Timeline



Oct 26, 2014:

11:51 PM Changeset in webkit [175215] by Carlos Garcia Campos
  • 2 edits in releases/WebKitGTK/webkit-2.6

Merge r173999 - [GTK] Enable CSS_IMAGE_SET in production builds
​https://bugs.webkit.org/show_bug.cgi?id=137142

Reviewed by Alejandro G. Castro.

This is required by the inspector to show some of the icons that
has a HiDPI variant.

  • Source/cmake/OptionsGTK.cmake:
11:29 PM Changeset in webkit [175214] by bshafiei@apple.com
  • 5 edits in branches/safari-600.1.4.12-branch/Source

Versioning.

9:28 PM Changeset in webkit [175213] by benjamin@webkit.org
  • 9 edits
    2 adds in trunk

Images with usemap should not have a link state
​https://bugs.webkit.org/show_bug.cgi?id=138071

Reviewed by Andreas Kling.

Source/WebCore:

Image tags were incorrectly getting a link state if a usemap is defined.
The <area> of the usemap is supposed to get the link state, but never the image.

Unfortunately, the Accessibility code was relying on the wrong behavior
to set the ImageMapRole, most of the patch is updating that code to get
the correct role without the link state.

Test: fast/selectors/images-with-usemap-should-not-match-link.html

The accessibility code is covered by a few tests in accessibility/.
The test accessibility/aria-used-on-image-maps.html ensure the <area>
elements also get the correct role.

  • dom/Node.h:

(WebCore::Node::setIsLink):
(WebCore::Node::clearIsLink): Deleted.
Unused code.

(WebCore::Node::isLink):
Move with its setter for clarity.

  • html/HTMLImageElement.cpp:

(WebCore::HTMLImageElement::parseAttribute):
This is the bug.

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::determineAccessibilityRole):
Here the branches for the role ImageMap where behind isLink() which would
never be valid after the fix.

My fix here was to use the same definition previously set by HTMLImageElement::parseAttribute(),
any usemap attribute qualifies the node for ImageMapRole.

(WebCore::AccessibilityRenderObject::anchorElement):
(WebCore::AccessibilityRenderObject::linkedUIElements):
(WebCore::AccessibilityRenderObject::url):

  • accessibility/AccessibilityObject.h:

(WebCore::AccessibilityObject::isAnchor): Deleted.
Something funny I discovered is the Accessibility code worked around the bug
by defining isAnchor() that exclude images from isLink().

Since isLink() will no longer be true for images, the hack can be removed.

  • accessibility/AccessibilityNodeObject.cpp:

(WebCore::AccessibilityNodeObject::anchorElement):
(WebCore::AccessibilityNodeObject::isAnchor): Deleted.

  • accessibility/AccessibilityNodeObject.h:
  • accessibility/mac/WebAccessibilityObjectWrapperMac.mm:

(-[WebAccessibilityObjectWrapper accessibilityAttributeNames]):

LayoutTests:

  • fast/selectors/images-with-usemap-should-not-match-link-expected.txt: Added.
  • fast/selectors/images-with-usemap-should-not-match-link.html: Added.
9:20 PM Changeset in webkit [175212] by benjamin@webkit.org
  • 11 edits in trunk/Source/WebCore

Replace Element::didAffectSelector() by setNeedsStyleRecalc()
​https://bugs.webkit.org/show_bug.cgi?id=138080

Reviewed by Andreas Kling.

Element::didAffectSelector() was probably an optimization at some point,
with the current code is does nothing more than setNeedsStyleRecalc().

  • dom/Document.cpp:

(WebCore::Document::setCSSTarget):

  • dom/Element.cpp:

(WebCore::Element::didAffectSelector): Deleted.

  • dom/Element.h:
  • html/HTMLAnchorElement.cpp:

(WebCore::HTMLAnchorElement::parseAttribute):

  • html/HTMLFormControlElement.cpp:

(WebCore::HTMLFormControlElement::disabledStateChanged):

  • html/HTMLInputElement.cpp:

(WebCore::HTMLInputElement::setChecked):
(WebCore::HTMLInputElement::setIndeterminate):

  • html/HTMLLinkElement.cpp:

(WebCore::HTMLLinkElement::parseAttribute):

  • html/HTMLOptGroupElement.cpp:

(WebCore::HTMLOptGroupElement::parseAttribute):

  • html/HTMLOptionElement.cpp:

(WebCore::HTMLOptionElement::parseAttribute):
(WebCore::HTMLOptionElement::setSelectedState):

  • html/HTMLProgressElement.cpp:

(WebCore::HTMLProgressElement::didElementStateChange):

3:25 PM Changeset in webkit [175211] by akling@apple.com
  • 10 edits in trunk/Source/WebCore

Move some RenderObject member functions to RenderElement.
<​https://webkit.org/b/138085>

Reviewed by Chris Dumez.

Move these two functions to RenderElement:

  • handleDynamicFloatPositionChange()
  • removeAnonymousWrappersForInlinesIfNecessary()

Also make childBecameNonInline() take a RenderElement& since
that's all we ever pass to it.

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::childBecameNonInline):

  • rendering/RenderBlock.h:
  • rendering/RenderBoxModelObject.h:

(WebCore::RenderBoxModelObject::childBecameNonInline):

  • rendering/RenderElement.cpp:

(WebCore::RenderElement::handleDynamicFloatPositionChange):
(WebCore::RenderElement::removeAnonymousWrappersForInlinesIfNecessary):

  • rendering/RenderElement.h:
  • rendering/RenderInline.cpp:

(WebCore::RenderInline::childBecameNonInline):

  • rendering/RenderInline.h:
  • rendering/RenderObject.cpp:

(WebCore::RenderObject::handleDynamicFloatPositionChange): Deleted.
(WebCore::RenderObject::removeAnonymousWrappersForInlinesIfNecessary): Deleted.

  • rendering/RenderObject.h:
1:20 PM Changeset in webkit [175210] by akling@apple.com
  • 7 edits in trunk/Source/WebCore

Move continuation teardown from subclasses to RenderBoxModelObject.
<​https://webkit.org/b/138081>

Reviewed by Antti Koivisto.

Let RenderBoxModelObject::willBeDestroyed() tear down any continuation
instead of having every subclass do this themselves.

Also added a RenderElement bit tracking whether the renderer has a
continuation. This avoids a hash lookup every time we destroy a
RenderBoxModelObject that didn't have a continuation.

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::willBeDestroyed):

  • rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::willBeDestroyed):

  • rendering/RenderBoxModelObject.cpp:

(WebCore::continuationMap):
(WebCore::RenderBoxModelObject::willBeDestroyed):
(WebCore::RenderBoxModelObject::continuation):
(WebCore::RenderBoxModelObject::setContinuation):

  • rendering/RenderElement.cpp:

(WebCore::RenderElement::RenderElement):

  • rendering/RenderElement.h:

(WebCore::RenderElement::setHasContinuation):
(WebCore::RenderElement::hasContinuation):

  • rendering/RenderInline.cpp:

(WebCore::RenderInline::willBeDestroyed):

Oct 25, 2014:

10:47 PM Changeset in webkit [175209] by benjamin@webkit.org
  • 2 edits in trunk/Source/WebCore

Remove a useless variable from SelectorCodeGenerator::generateElementMatchesSelectorList()
​https://bugs.webkit.org/show_bug.cgi?id=138076

Reviewed by Alexey Proskuryakov.

  • cssjit/SelectorCompiler.cpp:

(WebCore::SelectorCompiler::SelectorCodeGenerator::generateElementMatchesSelectorList):
The variable elementAddressRegisterIndex was just needed for debugging, it is useless now.

9:47 PM WebKitGTK/KeepingTheTreeGreen edited by yoon@igalia.com
(diff)
9:44 PM Changeset in webkit [175208] by yoon@igalia.com
  • 2 edits in trunk/LayoutTests

[GTK] Unreviewed GTK gardening after r174685.

  • platform/gtk/TestExpectations: Report new failures and update expectations.
9:12 PM Changeset in webkit [175207] by commit-queue@webkit.org
  • 2 edits in trunk/Tools

Fix link to Arch Linux Wiki
​https://bugs.webkit.org/show_bug.cgi?id=138010

Patch by Raniere Silva <​raniere@ime.unicamp.br> on 2014-10-25
Reviewed by Benjamin Poulain.

  • gtk/install-dependencies:
8:52 PM Changeset in webkit [175206] by Brian Burg
  • 4 edits in trunk/Source

Web Inspector: timelines should not count time elapsed while paused in the debugger
​https://bugs.webkit.org/show_bug.cgi?id=136351

Unreviewed, follow-up fix after r175203. The debugger agent should not assume
that the inspector environment's stopwatch has already been started.

Source/JavaScriptCore:

  • inspector/agents/InspectorDebuggerAgent.cpp:

(Inspector::InspectorDebuggerAgent::didPause): Check if the stopwatch isActive() before stopping.

Source/WTF:

  • wtf/Stopwatch.h:

(WTF::Stopwatch::isActive): Added. Allow peeking at the stopwatch state.

8:25 PM Changeset in webkit [175205] by ap@apple.com
  • 7 edits
    6 copies
    6 adds in trunk/LayoutTests

Land Yosemite results for six editing/selection/vertical* tests.

These failures used to be mostly hidden because of bug 138070, now they are visible.
The change is due to different fonts.

  • platform/mac-mavericks/editing/selection/vertical-lr-ltr-extend-line-backward-br-expected.txt: Copied from LayoutTests/platform/mac/editing/selection/vertical-lr-ltr-extend-line-backward-br-expected.txt.
  • platform/mac-mavericks/editing/selection/vertical-lr-ltr-extend-line-forward-br-expected.txt: Copied from LayoutTests/platform/mac/editing/selection/vertical-lr-ltr-extend-line-forward-br-expected.txt.
  • platform/mac-mavericks/editing/selection/vertical-rl-ltr-extend-line-backward-br-expected.txt: Copied from LayoutTests/platform/mac/editing/selection/vertical-rl-ltr-extend-line-backward-br-expected.txt.
  • platform/mac-mavericks/editing/selection/vertical-rl-ltr-extend-line-backward-p-expected.txt: Copied from LayoutTests/platform/mac/editing/selection/vertical-rl-ltr-extend-line-backward-p-expected.txt.
  • platform/mac-mavericks/editing/selection/vertical-rl-ltr-extend-line-forward-br-expected.txt: Copied from LayoutTests/platform/mac/editing/selection/vertical-rl-ltr-extend-line-forward-br-expected.txt.
  • platform/mac-mavericks/editing/selection/vertical-rl-ltr-extend-line-forward-p-expected.txt: Copied from LayoutTests/platform/mac/editing/selection/vertical-rl-ltr-extend-line-forward-p-expected.txt.
  • platform/mac/editing/selection/vertical-lr-ltr-extend-line-backward-br-expected.png: Added.
  • platform/mac/editing/selection/vertical-lr-ltr-extend-line-backward-br-expected.txt:
  • platform/mac/editing/selection/vertical-lr-ltr-extend-line-forward-br-expected.png: Added.
  • platform/mac/editing/selection/vertical-lr-ltr-extend-line-forward-br-expected.txt:
  • platform/mac/editing/selection/vertical-rl-ltr-extend-line-backward-br-expected.png: Added.
  • platform/mac/editing/selection/vertical-rl-ltr-extend-line-backward-br-expected.txt:
  • platform/mac/editing/selection/vertical-rl-ltr-extend-line-backward-p-expected.png: Added.
  • platform/mac/editing/selection/vertical-rl-ltr-extend-line-backward-p-expected.txt:
  • platform/mac/editing/selection/vertical-rl-ltr-extend-line-forward-br-expected.png: Added.
  • platform/mac/editing/selection/vertical-rl-ltr-extend-line-forward-br-expected.txt:
  • platform/mac/editing/selection/vertical-rl-ltr-extend-line-forward-p-expected.png: Added.
  • platform/mac/editing/selection/vertical-rl-ltr-extend-line-forward-p-expected.txt:
4:29 PM Changeset in webkit [175204] by ap@apple.com
  • 5 edits in trunk/Tools

Test regressions are not detected when image result is missing
​https://bugs.webkit.org/show_bug.cgi?id=138070

Reviewed by Simon Fraser.

  • Scripts/webkitpy/layout_tests/models/test_run_results.py:
  • Scripts/webkitpy/layout_tests/views/buildbot_results.py:

Count these as regressions, not as flaky tests.

  • Scripts/webkitpy/port/test.py:
  • Scripts/webkitpy/layout_tests/run_webkit_tests_integrationtest.py:

Test it.

3:05 PM Changeset in webkit [175203] by Brian Burg
  • 39 edits
    1 add in trunk/Source

Web Inspector: timelines should not count time elapsed while paused in the debugger
​https://bugs.webkit.org/show_bug.cgi?id=136351

Reviewed by Timothy Hatcher.

Source/JavaScriptCore:

Now that we have a stopwatch to provide pause-aware timing data, we can remove the
profiler's handling of debugger pause/continue callbacks. The debugger agent accounts
for suspended execution by pausing and resuming the stopwatch.

  • API/JSProfilerPrivate.cpp:

(JSStartProfiling): Use a fresh stopwatch when profiling from the JSC API.

  • inspector/InspectorEnvironment.h:
  • inspector/JSGlobalObjectInspectorController.cpp:

(Inspector::JSGlobalObjectInspectorController::JSGlobalObjectInspectorController):
(Inspector::JSGlobalObjectInspectorController::executionStopwatch):

  • inspector/JSGlobalObjectInspectorController.h:
  • inspector/ScriptDebugServer.cpp:

(Inspector::ScriptDebugServer::handlePause):

  • inspector/agents/InspectorDebuggerAgent.cpp:

(Inspector::InspectorDebuggerAgent::didPause):
(Inspector::InspectorDebuggerAgent::breakpointActionProbe):
(Inspector::InspectorDebuggerAgent::didContinue):

  • inspector/agents/InspectorDebuggerAgent.h:
  • profiler/LegacyProfiler.cpp:

(JSC::LegacyProfiler::profiler): Use nullptr.
(JSC::LegacyProfiler::startProfiling): Hand off a stopwatch to the profile generator.
(JSC::LegacyProfiler::stopProfiling): Use nullptr.
(JSC::LegacyProfiler::didPause): Deleted.
(JSC::LegacyProfiler::didContinue): Deleted.

  • profiler/LegacyProfiler.h:
  • profiler/Profile.cpp: The root node should always have a start time of 0.0.

(JSC::Profile::Profile):

  • profiler/ProfileGenerator.cpp: Remove debugger pause/continue callbacks and the

timestamp member that was used to track time elapsed by the debugger. Just use the
stopwatch's elapsed times to generate start/elapsed times for function calls.

(JSC::ProfileGenerator::create):
(JSC::ProfileGenerator::ProfileGenerator):
(JSC::AddParentForConsoleStartFunctor::operator()): The parent node of |console.profile|
should have a start time of 0.0, since it represents the starting node of profiling.

(JSC::ProfileGenerator::beginCallEntry):
(JSC::ProfileGenerator::endCallEntry):
(JSC::ProfileGenerator::didPause): Deleted.
(JSC::ProfileGenerator::didContinue): Deleted.

  • profiler/ProfileGenerator.h:

Source/WebCore:

To avoid counting time elapsed while the debugger is paused, timeline records should
keep track of time elapsed since the start of timeline capturing, rather than wall clock
timestamps. We can easily compute elapsed time by sharing a Stopwatch instance through the
inspector environment. The stopwatch runs with timelines and is paused with the debugger,
so subsequent time measurements will not include time elapsed while the debugger is paused.

This refactoring is safe because start and end times are only used to graph records; the
timestamp's actual value is irrelevant and is not displayed in the user interface. Date
timestamps are still included with network-related records as part of their header data.

No new tests, because we cannot reliably test timing changes induced by debugger pauses.
It is possible for records to accrue time before the debugger pauses or after it resumes.

  • inspector/InspectorCSSAgent.cpp: Remove unnecessary include.
  • inspector/InspectorController.cpp:

(WebCore::InspectorController::InspectorController):
(WebCore::InspectorController::executionStopwatch): Add a shared stopwatch.

  • inspector/InspectorController.h:
  • inspector/InspectorPageAgent.cpp:

(WebCore::InspectorPageAgent::timestamp): Redirect to the shared stopwatch.
(WebCore::InspectorPageAgent::domContentEventFired):
(WebCore::InspectorPageAgent::loadEventFired):

  • inspector/InspectorPageAgent.h:
  • inspector/InspectorResourceAgent.cpp:

(WebCore::InspectorResourceAgent::timestamp): Redirect to the shared stopwatch.
(WebCore::InspectorResourceAgent::willSendRequest):
(WebCore::InspectorResourceAgent::didReceiveResponse):
(WebCore::InspectorResourceAgent::didReceiveData):
(WebCore::InspectorResourceAgent::didFinishLoading):
(WebCore::InspectorResourceAgent::didFailLoading):
(WebCore::InspectorResourceAgent::didLoadResourceFromMemoryCache):
(WebCore::InspectorResourceAgent::willSendWebSocketHandshakeRequest):
(WebCore::InspectorResourceAgent::didReceiveWebSocketHandshakeResponse):
(WebCore::InspectorResourceAgent::didCloseWebSocket):
(WebCore::InspectorResourceAgent::didReceiveWebSocketFrame):
(WebCore::InspectorResourceAgent::didSendWebSocketFrame):
(WebCore::InspectorResourceAgent::didReceiveWebSocketFrameError):

  • inspector/InspectorResourceAgent.h:
  • inspector/InspectorTimelineAgent.cpp:

(WebCore::InspectorTimelineAgent::internalStart): Start and stop the stopwatch with timelines.
(WebCore::InspectorTimelineAgent::internalStop):
(WebCore::InspectorTimelineAgent::timestamp): Redirect to the shared stopwatch.
(WebCore::startProfiling):
(WebCore::InspectorTimelineAgent::startFromConsole):
(WebCore::InspectorTimelineAgent::willCallFunction):
(WebCore::InspectorTimelineAgent::willEvaluateScript):
(WebCore::TimelineTimeConverter::reset): Deleted.

  • inspector/InspectorTimelineAgent.h:

(WebCore::TimelineTimeConverter::TimelineTimeConverter): Deleted.
(WebCore::TimelineTimeConverter::fromMonotonicallyIncreasingTime): Deleted.
(WebCore::InspectorTimelineAgent::timeConverter): Deleted.

  • inspector/TimelineRecordFactory.cpp:
  • inspector/WorkerInspectorController.cpp:

(WebCore::WorkerInspectorController::WorkerInspectorController):
(WebCore::WorkerInspectorController::executionStopwatch): Add a shared stopwatch.

  • inspector/WorkerInspectorController.h:

Source/WebInspectorUI:

Don't update the timeline's current time when the debugger is paused.

Start and end times for timeline records are now in seconds elapsed since timeline
recording started, rather than milliseconds since the epoch. Also convert code that
tracks page/resource load timings to use elapsed times rather than timestamps.

Add a workaround to preserve compatibility with old backends. Convert legacy timestamps
in multiple agents to elapsed times.

  • UserInterface/Controllers/FrameResourceManager.js:

(WebInspector.FrameResourceManager.prototype.resourceRequestWillBeSent):
(WebInspector.FrameResourceManager.prototype.resourceRequestWasServedFromMemoryCache):
(WebInspector.FrameResourceManager.prototype.resourceRequestDidReceiveResponse):
(WebInspector.FrameResourceManager.prototype.resourceRequestDidReceiveData):
(WebInspector.FrameResourceManager.prototype.resourceRequestDidFinishLoading):
(WebInspector.FrameResourceManager.prototype.resourceRequestDidFailLoading):
(WebInspector.FrameResourceManager.prototype._addNewResourceToFrame):

  • UserInterface/Controllers/ProbeManager.js:
  • UserInterface/Controllers/TimelineManager.js:

(WebInspector.TimelineManager.prototype.computeElapsedTime): Forward to the active TimelineRecording.
(WebInspector.TimelineManager.prototype.eventRecorded.processRecord):
(WebInspector.TimelineManager.prototype.eventRecorded):
(WebInspector.TimelineManager.prototype.pageDidLoad):
(WebInspector.TimelineManager.prototype._loadNewRecording):

  • UserInterface/Models/Probe.js:

(WebInspector.ProbeSample):

  • UserInterface/Models/Resource.js:

(WebInspector.Resource.prototype.updateForRedirectResponse):
(WebInspector.Resource.prototype.updateForResponse):
(WebInspector.Resource.prototype.increaseSize):
(WebInspector.Resource.prototype.markAsFinished):
(WebInspector.Resource.prototype.markAsFailed):
(WebInspector.Resource.prototype.revertMarkAsFinished):

  • UserInterface/Models/TimelineRecording.js:

(WebInspector.TimelineRecording.prototype.computeElapsedTime):

  • UserInterface/Views/TimelineContentView.js:

(WebInspector.TimelineContentView.prototype._debuggerPaused):
(WebInspector.TimelineContentView.prototype._debuggerResumed):

Source/WTF:

  • WTF.vcxproj/WTF.vcxproj:
  • WTF.vcxproj/WTF.vcxproj.filters:
  • WTF.xcodeproj/project.pbxproj:
  • wtf/CMakeLists.txt:
  • wtf/Stopwatch.h: Added. This implements a refcounted monotonic stopwatch.

(WTF::Stopwatch::create):
(WTF::Stopwatch::Stopwatch):
(WTF::Stopwatch::reset):
(WTF::Stopwatch::start):
(WTF::Stopwatch::stop):
(WTF::Stopwatch::elapsedTime):

12:58 PM Changeset in webkit [175202] by ap@apple.com
  • 1 edit
    1 add in trunk/LayoutTests

Add Mavericks results for platform/mac/editing/spelling/delete-autocorrected-word-2.html, too.

Yosemite, Mavericks and Mountain Lion all have different results.

  • platform/mac-mavericks/platform/mac/editing/spelling/delete-autocorrected-word-2-expected.txt: Added.
12:25 PM Changeset in webkit [175201] by ap@apple.com
  • 3 edits in trunk/LayoutTests

Update Yosemite results for platform/mac/editing/spelling/delete-autocorrected-word-2.html.

Looks like this got confused during Yosemite tests landing.

This test is currently marked as failing on WK2, but it passed for me, so maybe we
can mark it as passing there too. Tracked as bug 105616.

  • platform/mac/editing/spelling/delete-autocorrected-word-2-expected.png:
  • platform/mac/editing/spelling/delete-autocorrected-word-2-expected.txt:
12:21 PM Changeset in webkit [175200] by mitz@apple.com
  • 2 edits in trunk/Source/WebCore

Fix builds using the public SDK.

  • platform/spi/cocoa/NSExtensionSPI.h:
12:10 PM Changeset in webkit [175199] by ap@apple.com
  • 2 edits
    1 copy
    2 deletes in trunk/LayoutTests

Update results for compositing/rtl/rtl-fixed-overflow-scrolled.html.

Prior to Yosemite results landing, this test was marked as failing on WK2, and its
results matched WK1. It got changed several times during the merge, so it ended
up being marked as failing on WK1, but with WK1 results landed.

Let's land WK2 results and keep testing on WK2. Also, Mavericks results were no different
from Yosemite.

  • platform/mac-mavericks/compositing/rtl/rtl-fixed-overflow-scrolled-expected.png: Removed.
  • platform/mac-mavericks/compositing/rtl/rtl-fixed-overflow-scrolled-expected.txt: Removed.
  • platform/mac/compositing/rtl/rtl-fixed-overflow-scrolled-expected.png: Copied from LayoutTests/compositing/rtl/rtl-fixed-overflow-scrolled-expected.png.
  • platform/mac/compositing/rtl/rtl-fixed-overflow-scrolled-expected.txt:
2:42 AM WebKitGTK/KeepingTheTreeGreen edited by yoon@igalia.com
(diff)
Note: See TracTimeline for information about the timeline view.