⚠ Archived content — this site is no longer maintained.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

Timeline



Jul 20, 2015:

11:18 PM Changeset in webkit [187101] by Michael Catanzaro
  • 2 edits in trunk/Source/WebCore

REGRESSION(r184376): [SOUP] Multiple assertions when downloading files
​https://bugs.webkit.org/show_bug.cgi?id=147039

Reviewed by Darin Adler.

No new tests; this will be covered once we enable the network process for API tests.

  • platform/network/soup/ResourceHandleSoup.cpp:

(WebCore::ResourceHandle::releaseForDownload): Call RefPtrBase::relaxAdoptionRequirement so
that we don't assert when storing the non-adopted ResourceHandle in a RefPtr. The ref will
be dropped in ResourceHandle::cleanupSoupOperation, which as the comment says should always
run. HOPEFULLY.
(WebCore::ResourceHandle::continueWillSendRequest): Don't assert that client() is nonnull,
because the code clearly expects and handles the case where it is null.
(WebCore::ResourceHandle::continueDidReceiveResponse): Ditto; note that here client() will
always be null for a download.

11:06 PM Changeset in webkit [187100] by Carlos Garcia Campos
  • 20 edits
    3 adds in trunk

[GTK] Add API to be notified about editor state changes
​https://bugs.webkit.org/show_bug.cgi?id=145875

Reviewed by Gustavo Noronha Silva.

Source/WebKit2:

Add WebKitEditorState object, that is created on demand by the
WebKitWebView and can be used to get the typing attributes of the
editor at the current position or selection.

  • PlatformGTK.cmake:
  • Shared/EditorState.cpp: Use part of the PostLayoutData struct

for the GTK port too.
(WebKit::EditorState::encode): Encode PostLayoutData for GTK and
remove our custom cursorRect.
(WebKit::EditorState::decode): Decode PostLayoutData for GTK and
remove our custom cursorRect.
(WebKit::EditorState::PostLayoutData::encode): Reorder it to
reduce the amonut of ifdefs.
(WebKit::EditorState::PostLayoutData::decode): Ditto.

  • Shared/EditorState.h: Add AttributeStrikeThrough to

TypingAttributes enum.

  • UIProcess/API/gtk/PageClientImpl.cpp:

(WebKit::PageClientImpl::selectionDidChange): Rename
updateTextInputState() to selectionDidChange() and also notify the
WebKitWebView.

  • UIProcess/API/gtk/PageClientImpl.h:
  • UIProcess/API/gtk/WebKitEditorState.cpp: Added.

(webkitEditorStateGetProperty):
(webkit_editor_state_class_init):
(webkitEditorStateSetTypingAttributes):
(webkitEditorStateCreate):
(webkitEditorStateChanged):
(webkit_editor_state_get_typing_attributes):

  • UIProcess/API/gtk/WebKitEditorState.h: Added.
  • UIProcess/API/gtk/WebKitEditorStatePrivate.h: Added.
  • UIProcess/API/gtk/WebKitWebView.cpp:

(webkitWebViewSelectionDidChange): Notify the WebKitEditorState if
it has already been created.
(webkit_web_view_get_editor_state): Ensure a WebKitEditorState and
return it.

  • UIProcess/API/gtk/WebKitWebView.h:
  • UIProcess/API/gtk/WebKitWebViewBase.cpp:

(webkitWebViewBaseUpdateTextInputState): Get the caret cursor rect
from PostLayoutData.

  • UIProcess/API/gtk/WebKitWebViewPrivate.h:
  • UIProcess/API/gtk/docs/webkit2gtk-4.0-sections.txt: Add new symbols.
  • UIProcess/API/gtk/docs/webkit2gtk-4.0.types: Add webkit_editor_state_get_type.
  • UIProcess/API/gtk/docs/webkit2gtk-docs.sgml: Add new section for WebKitEditorState.
  • UIProcess/API/gtk/webkit2.h: Include WebKitEditorState.h.
  • UIProcess/PageClient.h:
  • UIProcess/gtk/WebPageProxyGtk.cpp:

(WebKit::WebPageProxy::editorStateChanged): Call PageClientImpl::selectionDidChange().

  • WebProcess/WebPage/gtk/WebPageGtk.cpp:

(WebKit::WebPage::platformEditorState): Add typing attributes to EditorState.

Tools:

Update the typing attributes toggle buttons state according to the
editor state in MiniBrowser, and add a test case to the
WebViewEditor unit tests.

  • MiniBrowser/gtk/BrowserWindow.c:

(browserWindowEditingCommandToggleButtonSetActive):
(typingAttributesChanged):
(browserWindowSetupEditorToolbar):
(browserWindowConstructed):
(browser_window_init):

  • TestWebKitAPI/Tests/WebKit2Gtk/TestWebViewEditor.cpp:

(testWebViewEditorEditorStateTypingAttributes):
(beforeAll):

10:56 PM Changeset in webkit [187099] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebCore

Merge r187098. rdar://problem/21878275

10:39 PM Changeset in webkit [187098] by adachan@apple.com
  • 2 edits in trunk/Source/WebCore

Follow-up to my earlier fix for r147085
​https://bugs.webkit.org/show_bug.cgi?id=147085

Reviewed by Eric Carlson.

  • Modules/webaudio/AudioContext.cpp:

(WebCore::AudioContext::isPlayingAudioDidChange):
Null-check document() before dereferencing it in case the audio context's document is destroyed
by the time the code block is called on the main thread.

10:35 PM Changeset in webkit [187097] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebKit2

Merge r187072. rdar://problem/21582858

10:35 PM Changeset in webkit [187096] by matthew_hanson@apple.com
  • 5 edits in branches/safari-601.1-branch/Source/WebKit2

Merge r187050. rdar://problem/21582858

10:33 PM Changeset in webkit [187095] by Carlos Garcia Campos
  • 10 edits in trunk/Source/WebCore

Remove RenderTheme::shouldShowPlaceholderWhenFocused()
​https://bugs.webkit.org/show_bug.cgi?id=147104

Reviewed by Martin Robinson.

GTK+ is the only port that returns false in
shouldShowPlaceholderWhenFocused(). That's inconsistent with all
other browsers that show the placeholder text even for focused
entries in all platforms. We should change the GTK+ port
behaviour, but that would leave all implementations of
shouldShowPlaceholderWhenFocused returning true, so let's just
remove it.

  • html/HTMLTextFormControlElement.cpp:

(WebCore::HTMLTextFormControlElement::placeholderShouldBeVisible):
Do not consider whether the entry is focused or not.

  • platform/efl/RenderThemeEfl.h:
  • rendering/RenderTheme.h:

(WebCore::RenderTheme::shouldShowPlaceholderWhenFocused): Deleted.

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

(WebCore::RenderThemeIOS::shouldShowPlaceholderWhenFocused): Deleted.

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

(WebCore::RenderThemeMac::shouldShowPlaceholderWhenFocused): Deleted.

  • rendering/RenderThemeSafari.h:

(WebCore::RenderThemeSafari::shouldShowPlaceholderWhenFocused): Deleted.

  • rendering/RenderThemeWin.h:
10:27 PM Changeset in webkit [187094] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebKit2

Merge r187054. rdar://problem/21823349

10:27 PM Changeset in webkit [187093] by matthew_hanson@apple.com
  • 3 edits
    2 adds in branches/safari-601.1-branch

Merge r187053. rdar://problem/21778212

10:27 PM Changeset in webkit [187092] by matthew_hanson@apple.com
  • 14 edits in branches/safari-601.1-branch/Source

Merge r187044. rdar://problem/21661808

10:27 PM Changeset in webkit [187091] by matthew_hanson@apple.com
  • 4 edits in branches/safari-601.1-branch/Source

Merge r187039. rdar://problem/21474317

10:27 PM Changeset in webkit [187090] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebKit

Merge r187027. rdar://problem/21882777

10:27 PM Changeset in webkit [187089] by matthew_hanson@apple.com
  • 4 edits in branches/safari-601.1-branch/Source/JavaScriptCore

Merge r186819. rdar://problem/21729083

10:16 PM Changeset in webkit [187088] by zandobersek@gmail.com
  • 2 edits in trunk/Source/WebCore

[CoordinatedGraphics] CoordinatedGraphicsLayer::setContentsToImage() should avoid scheduling unnecessary flushes
​https://bugs.webkit.org/show_bug.cgi?id=147118

Reviewed by Martin Robinson.

Have CoordinatedGraphicsLayer::setContentsToImage() return early if the new
passed-in Image and the corresponding native image pointer are equal to the
currently set values.

This specifically avoids scheduling unnecessary flushes when setContentsToImage()
is repeatedly called with a null Image pointer, which in previous code would
unconditionally result in a scheduled flush even if there was no Image assigned
as the content of this layer before. Until now the layer flush scheduling was only
avoided if the two non-null native image pointers were equal.

  • platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:

(WebCore::CoordinatedGraphicsLayer::setContentsToImage):

10:08 PM Changeset in webkit [187087] by matthew_hanson@apple.com
  • 12 edits
    1 add in branches/safari-601.1-branch

Merge r186910. rdar://problem/21863296

10:08 PM Changeset in webkit [187086] by matthew_hanson@apple.com
  • 4 edits in branches/safari-601.1-branch/Source/WebCore

Merge r186715. rdar://problem/21863296

9:38 PM Changeset in webkit [187085] by matthew_hanson@apple.com
  • 20 edits
    1 copy
    4 adds in branches/safari-601.1-branch

Merge r186955. rdar://problem/14489674

9:38 PM Changeset in webkit [187084] by matthew_hanson@apple.com
  • 7 edits in branches/safari-601.1-branch/Source/WebKit2

Merge r187047. rdar://problem/20860410

9:38 PM Changeset in webkit [187083] by matthew_hanson@apple.com
  • 4 edits in branches/safari-601.1-branch/WebKitLibraries

Merge r187038. rdar://problem/20860410

9:38 PM Changeset in webkit [187082] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebCore

Merge r187035. rdar://problem/21712311

9:38 PM Changeset in webkit [187081] by matthew_hanson@apple.com
  • 9 edits in branches/safari-601.1-branch

Merge r187031. rdar://problem/21712311

9:37 PM Changeset in webkit [187080] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/JavaScriptCore

Merge r187028. rdar://problem/21869970

9:37 PM Changeset in webkit [187079] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebCore

Merge r187025. rdar://problem/21878275

9:37 PM Changeset in webkit [187078] by matthew_hanson@apple.com
  • 3 edits in branches/safari-601.1-branch/Source/WebKit2

Merge r187023. rdar://problem/21799011

9:37 PM Changeset in webkit [187077] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebKit2

Merge r187001. rdar://problem/21769577

9:37 PM Changeset in webkit [187076] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebCore

Merge r186991. rdar://problem/21643094

9:37 PM Changeset in webkit [187075] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebCore

Merge r186990. rdar://problem/21643094

9:37 PM Changeset in webkit [187074] by matthew_hanson@apple.com
  • 3 edits in branches/safari-601.1-branch/LayoutTests

Merge r186988. rdar://problem/21567820

9:37 PM Changeset in webkit [187073] by matthew_hanson@apple.com
  • 12 edits
    17 adds in branches/safari-601.1-branch

Merge r186982. rdar://problem/21567820

9:37 PM Changeset in webkit [187072] by Simon Fraser
  • 2 edits in trunk/Source/WebKit2

Fix the iOS build.

  • WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.h:
9:37 PM Changeset in webkit [187071] by matthew_hanson@apple.com
  • 3 edits in branches/safari-601.1-branch/Source/WebCore

Merge r186979. rdar://problem/19192076

9:37 PM Changeset in webkit [187070] by matthew_hanson@apple.com
  • 6 edits in branches/safari-601.1-branch/Source

Merge r186978. rdar://problem/21643094

9:37 PM Changeset in webkit [187069] by matthew_hanson@apple.com
  • 9 edits in branches/safari-601.1-branch

Merge r186976. rdar://problem/21643094

9:37 PM Changeset in webkit [187068] by matthew_hanson@apple.com
  • 3 edits
    2 adds in branches/safari-601.1-branch

Merge r186974. rdar://problem/21106945

9:37 PM Changeset in webkit [187067] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebKit2

Merge r186969. rdar://problem/21803781

9:37 PM Changeset in webkit [187066] by matthew_hanson@apple.com
  • 3 edits
    2 adds in branches/safari-601.1-branch

Merge r186968. rdar://problem/21359811

9:37 PM Changeset in webkit [187065] by matthew_hanson@apple.com
  • 10 edits
    1 add in branches/safari-601.1-branch

Merge r186964. rdar://problem/21803781

9:37 PM Changeset in webkit [187064] by matthew_hanson@apple.com
  • 7 edits
    9 adds in branches/safari-601.1-branch

Merge r186957. rdar://problem/21867831

9:37 PM Changeset in webkit [187063] by matthew_hanson@apple.com
  • 9 edits in branches/safari-601.1-branch/Source

Merge r186956. rdar://problem/21643094

9:36 PM Changeset in webkit [187062] by matthew_hanson@apple.com
  • 10 edits in branches/safari-601.1-branch/Source

Merge r186916. rdar://problem/21643094

9:36 PM Changeset in webkit [187061] by matthew_hanson@apple.com
  • 8 edits in branches/safari-601.1-branch

Merge r186912. rdar://problem/21866271

9:36 PM Changeset in webkit [187060] by matthew_hanson@apple.com
  • 3 edits in branches/safari-601.1-branch/Source/WebCore

Merge r186890. rdar://problem/21643094

9:36 PM Changeset in webkit [187059] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebCore

Merge r186875. rdar://problem/21643094

9:36 PM Changeset in webkit [187058] by matthew_hanson@apple.com
  • 3 edits in branches/safari-601.1-branch/Source/WebCore

Merge r186867. rdar://problem/21643094

9:36 PM Changeset in webkit [187057] by matthew_hanson@apple.com
  • 2 edits in branches/safari-601.1-branch/Source/WebCore

Merge r186863. rdar://problem/21715050

9:36 PM Changeset in webkit [187056] by matthew_hanson@apple.com
  • 12 edits
    1 copy
    6 adds in branches/safari-601.1-branch

Merge r186858. rdar://problem/21643094

9:36 PM Changeset in webkit [187055] by matthew_hanson@apple.com
  • 6 edits in branches/safari-601.1-branch/Source/WebCore

Merge r186854. rdar://problem/21840147

7:00 PM Changeset in webkit [187054] by Simon Fraser
  • 2 edits in trunk/Source/WebKit2

Facebook in tiled fullscreen is slow
​https://bugs.webkit.org/show_bug.cgi?id=147134
rdar://problem/21823349

Reviewed by Tim Horton.

TiledCoreAnimationDrawingArea::scaleViewToFitDocumentIfNeeded() could cause a page to toggle
between two fixed layout sizes differing by a pixel, because of rounding. This would cause
lots of extra layouts and painting.

This happened because the the fixed layout size was computed using ceil(m_webPage.size().width() / viewScale)

  • WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.mm:

(WebKit::TiledCoreAnimationDrawingArea::scaleViewToFitDocumentIfNeeded):

6:08 PM Changeset in webkit [187053] by commit-queue@webkit.org
  • 3 edits
    2 adds in trunk

AX: Selection change as a result of focusing an element may cause Safari to crash
​https://bugs.webkit.org/show_bug.cgi?id=147052
<rdar://problem/21778212>

Patch by Nan Wang <​n_wang@apple.com> on 2015-07-20
Reviewed by Chris Fleizach.

Source/WebCore:

When focusing an element, it may trigger a deferred layout that invalidates the render
element, which will cause axObjectCache() to be a nullptr, and lead to a crash. Fix that
by using a RefPtr to hold the object and also caching the axObjectCache().

Test: platform/mac/accessibility/focus-crash.html

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::setFocused):
(WebCore::AccessibilityRenderObject::setSelectedRows):

LayoutTests:

  • platform/mac/accessibility/focus-crash-expected.txt: Added.
  • platform/mac/accessibility/focus-crash.html: Added.
5:58 PM Changeset in webkit [187052] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebInspectorUI

Web Inspector: console.assert(false, "Message") message is not visible in console
​https://bugs.webkit.org/show_bug.cgi?id=147130

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2015-07-20
Reviewed by Timothy Hatcher.

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

(WebInspector.ConsoleMessageView.prototype._appendMessageTextAndArguments):
We were forgetting to assign the result of args.concat back into args.
In re-addressing this, improve the formatting of output if there is
a string message or not as the second argument to console.assert(), as
that is the common usage.

5:27 PM Changeset in webkit [187051] by achristensen@apple.com
  • 2 edits in trunk/Source/WebCore

Unreviewed build fix after r187049.

  • contentextensions/ContentExtension.cpp:

(WebCore::ContentExtensions::ContentExtension::populateDomainCacheIfNeeded):
Can't shrinkToFit HashSets.

5:23 PM Changeset in webkit [187050] by Wenson Hsieh
  • 5 edits in trunk/Source/WebKit2

Fix crash due to RemoteLayerTreeDisplayRefreshMonitor outliving RemoteLayerTreeDrawingArea
​https://bugs.webkit.org/show_bug.cgi?id=147124
<rdar://problem/21582858>

Reviewed by Simon Fraser.

Refactors RemoteLayerTreeDisplayRefreshMonitor to use a weak pointer rather than a reference
to its RemoteLayerTreeDrawingArea, since the drawing area may be deallocated before the monitor
in some rare cases. This rarely caused pages using requestAnimationFrame to crash on iOS. However,
this should not be the case: logically, a RemoteLayerTreeDrawingArea should always outlive its
refresh monitors. Refer to ​https://bugs.webkit.org/show_bug.cgi?id=147128 for more details.

  • WebProcess/WebPage/Cocoa/RemoteLayerTreeDisplayRefreshMonitor.h:
  • WebProcess/WebPage/Cocoa/RemoteLayerTreeDisplayRefreshMonitor.mm:

(WebKit::RemoteLayerTreeDisplayRefreshMonitor::RemoteLayerTreeDisplayRefreshMonitor):
(WebKit::RemoteLayerTreeDisplayRefreshMonitor::~RemoteLayerTreeDisplayRefreshMonitor): On destruction, checks

first to see whether or not the drawing area has been deallocated before telling it to update its monitors.

(WebKit::RemoteLayerTreeDisplayRefreshMonitor::requestRefreshCallback):

  • WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.h:

(WebKit::RemoteLayerTreeDrawingArea::createWeakPtr): Creates and returns a new weak pointer to itself.

  • WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.mm:

(WebKit::RemoteLayerTreeDrawingArea::RemoteLayerTreeDrawingArea):

5:12 PM Changeset in webkit [187049] by achristensen@apple.com
  • 16 edits in trunk/Source

[Content Extensions] Cache actions with domains that match everything
​https://bugs.webkit.org/show_bug.cgi?id=147050

Reviewed by Benjamin Poulain.

Source/WebCore:

If you have lots of rules with triggers like {"url-filter":".*","if-domain":*webkit.org}
we will spend a lot of time adding unnecessary actions to HashSets when you are not on webkit.org.
Caching all the rules and only adding them to a collection once when the domain changes saves a lot of URL interpreting time.

We already had an optimization for the css-display-none actions with triggers that matched everything using a special bytecode operation.
This removes the need for a special bytecode operation by caching the triggers that match everything with and without domains,
then from those cached actions we compile a stylesheet, and create cached actions for every new domain we visit.

All functionality is covered by existing API tests.

  • contentextensions/CompiledContentExtension.cpp:

(WebCore::ContentExtensions::CompiledContentExtension::~CompiledContentExtension):
(WebCore::ContentExtensions::CompiledContentExtension::globalDisplayNoneSelectors): Deleted.

  • contentextensions/CompiledContentExtension.h:
  • contentextensions/ContentExtension.cpp:

(WebCore::ContentExtensions::ContentExtension::ContentExtension):
(WebCore::ContentExtensions::ContentExtension::findFirstIgnorePreviousRules):
(WebCore::ContentExtensions::ContentExtension::globalDisplayNoneStyleSheet):
(WebCore::ContentExtensions::ContentExtension::compileGlobalDisplayNoneStyleSheet):
(WebCore::ContentExtensions::ContentExtension::populateDomainCacheIfNeeded):
(WebCore::ContentExtensions::ContentExtension::cachedDomainActions):
(WebCore::ContentExtensions::ContentExtension::universalActionsWithDomains):

  • contentextensions/ContentExtension.h:

(WebCore::ContentExtensions::ContentExtension::compiledExtension):
(WebCore::ContentExtensions::ContentExtension::universalActionsWithoutDomains):

  • contentextensions/ContentExtensionCompiler.cpp:

(WebCore::ContentExtensions::compileRuleList):

  • contentextensions/ContentExtensionRule.cpp:

(WebCore::ContentExtensions::ContentExtensionRule::ContentExtensionRule):
(WebCore::ContentExtensions::Action::deserialize):
(WebCore::ContentExtensions::Action::deserializeType):
(WebCore::ContentExtensions::Action::serializedLength):

  • contentextensions/ContentExtensionRule.h:

(WebCore::ContentExtensions::Action::operator==):
(WebCore::ContentExtensions::Action::setExtensionIdentifier):
(WebCore::ContentExtensions::Action::extensionIdentifier):

  • contentextensions/ContentExtensionsBackend.cpp:

(WebCore::ContentExtensions::ContentExtensionsBackend::actionsForResourceLoad):

  • contentextensions/DFABytecode.h:

(WebCore::ContentExtensions::instructionSizeWithArguments):

  • contentextensions/DFABytecodeCompiler.cpp:

(WebCore::ContentExtensions::DFABytecodeCompiler::emitAppendAction):

  • contentextensions/DFABytecodeInterpreter.cpp:

(WebCore::ContentExtensions::DFABytecodeInterpreter::interpretAppendAction):
(WebCore::ContentExtensions::DFABytecodeInterpreter::interpretTestFlagsAndAppendAction):
(WebCore::ContentExtensions::DFABytecodeInterpreter::actionsMatchingEverything):
(WebCore::ContentExtensions::DFABytecodeInterpreter::interpret):
(WebCore::ContentExtensions::DFABytecodeInterpreter::actionsForDefaultStylesheetFromDFARoot): Deleted.

  • contentextensions/DFABytecodeInterpreter.h:
  • loader/ResourceLoadInfo.h:

Source/WebKit2:

  • UIProcess/API/APIUserContentExtensionStore.h:
4:48 PM Changeset in webkit [187048] by Lucas Forschler
  • 1 edit in branches/safari-600.1.4.17-branch/Source/WebCore/platform/network/cf/CookieJarCFNet.cpp

Follow-up fix after r187042.

4:45 PM Changeset in webkit [187047] by andersca@apple.com
  • 7 edits in trunk/Source/WebKit2

Support blocking a plug-in for non-security reasons
​https://bugs.webkit.org/show_bug.cgi?id=145009

Patch by Gordon Sheridan <​gordon_sheridan@apple.com> on 2015-07-20
Reviewed by Anders Carlsson.

  • Shared/Plugins/PluginModuleInfo.h:

Replace PluginModuleBlocked with separate enum values for blocking for security and compatibility.

  • UIProcess/API/C/WKAPICast.h:

(WebKit::toWKPluginLoadPolicy):
Modify to handle change to PluginModuleBlocked enum.

(WebKit::toPluginModuleLoadPolicy):
Ditto.

  • UIProcess/API/C/WKPluginLoadPolicy.h:

Added enum value for blocking a plugin for compatibility reasons.

  • UIProcess/Plugins/mac/PluginInfoStoreMac.mm:

(WebKit::shouldBlockPlugin):
Check load policy for both reasons a plugin may be blocked.

(WebKit::WKPlugInModuleLoadPolicyToPluginModuleLoadPolicy):
Added function to convert between load policy enum types.

(WebKit::PluginInfoStore::defaultLoadPolicyForPlugin):
Modified to call WKLoadPolicyForPluginVersion() which can distinguish between blocked for security or compatibility.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::findPlugin):
Updated to recognize both PluginModuleBlocked enum values that indicate a plugin is blocked.

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::createPlugin):
Ditto.

(WebKit::WebPage::canPluginHandleResponse):
Ditto.

4:35 PM Changeset in webkit [187046] by Brian Burg
  • 4 edits
    4 adds in trunk/LayoutTests

Web Inspector: start using Promises to handle asynchronous steps in protocol tests
​https://bugs.webkit.org/show_bug.cgi?id=147096

Reviewed by Joseph Pecoraro.

Existing protocol tests use lots of reinvented wheels to run
multiple asynchronous tests. This patch begins consolidation of
these redundant wheels, using a tiny test framework based on
promises.

The first change is to introduce InspectorTest.awaitCommand,
which returns a promise that is fulfilled or rejected if the command
succeeds or fails. This is the building block for writing protocol
tests with promises.

The second change is the introduction of InspectorTest.AsyncTestSuite.
This properly chains together multiple asynchronous test methods
using the promise interface. A synchronous version is also added.

To demonstrate how async tests can be written using this new approach,
this patch converts existing protocol test. Test coverage for the
new frameworks also demonstrates their use.

  • http/tests/inspector-protocol/resources/InspectorTest.js:

(InspectorTest.sendCommand):
(InspectorTest.awaitCommand):
(InspectorTest.sendMessage):
(InspectorFrontendAPI.dispatchMessageAsync):
(InspectorTest.AsyncTestSuite): Add tests for this micro framework.
(InspectorTest.AsyncTestSuite.prototype.get passCount):
(InspectorTest.AsyncTestSuite.prototype.get skipCount):
(InspectorTest.AsyncTestSuite.prototype.addTestCase):
(InspectorTest.AsyncTestSuite.prototype.runTestCasesAndFinish.finish):
(InspectorTest.AsyncTestSuite.prototype.runTestCasesAndFinish):
(InspectorTest.AsyncTestSuite.prototype.runTestCases):
(InspectorTest.SyncTestSuite): Add tests for this micro framework.
(InspectorTest.SyncTestSuite.prototype.get passCount):
(InspectorTest.SyncTestSuite.prototype.get skipCount):
(InspectorTest.SyncTestSuite.prototype.addTestCase):
(InspectorTest.SyncTestSuite.prototype.runTestCasesAndFinish):
(InspectorTest.SyncTestSuite.prototype.runTestCases):
(InspectorTest.log):
(InspectorTest.assert):

  • inspector-protocol/runtime/getProperties-expected.txt:
  • inspector-protocol/runtime/getProperties.html: Rewrite.
4:31 PM Changeset in webkit [187045] by Antti Koivisto
  • 2 edits in trunk/Source/WebKit2

NSURLCache fallback does not work on iOS due to sandboxing
​https://bugs.webkit.org/show_bug.cgi?id=146314
<rdar://problem/21433691>

Reviewed by Anders Carlsson.

This broke with earlier cache path computation changes.

  • NetworkProcess/cocoa/NetworkProcessCocoa.mm:

(WebKit::NetworkProcess::platformInitializeNetworkProcessCocoa):

NSURLCache path is relative to the container cache path so "." is sufficient.
This puts the cache files under
<container>/Library/Caches/com.apple.WebKit.Networking/

4:27 PM Changeset in webkit [187044] by commit-queue@webkit.org
  • 14 edits in trunk/Source

Adopt AVPlayerLayerView
​https://bugs.webkit.org/show_bug.cgi?id=146862

Patch by Jeremy Jones <​jeremyj@apple.com> on 2015-07-20
Source/WebCore:

Reviewed by Jer Noble.

The CALayerHost is replaced with WebLayerHostView to be compatible with UIView animations.
Some animation is improved in the conversion.

WebAVPlayerLayerView and WebAVPictureInPicturePlayerLayerView derive from AVKit and UIKit respectively.
Because these frameworks are loaded at runtime, these classes must be generate using objc/runtime.h to
register them from c functions at runtime. The most important part of these UIViews is that their
backing layer is a WebAVPlayerLayer.

WebCALayerHostWrapper and WebAVVideoLayer are combined into WebAVPlayerLayer to simplify the hierarchy.
WebAVPlayerLayer is a stand-in for an AVPlayerLayer.

  • platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:

(WebCore::MediaPlayerPrivateAVFoundationObjC::setVideoFullscreenLayer): -removeFromSuperlayer is redundant.
(WebCore::MediaPlayerPrivateAVFoundationObjC::setVideoFullscreenGravity): syncTextTrackBounds on change.
(WebCore::MediaPlayerPrivateAVFoundationObjC::setVideoFullscreenFrame):
We never want animation here, since all animation will happen in UIViews.
This is just for going into the final size after a transform based animation.

  • platform/ios/WebVideoFullscreenControllerAVKit.mm:

(WebVideoFullscreenControllerContext::didSetupFullscreen): layer -> view
(WebVideoFullscreenControllerContext::didCleanupFullscreen): layer -> view
(WebVideoFullscreenControllerContext::setUpFullscreen): layer -> view

(WebVideoFullscreenControllerContext::setVideoLayerFrame): layer -> view
Use fence port to synchronize between the UIThread and the WebThread,
the same way WebKit2 uses a fence port to synchronize between processes.

(WebVideoFullscreenControllerContext::setVideoLayerGravity):
No longer necessary to cache videoGravity at this level.

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

(-[WebAVPlayerLayer init]):
(-[WebAVPlayerLayer dealloc]):
(-[WebAVPlayerLayer playerController]):
(-[WebAVPlayerLayer setBounds:]):
(-[WebAVPlayerLayer resolveBounds]):
(-[WebAVPlayerLayer setVideoGravity:]):
(-[WebAVPlayerLayer videoGravity]):
(-[WebAVPlayerLayer videoRect]):
(+[WebAVPlayerLayer keyPathsForValuesAffectingVideoRect]):
Added class WebAVPlayerLayer, replacing WebAVVideoLayer and WebCALayerHostWrapper.

(WebAVPictureInPicturePlayerLayerView_layerClass):
(getWebAVPictureInPicturePlayerLayerViewClass):
Added runtime class WebAVPictureInPicturePlayerLayerView

(WebAVPlayerLayerView_layerClass):
(WebAVPlayerLayerView_playerController):
(WebAVPlayerLayerView_setPlayerController):
(WebAVPlayerLayerView_videoView):
(WebAVPlayerLayerView_setVideoView):
(WebAVPlayerLayerView_startRoutingVideoToPictureInPicturePlayerLayerView):
(WebAVPlayerLayerView_stopRoutingVideoToPictureInPicturePlayerLayerView):
(WebAVPlayerLayerView_pictureInPicturePlayerLayerView):
(getWebAVPlayerLayerViewClass):
Added runtime class WebAVPlayerLayerView

(WebVideoFullscreenInterfaceAVKit::setVideoDimensions):
Dimensions are also stored in WebAVPlayerLayer so it can make decisions about
animating the video layer.

(WebVideoFullscreenInterfaceAVKit::setExternalPlayback):
(WebVideoFullscreenInterfaceAVKit::enterFullscreen):
(WebVideoFullscreenInterfaceAVKit::didStopPictureInPicture):
(WebVideoFullscreenInterfaceAVKit::cleanupFullscreen):
Straightforward layer to view conversion.

(WebVideoFullscreenInterfaceAVKit::preparedToReturnToInline):
Set view frame using the view hierarchy instead of assuming it is directly in a window.

(WebVideoFullscreenInterfaceAVKit::exitFullscreen):
Set view frame using the view hierarchy instead of assuming it is directly in a window.
dispatch_async before calling didExitFullscreen() to allows CATransactions to complete.

(WebVideoFullscreenInterfaceAVKit::setupFullscreen):
Set view frame using the view hierarchy instead of assuming it is directly in a window.
dispatch_async before calling didSetupFullscreen() to allows CATransactions to complete.

(-[WebCALayerHostWrapper dealloc]): Deleted.
(-[WebCALayerHostWrapper setVideoSublayer:]): Deleted.
(-[WebCALayerHostWrapper videoSublayer]): Deleted.
(-[WebCALayerHostWrapper setBounds:]): Deleted.
(-[WebCALayerHostWrapper resolveBounds]): Deleted.
Class WebCALayerHostWrapper deleted. Functionality rolled into WebAVPlayerLayer.

(+[WebAVVideoLayer videoLayer]): Deleted.
(-[WebAVVideoLayer init]): Deleted.
(-[WebAVVideoLayer setPlayerViewController:]): Deleted.
(-[WebAVVideoLayer setVideoSublayer:]): Deleted.
(-[WebAVVideoLayer setBounds:]): Deleted.
(-[WebAVVideoLayer setVideoLayerGravity:]): Deleted.
(-[WebAVVideoLayer videoLayerGravity]): Deleted.
(-[WebAVVideoLayer enterPIPModeRedirectingVideoToLayer:]): Deleted.
(-[WebAVVideoLayer leavePIPMode]): Deleted.
Class WebAVVideoLayer deleted. Functionality rolled into WebAVPlayerLayer.

  • platform/ios/WebVideoFullscreenModel.h:
  • platform/ios/WebVideoFullscreenModelVideoElement.h:
  • platform/ios/WebVideoFullscreenModelVideoElement.mm:

No need to store frame and gravity in the model. It is stored in the UI where it is used.

(WebVideoFullscreenModelVideoElement::videoLayerFrame): Deleted.
(WebVideoFullscreenModelVideoElement::videoLayerGravity): Deleted.

  • platform/spi/cocoa/AVKitSPI.h: Add AVPlayerLayerView.

Source/WebKit2:

Reviewed by Simon Fraser.

The CALayerHost is replaced with WebLayerHostView to be compatible with UIView animations.
videoLayerFrame and videoLayerGravity no longer need to be stored because they are stored
where they are used in the interface. Some animation is improved in the conversion.

  • UIProcess/ios/WebVideoFullscreenManagerProxy.h:
  • UIProcess/ios/WebVideoFullscreenManagerProxy.mm:

(+[WebLayerHostView layerClass]): Add class WebLayerHostView.
(-[WebLayerHostView contextID]):
(-[WebLayerHostView setContextID:]):
(-[WebLayerHostView layerHost]):
(WebKit::WebVideoFullscreenManagerProxy::invalidate): layer -> view
(WebKit::WebVideoFullscreenManagerProxy::setupFullscreenWithID):
Apply the hostingDeviceScaleFactor transform to -sublayerTransform instead of to
-transform. This more directly inverts the tranform WebProcess and allows -transform
to be used for animation in the UIProcess. This is important because UIView's actions
animate -transform, but not -sublayerTrasform.

(WebKit::WebVideoFullscreenManagerProxy::didCleanupFullscreen): layer -> view
(WebKit::WebVideoFullscreenModelContext::setVideoLayerFrame): Deleted.
(WebKit::WebVideoFullscreenModelContext::videoLayerFrame): Deleted.
(WebKit::WebVideoFullscreenModelContext::setVideoLayerGravity): Deleted.
(WebKit::WebVideoFullscreenModelContext::videoLayerGravity): Deleted.

  • WebProcess/ios/WebVideoFullscreenManager.mm:

(WebKit::WebVideoFullscreenManager::enterVideoFullscreenForVideoElement):
Set initial video layer frame to fix start point of animation.

(WebKit::WebVideoFullscreenManager::didSetupFullscreen):
dispatch_async allows the CATransaction to complete before continuing with the animation.
This prevents a flash during animation.

4:26 PM Changeset in webkit [187043] by andersca@apple.com
  • 7 edits in trunk/Source/WebCore

Add PLATFORM #ifdefs for Mac SPI headers
​https://bugs.webkit.org/show_bug.cgi?id=147129

Reviewed by Tim Horton.

  • Configurations/WebCore.xcconfig:

Skip NPAPI headers on iOS.

  • platform/mac/WebVideoFullscreenController.h:
  • platform/spi/mac/DataDetectorsSPI.h:
  • platform/spi/mac/NSEventSPI.h:
  • platform/spi/mac/NSFontSPI.h:
  • platform/spi/mac/NSMenuSPI.h:
4:24 PM WebInspectorCodingStyleGuide edited by Brian Burg
tweaks to style guide (diff)
4:23 PM Changeset in webkit [187042] by Lucas Forschler
  • 2 edits in branches/safari-600.1.4.17-branch/Source/WebCore

Merged r186895. rdar://problem/21714851

3:52 PM Changeset in webkit [187041] by Lucas Forschler
  • 2 edits in branches/safari-600.1.4.17-branch/Source/WebKit2

Merged r186812. rdar://problem/21877914

3:36 PM Changeset in webkit [187040] by matthew_hanson@apple.com
  • 1 edit in branches/safari-600.1.4.17-branch/Tools/WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp

Build fix after r186983.

3:33 PM Changeset in webkit [187039] by timothy_horton@apple.com
  • 4 edits in trunk/Source

REGRESSION (r174287): Flash of black when opening a new web view or navigating to a new page
​https://bugs.webkit.org/show_bug.cgi?id=147127
<rdar://problem/21474317>

Reviewed by Simon Fraser.

  • rendering/RenderView.cpp:

(WebCore::RenderView::paintBoxDecorations):

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::snapshotAtSize):
Avoid using an invalid documentBackgroundColor, fall back to baseBackgroundColor
like we did before r174287.

2:48 PM Changeset in webkit [187038] by commit-queue@webkit.org
  • 4 edits in trunk/WebKitLibraries

WKSI support for blocking a plug-in for non-security reasons.
​https://bugs.webkit.org/show_bug.cgi?id=145009

Patch by Gordon Sheridan <​gordon_sheridan@apple.com> on 2015-07-20
Reviewed by Anders Carlsson.

  • WebKitSystemInterface.h:
  • libWebKitSystemInterfaceMavericks.a:
  • libWebKitSystemInterfaceYosemite.a:
2:43 PM Changeset in webkit [187037] by achristensen@apple.com
  • 2 edits in trunk/Source/WebCore

[Win] Unreviewed build fix after r187022.

  • platform/win/WindowsTouch.h:

Undid changes from r187022

2:37 PM Changeset in webkit [187036] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

Allow video to rotate when app doesnt allow rotation.
​https://bugs.webkit.org/show_bug.cgi?id=147121

Patch by Jeremy Jones <​jeremyj@apple.com> on 2015-07-20
Reviewed by Jer Noble.

Set an SPI bool on the fullscreen video root view controller to allow it to override app rotation restrictions.
This allows video to be played in landscape in portrait only apps.

  • platform/ios/WebVideoFullscreenInterfaceAVKit.mm:

(WebVideoFullscreenInterfaceAVKit::setupFullscreen):

2:28 PM Changeset in webkit [187035] by akling@apple.com
  • 2 edits in trunk/Source/WebCore

Unreviewed assertion fix following r187031.

Just have GenericEventQueue::resume() return early if the queue isn't suspended.
Suspend/Resume calls are not always symmetrical in HTMLMediaElement.

  • dom/GenericEventQueue.cpp:

(WebCore::GenericEventQueue::resume):

2:20 PM Changeset in webkit [187034] by ddkilzer@apple.com
  • 2 edits in trunk/Source/WebCore

Unexpected ABI diffference between armv7 and armv7s in WebCore::GraphicsLayerCA::~GraphicsLayerCA()
<​http://webkit.org/b/147101>

Reviewed by Michael Saboff.

  • platform/graphics/transforms/TransformationMatrix.h: Align

Matrix4 on both armv7 and armv7s.

2:16 PM Changeset in webkit [187033] by saambarati1@gmail.com
  • 12 edits in trunk/Source/JavaScriptCore

"let" scoping introduced incoherent story about symbol table cloning
​https://bugs.webkit.org/show_bug.cgi?id=147046

Reviewed by Filip Pizlo.

This patch now establishes a clear set of rules for how SymbolTables
are owned by CodeBlock. Every SymbolTable that is used by a bytecode
instruction must live in CodeBlock's constant register pool. When CodeBlock
is being linked, it ensures that every SymbolTable in the constant pool is cloned.
This leaves no room for an un-cloned symbol table to be used by a bytecode instruction.
Some instructions may refer to SymbolTable's indirectly through a JSLexicalEnvironment.
This is fine, all JSLexicalEnvironment's are allocated with references to cloned symbol tables.

Another goal of this patch is to remove the notion that a SymbolTable is 1 to 1
with a CodeBlock. With lexical scoping, this view of the world is no longer
correct. This patch begins to remove this assumption by making CodeBlock's
symbolTable() getter method private. There is still one place where we need
to purge our codebase of this assumption and that is the type profiler. It
has not been updated for lexical scoping. After it is updated in
​https://bugs.webkit.org/show_bug.cgi?id=145438
we will be able to remove CodeBlock's symbolTable() getter entirely.

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::CodeBlock):
(JSC::CodeBlock::nameForRegister):

  • bytecode/CodeBlock.h:

(JSC::CodeBlock::addStringSwitchJumpTable):
(JSC::CodeBlock::stringSwitchJumpTable):
(JSC::CodeBlock::evalCodeCache):
(JSC::CodeBlock::symbolTable):

  • bytecode/UnlinkedCodeBlock.cpp:

(JSC::UnlinkedFunctionExecutable::visitChildren):
(JSC::UnlinkedFunctionExecutable::link):
(JSC::UnlinkedFunctionExecutable::codeBlockFor):

  • bytecode/UnlinkedCodeBlock.h:

(JSC::UnlinkedCodeBlock::addExceptionHandler):
(JSC::UnlinkedCodeBlock::exceptionHandler):
(JSC::UnlinkedCodeBlock::setSymbolTableConstantIndex):
(JSC::UnlinkedCodeBlock::symbolTableConstantIndex):
(JSC::UnlinkedCodeBlock::symbolTable): Deleted.
(JSC::UnlinkedCodeBlock::setSymbolTable): Deleted.

  • bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::generate):
(JSC::BytecodeGenerator::BytecodeGenerator):
(JSC::BytecodeGenerator::pushLexicalScope):
(JSC::BytecodeGenerator::variableForLocalEntry):
(JSC::BytecodeGenerator::createVariable):
(JSC::BytecodeGenerator::resolveType):
(JSC::BytecodeGenerator::emitResolveScope):

  • bytecompiler/BytecodeGenerator.h:

(JSC::BytecodeGenerator::thisRegister):
(JSC::BytecodeGenerator::instructions):
(JSC::BytecodeGenerator::symbolTable): Deleted.

  • dfg/DFGGraph.h:

(JSC::DFG::Graph::baselineCodeBlockFor):
(JSC::DFG::Graph::isStrictModeFor):
(JSC::DFG::Graph::symbolTableFor): Deleted.

  • jit/AssemblyHelpers.h:

(JSC::AssemblyHelpers::baselineCodeBlock):
(JSC::AssemblyHelpers::argumentsStart):
(JSC::AssemblyHelpers::symbolTableFor): Deleted.

  • runtime/CommonSlowPaths.cpp:

(JSC::SLOW_PATH_DECL):

  • runtime/Executable.cpp:

(JSC::FunctionExecutable::visitChildren):
(JSC::FunctionExecutable::clearUnlinkedCodeForRecompilation):
(JSC::FunctionExecutable::symbolTable): Deleted.

  • runtime/Executable.h:
1:45 PM Changeset in webkit [187032] by andersca@apple.com
  • 4 edits in trunk/Source/WebCore

Add the necessary platform checks to iOS SPI headers
​https://bugs.webkit.org/show_bug.cgi?id=147120

Reviewed by Tim Horton.

  • platform/spi/ios/GraphicsServicesSPI.h:
  • platform/spi/ios/MediaPlayerSPI.h:
  • platform/spi/ios/MobileGestaltSPI.h:
1:16 PM Changeset in webkit [187031] by akling@apple.com
  • 9 edits in trunk

Improve behavior of media elements in page cache.
<​https://webkit.org/b/147020>
<rdar://problem/21712311>

Reviewed by Chris Dumez.

Source/WebCore:

Make improvements for media elements when transitioning in/out of page cache:

  • Events that were scheduled when going into cache will now be delivered when the page is restored from cache.
  • Data buffering is turned off while in the cache. This reduces the memory cost of cached pages with media elements on iOS (where mediaserverd would keep upcoming video frames in memory for cached pages.)

Test: media/restore-from-page-cache.html (amended)

  • dom/GenericEventQueue.h:
  • dom/GenericEventQueue.cpp:

(WebCore::GenericEventQueue::enqueueEvent):
(WebCore::GenericEventQueue::suspend):
(WebCore::GenericEventQueue::resume):

Add a simple suspend/resume mechanism to GenericEventQueue that can
be used to support page caching.

  • html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::stop):
(WebCore::HTMLMediaElement::suspend):
(WebCore::HTMLMediaElement::resume):
(WebCore::HTMLMediaElement::stopWithoutDestroyingMediaPlayer):

Adapt to event queueing changes and add calls to setShouldBufferData().

  • html/HTMLSourceElement.h:
  • html/HTMLSourceElement.cpp:

(WebCore::HTMLSourceElement::HTMLSourceElement):
(WebCore::HTMLSourceElement::create):
(WebCore::HTMLSourceElement::activeDOMObjectName):
(WebCore::HTMLSourceElement::canSuspendForPageCache):
(WebCore::HTMLSourceElement::suspend):
(WebCore::HTMLSourceElement::resume):
(WebCore::HTMLSourceElement::stop):

Turn HTMLSourceElement into an ActiveDOMObject so it gets all the
appropriate page cache notifications directly. Suspend the delayed
error event delivery timer when cached.

LayoutTests:

Add some coverage for suspend/resume of queued events on cached media elements.

  • media/restore-from-page-cache-expected.txt:
  • media/restore-from-page-cache.html:
1:06 PM Changeset in webkit [187030] by Michael Catanzaro
  • 5 edits in trunk/Source/WebKit2

[Seccomp] Should be easier to debug blocked syscalls
​https://bugs.webkit.org/show_bug.cgi?id=142980

These should be printed even when not running in debug mode. There is no
value in hiding errors from release build users.

Reviewed by Žan Doberšek.

  • Shared/linux/SeccompFilters/SeccompBroker.cpp:

(WebKit::SeccompBroker::runLoop): Don't close stderr et. al. in release builds.

  • Shared/linux/SeccompFilters/Syscall.cpp:

(WebKit::write_uint): Added.
(WebKit::reportUnexpectedSyscall): Added.
(WebKit::Syscall::createFromContext): Call reportUnexpectedSyscall. Also, no need to crash
here in release builds.

  • Shared/linux/SeccompFilters/SyscallPolicy.cpp:

(WebKit::SyscallPolicy::hasPermissionForPath): Print a warning when access is denied.
(WebKit::SyscallPolicy::permissionToString): Added.

  • Shared/linux/SeccompFilters/SyscallPolicy.h: Add permissionToString.
12:50 PM Changeset in webkit [187029] by Csaba Osztrogonác
  • 2 edits in trunk/Source/WebKit2

Fix the !ENABLE(VIDEO) build after r186396
​https://bugs.webkit.org/show_bug.cgi?id=147116

Reviewed by Brent Fulgham.

  • WebProcess/WebProcess.cpp:

(WebKit::WebProcess::didClose):

12:04 PM Changeset in webkit [187028] by fpizlo@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

REGRESSION(186691): OSR entry is broken on loop headers that have no live variables
​https://bugs.webkit.org/show_bug.cgi?id=147074
rdar://problem/21869970

Reviewed by Michael Saboff.

The OSR entry must-handle block/value widening introduced in r186691 would cause the
CFA to reexecute if it caused any live local variables to change value. But this fails
if the must-handle block has no live local variables, and the entry block otherwise
appears to be unreachable.

This fixes the bug by having the change detection include whether the block hadn't been
visited in addition to whether any local variable values got widened.

This is a ~4% speed-up on SunSpider in browser.

  • dfg/DFGCFAPhase.cpp:

(JSC::DFG::CFAPhase::run):

11:51 AM Changeset in webkit [187027] by Brent Fulgham
  • 2 edits in trunk/Source/WebKit

[Win] Certain Debug builds can take close to 3 hours to link
​https://bugs.webkit.org/show_bug.cgi?id=147114
<rdar://problem/21882777>

Reviewed by Tim Horton.

Correct the 64-bit DebugSuffix link target to match the 32-bit version.

  • WebKit.vcxproj/WebKit/WebKit.vcxproj: Turn off the REF and ICF

flags for the 64-bit DebugSuffix link target.

11:12 AM Changeset in webkit [187026] by mark.lam@apple.com
  • 10 edits in trunk/Source

Rollout r187020 and r187021: breaks JSC API tests on debug builds.
​https://bugs.webkit.org/show_bug.cgi?id=147110

Source/JavaScriptCore:

  • heap/MachineStackMarker.cpp:

(JSC::MachineThreads::addCurrentThread):

  • runtime/JSLock.cpp:

(JSC::JSLockHolder::~JSLockHolder):
(JSC::JSLock::JSLock):
(JSC::JSLock::willDestroyVM):
(JSC::JSLock::setExclusiveThread):
(JSC::JSLock::lock):
(JSC::JSLock::unlock):
(JSC::JSLock::currentThreadIsHoldingLock):
(JSC::JSLock::dropAllLocks):

  • runtime/JSLock.h:

(JSC::JSLock::vm):
(JSC::JSLock::hasExclusiveThread):
(JSC::JSLock::exclusiveThread):

  • runtime/VM.h:

(JSC::VM::hasExclusiveThread):
(JSC::VM::exclusiveThread):
(JSC::VM::setExclusiveThread):

Source/WebCore:

  • bindings/js/JSDOMWindowBase.cpp:

(WebCore::JSDOMWindowBase::commonVM):

Source/WTF:

  • wtf/WTFThreadData.cpp:

(WTF::WTFThreadData::WTFThreadData):

  • wtf/WTFThreadData.h:

(WTF::WTFThreadData::stack):

11:05 AM Changeset in webkit [187025] by adachan@apple.com
  • 2 edits in trunk/Source/WebCore

Fix deadlock between -[AVPlayerItem currentTime] and -[AVPlayer isExternalPlaybackActive]
​https://bugs.webkit.org/show_bug.cgi?id=147085
<rdar://problem/21878275>

Reviewed by Jer Noble.

  • Modules/webaudio/AudioContext.cpp:

(WebCore::AudioContext::isPlayingAudioDidChange):
Call Document::updateIsPlayingMedia() on the main thread, since we could be on the audio I/O
thread here and the Document::updateIsPlayingMedia() call could block, causing a deadlock.

10:42 AM Changeset in webkit [187024] by commit-queue@webkit.org
  • 9 edits
    2 copies
    3 adds in trunk

[GTK] Add selection-changed signal to the WebKit2 API
​https://bugs.webkit.org/show_bug.cgi?id=137116

Patch by Tomas Popela <​tpopela@redhat.com> on 2015-07-20
Reviewed by Carlos Garcia Campos.

Source/WebKit2:

Add a new object WebKitWebEditor where the editing capabilities of a
WebPage will be grouped and exposed. Add the selection-changed signal
there (we used the same name as in WK1 API).

  • PlatformGTK.cmake:
  • UIProcess/API/gtk/docs/webkit2gtk-4.0-sections.txt:
  • WebProcess/InjectedBundle/API/gtk/WebKitWebEditor.cpp: Added.

(didChangeSelection):
(webkitWebEditorCreate):
(webkit_web_editor_get_page):

  • WebProcess/InjectedBundle/API/gtk/WebKitWebEditor.h: Added.
  • WebProcess/InjectedBundle/API/gtk/WebKitWebEditorPrivate.h: Added.
  • WebProcess/InjectedBundle/API/gtk/WebKitWebPage.cpp:

(webkitWebPageGetPage):
(webkit_web_page_get_editor):

  • WebProcess/InjectedBundle/API/gtk/WebKitWebPage.h:
  • WebProcess/InjectedBundle/API/gtk/WebKitWebPagePrivate.h:
  • WebProcess/InjectedBundle/API/gtk/webkit-web-extension.h:

Tools:

Add a new WebProcess test TestEditor that is testing the WebKitWebEditor
(namely the selection-changed signal).

  • TestWebKitAPI/Tests/WebKit2Gtk/CMakeLists.txt:
  • TestWebKitAPI/Tests/WebKit2Gtk/EditorTest.cpp: Added.

(WebKitWebEditorTest::create):
(WebKitWebEditorTest::selectionChangedCallback):
(WebKitWebEditorTest::testSelectionSelectAll):
(WebKitWebEditorTest::testSelectionCollapse):
(WebKitWebEditorTest::testSelectionModifyMove):
(WebKitWebEditorTest::testSelectionModifyExtend):
(WebKitWebEditorTest::testSelectionUnselect):
(registerTests):

  • TestWebKitAPI/Tests/WebKit2Gtk/TestEditor.cpp: Added.

(testWebKitWebEditorSelectionChanged):
(beforeAll):
(afterAll):

10:08 AM Changeset in webkit [187023] by andersca@apple.com
  • 3 edits in trunk/Source/WebKit2

WebsiteDataStore operations need to grab background assertions
​https://bugs.webkit.org/show_bug.cgi?id=147015
rdar://problem/21799011

Reviewed by Sam Weinig.

  • UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::fetchWebsiteData):
(WebKit::NetworkProcessProxy::deleteWebsiteData):
(WebKit::NetworkProcessProxy::deleteWebsiteDataForOrigins):

  • UIProcess/WebProcessProxy.cpp:

(WebKit::WebProcessProxy::fetchWebsiteData):
(WebKit::WebProcessProxy::deleteWebsiteData):
(WebKit::WebProcessProxy::deleteWebsiteDataForOrigins):

8:55 AM Changeset in webkit [187022] by achristensen@apple.com
  • 19 edits
    2 adds in trunk

Resurrect CMake build on Windows.
​https://bugs.webkit.org/show_bug.cgi?id=147083

Reviewed by Gyuyoung Kim.

.:

  • Source/cmake/OptionsCommon.cmake:
  • Source/cmake/OptionsWindows.cmake:

Change features to get it to compile. Still not a complete feature set.

Source/WebCore:

  • CMakeLists.txt:
  • PlatformWin.cmake:
  • dom/Document.cpp:

(WebCore::Document::~Document):

  • dom/Range.cpp:
  • dom/Text.cpp:
  • editing/CompositeEditCommand.cpp:
  • platform/graphics/texmap/BitmapTexturePool.cpp:

(WebCore::BitmapTexturePool::createTexture):

  • platform/win/WindowsTouch.h:
  • rendering/RenderThemeWin.h:

Fixed some including and compiling quirks needed for different configurations.

Source/WebKit:

  • PlatformWin.cmake: Added.

Based on original work by Patrick Gansterer and Mark Salisbury
in ​https://bugs.webkit.org/show_bug.cgi?id=72816

Source/WTF:

  • wtf/PlatformWin.cmake:

Tools:

  • DumpRenderTree/CMakeLists.txt:
  • DumpRenderTree/PlatformWin.cmake: Added.
  • WinLauncher/CMakeLists.txt:
5:04 AM Changeset in webkit [187021] by peavo@outlook.com
  • 2 edits in trunk/Source/JavaScriptCore

Unreviewed debug build fix after r187020.

  • heap/MachineStackMarker.cpp:

(JSC::MachineThreads::addCurrentThread):
VM::exclusiveThread() has changed return type to ThreadIdentifier.

2:14 AM Changeset in webkit [187020] by peavo@outlook.com
  • 9 edits in trunk/Source

JavaScriptCore performance is very bad on Windows
​https://bugs.webkit.org/show_bug.cgi?id=146448

Reviewed by Mark Lam.

Source/JavaScriptCore:

Profiling shows that std::this_thread::get_id() is slow on Windows.
Use WTF::currentThread() instead, which calls GetCurrentThreadId().
This is faster on Windows. The issue has been reported to Microsoft,
​https://connect.microsoft.com/VisualStudio/feedback/details/1558211.

  • runtime/JSLock.cpp:

(JSC::JSLockHolder::~JSLockHolder):
(JSC::JSLock::JSLock):
(JSC::JSLock::willDestroyVM):
(JSC::JSLock::setExclusiveThread):
(JSC::JSLock::lock):
(JSC::JSLock::unlock):
(JSC::JSLock::currentThreadIsHoldingLock):

  • runtime/JSLock.h:

(JSC::JSLock::vm):
(JSC::JSLock::hasExclusiveThread):
(JSC::JSLock::exclusiveThread):

  • runtime/VM.h:

(JSC::VM::hasExclusiveThread):
(JSC::VM::exclusiveThread):
(JSC::VM::setExclusiveThread):

Source/WebCore:

  • bindings/js/JSDOMWindowBase.cpp:

(WebCore::JSDOMWindowBase::commonVM): Compile fix.

Source/WTF:

Updating the stack bounds is time consuming.
Only update the stack bounds when a new fiber is running.

  • wtf/WTFThreadData.cpp:

(WTF::WTFThreadData::WTFThreadData):

  • wtf/WTFThreadData.h:

(WTF::WTFThreadData::stack):

12:19 AM Changeset in webkit [187019] by Julien Brianceau
  • 2 edits in trunk/Source/WTF

Use isnan from std namespace in Stopwatch.h.
​https://bugs.webkit.org/show_bug.cgi?id=146911

Reviewed by Žan Doberšek.

  • wtf/Stopwatch.h:

(WTF::Stopwatch::isActive):
(WTF::Stopwatch::start):
(WTF::Stopwatch::stop):
(WTF::Stopwatch::elapsedTime):

Jul 19, 2015:

11:38 PM Changeset in webkit [187018] by timothy_horton@apple.com
  • 8 edits
    1 add
    2 deletes in trunk

Make shrink-wrapping test a ref-test instead of pixel-test
​https://bugs.webkit.org/show_bug.cgi?id=147081

Reviewed by Sam Weinig.

Source/WebCore:

Nobody runs pixel tests.

  • svg/SVGPathUtilities.cpp:

(WebCore::pathIteratorForBuildingString):
(WebCore::buildStringFromPath):

  • svg/SVGPathUtilities.h:

Add a helper that turns a Path into a SVG path string.

  • testing/Internals.cpp:

(WebCore::Internals::pathStringWithShrinkWrappedRects):

  • testing/Internals.h:
  • testing/Internals.idl:

Have the internals shrink-wrap method return an SVG path instead of a DOMPath.

LayoutTests:

  • fast/shrink-wrap/rect-shrink-wrap-expected.html: Added.
  • fast/shrink-wrap/rect-shrink-wrap-expected.png: Removed.
  • fast/shrink-wrap/rect-shrink-wrap-expected.txt: Removed.
  • fast/shrink-wrap/rect-shrink-wrap.html:

SVG all the things.

4:37 PM Changeset in webkit [187017] by Yusuke Suzuki
  • 2 edits
    1 add in trunk/Source/JavaScriptCore

In strict mode, Object.keys(arguments) includes "length"
​https://bugs.webkit.org/show_bug.cgi?id=147071

Reviewed by Darin Adler.

ClonedAguments didn't set the "length" with DontEnum.

  • runtime/ClonedArguments.cpp:

(JSC::ClonedArguments::createWithInlineFrame):
(JSC::ClonedArguments::createByCopyingFrom):

  • tests/stress/arguments-length-always-dont-enum.js: Added.

(shouldBe):
(argsSloppy):
(argsStrict):

4:10 PM Changeset in webkit [187016] by commit-queue@webkit.org
  • 5 edits in trunk

new Date(NaN).toJSON() must return null instead of throwing a TypeError
​https://bugs.webkit.org/show_bug.cgi?id=141115

Patch by Jordan Harband <​ljharb@gmail.com> on 2015-07-19
Reviewed by Yusuke Suzuki.

Source/JavaScriptCore:

  • runtime/DatePrototype.cpp:

(JSC::dateProtoFuncToJSON):

LayoutTests:

  • js/dom/JSON-stringify-expected.txt:
  • js/resources/JSON-stringify.js:
3:27 PM Changeset in webkit [187015] by Michael Catanzaro
  • 2 edits in trunk/Tools

[GTK] Enable seccomp filter API tests
​https://bugs.webkit.org/show_bug.cgi?id=140071

Reviewed by Žan Doberšek.

These tests will now be run by the run-gtk-tests script.

  • TestWebKitAPI/PlatformGTK.cmake:
12:31 PM Changeset in webkit [187014] by saambarati1@gmail.com
  • 3 edits
    3 adds in trunk

Parser::parseFunctionInfo hits RELEASE_ASSERT for Arrow Functions
​https://bugs.webkit.org/show_bug.cgi?id=147090

Reviewed by Yusuke Suzuki.

Source/JavaScriptCore:

ArrowFunction's have there ParserFunctionInfo "name" field to
be a non-null pointer. This is obviously allowed and valid except we
had a RELEASE_ASSERT that claimed otherwise. This is a mistake.

Note: ArrowFunction's will never actually have a function name;
there ParserFunctionInfo "name" field will be the empty string.
This is not be mistaken with the name field being a null pointer.

  • parser/Parser.cpp:

(JSC::Parser<LexerType>::parseFunctionInfo):

LayoutTests:

  • js/arrowfunction-strict-mode-expected.txt: Added.
  • js/arrowfunction-strict-mode.html: Added.
  • js/script-tests/arrowfunction-strict-mode.js: Added.

(foo):

12:14 PM Changeset in webkit [187013] by Michael Catanzaro
  • 2 edits in trunk/Source/WebKit2

Unreviewed, placate -Wmismatched-tags after r187011

It's basically bug #146990 again. Oops.

  • WebProcess/gtk/SeccompFiltersWebProcessGtk.h:
9:57 AM Changeset in webkit [187012] by saambarati1@gmail.com
  • 77 edits
    6 adds in trunk

[ES6] Add support for block scope const
​https://bugs.webkit.org/show_bug.cgi?id=31813

Reviewed by Filip Pizlo.

Source/JavaScriptCore:

'const' is now implemented in an ES6 spec compliant manner.
'const' variables are always block scoped and always live
either on the stack or in a JSLexicalEnvironment. 'const'
variables never live on the global object.

Inside the BytecodeGenerator, when assigning to a stack
'const' variable or a LocalClosureVar 'const' variable,
we will emit code that just throws a type error.
When assigning to a ClosureVar const variable, CodeBlock linking
will ensure that we perform a dynamic lookup of that variable so
that put_to_scope's slow path throws a type error.

The old 'const' implementation has been removed in this patch.

  • bytecode/BytecodeList.json:
  • bytecode/BytecodeUseDef.h:

(JSC::computeUsesForBytecodeOffset):
(JSC::computeDefsForBytecodeOffset):

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::dumpBytecode):
(JSC::CodeBlock::CodeBlock):

  • bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::BytecodeGenerator):
(JSC::BytecodeGenerator::pushLexicalScope):
(JSC::BytecodeGenerator::prepareLexicalScopeForNextForLoopIteration):
(JSC::BytecodeGenerator::variable):
(JSC::BytecodeGenerator::variableForLocalEntry):
(JSC::BytecodeGenerator::createVariable):
(JSC::BytecodeGenerator::emitResolveScope):
(JSC::BytecodeGenerator::emitInstanceOf):
(JSC::BytecodeGenerator::emitGetById):
(JSC::BytecodeGenerator::isArgumentNumber):
(JSC::BytecodeGenerator::emitReadOnlyExceptionIfNeeded):
(JSC::BytecodeGenerator::emitEnumeration):
(JSC::BytecodeGenerator::variablePerSymbolTable): Deleted.
(JSC::BytecodeGenerator::emitInitGlobalConst): Deleted.

  • bytecompiler/BytecodeGenerator.h:

(JSC::Variable::Variable):
(JSC::Variable::isReadOnly):
(JSC::Variable::isSpecial):
(JSC::Variable::isConst):
(JSC::BytecodeGenerator::thisRegister):
(JSC::BytecodeGenerator::emitTypeOf):
(JSC::BytecodeGenerator::emitIn):

  • bytecompiler/NodesCodegen.cpp:

(JSC::PostfixNode::emitResolve):
(JSC::PrefixNode::emitResolve):
(JSC::ReadModifyResolveNode::emitBytecode):
(JSC::AssignResolveNode::emitBytecode):
(JSC::CommaNode::emitBytecode):
(JSC::BindingNode::bindValue):
(JSC::ConstDeclNode::emitCodeSingle): Deleted.
(JSC::ConstDeclNode::emitBytecode): Deleted.
(JSC::ConstStatementNode::emitBytecode): Deleted.

  • dfg/DFGByteCodeParser.cpp:

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

  • dfg/DFGCapabilities.cpp:

(JSC::DFG::capabilityLevel):

  • jit/JIT.cpp:

(JSC::JIT::privateCompileMainPass):

  • jit/JIT.h:
  • jit/JITPropertyAccess.cpp:

(JSC::JIT::emit_op_put_to_arguments):
(JSC::JIT::emit_op_init_global_const): Deleted.

  • jit/JITPropertyAccess32_64.cpp:

(JSC::JIT::emit_op_put_to_arguments):
(JSC::JIT::emit_op_init_global_const): Deleted.

  • llint/LowLevelInterpreter.asm:
  • llint/LowLevelInterpreter32_64.asm:
  • llint/LowLevelInterpreter64.asm:
  • parser/ASTBuilder.h:

(JSC::ASTBuilder::createDeclarationStatement):
(JSC::ASTBuilder::createEmptyVarExpression):
(JSC::ASTBuilder::createDebugger):
(JSC::ASTBuilder::appendStatement):
(JSC::ASTBuilder::createVarStatement): Deleted.
(JSC::ASTBuilder::createLetStatement): Deleted.
(JSC::ASTBuilder::createConstStatement): Deleted.
(JSC::ASTBuilder::appendConstDecl): Deleted.

  • parser/NodeConstructors.h:

(JSC::CommaNode::CommaNode):
(JSC::SourceElements::SourceElements):
(JSC::SwitchNode::SwitchNode):
(JSC::BlockNode::BlockNode):
(JSC::ConstStatementNode::ConstStatementNode): Deleted.
(JSC::ConstDeclNode::ConstDeclNode): Deleted.

  • parser/Nodes.h:

(JSC::ConstDeclNode::hasInitializer): Deleted.
(JSC::ConstDeclNode::ident): Deleted.

  • parser/Parser.cpp:

(JSC::Parser<LexerType>::parseStatementListItem):
(JSC::Parser<LexerType>::parseVariableDeclaration):
(JSC::Parser<LexerType>::parseWhileStatement):
(JSC::Parser<LexerType>::parseVariableDeclarationList):
(JSC::Parser<LexerType>::createBindingPattern):
(JSC::Parser<LexerType>::parseDestructuringPattern):
(JSC::Parser<LexerType>::parseDefaultValueForDestructuringPattern):
(JSC::Parser<LexerType>::parseForStatement):
(JSC::Parser<LexerType>::parseTryStatement):
(JSC::Parser<LexerType>::parseFunctionInfo):
(JSC::Parser<LexerType>::parseFunctionDeclaration):
(JSC::Parser<LexerType>::parseClass):
(JSC::Parser<LexerType>::parseConstDeclaration): Deleted.
(JSC::Parser<LexerType>::parseConstDeclarationList): Deleted.

  • parser/Parser.h:

(JSC::isEvalNode):
(JSC::isEvalNode<EvalNode>):
(JSC::isArguments):
(JSC::isEval):
(JSC::isEvalOrArgumentsIdentifier):
(JSC::Scope::Scope):
(JSC::Scope::declareCallee):
(JSC::Scope::declareVariable):
(JSC::Scope::declareLexicalVariable):
(JSC::Scope::hasDeclaredVariable):
(JSC::Scope::allowsVarDeclarations):
(JSC::Scope::allowsLexicalDeclarations):
(JSC::Scope::declareParameter):
(JSC::Scope::declareBoundParameter):
(JSC::Parser::destructuringKindFromDeclarationType):
(JSC::Parser::assignmentContextFromDeclarationType):
(JSC::Parser::isEvalOrArguments):
(JSC::Parser::currentScope):
(JSC::Parser::popScope):
(JSC::Parser::declareVariable):
(JSC::Parser::hasDeclaredVariable):
(JSC::Parser::setStrictMode):
(JSC::Parser::strictMode):
(JSC::Parser::isValidStrictMode):
(JSC::Parser::declareParameter):
(JSC::Parser::declareBoundParameter):
(JSC::Parser::breakIsValid):

  • parser/SyntaxChecker.h:

(JSC::SyntaxChecker::createForInLoop):
(JSC::SyntaxChecker::createForOfLoop):
(JSC::SyntaxChecker::createEmptyStatement):
(JSC::SyntaxChecker::createDeclarationStatement):
(JSC::SyntaxChecker::createReturnStatement):
(JSC::SyntaxChecker::createBreakStatement):
(JSC::SyntaxChecker::createVarStatement): Deleted.
(JSC::SyntaxChecker::createLetStatement): Deleted.

  • parser/VariableEnvironment.h:

(JSC::VariableEnvironmentEntry::isCaptured):
(JSC::VariableEnvironmentEntry::isConst):
(JSC::VariableEnvironmentEntry::isVar):
(JSC::VariableEnvironmentEntry::isLet):
(JSC::VariableEnvironmentEntry::setIsCaptured):
(JSC::VariableEnvironmentEntry::setIsConst):
(JSC::VariableEnvironmentEntry::setIsVar):
(JSC::VariableEnvironmentEntry::setIsLet):
(JSC::VariableEnvironmentEntry::isConstant): Deleted.
(JSC::VariableEnvironmentEntry::setIsConstant): Deleted.

  • runtime/Executable.cpp:

(JSC::ProgramExecutable::initializeGlobalProperties):

  • runtime/JSGlobalObject.cpp:

(JSC::JSGlobalObject::defineOwnProperty):
(JSC::JSGlobalObject::addGlobalVar):
(JSC::JSGlobalObject::addFunction):
(JSC::lastInPrototypeChain):

  • runtime/JSGlobalObject.h:

(JSC::JSGlobalObject::finishCreation):
(JSC::JSGlobalObject::addVar):
(JSC::JSGlobalObject::addConst): Deleted.

  • runtime/JSLexicalEnvironment.cpp:

(JSC::JSLexicalEnvironment::symbolTablePut):

  • tests/stress/const-and-with-statement.js: Added.

(truth):
(assert):
(shouldThrowInvalidConstAssignment):
(.):

  • tests/stress/const-exception-handling.js: Added.

(truth):
(assert):
(.):

  • tests/stress/const-loop-semantics.js: Added.

(truth):
(assert):
(shouldThrowInvalidConstAssignment):
(.):

  • tests/stress/const-not-strict-mode.js: Added.

(truth):
(assert):
(shouldThrowTDZ):
(.):

  • tests/stress/const-semantics.js: Added.

(truth):
(assert):
(shouldThrowInvalidConstAssignment):
(.):

  • tests/stress/const-tdz.js: Added.

(truth):
(assert):
(shouldThrowTDZ):
(.):

Source/WebInspectorUI:

"const" variables do not live on the global object and are only
accessible within the "Program" they're defined in. Therefore,
the WebInspector global must be defined as "var" and not "const".

  • UserInterface/Base/WebInspector.js:

LayoutTests:

"const" variables do not live on the global object. They
are only available in the "Program" (read: JavaScript file or
script tag) that they're defined in. Tests have been updated
accordingly to switch the "const" variables assumed to be globals
into "var"s. "var" declared variables in the top level scope
of a program do live on the global object.

  • fast/canvas/webgl/compressed-tex-image.html:
  • fast/dom/event-handler-attributes.html:
  • fast/forms/listbox-visible-size.html:
  • js/arguments-expected.txt:
  • js/arrowfunction-syntax-errors-expected.txt:
  • js/const-expected.txt:
  • js/const-without-initializer-expected.txt:
  • js/constant-count-expected.txt:
  • js/dom/inc-const-valueOf-expected.txt:
  • js/dom/script-tests/inc-const-valueOf.js:

(testPreIncConstVarWithAssign):

  • js/function-toString-parentheses-expected.txt:
  • js/kde/const-expected.txt:
  • js/kde/resources/const.js:
  • js/parser-syntax-check-expected.txt:
  • js/script-tests/arguments.js:

(argumentsVarUndefined):
(argumentsConst):
(argumentCalleeInException):
(argumentsConstUndefined): Deleted.

  • js/script-tests/class-syntax-declaration.js:

(A):

  • js/script-tests/class-syntax-expression.js:
  • js/script-tests/const-without-initializer.js:
  • js/script-tests/const.js:

(shouldThrowInvalidConstAssignment):
(assert):
(f):
(tryCatch1):
(tryCatch2):
(with1):
(with2):
(.):

  • js/script-tests/constant-count.js:

(f):

  • js/script-tests/function-dot-arguments.js:

(assignConstInitTest2.g):
(assignConstInitTest2):

  • js/script-tests/function-toString-parentheses.js:
  • js/script-tests/parser-syntax-check.js:
  • sputnik/Conformance/07_Lexical_Conventions/7.5_Tokens/7.5.3_Future_Reserved_Words/S7.5.3_A1.6-expected.txt:
  • sputnik/Conformance/07_Lexical_Conventions/7.8_Literals/7.8.5_Regular_Expression_Literals/S7.8.5_A3.1_T7-expected.txt:
  • sputnik/Conformance/07_Lexical_Conventions/7.8_Literals/7.8.5_Regular_Expression_Literals/S7.8.5_A3.1_T8-expected.txt:
  • sputnik/Conformance/07_Lexical_Conventions/7.8_Literals/7.8.5_Regular_Expression_Literals/S7.8.5_A3.1_T9-expected.txt:
  • sputnik/Conformance/08_Types/8.4_The_String_Type/S8.4_A13_T3-expected.txt:
  • sputnik/Conformance/08_Types/8.4_The_String_Type/S8.4_A14_T3-expected.txt:
  • sputnik/Conformance/12_Statement/12.2_Variable_Statement/S12.2_A8_T1-expected.txt:
  • sputnik/Conformance/12_Statement/12.2_Variable_Statement/S12.2_A8_T2-expected.txt:
  • sputnik/Conformance/12_Statement/12.2_Variable_Statement/S12.2_A8_T3-expected.txt:
  • sputnik/Conformance/12_Statement/12.2_Variable_Statement/S12.2_A8_T4-expected.txt:
  • sputnik/Conformance/12_Statement/12.2_Variable_Statement/S12.2_A8_T6-expected.txt:
  • sputnik/Conformance/12_Statement/12.2_Variable_Statement/S12.2_A8_T7-expected.txt:
  • sputnik/Conformance/12_Statement/12.2_Variable_Statement/S12.2_A8_T8-expected.txt:
  • transforms/3d/hit-testing/composited-hit-test.html:
  • transforms/3d/hit-testing/coplanar-with-camera.html:
  • transforms/3d/hit-testing/hover-rotated-negative-z.html:
  • transforms/3d/hit-testing/hover-rotated-with-children-negative-z.html:
  • transforms/3d/hit-testing/negative-zoffset-hit-test.html:
  • transforms/3d/hit-testing/overlapping-layers-hit-test.html:
  • transforms/3d/hit-testing/perspective-clipped.html:
  • transforms/3d/hit-testing/rotated-hit-test-with-child.html:
  • transforms/3d/hit-testing/rotated-hit-test.html:
  • transforms/3d/hit-testing/rotated-hit-test2.html:
  • transitions/resources/transition-test-helpers.js:

(roundNumber):

9:32 AM Changeset in webkit [187011] by Michael Catanzaro
  • 7 edits
    2 adds in trunk

[GTK] Add seccomp filters support
​https://bugs.webkit.org/show_bug.cgi?id=110014

Reviewed by Žan Doberšek.

.:

Find needed compiler and linker flags for libseccomp.

  • Source/cmake/OptionsGTK.cmake:

Source/WebKit2:

Allow building with ENABLE_SECCOMP_FILTERS=ON. Based on work by Thiago Marcos P. Santos.

  • PlatformGTK.cmake: Support ENABLE_SECCOMP_FILTERS build option.
  • WebProcess/gtk/SeccompFiltersWebProcessGtk.cpp: Added.

(WebKit::SeccompFiltersWebProcessGtk::SeccompFiltersWebProcessGtk):
(WebKit::SeccompFiltersWebProcessGtk::platformInitialize):

  • WebProcess/gtk/SeccompFiltersWebProcessGtk.h: Added.
  • WebProcess/soup/WebProcessSoup.cpp:

(WebKit::WebProcess::platformInitializeWebProcess): Initialize default
GTK+ web process seccomp filters.

Tools:

Add libseccomp to jhbuild modulesets.

  • gtk/jhbuild.modules:
6:16 AM Changeset in webkit [187010] by ddkilzer@apple.com
  • 5 edits in trunk/Source/WebKit/win

REGRESSION (r187002): Broke the Windows build: Reduce PassRefPtr in WebKit2 - 3
<​https://bugs.webkit.org/show_bug.cgi?id=146995>

Attempt to fix the Windows build.

  • WebCoreSupport/WebFrameLoaderClient.cpp:

(WebFrameLoaderClient::createDocumentLoader):
(WebFrameLoaderClient::createFrame):
(WebFrameLoaderClient::createPlugin):

  • WebCoreSupport/WebFrameLoaderClient.h:
  • WebDocumentLoader.cpp:

(WebDocumentLoader::create):

  • WebDocumentLoader.h:
5:35 AM Changeset in webkit [187009] by ddkilzer@apple.com
  • 2 edits in trunk/Source/WebCore

REGRESSION (r187008): CGPathCreateMutableCopyByTransformingPath() is not available on Windows
<​https://bugs.webkit.org/show_bug.cgi?id=147077>

  • platform/graphics/cg/PathCG.cpp:

(WebCore::Path::transform): Restore less efficient code path
for Windows.

Note: See TracTimeline for information about the timeline view.