WebKit-https.git
4 years ago[jhbuild] Fix pixman build with clang
ossy@webkit.org [Fri, 27 Nov 2015 13:56:20 +0000 (13:56 +0000)]
[jhbuild] Fix pixman build with clang
https://bugs.webkit.org/show_bug.cgi?id=151441

Reviewed by Carlos Garcia Campos.

* efl/jhbuild.modules:
* gtk/jhbuild.modules:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192780 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFix build warning in bignum.cc
ossy@webkit.org [Fri, 27 Nov 2015 13:24:16 +0000 (13:24 +0000)]
Fix build warning in bignum.cc
https://bugs.webkit.org/show_bug.cgi?id=150797

Reviewed by Geoffrey Garen.

* wtf/dtoa/bignum.cc:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192779 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUse mark pragmas only if it is supported
ossy@webkit.org [Fri, 27 Nov 2015 13:01:54 +0000 (13:01 +0000)]
Use mark pragmas only if it is supported
https://bugs.webkit.org/show_bug.cgi?id=151621

Reviewed by Mark Lam.

* b3/air/AirIteratedRegisterCoalescing.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192778 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFix the ENABLE(B3_JIT) build with GCC in B3Procedure.h
ossy@webkit.org [Fri, 27 Nov 2015 13:01:22 +0000 (13:01 +0000)]
Fix the ENABLE(B3_JIT) build with GCC in B3Procedure.h
https://bugs.webkit.org/show_bug.cgi?id=151620

Reviewed by Mark Lam.

* b3/B3Procedure.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192777 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[cmake] Add new B3 source files to the build system
ossy@webkit.org [Fri, 27 Nov 2015 12:59:34 +0000 (12:59 +0000)]
[cmake] Add new B3 source files to the build system
https://bugs.webkit.org/show_bug.cgi?id=151618

Reviewed by Gyuyoung Kim.

* CMakeLists.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192776 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Do not use the WebCore garbage collector timer
carlosgc@webkit.org [Fri, 27 Nov 2015 08:37:44 +0000 (08:37 +0000)]
[GTK] Do not use the WebCore garbage collector timer
https://bugs.webkit.org/show_bug.cgi?id=151623

Reviewed by Martin Robinson.

Now that garbage collector timers have been implemented in
JavaScriptCore for glib, we don't need to use another Timer in WebCore.

* bindings/js/GCController.cpp:
(WebCore::GCController::garbageCollectSoon):
(WebCore::GCController::garbageCollectNowIfNotDoneRecently):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192775 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoExtract prependToEnvironmentVariableList
ddkilzer@apple.com [Thu, 26 Nov 2015 16:43:22 +0000 (16:43 +0000)]
Extract prependToEnvironmentVariableList
<http://webkit.org/b/151536>

Reviewed by Daniel Bates.

Tests: webkitdirs_unittest/appendToEnvironmentVariableList.pl
       webkitdirs_unittest/prependToEnvironmentVariableList.pl

* Scripts/webkitdirs.pm: Export appendToEnvironmentVariableList
and prependToEnvironmentVariableList.
(appendToEnvironmentVariableList): Simplify variable name.
Switch to use $Config{path_sep}.
(prependToEnvironmentVariableList): Add new method.
(setupMacWebKitEnvironment): Switch to use
prependToEnvironmentVariableList().
(setupIOSWebKitEnvironment): Ditto.

* Scripts/webkitperl/webkitdirs_unittest/appendToEnvironmentVariableList.pl: Added.
* Scripts/webkitperl/webkitdirs_unittest/prependToEnvironmentVariableList.pl: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192774 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GLIB] Implement garbage collector timers
carlosgc@webkit.org [Thu, 26 Nov 2015 13:52:45 +0000 (13:52 +0000)]
[GLIB] Implement garbage collector timers
https://bugs.webkit.org/show_bug.cgi?id=151391

Reviewed by Žan Doberšek.

Add GLib implementation using GSource.

* heap/EdenGCActivityCallback.cpp:
* heap/FullGCActivityCallback.cpp:
* heap/GCActivityCallback.cpp:
(JSC::GCActivityCallback::GCActivityCallback):
(JSC::GCActivityCallback::scheduleTimer):
(JSC::GCActivityCallback::cancelTimer):
* heap/GCActivityCallback.h:
* heap/Heap.cpp:
(JSC::Heap::Heap):
* heap/HeapTimer.cpp:
(JSC::HeapTimer::HeapTimer):
(JSC::HeapTimer::~HeapTimer):
(JSC::HeapTimer::timerDidFire):
* heap/HeapTimer.h:
* heap/IncrementalSweeper.cpp:
(JSC::IncrementalSweeper::IncrementalSweeper):
(JSC::IncrementalSweeper::scheduleTimer):
(JSC::IncrementalSweeper::cancelTimer):
* heap/IncrementalSweeper.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192773 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Content Filtering] Crash in DocumentLoader::notifyFinished() when allowing a media...
aestes@apple.com [Thu, 26 Nov 2015 07:08:43 +0000 (07:08 +0000)]
[Content Filtering] Crash in DocumentLoader::notifyFinished() when allowing a media document to load
https://bugs.webkit.org/show_bug.cgi?id=151433
rdar://problem/23506594

Reviewed by Alexey Proskuryakov.

Source/WebCore:

When the main resource of a media document commits, WebKit cancels its load since the plug-in or media engine
will do its own loading. If content filtering is enabled, and the filter waits allow the load until the entire
resource is downloaded, then ContentFilter will attempt to call DocumentLoader::notifyFinished() immediately
after delivering the buffered resource data to DocumentLoader. However, delivering the data will have nulled out
DocumentLoader's m_mainResource when the load was cancelled, leading to a crash in notifyFinished().

To resolve this, add a new Stopped state to ContentFilter. Set this state if DocumentLoader clears its main
resource or detaches from its frame. If ContentFilter is in the Stopped state after calling
DocumentLoader::dataReceived(), do not proceed to call DocumentLoader::notifyFinished().

Test: contentfiltering/allow-media-document.html

* loader/ContentFilter.cpp:
(WebCore::ContentFilter::stopFilteringMainResource): Set m_state to Stopped. If m_mainResource is non-null,
removed ContentFilter as a client and set m_mainResource to null.
(WebCore::ContentFilter::notifyFinished): Stopped calling DocumentLoader::notifyFinished() if m_state is Stopped
after calling DocumentLoader::dataReceived().
* loader/ContentFilter.h:
* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::detachFromFrame): Called ContentFilter::stopFilteringMainResource() instead of setting
m_contentFilter to null.
(WebCore::DocumentLoader::clearMainResource): Ditto.

LayoutTests:

* contentfiltering/allow-media-document-expected.txt: Added.
* contentfiltering/allow-media-document.html: Added.
* contentfiltering/resources/test.mp4: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192772 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: save Inspector's breakpoints to localStorage whenever they are modified
bburg@apple.com [Wed, 25 Nov 2015 20:53:51 +0000 (20:53 +0000)]
Web Inspector: save Inspector's breakpoints to localStorage whenever they are modified
https://bugs.webkit.org/show_bug.cgi?id=151581

Reviewed by Timothy Hatcher.

Serialize all breakpoints to the "breakpoints" Setting in local storage
whenever any breakpoint model object is added, removed, or modified.

Remove the old listener that attempted to save breakpoints on the
pagehide event. It did not fire in important scenarios like exiting
the browser via Cmd-Q or killing the process via Ctrl-C / SIGKILL.

This is not expected to be a performance problem because most people
do not keep thousands of breakpoints active, and breakpoints are not
set very often. If it's a problem, we can mitigate it with coalescing.

* UserInterface/Controllers/DebuggerManager.js:
(WebInspector.DebuggerManager.prototype.addBreakpoint):
(WebInspector.DebuggerManager.prototype.removeBreakpoint):
(WebInspector.DebuggerManager.prototype._breakpointDisabledStateDidChange):
(WebInspector.DebuggerManager.prototype._saveBreakpoints):
(WebInspector.DebuggerManager.prototype._inspectorClosing): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192771 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoChecks for buffer-overflows when reading characters from textRun
mmaxfield@apple.com [Wed, 25 Nov 2015 19:04:03 +0000 (19:04 +0000)]
Checks for buffer-overflows when reading characters from textRun
https://bugs.webkit.org/show_bug.cgi?id=151055
<rdar://problem/23251789>

Patch by Pranjal Jumde <pjumde@apple.com> on 2015-11-25
Reviewed by Myles C. Maxfield.

Source/WebCore:

Prevents an off by one error when adding the last font data to the GlyphBuffer.

* Source/WebCore/platform/graphics/WidthIterator.cpp:
* Source/WebCore/platform/graphics/FontCascade.cpp:

LayoutTests:

* dom/html/level1/core/151055_asan.html:
* dom/html/level1/core/151055_asan-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192770 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoTeach MiniBrowser how to enable the mock content filter
aestes@apple.com [Wed, 25 Nov 2015 08:48:36 +0000 (08:48 +0000)]
Teach MiniBrowser how to enable the mock content filter
https://bugs.webkit.org/show_bug.cgi?id=151540

Reviewed by Andreas Kling.

Source/WebCore:

Moved the implementation of MockContentFilterEnabler from TestWebKitAPI to here, renamed it to
WebMockContentFilterEnabler, and made it compatible with the legacy Objective-C runtime. Renamed Decision and
DecisionPoint to WebMockContentFilterDecision and WebMockContentFilterDecisionPoint, and changed them from enum
classes to CF_ENUMs so that they can be used by both C++ and Objective-C source files.

* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSMockContentFilterSettingsCustom.cpp:
(WebCore::JSMockContentFilterSettings::decisionPoint):
(WebCore::JSMockContentFilterSettings::setDecisionPoint):
(WebCore::toDecision):
(WebCore::JSMockContentFilterSettings::decision):
(WebCore::JSMockContentFilterSettings::setDecision):
(WebCore::JSMockContentFilterSettings::unblockRequestDecision):
(WebCore::JSMockContentFilterSettings::setUnblockRequestDecision):
(WebCore::toJSValue): Deleted.
* testing/MockContentFilter.cpp:
(WebCore::MockContentFilter::willSendRequest):
(WebCore::MockContentFilter::responseReceived):
(WebCore::MockContentFilter::addData):
(WebCore::MockContentFilter::finishedAddingData):
(WebCore::MockContentFilter::unblockHandler):
(WebCore::MockContentFilter::maybeDetermineStatus):
* testing/MockContentFilter.h:
* testing/MockContentFilterEnabler.h: Added.
* testing/MockContentFilterEnabler.mm: Added.
(-[WebMockContentFilterEnabler initWithDecision:decisionPoint:blockedString:]):
(-[WebMockContentFilterEnabler initWithCoder:]):
(-[WebMockContentFilterEnabler encodeWithCoder:]):
(-[WebMockContentFilterEnabler enable]):
(-[WebMockContentFilterEnabler dealloc]):
(+[WebMockContentFilterEnabler supportsSecureCoding]):
(-[WebMockContentFilterEnabler copyWithZone:]):
* testing/MockContentFilterSettings.h:
(WebCore::MockContentFilterSettings::decisionPoint):
(WebCore::MockContentFilterSettings::setDecisionPoint):
(WebCore::MockContentFilterSettings::decision):
(WebCore::MockContentFilterSettings::setDecision):
(WebCore::MockContentFilterSettings::unblockRequestDecision):
(WebCore::MockContentFilterSettings::setUnblockRequestDecision):

Tools:

* MiniBrowser/Configurations/Base.xcconfig: Added WebCoreTestSupport and WTF to HEADER_SEARCH_PATHS.
* MiniBrowser/Configurations/MiniBrowser.xcconfig: Linked against libWebCoreTestSupport.
* MiniBrowser/Configurations/MiniBrowserBundle.xcconfig: Ditto.
* MiniBrowser/MiniBrowser.xcodeproj/project.pbxproj: Added blocked-page.html.
* MiniBrowser/MiniBrowserWebProcessPlugIn.m:
(-[MiniBrowserWebProcessPlugIn webProcessPlugIn:initializeWithObject:]): Observed the WebMockContentFilterEnabler key path.
(-[MiniBrowserWebProcessPlugIn dealloc]):
(-[MiniBrowserWebProcessPlugIn observeValueForKeyPath:ofObject:change:context:]): Stored the value in _contentFilterEnabler.
* MiniBrowser/blocked-page.html: Added.
* MiniBrowser/mac/AppDelegate.m:
(defaultConfiguration): Created a _WKProcessPoolConfiguration with MiniBrowser.wkbundle as the injected bundle.
* MiniBrowser/mac/BrowserWindowController.h:
* MiniBrowser/mac/BrowserWindowController.m:
(+[BrowserWindowController contentFilteringBlockedString]): Returned an NSString containing blocked-page.html.
* MiniBrowser/mac/SettingsController.h:
* MiniBrowser/mac/SettingsController.m:
(-[SettingsController _populateMenu]): Added [self _contentFilteringMenuItem] to the Settings menu.
(-[SettingsController validateMenuItem:]): Validated the new menu items. Disabled Decision and Decision Point items if filtering is disabled.
(-[SettingsController _contentFilteringMenuItem]): Returned a new Content Filtering menu item with a submenu.
(-[SettingsController toggleContentFilteringEnabled:]): Toggled ContentFilteringEnabledKey.
(-[SettingsController contentFilteringEnabled]): Returned value of ContentFilteringEnabledKey.
(-[SettingsController setContentFilteringDecision:]): Set ContentFilterDecisionKey and called -[BrowserAppDelegate didChangeSettings].
(-[SettingsController contentFilteringDecision]): Returned the value of ContentFilterDecisionKey as a WebMockContentFilterDecision.
Defaulted to WebMockContentFilterDecisionAllow if the value is invalid.
(-[SettingsController setContentFilteringDecisionPoint:]): Set ContentFilterDecisionPointKey and called -[BrowserAppDelegate didChangeSettings].
(-[SettingsController contentFilteringDecisionPoint]): Returned the value of ContentFilterDecisionPointKey as a WebMockContentFilterDecisionPoint.
Defaulted to WebMockContentFilterDecisionPointAfterWillSendRequest if the value is invalid.
* MiniBrowser/mac/WK1BrowserWindowController.h:
* MiniBrowser/mac/WK1BrowserWindowController.m:
(-[WK1BrowserWindowController dealloc]):
(-[WK1BrowserWindowController didChangeSettings]): Created a WebMockContentFilterEnabler with the current settings and stored it in _contentFilterEnabler.
* MiniBrowser/mac/WK2BrowserWindowController.m:
(-[WK2BrowserWindowController didChangeSettings]): Created a WebMockContentFilterEnabler and set it as the object for the eponymous bundle parameter.
* TestWebKitAPI/Configurations/TestWebKitAPI.xcconfig: Linked against libWebCoreTestSupport.
* TestWebKitAPI/Tests/WebKit2Cocoa/ContentFiltering.mm:
(configurationWithContentFilterSettings): Converted to use WebMockContentFilterEnabler, WebMockContentFilterDecision, and WebMockContentFilterDecisionPoint.
(TEST): Ditto.
(downloadTest): Ditto.
(+[MockContentFilterEnabler supportsSecureCoding]): Deleted.
(-[MockContentFilterEnabler copyWithZone:]): Deleted.
(-[MockContentFilterEnabler initWithCoder:]): Deleted.
(-[MockContentFilterEnabler initWithDecision:decisionPoint:]): Deleted.
(-[MockContentFilterEnabler encodeWithCoder:]): Deleted.
* TestWebKitAPI/Tests/WebKit2Cocoa/ContentFilteringPlugIn.mm:
(-[ContentFilteringPlugIn webProcessPlugIn:initializeWithObject:]): Converted to use WebMockContentFilterEnabler.
(-[ContentFilteringPlugIn dealloc]): Ditto.
(-[ContentFilteringPlugIn observeValueForKeyPath:ofObject:change:context:]): Ditto.
(+[MockContentFilterEnabler supportsSecureCoding]): Deleted.
(-[MockContentFilterEnabler copyWithZone:]): Deleted.
(-[MockContentFilterEnabler initWithCoder:]): Deleted.
(-[MockContentFilterEnabler dealloc]): Deleted.
(-[MockContentFilterEnabler encodeWithCoder:]): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192769 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[JSC] support Computed Property Names in destructuring Patterns
commit-queue@webkit.org [Wed, 25 Nov 2015 01:43:14 +0000 (01:43 +0000)]
[JSC] support Computed Property Names in destructuring Patterns
https://bugs.webkit.org/show_bug.cgi?id=151494

Patch by Caitlin Potter <caitp@igalia.com> on 2015-11-24
Reviewed by Saam Barati.

Add support for computed property names in destructuring BindingPatterns
and AssignmentPatterns.

Productions BindingProperty(1) and AssignmentProperty(2) allow for any valid
PropertName(3), including ComputedPropertyName(4)

1: http://tc39.github.io/ecma262/#prod-BindingProperty
2: http://tc39.github.io/ecma262/#prod-AssignmentProperty
3: http://tc39.github.io/ecma262/#prod-PropertyName
4: http://tc39.github.io/ecma262/#prod-ComputedPropertyName

* bytecompiler/NodesCodegen.cpp:
(JSC::ObjectPatternNode::bindValue):
* parser/ASTBuilder.h:
(JSC::ASTBuilder::appendObjectPatternEntry):
* parser/Nodes.h:
(JSC::ObjectPatternNode::appendEntry):
* parser/Parser.cpp:
(JSC::Parser<LexerType>::parseDestructuringPattern):
* parser/SyntaxChecker.h:
(JSC::SyntaxChecker::operatorStackPop):
* tests/es6.yaml:
* tests/es6/destructuring_assignment_computed_properties.js: Added.
(test):
(test.computeName):
(test.loadValue):
(test.out.get a):
(test.out.set a):
(test.out.get b):
(test.out.set b):
(test.out.get c):
(test.out.set c):
(test.get var):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192768 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION(r192053): MiniBrowser doesn't exit when clicking on the close-window button
gyuyoung.kim@webkit.org [Wed, 25 Nov 2015 00:20:42 +0000 (00:20 +0000)]
REGRESSION(r192053): MiniBrowser doesn't exit when clicking on the close-window button
https://bugs.webkit.org/show_bug.cgi?id=151567

Reviewed by Darin Adler.

* MiniBrowser/efl/main.c: Call window_close() instead of ewk_view_try_close().
(on_window_deletion):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192767 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, rolling out r192536, r192722, and r192743.
commit-queue@webkit.org [Tue, 24 Nov 2015 21:37:37 +0000 (21:37 +0000)]
Unreviewed, rolling out r192536, r192722, and r192743.
https://bugs.webkit.org/show_bug.cgi?id=151593

Still causing trouble. (Requested by kling on #webkit).

Reverted changesets:

"[JSC] JSPropertyNameEnumerator could be destructorless."
https://bugs.webkit.org/show_bug.cgi?id=151242
http://trac.webkit.org/changeset/192536

"REGRESSION(r192536): Null pointer dereference in
JSPropertyNameEnumerator::visitChildren()."
https://bugs.webkit.org/show_bug.cgi?id=151495
http://trac.webkit.org/changeset/192722

"REGRESSION(r192536): Null pointer dereference in
JSPropertyNameEnumerator::visitChildren()."
https://bugs.webkit.org/show_bug.cgi?id=151495
http://trac.webkit.org/changeset/192743

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192766 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Streams API] Implement pipeTo method in readable Stream
calvaris@igalia.com [Tue, 24 Nov 2015 18:47:19 +0000 (18:47 +0000)]
[Streams API] Implement pipeTo method in readable Stream
https://bugs.webkit.org/show_bug.cgi?id=151588

Reviewed by Darin Adler.

Source/WebCore:

Implemented pipeTo method according to the reference implementation in the spec as the spec is not written
yet. It can be found at https://github.com/whatwg/streams/blob/632b26a05f3106650b1ec91239ad5b012e6c64af/reference-implementation/lib/readable-stream.js#L75.

Tests: streams/pipe-to.html
       streams/reference-implementation/brand-checks.html
       streams/reference-implementation/pipe-through.html
       streams/reference-implementation/pipe-to.html
       streams/reference-implementation/pipe-to-options.html
       streams/reference-implementation/readable-stream-templated

* Modules/streams/ReadableStream.js:
(doPipe): Internal function of pipeTo.
(closeDest): Internal function of pipeTo.
(abortDest): Internal function of pipeTo.
(pipeTo): Implemented as per spec with some other internal functions as helpers.

LayoutTests:

Test "Piping to a writable stream that does not consume the writes fast enough exerts backpressure on the
source" was moved to its own file because it causes timing issues.

* streams/reference-implementation/pipe-to.html: Moved "Piping to a writable stream that does not consume the
writes fast enough exerts backpressure on the source" test to its own file.
* streams/pipe-to.html: Added with "Piping to a writable stream that does not consume the writes fast enough
exerts backpressure on the source" test.
* streams/reference-implementation/brand-checks.html: Fixed issue with the creation of a ReadableStreamReader.
* streams/pipe-to-expected.txt:
* streams/reference-implementation/brand-checks-expected.txt:
* streams/reference-implementation/pipe-through-expected.txt:
* streams/reference-implementation/pipe-to-expected.txt:
* streams/reference-implementation/pipe-to-options-expected.txt:
* streams/reference-implementation/readable-stream-templated-expected.txt: Expectations.
* platform/mac/TestExpectations:
* platform/win/TestExpectations: Flagged pipe-to test because of webkit.org/b/147933.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192765 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: Cmd-1 to Cmd-9 shortcuts should select tabs by ordinal
bburg@apple.com [Tue, 24 Nov 2015 18:22:03 +0000 (18:22 +0000)]
Web Inspector: Cmd-1 to Cmd-9 shortcuts should select tabs by ordinal
https://bugs.webkit.org/show_bug.cgi?id=151577

Reviewed by Timothy Hatcher.

The shortcuts only work if a tab at the specified ordinal exists.

* UserInterface/Base/Main.js:
(WebInspector.contentLoaded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192764 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION (r190983): Non-element, non-text nodes should not be distributed to slots
antti@apple.com [Tue, 24 Nov 2015 12:40:57 +0000 (12:40 +0000)]
REGRESSION (r190983): Non-element, non-text nodes should not be distributed to slots
https://bugs.webkit.org/show_bug.cgi?id=151566
rdar://problem/23430177

Reviewed by Zalan Bujtas.

Source/WebCore:

We don't invalidate slot assignments except for text or element children. Fix by not
not assigning other nodes to slots as it is not useful.

Test: fast/html/details-comment-crash.html

* dom/SlotAssignment.cpp:
(WebCore::slotNameFromSlotAttribute):
(WebCore::SlotAssignment::findAssignedSlot):
(WebCore::SlotAssignment::assignSlots):

LayoutTests:

* fast/html/details-comment-crash-expected.html: Added.
* fast/html/details-comment-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192763 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Win] Implement methods to get and set inner html.
peavo@outlook.com [Tue, 24 Nov 2015 08:37:31 +0000 (08:37 +0000)]
[Win] Implement methods to get and set inner html.
https://bugs.webkit.org/show_bug.cgi?id=151569

Reviewed by Alex Christensen.

* DOMHTMLClasses.cpp:
(DOMHTMLElement::innerHTML):
(DOMHTMLElement::setInnerHTML):
(DOMHTMLElement::innerText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192762 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Use the network process unconditionally
carlosgc@webkit.org [Tue, 24 Nov 2015 08:01:35 +0000 (08:01 +0000)]
[GTK] Use the network process unconditionally
https://bugs.webkit.org/show_bug.cgi?id=151541

Reviewed by Alex Christensen.

Source/WebKit2:

Make the shared secondary process model become multiple secondary
process model with a limit of 1 web process. Use the same options
when creating a context with legacy configuration (unit tests and
inspector proxy).

* NetworkProcess/gtk/NetworkProcessMainGtk.cpp:
* UIProcess/API/APIProcessPoolConfiguration.cpp:
(API::ProcessPoolConfiguration::createWithLegacyOptions):
* UIProcess/API/gtk/WebKitWebContext.cpp:
(webkitWebContextConstructed):
(webkit_web_context_set_process_model):
(webkit_web_context_get_process_model):
(webkit_web_context_set_web_process_count_limit):
(webkit_web_context_get_web_process_count_limit):
(toWebKitProcessModel): Deleted.
* UIProcess/gtk/WebInspectorProxyGtk.cpp:
* UIProcess/gtk/WebProcessPoolGtk.cpp:
(WebKit::WebProcessPool::setIgnoreTLSErrors):
(WebKit::WebProcessPool::platformInitializeWebProcess):

Tools:

Fix TestInspectorServer test. Do not assume we already have the
title we want when the page has been loaded, since the title is
changed afterwards. So, check if the title has already been set,
and if not wait for it.

* TestWebKitAPI/Tests/WebKit2Gtk/TestInspectorServer.cpp:
(openRemoteDebuggingSession):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192761 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, rolling out r192716 and r192717.
commit-queue@webkit.org [Tue, 24 Nov 2015 07:13:03 +0000 (07:13 +0000)]
Unreviewed, rolling out r192716 and r192717.
https://bugs.webkit.org/show_bug.cgi?id=151582

The patch was incorrect. We always need at least one data
point in each configuration (Requested by rniwa on #webkit).

Reverted changesets:

"Perf dashboard's should not include results more than 366
days old in JSON"
https://bugs.webkit.org/show_bug.cgi?id=151529
http://trac.webkit.org/changeset/192716

"Build fix for old version of PHP."
http://trac.webkit.org/changeset/192717

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192760 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, fix the Mac CMake build after r192793.
bburg@apple.com [Tue, 24 Nov 2015 07:12:19 +0000 (07:12 +0000)]
Unreviewed, fix the Mac CMake build after r192793.

* PlatformMac.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192759 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoHardening against CSSSelector double frees
ddkilzer@apple.com [Tue, 24 Nov 2015 03:30:03 +0000 (03:30 +0000)]
Hardening against CSSSelector double frees
<http://webkit.org/b/56124>
<rdar://problem/9119036>

Reviewed by Antti Koivisto.

Add some security assertions to catch this issue if it ever
happens in Debug builds, and make changes in
CSSSelector::~CSSSelector() and
CSSSelectorList::deleteSelectors() to prevent obvious issues if
they're ever called twice in Release builds.

No new tests because we don't know how to reproduce this.

* css/CSSSelector.cpp:
(WebCore::CSSSelector::CSSSelector): Initialize
m_destructorHasBeenCalled.
* css/CSSSelector.h:
(WebCore::CSSSelector::m_destructorHasBeenCalled): Add bitfield.
(WebCore::CSSSelector::CSSSelector): Initialize
m_destructorHasBeenCalled.
(WebCore::CSSSelector::~CSSSelector): Add security assertion
that this is never called twice.  Clear out any fields that
would have caused us to dereference an object twice.

* css/CSSSelectorList.cpp:
(WebCore::CSSSelectorList::deleteSelectors): Clear
m_selectorArray when freeing the memory to which it was
pointing.  This prevents re-entrancy issues or calling this
method twice on the same thread.  Also restructure the for()
loop to prevent calling CSSSelector::isLastInSelectorList()
after CSSSelector::~CSSSelector() has been called (via CRBug
241892).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192758 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: Add a keyboard shortcut to restore the default zoom level
bburg@apple.com [Tue, 24 Nov 2015 02:53:28 +0000 (02:53 +0000)]
Web Inspector: Add a keyboard shortcut to restore the default zoom level
https://bugs.webkit.org/show_bug.cgi?id=151237

Reviewed by Timothy Hatcher.

The Cmd-0 shortcut matches the keybindings of Safari and other browsers.

* UserInterface/Base/Main.js:
(WebInspector.contentLoaded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192757 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: inspector settings should not be shared between different inspection...
bburg@apple.com [Tue, 24 Nov 2015 02:49:51 +0000 (02:49 +0000)]
Web Inspector: inspector settings should not be shared between different inspection levels
https://bugs.webkit.org/show_bug.cgi?id=151151

Reviewed by Timothy Hatcher.

Source/WebCore:

InspectorController and InspectorFrontendClient should know how to compute their
inspection levels, so that this level can be exposed to the Inspector frontend for
preference disambiguation by inspection level.

 - A Page with normal web content has an inspection level of 0.
 - The first Web Inspector has an inspection level of 1.
 - The second Web Inspector that inspects the Inspector has an inspection level of 2.
 - And so forth...

For local frontend clients (WK1 and the ProtocolTestStub), the inspection level of
a Page computed as follows:

- If the page's inspector controller has no frontend client, then the Page
is normal web content (level 0).
- If the page's inspector controller has a frontend client, ask for its inspection level.
A frontend client looks at the inspection level reported by inspected page's inspector
controller and increments by one to account for the crossed inspection boundary.

No new tests, this is blocked by test infrastructure fixes (https://webkit.org/b/151573).

* inspector/InspectorController.cpp:
(WebCore::InspectorController::inspectionLevel): Added.
(WebCore::InspectorController::hasInspectorFrontendClient): Deleted.
* inspector/InspectorController.h:
* inspector/InspectorFrontendClient.h:
* inspector/InspectorFrontendClientLocal.cpp:
(WebCore::InspectorFrontendClientLocal::canAttachWindow): Check inspectionLevel().
(WebCore::InspectorFrontendClientLocal::inspectionLevel): Added.
* inspector/InspectorFrontendClientLocal.h:
* inspector/InspectorFrontendHost.cpp:
(WebCore::InspectorFrontendHost::inspectionLevel): Added.
* inspector/InspectorFrontendHost.h:
* inspector/InspectorFrontendHost.idl:
* page/ContextMenuController.cpp:
(WebCore::ContextMenuController::populate): Check inspectionLevel().

Source/WebInspectorUI:

It's really annoying to have shared settings between Inspector^1 and Inspector^2, because
they are not designed to respond to external changes to local storage keys. As a result,
the two inspectors seemingly have different settings while both are open, but the settings
clobber each other depending on which inspector is closed first. On the next inspector
being opened, it has settings that reflect whichever inspector closed last.

To fix this, alter the local storage key prefix to incude the inspection level when it is
greater than 1 (i.e., inspecting the inspector). The storage prefix becomes
"com.apple.WebInspector-${inspectionLevel}." in this case.
This is backwards-compatible for normal Inspector, who keeps the same storage prefix.

Alternate approaches that were considered and abandoned:

 - Use separate WKWebSiteDataStores for each inspection level. This API (as it currently is)
does not support multiple on-disk storage locations. We don't want to add an entirely new
Library directory for each inspector level, and sharing a single location doesn't fix the
shared-settings bug. Changing storage location would lose all existing Inspector settings.

 - Sync settings between multiple open Inspectors. This would be catastrophic, as changing
the active tab on one inspector would immediately propagate to the other open inspectors.

* UserInterface/Models/Setting.js:
(WebInspector.Setting):
* UserInterface/Protocol/InspectorFrontendHostStub.js:
(window.InspectorFrontendHost.WebInspector.InspectorFrontendHostStub.prototype.inspectionLevel):

Source/WebKit2:

In WebKit2, we already track the inspection level of WebPages in order to give inspectors
different page groups. Send the inspection level to WebInspectorUI WebProcess when
establishing a connection from the UIProcess. Use this number in the FrontendClient.

* UIProcess/WebInspectorProxy.cpp:
(WebKit::WebInspectorProxy::inspectionLevel): Renamed from inspectorLevel to be consistent.
(WebKit::WebInspectorProxy::inspectorPageGroupIdentifier):
(WebKit::WebInspectorProxy::didRelaunchInspectorPageProcess):
(WebKit::WebInspectorProxy::eagerlyCreateInspectorPage):
(WebKit::WebInspectorProxy::createInspectorPage):
(WebKit::WebInspectorProxy::inspectorLevel): Deleted.
* UIProcess/WebInspectorProxy.h:
* UIProcess/mac/WebInspectorProxyMac.mm:
(WebKit::WebInspectorProxy::updateInspectorWindowTitle):
* WebProcess/WebPage/WebInspectorUI.cpp:
(WebKit::WebInspectorUI::establishConnection):
* WebProcess/WebPage/WebInspectorUI.h:
* WebProcess/WebPage/WebInspectorUI.messages.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192756 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: use Cmd-Option-L and Cmd-Option-R for toggling navigation and details...
bburg@apple.com [Tue, 24 Nov 2015 02:43:36 +0000 (02:43 +0000)]
Web Inspector: use Cmd-Option-L and Cmd-Option-R for toggling navigation and details sidebars
https://bugs.webkit.org/show_bug.cgi?id=151572

Reviewed by Timothy Hatcher.

The previous shortcuts were Cmd-0 and Cmd-Option-0, but these didn't match Safari's
sidebar shortcuts and conflict with Cmd-0 as the cross-browser way to reset page zoom.

* UserInterface/Base/Main.js:
(WebInspector.contentLoaded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192755 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION(r188206): [EFL] Missing to adjust scrollbar size to ewk_view_contents_size...
gyuyoung.kim@webkit.org [Tue, 24 Nov 2015 01:17:50 +0000 (01:17 +0000)]
REGRESSION(r188206): [EFL] Missing to adjust scrollbar size to ewk_view_contents_size_get API test
https://bugs.webkit.org/show_bug.cgi?id=148735

Reviewed by Csaba Osztrogonác.

r189256 missed to adjust scrollbar size to other tests in ewk_view_contents_size_get() API test.
In this case we need to apply scrollbar size to both width and height unlike previous fix. Because
horizontal scrollbar is also shown since device pixel ratio(= 2.0) is adjusted.

* UIProcess/API/efl/tests/test_ewk2_view.cpp:
(TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: RemoteInspector should track targets and connections for remote automation
bburg@apple.com [Mon, 23 Nov 2015 23:51:48 +0000 (23:51 +0000)]
Web Inspector: RemoteInspector should track targets and connections for remote automation
https://bugs.webkit.org/show_bug.cgi?id=151042

Reviewed by Joseph Pecoraro.

Source/JavaScriptCore:

Refactor RemoteInspector so it can be used to send listings of different target types.
First, rename Debuggable to RemoteInspectionTarget, and pull things not specific to
remote inspection into the base class RemoteControllableTarget and its Connection class.

Add a new RemoteControllableTarget called RemoteAutomationTarget, used by UIProcess
to support remote UI automation via webinspectord. On the protocol side, this target
uses a new WIRTypeKey called WIRTypeAutomation to distiguish the listing from
Web and JavaScript listings and avoid inventing a new listing mechanism.

* API/JSContextRef.cpp:
(JSGlobalContextGetDebuggerRunLoop):
(JSGlobalContextSetDebuggerRunLoop):
* JavaScriptCore.xcodeproj/project.pbxproj:
* inspector/InspectorFrontendChannel.h:
* inspector/remote/RemoteAutomationTarget.cpp: Added.
(Inspector::RemoteAutomationTarget::setAutomationAllowed): Added.
* inspector/remote/RemoteAutomationTarget.h: Added.
* inspector/remote/RemoteConnectionToTarget.h: Renamed from Source/JavaScriptCore/inspector/remote/RemoteInspectorDebuggableConnection.h.
(Inspector::RemoteTargetBlock::RemoteTargetBlock):
(Inspector::RemoteTargetBlock::~RemoteTargetBlock):
(Inspector::RemoteTargetBlock::operator=):
(Inspector::RemoteTargetBlock::operator()):
* inspector/remote/RemoteConnectionToTarget.mm: Renamed from Source/JavaScriptCore/inspector/remote/RemoteInspectorDebuggableConnection.mm.
(Inspector::RemoteTargetHandleRunSourceGlobal):
(Inspector::RemoteTargetQueueTaskOnGlobalQueue):
(Inspector::RemoteTargetInitializeGlobalQueue):
(Inspector::RemoteTargetHandleRunSourceWithInfo):
(Inspector::RemoteConnectionToTarget::RemoteConnectionToTarget):
(Inspector::RemoteConnectionToTarget::~RemoteConnectionToTarget):
(Inspector::RemoteConnectionToTarget::destination):
(Inspector::RemoteConnectionToTarget::connectionIdentifier):
(Inspector::RemoteConnectionToTarget::dispatchAsyncOnTarget):
(Inspector::RemoteConnectionToTarget::setup):
(Inspector::RemoteConnectionToTarget::targetClosed):
(Inspector::RemoteConnectionToTarget::close):
(Inspector::RemoteConnectionToTarget::sendMessageToTarget):
(Inspector::RemoteConnectionToTarget::sendMessageToFrontend):
(Inspector::RemoteConnectionToTarget::setupRunLoop):
(Inspector::RemoteConnectionToTarget::teardownRunLoop):
(Inspector::RemoteConnectionToTarget::queueTaskOnPrivateRunLoop):
* inspector/remote/RemoteControllableTarget.cpp: Added.
(Inspector::RemoteControllableTarget::~RemoteControllableTarget):
(Inspector::RemoteControllableTarget::init):
(Inspector::RemoteControllableTarget::update):
* inspector/remote/RemoteControllableTarget.h: Added.
* inspector/remote/RemoteInspectionTarget.cpp: Renamed from Source/JavaScriptCore/inspector/remote/RemoteInspectorDebuggable.cpp.
(Inspector::RemoteInspectionTarget::remoteControlAllowed):
(Inspector::RemoteInspectionTarget::setRemoteDebuggingAllowed):
(Inspector::RemoteInspectionTarget::pauseWaitingForAutomaticInspection):
(Inspector::RemoteInspectionTarget::unpauseForInitializedInspector):
* inspector/remote/RemoteInspectionTarget.h: Renamed from Source/JavaScriptCore/inspector/remote/RemoteInspectorDebuggable.h.
(isType):
* inspector/remote/RemoteInspector.h:

    Code to manage Debuggables now works with RemoteControllableTargets and doesn't
    care whether the target is for Inspection or Automation. Listing data with target-
    and type-specific information are captured when clients call into RemoteInspector
    since that's the easiest time to gather this information on the right thread.
    Use the is<> / downcast<> machinery when we need a concrete Target type.

* inspector/remote/RemoteInspector.mm:
(Inspector::RemoteInspector::nextAvailableIdentifier):
(Inspector::RemoteInspector::registerTarget): renamed from registerDebuggable.
(Inspector::RemoteInspector::unregisterTarget): renamed from unregisterDebuggable.
(Inspector::RemoteInspector::updateTarget): renamed from updateDebuggable.
(Inspector::RemoteInspector::updateAutomaticInspectionCandidate):
(Inspector::RemoteInspector::sendMessageToRemote):
(Inspector::RemoteInspector::setupFailed):
(Inspector::RemoteInspector::stopInternal):
(Inspector::RemoteInspector::setupXPCConnectionIfNeeded):
(Inspector::RemoteInspector::xpcConnectionFailed):
(Inspector::RemoteInspector::listingForTarget):
(Inspector::RemoteInspector::listingForInspectionTarget):
(Inspector::RemoteInspector::listingForAutomationTarget):
(Inspector::RemoteInspector::pushListingsNow):
(Inspector::RemoteInspector::pushListingsSoon):
(Inspector::RemoteInspector::receivedSetupMessage):
(Inspector::RemoteInspector::receivedDataMessage):
(Inspector::RemoteInspector::receivedDidCloseMessage):
(Inspector::RemoteInspector::receivedGetListingMessage):
(Inspector::RemoteInspector::receivedIndicateMessage):
(Inspector::RemoteInspector::receivedConnectionDiedMessage):
(Inspector::RemoteInspector::RemoteInspector): Deleted.
(Inspector::RemoteInspector::registerDebuggable): Deleted.
(Inspector::RemoteInspector::unregisterDebuggable): Deleted.
(Inspector::RemoteInspector::updateDebuggable): Deleted.
(Inspector::RemoteInspector::updateDebuggableAutomaticInspectCandidate): Deleted.
(Inspector::RemoteInspector::sendMessageToRemoteFrontend): Deleted.
(Inspector::RemoteInspector::listingForDebuggable): Deleted.
(Inspector::RemoteInspector::pushListingNow): Deleted.
(Inspector::RemoteInspector::pushListingSoon): Deleted.
* inspector/remote/RemoteInspectorConstants.h:
* runtime/JSGlobalObjectDebuggable.cpp:
(JSC::JSGlobalObjectDebuggable::dispatchMessageFromRemote):
(JSC::JSGlobalObjectDebuggable::pauseWaitingForAutomaticInspection):
(JSC::JSGlobalObjectDebuggable::dispatchMessageFromRemoteFrontend): Deleted.
* runtime/JSGlobalObjectDebuggable.h:

Source/WebCore:

Use the new RemoteControllableTarget API method names.

No new tests, no behavior change.

* page/PageDebuggable.cpp:
(WebCore::PageDebuggable::connect):
(WebCore::PageDebuggable::dispatchMessageFromRemote):
(WebCore::PageDebuggable::dispatchMessageFromRemoteFrontend): Deleted.
* page/PageDebuggable.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192753 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoModern IDB: Unskip all indexeddb/mozilla tests that pass.
beidson@apple.com [Mon, 23 Nov 2015 23:12:03 +0000 (23:12 +0000)]
Modern IDB: Unskip all indexeddb/mozilla tests that pass.
https://bugs.webkit.org/show_bug.cgi?id=151568
Source/WebCore:

Reviewed by Alex Christensen.

No new tests (Lots of skipped existing tests now pass).

* Modules/indexeddb/shared/IDBError.h:

LayoutTests:

Reviewed by Alex Christensen.

* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192752 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRename JavaScriptCore builtins files to match exposed object names
bburg@apple.com [Mon, 23 Nov 2015 22:10:05 +0000 (22:10 +0000)]
Rename JavaScriptCore builtins files to match exposed object names
https://bugs.webkit.org/show_bug.cgi?id=151549

Reviewed by Youenn Fablet.

As a subtask of unifying code generation for WebCore and JSC builtins, we need to get rid of
differences between builtins filenames (e.g., Promise.prototype.js) and the name of the
generated Builtin object (PromisePrototype).

If we don't do this, then both build systems need special hacks to normalize the object name
from the file name. It's easier to just normalize the filename.

* CMakeLists.txt:
* DerivedSources.make:
* JavaScriptCore.xcodeproj/project.pbxproj:
* builtins/ArrayIteratorPrototype.js: Renamed from Source/JavaScriptCore/builtins/ArrayIterator.prototype.js.
* builtins/ArrayPrototype.js: Renamed from Source/JavaScriptCore/builtins/Array.prototype.js.
* builtins/FunctionPrototype.js: Renamed from Source/JavaScriptCore/builtins/Function.prototype.js.
* builtins/IteratorPrototype.js: Renamed from Source/JavaScriptCore/builtins/Iterator.prototype.js.
* builtins/PromiseOperations.js: Renamed from Source/JavaScriptCore/builtins/Operations.Promise.js.
* builtins/PromisePrototype.js: Renamed from Source/JavaScriptCore/builtins/Promise.prototype.js.
* builtins/StringIteratorPrototype.js: Renamed from Source/JavaScriptCore/builtins/StringIterator.prototype.js.
* builtins/TypedArrayPrototype.js: Renamed from Source/JavaScriptCore/builtins/TypedArray.prototype.js.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd WebCore namespace for generated bindings for supplemental method calls
commit-queue@webkit.org [Mon, 23 Nov 2015 19:57:28 +0000 (19:57 +0000)]
Add WebCore namespace for generated bindings for supplemental method calls
https://bugs.webkit.org/show_bug.cgi?id=151431

Patch by Nikos Andronikos <nikos.andronikos-webkit@cisra.canon.com.au> on 2015-11-23
Reviewed by Alex Christensen.

Address build failures for generated supplemental method calls which are in the WebCore namespace.
The bindings are outside the WebCore namespace - so add WebCore namespace specifier where required.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateHeader):
(GenerateImplementation):
(GenerateParametersCheck):
* bindings/scripts/CodeGeneratorObjC.pm:
(GenerateImplementation):
* bindings/scripts/test/JS/JSTestInterface.cpp:
(WebCore::jsTestInterfaceConstructorSupplementalStaticReadOnlyAttr):
(WebCore::jsTestInterfaceConstructorSupplementalStaticAttr):
(WebCore::jsTestInterfaceSupplementalStr1):
(WebCore::jsTestInterfaceSupplementalStr2):
(WebCore::jsTestInterfaceSupplementalNode):
(WebCore::setJSTestInterfaceConstructorSupplementalStaticAttr):
(WebCore::setJSTestInterfaceSupplementalStr2):
(WebCore::setJSTestInterfaceSupplementalNode):
(WebCore::jsTestInterfacePrototypeFunctionSupplementalMethod1):
(WebCore::jsTestInterfacePrototypeFunctionSupplementalMethod2):
(WebCore::jsTestInterfaceConstructorFunctionSupplementalMethod4):
* bindings/scripts/test/ObjC/DOMTestInterface.mm:
(-[DOMTestInterface supplementalStr1]):
(-[DOMTestInterface supplementalStr2]):
(-[DOMTestInterface setSupplementalStr2:]):
(-[DOMTestInterface supplementalStr3]):
(-[DOMTestInterface setSupplementalStr3:]):
(-[DOMTestInterface supplementalNode]):
(-[DOMTestInterface setSupplementalNode:]):
(-[DOMTestInterface builtinAttribute]):
(-[DOMTestInterface setBuiltinAttribute:]):
(-[DOMTestInterface supplementalMethod1]):
(-[DOMTestInterface supplementalMethod2:objArg:]):
(-[DOMTestInterface supplementalMethod3]):
(-[DOMTestInterface supplementalMethod4]):
(-[DOMTestInterface builtinFunction]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoModern IDB: When a transaction is aborted, call onerror handlers for all in-progress...
beidson@apple.com [Mon, 23 Nov 2015 19:48:50 +0000 (19:48 +0000)]
Modern IDB: When a transaction is aborted, call onerror handlers for all in-progress requests.
https://bugs.webkit.org/show_bug.cgi?id=151550

Reviewed by Alex Christensen.

Source/WebCore:

Test: storage/indexeddb/modern/abort-requests-cancelled.html
      storage/indexeddb/modern/idbtransaction-objectstore-failures.html (with changes)
      storage/indexeddb/modern/index-5.html (with changes)
      Various (currently skipped) legacy IDB tests.

* Modules/indexeddb/client/IDBDatabaseImpl.cpp:
(WebCore::IDBClient::IDBDatabase::transaction):
(WebCore::IDBClient::IDBDatabase::startVersionChangeTransaction):
(WebCore::IDBClient::IDBDatabase::didStartTransaction):
(WebCore::IDBClient::IDBDatabase::willCommitTransaction):
(WebCore::IDBClient::IDBDatabase::didCommitTransaction):
(WebCore::IDBClient::IDBDatabase::willAbortTransaction):
(WebCore::IDBClient::IDBDatabase::didAbortTransaction):
(WebCore::IDBClient::IDBDatabase::didCommitOrAbortTransaction):

* Modules/indexeddb/client/IDBTransactionImpl.cpp:
(WebCore::IDBClient::IDBTransaction::abort):
(WebCore::IDBClient::IDBTransaction::abortOnServerAndCancelRequests):
(WebCore::IDBClient::IDBTransaction::didCreateObjectStoreOnServer):
(WebCore::IDBClient::IDBTransaction::didCreateIndexOnServer):
(WebCore::IDBClient::IDBTransaction::didGetRecordOnServer):
(WebCore::IDBClient::IDBTransaction::didDeleteObjectStoreOnServer):
(WebCore::IDBClient::IDBTransaction::didDeleteIndexOnServer):
(WebCore::IDBClient::IDBTransaction::immediateAbort): Deleted.
(WebCore::IDBClient::IDBTransaction::abortOnServer): Deleted.
* Modules/indexeddb/client/IDBTransactionImpl.h:

* Modules/indexeddb/server/MemoryIDBBackingStore.cpp:
(WebCore::IDBServer::MemoryIDBBackingStore::abortTransaction):
(WebCore::IDBServer::MemoryIDBBackingStore::commitTransaction):

* Modules/indexeddb/shared/IDBError.cpp:
(WebCore::idbErrorName):
(WebCore::idbErrorDescription):
* Modules/indexeddb/shared/IDBError.h:

* Modules/indexeddb/shared/IDBResourceIdentifier.cpp:
(WebCore::IDBResourceIdentifier::loggingString):
* Modules/indexeddb/shared/IDBResourceIdentifier.h:

LayoutTests:

* storage/indexeddb/modern/abort-requests-cancelled-expected.txt: Added.
* storage/indexeddb/modern/abort-requests-cancelled.html: Added.
* storage/indexeddb/modern/idbtransaction-objectstore-failures-expected.txt:
* storage/indexeddb/modern/idbtransaction-objectstore-failures.html:
* storage/indexeddb/modern/index-5-expected.txt:
* storage/indexeddb/modern/index-5.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoModern IDB: Unskip storage/indexeddb/mozilla/global-data.html.
beidson@apple.com [Mon, 23 Nov 2015 19:41:05 +0000 (19:41 +0000)]
Modern IDB: Unskip storage/indexeddb/mozilla/global-data.html.
https://bugs.webkit.org/show_bug.cgi?id=151557

Reviewed by Alex Christensen.

Source/WebCore:

No new tests (Unskipping existing test storage/indexeddb/mozilla/global-data.html).

- Reworking some invalid ASSERTS
- Actually opening pending open-database-requests after a version change transaction completes
- Allow starting new transactions when the version change transaction has *started* finishing,
  but before it finishes finishing.

* Modules/indexeddb/client/IDBDatabaseImpl.cpp:
(WebCore::IDBClient::IDBDatabase::transaction):

* Modules/indexeddb/client/IDBTransactionImpl.h:

* Modules/indexeddb/server/MemoryObjectStore.cpp:
(WebCore::IDBServer::MemoryObjectStore::~MemoryObjectStore):

* Modules/indexeddb/server/MemoryObjectStoreCursor.cpp:
(WebCore::IDBServer::MemoryObjectStoreCursor::keyAdded): Deleted.

* Modules/indexeddb/server/UniqueIDBDatabase.cpp:
(WebCore::IDBServer::UniqueIDBDatabase::UniqueIDBDatabase):
(WebCore::IDBServer::UniqueIDBDatabase::handleOpenDatabaseOperations):
(WebCore::IDBServer::UniqueIDBDatabase::commitTransaction):
* Modules/indexeddb/server/UniqueIDBDatabase.h:

LayoutTests:

* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192748 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFix crash in ~WebProcessPool when using Geolocation with useNetworkProcess=true
achristensen@apple.com [Mon, 23 Nov 2015 18:39:16 +0000 (18:39 +0000)]
Fix crash in ~WebProcessPool when using Geolocation with useNetworkProcess=true
https://bugs.webkit.org/show_bug.cgi?id=151532

Reviewed by Benjamin Poulain.

Source/WebKit2:

* UIProcess/WebGeolocationManagerProxy.cpp:
(WebKit::WebGeolocationManagerProxy::processPoolDestroyed):
(WebKit::WebGeolocationManagerProxy::processDidClose):
When a WebProcessPool is destroyed, only call stopUpdating if m_updateRequesters.clear()
stopped the updating, like we do in WebGeolocationManagerProxy::removeRequester.
Otherwise, call setEnableHighAccuracy if needed, also like we do in WebGeolocationManagerProxy::removeRequester.

Tools:

* TestWebKitAPI/Tests/WebKit2/Geolocation.cpp:
(TestWebKitAPI::GeolocationTransitionToHighAccuracyStateTracker::eventsChanged):
(TestWebKitAPI::TEST):
(TestWebKitAPI::GeolocationTransitionToLowAccuracyStateTracker::eventsChanged):
(TestWebKitAPI::GeolocationTransitionToHighAccuracyStateTracker::GeolocationTransitionToHighAccuracyStateTracker): Deleted.
(TestWebKitAPI::GeolocationTransitionToLowAccuracyStateTracker::GeolocationTransitionToLowAccuracyStateTracker): Deleted.
Properly load about:blank in all WebViews to clean up.  Without this change, we had a
Geolocation provider stopping after its state tracker was destroyed with its stack frame,
so it was calling a function on a test object that had gone out of scope.
Also, call WKContextSetUsesNetworkProcess(context, true) to show what crash this fixed,
but that will become the default soon and that call will be removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove DOMPromiseWithCallback
youenn.fablet@crf.canon.fr [Mon, 23 Nov 2015 18:31:58 +0000 (18:31 +0000)]
Remove DOMPromiseWithCallback
https://bugs.webkit.org/show_bug.cgi?id=151565

Reviewed by Eric Carlson.

DOMPromiseWithCallback is no longer used as WebCore JS built-ins are used to the cases where DOMPromiseWithCallback made sense.

No change in behavior.

* Modules/mediastream/MediaDevices.h: Moved from DOMPromiseWithCallback to DOMPromise since the API is now called by JS built-ins.
* bindings/js/JSDOMPromise.h:
(WebCore::DOMPromiseWithCallback::DOMPromiseWithCallback): Deleted.
(WebCore::DOMPromiseIteratorWithCallback::DOMPromiseIteratorWithCallback): Deleted.
(WebCore::Error>::resolve): Deleted.
(WebCore::Error>::reject): Deleted.
(WebCore::Error>::resolveEnd): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: when inspecting the inspector, add the inspection level to the title bar
bburg@apple.com [Mon, 23 Nov 2015 17:00:34 +0000 (17:00 +0000)]
Web Inspector: when inspecting the inspector, add the inspection level to the title bar
https://bugs.webkit.org/show_bug.cgi?id=151555

Reviewed by Timothy Hatcher.

Source/WebCore:

* English.lproj/Localizable.strings: add new localized string for alternate inspector title.

Source/WebInspectorUI:

Drive-by update to remove some localization strings that are no longer used.

* Localizations/en.lproj/localizedStrings.js:

Source/WebKit2:

If the inspection level says we are inspecting an inspector, include the level
in the title bar to easily disambiguate it from the base level inspector.

* UIProcess/mac/WebInspectorProxyMac.mm:
(WebKit::WebInspectorProxy::updateInspectorWindowTitle):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GStreamer] No need to assert the pipeline's bus presence in MediaPlayerPrivateGStrea...
zandobersek@gmail.com [Mon, 23 Nov 2015 13:38:49 +0000 (13:38 +0000)]
[GStreamer] No need to assert the pipeline's bus presence in MediaPlayerPrivateGStreamerBase dtor
https://bugs.webkit.org/show_bug.cgi?id=151558

Reviewed by Carlos Garcia Campos.

In the MediaPlayerPrivateGStreamerBase destructor, there's some leftover code
that acquires a reference to the GStreamer pipeline's bus object and then
just asserts that it's not null. Not very useful, so remove it.

Also use nullptr to null out the m_player member.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION(r192536): Null pointer dereference in JSPropertyNameEnumerator::visitChild...
akling@apple.com [Mon, 23 Nov 2015 12:44:24 +0000 (12:44 +0000)]
REGRESSION(r192536): Null pointer dereference in JSPropertyNameEnumerator::visitChildren().
<https://webkit.org/b/151495>

Reviewed by Mark Lam

The test I added when fixing this bug the first time caught another bug when
run on 32-bit: jsString() can also cause GC, so we have to make sure that
JSPropertyNameEnumerator::m_propertyNames is null until after the array it
points to has been populated.

Test: property-name-enumerator-gc-151495.js

* runtime/JSPropertyNameEnumerator.cpp:
(JSC::JSPropertyNameEnumerator::finishCreation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed speculative buildfix after r192701.
ossy@webkit.org [Mon, 23 Nov 2015 11:25:13 +0000 (11:25 +0000)]
Unreviewed speculative buildfix after r192701.

* PlatformMac.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoASSERTION FAILED: freeSpace >= 0 in WebCore::RenderGrid::computeTrackSizesForDirection
svillar@igalia.com [Mon, 23 Nov 2015 09:03:30 +0000 (09:03 +0000)]
ASSERTION FAILED: freeSpace >= 0 in WebCore::RenderGrid::computeTrackSizesForDirection
https://bugs.webkit.org/show_bug.cgi?id=151254

Reviewed by Darin Adler.

Source/WebCore:

When in quirks mode, abnormally huge margins could lead to
negative computations of available logical sizes. We could add
an !document.inQuirksMode() check in the assertion but since
negative freeSpace values are valid (i.e. the condition is not
really part of the contract and everything will work as
expected) I've decided to simply remove it.

Test: fast/css-grid-layout/grid-quirks-mode-huge-margin-crash.html

* rendering/RenderGrid.cpp:
(WebCore::RenderGrid::computeTrackSizesForDirection): Deleted.

LayoutTests:

* fast/css-grid-layout/grid-quirks-mode-huge-margin-crash-expected.txt: Added.
* fast/css-grid-layout/grid-quirks-mode-huge-margin-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed. Update OptionsGTK.cmake and NEWS for 2.11.2 release.
carlosgc@webkit.org [Mon, 23 Nov 2015 08:47:39 +0000 (08:47 +0000)]
Unreviewed. Update OptionsGTK.cmake and NEWS for 2.11.2 release.

.:

* Source/cmake/OptionsGTK.cmake: Bump version numbers.

Source/WebKit2:

* gtk/NEWS: Add release notes for 2.11.2.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agorun-webkit-tests: http server for imported W3C tests doesn't work with --layout-tests...
ddkilzer@apple.com [Mon, 23 Nov 2015 04:02:48 +0000 (04:02 +0000)]
run-webkit-tests: http server for imported W3C tests doesn't work with --layout-tests-directory switch
<http://webkit.org/b/151542>

Reviewed by Daniel Bates.

* Scripts/webkitpy/layout_tests/servers/web_platform_test_server.py:
(base_url): Use the Port object (already passed in) to give us
the path to the LayoutTests directory, which already checks for a
--layout-tests-directory command-line switch.
(WebPlatformTestServer.__init__): Remove layout_test_results_dir
argument since we can get this from the Port object already
passed in via Port.results_directory().  Also switch to use
Port.layout_tests_dir() to get the LayoutTests directory.

* Scripts/webkitpy/layout_tests/servers/web_platform_test_server_unittest.py:
(TestWebPlatformTestServer.test_custom_layout_tests_directory):
Add test case for custom LayoutTests directory.
(TestWebPlatformTestServer.test_previously_spawned_instance):
Update Port object to set mock results directory as if it was
set on the command-line.  Remove unneeded argument from
WebPlatformTestServer constructor.
(TestWebPlatformTestServer.test_corrupted_subserver_files): Ditto.

* Scripts/webkitpy/port/base.py:
(Port.to.start_web_platform_test_server): Remove unneeded
argument from WebPlatformTestServer constuctor.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] [l10n] Updated Italian translation of WebKitGTK+
mcatanzaro@igalia.com [Sun, 22 Nov 2015 22:09:32 +0000 (22:09 +0000)]
[GTK] [l10n] Updated Italian translation of WebKitGTK+
https://bugs.webkit.org/show_bug.cgi?id=151543

Unreviewed.

Patch by Milo Casagrande <milo@milo.name> on 2015-11-22

* it.po:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoReverted r192734. It broke several Production builds.
aestes@apple.com [Sun, 22 Nov 2015 20:48:45 +0000 (20:48 +0000)]
Reverted r192734. It broke several Production builds.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: Clean up FolderizedTreeElement folder settings
mattbaker@apple.com [Sun, 22 Nov 2015 19:24:11 +0000 (19:24 +0000)]
Web Inspector: Clean up FolderizedTreeElement folder settings
https://bugs.webkit.org/show_bug.cgi?id=151539

Reviewed by Brian Burg.

The expanded state for each folder was stored as an external property on the folder
tree element. Now FolderizedTreeElement keeps a map of folders to settings.

* UserInterface/Views/FolderizedTreeElement.js:
(WebInspector.FolderizedTreeElement):
(WebInspector.FolderizedTreeElement.prototype.removeChildren):
(WebInspector.FolderizedTreeElement.prototype._parentTreeElementForRepresentedObject):
(WebInspector.FolderizedTreeElement.prototype._folderTreeElementExpandedStateChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoTeach MiniBrowser how to enable the mock content filter
aestes@apple.com [Sun, 22 Nov 2015 17:45:50 +0000 (17:45 +0000)]
Teach MiniBrowser how to enable the mock content filter
https://bugs.webkit.org/show_bug.cgi?id=151540

Reviewed by Andreas Kling.

Source/WebCore:

Moved the implementation of MockContentFilterEnabler from TestWebKitAPI to here, renamed it to
WebMockContentFilterEnabler, and made it compatible with the legacy Objective-C runtime. Renamed Decision and
DecisionPoint to WebMockContentFilterDecision and WebMockContentFilterDecisionPoint, and changed them from enum
classes to CF_ENUMs so that they can be used by both C++ and Objective-C source files.

* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSMockContentFilterSettingsCustom.cpp:
(WebCore::JSMockContentFilterSettings::decisionPoint):
(WebCore::JSMockContentFilterSettings::setDecisionPoint):
(WebCore::toDecision):
(WebCore::JSMockContentFilterSettings::decision):
(WebCore::JSMockContentFilterSettings::setDecision):
(WebCore::JSMockContentFilterSettings::unblockRequestDecision):
(WebCore::JSMockContentFilterSettings::setUnblockRequestDecision):
(WebCore::toJSValue): Deleted.
* testing/MockContentFilter.cpp:
(WebCore::MockContentFilter::willSendRequest):
(WebCore::MockContentFilter::responseReceived):
(WebCore::MockContentFilter::addData):
(WebCore::MockContentFilter::finishedAddingData):
(WebCore::MockContentFilter::unblockHandler):
(WebCore::MockContentFilter::maybeDetermineStatus):
* testing/MockContentFilter.h:
* testing/MockContentFilterEnabler.h: Added.
* testing/MockContentFilterEnabler.mm: Added.
(-[WebMockContentFilterEnabler initWithDecision:decisionPoint:blockedString:]):
(-[WebMockContentFilterEnabler initWithCoder:]):
(-[WebMockContentFilterEnabler encodeWithCoder:]):
(-[WebMockContentFilterEnabler enable]):
(-[WebMockContentFilterEnabler dealloc]):
(+[WebMockContentFilterEnabler supportsSecureCoding]):
(-[WebMockContentFilterEnabler copyWithZone:]):
* testing/MockContentFilterSettings.h:
(WebCore::MockContentFilterSettings::decisionPoint):
(WebCore::MockContentFilterSettings::setDecisionPoint):
(WebCore::MockContentFilterSettings::decision):
(WebCore::MockContentFilterSettings::setDecision):
(WebCore::MockContentFilterSettings::unblockRequestDecision):
(WebCore::MockContentFilterSettings::setUnblockRequestDecision):

Tools:

* MiniBrowser/Configurations/Base.xcconfig: Added WebCoreTestSupport and WTF to HEADER_SEARCH_PATHS.
* MiniBrowser/Configurations/MiniBrowser.xcconfig: Linked against libWebCoreTestSupport.
* MiniBrowser/Configurations/MiniBrowserBundle.xcconfig: Ditto.
* MiniBrowser/MiniBrowser.xcodeproj/project.pbxproj: Added blocked-page.html.
* MiniBrowser/MiniBrowserWebProcessPlugIn.m:
(-[MiniBrowserWebProcessPlugIn webProcessPlugIn:initializeWithObject:]): Observed the WebMockContentFilterEnabler key path.
(-[MiniBrowserWebProcessPlugIn dealloc]):
(-[MiniBrowserWebProcessPlugIn observeValueForKeyPath:ofObject:change:context:]): Stored the value in _contentFilterEnabler.
* MiniBrowser/blocked-page.html: Added.
* MiniBrowser/mac/AppDelegate.m:
(defaultConfiguration): Created a _WKProcessPoolConfiguration with MiniBrowser.wkbundle as the injected bundle.
* MiniBrowser/mac/BrowserWindowController.h:
* MiniBrowser/mac/BrowserWindowController.m:
(+[BrowserWindowController contentFilteringBlockedString]): Returned an NSString containing blocked-page.html.
* MiniBrowser/mac/SettingsController.h:
* MiniBrowser/mac/SettingsController.m:
(-[SettingsController _populateMenu]): Added [self _contentFilteringMenuItem] to the Settings menu.
(-[SettingsController validateMenuItem:]): Validated the new menu items. Disabled Decision and Decision Point items if filtering is disabled.
(-[SettingsController _contentFilteringMenuItem]): Returned a new Content Filtering menu item with a submenu.
(-[SettingsController toggleContentFilteringEnabled:]): Toggled ContentFilteringEnabledKey.
(-[SettingsController contentFilteringEnabled]): Returned value of ContentFilteringEnabledKey.
(-[SettingsController setContentFilteringDecision:]): Set ContentFilterDecisionKey and called -[BrowserAppDelegate didChangeSettings].
(-[SettingsController contentFilteringDecision]): Returned the value of ContentFilterDecisionKey as a WebMockContentFilterDecision.
Defaulted to WebMockContentFilterDecisionAllow if the value is invalid.
(-[SettingsController setContentFilteringDecisionPoint:]): Set ContentFilterDecisionPointKey and called -[BrowserAppDelegate didChangeSettings].
(-[SettingsController contentFilteringDecisionPoint]): Returned the value of ContentFilterDecisionPointKey as a WebMockContentFilterDecisionPoint.
Defaulted to WebMockContentFilterDecisionPointAfterWillSendRequest if the value is invalid.
* MiniBrowser/mac/WK1BrowserWindowController.h:
* MiniBrowser/mac/WK1BrowserWindowController.m:
(-[WK1BrowserWindowController dealloc]):
(-[WK1BrowserWindowController didChangeSettings]): Created a WebMockContentFilterEnabler with the current settings and stored it in _contentFilterEnabler.
* MiniBrowser/mac/WK2BrowserWindowController.m:
(-[WK2BrowserWindowController didChangeSettings]): Created a WebMockContentFilterEnabler and set it as the object for the eponymous bundle parameter.
* TestWebKitAPI/Configurations/TestWebKitAPI.xcconfig: Linked against libWebCoreTestSupport.
* TestWebKitAPI/Tests/WebKit2Cocoa/ContentFiltering.mm:
(configurationWithContentFilterSettings): Converted to use WebMockContentFilterEnabler, WebMockContentFilterDecision, and WebMockContentFilterDecisionPoint.
(TEST): Ditto.
(downloadTest): Ditto.
(+[MockContentFilterEnabler supportsSecureCoding]): Deleted.
(-[MockContentFilterEnabler copyWithZone:]): Deleted.
(-[MockContentFilterEnabler initWithCoder:]): Deleted.
(-[MockContentFilterEnabler initWithDecision:decisionPoint:]): Deleted.
(-[MockContentFilterEnabler encodeWithCoder:]): Deleted.
* TestWebKitAPI/Tests/WebKit2Cocoa/ContentFilteringPlugIn.mm:
(-[ContentFilteringPlugIn webProcessPlugIn:initializeWithObject:]): Converted to use WebMockContentFilterEnabler.
(-[ContentFilteringPlugIn dealloc]): Ditto.
(-[ContentFilteringPlugIn observeValueForKeyPath:ofObject:change:context:]): Ditto.
(+[MockContentFilterEnabler supportsSecureCoding]): Deleted.
(-[MockContentFilterEnabler copyWithZone:]): Deleted.
(-[MockContentFilterEnabler initWithCoder:]): Deleted.
(-[MockContentFilterEnabler dealloc]): Deleted.
(-[MockContentFilterEnabler encodeWithCoder:]): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd a test to make sure that font-variant: inherit works correctly.
mmaxfield@apple.com [Sun, 22 Nov 2015 17:19:38 +0000 (17:19 +0000)]
Add a test to make sure that font-variant: inherit works correctly.

Unreviewed.

* fast/text/font-variant-inherit-expected.html: Added.
* fast/text/font-variant-inherit.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFont selection should not consult font-variant property
mmaxfield@apple.com [Sun, 22 Nov 2015 17:06:35 +0000 (17:06 +0000)]
Font selection should not consult font-variant property
https://bugs.webkit.org/show_bug.cgi?id=151537

Reviewed by Simon Fraser.

Source/WebCore:

In section 4.7 of the CSS Fonts Level 3 spec, it says "[The font-variant and
font-feature-settings] do not affect font selection."

All the other browsers (Chrome, Firefox, and Edge) all obey the spec here. We
are the only one who misbehaves. This patch aligns our behavior with the other
browsers.

Test: fast/text/font-selection-font-variant.html

* css/CSSFontSelector.cpp:
(WebCore::computeTraitsMask): Deleted.
(WebCore::compareFontFaces): Deleted.
(WebCore::CSSFontSelector::getFontFace): Deleted.
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
(WebCore::isKeywordPropertyID):
(WebCore::CSSParser::createFontFaceRule):
(WebCore::CSSParser::CSSParser): Deleted.
(WebCore::CSSParser::parseValue): Deleted.
(WebCore::CSSParser::parseDeclaration): Deleted.
(WebCore::CSSParser::clearProperties): Deleted.
(WebCore::CSSParser::parseFontVariant): Deleted.
(WebCore::CSSParser::createStyleRule): Deleted.
(WebCore::CSSParser::deleteFontFaceOnlyValues): Deleted.
* css/CSSParser.h:
* platform/graphics/FontDescription.cpp:
(WebCore::FontDescription::traitsMask): Deleted.
* platform/graphics/win/FontCacheWin.cpp:
(WebCore::traitsInFamilyEnumProc):
* platform/graphics/cocoa/FontCacheCoreText.cpp:
(WebCore::toTraitsMask):
* platform/text/TextFlags.h:

LayoutTests:

* fast/text/font-selection-font-variant-expected.html: Added.
* fast/text/font-selection-font-variant.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, rolling out r192727.
carlosgc@webkit.org [Sun, 22 Nov 2015 10:48:48 +0000 (10:48 +0000)]
Unreviewed, rolling out r192727.

It made the selections transparent again and broke
/webkit2/WebKitWebView/snapshot

Reverted changeset:

"[GTK] RenderThemeGtk::platformActiveSelectionBackgroundColor,
et. al. should not clobber state of cached GtkStyleContexts"
https://bugs.webkit.org/show_bug.cgi?id=151533
http://trac.webkit.org/changeset/192727

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed. Add missing inspector file to GTK+ compilation.
carlosgc@webkit.org [Sun, 22 Nov 2015 10:33:34 +0000 (10:33 +0000)]
Unreviewed. Add missing inspector file to GTK+ compilation.

* PlatformGTK.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Some unit tests fail when using the network process
carlosgc@webkit.org [Sun, 22 Nov 2015 09:00:45 +0000 (09:00 +0000)]
[GTK] Some unit tests fail when using the network process
https://bugs.webkit.org/show_bug.cgi?id=151490

Reviewed by Martin Robinson.

Run the soup server in a separate thread in TestResources test to
avoid deadlocks.

This fixes /webkit2/WebKitWebView/sync-request-on-max-conns and
/webkit2/WebKitWebResource/get-data when using the network process.

* TestWebKitAPI/Tests/WebKit2Gtk/TestResources.cpp:
(beforeAll): Create the WebKitTestServer with ServerRunInThread flag.
* TestWebKitAPI/gtk/WebKit2Gtk/WebKitTestServer.cpp:
(WebKitTestServer::WebKitTestServer): When ServerRunInThread is
present, create a WorkQueue to run the server.
(WebKitTestServer::run): Run the server in the work queue if it
has been created.
* TestWebKitAPI/gtk/WebKit2Gtk/WebKitTestServer.h: Convert server
type into server options as flags.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] ImageDiff should normalize the diff image
carlosgc@webkit.org [Sun, 22 Nov 2015 08:53:56 +0000 (08:53 +0000)]
[GTK] ImageDiff should normalize the diff image
https://bugs.webkit.org/show_bug.cgi?id=151261

Reviewed by Sergio Villar Senin.

* ImageDiff/gtk/ImageDiff.cpp:
(readPixbufFromStdin): Fix memory leak.
(differenceImageFromDifferenceBuffer): Normalize diff buffer.
(calculateDifference): Pass max distance to differenceImageFromDifferenceBuffer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] RenderThemeGtk::platformActiveSelectionBackgroundColor, et. al. should not...
mcatanzaro@igalia.com [Sun, 22 Nov 2015 08:50:40 +0000 (08:50 +0000)]
[GTK] RenderThemeGtk::platformActiveSelectionBackgroundColor, et. al. should not clobber state of cached GtkStyleContexts
https://bugs.webkit.org/show_bug.cgi?id=151533

Reviewed by Carlos Garcia Campos.

platformActiveSelectionBackgroundColor(), platformInactiveSelectionBackgroundColor(), etc.
are const functions intended only to return a color used for painting, but since r174929
they also change the state of the cached style contexts we use for GTK_TYPE_ENTRY and
GTK_TYPE_TREE_VIEW. That's wrong; those style contexts should not have any state set. This
could cause theme colors returned by those GtkStyleContexts to change unexpectedly,
depending on whether the state is explicitly set before each use, or whether the theme
actually uses the states.

This didn't cause any regression only because every place using these style contexts
explicitly sets the state of the style contexts before use. In fact, the GtkTreeView style
context is not used anywhere else, and the GtkEntry style context is only used in
paintTextField, which does set the state before use (and then reverts it using
save/restore), so this cannot have broken anything in practice. But it's a landmine waiting
for the next programmer to trip it.

Fix this with a gtk_style_context_save()/gtk_style_context_restore() pair.

* rendering/RenderThemeGtk.cpp:
(WebCore::styleColor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoTiny cleanup in ComplexTextController::collectComplexTextRuns()
mmaxfield@apple.com [Sat, 21 Nov 2015 23:06:46 +0000 (23:06 +0000)]
Tiny cleanup in ComplexTextController::collectComplexTextRuns()
https://bugs.webkit.org/show_bug.cgi?id=151534

Reviewed by Zalan Bujtas.

The isMissingGlyph boolean is completely unnecessary. Its entire
responsiblity is duplicated by the "font" pointer.

No new tests because there is no behavior change.

* platform/graphics/mac/ComplexTextController.cpp:
(WebCore::ComplexTextController::collectComplexTextRuns):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago== Rolled over to ChangeLog-2015-11-21 ==
mcatanzaro@igalia.com [Sat, 21 Nov 2015 20:02:34 +0000 (20:02 +0000)]
== Rolled over to ChangeLog-2015-11-21 ==

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Off-by-one error in getStyleContext()
mcatanzaro@igalia.com [Sat, 21 Nov 2015 17:00:00 +0000 (17:00 +0000)]
[GTK] Off-by-one error in getStyleContext()
https://bugs.webkit.org/show_bug.cgi?id=151524

Reviewed by Carlos Garcia Campos.

GtkWidgetPath* path = gtk_widget_path_new();
gtk_widget_path_append_type(path, widgetType);
// ...
gtk_widget_path_iter_add_class(path, 0, GTK_STYLE_CLASS_BUTTON);
gtk_widget_path_iter_add_class(path, 1, "text-button");

Only one widget type was appended to the widget path, so the maximum valid index is 0. This
code means to add both style classes to the first widget type in the widget path, so the
second call should use index 0 rather than index 1.

This caused no bug in practice, because when the index is invalid,
gtk_widget_path_iter_add_class() automatically changes the index to the last valid position
in the widget path -- in this case, 0. This is routinely done with -1 as a convention for
specifying the last position in the widget path.

* rendering/RenderThemeGtk.cpp:
(WebCore::getStyleContext):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Warning spam from GtkStyleContext
mcatanzaro@igalia.com [Sat, 21 Nov 2015 16:57:47 +0000 (16:57 +0000)]
[GTK] Warning spam from GtkStyleContext
https://bugs.webkit.org/show_bug.cgi?id=151520

Reviewed by Carlos Garcia Campos.

Audit every use of gtk_style_context_get_* to fix compatibility with GTK+ 3.19. Some of
these were already fine and are only changed for clarity.

Company:  gtk_style_context_get() (and _get_padding/border/color()) should only ever be
          called with the same state as gtk_style_context_get_state()
Company:  usually that's a simple replacing of the old state (like in the trace you posted)
Company:  sometimes it requires calling gtk_style_context_set_sate() with the right state
          first
Company:  and in very rare cases it needs a gtk_style_context_save() before the set_state(),
          too

* platform/gtk/ScrollbarThemeGtk.cpp:
(WebCore::adjustRectAccordingToMargin):
* rendering/RenderThemeGtk.cpp:
(gtk_css_section_print):
(WebCore::getStyleContext):
(WebCore::RenderThemeGtk::initMediaColors):
(WebCore::renderButton):
(WebCore::getComboBoxMetrics):
(WebCore::RenderThemeGtk::paintMenuList):
(WebCore::RenderThemeGtk::paintTextField):
(WebCore::RenderThemeGtk::paintProgressBar):
(WebCore::spinButtonArrowSize):
(WebCore::RenderThemeGtk::adjustInnerSpinButtonStyle):
(WebCore::styleColor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION(r192536): Null pointer dereference in JSPropertyNameEnumerator::visitChild...
akling@apple.com [Sat, 21 Nov 2015 06:07:40 +0000 (06:07 +0000)]
REGRESSION(r192536): Null pointer dereference in JSPropertyNameEnumerator::visitChildren().
<https://webkit.org/b/151495>

Reviewed by Mark Lam.

Source/JavaScriptCore:

The copied space allocation in JSPropertyNameEnumerator::finishCreation()
may end up triggering a GC, and so JSPropertyNameEnumerator::visitChildren()
would get called while m_propertyNames was still null.

This patch fixes that by having visitChildren() check for pointer nullity
instead of the number of names, since that is non-zero even before the
allocation is made.

Added a test that induces GC during JSPropertyNameEnumerator construction
to cover this bug.

Test: property-name-enumerator-gc-151495.js

* runtime/JSPropertyNameEnumerator.cpp:
(JSC::JSPropertyNameEnumerator::visitChildren):

LayoutTests:

* js/property-name-enumerator-gc-151495.html: Added.
* js/property-name-enumerator-gc-151495-expected.txt: Added.
* js/script-tests/property-name-enumerator-gc-151495.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoGC timers should carry on gracefully when Heap claims it grew from GC.
akling@apple.com [Sat, 21 Nov 2015 04:40:40 +0000 (04:40 +0000)]
GC timers should carry on gracefully when Heap claims it grew from GC.
<https://webkit.org/b/151521>

Reviewed by Mark Lam.

TL;DR the Heap "extra memory" reporting APIs are hard to use 100% correctly
and GC scheduling shouldn't break if someone makes a mistake with it.

The JSC::Heap allows you to report an extra memory cost for any GC object.
This is reported first when allocating the memory, and then each time the
object is visited during the marking phase.

When reporting an allocation, it's added to the Heap's "bytes allocated in
this cycle" counter. This contributes to the computed heap size at the start
of a collection.

When visiting a GC object that reports extra memory, it's added to the Heap's
"extra memory visited in this collection" counter. This contributes to the
computed heap size at the end of a collection.

As you can see, this means that visiting more memory than we said we allocated
can lead to the Heap thinking it's bigger after a collection than it was before.

Clients of this API do some sketchy things to compute costs, for instance
StringImpl cost is determined by dividing the number of bytes used for the
characters, and dividing it by the StringImpl's ref count. Since a JSString
could be backed by any StringImpl, any code that modifies a StringImpl's
ref count during collection will change the extra memory reported by all
JSString objects that wrap that StringImpl.

So anyways...

The object death rate, which is the basis for when to schedule the next
collection is computed like so:

    deathRate = (sizeBeforeGC - sizeAfterGC) / sizeBeforeGC

This patch adds a safety mechanism that returns a zero death rate when the Heap
claims it grew from collection.

* heap/EdenGCActivityCallback.cpp:
(JSC::EdenGCActivityCallback::deathRate):
* heap/FullGCActivityCallback.cpp:
(JSC::FullGCActivityCallback::deathRate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoModern IDB: After versionchange transactions complete, fire onsuccess on the original...
beidson@apple.com [Sat, 21 Nov 2015 03:57:19 +0000 (03:57 +0000)]
Modern IDB: After versionchange transactions complete, fire onsuccess on the original IDBOpenDBRequest
https://bugs.webkit.org/show_bug.cgi?id=151522

Reviewed by Alex Christensen.

Source/WebCore:

Test: storage/indexeddb/modern/opendatabase-success-after-versionchange.html (And changes to other existing tests)

* Modules/indexeddb/client/IDBDatabaseImpl.cpp:
(WebCore::IDBClient::IDBDatabase::startVersionChangeTransaction):
* Modules/indexeddb/client/IDBDatabaseImpl.h:

* Modules/indexeddb/client/IDBOpenDBRequestImpl.cpp:
(WebCore::IDBClient::IDBOpenDBRequest::fireSuccessAfterVersionChangeCommit):
(WebCore::IDBClient::IDBOpenDBRequest::onUpgradeNeeded):
* Modules/indexeddb/client/IDBOpenDBRequestImpl.h:

* Modules/indexeddb/client/IDBTransactionImpl.cpp:
(WebCore::IDBClient::IDBTransaction::create):
(WebCore::IDBClient::IDBTransaction::IDBTransaction):
(WebCore::IDBClient::IDBTransaction::dispatchEvent):
* Modules/indexeddb/client/IDBTransactionImpl.h:

LayoutTests:

* storage/indexeddb/modern/deletedatabase-1-expected.txt:
* storage/indexeddb/modern/deletedatabase-1.html:
* storage/indexeddb/modern/deletedatabase-2-expected.txt:
* storage/indexeddb/modern/deletedatabase-2.html:
* storage/indexeddb/modern/opendatabase-success-after-versionchange-expected.txt: Added.
* storage/indexeddb/modern/opendatabase-success-after-versionchange.html: Added.
* storage/indexeddb/modern/opendatabase-versions-expected.txt:
* storage/indexeddb/modern/opendatabase-versions.html:
* storage/indexeddb/modern/versionchange-event-expected.txt:
* storage/indexeddb/modern/versionchange-event.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: Timeline does not immediately update current time after switching...
commit-queue@webkit.org [Sat, 21 Nov 2015 03:31:08 +0000 (03:31 +0000)]
Web Inspector: Timeline does not immediately update current time after switching tabs while recording
https://bugs.webkit.org/show_bug.cgi?id=151528

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2015-11-20
Reviewed by Timothy Hatcher.

* UserInterface/Views/TimelineRecordingContentView.js:
(WebInspector.TimelineRecordingContentView.prototype.shown):
Provide a number so that _startUpdatingCurrentTime doesn't decide
to wait until the next event to update current time.

(WebInspector.TimelineRecordingContentView.prototype._update):
When switching tabs we stop updating and hit this code. However
we want to keep _lastUpdateTimestamp so that if we switch back
to the tab we can re-compute the current time. So don't clear it
in cases where the view was hidden, as it might need to use
it when it is shown again.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: Remove "Show only resources with breakpoints" filter button
mattbaker@apple.com [Sat, 21 Nov 2015 03:29:31 +0000 (03:29 +0000)]
Web Inspector: Remove "Show only resources with breakpoints" filter button
https://bugs.webkit.org/show_bug.cgi?id=151517

Reviewed by Timothy Hatcher.

Removed filter button in preparation for https://bugs.webkit.org/show_bug.cgi?id=151119.

* UserInterface/Views/DebuggerSidebarPanel.js:
(WebInspector.DebuggerSidebarPanel.showResourcesWithBreakpointsOnlyFilterFunction): Deleted.
(WebInspector.DebuggerSidebarPanel): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoBuild fix for old version of PHP.
rniwa@webkit.org [Sat, 21 Nov 2015 03:16:11 +0000 (03:16 +0000)]
Build fix for old version of PHP.

* public/api/runs.php:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoPerf dashboard's should not include results more than 366 days old in JSON
rniwa@webkit.org [Sat, 21 Nov 2015 03:12:14 +0000 (03:12 +0000)]
Perf dashboard's should not include results more than 366 days old in JSON
https://bugs.webkit.org/show_bug.cgi?id=151529

Reviewed by Timothy Hatcher.

Don't return results more than 366 days old in /api/runs/ JSON API.
This is a ~5% runtime improvement and reduces the JSON file size by 20-50% in the internal perf dashboard.

* public/api/runs.php:
(main): Added the support for "?noResults" to avoid echoing results. This is useful for debugging.
Also instantiate RunsGenerator before issuing the query to find all configurations so that the runtime cost
of doing so will be included in elapsedTime.
(RunsGenerator::fetch_runs): Skip a row when its build and commit times are more than 366 days old.
(RunsGenerator::format_run): Takes build_time and revisions as arguments since fetch_runs uses them now.
(RunsGenerator::parse_revisions_array): Compute the max of commit times.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAnimometer: graphs should not do interpolation
simon.fraser@apple.com [Sat, 21 Nov 2015 03:05:50 +0000 (03:05 +0000)]
Animometer: graphs should not do interpolation
https://bugs.webkit.org/show_bug.cgi?id=151526

Reviewed by Simon Fraser.

Having the graphs do interpolation is misleading, because you can't see the actual data.

Also remove "shape-rendering: crispEdges;" so the lines get antialiased.

* Animometer/runner/resources/animometer.css:
(section#test-graph > data > svg): Deleted.
* Animometer/runner/resources/graph.js:
(graph): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, add watchlist for PerformanceTests/Animometer and add myself.
said@apple.com [Sat, 21 Nov 2015 02:51:51 +0000 (02:51 +0000)]
Unreviewed, add watchlist for PerformanceTests/Animometer and add myself.

* Scripts/webkitpy/common/config/watchlist:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, mark fast/replaced/replaced-breaking.html as failing on ElCapitan-wk1.
cdumez@apple.com [Sat, 21 Nov 2015 02:29:23 +0000 (02:29 +0000)]
Unreviewed, mark fast/replaced/replaced-breaking.html as failing on ElCapitan-wk1.

The test gives a slightly different output of this configuration only
and I don't think we can have platform-specific test expectations for
this particular configuration.

* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoMore deviceRGB color cleanup
simon.fraser@apple.com [Sat, 21 Nov 2015 01:51:48 +0000 (01:51 +0000)]
More deviceRGB color cleanup
https://bugs.webkit.org/show_bug.cgi?id=151523
<rdar://problem/23638597>

Reviewed by Tim Horton.

Replace calls to deviceRGBColorSpaceRef() with sRGBColorSpaceRef(), and use
sRGBColorSpaceRef() in a few places that were manually creating the colorspace.

Also use cachedCGColor() in a more places that were manually constructing CGColorRefs
from Colors.

Source/WebCore:

* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::createImageForTimeInRect):
(WebCore::createImageFromPixelBuffer):
* platform/graphics/ca/cocoa/PlatformCAFiltersCocoa.mm:
* platform/graphics/ca/cocoa/PlatformCALayerCocoa.mm:
(PlatformCALayerCocoa::setBackgroundColor):
(PlatformCALayerCocoa::setBorderColor):
* platform/graphics/ca/cocoa/WebSystemBackdropLayer.mm:
(-[WebLightSystemBackdropLayer init]):
(-[WebDarkSystemBackdropLayer init]):
* platform/graphics/cg/GradientCG.cpp:
(WebCore::Gradient::platformGradient):
* platform/graphics/cg/GraphicsContext3DCG.cpp:
(WebCore::GraphicsContext3D::ImageExtractor::extractImage):
(WebCore::GraphicsContext3D::paintToCanvas):
* platform/graphics/cg/ImageBufferCG.cpp:
(WebCore::ImageBuffer::copyImage):
(WebCore::ImageBuffer::toDataURL):
(WebCore::ImageDataToDataURL):
* platform/graphics/mac/GraphicsContextMac.mm:
(WebCore::linearRGBColorSpaceRef):
* platform/graphics/mac/WebGLLayer.mm:
(-[WebGLLayer copyImageSnapshotWithColorSpace:]):
* platform/mediastream/mac/AVVideoCaptureSource.mm:
(WebCore::AVVideoCaptureSource::currentFrameCGImage):
* rendering/RenderThemeIOS.mm:
(WebCore::drawRadialGradient):
* rendering/RenderThemeMac.mm:
(WebCore::RenderThemeMac::paintMenuListButtonGradients):
(WebCore::RenderThemeMac::paintSliderTrack):

Source/WebKit2:

* Shared/cg/ShareableBitmapCG.cpp:
(WebKit::ShareableBitmap::createGraphicsContext):
(WebKit::ShareableBitmap::createCGImage):
* Shared/mac/RemoteLayerTreePropertyApplier.mm:
(WebKit::cgColorFromColor):
* UIProcess/mac/RemoteLayerTreeDrawingAreaProxy.mm:
(WebKit::RemoteLayerTreeDrawingAreaProxy::initializeDebugIndicator):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRenaming PageCache suspension code to support more reasons for suspension.
commit-queue@webkit.org [Sat, 21 Nov 2015 01:42:11 +0000 (01:42 +0000)]
Renaming PageCache suspension code to support more reasons for suspension.
https://bugs.webkit.org/show_bug.cgi?id=151527

Patch by Katlyn Graff <kgraff@apple.com> on 2015-11-20
Reviewed by Ryosuke Niwa.

No new tests because this is simply a refactor.

Renamed Element:: and Document:: documentWillSuspendForPageCache(),
documentDidResumeFromPageCache(),
registerForPageCacheSuspensionCallbacks(),
unregisterForPageCacheSuspensionCallbacks() to prepare to support
alternate reasons for document-level suspension.

* dom/Document.cpp:
(WebCore::Document::suspend):
(WebCore::Document::resume):
(WebCore::Document::registerForDocumentSuspensionCallbacks):
(WebCore::Document::unregisterForDocumentSuspensionCallbacks):
(WebCore::Document::documentWillSuspendForPageCache): Deleted.
(WebCore::Document::documentDidResumeFromPageCache): Deleted.
(WebCore::Document::registerForPageCacheSuspensionCallbacks): Deleted.
(WebCore::Document::unregisterForPageCacheSuspensionCallbacks): Deleted.
* dom/Document.h:
* dom/Element.h:
(WebCore::Element::prepareForDocumentSuspension):
(WebCore::Element::resumeFromDocumentSuspension):
(WebCore::Element::documentWillSuspendForPageCache): Deleted.
(WebCore::Element::documentDidResumeFromPageCache): Deleted.
* history/CachedFrame.cpp:
(WebCore::CachedFrameBase::restore):
(WebCore::CachedFrame::CachedFrame):
* html/HTMLFormElement.cpp:
(WebCore::HTMLFormElement::~HTMLFormElement):
(WebCore::HTMLFormElement::parseAttribute):
(WebCore::HTMLFormElement::resumeFromDocumentSuspension):
(WebCore::HTMLFormElement::didMoveToNewDocument):
(WebCore::HTMLFormElement::documentDidResumeFromPageCache): Deleted.
* html/HTMLFormElement.h:
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::~HTMLInputElement):
(WebCore::HTMLInputElement::registerForSuspensionCallbackIfNeeded):
(WebCore::HTMLInputElement::unregisterForSuspensionCallbackIfNeeded):
(WebCore::HTMLInputElement::resumeFromDocumentSuspension):
(WebCore::HTMLInputElement::prepareForDocumentSuspension):
(WebCore::HTMLInputElement::didMoveToNewDocument):
(WebCore::HTMLInputElement::documentDidResumeFromPageCache): Deleted.
(WebCore::HTMLInputElement::documentWillSuspendForPageCache): Deleted.
* html/HTMLInputElement.h:
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::registerWithDocument):
(WebCore::HTMLMediaElement::unregisterWithDocument):
(WebCore::HTMLMediaElement::prepareForDocumentSuspension):
(WebCore::HTMLMediaElement::resumeFromDocumentSuspension):
(WebCore::HTMLMediaElement::documentWillSuspendForPageCache): Deleted.
(WebCore::HTMLMediaElement::documentDidResumeFromPageCache): Deleted.
* html/HTMLMediaElement.h:
* html/HTMLPlugInImageElement.cpp:
(WebCore::HTMLPlugInImageElement::~HTMLPlugInImageElement):
(WebCore::HTMLPlugInImageElement::createElementRenderer):
(WebCore::HTMLPlugInImageElement::didMoveToNewDocument):
(WebCore::HTMLPlugInImageElement::prepareForDocumentSuspension):
(WebCore::HTMLPlugInImageElement::resumeFromDocumentSuspension):
(WebCore::HTMLPlugInImageElement::documentWillSuspendForPageCache): Deleted.
(WebCore::HTMLPlugInImageElement::documentDidResumeFromPageCache): Deleted.
* html/HTMLPlugInImageElement.h:
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::commitProvisionalLoad):
* svg/SVGSVGElement.cpp:
(WebCore::SVGSVGElement::SVGSVGElement):
(WebCore::SVGSVGElement::~SVGSVGElement):
(WebCore::SVGSVGElement::didMoveToNewDocument):
(WebCore::SVGSVGElement::prepareForDocumentSuspension):
(WebCore::SVGSVGElement::resumeFromDocumentSuspension):
(WebCore::SVGSVGElement::documentWillSuspendForPageCache): Deleted.
(WebCore::SVGSVGElement::documentDidResumeFromPageCache): Deleted.
* svg/SVGSVGElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, rolling out r192139.
cdumez@apple.com [Sat, 21 Nov 2015 00:20:53 +0000 (00:20 +0000)]
Unreviewed, rolling out r192139.
https://bugs.webkit.org/show_bug.cgi?id=151525

as it may cause WKTR to crash when the Flash plugin is
installed (Requested by cdumez on #webkit).

Reverted changeset:

"Unreviewed gardening, update fast/replaced/replaced-
breaking.html"
http://trac.webkit.org/changeset/192139

Patch by Commit Queue <commit-queue@webkit.org> on 2015-11-20

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFix the Windows build.
simon.fraser@apple.com [Sat, 21 Nov 2015 00:04:20 +0000 (00:04 +0000)]
Fix the Windows build.

* platform/graphics/cg/IOSurfacePool.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoNew JSC tests introduced in r192664 fail on ARM.
mark.lam@apple.com [Fri, 20 Nov 2015 23:45:40 +0000 (23:45 +0000)]
New JSC tests introduced in r192664 fail on ARM.
https://bugs.webkit.org/show_bug.cgi?id=151485

Reviewed by Geoffrey Garen.

The newly added tests are exposing some pre-existing bugs.  The bugs are tracked here:
    https://bugs.webkit.org/show_bug.cgi?id=151514
    https://bugs.webkit.org/show_bug.cgi?id=151515

Skipping the tests for now.

* tests/stress/op_div.js:
* tests/stress/op_rshift.js:
* tests/stress/op_urshift.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWent a bit too far, revert part of the previous patch.
ap@apple.com [Fri, 20 Nov 2015 23:37:23 +0000 (23:37 +0000)]
Went a bit too far, revert part of the previous patch.

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Images/Mavericks.png: Added.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Images/Mavericks@2x.png: Added.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Dashboard.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Styles/Main.css:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove Mavericks bots.
ap@apple.com [Fri, 20 Nov 2015 23:16:44 +0000 (23:16 +0000)]
Remove Mavericks bots.

Rubber-stamped by Lucas Forschler.

* BuildSlaveSupport/build.webkit.org-config/config.json:
* BuildSlaveSupport/build.webkit.org-config/mastercfg_unittest.py:
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Images/Mavericks.png: Removed.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Images/Mavericks@2x.png: Removed.
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BubbleQueueServer.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Dashboard.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Styles/Main.css:
* BuildSlaveSupport/build.webkit.org-config/templates/root.html:
* BuildSlaveSupport/build.webkit.org-config/wkbuild.py:
* BuildSlaveSupport/build.webkit.org-config/wkbuild_unittest.py:
* Scripts/copy-webkitlibraries-to-product-directory:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: Add support for Gradients in the Visual sidebar background editor
commit-queue@webkit.org [Fri, 20 Nov 2015 23:16:18 +0000 (23:16 +0000)]
Web Inspector: Add support for Gradients in the Visual sidebar background editor
https://bugs.webkit.org/show_bug.cgi?id=150494

Patch by Devin Rousso <dcrousso+webkit@gmail.com> on 2015-11-20
Reviewed by Timothy Hatcher.

Allows the editors in the Visual sidebar Background Style section to
work with gradients and data URIs.

* Localizations/en.lproj/localizedStrings.js:
* UserInterface/Controllers/CodeMirrorGradientEditingController.css: Deleted.
Some styling was reused in VisualStyleBackgroundPicker.css.

* UserInterface/Controllers/CodeMirrorGradientEditingController.js:
(WebInspector.CodeMirrorGradientEditingController):
(WebInspector.CodeMirrorGradientEditingController.prototype.popoverWillPresent):
(WebInspector.CodeMirrorGradientEditingController.prototype.popoverDidPresent):
(WebInspector.CodeMirrorGradientEditingController.prototype._gradientEditorGradientChanged):
(WebInspector.CodeMirrorGradientEditingController.prototype.handleEvent): Deleted.
(WebInspector.CodeMirrorGradientEditingController.prototype.gradientSliderStopsDidChange): Deleted.
(WebInspector.CodeMirrorGradientEditingController.prototype.gradientSliderStopWasSelected): Deleted.
(WebInspector.CodeMirrorGradientEditingController.prototype.dragToAdjustControllerWasAdjustedByAmount): Deleted.
(WebInspector.CodeMirrorGradientEditingController.prototype._handleInputEvent): Deleted.
(WebInspector.CodeMirrorGradientEditingController.prototype._angleInputValueDidChange): Deleted.
(WebInspector.CodeMirrorGradientEditingController.prototype._handleChangeEvent): Deleted.
(WebInspector.CodeMirrorGradientEditingController.prototype._colorPickerColorChanged): Deleted.
(WebInspector.CodeMirrorGradientEditingController.prototype._updateCSSClassForGradientType): Deleted.
Reworked gradient editor popup into WebInspector.GradientEditor.

* UserInterface/Main.html:
* UserInterface/Models/Gradient.js:
(WebInspector.Gradient.fromString):
(WebInspector.Gradient.stopsWithComponents):
(WebInspector.LinearGradient.linearGradientWithComponents):
Removed console.error statements as they didn't do anything but clog the console.

* UserInterface/Views/GradientEditor.css: Added.
(.gradient-editor):
(.gradient-editor.radial-gradient):
(.gradient-editor.editing-color):
(.gradient-editor.radial-gradient.editing-color):
(.gradient-editor > .gradient-type-select):
(.gradient-editor > .gradient-slider):
(.gradient-editor > .color-picker):
(.gradient-editor > .color-picker > .slider):
(.gradient-editor > .color-picker > .brightness):
(.gradient-editor > .color-picker > .opacity):
(.gradient-editor > .gradient-angle):
(.gradient-editor.radial-gradient > .gradient-angle):
(.gradient-editor > .gradient-angle > input):

* UserInterface/Views/GradientEditor.js: Added.
(WebInspector.GradientEditor):
(WebInspector.GradientEditor.prototype.get element):
(WebInspector.GradientEditor.prototype.set gradient):
(WebInspector.GradientEditor.prototype.get gradient):
(WebInspector.GradientEditor.prototype.gradientSliderStopsDidChange):
(WebInspector.GradientEditor.prototype.gradientSliderStopWasSelected):
(WebInspector.GradientEditor.prototype.dragToAdjustControllerWasAdjustedByAmount):
(WebInspector.GradientEditor.prototype._updateCSSClassForGradientType):
(WebInspector.GradientEditor.prototype._gradientTypeChanged):
(WebInspector.GradientEditor.prototype._colorPickerColorChanged):
(WebInspector.GradientEditor.prototype._angleChanged):
(WebInspector.GradientEditor.prototype._angleInputValueDidChange):
New standalone editor for CSS Gradients.

* UserInterface/Views/VisualStyleBackgroundPicker.css: Added.
(.visual-style-property-container.background-picker > .visual-style-property-value-container):
(.visual-style-property-container.background-picker > .visual-style-property-value-container > .color-gradient-swatch):
(.visual-style-property-container.background-picker > .visual-style-property-value-container > .color-gradient-swatch:hover):
(.visual-style-property-container.background-picker > .visual-style-property-value-container > .color-gradient-swatch:active):
(.visual-style-property-container.background-picker > .visual-style-property-value-container > .color-gradient-swatch > span):
(.visual-style-property-container.background-picker > .visual-style-property-value-container.gradient-value > .color-gradient-swatch):
(.visual-style-property-container.background-picker > .visual-style-property-value-container.gradient-value > .color-gradient-swatch + .value-input):
(.visual-style-property-container.background-picker > .visual-style-property-value-container > .value-input):
(.visual-style-property-container.background-picker > .visual-style-property-value-container > .value-input[disabled]):
(.visual-style-property-container.background-picker > .visual-style-property-value-container > .value-type-picker-select):

* UserInterface/Views/VisualStyleBackgroundPicker.js: Added.
(WebInspector.VisualStyleBackgroundPicker):
(WebInspector.VisualStyleBackgroundPicker.prototype.get value):
(WebInspector.VisualStyleBackgroundPicker.prototype.set value):
(WebInspector.VisualStyleBackgroundPicker.prototype.get synthesizedValue):
(WebInspector.VisualStyleBackgroundPicker.prototype.parseValue):
(WebInspector.VisualStyleBackgroundPicker.prototype._updateValueInput):
(WebInspector.VisualStyleBackgroundPicker.prototype._updateGradientSwatch):
(WebInspector.VisualStyleBackgroundPicker.prototype._gradientSwatchClicked):
(WebInspector.VisualStyleBackgroundPicker.prototype._gradientEditorGradientChanged):
(WebInspector.VisualStyleBackgroundPicker.prototype._valueInputValueChanged):
(WebInspector.VisualStyleBackgroundPicker.prototype._keywordSelectMouseDown):
(WebInspector.VisualStyleBackgroundPicker.prototype._handleKeywordChanged):
(WebInspector.VisualStyleBackgroundPicker.prototype._createValueOptions):
(WebInspector.VisualStyleBackgroundPicker.prototype._addAdvancedValues):
(WebInspector.VisualStyleBackgroundPicker.prototype._removeAdvancedValues):
(WebInspector.VisualStyleBackgroundPicker.prototype._toggleTabbingOfSelectableElements):
Visual property editor for the CSS background-image property. Supports
limited keywords, as well as url() and gradients.

* UserInterface/Views/VisualStyleCommaSeparatedKeywordEditor.js:
(WebInspector.VisualStyleCommaSeparatedKeywordEditor.prototype.set value):
Fixed regular expression parsing to not match commas inside parenthesis.

* UserInterface/Views/VisualStyleDetailsPanel.js:
(WebInspector.VisualStyleDetailsPanel.prototype._populateBackgroundStyleSection):

* UserInterface/Views/VisualStylePropertyEditorLink.css:
(.visual-style-property-editor-link:not(.link-all) > .visual-style-property-editor-link-border):
Fixed spacing of link line.

* UserInterface/Views/VisualStyleURLInput.js: Removed.
Reworked into VisualStyleBackgroundPicker.js.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUpdate the patch and the plan files of the graphics benchmark with the latest revision
commit-queue@webkit.org [Fri, 20 Nov 2015 23:14:52 +0000 (23:14 +0000)]
Update the patch and the plan files of the graphics benchmark with the latest revision
https://bugs.webkit.org/show_bug.cgi?id=151503

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2015-11-20
Reviewed by Ryosuke Niwa.

The time for each graphics test has been shortened from 30 seconds to 10
seconds. We need to update the graphics benchmark patch and plan files so
we can get the results faster. Also this will allow running the benchmark
also on the iOS performance bots.

* Scripts/webkitpy/benchmark_runner/data/patches/Animometer.patch:
* Scripts/webkitpy/benchmark_runner/data/plans/animometer.plan:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoA window with a hung tab waits 5s before becoming active
ggaren@apple.com [Fri, 20 Nov 2015 23:10:57 +0000 (23:10 +0000)]
A window with a hung tab waits 5s before becoming active
https://bugs.webkit.org/show_bug.cgi?id=151426

Reviewed by Sam Weinig.

This patch adds an optimization to skip the synchronous web process
message to check for a legacy scrollbar when we know that legacy
scrollbars are not enabled.

(Note that legacy scrollbars don't work quite right, due to
<rdar://problem/23585420> and <rdar://problem/23605296>. Still, I verified
with logging that we do the synchronous message when legacy scrollbars
are enabled.)

A consequence of this change is that we will no longer support
click-to-scroll-while-inactive behavior for scrollbars with custom looks
on systems with modern scrollbars. I spoke with Beth and Dan, and they
agreed that this is a reasonable change to make, since we don't support
click-to-scroll-while-inactive behavior for fully custom scrollbars either,
and since systems with modern scrollbars typically use swipe to scroll.

* UIProcess/Cocoa/WebViewImpl.h: Added some helper functions to explain
the behaviors we're checking for.

* UIProcess/Cocoa/WebViewImpl.mm:
(WebKit::WebViewImpl::mightBeginDragWhileInactive): Factored out from
shouldDelayWindowOrderingForEvent.

(WebKit::WebViewImpl::mightBeginScrollWhileInactive): New function.

(WebKit::WebViewImpl::acceptsFirstMouse): Moved this function next to
shouldDelayWindowOrderingForEvent because their responsibilities are
very similar. Added a fast path check for when we know that we will
not accept first mouse because we can't start a drag or scroll by
clicking while inactive.

(WebKit::WebViewImpl::shouldDelayWindowOrderingForEvent): Refactored
to use the helper function. Behavior unchanged.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoA hung webpage pretends to be responsive if you scroll
ggaren@apple.com [Fri, 20 Nov 2015 23:10:02 +0000 (23:10 +0000)]
A hung webpage pretends to be responsive if you scroll
https://bugs.webkit.org/show_bug.cgi?id=151518

Reviewed by Sam Weinig.

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::sendWheelEvent):
(WebKit::WebPageProxy::didReceiveEvent): Don't treat wheel events as
starting or stopping the responsiveness timer. Wheel events usually
process on the event dispatch thread, which responds even if the main
thread is hung.

Instead, send an out-of-band ping to the main thread to verify that
it is still responsive and we'll be able to paint and respond to clicks
after scrolling.

* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::sendMainThreadPing):
(WebKit::WebProcessProxy::didReceiveMainThreadPing):
* UIProcess/WebProcessProxy.h:
* UIProcess/WebProcessProxy.messages.in: UI process support for pinging
the main thread in the web process.

* WebProcess/WebProcess.cpp:
(WebKit::WebProcess::mainThreadPing):
* WebProcess/WebProcess.h:
* WebProcess/WebProcess.messages.in: Web process support for responding
to pings.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAllow more buffer formats in the IOSurface pool
simon.fraser@apple.com [Fri, 20 Nov 2015 22:55:56 +0000 (22:55 +0000)]
Allow more buffer formats in the IOSurface pool
https://bugs.webkit.org/show_bug.cgi?id=151516

Reviewed by Tim Horton.
Source/WebCore:

Previously IOSurface::create was only looking in the pool for RGBA-format surfaces. Change that to
always look in the pool, and to cache all format types. We include format in the criteria used
to pick a surface from the pool.

* platform/graphics/cg/IOSurfacePool.cpp:
(WebCore::surfaceMatchesParameters):
(WebCore::IOSurfacePool::takeSurface):
(WebCore::IOSurfacePool::shouldCacheFormat):
(WebCore::IOSurfacePool::shouldCacheSurface):
(WebCore::IOSurfacePool::addSurface):
* platform/graphics/cg/IOSurfacePool.h:
* platform/graphics/cocoa/IOSurface.h:
* platform/graphics/cocoa/IOSurface.mm:
(IOSurface::surfaceFromPool):
(IOSurface::create):
(IOSurface::IOSurface):

Source/WebKit2:

Have RemoteLayerBackingStore go through a static function on IOSurface to return
a surface to the pool, rather than knowing about IOSurfacePool directly.

* Shared/mac/RemoteLayerBackingStore.mm:
(WebKit::RemoteLayerBackingStore::Buffer::discard):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Win] Support High DPI drawing with CACFLayers
bfulgham@apple.com [Fri, 20 Nov 2015 22:35:01 +0000 (22:35 +0000)]
[Win] Support High DPI drawing with CACFLayers
https://bugs.webkit.org/show_bug.cgi?id=147242
<rdar://problem/19861992>

Reviewed by Simon Fraser.

* platform/graphics/ca/win/WKCACFViewLayerTreeHost.cpp:
(WebCore::WKCACFViewLayerTreeHost::initializeContext): Set correct content scale factor
for current screen, and apply an appropriate base transform to the CACFLayer so drawing
operations are done properly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoB3 should have a Select opcode
fpizlo@apple.com [Fri, 20 Nov 2015 22:31:19 +0000 (22:31 +0000)]
B3 should have a Select opcode
https://bugs.webkit.org/show_bug.cgi?id=150762

Reviewed by Benjamin Poulain.

This cleans up our conditional move implementation - specifically so that it distinguishes between
comparing the low 32 bits of a GPR and all bits of a GPR - and fixes bugs with operand ordering. It
then adds a Select opcode to B3 and adds all of the strength reduction and lowering magic that it
needs. Finally this change implements FTL::Output::select() in terms of B3::Select.

This patch lets us run Kraken/imaging-gaussian-blur. Running that benchmark using FTL+B3 is a 17%
speed-up. The compile times go down dramatically (by about 7x) and code quality stays about the same.

* assembler/MacroAssembler.h:
(JSC::MacroAssembler::moveDoubleConditionally32):
(JSC::MacroAssembler::moveDoubleConditionally64):
(JSC::MacroAssembler::moveDoubleConditionallyTest32):
(JSC::MacroAssembler::moveDoubleConditionallyTest64):
(JSC::MacroAssembler::moveDoubleConditionallyDouble):
(JSC::MacroAssembler::lea):
* assembler/MacroAssemblerX86Common.h:
(JSC::MacroAssemblerX86Common::move):
(JSC::MacroAssemblerX86Common::moveConditionallyDouble):
(JSC::MacroAssemblerX86Common::zeroExtend32ToPtr):
(JSC::MacroAssemblerX86Common::moveConditionally32):
(JSC::MacroAssemblerX86Common::moveConditionallyTest32):
(JSC::MacroAssemblerX86Common::set32):
(JSC::MacroAssemblerX86Common::cmov):
(JSC::MacroAssemblerX86Common::moveConditionally): Deleted.
(JSC::MacroAssemblerX86Common::moveConditionallyTest): Deleted.
* assembler/MacroAssemblerX86_64.h:
(JSC::MacroAssemblerX86_64::branchNeg64):
(JSC::MacroAssemblerX86_64::moveConditionally64):
(JSC::MacroAssemblerX86_64::moveConditionallyTest64):
(JSC::MacroAssemblerX86_64::abortWithReason):
* assembler/X86Assembler.h:
(JSC::X86Assembler::cmovl_rr):
(JSC::X86Assembler::cmovl_mr):
(JSC::X86Assembler::cmovel_rr):
(JSC::X86Assembler::cmovnel_rr):
(JSC::X86Assembler::cmovpl_rr):
(JSC::X86Assembler::cmovnpl_rr):
(JSC::X86Assembler::cmovq_rr):
(JSC::X86Assembler::cmovq_mr):
(JSC::X86Assembler::cmoveq_rr):
(JSC::X86Assembler::cmovneq_rr):
(JSC::X86Assembler::cmovpq_rr):
(JSC::X86Assembler::cmovnpq_rr):
* b3/B3LowerToAir.cpp:
(JSC::B3::Air::LowerToAir::createCompare):
(JSC::B3::Air::LowerToAir::createSelect):
(JSC::B3::Air::LowerToAir::marshallCCallArgument):
(JSC::B3::Air::LowerToAir::lower):
* b3/B3MoveConstants.cpp:
* b3/B3Opcode.cpp:
(WTF::printInternal):
* b3/B3Opcode.h:
* b3/B3ReduceStrength.cpp:
* b3/B3Validate.cpp:
* b3/B3Value.cpp:
(JSC::B3::Value::effects):
(JSC::B3::Value::key):
(JSC::B3::Value::checkOpcode):
(JSC::B3::Value::typeFor):
* b3/B3Value.h:
* b3/B3ValueKey.cpp:
(JSC::B3::ValueKey::dump):
(JSC::B3::ValueKey::materialize):
* b3/B3ValueKey.h:
(JSC::B3::ValueKey::ValueKey):
(JSC::B3::ValueKey::hash):
(JSC::B3::ValueKey::operator bool):
* b3/B3ValueKeyInlines.h:
(JSC::B3::ValueKey::ValueKey):
(JSC::B3::ValueKey::child):
* b3/air/AirOpcode.opcodes:
* b3/testb3.cpp:
(JSC::B3::testTruncSExt32):
(JSC::B3::testBasicSelect):
(JSC::B3::testSelectTest):
(JSC::B3::testSelectCompareDouble):
(JSC::B3::testSelectDouble):
(JSC::B3::testSelectDoubleTest):
(JSC::B3::testSelectDoubleCompareDouble):
(JSC::B3::zero):
(JSC::B3::run):
* ftl/FTLB3Output.h:
(JSC::FTL::Output::testIsZeroPtr):
(JSC::FTL::Output::testNonZeroPtr):
(JSC::FTL::Output::select):
(JSC::FTL::Output::extractValue):
(JSC::FTL::Output::fence):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoModern IDB: In the VersionChangeEvent for delete database calls, oldVersion should...
beidson@apple.com [Fri, 20 Nov 2015 22:14:56 +0000 (22:14 +0000)]
Modern IDB: In the VersionChangeEvent for delete database calls, oldVersion should be null instead of 0.
https://bugs.webkit.org/show_bug.cgi?id=151481

Reviewed by Sam Weinig.

Source/WebCore:

No new tests, covered by changes to:
    storage/indexeddb/modern/deletedatabase-1.html
    storage/indexeddb/modern/deletedatabase-2.html

* Modules/indexeddb/IDBVersionChangeEvent.h:
* Modules/indexeddb/IDBVersionChangeEvent.idl:

* Modules/indexeddb/client/IDBVersionChangeEventImpl.cpp:
(WebCore::IDBClient::IDBVersionChangeEvent::newVersion):
* Modules/indexeddb/client/IDBVersionChangeEventImpl.h:

* Modules/indexeddb/legacy/LegacyVersionChangeEvent.cpp:
(WebCore::LegacyVersionChangeEvent::newVersion):
* Modules/indexeddb/legacy/LegacyVersionChangeEvent.h:

LayoutTests:

* storage/indexeddb/modern/deletedatabase-1-expected.txt:
* storage/indexeddb/modern/deletedatabase-2-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove NETWORK_PROCESS compile flag
achristensen@apple.com [Fri, 20 Nov 2015 22:10:52 +0000 (22:10 +0000)]
Remove NETWORK_PROCESS compile flag
https://bugs.webkit.org/show_bug.cgi?id=151512

Reviewed by Tim Horton.

.:

* Source/cmake/OptionsEfl.cmake:
* Source/cmake/OptionsGTK.cmake:
* Source/cmake/OptionsMac.cmake:
* Source/cmake/WebKitFeatures.cmake:

Source/WebKit2:

* CMakeLists.txt:
* NetworkProcess/EntryPoint/mac/LegacyProcess/NetworkProcessMain.mm:
(NetworkProcessMain):
* NetworkProcess/EntryPoint/mac/XPCService/NetworkServiceEntryPoint.mm:
(NetworkServiceInitializer):
* NetworkProcess/FileAPI/NetworkBlobRegistry.cpp:
(WebKit::NetworkBlobRegistry::filesInBlob):
* NetworkProcess/FileAPI/NetworkBlobRegistry.h:
* NetworkProcess/NetworkConnectionToWebProcess.cpp:
(WebKit::NetworkConnectionToWebProcess::blobSize):
* NetworkProcess/NetworkConnectionToWebProcess.h:
* NetworkProcess/NetworkConnectionToWebProcess.messages.in:
* NetworkProcess/NetworkLoad.cpp:
* NetworkProcess/NetworkLoad.h:
* NetworkProcess/NetworkLoadClient.h:
* NetworkProcess/NetworkLoadParameters.cpp:
(WebKit::NetworkLoadParameters::NetworkLoadParameters):
* NetworkProcess/NetworkLoadParameters.h:
* NetworkProcess/NetworkProcess.cpp:
* NetworkProcess/NetworkProcess.h:
* NetworkProcess/NetworkProcess.messages.in:
* NetworkProcess/NetworkResourceLoader.cpp:
* NetworkProcess/NetworkResourceLoader.h:
* NetworkProcess/NetworkResourceLoader.messages.in:
* NetworkProcess/cocoa/NetworkProcessCocoa.mm:
(WebKit::NetworkProcess::clearDiskCache):
* NetworkProcess/efl/NetworkProcessMainEfl.cpp:
(WebKit::NetworkProcessMainUnix):
* NetworkProcess/gtk/NetworkProcessMainGtk.cpp:
(WebKit::NetworkProcessMainUnix):
* NetworkProcess/ios/NetworkProcessIOS.mm:
* NetworkProcess/mac/NetworkLoadMac.mm:
* NetworkProcess/mac/NetworkProcessMac.mm:
* NetworkProcess/mac/NetworkResourceLoaderMac.mm:
(WebKit::NetworkResourceLoader::willCacheResponseAsync):
* NetworkProcess/soup/NetworkProcessSoup.cpp:
(WebKit::NetworkProcess::platformTerminate):
* NetworkProcess/soup/RemoteNetworkingContextSoup.cpp:
(WebKit::RemoteNetworkingContext::storageSession):
* Shared/Authentication/AuthenticationManager.cpp:
(WebKit::AuthenticationManager::didReceiveAuthenticationChallenge):
* Shared/Network/CustomProtocols/Cocoa/CustomProtocolManagerCocoa.mm:
(WebKit::CustomProtocolManager::initialize):
(WebKit::CustomProtocolManager::addCustomProtocol):
* Shared/Network/CustomProtocols/CustomProtocolManager.h:
* Shared/Network/CustomProtocols/soup/CustomProtocolManagerSoup.cpp:
(WebKit::CustomProtocolManager::supplementName):
(WebKit::CustomProtocolManager::initialize):
(WebKit::CustomProtocolManager::registerScheme):
* Shared/Network/NetworkProcessCreationParameters.cpp:
(WebKit::NetworkProcessCreationParameters::decode):
* Shared/Network/NetworkProcessCreationParameters.h:
* Shared/Network/NetworkProcessSupplement.h:
(WebKit::NetworkProcessSupplement::initialize):
* Shared/Network/NetworkResourceLoadParameters.cpp:
(WebKit::NetworkResourceLoadParameters::decode):
* Shared/Network/NetworkResourceLoadParameters.h:
* Shared/ProcessExecutablePath.h:
* Shared/WebProcessCreationParameters.cpp:
(WebKit::WebProcessCreationParameters::WebProcessCreationParameters):
(WebKit::WebProcessCreationParameters::encode):
(WebKit::WebProcessCreationParameters::decode):
* Shared/WebProcessCreationParameters.h:
* Shared/efl/ProcessExecutablePathEfl.cpp:
(WebKit::executablePathOfPluginProcess):
(WebKit::executablePathOfNetworkProcess):
(WebKit::executablePathOfDatabaseProcess):
* Shared/gtk/ProcessExecutablePathGtk.cpp:
(WebKit::executablePathOfPluginProcess):
(WebKit::executablePathOfNetworkProcess):
(WebKit::executablePathOfDatabaseProcess):
* Shared/mac/CookieStorageShim.h:
* Shared/mac/CookieStorageShim.mm:
(-[WKNSURLSessionLocal _getCookieHeadersForTask:completionHandler:]):
* Shared/mac/CookieStorageShimLibrary.cpp:
* Shared/mac/CookieStorageShimLibrary.h:
* UIProcess/API/APISession.cpp:
(API::Session::~Session):
* UIProcess/API/C/mac/WKContextPrivateMac.mm:
(WKContextGetNetworkProcessIdentifier):
* UIProcess/API/efl/ewk_context.cpp:
(ewk_context_process_model_set):
(ewk_context_process_model_get):
(ewk_context_tls_error_policy_get):
* UIProcess/Cocoa/WebProcessPoolCocoa.mm:
(WebKit::WebProcessPool::updateProcessSuppressionState):
(WebKit::WebProcessPool::platformInitializeWebProcess):
(WebKit::WebProcessPool::platformInitializeNetworkProcess):
(WebKit::WebProcessPool::platformInvalidateContext):
* UIProcess/Downloads/DownloadProxy.cpp:
(WebKit::DownloadProxy::cancel):
* UIProcess/Launcher/ProcessLauncher.cpp:
(WebKit::ProcessLauncher::processTypeAsString):
(WebKit::ProcessLauncher::getProcessTypeFromString):
* UIProcess/Launcher/ProcessLauncher.h:
* UIProcess/Launcher/efl/ProcessLauncherEfl.cpp:
(WebKit::ProcessLauncher::launchProcess):
* UIProcess/Launcher/gtk/ProcessLauncherGtk.cpp:
(WebKit::ProcessLauncher::launchProcess):
* UIProcess/Launcher/mac/ProcessLauncherMac.mm:
(WebKit::computeProcessShimPath):
(WebKit::serviceName):
(WebKit::shouldLeakBoost):
(WebKit::createProcess):
* UIProcess/Network/NetworkProcessProxy.cpp:
(WebKit::NetworkProcessProxy::setIsHoldingLockedFiles):
* UIProcess/Network/NetworkProcessProxy.h:
* UIProcess/Network/NetworkProcessProxy.messages.in:
* UIProcess/Network/mac/NetworkProcessProxyMac.mm:
(WebKit::NetworkProcessProxy::platformGetLaunchOptions):
* UIProcess/Network/soup/NetworkProcessProxySoup.cpp:
(WebKit::NetworkProcessProxy::platformGetLaunchOptions):
* UIProcess/WebCookieManagerProxy.cpp:
(WebKit::WebCookieManagerProxy::setHTTPCookieAcceptPolicy):
(WebKit::WebCookieManagerProxy::getHTTPCookieAcceptPolicy):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::WebPageProxy):
(WebKit::WebPageProxy::setSessionID):
(WebKit::WebPageProxy::initializeWebPage):
* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::WebProcessPool):
(WebKit::WebProcessPool::~WebProcessPool):
(WebKit::WebProcessPool::initializeClient):
(WebKit::WebProcessPool::networkingProcessConnection):
(WebKit::WebProcessPool::languageChanged):
(WebKit::WebProcessPool::setUsesNetworkProcess):
(WebKit::WebProcessPool::usesNetworkProcess):
(WebKit::WebProcessPool::ensureNetworkProcess):
(WebKit::WebProcessPool::getNetworkProcessConnection):
(WebKit::WebProcessPool::ensureDatabaseProcess):
(WebKit::WebProcessPool::setAnyPageGroupMightHavePrivateBrowsingEnabled):
(WebKit::WebProcessPool::createNewWebProcess):
(WebKit::WebProcessPool::download):
(WebKit::WebProcessPool::resumeDownload):
(WebKit::WebProcessPool::setAdditionalPluginsDirectory):
(WebKit::WebProcessPool::networkProcessIdentifier):
(WebKit::WebProcessPool::setAlwaysUsesComplexTextCodePath):
(WebKit::WebProcessPool::setCanHandleHTTPSServerTrustEvaluation):
(WebKit::WebProcessPool::registerURLSchemeAsLocal):
(WebKit::WebProcessPool::setCacheModel):
(WebKit::WebProcessPool::setDefaultRequestTimeoutInterval):
(WebKit::WebProcessPool::createDownloadProxy):
(WebKit::WebProcessPool::useTestingNetworkSession):
(WebKit::WebProcessPool::allowSpecificHTTPSCertificateForHost):
(WebKit::WebProcessPool::requestNetworkingStatistics):
(WebKit::WebProcessPool::handleMessage):
* UIProcess/WebProcessPool.h:
(WebKit::WebProcessPool::sendToNetworkingProcess):
(WebKit::WebProcessPool::sendToNetworkingProcessRelaunchingIfNecessary):
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::getPluginProcessConnection):
(WebKit::WebProcessProxy::getNetworkProcessConnection):
(WebKit::WebProcessProxy::getDatabaseProcessConnection):
(WebKit::WebProcessProxy::createDownloadProxy):
(WebKit::WebProcessProxy::didCancelProcessSuspension):
(WebKit::WebProcessProxy::reinstateNetworkProcessAssertionState):
(WebKit::WebProcessProxy::didSetAssertionState):
* UIProcess/WebProcessProxy.h:
(WebKit::WebProcessProxy::throttler):
* UIProcess/WebProcessProxy.messages.in:
* UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::~WebsiteDataStore):
(WebKit::WebsiteDataStore::cloneSessionData):
* UIProcess/gtk/WebProcessPoolGtk.cpp:
(WebKit::initInspectorServer):
(WebKit::WebProcessPool::setIgnoreTLSErrors):
* UIProcess/soup/WebProcessPoolSoup.cpp:
(WebKit::WebProcessPool::platformInitializeNetworkProcess):
* WebKit2Prefix.h:
* WebProcess/FileAPI/BlobRegistryProxy.cpp:
(WebKit::BlobRegistryProxy::blobSize):
* WebProcess/FileAPI/BlobRegistryProxy.h:
* WebProcess/Network/NetworkProcessConnection.cpp:
* WebProcess/Network/NetworkProcessConnection.h:
* WebProcess/Network/NetworkProcessConnection.messages.in:
* WebProcess/Network/WebResourceLoadScheduler.cpp:
(WebKit::WebResourceLoadScheduler::networkProcessCrashed):
* WebProcess/Network/WebResourceLoadScheduler.h:
* WebProcess/Network/WebResourceLoader.cpp:
* WebProcess/Network/WebResourceLoader.h:
* WebProcess/Network/WebResourceLoader.messages.in:
* WebProcess/Plugins/Netscape/NetscapeBrowserFuncs.cpp:
(WebKit::initializeBrowserFuncs):
(WebKit::netscapeBrowserFuncs):
* WebProcess/Plugins/Netscape/NetscapePlugin.cpp:
(WebKit::NetscapePlugin::registerRedirect):
(WebKit::NetscapePlugin::urlRedirectResponse):
* WebProcess/WebCoreSupport/WebPlatformStrategies.cpp:
(WebKit::WebPlatformStrategies::cookiesForDOM):
(WebKit::WebPlatformStrategies::setCookiesFromDOM):
(WebKit::WebPlatformStrategies::cookiesEnabled):
(WebKit::WebPlatformStrategies::cookieRequestHeaderFieldValue):
(WebKit::WebPlatformStrategies::getRawCookies):
(WebKit::WebPlatformStrategies::deleteCookie):
(WebKit::WebPlatformStrategies::resourceLoadScheduler):
(WebKit::WebPlatformStrategies::createBlobRegistry):
* WebProcess/WebCoreSupport/WebPlatformStrategies.h:
* WebProcess/WebPage/WebFrame.cpp:
(WebKit::WebFrame::startDownload):
(WebKit::WebFrame::convertMainResourceLoadToDownload):
* WebProcess/WebProcess.cpp:
(WebKit::WebProcess::WebProcess):
(WebKit::WebProcess::initializeWebProcess):
(WebKit::WebProcess::ensureNetworkProcessConnection):
(WebKit::WebProcess::registerURLSchemeAsEmptyDocument):
(WebKit::WebProcess::usesNetworkProcess):
(WebKit::WebProcess::networkConnection):
(WebKit::WebProcess::webResourceLoadScheduler):
(WebKit::WebProcess::webToDatabaseProcessConnectionClosed):
(WebKit::WebProcess::prefetchDNS):
* WebProcess/WebProcess.h:
* config.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[JSC] Add Air lowering to BitNot() for Xor(value, -1)
benjamin@webkit.org [Fri, 20 Nov 2015 22:07:05 +0000 (22:07 +0000)]
[JSC] Add Air lowering to BitNot() for Xor(value, -1)
https://bugs.webkit.org/show_bug.cgi?id=151474

Patch by Benjamin Poulain <bpoulain@apple.com> on 2015-11-20
Reviewed by Filip Pizlo.

* assembler/MacroAssemblerX86Common.h:
(JSC::MacroAssemblerX86Common::not32):
* assembler/MacroAssemblerX86_64.h:
(JSC::MacroAssemblerX86_64::not64):
* assembler/X86Assembler.h:
(JSC::X86Assembler::notq_r):
(JSC::X86Assembler::notq_m):
* b3/B3LowerToAir.cpp:
(JSC::B3::Air::LowerToAir::createGenericCompare):
(JSC::B3::Air::LowerToAir::lower):
* b3/B3ReduceStrength.cpp:
* b3/air/AirOpcode.opcodes:
* b3/testb3.cpp:
(JSC::B3::testNegValueSubOne):
(JSC::B3::testNegValueSubOne32):
(JSC::B3::testBitNotArg):
(JSC::B3::testBitNotImm):
(JSC::B3::testBitNotMem):
(JSC::B3::testBitNotArg32):
(JSC::B3::testBitNotImm32):
(JSC::B3::testBitNotMem32):
(JSC::B3::testBitNotOnBooleanAndBranch32):
(JSC::B3::int64Operands):
(JSC::B3::int32Operands):
(JSC::B3::run):
* ftl/FTLB3Output.h:
(JSC::FTL::Output::bitNot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192696 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoSuper use should be recorded in per-function scope
utatane.tea@gmail.com [Fri, 20 Nov 2015 21:19:57 +0000 (21:19 +0000)]
Super use should be recorded in per-function scope
https://bugs.webkit.org/show_bug.cgi?id=151500

Reviewed by Geoffrey Garen.

"super" use is prohibited under the non-constructor / non-class-method-related functions.
This "super" use should be recorded in per-function scope to check its incorrect use after
parsing a function.
Currently, we accidentally record it to a lexical current scope. So when using "super" inside
a block scope, our "super" use guard miss it.

* parser/Parser.cpp:
(JSC::Parser<LexerType>::parseMemberExpression):
* parser/Parser.h:
(JSC::Parser::currentVariableScope):
(JSC::Parser::currentFunctionScope):
(JSC::Parser::declareVariable):
* tests/stress/super-in-lexical-scope.js: Added.
(testSyntax):
(testSyntaxError):
(testSyntaxError.test):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAddressing missed review feedback for:
beidson@apple.com [Fri, 20 Nov 2015 21:14:58 +0000 (21:14 +0000)]
Addressing missed review feedback for:
Modern IDB: Make in-memory ObjectStore cursors work.
https://bugs.webkit.org/show_bug.cgi?id=151196

Reviewed by Darin Adler.

* Modules/indexeddb/server/MemoryObjectStore.cpp:
(WebCore::IDBServer::MemoryObjectStore::maybeOpenCursor):
* Modules/indexeddb/server/MemoryObjectStoreCursor.cpp:
(WebCore::IDBServer::MemoryObjectStoreCursor::create): Deleted.
* Modules/indexeddb/server/MemoryObjectStoreCursor.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192694 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoCaching of properties on objects that have named property getters is sometimes incorrect
cdumez@apple.com [Fri, 20 Nov 2015 20:45:12 +0000 (20:45 +0000)]
Caching of properties on objects that have named property getters is sometimes incorrect
https://bugs.webkit.org/show_bug.cgi?id=151453
<rdar://problem/23049343>

Reviewed by Gavin Barraclough.

Source/JavaScriptCore:

Add new GetOwnPropertySlotIsImpureForPropertyAbsence TypeInfo flag to be
used by objects that have a non-'OverrideBuiltins' named property getter.
This flag prevents caching of properties that are missing as a named
property with this name may later become available.

Objects with an 'OverrideBuiltins' named property getter will keep using
the GetOwnPropertySlotIsImpure TypeInfo flag, which prevents all property
caching since named properties can override own properties or properties
on the prototype.

* bytecode/ComplexGetStatus.cpp:
(JSC::ComplexGetStatus::computeFor):
* bytecode/PropertyCondition.cpp:
(JSC::PropertyCondition::isStillValid):
* jit/Repatch.cpp:
(JSC::tryCacheGetByID):
(JSC::tryRepatchIn):
* jsc.cpp:
* runtime/JSTypeInfo.h:
(JSC::TypeInfo::getOwnPropertySlotIsImpure):
(JSC::TypeInfo::getOwnPropertySlotIsImpureForPropertyAbsence):
(JSC::TypeInfo::prohibitsPropertyCaching): Deleted.
* runtime/Structure.h:

Source/WebCore:

In r188590, we dropped the JSC::GetOwnPropertySlotIsImpure TypeInfo flag for
interfaces that have a non-'OverrideBuiltins' named property getter in order
to allow caching of properties returns by GetOwnPropertySlot(). We assumed
this was safe as it was no longer possible for named properties to override
own properties (or properties on the prototype).

However, there is an issue when we cache the non-existence of a property.
Even though at one point the property did not exist, a named property with
this name may later become available. In such case, caching would cause us
to wrongly report a property as missing.

To address the problem, this patch introduces a new
GetOwnPropertySlotIsImpureForPropertyAbsence TypeInfo flag and uses it for
interfaces that have a non-'OverrideBuiltins' named property getter. This
will cause us to not cache the fact that a property is missing on such
objects, while maintaining the performance win from r188590 in the common
case.

Test: fast/dom/NamedNodeMap-named-getter-caching.html

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateHeader):
* bindings/scripts/test/JS/JSTestCustomNamedGetter.h:
* bindings/scripts/test/JS/JSTestEventTarget.h:
* bindings/scripts/test/JS/JSTestOverrideBuiltins.h:

LayoutTests:

Add layout test to make sure caching does not cause NamedNodeMap's
named property getter to sometimes wrongly report a property as
missing.

* fast/dom/NamedNodeMap-named-getter-caching-expected.txt: Added.
* fast/dom/NamedNodeMap-named-getter-caching.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove chartPointRadius from interactive chart component
rniwa@webkit.org [Fri, 20 Nov 2015 20:40:26 +0000 (20:40 +0000)]
Remove chartPointRadius from interactive chart component
https://bugs.webkit.org/show_bug.cgi?id=151480

Reviewed by Darin Adler.

Replaced the parameter by CSS rules.

* public/v2/chart-pane.css:
(.chart .dot):
(.chart .dot.foreground):
(.chart .highlight):
(.chart .extent):
* public/v2/index.html:
* public/v2/interactive-chart.js:
(App.InteractiveChartComponent.Ember.Component.extend._constructGraphIfPossible):
(App.InteractiveChartComponent.Ember.Component.extend._highlightedItemsChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoPerf dashboard's runs API uses more than 128MB of memory
rniwa@webkit.org [Fri, 20 Nov 2015 20:39:34 +0000 (20:39 +0000)]
Perf dashboard's runs API uses more than 128MB of memory
https://bugs.webkit.org/show_bug.cgi?id=151478

Reviewed by Andreas Kling.

Don't fetch all query results at once to avoid using twice as much memory as needed.
Use iterative API to format each result at a time.

This change is also a 5% runtime performance gain.

* public/api/runs.php:
(RunsGenerator::__construct): Takes a Database instance instead of a list of configurations. The latter is
no longer needed as we pass in each configuration type explicitly to fetch_runs.
(RunsGenerator::fetch_runs): Renamed from add_runs since it now executes the database query via execute_query.
Also moved the logic to compute the last modified time here.
(RunsGenerator::execute_query): Moved from fetch_runs_for_config. Use Database::query instead of query_and_fetch_all.
(RunsGeneratorForTestGroup):
(RunsGeneratorForTestGroup::__construct):
(RunsGeneratorForTestGroup::execute_query): Moved from fetch_runs_for_config_and_test_group.

* public/include/db.php:
(generate_data_file): Lock the file to avoid corruption.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoPerf dashboard always fetches charts JSON twice
rniwa@webkit.org [Fri, 20 Nov 2015 20:35:18 +0000 (20:35 +0000)]
Perf dashboard always fetches charts JSON twice
https://bugs.webkit.org/show_bug.cgi?id=151483

Reviewed by Andreas Kling.

Only re-generate "runs" JSON via /api/runs/ when the cache doesn't exist in /data/ or the cached JSON is
obsolete (shouldRefetch is set true) or corrupt (the second closure).

* public/v2/app.js:
(App.Pane._fetch):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoBack-buffer to front-buffer copy fails for some buffer formats
simon.fraser@apple.com [Fri, 20 Nov 2015 20:05:05 +0000 (20:05 +0000)]
Back-buffer to front-buffer copy fails for some buffer formats
https://bugs.webkit.org/show_bug.cgi?id=151475
rdar://problem/23617899

Reviewed by Tim Horton.

Source/WebCore:

Fix some fo the bitsPerComponent/bitsPerPixel options in IOSurface::ensurePlatformContext()
for RGB10 buffers. Fix IOSurface::format() to return the new formats.

Implement IOSurface::copyToSurface(), which does a synchronous copy between
surfaces.

* platform/graphics/cocoa/IOSurface.h:
* platform/graphics/cocoa/IOSurface.mm:
(IOSurface::create):
(IOSurface::ensurePlatformContext):
(IOSurface::format):
(IOSurface::copyToSurface):

Source/WebKit2:

When displaying RemoteLayerBackingStore, we copy the back buffer to the front
buffer before painting the updated regions. For buffers using Format::RGB10A8,
the CGImage-based copy fails, so in this case, use IOSurface::copyToSurface().

Reorganized RemoteLayerBackingStore::drawInContext() to make this a bit easier
to understand. First, we either copy the entire surface over, or paint the backImage.
Then we clip to the dirty rects, and clear them, then paint the layer contents.

* Shared/mac/RemoteLayerBackingStore.mm:
(WebKit::RemoteLayerBackingStore::display):
(WebKit::RemoteLayerBackingStore::drawInContext):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoSimple line layout: Add text-indent support.
zalan@apple.com [Fri, 20 Nov 2015 19:20:57 +0000 (19:20 +0000)]
Simple line layout: Add text-indent support.
https://bugs.webkit.org/show_bug.cgi?id=151472

Reviewed by Simon Fraser.

This enables us to use simple line layout on text-indent content.

Source/WebCore:

Test: fast/text/simple-line-text-indent.html

* rendering/SimpleLineLayout.cpp:
(WebCore::SimpleLineLayout::updateLineConstrains):
(WebCore::SimpleLineLayout::createTextRuns):
(WebCore::SimpleLineLayout::canUseForStyle): Deleted.

LayoutTests:

* fast/text/simple-line-text-indent-expected.html: Added.
* fast/text/simple-line-text-indent.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192688 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoModern IDB: IDBFactory.deleteDatabase() support.
beidson@apple.com [Fri, 20 Nov 2015 18:55:42 +0000 (18:55 +0000)]
Modern IDB: IDBFactory.deleteDatabase() support.
https://bugs.webkit.org/show_bug.cgi?id=151456

Reviewed by Alex Christensen.

Source/WebCore:

Tests: storage/indexeddb/modern/deletedatabase-1.html
       storage/indexeddb/modern/deletedatabase-2.html

* Modules/indexeddb/client/IDBDatabaseImpl.cpp:
(WebCore::IDBClient::IDBDatabase::hasPendingActivity):
(WebCore::IDBClient::IDBDatabase::didCommitOrAbortTransaction):
* Modules/indexeddb/client/IDBDatabaseImpl.h:

* Modules/indexeddb/client/IDBOpenDBRequestImpl.cpp:
(WebCore::IDBClient::IDBOpenDBRequest::onDeleteDatabaseSuccess):
(WebCore::IDBClient::IDBOpenDBRequest::requestCompleted):
* Modules/indexeddb/client/IDBOpenDBRequestImpl.h:

* Modules/indexeddb/client/IDBTransactionImpl.cpp:
(WebCore::IDBClient::IDBTransaction::hasPendingActivity): Deleted.

* Modules/indexeddb/server/IDBServer.cpp:
(WebCore::IDBServer::IDBServer::deleteDatabase):
(WebCore::IDBServer::IDBServer::deleteUniqueIDBDatabase):
* Modules/indexeddb/server/IDBServer.h:

* Modules/indexeddb/server/UniqueIDBDatabase.cpp:
(WebCore::IDBServer::UniqueIDBDatabase::UniqueIDBDatabase):
(WebCore::IDBServer::UniqueIDBDatabase::hasAnyPendingCallbacks):
(WebCore::IDBServer::UniqueIDBDatabase::maybeDeleteDatabase):
(WebCore::IDBServer::UniqueIDBDatabase::handleOpenDatabaseOperations):
(WebCore::IDBServer::UniqueIDBDatabase::handleDelete):
(WebCore::IDBServer::UniqueIDBDatabase::notifyConnectionsOfVersionChangeForUpgrade):
(WebCore::IDBServer::UniqueIDBDatabase::notifyConnectionsOfVersionChange):
(WebCore::IDBServer::UniqueIDBDatabase::connectionClosedFromClient):
(WebCore::IDBServer::UniqueIDBDatabase::enqueueTransaction):
(WebCore::IDBServer::UniqueIDBDatabase::invokeDeleteOrRunTransactionTimer):
(WebCore::IDBServer::UniqueIDBDatabase::deleteOrRunTransactionsTimerFired):
(WebCore::IDBServer::UniqueIDBDatabase::didPerformActivateTransactionInBackingStore):
(WebCore::IDBServer::UniqueIDBDatabase::inProgressTransactionCompleted):
(WebCore::IDBServer::UniqueIDBDatabase::invokeTransactionScheduler): Deleted.
(WebCore::IDBServer::UniqueIDBDatabase::transactionSchedulingTimerFired): Deleted.
* Modules/indexeddb/server/UniqueIDBDatabase.h:
(WebCore::IDBServer::UniqueIDBDatabase::identifier):
(WebCore::IDBServer::UniqueIDBDatabase::deletePending):

* Modules/indexeddb/shared/IDBResultData.cpp:
(WebCore::IDBResultData::deleteDatabaseSuccess):
* Modules/indexeddb/shared/IDBResultData.h:

LayoutTests:

* storage/indexeddb/modern/deletedatabase-1-expected.txt: Added.
* storage/indexeddb/modern/deletedatabase-1.html: Added.
* storage/indexeddb/modern/deletedatabase-2-expected.txt: Added.
* storage/indexeddb/modern/deletedatabase-2.html: Added.
* storage/indexeddb/modern/deletedatabase-request-event-expected.txt:
* storage/indexeddb/modern/deletedatabase-request-event.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoModern IDB: Get IDBRequest.readyState right.
beidson@apple.com [Fri, 20 Nov 2015 18:55:11 +0000 (18:55 +0000)]
Modern IDB: Get IDBRequest.readyState right.
https://bugs.webkit.org/show_bug.cgi?id=151484

Reviewed by Alex Christensen.

Source/WebCore:

Test: storage/indexeddb/modern/request-readystate.html

* Modules/indexeddb/client/IDBRequestImpl.cpp:
(WebCore::IDBClient::IDBRequest::readyState):

LayoutTests:

* storage/indexeddb/modern/request-readystate-expected.txt: Added.
* storage/indexeddb/modern/request-readystate.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192686 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, fix installation of ninja-build after the previous commit
mcatanzaro@igalia.com [Fri, 20 Nov 2015 15:17:36 +0000 (15:17 +0000)]
Unreviewed, fix installation of ninja-build after the previous commit

* gtk/install-dependencies:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192685 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoChanging some imported/w3c/web-platform-tests/html/semantics expectations from Timeou...
youenn.fablet@crf.canon.fr [Fri, 20 Nov 2015 15:01:27 +0000 (15:01 +0000)]
Changing some imported/w3c/web-platform-tests/html/semantics expectations from Timeout/Failure to Pass
https://bugs.webkit.org/show_bug.cgi?id=151492

Unreviewed.

* TestExpectations: Removing timeout/failure expectations.
* platform/efl/TestExpectations: Removing no longer needed pass expectation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192684 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] install-dependencies should run installer exactly once
mcatanzaro@igalia.com [Fri, 20 Nov 2015 14:41:50 +0000 (14:41 +0000)]
[GTK] install-dependencies should run installer exactly once
https://bugs.webkit.org/show_bug.cgi?id=151477

Reviewed by Carlos Garcia Campos.

* gtk/install-dependencies:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192683 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[EFL] Enable FTL JIT by default on X86_64
ossy@webkit.org [Fri, 20 Nov 2015 11:38:10 +0000 (11:38 +0000)]
[EFL] Enable FTL JIT by default on X86_64
https://bugs.webkit.org/show_bug.cgi?id=143822

Reviewed by Carlos Garcia Campos.

.:

* Source/cmake/OptionsEfl.cmake:

Tools:

* Scripts/build-jsc:
* Scripts/run-javascriptcore-tests:
* Scripts/webkitperl/FeatureList.pm:
* efl/jhbuild.modules:
* gtk/jhbuildrc:
* jhbuild/jhbuildrc_common.py:
(init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192682 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnskip mediastream tests after crash was fixed
ddkilzer@apple.com [Fri, 20 Nov 2015 09:33:23 +0000 (09:33 +0000)]
Unskip mediastream tests after crash was fixed
<https://bugs.webkit.org/show_bug.cgi?id=151353>

* platform/mac/TestExpectations: Bug 151462 fixed the crash, so
unskip the tests again.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192681 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUse HTTPHeaderName as much as possible in XMLHttpRequest
youenn.fablet@crf.canon.fr [Fri, 20 Nov 2015 09:25:56 +0000 (09:25 +0000)]
Use HTTPHeaderName as much as possible in XMLHttpRequest
https://bugs.webkit.org/show_bug.cgi?id=151438

Reviewed by Darin Adler.

Removing XMLHttpRequest::setRequestHeaderInternal and XMLHttpRequest::getRequestHeader.
Using directly HTTPHeaderMap.add, HTTPHeaderMap.set and HTTPHeaderMap.get.

No change in behavior.

* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::send):
(WebCore::XMLHttpRequest::setRequestHeader):
(WebCore::XMLHttpRequest::getAllResponseHeaders):
* xml/XMLHttpRequest.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@192680 268f45cc-cd09-0410-ab3c-d52691b4dbfc