Timeline

Apr 14, 2014:

11:57 PM Changeset in webkit [167302] by psolanki@apple.com

• 2 edits in trunk/Source/WebCore

Unreviewed. Attempt to fix Windows build after r167277.

• page/FrameView.cpp:

(WebCore::FrameView::willPaintContents):

11:51 PM Changeset in webkit [167301] by ap@apple.com

• 2 edits in trunk/Tools

Remove Gtk WebKit1 tester.

Gtk WebKit1 port no longer exists, and dashboard was trying to load its tester
results over and over as fast as it could.

• BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:

(WebKitBuildbot):

11:40 PM Changeset in webkit [167300] by ap@apple.com

• 2 edits in trunk/Tools

Wrong link for webkitpy tests results in dashboard popover
​https://bugs.webkit.org/show_bug.cgi?id=131664

Reviewed by Timothy Hatcher.

• BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotTesterQueueView.js:

(BuildbotTesterQueueView.prototype._presentPopoverForMultipleFailureKinds):
Use a correct link.

11:24 PM Changeset in webkit [167299] by commit-queue@webkit.org

• 4 edits
  1 delete in trunk

Unreviewed, rolling out r167261.
​https://bugs.webkit.org/show_bug.cgi?id=131667

broke many navigation tests (Requested by ap on #webkit).

Reverted changeset:

"Web Replay: memoize fallback time values for
document.lastModified"
​https://bugs.webkit.org/show_bug.cgi?id=131318
​http://trac.webkit.org/changeset/167261

11:13 PM Changeset in webkit [167298] by Simon Fraser

• 10 edits in trunk/Source

[iOS WK2] Pages often blank on first load if page loaded by typing the URL
​https://bugs.webkit.org/show_bug.cgi?id=131665

Reviewed by Tim Horton.

The document overlay-related code in RemoteLayerTreeDrawingArea::setRootCompositingLayer()
was triggering a compositing layer flush when called with a null rootLayer, which happens
for pages going into the page cache. This would trigger a layer flush that would clobber
the root layer for the visible page, resulting in missing content.

Also, rebuildCompositingLayerTree() is called recursively and the m_documentOverlayRootLayer
was being added to (and then removed from) every single compositing layers.

Fix both these by changing to a pull model, where RenderLayerCompositor requests
the overlay layer via ChromeClient, and gets it at the end of every flush,
adding to the children of the root layer.

Source/WebCore:

• WebCore.exp.in:
• page/ChromeClient.h:

(WebCore::ChromeClient::documentOverlayLayerForFrame):

• rendering/RenderLayerCompositor.cpp:

(WebCore::RenderLayerCompositor::RenderLayerCompositor):
(WebCore::RenderLayerCompositor::flushPendingLayerChanges): Put visibleRect
into a variable for ease of debugging.
(WebCore::RenderLayerCompositor::updateCompositingLayers): Asser
that we're not in the page cache (this would have caught the bug).
(WebCore::RenderLayerCompositor::appendOverlayLayers):
(WebCore::RenderLayerCompositor::rebuildCompositingLayerTree):
(WebCore::RenderLayerCompositor::setDocumentOverlayRootLayer): Deleted.

• rendering/RenderLayerCompositor.h:

Source/WebKit2:

• WebProcess/WebCoreSupport/WebChromeClient.cpp:

(WebKit::WebChromeClient::documentOverlayLayerForFrame):

• WebProcess/WebCoreSupport/WebChromeClient.h:
• WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.mm:

(WebKit::RemoteLayerTreeDrawingArea::setRootCompositingLayer):

11:13 PM Changeset in webkit [167297] by commit-queue@webkit.org

• 5 edits
  3 deletes in trunk

Unreviewed, rolling out r167272.
​https://bugs.webkit.org/show_bug.cgi?id=131666

Broke multiple tests (Requested by ap on #webkit).

Reverted changeset:

"Function.bind itself is too slow"
​https://bugs.webkit.org/show_bug.cgi?id=131636
​http://trac.webkit.org/changeset/167272

10:47 PM Changeset in webkit [167296] by ap@apple.com

• 2 edits in trunk/Tools

Update webkitpy regression test result that started to fail after ​http://trac.webkit.org/r167243

I'm not sure whether the changes are desirable, but am not suspicious enough to roll out.

• Scripts/webkitpy/common/checkout/checkout_unittest.py:

(CommitMessageForThisCommitTest):

10:23 PM Changeset in webkit [167295] by jhoneycutt@apple.com

• 7 edits
  2 adds in trunk

Assertion failure under FEImage::determineAbsolutePaintRect()

<​https://bugs.webkit.org/show_bug.cgi?id=131660>
<rdar://problem/15669294>

Source/WebCore:
This patch merges Chromium r149536 (see
<​https://chromiumcodereview.appspot.com/14701012>), which moves
m_absoluteTransform out of SVGFilter and into the base Filter class, so
that it isn't necessary to cast a Filter to SVGFilter to get the
absolute transform.

Reviewed by Geoffrey Garen.

Test: svg/filters/feImage-filter-assertion.html

• platform/graphics/filters/Filter.h:

(WebCore::Filter::Filter):
Changed to take the absolute transform.
(WebCore::Filter::absoluteTransform):
Moved from SVGFilter.
(WebCore::Filter::mapAbsolutePointToLocalPoint):
Ditto.

• rendering/FilterEffectRenderer.cpp:

(WebCore::FilterEffectRenderer::FilterEffectRenderer):
Pass a default AffineTransform() to the Filter base class.

• svg/graphics/filters/SVGFEImage.cpp:

(WebCore::FEImage::determineAbsolutePaintRect):
Use the Filter without casting it to SVGFilter.
(WebCore::FEImage::platformApplySoftware):
Ditto.

• svg/graphics/filters/SVGFilter.cpp:

(WebCore::SVGFilter::SVGFilter):
Pass the transform to the base class, and remove initialization of a
removed member var.

• svg/graphics/filters/SVGFilter.h:

Member var moved to Filter.h.

LayoutTests:
Reviewed by Geoffrey Garen.

• svg/filters/feImage-filter-assertion-expected.txt: Added.
• svg/filters/feImage-filter-assertion.html: Added.

10:01 PM April 2014 Meeting edited by simon.pena@samsung.com

(diff)

9:31 PM Changeset in webkit [167294] by commit-queue@webkit.org

• 29 edits in trunk/Source/WebInspectorUI

Update to CodeMirror 4.0.
​https://bugs.webkit.org/show_bug.cgi?id=130019

The CodeMirror 4.0 library adds multiple selections and the ability to
undo/redo selections. Selections are made by holding the Command (Cmd) key
and clicking into an editor, or by holding Option (Alt) and making a block
selection followed by entering a character or moving the cursor.

Patch by Jono Wells <​jonowells@apple.com> on 2014-04-14
Reviewed by Timothy Hatcher.

• Scripts/update-codemirror-resources.rb:
• Scripts/update-pretty-printer.rb:

Update scripts to reflect updated locations for CodeMirror files.
No longer copies LESS mode file as this has been integrated into CSS mode
in CodeMirror 4.0.

• Tools/PrettyPrinting/codemirror.css:
• Tools/PrettyPrinting/codemirror.js:
• Tools/PrettyPrinting/css.js:
• Tools/PrettyPrinting/javascript.js:
• UserInterface/External/CodeMirror/clojure.js:
• UserInterface/External/CodeMirror/closebrackets.js:
• UserInterface/External/CodeMirror/codemirror.css:
• UserInterface/External/CodeMirror/codemirror.js:
• UserInterface/External/CodeMirror/coffeescript.js:
• UserInterface/External/CodeMirror/comment.js:
• UserInterface/External/CodeMirror/css.js:
• UserInterface/External/CodeMirror/htmlmixed.js:
• UserInterface/External/CodeMirror/javascript.js:
• UserInterface/External/CodeMirror/livescript.js:
• UserInterface/External/CodeMirror/matchbrackets.js:
• UserInterface/External/CodeMirror/overlay.js:
• UserInterface/External/CodeMirror/placeholder.js:
• UserInterface/External/CodeMirror/runmode.js:
• UserInterface/External/CodeMirror/sass.js:
• UserInterface/External/CodeMirror/searchcursor.js:
• UserInterface/External/CodeMirror/sql.js:
• UserInterface/External/CodeMirror/xml.js:

Update to CodeMirror 4.0.

• UserInterface/Main.html: Remove less.js which is now part of css.js.

• UserInterface/Views/CSSStyleDeclarationTextEditor.css:
• UserInterface/Views/CSSStyleDeclarationTextEditor.js:

(WebInspector.CSSStyleDeclarationTextEditor.prototype.):
(WebInspector.CSSStyleDeclarationTextEditor.prototype._createColorSwatches):
(WebInspector.CSSStyleDeclarationTextEditor.prototype._updateJumpToSymbolTrackingMode):

• UserInterface/Views/SyntaxHighlightingDefaultTheme.css:

Update styles to match CodeMirror changes. Update CSSStyleDeclarationTextEditor.js
to match CodeMirror API updates (doc.removeLine() has been removed, so
replaceRange() is used instead). Also the tokenTrackingController is now enabled in the
CSSStyleDeclarationTextEditor and SourceCodeTextEditor when the Option (Alt) key
is pressed instead of the Command (Cmd) key so as not to conflict with multiple
cursor placement in the CodeMirror update.

8:53 PM Changeset in webkit [167293] by ggaren@apple.com

• 2 edits in trunk/Source/JavaScriptCore

ASSERT when firing low memory warning
​https://bugs.webkit.org/show_bug.cgi?id=131659

Reviewed by Mark Hahnenberg.

• heap/Heap.cpp:

(JSC::Heap::deleteAllCompiledCode): Allow deleteAllCompiledCode to be
called when no GC is happening because that is what we do when a low
memory warning fires, and it is harmless.

8:52 PM Changeset in webkit [167292] by ggaren@apple.com

• 16 edits in trunk

MallocBench should scavenge explicitly instead of waiting
​https://bugs.webkit.org/show_bug.cgi?id=131661

Reviewed by Andreas Kling.

PerformanceTests:

• MallocBench/MallocBench.xcodeproj/project.pbxproj: Don't build mbmalloc

by default because it will overwrite any other mbmalloc you're working
with in the WebKitBuild directory.

• MallocBench/MallocBench/Benchmark.cpp:

(Benchmark::run): Scavenge explicitly instead of waiting. This is faster,
and it's the only way to get FastMalloc to scavenge. (That's a bug in
FastMalloc, but we don't want it to interfere with broader testing.)

• MallocBench/MallocBench/mbmalloc.cpp:
• MallocBench/MallocBench/mbmalloc.h: Added a scavenge implementation

for system malloc.

Source/bmalloc:
Added explicit scavenge support to bmalloc. This isn't a memory win,
since bmalloc's per-thread cache is so small. But it makes testing
simpler.

• bmalloc/Allocator.cpp:

(bmalloc::Allocator::~Allocator):
(bmalloc::Allocator::scavenge):

• bmalloc/Allocator.h:
• bmalloc/Cache.cpp:

(bmalloc::Cache::operator new):
(bmalloc::Cache::operator delete):
(bmalloc::Cache::Cache):
(bmalloc::Cache::scavenge):

• bmalloc/Cache.h:
• bmalloc/Deallocator.cpp:

(bmalloc::Deallocator::~Deallocator):
(bmalloc::Deallocator::scavenge):

• bmalloc/Deallocator.h: Factored existing scavenging code into helper

functions, for reuse.

• bmalloc/Heap.cpp:

(bmalloc::sleep):
(bmalloc::Heap::concurrentScavenge):
(bmalloc::Heap::scavenge):
(bmalloc::Heap::scavengeSmallPages):
(bmalloc::Heap::scavengeMediumPages):
(bmalloc::Heap::scavengeLargeRanges):

• bmalloc/Heap.h: Made scavenge sleep duration a parameter. Forced

scavenging -- in response to a benchmark or a low memory warning --
wants to complete as soon as possible, so its sleep duration is 0.

• bmalloc/bmalloc.h:

(bmalloc::api::scavenge):

• bmalloc/mbmalloc.cpp: Exported the scavenge API for MallocBench's use.

8:06 PM Changeset in webkit [167291] by Darin Adler

• 2 edits in trunk/Source/WebCore

REGRESSION (r158617): Find on Page can get stuck in a loop when the search string occurs in an <input> in a <fieldset>
​https://bugs.webkit.org/show_bug.cgi?id=126322

Reviewed by Ryosuke Niwa.

One additional tweak to the fix for the bug above.
Fixes crash in editing/editability/ignored-content.html test.

• html/HTMLObjectElement.cpp:

(WebCore::HTMLObjectElement::canContainRangeEndPoint): Call through to
HTMLElement::canContainRangeEndPoint, bypassing HTMLPlugInElement override
that always returns false. Without this change, this function was always
returning false.

7:54 PM Changeset in webkit [167290] by Simon Fraser

• 4 edits in trunk/Source/WebCore

Crash in TileController::tileRevalidationTimerFired
​https://bugs.webkit.org/show_bug.cgi?id=131656
<rdar://problem/16583166>

Reviewed by Sam Weinig.

It's possible for the TileController revalidation timer to fire after
the GraphicsLayer has been destroyed, so the PlatformCALayer no longer
has an owningGraphicsLayer.

Bail from the timer callback if owningGraphicsLayer() is null.

Also some drive-by 0 -> nullptr changes.

• platform/graphics/ca/GraphicsLayerCA.cpp:

(WebCore::GraphicsLayerCA::willBeDestroyed):

• platform/graphics/ca/PlatformCALayer.cpp:

(WebCore::PlatformCALayer::~PlatformCALayer):

• platform/graphics/ca/mac/TileController.mm:

(WebCore::TileController::tileRevalidationTimerFired):

7:44 PM Changeset in webkit [167289] by ggaren@apple.com

• 3 edits
  1 add in trunk/Source/bmalloc

Use 4kB pages on Mac
​https://bugs.webkit.org/show_bug.cgi?id=131658

Reviewed by Sam Weinig.

This reduces memory use a lot on Membuster:

base patch Δ

Execution Time:

reddit_memory_warning 18ms 17ms ^{1.06x faster
flickr_memory_warning 34ms 36ms ! 1.06x slower
theverge_memory_warning 39ms 41ms ! 1.05x slower}

<geometric mean> 29ms 29ms ! 1.02x slower
<arithmetic mean> 30ms 31ms ! 1.03x slower
<harmonic mean> 27ms 27ms ^{1.0x faster}

Peak Memory:

reddit_memory_warning 16,412kB 16,436kB ! 1.0x bigger
flickr_memory_warning 30,120kB 30,184kB ! 1.0x bigger
theverge_memory_warning 33,408kB 33,420kB ! 1.0x bigger

<geometric mean> 25,466kB 25,499kB ! 1.0x bigger
<arithmetic mean> 26,647kB 26,680kB ! 1.0x bigger
<harmonic mean> 24,181kB 24,214kB ! 1.0x bigger

Memory at End:

reddit_memory_warning 2,404kB 1,920kB ^{1.25x smaller
flickr_memory_warning 3,764kB 3,072kB} 1.23x smaller
theverge_memory_warning 3,648kB 3,132kB ^{1.16x smaller}

<geometric mean> 3,208kB 2,644kB ^{1.21x smaller
<arithmetic mean> 3,272kB 2,708kB} 1.21x smaller
<harmonic mean> 3,139kB 2,574kB ^{1.22x smaller}

• bmalloc.xcodeproj/project.pbxproj:
• bmalloc/BPlatform.h: Added.
• bmalloc/VMAllocate.h: Only use 16kB pages on iOS because the page size

is 4kB on Mac.

7:32 PM April 2014 Meeting edited by Simon Fraser

(diff)

7:20 PM Changeset in webkit [167288] by mhahnenberg@apple.com

• 3 edits in trunk/Source/JavaScriptCore

emit_op_put_by_id should not emit a write barrier that filters on value
​https://bugs.webkit.org/show_bug.cgi?id=131654

Reviewed by Filip Pizlo.

The 32-bit implementation does this, and it can cause crashes if we later repatch the
code to allocate and store new Butterflies.

• jit/JITPropertyAccess.cpp:

(JSC::JIT::emitWriteBarrier): We also weren't verifying that the base was a cell on
32-bit if we were passed ShouldFilterBase. I also took the liberty of sinking the tag
load down into the if statement so that we don't do it if we're not filtering on the value.

• jit/JITPropertyAccess32_64.cpp:

(JSC::JIT::emit_op_put_by_id):

7:03 PM April 2014 Meeting edited by sam@webkit.org

(diff)

7:02 PM April 2014 Meeting edited by sam@webkit.org

(diff)

7:00 PM April 2014 Meeting edited by sam@webkit.org

(diff)

5:55 PM Changeset in webkit [167287] by timothy_horton@apple.com

• 2 edits in trunk/Source/WebKit2

Fix the 32-bit build.

• UIProcess/mac/ViewGestureControllerMac.mm:

5:47 PM Changeset in webkit [167286] by Simon Fraser

• 2 edits in trunk/Source/WebKit2

REGRESSION (WebKit2): Programmatic scrolls in overflow-scrolling:touch don't work
​https://bugs.webkit.org/show_bug.cgi?id=131649

Reviewed by Tim Horton.

Update the UIScrollView's contentOffset when we're told that the scroll position
changed.

Also refactor slightly to have a single exception-protected block.

• UIProcess/Scrolling/ios/ScrollingTreeOverflowScrollingNodeIOS.mm:

(WebKit::ScrollingTreeOverflowScrollingNodeIOS::updateAfterChildren):

5:47 PM Changeset in webkit [167285] by timothy_horton@apple.com

• 2 edits in trunk/Source/WebKit2

Fix the 32-bit build.

• UIProcess/mac/ViewGestureControllerMac.mm:

5:41 PM Changeset in webkit [167284] by timothy_horton@apple.com

• 3 edits in trunk/Source/WebKit2

Sometimes-crash under handleSwipeGesture after closing a window or quitting
​https://bugs.webkit.org/show_bug.cgi?id=131648
<rdar://problem/15966106>

Reviewed by Simon Fraser.

• UIProcess/mac/ViewGestureController.h:
• UIProcess/mac/ViewGestureControllerMac.mm:

(WebKit::ViewGestureController::~ViewGestureController):
(WebKit::ViewGestureController::trackSwipeGesture):
Keep a Objective C object with a single boolean property, isCancelled, on
the ViewGestureController, and also retained by the swipe-tracking block.
When the ViewGestureController is destroyed, we set isCancelled to YES,
and the next time the block is invoked, we will cancel the swipe without
touching the destroyed ViewGestureController.

(WebKit::ViewGestureController::handleSwipeGesture):
Don't try to handle a swipe gesture if the drawing area is missing.

(WebKit::ViewGestureController::endSwipeGesture):
Clear the swipe cancellation tracker when the gesture completes.

5:27 PM Changeset in webkit [167283] by Bem Jones-Bey

• 5 edits in trunk/Source/WebCore

[CSS Shapes] Remove some leftover shape-inside code
​https://bugs.webkit.org/show_bug.cgi?id=131641

Reviewed by Dean Jackson.

I discovered that some code had been leftover from the shape-inside
removal. This removes that leftover code.

No new tests, no behavior change.

• platform/text/BidiResolver.h:
• rendering/BidiRun.cpp:

(WebCore::BidiRun::BidiRun):

• rendering/RenderBlockFlow.h:
• rendering/RenderBlockLineLayout.cpp:

(WebCore::RenderBlockFlow::createLineBoxes):
(WebCore::RenderBlockFlow::constructLine):
(WebCore::computeExpansionForJustifiedText):
(WebCore::RenderBlockFlow::computeInlineDirectionPositionsForSegment):

5:14 PM Changeset in webkit [167282] by jer.noble@apple.com

• 2 edits in trunk/Source/WebCore

[MSE][Mac] video.currentTime is sometimes negative.
​https://bugs.webkit.org/show_bug.cgi?id=131644

Reviewed by Eric Carlson.

AVSampleBufferRenderSynchronizer will occasionally return slightly negative values
when beginning playback. Clamp the return value to 0.

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

(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::currentTimeDouble):

5:07 PM Changeset in webkit [167281] by Bem Jones-Bey

• 2 edits in trunk/Source/WebCore

[CSSExclusions] Remove FIXME referencing closed bug
​https://bugs.webkit.org/show_bug.cgi?id=131645

Reviewed by Dean Jackson.

This is silly, but the comment is really confusing as it's entirely
wrong now.

No new tests, no behavior change.

• rendering/style/RenderStyle.cpp:

(WebCore::RenderStyle::changeRequiresRepaint):

5:06 PM April 2014 Meeting edited by akling@apple.com

(diff)

4:47 PM Changeset in webkit [167280] by enrica@apple.com

• 3 edits in trunk/Source/WebKit2

REGRESSION(r166027) Menu is shown and immediately hidden after doubletap gesture.
​https://bugs.webkit.org/show_bug.cgi?id=131646
<rdar://problem/16614374>

Reviewed by Benjamin Poulain.

This is a problem with every gesture that changes the selection
We don't need to delay updating the selection if the selection is
being changed with a gesture.

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

(-[WKContentView _didUpdateBlockSelectionWithTouch:withFlags:growThreshold:shrinkThreshold:]):
(-[WKContentView changeSelectionWithGestureAt:withGesture:withState:]):
(-[WKContentView changeSelectionWithTouchAt:withSelectionTouch:baseIsStart:]):
(-[WKContentView changeSelectionWithTouchesFrom:to:withGesture:withState:]):
(-[WKContentView changeBlockSelectionWithTouchAt:withSelectionTouch:forHandle:]):
(-[WKContentView _selectionChanged]):

4:38 PM April 2014 Meeting edited by abucur@adobe.com

(diff)

4:22 PM Changeset in webkit [167279] by akling@apple.com

• 2 edits in trunk/Source/WebCore

Build fixage.

• page/FrameView.cpp:

4:14 PM April 2014 Meeting edited by rniwa@webkit.org

(diff)

4:08 PM Changeset in webkit [167278] by jhoneycutt@apple.com

• 3 edits in trunk/Source/WebCore

Assertion failure !node

node->isElementNode() in

WebCore::RenderBlock::inlineElementContinuation

​https://bugs.webkit.org/show_bug.cgi?id=108829
<rdar://problem/13666405>

I can't reproduce this assertion failure, but there seems to be an
invalid assumption in RenderBlock::inlineElementContinuation() that
anything with the "isInline()" bit set is a RenderInline.

No new test because the test case in the bug does not repro for me.

Reviewed by Brent Fulgham.

• rendering/RenderBlock.cpp:

(WebCore::RenderBlock::inlineElementContinuation):
Dave Hyatt says that this function should only return RenderInline
objects (not non-RenderInline inline objects), so update the checks
from isInline() to isRenderInline() before casting with
toRenderInline().

• rendering/RenderInline.cpp:

(WebCore::RenderInline::inlineElementContinuation):
Ditto.

4:08 PM April 2014 Meeting edited by rniwa@webkit.org

(diff)

4:03 PM Changeset in webkit [167277] by akling@apple.com

• 14 edits in trunk/Source

Do more things under memory pressure on non-iOS platforms.
<​https://webkit.org/b/131625>

Reviewed by Antti Koivisto.

Rename hasReceivedMemoryPressure() to isUnderMemoryPressure() and
make it use std::atomic<bool> instead of OSAtomic primitives.

Unmask most of the PLATFORM(IOS) blocks so all platforms can take
advantage of optimizations done while under pressure. Note that
isUnderMemoryPressure() will still always return false on platforms
other than iOS/WK1, but this will change soon.

• history/PageCache.cpp:

(WebCore::PageCache::canCache):

• loader/FrameLoader.cpp:

(WebCore::FrameLoader::commitProvisionalLoad):

• page/FrameView.cpp:

(WebCore::FrameView::willPaintContents):
(WebCore::FrameView::didPaintContents):

• platform/MemoryPressureHandler.cpp:

(WebCore::MemoryPressureHandler::MemoryPressureHandler):

• platform/MemoryPressureHandler.h:

(WebCore::MemoryPressureHandler::isUnderMemoryPressure):

• platform/cocoa/MemoryPressureHandlerCocoa.mm:

(WebCore::MemoryPressureHandler::setReceivedMemoryPressure):
(WebCore::MemoryPressureHandler::clearMemoryPressure):
(WebCore::MemoryPressureHandler::respondToMemoryPressureIfNeeded):
(WebCore::MemoryPressureHandler::hasReceivedMemoryPressure): Deleted.

• platform/graphics/FontCache.cpp:

(WebCore::FontCache::purgeInactiveFontDataIfNeeded):

• platform/ios/LegacyTileCache.mm:

(WebCore::LegacyTileCache::createTilesInActiveGrid):

• platform/ios/LegacyTileGrid.mm:

(WebCore::LegacyTileGrid::shouldUseMinimalTileCoverage):

• platform/ios/LegacyTileLayerPool.mm:

(WebCore::LegacyTileLayerPool::addLayer):

• platform/ios/TileControllerMemoryHandlerIOS.cpp:

(WebCore::TileControllerMemoryHandler::tileControllerGainedUnparentedTiles):

3:52 PM April 2014 Meeting edited by Simon Fraser

(diff)

3:46 PM Changeset in webkit [167276] by mitz@apple.com

• 7 edits in trunk/Source/WebKit2

Let the bundle form client know whether a change in a text field was initiated by user typing
​https://bugs.webkit.org/show_bug.cgi?id=131643

Reviewed by Sam Weinig.

• WebProcess/InjectedBundle/API/Cocoa/WKWebProcessPlugInFormDelegatePrivate.h: Added

initiatedByUserTyping parameter to the delegate method.

• WebProcess/InjectedBundle/API/mac/WKWebProcessPlugInBrowserContextController.mm:

(-[WKWebProcessPlugInBrowserContextController _setFormDelegate:]): Updated overload of
textDidChangeInTextField to take the initiatedByUserTyping parameter and pass it along to
the delegate.

• WebProcess/InjectedBundle/APIInjectedBundleFormClient.h:

(API::InjectedBundle::FormClient::textDidChangeInTextField): Added initiatedByUserTyping
parameter.

• WebProcess/InjectedBundle/InjectedBundlePageFormClient.cpp:

(WebKit::InjectedBundlePageFormClient::textDidChangeInTextField): Maintained the existing
behavior of not calling the C SPI client if the change was not initiated by user typing.

• WebProcess/InjectedBundle/InjectedBundlePageFormClient.h: Updated for added parameter.

• WebProcess/WebCoreSupport/WebEditorClient.cpp:

(WebKit::WebEditorClient::textDidChangeInTextField): Changed to call the bundle client
function unconditionally, but pass along whether the change was initiated by user typing.

3:44 PM Changeset in webkit [167275] by mitz@apple.com

• 3 edits in trunk/Source/WebKit2

[Cocoa] Add a WKWebProcessPlugInFormDelegate method corresponding to willSendSubmitEvent
​https://bugs.webkit.org/show_bug.cgi?id=131639

Reviewed by Tim Horton.

• WebProcess/InjectedBundle/API/Cocoa/WKWebProcessPlugInFormDelegatePrivate.h: Declared

new delegate method.

• WebProcess/InjectedBundle/API/mac/WKWebProcessPlugInBrowserContextController.mm:

(-[WKWebProcessPlugInBrowserContextController _setFormDelegate:]): Added an override
of API::InjectedBundle::FormClient::willSendSubmitEvent which calls the new delegate
method.

3:39 PM Changeset in webkit [167274] by benjamin@webkit.org

• 3 edits in trunk/Source/WebKit2

[iOS][WK2] Change the SPI used when starting the rotation animation
​https://bugs.webkit.org/show_bug.cgi?id=131638

Patch by Benjamin Poulain <​bpoulain@apple.com> on 2014-04-14
Reviewed by Tim Horton.

Having an update block where all the properties are changed is more convenient for Safari.

• UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _beginAnimatedResizeWithUpdates:]):
(-[WKWebView _beginAnimatedResizeToSize:obscuredInsets:minimumLayoutSizeOverride:]): Deleted.

• UIProcess/API/Cocoa/WKWebViewPrivate.h:

3:37 PM Changeset in webkit [167273] by Simon Fraser

• 5 edits in trunk/Source/WebKit2

[iOS WK2] Hash table assertion closing a tab
​https://bugs.webkit.org/show_bug.cgi?id=131640

Reviewed by Tim Horton.

Have RemoteLayerTreeContext keep track of all PlatformCALayerRemotes,
and clear their context pointer when it is being destroyed to avoid
calling into a deleted object later.

• WebProcess/WebPage/mac/PlatformCALayerRemote.cpp:

(WebKit::PlatformCALayerRemote::~PlatformCALayerRemote):
(WebKit::PlatformCALayerRemote::addAnimationForKey):

• WebProcess/WebPage/mac/PlatformCALayerRemote.h:

(WebKit::PlatformCALayerRemote::clearContext):

• WebProcess/WebPage/mac/RemoteLayerTreeContext.h:
• WebProcess/WebPage/mac/RemoteLayerTreeContext.mm:

(WebKit::RemoteLayerTreeContext::~RemoteLayerTreeContext):
(WebKit::RemoteLayerTreeContext::layerWasCreated):
(WebKit::RemoteLayerTreeContext::layerWillBeDestroyed):

3:05 PM Changeset in webkit [167272] by oliver@apple.com

• 5 edits
  3 adds in trunk

Function.bind itself is too slow
​https://bugs.webkit.org/show_bug.cgi?id=131636

Reviewed by Geoffrey Garen.

Source/JavaScriptCore:
Rather than forcing creation of an activation, we now store
bound function properties directly on the returned closure.
This is necessary to deal with code that creates many function
bindings, but does not call them very often.

This is a 60% speed up in the included js/regress test.

• builtins/BuiltinExecutables.cpp:

(JSC::BuiltinExecutables::createBuiltinExecutable):

• builtins/Function.prototype.js:

(bind.bindingFunction):
(bind.else.switch.case.1.bindingFunction.bindingFunction.bindingFunction.boundOversizedCallThunk):
(bind.else.switch.case.1.bindingFunction):
(bind.else.switch.case.2.bindingFunction.bindingFunction.bindingFunction.boundOversizedCallThunk):
(bind.else.switch.case.2.bindingFunction):
(bind.else.switch.case.3.bindingFunction.bindingFunction.bindingFunction.boundOversizedCallThunk):
(bind.else.switch.case.3.bindingFunction):
(bind.else.switch.bindingFunction):
(bind):
(bind.else.switch.case.1.bindingFunction.oversizedCall): Deleted.
(bind.else.switch.case.2.bindingFunction.oversizedCall): Deleted.
(bind.else.switch.case.3.bindingFunction.oversizedCall): Deleted.

• runtime/CommonIdentifiers.h:

LayoutTests:
New test, and fix bogus log in old one

• js/regress/function-bind-create-expected.html: Added.
• js/regress/function-bind-create.html: Added.
• js/regress/script-tests/function-bind-create.js: Added.

(test):

• js/regress/script-tests/function-bind.js:

2:51 PM Changeset in webkit [167271] by timothy_horton@apple.com

• 3 edits in trunk/Source/WebKit2

Allow dynamic changes of WKThumbnailView snapshot scale
​https://bugs.webkit.org/show_bug.cgi?id=131628
<rdar://problem/16584156>

Reviewed by Beth Dakin.

• UIProcess/API/Cocoa/_WKThumbnailView.mm:

(-[_WKThumbnailView _requestSnapshotIfNeeded]):
Don't bail from requesting a snapshot just because we already had one; we want to resnapshot.
Keep track of when we bail from requesting a snapshot because we have one in-flight, so we can request it later.

(-[_WKThumbnailView _didTakeSnapshot:]):
Apply the scale to the bitmap size (whoops!).
Make CA always resize the content to fill the thumbnail view's layer, respecting aspect ratio.
Re-snapshot if we previously deferred a snapshot.

(-[_WKThumbnailView setScale:]):
If we get a scale change, request a new snapshot.

(-[_WKThumbnailView setUsesSnapshot:]):
Never apply thumbnail scale to the page if we're using snapshots. It's unnecessary,
because scale will be applied simply when painting the software snapshot.
Apply the thumbnail scale if we're going from usesSnapshot -> !usesSnapshot.

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::snapshotAtSize):
Snapshots should use the greater of the two scales, so as to never leave a portion of the snapshot unpainted.

2:48 PM Changeset in webkit [167270] by Beth Dakin

• 2 edits in trunk/Source/WebCore

RenderLayerCompositor's m_layerForOverhangAreas should be offset by the
topContentInset
​https://bugs.webkit.org/show_bug.cgi?id=131632
-and corresponding-
<rdar://problem/16609602>

Reviewed by Tim Horton.

Offset m_layerForOverhangAreas by the topContentInset.

• rendering/RenderLayerCompositor.cpp:

(WebCore::RenderLayerCompositor::updateOverflowControlsLayers):

2:46 PM Changeset in webkit [167269] by Julien Brianceau

• 3 edits in trunk/Source/JavaScriptCore

[sh4] Allow use of SubImmediates in LLINT.
​https://bugs.webkit.org/show_bug.cgi?id=131608

Reviewed by Mark Lam.

Allow use of SubImmediates with const pool so the sh4 architecture can
share the arm path for setEntryAddress macro. It reduces architecture
specific code and lead to a more optimal generated code for sh4.

• llint/LowLevelInterpreter.asm:
• offlineasm/sh4.rb:

2:38 PM Changeset in webkit [167268] by timothy_horton@apple.com

• 2 edits in trunk/Source/WebKit2

REGRESSION (WebKit2 View Gestures): Double-tap doesn't work properly when there's no element underneath the cursor
​https://bugs.webkit.org/show_bug.cgi?id=131629
<rdar://problem/16192821>

Reviewed by Dan Bernstein.

• UIProcess/mac/ViewGestureControllerMac.mm:

(WebKit::ViewGestureController::didCollectGeometryForSmartMagnificationGesture):
If there's nothing under the cursor, zoom towards the cursor instead of towards 0,0.

2:24 PM Changeset in webkit [167267] by ap@apple.com

• 5 edits in trunk/Source/WebCore

Eliminate CachedFrame::m_mousePressNode
​https://bugs.webkit.org/show_bug.cgi?id=131626

Reviewed by Brady Eidson.

I couldn't find any observable effect of this change.

• history/CachedFrame.cpp:

(WebCore::CachedFrameBase::CachedFrameBase):
(WebCore::CachedFrameBase::restore):
(WebCore::CachedFrame::clear):

• history/CachedFrame.h:

(WebCore::CachedFrame::documentLoader):
(WebCore::CachedFrame::mousePressNode): Deleted.
Eliminated m_mousePressNode, accessor, and code that reached out to EventHandler.

• page/EventHandler.cpp:

(WebCore::EventHandler::mousePressNode): Deleted.
(WebCore::EventHandler::setMousePressNode): Deleted.

• page/EventHandler.h:

(WebCore::EventHandler::setMousePressed): Deleted. This function was already unused.

1:46 PM Changeset in webkit [167266] by Simon Fraser

• 11 edits in trunk/Source

Run filter animations in the UI process with UI-side compositing
​https://bugs.webkit.org/show_bug.cgi?id=131199
<rdar://problem/16479487>

Source/WebCore:

Reviewed by Sam Weinig.

Add FilterOperation::clone() which is used during decoding.
Export some things.

• WebCore.exp.in:
• platform/graphics/filters/FilterOperation.h:

Source/WebKit2:

Reviewed by Sam Weinig.

• Shared/WebCoreArgumentCoders.cpp: Removed lots of WebCore::

(IPC::ArgumentCoder<SelectionRect>::decode):
(IPC::ArgumentCoder<PasteboardWebContent>::encode):
(IPC::ArgumentCoder<PasteboardWebContent>::decode):
(IPC::ArgumentCoder<PasteboardImage>::encode):
(IPC::ArgumentCoder<PasteboardImage>::decode):
(IPC::ArgumentCoder<URL>::decode):
(IPC::ArgumentCoder<UserStyleSheet>::encode):
(IPC::ArgumentCoder<UserStyleSheet>::decode):
(IPC::ArgumentCoder<UserScript>::encode):
(IPC::ArgumentCoder<UserScript>::decode):
(IPC::ArgumentCoder<ScrollableAreaParameters>::encode):
(IPC::ArgumentCoder<ScrollableAreaParameters>::decode):
(IPC::ArgumentCoder<FixedPositionViewportConstraints>::encode):
(IPC::ArgumentCoder<FixedPositionViewportConstraints>::decode):
(IPC::ArgumentCoder<StickyPositionViewportConstraints>::encode):
(IPC::ArgumentCoder<StickyPositionViewportConstraints>::decode):
(IPC::ArgumentCoder<FilterOperation>::encode):
(IPC::decodeFilterOperation):
(IPC::ArgumentCoder<FilterOperations>::encode):
(IPC::ArgumentCoder<WebCore::UserStyleSheet>::encode): Deleted.
(IPC::ArgumentCoder<WebCore::UserStyleSheet>::decode): Deleted.
(IPC::ArgumentCoder<WebCore::UserScript>::encode): Deleted.
(IPC::ArgumentCoder<WebCore::UserScript>::decode): Deleted.
(IPC::ArgumentCoder<WebCore::ScrollableAreaParameters>::encode): Deleted.
(IPC::ArgumentCoder<WebCore::ScrollableAreaParameters>::decode): Deleted.
(IPC::ArgumentCoder<WebCore::FixedPositionViewportConstraints>::encode): Deleted.
(IPC::ArgumentCoder<WebCore::FixedPositionViewportConstraints>::decode): Deleted.
(IPC::ArgumentCoder<WebCore::StickyPositionViewportConstraints>::encode): Deleted.
(IPC::ArgumentCoder<WebCore::StickyPositionViewportConstraints>::decode): Deleted.
(IPC::encodeFilterOperation): Deleted.
Add encoding/decoding support for FilterOperation.
decodeFilterOperation() has to be a bare function because of the RefPtr
out parameter.
REFERENCE filters should never be encoded, since they have CachedSVGDocumentReferences.

• Shared/WebCoreArgumentCoders.h:

• Shared/mac/RemoteLayerTreeTransaction.mm:

(WebKit::RemoteLayerTreeTextStream::operator<<):
Add logging for filter animations to the RemoteLayerTree log output.

• WebProcess/WebPage/mac/GraphicsLayerCARemote.h:
• WebProcess/WebPage/mac/GraphicsLayerCARemote.cpp:

(WebKit::GraphicsLayerCARemote::addAnimation): Deleted. We can run
filter animations now.

• WebProcess/WebPage/mac/PlatformCAAnimationRemote.h:

KeyValues have a RefPtr<WebCore::FilterOperation> now (which can't be part of
the union because we need its constructor to get called).

• WebProcess/WebPage/mac/PlatformCAAnimationRemote.mm:

(WebKit::PlatformCAAnimationRemote::KeyframeValue::encode): Encode the filter.
(WebKit::PlatformCAAnimationRemote::KeyframeValue::decode): Decode the filter.
(WebKit::PlatformCAAnimationRemote::setFromValue): Implement.
(WebKit::PlatformCAAnimationRemote::setToValue): Implement.
(WebKit::PlatformCAAnimationRemote::setValues): Implement.
(WebKit::animationValueFromKeyframeValue): Handle filters.

1:38 PM Changeset in webkit [167265] by Simon Fraser

• 2 edits in trunk/Source/WebKit2

Fix the build.

• UIProcess/mac/PageClientImpl.h:

1:37 PM Changeset in webkit [167264] by jer.noble@apple.com

• 4 edits in trunk/Source/WebCore

Use after free in WebCore::CachedResourceHandleBase::~CachedResourceHandleBase / WebCore::removeDetachedChildrenInContainer
​https://bugs.webkit.org/show_bug.cgi?id=131169

Reviewed by Eric Carlson.

Invalidate the WebCoreAVFResourceLoader owned by MediaPlayerPrivateAVFoundationObjC
in its destructor, to prevent a private function being called in response to the
WebCoreAVFResourceLoader being stopped.

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

(WebCore::MediaPlayerPrivateAVFoundationObjC::~MediaPlayerPrivateAVFoundationObjC):

• platform/graphics/avfoundation/objc/WebCoreAVFResourceLoader.h:
• platform/graphics/avfoundation/objc/WebCoreAVFResourceLoader.mm:

(WebCore::WebCoreAVFResourceLoader::invalidate):

1:31 PM Changeset in webkit [167263] by mark.lam@apple.com

• 2 edits in trunk/Source/WTF

monotonicallyIncreasingTime() should only initialize its static timebaseInfo once.
<​https://webkit.org/b/131630>

Reviewed by Filip Pizlo.

The current initialization of the static field is not thread safe.

• wtf/CurrentTime.cpp:

(WTF::monotonicallyIncreasingTime):

12:49 PM Changeset in webkit [167262] by Simon Fraser

• 21 edits in trunk/Source

[WK2 iOS] Scrolling to anchor links is broken
​https://bugs.webkit.org/show_bug.cgi?id=131618
<rdar://problem/16599144>

Source/WebCore:

Reviewed by Tim Horton.

Have ScrollingTreeScrollingNode pass RequestedScrollPosition updates
to the scrolling tree, so that the scrolling tree can have custom behavior
for them if necessary.

• page/scrolling/ScrollingTree.h:

(WebCore::ScrollingTree::scrollingTreeNodeRequestsScroll):

• page/scrolling/ScrollingTreeScrollingNode.cpp:

(WebCore::ScrollingTreeScrollingNode::updateAfterChildren):

• page/scrolling/ScrollingTreeScrollingNode.h:

Source/WebKit2:

Reviewed by Tim Horton.

The RemoteScrollingTree implements scrollingTreeNodeRequestsScroll
to get informed about requested scroll position updates, and passes
them along via the RemoteScrollingCoordinatorProxy, WebPageProxy and PageClient
to the WKWebView, which performs a scroll.

• UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _scrollToContentOffset:WebCore::]): Scroll to content offset,
taking page scale and insets into account.

• UIProcess/API/Cocoa/WKWebViewInternal.h:
• UIProcess/CoordinatedGraphics/WebView.cpp:

(WebKit::WebView::requestScroll):

• UIProcess/CoordinatedGraphics/WebView.h:
• UIProcess/PageClient.h:
• UIProcess/Scrolling/RemoteScrollingCoordinatorProxy.cpp:

(WebKit::RemoteScrollingCoordinatorProxy::scrollingTreeNodeRequestsScroll):
Pass scrolls along to the WebPageProxy for the root node. We will also need
to handle programmatic scrolls for overflow soon.

• UIProcess/Scrolling/RemoteScrollingCoordinatorProxy.h:
• UIProcess/Scrolling/RemoteScrollingTree.cpp:

(WebKit::RemoteScrollingTree::scrollingTreeNodeRequestsScroll):

• UIProcess/Scrolling/RemoteScrollingTree.h:
• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::requestScroll):

• UIProcess/WebPageProxy.h:
• UIProcess/ios/PageClientImplIOS.h:
• UIProcess/ios/PageClientImplIOS.mm:

(WebKit::PageClientImpl::canScrollView):
(WebKit::PageClientImpl::requestScroll):

• UIProcess/mac/PageClientImpl.mm:

(WebKit::PageClientImpl::requestScroll):

12:45 PM Changeset in webkit [167261] by Brian Burg

• 4 edits
  1 add in trunk

Web Replay: memoize fallback time values for document.lastModified
​https://bugs.webkit.org/show_bug.cgi?id=131318

Reviewed by Joseph Pecoraro.

Source/WebCore:
If a document's Last-Modified header can't be found or used, then
document.lastModified is derived from the current system time or
from filesystem data, which is obviously nondeterministic.

It's better to handle this inside Document::lastModified rather than using
MemoizedDOMResult, because only the fallback case is nondeterministic.

Test: http/tests/inspector/replay/document-last-modified-fallback-value.html

• dom/Document.cpp:

(WebCore::Document::lastModified): Save or reuse memoized fallback value.

• replay/WebInputs.json: Add input DocumentLastModifiedDate.

LayoutTests:

• http/tests/inspector/replay/document-last-modified-fallback-value.html: Added.

12:11 PM Changeset in webkit [167260] by ggaren@apple.com

• 8 edits in trunk/PerformanceTests

A few MallocBench record/replay fixes
​https://bugs.webkit.org/show_bug.cgi?id=131627

Reviewed by Andreas Kling.

• MallocBench/MallocBench/Interpreter.cpp:

(Interpreter::run): Accept 0-sized allocations without asserting because
WebKit does that sometimes.

• MallocBench/MallocBench/flickr.ops:
• MallocBench/MallocBench/flickr_memory_warning.ops:
• MallocBench/MallocBench/reddit.ops:
• MallocBench/MallocBench/reddit_memory_warning.ops:
• MallocBench/MallocBench/theverge.ops:
• MallocBench/MallocBench/theverge_memory_warning.ops: Updated these

recordings because a bug in the recording mechanism caused one out of
every few thousand slot values to be bogus.

12:02 PM Changeset in webkit [167259] by commit-queue@webkit.org

• 3 edits in trunk/LayoutTests

[GTK] Unreviewed GTK gardening.

Patch by Eduardo Lima Mitev <​elima@igalia.com> on 2014-04-14

• platform/gtk/TestExpectations: Update test expectations for new failing test 'editing/editability/ignored-content.html'.
• platform/gtk/inspector-protocol/dom/getAccessibilityPropertiesForNode-expected.txt: Rebaselined.

12:00 PM Changeset in webkit [167258] by bshafiei@apple.com

• 5 edits in trunk/Source

Versioning.

11:57 AM Changeset in webkit [167257] by bshafiei@apple.com

• 1 copy in tags/Safari-538.29

New tag.

11:40 AM Changeset in webkit [167256] by Antti Koivisto

• 9 edits in trunk/Source

Keep secondary tile grid for zoomed-out scale
​https://bugs.webkit.org/show_bug.cgi?id=131586

Reviewed by Darin Adler.

Source/WebCore:

• platform/graphics/ca/GraphicsLayerCA.cpp:

(WebCore::GraphicsLayerCA::updateContentsScale):

Don't repaint tiled backing with setNeedsDisplay, it invalidates itself correctly in setContentsScale.
Update custom child layers when tiled backing scale changes.

• platform/graphics/ca/mac/TileController.h:
• platform/graphics/ca/mac/TileController.mm:

(WebCore::TileController::TileController):
(WebCore::TileController::setNeedsDisplay):

Drop the whole zoomed-out grid on full repaint.

(WebCore::TileController::setNeedsDisplayInRect):

Drop changed zoomed-out tiles. A more sophisticated strategy is possible.

(WebCore::TileController::setContentsScale):

Swap the zoomed-out grid in and out as needed.
Repaint the active grid after scale change so the client does not have to.

(WebCore::TileController::contentsScale):

Get the content scale from the tile grid so it is not kept in two places.

(WebCore::TileController::zoomedOutContentsScale):
(WebCore::TileController::setZoomedOutContentsScale):

Drop the zoomed-out grid if it no longer matches the zoomed-out scale.

(WebCore::TileController::tileRevalidationTimerFired):
(WebCore::TileController::retainedTileBackingStoreMemory):
(WebCore::TileController::containerLayers):

Return both zoomed-out tiles and the active tiles. Active tiles are on top.

(WebCore::TileController::numberOfUnparentedTiles):
(WebCore::TileController::removeUnparentedTilesNow):

• platform/graphics/ca/mac/TileGrid.h:
• platform/graphics/ca/mac/TileGrid.mm:

(WebCore::TileGrid::dropTilesInRect):

Add a function for dropping tiles.

(WebCore::TileGrid::revalidateTiles):

Source/WebKit2:

• WebProcess/WebPage/mac/PlatformCALayerRemoteTiledBacking.cpp:

(WebKit::PlatformCALayerRemoteTiledBacking::PlatformCALayerRemoteTiledBacking):
(WebKit::PlatformCALayerRemoteTiledBacking::customSublayers):

Always request new sublayer list from tile controller.

• WebProcess/WebPage/mac/PlatformCALayerRemoteTiledBacking.h:

11:33 AM WebKitGTK/KeepingTheTreeGreen edited by elima@igalia.com

(diff)

11:17 AM WebKitGTK/KeepingTheTreeGreen edited by elima@igalia.com

(diff)

11:05 AM April 2014 Meeting edited by zoltan@webkit.org

(diff)

11:04 AM Changeset in webkit [167255] by akling@apple.com

• 2 edits in trunk/Source/JavaScriptCore

Array.prototype.concat should allocate output storage only once.
<​https://webkit.org/b/131609>

Do a first pass across 'this' and any arguments to compute the
final size of the resulting array from Array.prototype.concat.
This avoids having to grow the output incrementally as we go.

This also includes two other micro-optimizations:

• Mark getProperty() with ALWAYS_INLINE.

• Use JSArray::length() instead of taking the generic property lookup path when we know an argument is an Array.

My MBP says ~3% progression on Dromaeo/jslib-traverse-jquery.

Reviewed by Oliver & Darin.

• runtime/ArrayPrototype.cpp:

(JSC::getProperty):
(JSC::arrayProtoFuncConcat):

11:02 AM April 2014 Meeting edited by betravis@adobe.com

(diff)

11:00 AM April 2014 Meeting edited by Bem Jones-Bey

(diff)

10:51 AM April 2014 Meeting edited by ddkilzer@webkit.org

(diff)

10:49 AM April 2014 Meeting edited by Bem Jones-Bey

(diff)

10:33 AM Changeset in webkit [167254] by ap@apple.com

• 2 edits in trunk/Source/bmalloc

Fixed svn:ignore on bmalloc.xcodeproj, it had erroneous leading spaces.

• bmalloc.xcodeproj: Modified property svn:ignore.

10:27 AM Changeset in webkit [167253] by timothy_horton@apple.com

• 7 edits in trunk/Source/WebKit2

Make WK(Web)View magnification setters actually use view-relative positions
​https://bugs.webkit.org/show_bug.cgi?id=131611
<rdar://problem/15965239>

Reviewed by Darin Adler.

• UIProcess/API/mac/WKView.mm:

(-[WKView setMagnification:centeredAtPoint:]):
(-[WKView setMagnification:]):
Use scalePageInViewCoordinates instead.

• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::scalePageInViewCoordinates):

• UIProcess/WebPageProxy.h:
• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::scalePageInViewCoordinates):
(WebKit::WebPage::pageScaleFactor):

• WebProcess/WebPage/WebPage.h:
• WebProcess/WebPage/WebPage.messages.in:

Add scalePageInViewCoordinates, which turns the scale centerpoint within the view
into what scalePage expects: a post-scale scroll offset.

10:26 AM Changeset in webkit [167252] by akling@apple.com

• 2 edits in trunk/Source/JavaScriptCore

Unreviewed, rolling out r167249.
​https://bugs.webkit.org/show_bug.cgi?id=131621

broke 3 tests on cloop (Requested by kling on #webkit).

Reverted changeset:

"Array.prototype.concat should allocate output storage only
once."
​https://bugs.webkit.org/show_bug.cgi?id=131609
​http://trac.webkit.org/changeset/167249

Patch by Commit Queue <​commit-queue@webkit.org> on 2014-04-14

10:20 AM WebKitGTK/KeepingTheTreeGreen edited by elima@igalia.com

(diff)

10:17 AM Changeset in webkit [167251] by oliver@apple.com

• 2 edits in trunk/Source/WebCore

Update test result

9:52 AM WebKitGTK/KeepingTheTreeGreen edited by elima@igalia.com

(diff)

9:31 AM Changeset in webkit [167250] by commit-queue@webkit.org

• 2 edits in trunk/Source/JavaScriptCore

Fixed potential integer truncation.
​https://bugs.webkit.org/show_bug.cgi?id=131615

Patch by Alex Christensen <​achristensen@webkit.org> on 2014-04-14
Reviewed by Darin Adler.

• assembler/X86Assembler.h:

(JSC::X86Assembler::fillNops):
Truncate the size_t to an unsigned after it is limited to 15 instead of before.

9:19 AM Changeset in webkit [167249] by akling@apple.com

• 2 edits in trunk/Source/JavaScriptCore

Array.prototype.concat should allocate output storage only once.
<​https://webkit.org/b/131609>

Do a first pass across 'this' and any arguments to compute the
final size of the resulting array from Array.prototype.concat.
This avoids having to grow the output incrementally as we go.

This also includes two other micro-optimizations:

• Mark getProperty() with ALWAYS_INLINE.

• Use JSArray::length() instead of taking the generic property lookup path when we know an argument is an Array.

My MBP says ~3% progression on Dromaeo/jslib-traverse-jquery.

Reviewed by Darin Adler.

• runtime/ArrayPrototype.cpp:

(JSC::getProperty):
(JSC::arrayProtoFuncConcat):

9:02 AM Changeset in webkit [167248] by krit@webkit.org

• 5 edits in trunk

Optimize Canvas fill and drawImage with SourceIn, DestinationIn, SourceOut, and DestinationAtop using transparencyLayer.
​https://bugs.webkit.org/show_bug.cgi?id=79659

Reviewed by Darin Adler.

Source/WebCore:
Optimize fill() and fillRect() operations in Canvas on composited contexts by
10 to 20 times on CG.

Replacing the ImageBuffer code by transparency layers allows the
graphics library to optimize the drawing.

Doing the same for drawImage() would give performance regressions.

An inline function will create a transparency layer for CG. Cairo graphics
does not composite correctly when a transparency layer gets created.
The inline function is just a NOOP for Cairo.

This fixes bug 131303 as well.

Added performance tests with r167124 already.

• html/canvas/CanvasRenderingContext2D.cpp:

(WebCore::CanvasRenderingContext2D::fillInternal):
(WebCore::CanvasRenderingContext2D::strokeInternal):
(WebCore::CanvasRenderingContext2D::beginCompositeLayer):
(WebCore::CanvasRenderingContext2D::endCompositeLayer):
(WebCore::CanvasRenderingContext2D::fillRect):
(WebCore::CanvasRenderingContext2D::strokeRect):
(WebCore::CanvasRenderingContext2D::drawTextInternal):
(WebCore::CanvasRenderingContext2D::fullCanvasCompositedFill): Deleted.

• html/canvas/CanvasRenderingContext2D.h:

LayoutTests:
Unskip previously failing tests.

• platform/gtk/TestExpectations:

8:48 AM Changeset in webkit [167247] by Carlos Garcia Campos

• 1 copy in releases/WebKitGTK/webkit-2.4.1

Tagging the WebKitGTK+ 2.4.1 release

8:42 AM Changeset in webkit [167246] by timothy_horton@apple.com

• 2 edits in trunk/Source/WebCore

Lots of compositing test failures after r167152
​https://bugs.webkit.org/show_bug.cgi?id=131574

Reviewed by Darin Adler.

• platform/graphics/GraphicsLayer.cpp:

(WebCore::dumpChildren):
(WebCore::GraphicsLayer::dumpProperties):
Make child-dumping recursive so that we can easily skip layers up to any depth.

8:35 AM Changeset in webkit [167245] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.4

Unreviewed. Update NEWS and Versions.m4 for 2.4.1 release.

.:

• Source/autotools/Versions.m4: Bump version numbers.

Source/WebKit/gtk:

• NEWS: Added release notes for 2.4.1.

5:45 AM Changeset in webkit [167244] by pmolnar.u-szeged@partner.samsung.com

• 3 edits in trunk/Source/WebCore

Fix incorrect indentations in CodeGeneratorJS.pm introduced in r165521
​https://bugs.webkit.org/show_bug.cgi?id=131613

Reviewed by Csaba Osztrogonác.

• bindings/scripts/CodeGeneratorJS.pm:

(GenerateImplementation):
Fixed 5-space indentation.

• bindings/scripts/test/JS/JSTestNondeterministic.cpp:

Updated the tests accordingly.

5:24 AM Changeset in webkit [167243] by commit-queue@webkit.org

• 2 edits in trunk/Tools

Fix commit-log-editor bug revealed by r165447
​https://bugs.webkit.org/show_bug.cgi?id=130676

Patch by Jozsef Berta <​jberta.u-szeged@partner.samsung.com> on 2014-04-14
Reviewed by Csaba Osztrogonác.

• Scripts/commit-log-editor:

(createCommitMessage): Omitting empty Source/JavaScriptCore:... blocks.
Add \n before the first block too, because the longest common prefix now ends with only one newline.
(removeLongestCommonPrefixEndingInNewline): The longest common prefix ends with only one newline,
now the last block of the common prefix isn't duplicated below. Changing the function name accordingly.

4:41 AM WebKitGTK/2.4.x edited by Carlos Garcia Campos

(diff)

4:40 AM Changeset in webkit [167242] by Carlos Garcia Campos

• 9 edits
  4 adds in releases/WebKitGTK/webkit-2.4

Merge r167193 - [GStreamer] No CORS support for media elements
​https://bugs.webkit.org/show_bug.cgi?id=99037

Patch by Youenn Fablet <​youenn.fablet@crf.canon.fr> on 2014-04-13
Reviewed by Philippe Normand.

Source/WebCore:

Added CORS access control check to media sources when crossorigin attribute is set.

Added getter to CORS access control check status (used to compute whether the stream is tainted or not).
Related test is http/tests/security/video-cross-origin-readback.html.

Disabled access to cross-origin streams that fail CORS check when crossorigin attribute is set.
Related test is http/tests/security/video-cross-origin-accessfailure.html.

Tests: http/tests/security/video-cross-origin-accessfailure.html

http/tests/security/video-cross-origin-accesssameorigin.html

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:

(WebCore::MediaPlayerPrivateGStreamer::didPassCORSAccessCheck): Return whether media is cross-origin (tainted) or not by querying the gstreamer source layer.

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h: Added MediaPlayerPrivateGStreamer::didPassCORSAccessCheck declaration.
• platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:

(webKitWebSrcStart): Passed CORS mode parameter to the streaming client. In case of CORS check failure, stop the resource loading.
(webKitSrcPassedCORSAccessCheck): Return whether CORS access control check was done and successful.
(StreamingClient::handleResponseReceived): Take a parameter to assign the CORS access control check result.
(CachedResourceStreamingClient::CachedResourceStreamingClient): Updated setting of the ResourceLoaderOptions according CORS mode.
(CachedResourceStreamingClient::responseReceived): Check CORS and pass result to handleResponseReceived.
(ResourceHandleStreamingClient::didReceiveResponse): No CORS check.

• platform/graphics/gstreamer/WebKitWebSourceGStreamer.h: Added webKitSrcPassedCORSAccessCheck declaration.

LayoutTests:

http/tests/security/video-cross-origin-accessfailure.html verifies that cross-origin streams that fail CORS check
are not played when crossorigin attribute is set.

http/tests/security/video-cross-origin-accesssameorigin.html verifies that access to same-origin streams

are played when crossorigin attribute is set.

• http/tests/security/video-cross-origin-accessfailure-expected.txt: Added.
• http/tests/security/video-cross-origin-accessfailure.html: Added.
• http/tests/security/video-cross-origin-accesssameorigin-expected.txt: Added.
• http/tests/security/video-cross-origin-accesssameorigin.html: Added.
• platform/efl/TestExpectations: Enabled http/tests/security/video-cross-origin-readback.html.
• platform/gtk/TestExpectations: Ditto.
• platform/mac/TestExpectations: Disabled http/tests/security/video-cross-origin-accessfailure.html.

4:25 AM Changeset in webkit [167241] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.4/Source/WebCore

Merge r166304 - REGRESSION(r162679): Poster image visible under the video
​https://bugs.webkit.org/show_bug.cgi?id=130783

Reviewed by Simon Fraser.

In the listed revision, we started checking for isRenderImage()
instead of isImage(). RenderMedias return 'true' for the first
but 'false' for the second. Change the if() statement to check
for isRenderMedia() in addition to !isRenderImage().

• rendering/RenderLayerBacking.cpp:

(WebCore::RenderLayerBacking::isDirectlyCompositedImage):

4:21 AM Changeset in webkit [167240] by Manuel Rego Casasnovas

• 3 edits
  2 adds in trunk

[JSC] CSSStyleDeclaration report incorrect descriptor
​https://bugs.webkit.org/show_bug.cgi?id=89697

Reviewed by Benjamin Poulain.

Source/WebCore:

Change descriptor of CSSStyleDeclaration properties in order to have
writable and enumerable attributes set to true. Configurable is kept to
false since the property is not deleteable.

Test: fast/dom/CSSStyleDeclaration/cssstyledeclaration-properties-descriptor.html

• bindings/js/JSCSSStyleDeclarationCustom.cpp:

(WebCore::JSCSSStyleDeclaration::getOwnPropertySlotDelegate): Only set
DontDelete attribute when creating the descriptor for
CSSStyleDeclaration properties.

LayoutTests:

Add new test to check the descriptor of CSSStyleDeclaration properties.

• fast/dom/CSSStyleDeclaration/cssstyledeclaration-properties-descriptor-expected.txt: Added.
• fast/dom/CSSStyleDeclaration/cssstyledeclaration-properties-descriptor.html: Added.

4:20 AM Changeset in webkit [167239] by Carlos Garcia Campos

• 3 edits
  2 adds in releases/WebKitGTK/webkit-2.4

Merge r166090 - Source/WebCore: Fix a crash when assigning an object to document.location
​https://bugs.webkit.org/show_bug.cgi?id=130213

Reviewed by Geoffrey Garen.

Convert location to string before we make use the document.
This prevents us from attempting to navigate a frame that
has already been removed.

Test: fast/dom/navigation-with-sideeffects-crash.html

• bindings/js/JSDocumentCustom.cpp:

(WebCore::JSDocument::location):
(WebCore::JSDocument::setLocation):

LayoutTests: Fix semantics of JS execution when assigning an object to document.location
​https://bugs.webkit.org/show_bug.cgi?id=130213

Reviewed by Geoffrey Garen.

• fast/dom/navigation-with-sideeffects-expected.txt: Added.
• fast/dom/navigation-with-sideeffects.html: Added.

4:12 AM Changeset in webkit [167238] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.4/Source/WebKit2

Merge r166026 - Generalize WebInspector check in maybeInitializeSandboxExtensionHandle().
​https://bugs.webkit.org/show_bug.cgi?id=130079
<rdar://problem/16286683>

Reviewed by Anders Carlsson.

• UIProcess/WebPageProxy.cpp: (WebKit::WebPageProxy::maybeInitializeSandboxExtensionHandle):

Perform an early return for all paths that don't need a sandbox extension due to
access being assumed, not only for WebInspector pages.

• UIProcess/WebProcessProxy.cpp:

(WebKit::WebProcessProxy::hasAssumedReadAccessToURL):
(WebKit::WebProcessProxy::checkURLReceivedFromWebProcess):

• UIProcess/WebProcessProxy.h:

Factored out m_localPathsWithAssumedReadAccess iteration to a public function.

4:04 AM Changeset in webkit [167237] by Carlos Garcia Campos

• 4 edits
  2 adds in releases/WebKitGTK/webkit-2.4

Merge r165921 - Crash with long selector list
​https://bugs.webkit.org/show_bug.cgi?id=123006

Reviewed by Andreas Kling.

Source/WebCore:

Test: fast/css/long-selector-list-crash.html

• css/CSSSelectorList.cpp:

(WebCore::CSSSelectorList::CSSSelectorList):
(WebCore::CSSSelectorList::adoptSelectorVector):
(WebCore::CSSSelectorList::operator=):

• css/StyleRule.cpp:

(WebCore::StyleRule::create):

Add a bunch of asserts.

(WebCore::StyleRule::splitIntoMultipleRulesWithMaximumSelectorComponentCount):

This could produce a zero-length selector list.

LayoutTests:

• fast/css/long-selector-list-crash-expected.txt: Added.
• fast/css/long-selector-list-crash.html: Added.

3:53 AM Changeset in webkit [167236] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.4/Source/JavaScriptCore

Merge r165902 - Update type of local vars to match the type of String length.
<​https://webkit.org/b/130077>

Reviewed by Geoffrey Garen.

• runtime/JSStringJoiner.cpp:

(JSC::JSStringJoiner::join):

3:47 AM Changeset in webkit [167235] by Carlos Garcia Campos

• 5 edits
  4 adds in releases/WebKitGTK/webkit-2.4

Merge r165821 - Mutating rules returned by getMatchedCSSRules can result in crash
​https://bugs.webkit.org/show_bug.cgi?id=130209

Source/WebCore:

Reviewed by Andreas Kling.

The non-standard getMatchedCSSRules API returns CSSStyleRule objects that don't
have parent stylesheet pointer (as we don't know which sheet the rule originated from).
Mutating the rule via such wrapper can lead to crashes later as we fail to invalidate
the underlying stylesheet.

Fix by disallowing mutation of style rules that don't have parent sheet pointer. CSSStyleRule
has two mutable properties selectorText and style. The latter gives back CSSStyleDeclaration.
This patch disallows mutations in both cases for CSSStyleRules that don't have parent stylesheet
pointer.

While it is technically possible to have CSSRules that are legitimately disconnected
from stylesheet (by removing rule from sheet while holding a reference to it) it never
makes sense to mutate such rule as there is no way to do anything with it afterwards.

Tests: fast/css/getMatchedCSSProperties-rule-mutation.html

fast/css/getMatchedCSSRules-crash.html

• css/CSSStyleRule.cpp:

(WebCore::CSSStyleRule::setSelectorText):

Bail out if parent stylesheet is null.

• css/PropertySetCSSStyleDeclaration.cpp:

(WebCore::PropertySetCSSStyleDeclaration::setCssText):
(WebCore::PropertySetCSSStyleDeclaration::setProperty):
(WebCore::PropertySetCSSStyleDeclaration::removeProperty):
(WebCore::PropertySetCSSStyleDeclaration::setPropertyInternal):

Allow StyleRuleCSSStyleDeclaration subclass cancel the mutation via
boolean return value from willMutate.

(WebCore::StyleRuleCSSStyleDeclaration::willMutate):

Disallow mutation if the owning CSSStyleRule is null or has null stylesheet.

(WebCore::StyleRuleCSSStyleDeclaration::didMutate):

We never get here with null rule or stylesheet anymore.

• css/PropertySetCSSStyleDeclaration.h:

(WebCore::PropertySetCSSStyleDeclaration::willMutate):

LayoutTests:

Reviewed by Andreas Kling.

• fast/css/getMatchedCSSProperties-rule-mutation-expected.txt: Added.
• fast/css/getMatchedCSSProperties-rule-mutation.html: Added.
• fast/css/getMatchedCSSRules-crash-expected.txt: Added.
• fast/css/getMatchedCSSRules-crash.html: Added.

3:38 AM Changeset in webkit [167234] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.4/Source/WebCore

Merge r165548 - Don't send synchronous resize events when FrameView has auto-sizing enabled.
<​https://webkit.org/b/130198>
<rdar://problem/15991333>

Reviewed by Dan Bernstein.

• page/FrameView.cpp:

(WebCore::FrameView::sendResizeEventIfNeeded):

3:26 AM Changeset in webkit [167233] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.4/Source/WTF

Merge r165425 - Improve WeakPtr operators.
​https://bugs.webkit.org/show_bug.cgi?id=130053

Reviewed by Andreas Kling.

Replace the "operator!()" with an explicit bool operator. Add an "operator->()".

• wtf/WeakPtr.h:

(WTF::WeakPtr::operator bool):
(WTF::WeakPtr::operator->):

3:05 AM Changeset in webkit [167232] by Carlos Garcia Campos

• 5 edits
  3 adds in releases/WebKitGTK/webkit-2.4

Merge r165339 - SerializedScriptValue may move Identifiers between worlds
​https://bugs.webkit.org/show_bug.cgi?id=129979

Reviewed by Andreas Kling.

Source/WebCore:

Test: fast/workers/worker-copy-shared-blob-url.html

Don't use Strings to store blob URLs as String's may be Identifiers
and they can only exist in one world/thread at a time.

• Modules/indexeddb/IDBObjectStore.cpp:

(WebCore::IDBObjectStore::put):

• bindings/js/SerializedScriptValue.cpp:

(WebCore::CloneDeserializer::deserializeString):
(WebCore::SerializedScriptValue::addBlobURL):
(WebCore::SerializedScriptValue::SerializedScriptValue):

• bindings/js/SerializedScriptValue.h:

LayoutTests:

Add test cases

• fast/workers/resources/worker-copy-shared-blob-url-worker.js: Added.

(count.0.onmessage):

• fast/workers/worker-copy-shared-blob-url-expected.txt: Added.
• fast/workers/worker-copy-shared-blob-url.html: Added.

2:47 AM Changeset in webkit [167231] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.4/Source/WebCore

Merge r165145 - ASSERT(newestManifest) fails in WebCore::ApplicationCacheGroup::didFinishLoadingManifest()
​https://bugs.webkit.org/show_bug.cgi?id=129753
<rdar://problem/12069835>

Reviewed by Alexey Proskuryakov.

Fixes an issue where an assertion failure would occur when visiting a web site whose on-disk
app cache doesn't contain a manifest resource.

For some reason an app cache for a web site may be partially written to disk. In particular, the
app cache may only contain a CacheGroups entry. That is, the manifest resource and origin records
may not be persisted to disk. From looking over the code, we're unclear how such a situation can occur
and hence have been unable to create such an app cache. We were able to reproduce this issue using
an app cache database file that was provided by a person that was affected by this issue.

No test included because it's not straightforward to write a test for this change.

• loader/appcache/ApplicationCacheGroup.cpp:

(WebCore::ApplicationCacheGroup::checkIfLoadIsComplete): Assert that m_cacheBeingUpdated->manifestResource()
is non-null. Currently we only document this assumption in a code comment. Also separated a single assertion
expression into two assertion expressions to make it straightforward to identify the failing sub-expression
on failure.

• loader/appcache/ApplicationCacheStorage.cpp:

(WebCore::ApplicationCacheStorage::store): Modified to call ApplicationCacheStorage::deleteCacheGroupRecord()
to remove a cache group and associated cache records (if applicable) before inserting a cache group entry.
This replacement approach will ultimately repair incomplete app cache data for people affected by this bug.
(WebCore::ApplicationCacheStorage::loadCache): Log an error and return nullptr if the cache we loaded doesn't
have a manifest resource.
(WebCore::ApplicationCacheStorage::deleteCacheGroupRecord): Added.
(WebCore::ApplicationCacheStorage::deleteCacheGroup): Extracted deletion logic for cache group record into
ApplicationCacheStorage::deleteCacheGroupRecord().

• loader/appcache/ApplicationCacheStorage.h:

2:43 AM Changeset in webkit [167230] by Carlos Garcia Campos

• 4 edits
  3 adds in releases/WebKitGTK/webkit-2.4

Merge r165138 - Fix crash in CompositeEditCommand::cloneParagraphUnderNewElement()
<​http://webkit.org/b/129751>
<rdar://problem/16237965>

Reviewed by Jon Honeycutt.

Merged from Blink (patch by Yuta Kitamura):
​https://src.chromium.org/viewvc/blink?revision=168160&view=revision
​http://crbug.com/345005

The root cause is CompositeEditCommand::moveParagraphWithClones() passing
two positions |start| and |end| which do not follow the document order,
i.e. in some situations |start| is located after |end| because of
the difference in affinity.

This patch fixes this crash by normalizing |end| to |start| in such situations.
It also adds an ASSERT that checks the relationship between |start| and |end|.

Source/WebCore:

Test: editing/execCommand/format-block-crash.html

• editing/CompositeEditCommand.cpp:

(WebCore::CompositeEditCommand::cloneParagraphUnderNewElement):
(WebCore::CompositeEditCommand::moveParagraphWithClones):

• editing/CompositeEditCommand.h:

LayoutTests:

• editing/execCommand/format-block-crash-expected.txt: Added.
• editing/execCommand/format-block-crash.html: Added.
• editing/execCommand/resources/format-block-crash-iframe.html: Added.

2:37 AM Changeset in webkit [167229] by Carlos Garcia Campos

• 3 edits
  2 adds in releases/WebKitGTK/webkit-2.4

Merge r164933 - Ensure keySplines is valid in SMIL animations
<​http://webkit.org/b/129547>
<rdar://problem/15676128>

Reviewed by Darin Adler.

Merged from Blink (patch by Philip Rogers):
​https://src.chromium.org/viewvc/blink?revision=156452&view=revision
​http://crbug.com/276111

This patch fixes a crash in SMIL animations when keySplines are not
specified. The SMIL spec is clear on this:
​http://www.w3.org/TR/2001/REC-smil-animation-20010904/#AnimFuncCalcMode
"If there are any errors in the keyTimes specification (bad values,
too many or too few values), the animation will have no effect."

This patch simply checks that keyTimes is not empty. Previously,
splinesCount was set to be m_keySplines.size() + 1 in
SVGAnimationElement.cpp; this patch changes splinesCount to be equal
to m_keySplines.size() to make the logic easier to follow and to
match other checks in SVGAnimationElement::startedActiveInterval.

Source/WebCore:

Test: svg/animations/animate-keysplines-crash.html

• svg/SVGAnimationElement.cpp:

(WebCore::SVGAnimationElement::startedActiveInterval):

LayoutTests:

• svg/animations/animate-keysplines-crash-expected.txt: Added.
• svg/animations/animate-keysplines-crash.html: Added.

2:34 AM Changeset in webkit [167228] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.4/Source/WebCore

Merge r164876 - Properly clear m_logicallyLastRun to remove use-after-free possibility
​https://bugs.webkit.org/show_bug.cgi?id=129489

Reviewed by David Hyatt.

A use-after-free issue was caught in Blink because m_logicallyLastRun
is not cleared when the item it points to is deleted. Clearing it
turns the use-after-free into a segfault, and prevents any future
use-after-frees from happening.

• platform/text/BidiRunList.h:

(WebCore::BidiRunList<Run>::deleteRuns):

2:31 AM Changeset in webkit [167227] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.4/Source/WTF

Merge r164408 - Crash in WTF::StringBuilder::append()

​https://bugs.webkit.org/show_bug.cgi?id=125817
<rdar://problem/15671883>

Reviewed by Oliver Hunt.

• wtf/text/StringBuilder.cpp:

(WTF::expandedCapacity):
Ensure that we return a new capacity of at least 'requiredLength' in
the case where requiredLength is large. Also, use unsigned rather than
size_t for the parameters and the return value, as callers pass
unsigned arguments and treat the result as an unsigned int.

2:24 AM Changeset in webkit [167226] by Carlos Garcia Campos

• 3 edits
  2 adds in releases/WebKitGTK/webkit-2.4

Merge r164367 - Do not dispatch change event twice in single step action
​https://bugs.webkit.org/show_bug.cgi?id=116936
<rdar://problem/16086828>

Reviewed by Ryosuke Niwa.

Merged from Blink (patch by Kent Tamura):
​https://src.chromium.org/viewvc/blink?view=rev&revision=151175

Source/WebCore:

Test: fast/forms/number/number-type-update-by-change-event.html

• html/InputType.cpp:

(WebCore::InputType::stepUpFromRenderer):

LayoutTests:

• fast/forms/number/number-type-update-by-change-event-expected.txt: Added.
• fast/forms/number/number-type-update-by-change-event.html: Added.

2:17 AM Changeset in webkit [167225] by Carlos Garcia Campos

• 3 edits
  2 adds in releases/WebKitGTK/webkit-2.4

Merge r164249 - AX: Invalid cast in WebCore::AccessibilityTable::isDataTable (CRBug 280352)
<​https://webkit.org/b/128925>
<rdar://problem/16087351>

Merged from Blink (patch by Dominic Mazzoni):
​https://src.chromium.org/viewvc/blink?revision=159711&view=revision

Reviewed by Oliver Hunt.

Source/WebCore:

Don't cast to a table cell element unless we are sure it is one.

Test: accessibility/display-table-cell-causes-crash.html

• accessibility/AccessibilityTable.cpp:

(WebCore::AccessibilityTable::isDataTable):

LayoutTests:

• accessibility/display-table-cell-causes-crash-expected.txt: Added.
• accessibility/display-table-cell-causes-crash.html: Added.

2:11 AM Changeset in webkit [167224] by Carlos Garcia Campos

• 3 edits
  2 adds in releases/WebKitGTK/webkit-2.4

Merge r164204 - Ensure that removing an iframe from the DOM tree disconnects its Frame.
<​https://webkit.org/b/128889>
<rdar://problem/15671221>

Merged from Blink (patch by Adam Klein):
​https://src.chromium.org/viewvc/blink?revision=156174&view=revision

Source/WebCore:

SubframeLoadingDisabler wasn't catching the case when an <iframe> was,
in its unload handler, removed and re-added to the same parent.
Fix this by using a count of SubframeLoadingDisablers that are on the
stack for a given root, rather than a simple boolean.

Test: fast/frames/reattach-in-unload.html

• html/HTMLFrameOwnerElement.h:

(WebCore::SubframeLoadingDisabler::disabledSubtreeRoots):

LayoutTests:

• fast/frames/reattach-in-unload-expected.txt: Added.
• fast/frames/reattach-in-unload.html: Added.

2:00 AM Changeset in webkit [167223] by Carlos Garcia Campos

• 3 edits
  2 adds in releases/WebKitGTK/webkit-2.4

Merge r164170 - ASSERT_WITH_SECURITY_IMPLICATION in WebCore::toElement
​https://bugs.webkit.org/show_bug.cgi?id=128810

Patch by Renata Hodovan <​rhodovan.u-szeged@partner.samsung.com> on 2014-02-15
Reviewed by Ryosuke Niwa.

Source/WebCore:

Make CompositeEditCommand::cloneParagraphUnderNewElement() to work when |outerNode|
doesn't contain |start|.

Before this patch, CompositeEditCommand::cloneParagraphUnderNewElement() tried to copy
ancestry nodes from |start| to Document node when |start| position isn't in |outerNode|. This
patch changes CompositeEditCommand::cloneParagraphUnderNewElement() to copy |start| to
|outerNode| only if |outerNode| contains |start| position.

Merged from Blink ​https://src.chromium.org/viewvc/blink?revision=161762&view=revision by ​yosin@chromium.org.

Test: editing/execCommand/indent-with-uneditable-crash.html

• editing/CompositeEditCommand.cpp:

(WebCore::CompositeEditCommand::cloneParagraphUnderNewElement):

LayoutTests:

• editing/execCommand/indent-with-uneditable-crash-expected.txt: Added.
• editing/execCommand/indent-with-uneditable-crash.html: Added.

1:53 AM Changeset in webkit [167222] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.4

Merge r163599 - loadSubframe can return null in SubframeLoader::loadOrRedirectSubframe
​https://bugs.webkit.org/show_bug.cgi?id=128344

Reviewed by Ryosuke Niwa.

Source/WebCore:

• loader/SubframeLoader.cpp:

(WebCore::SubframeLoader::loadOrRedirectSubframe):

LayoutTests:

• TestExpectations: Unskipped tests that were previously crashing

1:47 AM Changeset in webkit [167221] by Carlos Garcia Campos

• 6 edits
  6 adds in releases/WebKitGTK/webkit-2.4

Merge r163579 - Make adoption agency use the task queue
​https://bugs.webkit.org/show_bug.cgi?id=109445

Reviewed by Ryosuke Niwa.

Source/WebCore:

Tests: fast/parser/adoption-agency-crash-01.html

fast/parser/adoption-agency-crash-02.html
fast/parser/adoption-agency-crash-03.html

• html/parser/HTMLConstructionSite.cpp:

(WebCore::insert):
(WebCore::executeInsertTask):
(WebCore::executeReparentTask):
(WebCore::executeInsertAlreadyParsedChildTask):
(WebCore::executeTakeAllChildrenTask):
(WebCore::executeTask):
(WebCore::HTMLConstructionSite::attachLater):
(WebCore::HTMLConstructionSite::executeQueuedTasks):
(WebCore::HTMLConstructionSite::insertTextNode):
(WebCore::HTMLConstructionSite::reparent):
(WebCore::HTMLConstructionSite::insertAlreadyParsedChild):
(WebCore::HTMLConstructionSite::takeAllChildren):
(WebCore::HTMLConstructionSite::fosterParent):

• html/parser/HTMLConstructionSite.h:

(WebCore::HTMLConstructionSiteTask::HTMLConstructionSiteTask):
(WebCore::HTMLConstructionSiteTask::oldParent):

• html/parser/HTMLTreeBuilder.cpp:

(WebCore::HTMLTreeBuilder::callTheAdoptionAgency):

LayoutTests:

• TestExpectations:
• fast/parser/adoption-agency-crash-01-expected.txt: Added.
• fast/parser/adoption-agency-crash-01.html: Added.
• fast/parser/adoption-agency-crash-02-expected.txt: Added.
• fast/parser/adoption-agency-crash-02.html: Added.
• fast/parser/adoption-agency-crash-03-expected.txt: Added.
• fast/parser/adoption-agency-crash-03.html: Added.

1:46 AM Changeset in webkit [167220] by benjamin@webkit.org

• 7 edits in trunk/Source

[JSC] Improve the call site of string comparison in some hot path
​https://bugs.webkit.org/show_bug.cgi?id=131605

Reviewed by Darin Adler.

Source/JavaScriptCore:

When resolved, the String of a JSString is never null. It can be empty but not null.
The null value is reserved for ropes but those would be resolved when getting the value.

Consequently, we should use the equal() operation that do not handle null values.
Using the StringImpl directly is already common in StringPrototype but it was not used here for some reason.

• jit/JITOperations.cpp:
• runtime/JSCJSValueInlines.h:

(JSC::JSValue::equalSlowCaseInline):
(JSC::JSValue::strictEqualSlowCaseInline):
(JSC::JSValue::pureStrictEqual):

Source/WebCore:

• dom/NodeRareData.h:

(WebCore::NodeListsNodeData::NodeListCacheMapEntryHash::equal):
We should use the right comparison operation depending on the Hash Traits.

Source/WTF:

• wtf/text/StringImpl.cpp:

(WTF::stringImplContentEqual):
Inline that function to reduce the call overhead for JSC.
This is only inlined twice, it is not catastrophic for our binary.

1:45 AM Changeset in webkit [167219] by akling@apple.com

• 2 edits
  1 move
  1 delete in trunk/Source/WebCore

Merge MemoryPressureHandler{Mac,IOS}.mm
<​https://webkit.org/b/131603>

Join the iOS and Mac platform implementations of MemoryPressureHandler
under the shared Cocoa banner. Each platform still has its own quirky
behavior, but this puts them in the same file so we can start sharing.

Reviewed by Darin Adler.

• WebCore.xcodeproj/project.pbxproj:
• platform/cocoa/MemoryPressureHandlerCocoa.mm: Renamed from Source/WebCore/platform/mac/MemoryPressureHandlerMac.mm.

(WebCore::MemoryPressureHandler::platformReleaseMemory):
(WebCore::MemoryPressureHandler::install):
(WebCore::MemoryPressureHandler::uninstall):
(WebCore::MemoryPressureHandler::holdOff):
(WebCore::MemoryPressureHandler::respondToMemoryPressure):
(WebCore::respondToMemoryPressureCallback):
(WebCore::MemoryPressureHandler::installMemoryReleaseBlock):
(WebCore::MemoryPressureHandler::setReceivedMemoryPressure):
(WebCore::MemoryPressureHandler::hasReceivedMemoryPressure):
(WebCore::MemoryPressureHandler::clearMemoryPressure):
(WebCore::MemoryPressureHandler::shouldWaitForMemoryClearMessage):
(WebCore::MemoryPressureHandler::respondToMemoryPressureIfNeeded):

• platform/ios/MemoryPressureHandlerIOS.mm: Removed.

1:42 AM Changeset in webkit [167218] by benjamin@webkit.org

• 10 edits
  4 adds in trunk

CSS JIT: compile the :nth-child() pseudo class
​https://bugs.webkit.org/show_bug.cgi?id=131602

Reviewed by Andreas Kling.

Source/WebCore:

Tests: fast/selectors/nth-child-bounds.html

fast/selectors/nth-child-with-backtracking.html

Compile the :nth-child() pseudo class function + some related clean up.

• css/CSSSelector.cpp:

(WebCore::CSSSelector::nthA):
(WebCore::CSSSelector::nthB):
Expose the parsed value of an+b filters. Those values are used to compile
the selector.

(WebCore::CSSSelector::RareData::parseNth):
While working on the patch, I discovered some severe issues with the parsing of large
values of a and/or b. The problem comes from the way the CSS parser handle the values:
the values are parsed as a double then converted to an AtomicString for CSSSelector.

There are many problems related to large values but we never got bug reports because
they are very uncommon. Fixing those problem would require changing the parser.

Here, CSSSelector::RareData::parseNth() is hardened a little bit to avoid absurd values
of a and b.

• css/CSSSelector.h:
• cssjit/RegisterAllocator.h:

It looks like I forgot RDX in the list of register. Add it now since it is required
for SelectorCodeGenerator::modulo().

• cssjit/SelectorCompiler.cpp:

(WebCore::SelectorCompiler::addPseudoType):
(WebCore::SelectorCompiler::SelectorCodeGenerator::SelectorCodeGenerator):
(WebCore::SelectorCompiler::SelectorCodeGenerator::modulo):
(WebCore::SelectorCompiler::SelectorCodeGenerator::moduloIsZero):
There is no modulo() operation exposed on the macro assemblers. This is a basic
implementation on top of idiv for x86_64.

Since idiv works exclusively with RAX and RDX, most of the code is about getting
those registers efficiently.

(WebCore::SelectorCompiler::SelectorCodeGenerator::generateElementMatching):
(WebCore::SelectorCompiler::setElementChildIndex):
(WebCore::SelectorCompiler::setElementChildIndexAndUpdateStyle):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateElementIsNthChild):
This is pretty much a straightforward implementation of :nth-child().
The first part counts the number of previous elements.
The second part updates the tree if this is style resolution.
The last part compares the number of previous siblings to an+b to find if the filter matches.

The only part that diverges from SelectorChecker is how childIndex is used. Instead of testing it
at every iteration, only the first iteration handle the cache.

• dom/ElementRareData.h:

(WebCore::ElementRareData::childIndexMemoryOffset):

• dom/Node.h:

(WebCore::Node::rareDataMemoryOffset):
(WebCore::Node::flagHasRareData):

• rendering/style/RenderStyle.h:

LayoutTests:

Add a couple of test for the new code:
-nth-child-with-backtracking tests the register pressure with backtracking.
-nth-child-bounds tests invalid selectors do not cause problems.

• fast/selectors/nth-child-bounds-expected.txt: Added.
• fast/selectors/nth-child-bounds.html: Added.
• fast/selectors/nth-child-with-backtracking-expected.txt: Added.
• fast/selectors/nth-child-with-backtracking.html: Added.

• http/tests/security/video-poster-cross-origin-crash.html:

Now that CSSSelector filters out ridiculously bad values, the pseudo class in this test
was no longer executed.
The particular value of nth-child is irrelevant for this test, all it needs it the tree marking
while not matching.

1:29 AM Changeset in webkit [167217] by Carlos Garcia Campos

• 5 edits
  2 adds in releases/WebKitGTK/webkit-2.4

Merge r163465 - Move characterAt index checks from InlineIterator to RenderText
​https://bugs.webkit.org/show_bug.cgi?id=128224

Source/WebCore:

Reviewed by Simon Fraser.

Move characterAt index checks from InlineIterator to RenderText
so that all RenderText calls are covered. Few safe instances are
now covered with uncheckedCharacterAt.

Merged from Blink:
​http://src.chromium.org/viewvc/blink?view=revision&revision=150830

Test: fast/text/character-at-crash.html

• rendering/InlineIterator.h:

(WebCore::InlineIterator::characterAt):

• rendering/RenderText.cpp:

(WebCore::RenderText::computePreferredLogicalWidths):

• rendering/RenderText.h:

(WebCore::RenderText::operator[]):
(WebCore::RenderText::uncheckedCharacterAt):
(WebCore::RenderText::characterAt):

LayoutTests:

Move characterAt index checks from InlineIterator to RenderText
so that all RenderText calls are covered. Few safe instances are
now covered with uncheckedCharacterAt.

Merged from Blink:
​http://src.chromium.org/viewvc/blink?view=revision&revision=150830

Reviewed by Simon Fraser.

• fast/text/character-at-crash-expected.txt: Added.
• fast/text/character-at-crash.html: Added.

1:10 AM Changeset in webkit [167216] by timothy_horton@apple.com

• 8 edits in trunk/Source

Support setting a background color on page overlays
​https://bugs.webkit.org/show_bug.cgi?id=131600

Reviewed by Darin Adler.

• WebProcess/WebPage/PageOverlay.cpp:

(WebKit::PageOverlay::PageOverlay):
(WebKit::PageOverlay::bounds):
(WebKit::PageOverlay::startFadeInAnimation):
(WebKit::PageOverlay::startFadeOutAnimation):
(WebKit::PageOverlay::startFadeAnimation):
(WebKit::PageOverlay::fadeAnimationTimerFired):
Minor style adjustments.
Use more references everywhere.

(WebKit::PageOverlay::setBackgroundColor):
(WebKit::PageOverlay::setNeedsDisplay):

• WebProcess/WebPage/PageOverlay.h:

(WebKit::PageOverlay::backgroundColor):
Keep track of our background color, and push it down to the page overlay controller if we have one.

• WebProcess/WebPage/PageOverlayController.cpp:

(WebKit::updateOverlayGeometry):
(WebKit::PageOverlayController::clearPageOverlay):
Page overlays will always have the right size, and just use drawsContents to determine
whether or not they should tile/have backing store/etc.

(WebKit::PageOverlayController::installPageOverlay):
(WebKit::PageOverlayController::didChangeOverlayBackgroundColor):
Set the background color of the layer.

• WebProcess/WebPage/PageOverlayController.h:

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::WebPage):
References!

• rendering/RenderLayerCompositor.cpp:

(WebCore::RenderLayerCompositor::setDocumentOverlayRootLayer):
Mark the compositing tree as needing a rebuild when we get a new document-relative
overlay layer; otherwise we were depending on something else coming along and
requiring a rebuild, which didn't always happen.

12:50 AM Changeset in webkit [167215] by mihnea@adobe.com

• 4 edits
  2 adds in trunk

[CSS Regions] Hit testing doesn't work in video
​https://bugs.webkit.org/show_bug.cgi?id=131485

Reviewed by Andrei Bucur.

Source/WebCore:

When hit testing flow thread layer through the region layer,
we have to pass the depth sorting information and take that
into account for the situation in which an ancestor of the region
has preserve-3d transform style.

Test: fast/regions/hit-test-region-preserve3d-container.html

• rendering/RenderLayer.cpp:

(WebCore::RenderLayer::hitTestLayer):
(WebCore::RenderLayer::hitTestFlowThreadIfRegionForFragments):

• rendering/RenderLayer.h:

LayoutTests:

• fast/regions/hit-test-region-preserve3d-container-expected.txt: Added.
• fast/regions/hit-test-region-preserve3d-container.html: Added.

12:28 AM WebKitGTK/2.4.x edited by Carlos Garcia Campos

(diff)

12:28 AM Changeset in webkit [167214] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.4/Source/WebKit2

Merge r167128 - wk2-gtk does not display anything
​https://bugs.webkit.org/show_bug.cgi?id=125558

Reviewed by Martin Robinson.

Remove fcntl call to set access mode flags on the duplicated files
descriptor. Those flags are ignored in Linux and make fcntl to
fail in FreeBSD. We should handle the case where the passed
protection is ReadOnly.
Thanks to Raphael Kubo da Costa who proposed the solution.

• Platform/unix/SharedMemoryUnix.cpp:

(WebKit::SharedMemory::createHandle):
(WebKit::accessModeFile): Deleted.

12:19 AM Changeset in webkit [167213] by commit-queue@webkit.org

• 51 edits
  9 adds in trunk/LayoutTests

Update html5lib test data to latest version
​https://bugs.webkit.org/show_bug.cgi?id=131588

Patch by Koji Ishii <​kojishi@gmail.com> on 2014-04-14
Reviewed by Darin Adler.

This patch imports the latest version of the "dat" files from
​https://github.com/html5lib/html5lib-tests
Also updated expected files for failures of new tests.

• html5lib/generated/run-adoption01-data-expected.txt: Passes now as a bug in the test was fixed.
• html5lib/generated/run-adoption01-write-expected.txt: Supressed new failures.
• html5lib/generated/run-inbody01-data-expected.txt: Added.
• html5lib/generated/run-inbody01-data.html: Added.
• html5lib/generated/run-inbody01-write-expected.txt: Added.
• html5lib/generated/run-inbody01-write.html: Added.
• html5lib/generated/run-main-element-data-expected.txt: Added.
• html5lib/generated/run-main-element-data.html: Added.
• html5lib/generated/run-main-element-write-expected.txt: Added.
• html5lib/generated/run-main-element-write.html: Added.
• html5lib/generated/run-template-data-expected.txt: Supressed new failures.
• html5lib/generated/run-template-write-expected.txt: Supressed new failures.
• html5lib/generated/run-tests21-data-expected.txt: Supressed new failures.
• html5lib/generated/run-tests21-write-expected.txt: Supressed new failures.
• html5lib/generated/run-tests25-data-expected.txt: Supressed new failures.
• html5lib/generated/run-tests25-write-expected.txt: Supressed new failures.
• html5lib/resources/adoption01.dat:
• html5lib/resources/adoption02.dat:
• html5lib/resources/comments01.dat:
• html5lib/resources/doctype01.dat:
• html5lib/resources/domjs-unsafe.dat:
• html5lib/resources/entities01.dat:
• html5lib/resources/entities02.dat:
• html5lib/resources/html5test-com.dat:
• html5lib/resources/inbody01.dat:
• html5lib/resources/isindex.dat:
• html5lib/resources/main-element.dat: Added.
• html5lib/resources/pending-spec-changes-plain-text-unsafe.dat:
• html5lib/resources/pending-spec-changes.dat:
• html5lib/resources/plain-text-unsafe.dat:
• html5lib/resources/scriptdata01.dat:
• html5lib/resources/tables01.dat:
• html5lib/resources/template.dat:
• html5lib/resources/tests1.dat:
• html5lib/resources/tests10.dat:
• html5lib/resources/tests14.dat:
• html5lib/resources/tests15.dat:
• html5lib/resources/tests16.dat:
• html5lib/resources/tests17.dat:
• html5lib/resources/tests18.dat:
• html5lib/resources/tests19.dat:
• html5lib/resources/tests2.dat:
• html5lib/resources/tests20.dat:
• html5lib/resources/tests21.dat:
• html5lib/resources/tests22.dat:
• html5lib/resources/tests23.dat:
• html5lib/resources/tests25.dat:
• html5lib/resources/tests26.dat:
• html5lib/resources/tests3.dat:
• html5lib/resources/tests4.dat:
• html5lib/resources/tests5.dat:
• html5lib/resources/tests6.dat:
• html5lib/resources/tests7.dat:
• html5lib/resources/tests8.dat:
• html5lib/resources/tests9.dat:
• html5lib/resources/tests_innerHTML_1.dat:
• html5lib/resources/tricky01.dat:
• html5lib/resources/webkit01.dat:
• html5lib/resources/webkit02.dat:

12:17 AM Changeset in webkit [167212] by Carlos Garcia Campos

• 8 edits in releases/WebKitGTK/webkit-2.4

Merge r167185 - [SOUP] Libsoup internal credential setting should be controlled by loader decision
​https://bugs.webkit.org/show_bug.cgi?id=130963

Patch by Youenn Fablet <​youenn.fablet@crf.canon.fr> on 2014-04-12
Reviewed by Darin Adler.

Source/WebCore:

Disabled libsoup internal authentication manager for messages for which no credential is available and no stored credentials should be used.
Updated synchronous loader to return whether using credentials or not according StoredCredential loader option parameter.
Unskipped test http/tests/xmlhttprequest/cross-origin-no-authorization.html covers the patch.

• platform/network/ResourceHandleInternal.h:

(WebCore::ResourceHandleInternal::ResourceHandleInternal): Added m_useAuthenticationManager boolean to control whether disable authentication manager or not.

• platform/network/soup/ResourceHandleSoup.cpp:

(WebCore::WebCoreSynchronousLoader::WebCoreSynchronousLoader): Added m_storedCredentials member.
(WebCore::WebCoreSynchronousLoader::shouldUseCredentialStorage): Return true if stored credentials are allowed.
(WebCore::applyAuthenticationToRequest): Set m_useAuthenticationManager value to disable authentication manager if cannot use stored credentials and ResourceHandleInternal has no username and password.
(WebCore::createSoupMessageForHandleAndRequest): Disable authentication mananger according m_useAuthenticationManager value.
(WebCore::ResourceHandle::platformLoadResourceSynchronously): Added StoredCredentials loader option to the sync loader constructor.

Source/WebKit/efl:

• WebCoreSupport/FrameLoaderClientEfl.cpp:

(WebCore::FrameLoaderClientEfl::shouldUseCredentialStorage): Similarly to GTK, let soup/loader layer handle when to use credential storage. Return always true

LayoutTests:

• platform/efl/TestExpectations: Unskipped http/tests/xmlhttprequest/cross-origin-no-authorization.html.
• platform/gtk/TestExpectations: Ditto.

Apr 13, 2014:

11:58 PM Changeset in webkit [167211] by Darin Adler

• 6 edits
  1 add in trunk

userVisibleString should not try to "encode" host names
​https://bugs.webkit.org/show_bug.cgi?id=131587
rdar://problem/14686849

Reviewed by Alexey Proskuryakov.

Source/WebCore:

• WebCore.exp.in: Updated for WebCoreNSURLExtras argument type changes.

• platform/mac/WebCoreNSURLExtras.h: Removed unneeded code to make this

Objective-C++ header compile in plain C++ files, which we never need to do.
Added missing argument name, baseURL, and changed mysterious CFIndex arguments
to the correct type, CFURLComponentType.

• platform/mac/WebCoreNSURLExtras.mm:

(WebCore::isLookalikeCharacter): Removed the inline keyword from this, and added
more lookalike characters from the Mozilla list referenced here.
(WebCore::URLByTruncatingOneCharacterBeforeComponent): Updated argument type.
(WebCore::dataForURLComponentType): Ditto.
(WebCore::userVisibleString): Only call mapHostNames if host name decoding is
needed; no encoding here.

Tools:

• TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added URLExtras.mm.

• TestWebKitAPI/Tests/mac/URLExtras.mm: Added. A few tests for WebCoreNSURLExtras.

Coverage is tiny at this point; we could add a lot more cases!

11:55 PM Changeset in webkit [167210] by Darin Adler

• 14 edits
  2 adds in trunk

REGRESSION (r158617): Find on Page can get stuck in a loop when the search string occurs in an <input> in a <fieldset>
​https://bugs.webkit.org/show_bug.cgi?id=126322

Reviewed by Ryosuke Niwa.

Source/WebCore:

• dom/Element.cpp:

(WebCore::Element::canContainRangeEndPoint): Now returns false when the role of the element
is "img". This is the same rule that's hard-coded in isRenderReplacedElement for the same
reason. Need more test coverage to make sure this role feature works consistently.

• dom/Element.h: Made canContainRangeEndPoint no longer inline since it's not just a

return statement any more.

• dom/Position.cpp:

(WebCore::Position::isCandidate): Took out code that calls isRendererReplacedElement
that was added in r158617; not needed now that we updated canContainRangeEndPoint.

• dom/Range.cpp:

(WebCore::Range::firstNode): Removed code here that called isRendererReplacedElement.
This was the wrong level to be adding editing logic, and there's a FIXME here to that
effect, which we are now deleting. This was the change that broke Find.

• editing/TextIterator.cpp: Added a comment about the redundancy between the

isRendererReplacedElement and editingIgnoresContent functions.

• html/HTMLHRElement.cpp:

(WebCore::HTMLHRElement::canContainRangeEndPoint): Call through to base class instead
of just returning true when we have child nodes. Lets Element::canContainRangeEndPoint
do its thing.

• html/HTMLHRElement.h: Ditto.

• html/HTMLObjectElement.cpp:

(WebCore::HTMLObjectElement::canContainRangeEndPoint): Call through to base class instead
of just returning true when we have fallback content. Lets Element::canContainRangeEndPoint
do its thing.

• html/HTMLObjectElement.h: Ditto.

• testing/Internals.cpp:

(WebCore::Internals::countMatchesForText): Set the limit to 1000 instead of infinite.

LayoutTests:

• editing/text-iterator/count-matches-in-form-expected.txt: Added.
• editing/text-iterator/count-matches-in-form.html: Added.

• fast/text/window-find.html: Tweaked the test a bit, making it a little easier to

see if the test hasn't even run.

11:55 PM WebKitGTK/2.4.x edited by Philippe Normand

(diff)

10:35 PM April 2014 Meeting edited by rniwa@webkit.org

Add my intentions to attend talks/hackathons (diff)

10:29 PM Changeset in webkit [167209] by ljaehun.lim@samsung.com

• 2 edits in trunk/Tools

Use #include instead of #import
​https://bugs.webkit.org/show_bug.cgi?id=131604

Reviewed by Darin Adler.

Fix "warning: #import is a deprecated GCC extension [-Wdeprecated]".

• TestWebKitAPI/Tests/WTF/StringView.cpp:

9:28 PM Changeset in webkit [167208] by Darin Adler

• 6 edits in trunk/Source/WebCore

Use unique_ptr for FillLayer::m_next
​https://bugs.webkit.org/show_bug.cgi?id=75222

Reviewed by Dan Bernstein.

• css/DeprecatedStyleBuilder.cpp:

(WebCore::ApplyPropertyFillLayer::applyInheritValue):
Renamed currChild to just child and prevChild to previousChild.
Changed code to pass ownership of the new FillLayer immediately.
Changed some loops to be for loops.
(WebCore::ApplyPropertyFillLayer::applyInitialValue): Ditto.
(WebCore::ApplyPropertyFillLayer::applyValue): Ditto.

• rendering/RenderBox.cpp:

(WebCore::RenderBox::backgroundHasOpaqueTopLayer): Use reference
instead of pointer.
(WebCore::RenderBox::paintFillLayers): Ditto.

• rendering/RenderBoxModelObject.cpp:

(WebCore::RenderBoxModelObject::paintFillLayerExtended): Ditto.

• rendering/style/FillLayer.cpp:

(WebCore::FillLayer::FillLayer): Removed m_next initializer since it is now an
OwnPtr and initializes automatically. In a couple other places, changed m_next
initializer to use make_unique.
(WebCore::FillLayer::~FillLayer): Wrote loop for deletion of m_next.
(WebCore::FillLayer::operator=): Removed unneeded explicit deletion of m_next.
(WebCore::FillLayer::cullEmptyLayers): Ditto.
(WebCore::clipMax): Marked inline.
(WebCore::FillLayer::computeClipMax): Rewrote to use a loop instead of recursion.
(WebCore::FillLayer::containsImage): Ditto.
(WebCore::FillLayer::imagesAreLoaded): Ditto.
(WebCore::FillLayer::hasOpaqueImage): Rewrote to use && instead of multiple if.
(WebCore::FillLayer::hasImage): Rewrote to use a loop instead of recursion.
(WebCore::FillLayer::hasFixedImage): Ditto.

• rendering/style/FillLayer.h: Changed m_next to be a unique_ptr.

7:57 PM Changeset in webkit [167207] by aestes@apple.com

• 15 edits
  1 add in trunk/Source

[QuickLook] Move file system-related code into WebKit
​https://bugs.webkit.org/show_bug.cgi?id=131597

Reviewed by Dan Bernstein.

Source/WebCore:

QuickLookHandle should not be responsible for saving a copy of the
original document to disk as it might be running in a process that
either can't write to disk or can only write into a sandboxed
container. To account for this, we need to separate the concern of
quick look conversion from that of original document saving so that
each activity can run in the appropriate process.

Created a new interface between WebCore and WebKit by adding a client
(QuickLookHandleClient) to QuickLookHandle which is notified of incoming
bytes. A new function on FrameLoaderClient tells WebKit when a new
QuickLookHandle is created, giving WebKit the opportunity to register a
handle client. Moved the existing file system-related code as well as
code only needed by WebKit1 in QuickLookHandle into a new WebKit1
QuickLookHandleClient subclass.

• WebCore.exp.in:
• WebCore.xcodeproj/project.pbxproj: Made QuickLookHandleClient.h Private.
• loader/FrameLoaderClient.h:

(WebCore::FrameLoaderClient::didCreateQuickLookHandle): Added.

• loader/ResourceLoader.cpp:

(WebCore::ResourceLoader::didCreateQuickLookHandle): Called FrameLoaderClient::didCreateQuickLookHandle().

• loader/ResourceLoader.h:
• platform/network/ResourceHandle.h: Made m_quickLook a unique_ptr.

(WebCore::ResourceHandle::setQuickLookHandle): Changed to take a unique_ptr.

• platform/network/ResourceHandleClient.h:

(WebCore::ResourceHandleClient::didCreateQuickLookHandle): Added.

• platform/network/ios/QuickLook.h: Added m_client, gave m_converter a stronger type, and made m_nsResponse a RetainPtr.

(WebCore::QuickLookHandle::setClient): Added.
(WebCore::QuickLookHandle::firstRequestURL): Added.
(WebCore::QuickLookHandle::converter): Added.

• platform/network/ios/QuickLook.mm:

(WebCore::registerQLPreviewConverterIfNeeded):
(WebCore::createTemporaryFileForQuickLook): Made non-static.
(WebCore::emptyClient): Returned a shared empty QuickLookHandleClient.
(WebCore::QuickLookHandle::QuickLookHandle): Removed file system and WebKit1-only code.
(WebCore::QuickLookHandle::create): Changed to return a unique_ptr.
(WebCore::QuickLookHandle::nsResponse):
(WebCore::QuickLookHandle::didReceiveDataArray): Removed file system code and called QuickLookHandleClient::didReceiveDataArray() instead.
(WebCore::QuickLookHandle::didReceiveData): Removed file system code and called QuickLookHandleClient::didReceiveData() instead.
(WebCore::QuickLookHandle::didFinishLoading): Removed file system code and called QuickLookHandleClient::didFinishLoading() instead.
(WebCore::QuickLookHandle::didFail): Removed file system and WebKit1-only code, calling QuickLookHandleClient::didFail() instead.
(WebCore::QuickLookHandle::~QuickLookHandle): Removed file system and WebKit1-only code. Cleared our reference to m_client.
(WebCore::QuickLookHandle::previewFileName): Retrieved from m_converter.
(WebCore::QuickLookHandle::previewRequestURL): Ditto.

• platform/network/ios/QuickLookHandleClient.h: Added.

(WebCore::QuickLookHandleClient::~QuickLookHandleClient):
(WebCore::QuickLookHandleClient::didReceiveDataArray):
(WebCore::QuickLookHandleClient::didReceiveData):
(WebCore::QuickLookHandleClient::didFinishLoading):
(WebCore::QuickLookHandleClient::didFail):

Source/WebKit/mac:

Moved file system and WebKit1-only code from QuickLookHandle into a new
QuickLookHandleClient subclass.

• WebCoreSupport/WebFrameLoaderClient.h:
• WebCoreSupport/WebFrameLoaderClient.mm:

Source/WebKit2:

• WebProcess/Network/WebResourceLoader.h: Made m_quickLookHandle a unique_ptr.
• WebProcess/ios/WebResourceLoaderIOS.mm:

(WebKit::WebResourceLoader::setUpQuickLookHandleIfNeeded):

6:35 PM Changeset in webkit [167206] by aestes@apple.com

• 2 edits in trunk/Source/WTF

Relax adoption requirements of RefCounted objects that are NeverDestroyed
​https://bugs.webkit.org/show_bug.cgi?id=131593

Reviewed by Dan Bernstein.

RefCounted objects that are created by NeverDestroyed<> won't have a
RefPtr adopting them, so call relaxAdoptionRequirements().

• wtf/NeverDestroyed.h:

(WTF::NeverDestroyed::NeverDestroyed):
(WTF::NeverDestroyed::MaybeRelax::MaybeRelax):

5:02 PM Changeset in webkit [167205] by ggaren@apple.com

• 3 edits
  12 adds in trunk/PerformanceTests

Added some website recordings to MallocBench -- taken from Membuster
​https://bugs.webkit.org/show_bug.cgi?id=131601

Reviewed by Ryosuke Niwa.

Added flickr, reddit, and theverge -- each recorded from Membuster's
cache, with and without sending Safari a low memory warning.

• MallocBench/MallocBench.xcodeproj/project.pbxproj:
• MallocBench/MallocBench/Benchmark.cpp:
• MallocBench/MallocBench/flickr.cpp: Added.

(benchmark_flickr):
(benchmark_flickr_memory_warning):

• MallocBench/MallocBench/flickr.h: Added.
• MallocBench/MallocBench/flickr.ops: Added.
• MallocBench/MallocBench/flickr_memory_warning.ops: Added.
• MallocBench/MallocBench/reddit.cpp: Added.

(benchmark_reddit):
(benchmark_reddit_memory_warning):

• MallocBench/MallocBench/reddit.h: Added.
• MallocBench/MallocBench/reddit.ops: Added.
• MallocBench/MallocBench/reddit_memory_warning.ops: Added.
• MallocBench/MallocBench/theverge.cpp: Added.

(benchmark_theverge):
(benchmark_theverge_memory_warning):

• MallocBench/MallocBench/theverge.h: Added.
• MallocBench/MallocBench/theverge.ops: Added.
• MallocBench/MallocBench/theverge_memory_warning.ops: Added.

4:28 PM Changeset in webkit [167204] by ggaren@apple.com

• 3 edits in trunk/Source/bmalloc

Fixed some mbmalloc exports
​https://bugs.webkit.org/show_bug.cgi?id=131599

Reviewed by Ryosuke Niwa.

• bmalloc.xcodeproj/project.pbxproj: Made some headers a private part

of the project, so we can call them from API.

• bmalloc/mbmalloc.cpp: Marked the mbmalloc functions with default

visibility, so they show up as exported in the .dylib.

4:21 PM Changeset in webkit [167203] by ggaren@apple.com

• 4 edits in trunk/PerformanceTests

MallocBench record/replay should support realloc
​https://bugs.webkit.org/show_bug.cgi?id=131598

Reviewed by Ryosuke Niwa.

• MallocBench/MallocBench.xcodeproj/project.pbxproj: Fixed some linkage

issues that caused us not to fully link to system malloc in the default
case. Also marked mbmalloc.dylib as required so the error message will
be clearer if we mess up.

• MallocBench/MallocBench/Interpreter.cpp:

(Interpreter::run):

• MallocBench/MallocBench/Interpreter.h: Added the realloc case, and

upgraded one-letter names to full words.

4:03 PM April 2014 Meeting edited by rniwa@webkit.org

Add a hackathon to import more w3c tests (diff)

1:02 PM Changeset in webkit [167202] by aestes@apple.com

• 2 edits in trunk/Source/WebKit2

REGRESSION (r167164): -[WKNSURLSessionLocal _getCookieHeadersForTask:completionHandler:] _block_invoke can access a deallocated completionHandler
​https://bugs.webkit.org/show_bug.cgi?id=131595

Reviewed by Dan Bernstein.

Make a copy of completionHandler and release it after we're done with it.

• Shared/mac/CookieStorageShim.mm:

(-[WKNSURLSessionLocal _getCookieHeadersForTask:completionHandler:]):

12:23 PM Changeset in webkit [167201] by psolanki@apple.com

• 5 edits in trunk/Source/WebKit2

Don't use ImportanceAssertion on iOS
​https://bugs.webkit.org/show_bug.cgi?id=131481
<rdar://problem/16575830>

Reviewed by Darin Adler.

We have other API to mark processes as being in use on iOS. No need to use ImportanceAssertion.

• Platform/IPC/MessageDecoder.cpp:
• Platform/IPC/MessageDecoder.h:
• Platform/IPC/mac/ConnectionMac.cpp:

(IPC::Connection::receiveSourceEventHandler):

• Platform/IPC/mac/ImportanceAssertion.h:

12:20 PM Changeset in webkit [167200] by psolanki@apple.com

• 3 edits in trunk/Source/WebCore

Move early return out of dispatch_async() block so we can return from willSendRequest quickly
​https://bugs.webkit.org/show_bug.cgi?id=131478
<rdar://problem/16575535>

Reviewed by Alexey Proskuryakov.

Do a quick check to see if we need to synthesize the redirect response on the dispatch queue
and return from willSendRequest callback quickly instead of always doing an effectively synchronous
call to the main thread. We can't call synthesizeRedirectResponseIfNecessary on the dispatch
queue since that accesses the ResourceRequest.

No new tests because no change in functionality.

• platform/network/cf/ResourceHandleCFURLConnectionDelegate.h:
• platform/network/cf/ResourceHandleCFURLConnectionDelegateWithOperationQueue.cpp:

(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::setupRequest): Save the
request scheme to use later for early return from willSendRequest.
(WebCore::ResourceHandleCFURLConnectionDelegateWithOperationQueue::willSendRequest):

11:01 AM Changeset in webkit [167199] by oliver@apple.com

• 70 edits
  3 adds
  2 deletes in trunk

Rewrite Function.bind as a builtin
​https://bugs.webkit.org/show_bug.cgi?id=131083

Reviewed by Geoffrey Garen.

Source/JavaScriptCore:

This change removes the existing function.bind implementation
entirely so JSBoundFunction is no more.

Instead we just return a regular JS closure with a few
private properties hanging off it that allow us to perform
the necessary bound function fakery. While most of this is
simple, a couple of key changes:

• The parser and lexer now directly track whether they're parsing code for call or construct and convert the private name @IsConstructor into TRUETOK or FALSETOK as appropriate. This automatically gives us the ability to vary behaviour from within the builtin. It also leaves a lot of headroom for trivial future improvements.
• The instanceof operator now uses the prototypeForHasInstance private name, and we have a helper function to ensure that all objects that need to can update their magical 'prototype' property pair correctly.

• API/JSScriptRef.cpp:

(parseScript):

• JavaScriptCore.xcodeproj/project.pbxproj:
• builtins/BuiltinExecutables.cpp:

(JSC::BuiltinExecutables::createBuiltinExecutable):

• builtins/Function.prototype.js:

(bind.bindingFunction):
(bind.else.bindingFunction):
(bind):

• bytecode/UnlinkedCodeBlock.cpp:

(JSC::generateFunctionCodeBlock):

• bytecompiler/NodesCodegen.cpp:

(JSC::InstanceOfNode::emitBytecode):

• interpreter/Interpreter.cpp:
• parser/Lexer.cpp:

(JSC::Lexer<T>::Lexer):
(JSC::Lexer<LChar>::parseIdentifier):
(JSC::Lexer<UChar>::parseIdentifier):

• parser/Lexer.h:
• parser/Parser.cpp:

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

• parser/Parser.h:

(JSC::parse):

• parser/ParserModes.h:
• runtime/CodeCache.cpp:

(JSC::CodeCache::getGlobalCodeBlock):
(JSC::CodeCache::getFunctionExecutableFromGlobalCode):

• runtime/CommonIdentifiers.h:
• runtime/Completion.cpp:

(JSC::checkSyntax):

• runtime/Executable.cpp:

(JSC::ProgramExecutable::checkSyntax):

• runtime/FunctionPrototype.cpp:

(JSC::FunctionPrototype::addFunctionProperties):
(JSC::functionProtoFuncBind): Deleted.

• runtime/JSBoundFunction.cpp: Removed.
• runtime/JSBoundFunction.h: Removed.
• runtime/JSFunction.cpp:

(JSC::RetrieveCallerFunctionFunctor::RetrieveCallerFunctionFunctor):
(JSC::RetrieveCallerFunctionFunctor::operator()):
(JSC::retrieveCallerFunction):
(JSC::JSFunction::getOwnPropertySlot):
(JSC::JSFunction::defineOwnProperty):

• runtime/JSGlobalObject.cpp:

(JSC::JSGlobalObject::reset):

• runtime/JSGlobalObjectFunctions.cpp:

(JSC::globalFuncSetTypeErrorAccessor):

• runtime/JSGlobalObjectFunctions.h:
• runtime/JSObject.h:

(JSC::JSObject::inlineGetOwnPropertySlot):

Source/WebCore:

Switch WebCore to use the helper functions when defining the
prototype properties on DOM constructors, and update bindings
tests accordingly.

• bindings/js/JSImageConstructor.cpp:

(WebCore::JSImageConstructor::finishCreation):

• bindings/scripts/CodeGeneratorJS.pm:

(GenerateConstructorHelperMethods):

• bindings/scripts/test/JS/JSTestActiveDOMObject.cpp:

(WebCore::JSTestActiveDOMObjectConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestCustomNamedGetter.cpp:

(WebCore::JSTestCustomNamedGetterConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestEventConstructor.cpp:

(WebCore::JSTestEventConstructorConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestEventTarget.cpp:

(WebCore::JSTestEventTargetConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestException.cpp:

(WebCore::JSTestExceptionConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestGenerateIsReachable.cpp:

(WebCore::JSTestGenerateIsReachableConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestInterface.cpp:

(WebCore::JSTestInterfaceConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:

(WebCore::JSTestMediaQueryListListenerConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestNamedConstructor.cpp:

(WebCore::JSTestNamedConstructorConstructor::finishCreation):
(WebCore::JSTestNamedConstructorNamedConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestNode.cpp:

(WebCore::JSTestNodeConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestObj.cpp:

(WebCore::JSTestObjConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestOverloadedConstructors.cpp:

(WebCore::JSTestOverloadedConstructorsConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:

(WebCore::JSTestSerializedScriptValueInterfaceConstructor::finishCreation):

• bindings/scripts/test/JS/JSTestTypedefs.cpp:

(WebCore::JSTestTypedefsConstructor::finishCreation):

• bindings/scripts/test/JS/JSattribute.cpp:

(WebCore::JSattributeConstructor::finishCreation):

• bindings/scripts/test/JS/JSreadonly.cpp:

(WebCore::JSreadonlyConstructor::finishCreation):

LayoutTests:

Testing.

• js/dom/function-bind-expected.txt:
• js/regress/function-bind-expected.txt: Added.
• js/regress/function-bind.html: Added.
• js/regress/script-tests/function-bind.js: Added.

(foo):

9:41 AM Changeset in webkit [167198] by Simon Fraser

• 2 edits in trunk/Source/WebCore

[iOS WK2] Hook up scrolling tree nodes when coming out of the page cache
​https://bugs.webkit.org/show_bug.cgi?id=131577

Reviewed by Tim Horton.

The call to scrollingCoordinator->frameViewRootLayerDidChange() was inside
a #if !PLATFORM(IOS) block, but now that we use the ScrollingCoordinator
for WK2 we want to call this.

• loader/HistoryController.cpp:

(WebCore::HistoryController::restoreScrollPositionAndViewState):

7:59 AM Changeset in webkit [167197] by zandobersek@gmail.com

• 2 edits in trunk/Source/WebCore

Unreviewed build fix after r167196.

• platform/RemoteCommandListener.cpp:

(WebCore::RemoteCommandListener::create): Fall back to using the new operator
for allocating RemoteCommandListener object. Using std::make_unique() requires
for the operator to be public, which doesn't work well with the static create()
method.

7:18 AM Changeset in webkit [167196] by zandobersek@gmail.com

• 10 edits in trunk/Source/WebCore

Remove unnecessary uses of std::move() in return statements
​https://bugs.webkit.org/show_bug.cgi?id=131457

Reviewed by Darin Adler.

Don't use std::move() in return statements unless necessary as it inhibits
named return value optimizations as performed by compilers.

• Modules/battery/BatteryManager.cpp:

(WebCore::BatteryManager::create):

• html/FormController.cpp:

(WebCore::FormController::createSavedFormStateMap):

• html/canvas/WebGLRenderingContext.cpp:

(WebCore::WebGLRenderingContext::create):

• platform/RemoteCommandListener.cpp:

(WebCore::RemoteCommandListener::create):

• platform/graphics/ca/GraphicsLayerCA.cpp:

(WebCore::GraphicsLayer::create):

• platform/ios/RemoteCommandListenerIOS.mm:

(WebCore::RemoteCommandListener::create):

• rendering/RenderGrid.cpp:

(WebCore::RenderGrid::GridIterator::nextEmptyGridArea):

• rendering/RenderLayer.cpp:

(WebCore::RenderLayer::setupFilters):

• rendering/style/CounterDirectives.cpp:

(WebCore::clone):

4:23 AM Changeset in webkit [167195] by commit-queue@webkit.org

• 11 edits in trunk/Source

Unreviewed, rolling out r167168 and r167194.
​https://bugs.webkit.org/show_bug.cgi?id=131589

Caused massive ASSERTION failures on the GTK Debug bot
(Requested by philn on #webkit).

Reverted changesets:

"[GTK] Add HighDPI support for non-accelerated compositing
contents"
​https://bugs.webkit.org/show_bug.cgi?id=131562
​http://trac.webkit.org/changeset/167168

Source/WebCore:

"Unreviewed. Fix GTK+ build with recent cairo and GTK+ after
r167168."
​http://trac.webkit.org/changeset/167194

Source/WebKit2:

"Unreviewed. Fix GTK+ build with recent cairo and GTK+ after
r167168."
​http://trac.webkit.org/changeset/167194

2:46 AM Changeset in webkit [167194] by Carlos Garcia Campos

• 2 edits in trunk/Source/WebKit2

Unreviewed. Fix GTK+ build with recent cairo and GTK+ after r167168.

• UIProcess/API/gtk/WebKitWebViewBase.cpp:

(webkitWebViewBaseCreateWebPage):

2:33 AM Changeset in webkit [167193] by commit-queue@webkit.org

• 9 edits
  4 adds in trunk

[GStreamer] No CORS support for media elements
​https://bugs.webkit.org/show_bug.cgi?id=99037

Patch by Youenn Fablet <​youenn.fablet@crf.canon.fr> on 2014-04-13
Reviewed by Philippe Normand.

Source/WebCore:

Added CORS access control check to media sources when crossorigin attribute is set.

Added getter to CORS access control check status (used to compute whether the stream is tainted or not).
Related test is http/tests/security/video-cross-origin-readback.html.

Disabled access to cross-origin streams that fail CORS check when crossorigin attribute is set.
Related test is http/tests/security/video-cross-origin-accessfailure.html.

Tests: http/tests/security/video-cross-origin-accessfailure.html

http/tests/security/video-cross-origin-accesssameorigin.html

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:

(WebCore::MediaPlayerPrivateGStreamer::didPassCORSAccessCheck): Return whether media is cross-origin (tainted) or not by querying the gstreamer source layer.

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h: Added MediaPlayerPrivateGStreamer::didPassCORSAccessCheck declaration.
• platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:

(webKitWebSrcStart): Passed CORS mode parameter to the streaming client. In case of CORS check failure, stop the resource loading.
(webKitSrcPassedCORSAccessCheck): Return whether CORS access control check was done and successful.
(StreamingClient::handleResponseReceived): Take a parameter to assign the CORS access control check result.
(CachedResourceStreamingClient::CachedResourceStreamingClient): Updated setting of the ResourceLoaderOptions according CORS mode.
(CachedResourceStreamingClient::responseReceived): Check CORS and pass result to handleResponseReceived.
(ResourceHandleStreamingClient::didReceiveResponse): No CORS check.

• platform/graphics/gstreamer/WebKitWebSourceGStreamer.h: Added webKitSrcPassedCORSAccessCheck declaration.

LayoutTests:

http/tests/security/video-cross-origin-accessfailure.html verifies that cross-origin streams that fail CORS check
are not played when crossorigin attribute is set.

http/tests/security/video-cross-origin-accesssameorigin.html verifies that access to same-origin streams

are played when crossorigin attribute is set.

• http/tests/security/video-cross-origin-accessfailure-expected.txt: Added.
• http/tests/security/video-cross-origin-accessfailure.html: Added.
• http/tests/security/video-cross-origin-accesssameorigin-expected.txt: Added.
• http/tests/security/video-cross-origin-accesssameorigin.html: Added.
• platform/efl/TestExpectations: Enabled http/tests/security/video-cross-origin-readback.html.
• platform/gtk/TestExpectations: Ditto.
• platform/mac/TestExpectations: Disabled http/tests/security/video-cross-origin-accessfailure.html.

1:05 AM Changeset in webkit [167192] by Darin Adler

• 5 edits in trunk

REGRESSION (r166860): ASSERTION FAILED: !isCalculated() on fast/css/image-set-value-not-removed-crash.html
​https://bugs.webkit.org/show_bug.cgi?id=131480

Reviewed by Andreas Kling.

Source/WebCore:

Fixes intermittent assertion failure in fast/css/image-set-value-not-removed-crash.html.

• css/CSSComputedStyleDeclaration.cpp:

(WebCore::valueForImageSliceSide): Added. Helper used below in valueForNinePieceImageSlice.
Handles calculated values by returning 0; incorrect but predictable.
(WebCore::valueForNinePieceImageSlice): Updated to call valueForImageSliceSide.
(WebCore::positionOffsetValue): Use nullptr.
(WebCore::ComputedStyleExtractor::propertyValue): Updated to call positionOffsetValue
by its new name. Removed "get" from the name.
(WebCore::positionOffsetValue): Renamed from getPositionOffsetValue.

• platform/Length.h: Made isCalculated public.

LayoutTests:

• platform/mac/TestExpectations: Unskip the test now that the assertion is fixed.