forward...
cdumez@apple.com [Wed, 24 Oct 2018 16:39:19 +0000 (16:39 +0000)]
[PSON] When navigating back and forth, 'about:blank' shows up in the back/forward list
https://bugs.webkit.org/show_bug.cgi?id=190846
<rdar://problem/45058938>

Reviewed by Antti Koivisto.

Source/WebCore:

When a page gets suspended after a process-swap, we navigate it to about:blank from inside the navigation
policy handler, by overriding the request URL. This normally works fine because we usually process-swap
on standard navigation. However, when we would process-swap on a back/forward navigation, we would end
up using the back/forward navigation load type to do the about:blank load. This would have repercussions
because history navigations update the current history item with the new URL (in this case 'about:blank').
To avoid the issue, switch to a standard load type whenever the client asks us to suspend and we load
'about:blank' as a result.

* loader/FrameLoader.cpp:
(WebCore::FrameLoader::continueLoadAfterNavigationPolicy):

Tools:

Add API test coverage.

* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237386 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoMarking Multiple tests as flakey at once
tsavell@apple.com [Wed, 24 Oct 2018 16:33:19 +0000 (16:33 +0000)]
Marking Multiple tests as flakey at once
https://bugs.webkit.org/show_bug.cgi?id=190847

Unreviewed Test Gardening

* platform/ios-wk2/TestExpectations:
* platform/mac-wk1/TestExpectations:
* platform/mac-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237385 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoREGRESSION (r237257): [iOS] Crashes in com.apple.WebKit: WebKit::RemoteScrollingCoord...
cdumez@apple.com [Wed, 24 Oct 2018 15:38:26 +0000 (15:38 +0000)]
REGRESSION (r237257): [iOS] Crashes in com.apple.WebKit: WebKit::RemoteScrollingCoordinator::scheduleTreeStateCommit
https://bugs.webkit.org/show_bug.cgi?id=190804

Reviewed by Alex Christensen.

Fix null-dereference of the drawing area in RemoteScrollingCoordinator::scheduleTreeStateCommit.
With process-swap-on-navigation enabled, suspended pages' drawing area gets torn down.

* WebProcess/WebPage/RemoteLayerTree/RemoteScrollingCoordinator.mm:
(WebKit::RemoteScrollingCoordinator::scheduleTreeStateCommit):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237384 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[GTK] TestSSL failing because of missing libsoup patch
csaavedra@igalia.com [Wed, 24 Oct 2018 14:32:43 +0000 (14:32 +0000)]
[GTK] TestSSL failing because of missing libsoup patch
https://bugs.webkit.org/show_bug.cgi?id=190869

Reviewed by Michael Catanzaro.

* gtk/jhbuild.modules: Add the patch below to fix the API test TestSSL
* gtk/patches/libsoup-soup-socket-fix-critical-warning-when-the-peer-certi.patch: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237383 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[Win] Add function call name information to stack traces
chris.reid@sony.com [Wed, 24 Oct 2018 14:27:13 +0000 (14:27 +0000)]
[Win] Add function call name information to stack traces
https://bugs.webkit.org/show_bug.cgi?id=190761

Reviewed by Fujii Hironori.

Add symbol information to stack traces using dbghelp.dll
This library will use symbols files from these sources:
    - The current working directory
    - The directory containing the application's executable
    - _NT_SYMBOL_PATH environment variable
    - _NT_ALTERNATE_SYMBOL_PATH environment variable

This functionality is currently only enabled in debug mode since dbghelp is not threadsafe.
The DbgHelper class attempts to synchronize calls to dbghelp.dll but external code
can still attempt to call the library at the same time as WebKit.

* wtf/CMakeLists.txt:
* wtf/PlatformWin.cmake:
* wtf/StackTrace.cpp:
* wtf/win/DbgHelperWin.cpp: Added.
* wtf/win/DbgHelperWin.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237382 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[macOS Debug WK2] Layout Test http/tests/ssl/applepay/ApplePayShippingAddressChangeEv...
aestes@apple.com [Wed, 24 Oct 2018 14:15:15 +0000 (14:15 +0000)]
[macOS Debug WK2] Layout Test http/tests/ssl/applepay/ApplePayShippingAddressChangeEventErrorsV3.https.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=190650
<rdar://problem/45341914>

Reviewed by Alex Christensen.

Source/WebCore:

In computeErrors, we intend to ignore errors when computing paymentMethodErrors. But we
weren't clearing exceptions generated from calling convert(), so they would end up being
logged as unhandled promise rejections.

Changed ApplePayPaymentHandler::computeErrors() to use a CatchScope to clear exceptions when
decoding paymentMethodErrors.

Also changed ApplePayShippingAddressChangeEventErrorsV3.https.html to catch promise
rejections from calling PaymentRequest.abort().

Covered by existing test.

* Modules/applepay/paymentrequest/ApplePayPaymentHandler.cpp:
(WebCore::ApplePayPaymentHandler::computeErrors const):

LayoutTests:

* http/tests/resources/payment-request.js:
(updateDetailsOnShippingAddressChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237381 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[WPE][GTK] Cleanups to the certificate decoder
csaavedra@igalia.com [Wed, 24 Oct 2018 13:45:26 +0000 (13:45 +0000)]
[WPE][GTK] Cleanups to the certificate decoder
https://bugs.webkit.org/show_bug.cgi?id=190867

Reviewed by Žan Doberšek.

* Shared/soup/WebCoreArgumentCodersSoup.cpp:
(IPC::ArgumentCoder<CertificateInfo>::decode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237380 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agotopPrivatelyControlledDomain is slow
antti@apple.com [Wed, 24 Oct 2018 10:20:34 +0000 (10:20 +0000)]
topPrivatelyControlledDomain is slow
https://bugs.webkit.org/show_bug.cgi?id=190792

Reviewed by Alex Christensen and Chris Dumez.

It calls into some slowish CFNetwork code and ends up showing up in profiles.

* platform/mac/PublicSuffixMac.mm:
(WebCore::topPrivatelyControlledDomain):

Add a cache that avoids calls into frameworks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237379 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUse WeakPtr and GenericTaskQueue within ObjC classes used by MediaPlayerPrivateAVFoun...
jer.noble@apple.com [Wed, 24 Oct 2018 10:03:31 +0000 (10:03 +0000)]
Use WeakPtr and GenericTaskQueue within ObjC classes used by MediaPlayerPrivateAVFoundationObjC
https://bugs.webkit.org/show_bug.cgi?id=190790

Reviewed by Alex Christensen.

Move towards using WeakPtr callbacks instead of raw pointers within the utility objects used by
MediaPlayerPrivateAVFoundationObjC. Additionally, accessing WeakPtr off the thread which created
the pointer is not allowed, so use a GenericTaskQueue to schedule callbacks instead. Make
GenericTaskQueue<Timer> thread-safe by locking around access to m_pendingTasks, and by making
incrementing the pending task count atomic.

* platform/GenericTaskQueue.cpp:
(WebCore::TaskDispatcher<Timer>::postTask):
(WebCore::TaskDispatcher<Timer>::sharedLock):
(WebCore::TaskDispatcher<Timer>::sharedTimerFired):
(WebCore::TaskDispatcher<Timer>::dispatchOneTask):
(WebCore::TaskDispatcher<Timer>::pendingDispatchers):
* platform/GenericTaskQueue.h:
* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.h:
* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::MediaPlayerPrivateAVFoundationObjC):
(WebCore::MediaPlayerPrivateAVFoundationObjC::~MediaPlayerPrivateAVFoundationObjC):
(-[WebCoreAVFMovieObserver initWithPlayer:]):
(-[WebCoreAVFMovieObserver disconnect]):
(-[WebCoreAVFMovieObserver metadataLoaded]):
(-[WebCoreAVFMovieObserver didEnd:]):
(-[WebCoreAVFMovieObserver observeValueForKeyPath:ofObject:change:context:]):
(-[WebCoreAVFMovieObserver legibleOutput:didOutputAttributedStrings:nativeSampleBuffers:forItemTime:]):
(-[WebCoreAVFMovieObserver outputSequenceWasFlushed:]):
(-[WebCoreAVFLoaderDelegate initWithPlayer:]):
(-[WebCoreAVFLoaderDelegate resourceLoader:shouldWaitForLoadingOfRequestedResource:]):
(-[WebCoreAVFLoaderDelegate resourceLoader:didCancelLoadingRequest:]):
(-[WebCoreAVFPullDelegate initWithPlayer:]):
(-[WebCoreAVFPullDelegate outputMediaDataWillChange:]):
(-[WebCoreAVFMovieObserver initWithCallback:]): Deleted.
(-[WebCoreAVFLoaderDelegate initWithCallback:]): Deleted.
(-[WebCoreAVFLoaderDelegate setCallback:]): Deleted.
(-[WebCoreAVFPullDelegate initWithCallback:]): Deleted.
(-[WebCoreAVFPullDelegate setCallback:]): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237378 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[WPE][GTK] Cleanups to the certificate encoder
csaavedra@igalia.com [Wed, 24 Oct 2018 09:10:24 +0000 (09:10 +0000)]
[WPE][GTK] Cleanups to the certificate encoder
https://bugs.webkit.org/show_bug.cgi?id=190865

Reviewed by Žan Doberšek.

* Shared/soup/WebCoreArgumentCodersSoup.cpp:
(IPC::ArgumentCoder<CertificateInfo>::encode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237377 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoTextTrack cues should be updated more often than every 250ms.
jer.noble@apple.com [Wed, 24 Oct 2018 09:05:05 +0000 (09:05 +0000)]
TextTrack cues should be updated more often than every 250ms.
https://bugs.webkit.org/show_bug.cgi?id=190827

Reviewed by Eric Carlson.

Source/WebCore:

Test: media/track/track-cue-timing.html

TextTracks cues are recalculated on the playback timer, which fires at least every 250ms.
In addition to this timer, add a method to MediaPlayer to provide a task which will be
performed at a particular media time, and use this new method to request cues be updated
at the next interesting media time. The next interesting time would be either when the
soonest current cue will end, or when the next non-current cue will start, whichever is
earlier.

(Determining the "next non-current cue" requires new API on PODIntervalTree, as that class
does not have iterators per-se.)

* html/HTMLMediaElement.cpp:
(WebCore::compareCueIntervalEndTime):
(WebCore::HTMLMediaElement::updateActiveTextTrackCues):
* platform/PODIntervalTree.h:
* platform/graphics/MediaPlayer.cpp:
(WebCore::MediaPlayer::performTaskAtMediaTime):
* platform/graphics/MediaPlayer.h:
* platform/graphics/MediaPlayerPrivate.h:
(WebCore::MediaPlayerPrivateInterface::performTaskAtMediaTime):
* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.h:
* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::performTaskAtMediaTime):
* platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaSourceAVFObjC.h:
* platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaSourceAVFObjC.mm:
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::performTaskAtMediaTime):

LayoutTests:

Add a LayoutTest which plays back 6s of captions, each 50 ms in duration, and uses
when the "enter" and "exit" were fired (in media time) to check whether they were missed
or not. The test succeeds if fewer than 50 of the 120 cues were missed.

* media/track/track-cue-missing-expected.txt: Added.
* media/track/track-cue-missing.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237376 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[Win] Assertion fails while destructing local static AtomicString of FontCache::lastR...
Hironori.Fujii@sony.com [Wed, 24 Oct 2018 04:58:27 +0000 (04:58 +0000)]
[Win] Assertion fails while destructing local static AtomicString of FontCache::lastResortFallbackFont
https://bugs.webkit.org/show_bug.cgi?id=190748

Reviewed by Myles C. Maxfield.

FontCache::lastResortFallbackFont had local static AtomicString
variables which are not capsuled by NeverDestroyed. Hence, those
variables are destructed on IPC thread when WebProcess exits.

Changed those AtomicString to NeverDestroyed<AtomicString> not to
be destructed on exit.

This change can't be tested automatically because WebKitTestRunner
doesn't support Windows port yet.

* platform/graphics/win/FontCacheWin.cpp:
(WebCore::FontCache::lastResortFallbackFont): Changed
fallbackFonts's type from AtomicString array to
NeverDestroyed<AtomicString> array.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237375 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[WebGPU] Rename old GPU* implementation files to GPULegacy* for WebMetal
justin_fan@apple.com [Tue, 23 Oct 2018 22:05:27 +0000 (22:05 +0000)]
[WebGPU] Rename old GPU* implementation files to GPULegacy* for WebMetal
https://bugs.webkit.org/show_bug.cgi?id=190817

Reviewed by Dean Jackson.

Source/WebCore:

No new tests. No change in behavior.

* [Large list of refactored classes and renamed files]: GPU* is now GPULegacy*

Tools:

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebCore/mac/GPULegacyBuffer.mm: Renamed from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUBuffer.mm.
(TestWebKitAPI::TEST_F):
* TestWebKitAPI/Tests/WebCore/mac/GPULegacyCommandQueue.mm: Copied from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUFunction.mm.
(TestWebKitAPI::TEST_F):
* TestWebKitAPI/Tests/WebCore/mac/GPULegacyDevice.mm: Renamed from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUDevice.mm.
(TestWebKitAPI::TEST_F):
* TestWebKitAPI/Tests/WebCore/mac/GPULegacyFunction.mm: Renamed from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUFunction.mm.
(TestWebKitAPI::TEST_F):
* TestWebKitAPI/Tests/WebCore/mac/GPULegacyLibrary.mm: Renamed from Tools/TestWebKitAPI/Tests/WebCore/mac/GPULibrary.mm.
(TestWebKitAPI::TEST_F):
* TestWebKitAPI/Tests/WebCore/mac/GPULegacyTest.h: Renamed from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUTest.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237368 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUnreviewed, rolling out r237261.
ryanhaddad@apple.com [Tue, 23 Oct 2018 21:07:14 +0000 (21:07 +0000)]
Unreviewed, rolling out r237261.

The layout test for this change crashes under GuardMalloc.

Reverted changeset:

"Handle MDNS resolution of candidates through libwebrtc
directly"
https://bugs.webkit.org/show_bug.cgi?id=190681
https://trac.webkit.org/changeset/237261

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237367 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUnreviewed, rolling out r237357.
tsavell@apple.com [Tue, 23 Oct 2018 20:53:20 +0000 (20:53 +0000)]
Unreviewed, rolling out r237357.

API test is now failing on all platforms.

Reverted changeset:

"topPrivatelyControlledDomain is slow"
https://bugs.webkit.org/show_bug.cgi?id=190792
https://trac.webkit.org/changeset/237357

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237366 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoFix false positive leaks when using custom -init methods that don't start with -init
ddkilzer@apple.com [Tue, 23 Oct 2018 20:36:37 +0000 (20:36 +0000)]
Fix false positive leaks when using custom -init methods that don't start with -init
<https://webkit.org/b/190818>
<rdar://problem/45476042>

Reviewed by Dan Bernstein.

Source/WebKit:

* UIProcess/Cocoa/WKWebViewContentProvider.h:
(-[WKWebViewContentProvider web_initWithFrame:webView:mimeType:]):
Add attribute to make this behave like an -init method.

Source/WebKitLegacy/mac:

* Misc/WebKitErrors.m: Move `descriptions` static global so that
it is accessible by both categories.
(-[NSError _webkit_initWithDomain:code:URL:]): Declare new
NSError (WebKitInternal) category.  Add attribute to make this
behave like an -init method.
(-[NSError _webkit_initWithDomain:code:URL:]): Move into new
NSError (WebKitInternal) category.  Modernize method.  Replace
constant NSString with `NSURLErrorFailingURLStringErrorKey`.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237365 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUnreviewed, rolling out r237350.
tsavell@apple.com [Tue, 23 Oct 2018 20:02:52 +0000 (20:02 +0000)]
Unreviewed, rolling out r237350.

Caused several Crashes cross multiple tests and platforms.

Reverted changeset:

"Use WeakPtr and GenericTaskQueue within ObjC classes used by
MediaPlayerPrivateAVFoundationObjC"
https://bugs.webkit.org/show_bug.cgi?id=190790
https://trac.webkit.org/changeset/237350

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237364 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoWeb Inspector: Remove unused css-rules pretty printing test expectation
commit-queue@webkit.org [Tue, 23 Oct 2018 19:44:46 +0000 (19:44 +0000)]
Web Inspector: Remove unused css-rules pretty printing test expectation
https://bugs.webkit.org/show_bug.cgi?id=190834

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2018-10-23
Reviewed by Matt Baker.

* inspector/codemirror/prettyprinting-css-rules-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237363 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoReally prevent PLATFORM(IOS) from being used
ap@apple.com [Tue, 23 Oct 2018 18:30:40 +0000 (18:30 +0000)]
Really prevent PLATFORM(IOS) from being used
https://bugs.webkit.org/show_bug.cgi?id=190813

Reviewed by Tim Horton.

* wtf/Platform.h: This relies on -Wundef, which we have enabled.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237361 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUnreviewed, rolling out r237280.
ryanhaddad@apple.com [Tue, 23 Oct 2018 18:17:11 +0000 (18:17 +0000)]
Unreviewed, rolling out r237280.

Causes fast/box-shadow/box-shadow-with-zero-radius.html to
fail on iOS Simulator.

Reverted changeset:

"[CG] Adopt CG SPI for non-even cornered rounded rects"
https://bugs.webkit.org/show_bug.cgi?id=190155
https://trac.webkit.org/changeset/237280

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237359 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agotopPrivatelyControlledDomain is slow
antti@apple.com [Tue, 23 Oct 2018 18:02:32 +0000 (18:02 +0000)]
topPrivatelyControlledDomain is slow
https://bugs.webkit.org/show_bug.cgi?id=190792

Reviewed by Alex Christensen.

It calls into some slowish CFNetwork code and ends up showing up in profiles.

* platform/mac/PublicSuffixMac.mm:
(WebCore::topPrivatelyControlledDomain):

Add a cache that avoids calls into frameworks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237357 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[macOS Debug WK1] Layout Test inspector/codemirror/prettyprinting-css.html is a flaky...
ryanhaddad@apple.com [Tue, 23 Oct 2018 17:54:57 +0000 (17:54 +0000)]
[macOS Debug WK1] Layout Test inspector/codemirror/prettyprinting-css.html is a flaky timeout
https://bugs.webkit.org/show_bug.cgi?id=190833

Unreviewed test gardening.

* platform/mac-wk1/TestExpectations: Mark test as slow.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237356 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[PSON] Add support for cross-site client-side redirects
cdumez@apple.com [Tue, 23 Oct 2018 17:54:34 +0000 (17:54 +0000)]
[PSON] Add support for cross-site client-side redirects
https://bugs.webkit.org/show_bug.cgi?id=190806
<rdar://problem/45047344>

Reviewed by Geoffrey Garen.

Source/WebCore:

Add support for cross-site client-side redirects so that it swaps process and so that the back
forward list looks as expected. To support this, the following changes had to be done:
- The NavigationAction now provides additional information so that the WebProcess can communicate
  things about the client-side redirect to the UIProcess: lockHistory / lockBackForwardList and
  clientRedirectSourceForHistory.
- If the UIProcess decides to process-swap on a client-side redirect, we pass the client-side
  redirect information to the new WebContent process via LoadRequest struct. WebCore then takes
  care of setting things up using this information so that it recognizes that it is continuing
  a load that is a client side redirect.
- We also need to pass the current BackForwardListItem / HistoryItem to the new WebContent
  process so that the new process can truly lock history and keep updating the *current*
  HistoryItem, instead of creating a new HistoryItem.
- After a process swap, when we re-construct the WebFrameProxy for the main frame in the new
  process, we now set the frame's URL in the UIProcess to the URL it had before we swapped.
  Clients such as Safari, rely on the main frame's URL being the expected value (the last
  committed load URL) until the next load is committed when receiving didPerformRedirect
  calls. Because we are destroying the main frame on process-swapping, we were losing the
  last committed URL and Safari would hit assertions.

With this model, the willPerformClientRedirect IPC is still sent from the previous WebProcess
and the didPerformClientRedirect IPC is now sent by the new WebProcess. No change should be
observable from the client's point of view.

* loader/FrameLoadRequest.h:
(WebCore::FrameLoadRequest::setLockHistory):
(WebCore::FrameLoadRequest::setlockBackForwardList):
(WebCore::FrameLoadRequest::clientRedirectSourceForHistory const):
(WebCore::FrameLoadRequest::setClientRedirectSourceForHistory):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::loadURL):
(WebCore::FrameLoader::load):
* loader/HistoryController.cpp:
(WebCore::HistoryController::updateForRedirectWithLockedBackForwardList):
* loader/HistoryController.h:
* loader/NavigationAction.h:
(WebCore::NavigationAction::lockHistory const):
(WebCore::NavigationAction::setLockHistory):
(WebCore::NavigationAction::lockBackForwardList const):
(WebCore::NavigationAction::setLockBackForwardList):

Source/WebKit:

* Shared/LoadParameters.cpp:
(WebKit::LoadParameters::encode const):
(WebKit::LoadParameters::decode):
* Shared/LoadParameters.h:
* Shared/NavigationActionData.cpp:
(WebKit::NavigationActionData::encode const):
(WebKit::NavigationActionData::decode):
* Shared/NavigationActionData.h:
* Shared/WebBackForwardListItem.cpp:
(WebKit::WebBackForwardListItem::setSuspendedPage):
* UIProcess/API/APINavigation.h:
(API::Navigation::setLockHistory):
(API::Navigation::lockHistory const):
(API::Navigation::setLockBackForwardList):
(API::Navigation::lockBackForwardList const):
(API::Navigation::setClientRedirectSourceForHistory):
(API::Navigation::clientRedirectSourceForHistory const):
* UIProcess/API/APINavigationClient.h:
(API::NavigationClient::willPerformClientRedirect):
(API::NavigationClient::didPerformClientRedirect):
* UIProcess/API/Cocoa/WKNavigationDelegatePrivate.h:
* UIProcess/Cocoa/NavigationState.h:
* UIProcess/Cocoa/NavigationState.mm:
(WebKit::NavigationState::setNavigationDelegate):
(WebKit::NavigationState::NavigationClient::didPerformClientRedirect):
* UIProcess/FrameLoadState.h:
(WebKit::FrameLoadState::setURL):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::loadRequestWithNavigation):
(WebKit::WebPageProxy::continueNavigationInNewProcess):
(WebKit::WebPageProxy::decidePolicyForNavigationAction):
(WebKit::WebPageProxy::didPerformClientRedirect):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::loadRequest):
(WebKit::WebPage::setCurrentHistoryItemForReattach):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:

Tools:

Add API test coverage.

* TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
(-[PSONNavigationDelegate _webView:willPerformClientRedirectToURL:delay:]):
(-[PSONNavigationDelegate _webView:didPerformClientRedirectFromURL:toURL:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237355 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[ MacOS Debug WK1 ] Layout Test fast/forms/state-restore-to-non-edited-controls.html...
tsavell@apple.com [Tue, 23 Oct 2018 17:26:55 +0000 (17:26 +0000)]
[ MacOS Debug WK1 ] Layout Test fast/forms/state-restore-to-non-edited-controls.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=190832

Unreviewed Test Gardening.

* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237354 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoLayout Test imported/w3c/web-platform-tests/html/semantics/embedded-content/media...
ryanhaddad@apple.com [Tue, 23 Oct 2018 17:22:43 +0000 (17:22 +0000)]
Layout Test imported/w3c/web-platform-tests/html/semantics/embedded-content/media-elements/ready-states/autoplay-with-slow-text-tracks.html is flaky.
https://bugs.webkit.org/show_bug.cgi?id=176929

Unreviewed test gardening.

* platform/mac/TestExpectations: Mark test as flaky.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237353 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoREGRESSION (r236501?): Layout Test compositing/geometry/limit-layer-bounds-opacity...
ryanhaddad@apple.com [Tue, 23 Oct 2018 16:58:40 +0000 (16:58 +0000)]
REGRESSION (r236501?): Layout Test compositing/geometry/limit-layer-bounds-opacity-transition.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=190210

Unreviewed test gardening.

* platform/mac/TestExpectations: Mark test as flaky.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237352 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months ago[WPE][GTK] Pass full certificate chain in CertificateInfo coder
csaavedra@igalia.com [Tue, 23 Oct 2018 15:52:11 +0000 (15:52 +0000)]
[WPE][GTK] Pass full certificate chain in CertificateInfo coder
https://bugs.webkit.org/show_bug.cgi?id=190789

Reviewed by Michael Catanzaro.

When the network process serializes certificate data to other
processes through the argument coders, the certificate chain, if
present, is lost. In practice this means that applications using
the public API to process certificate info have no details on the
certificate chain, other than the very basics included in the
certificate. Serialize the entire chain if available in the
certificate.

Source/WebKit:

* Shared/soup/WebCoreArgumentCodersSoup.cpp:
(IPC::ArgumentCoder<CertificateInfo>::encode): Encode the
certificate chain if present.
(IPC::ArgumentCoder<CertificateInfo>::decode): Decode the
entire certificate chain and rebuild it.

Tools:

* TestWebKitAPI/Tests/WebKitGLib/TestSSL.cpp:
(testSSL): Test that the self-signed certificate has no bogus
issuer certificate.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237351 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 months agoUse WeakPtr and GenericTaskQueue within ObjC classes used by MediaPlayerPrivateAVFoun...
jer.noble@apple.com [Tue, 23 Oct 2018 14:40:58 +0000 (14:40 +0000)]
Use WeakPtr and GenericTaskQueue within ObjC classes used by MediaPlayerPrivateAVFoundationObjC
https://bugs.webkit.org/show_bug.cgi?id=190790

Reviewed by Alex Christensen.

Move towards using WeakPtr callbacks instead of raw pointers within the utility objects used by
MediaPlayerPrivateAVFoundationObjC. Additionally, accessing WeakPtr off the thread which created
the pointer is not allowed, so use a GenericTaskQueue to schedule callbacks instead. Make
GenericTaskQueue<Timer> thread-safe by locking around access to m_pendingTasks, and by making
incrementing the pending task count atomic.

* platform/GenericTaskQueue.cpp:
(WebCore::TaskDispatcher<Timer>::postTask):
(WebCore::TaskDispatcher<Timer>::sharedLock):
(WebCore::TaskDispatcher<Timer>::sharedTimerFired):
(WebCore::TaskDispatcher<Timer>::dispatchOneTask):
(WebCore::TaskDispatcher<Timer>::pendingDispatchers):
* platform/GenericTaskQueue.h:
* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.h:
* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::MediaPlayerPrivateAVFoundationObjC):
(WebCore::MediaPlayerPrivateAVFoundationObjC::~MediaPlayerPrivateAVFoundationObjC):
(-[WebCoreAVFMovieObserver initWithPlayer:]):
(-[WebCoreAVFMovieObserver disconnect]):
(-[WebCoreAVFMovieObserver metadataLoaded]):
(-[WebCoreAVFMovieObserver didEnd:]):
(-[WebCoreAVFMovieObserver observeValueForKeyPath:ofObject:change:context:]):
(-[WebCoreAVFMovieObserver legibleOutput:didOutputAttributedStrings:nativeSampleBuffers:forItemTime:]):
(-[WebCoreAVFMovieObserver outputSequenceWasFlushed:]):
(-[WebCoreAVFLoaderDelegate initWithPlayer:]):
(-[WebCoreAVFLoaderDelegate resourceLoader:shouldWaitForLoadingOfRequestedResource:]):
(-[WebCoreAVFLoaderDelegate resourceLoader:didCancelLoadingRequest:]):
(-[WebCoreAVFPullDelegate initWithPlayer:]):
(-[WebCoreAVFPullDelegate outputMediaDataWillChange:]):
(-[WebCoreAVFMovieObserver initWithCallback:]): Deleted.
(-[WebCoreAVFLoaderDelegate initWithCallback:]): Deleted.
(-[WebCoreAVFLoaderDelegate setCallback:]): Deleted.
(-[WebCoreAVFPullDelegate initWithCallback:]): Deleted.
(-[WebCoreAVFPullDelegate setCallback:]): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237350 268f45cc-cd09-0410-ab3c-d52691b4dbfc