WebKit-https.git
5 years ago[EFL] Unreviewed gardening on 29th May
gyuyoung.kim@webkit.org [Fri, 29 May 2015 06:17:10 +0000 (06:17 +0000)]
[EFL] Unreviewed gardening on 29th May

Unskip passing tests for viewport-percentage, text, sub-pixel
and update new failing accessibility, fast/overflow and so on.

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184981 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUpdate results of WebKit1.StringTruncator after r184965. I missed one letter.
ap@apple.com [Fri, 29 May 2015 04:56:13 +0000 (04:56 +0000)]
Update results of WebKit1.StringTruncator after r184965. I missed one letter.

* TestWebKitAPI/Tests/mac/StringTruncator.mm: (TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184980 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoReplaced 0 with nullptr in WebCore/accessibility.
commit-queue@webkit.org [Fri, 29 May 2015 04:53:35 +0000 (04:53 +0000)]
Replaced 0 with nullptr in WebCore/accessibility.
https://bugs.webkit.org/show_bug.cgi?id=145309

Patch by Hunseop Jeong <hs85.jeong@samsung.com> on 2015-05-28
Reviewed by Chris Fleizach.

No new tests, no behavior change.

* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::mathPrescripts):
(WebCore::AccessibilityRenderObject::mathPostscripts):
* accessibility/AccessibilityScrollView.cpp:
(WebCore::AccessibilityScrollView::scrollBar):
* accessibility/atk/AXObjectCacheAtk.cpp:
(WebCore::notifyChildrenSelectionChange):
* accessibility/atk/WebKitAccessibleHyperlink.cpp:
(webkitAccessibleHyperlinkInit):
* accessibility/atk/WebKitAccessibleInterfaceDocument.cpp:
(webkitAccessibleDocumentGetAttributes):
* accessibility/atk/WebKitAccessibleInterfaceText.cpp:
(getAttributeSetForAccessibilityObject):
(attributeSetDifference):
* accessibility/atk/WebKitAccessibleUtil.h:
* accessibility/atk/WebKitAccessibleWrapperAtk.cpp:
(webkitAccessibleRefChild):
(webkitAccessibleGetAttributes):
(webkitAccessibleGetObjectLocale):
(getInterfaceMaskFromObject):
(objectFocusedAndCaretOffsetUnignored):
(cacheAndReturnAtkProperty):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184979 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: Update toolbar styles
nvasilyev@apple.com [Fri, 29 May 2015 04:50:59 +0000 (04:50 +0000)]
Web Inspector: Update toolbar styles
https://bugs.webkit.org/show_bug.cgi?id=145419

Also, use hairline borders (1 phisical pixel) on retina (DPI >= 2).

Reviewed by Timothy Hatcher.

* UserInterface/Views/ButtonToolbarItem.css:
(.toolbar .item.button): Deleted.
(.toolbar.icon-and-label-vertical .item.button): Deleted.
(.toolbar.icon-and-label-horizontal .item.button): Deleted.
(.toolbar .item.button > .glyph): Deleted.
(.toolbar.small-size .item.button > .glyph): Deleted.
* UserInterface/Views/DashboardContainerView.css:
(.toolbar .dashboard-container): Deleted.
(.toolbar.collapsed .dashboard-container): Deleted.
(.toolbar.small-size:matches(.icon-only, .icon-and-label-vertical, .icon-and-label-horizontal) .dashboard-container): Deleted.
(.toolbar .dashboard.slide-out-down): Deleted.
(.toolbar .dashboard.slide-in-up): Deleted.
* UserInterface/Views/Toolbar.css:
(.toolbar .search-bar > input[type="search"]):
(.toolbar .search-bar > input[type="search"]::-webkit-textfield-decoration-container):
(body:not(.unknown-mac) .toolbar .dashboard-container):
(body:not(.unknown-mac) .toolbar .search-bar > input[type="search"]):
(body:not(.unknown-mac) .toolbar .search-bar > input[type="search"]:focus):
(body:not(.unknown-mac) .toolbar .item.button:active):
(body.window-inactive:not(.unknown-mac) .toolbar .dashboard-container):
(body.unknown-mac .toolbar .dashboard-container):
(body.unknown-mac .toolbar .search-bar > input[type="search"]):
(body.unknown-mac .toolbar .search-bar > input[type="search"]:focus):
(@media (-webkit-min-device-pixel-ratio: 2)):
(body.unknown-mac .toolbar .item.button:active):
(body.unknown-mac.window-inactive .toolbar .dashboard-container):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184978 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: Jump from a computed style to the rule it came from
commit-queue@webkit.org [Fri, 29 May 2015 04:47:39 +0000 (04:47 +0000)]
Web Inspector: Jump from a computed style to the rule it came from
https://bugs.webkit.org/show_bug.cgi?id=120640

Patch by Devin Rousso <drousso@apple.com> on 2015-05-28
Reviewed by Timothy Hatcher.

* UserInterface/Views/CSSStyleDeclarationTextEditor.js:
(WebInspector.CSSStyleDeclarationTextEditor.prototype._createTextMarkerForPropertyIfNeeded):
If the delegate of CSSStyleDeclarationTextEditor has cssStyleDeclarationTextEditorShouldAddPropertyGoToArrows set to true, add a goToArrow to all CSS property entries.
(WebInspector.CSSStyleDeclarationTextEditor.prototype.highlightProperty.propertiesMatch):
(WebInspector.CSSStyleDeclarationTextEditor.prototype.highlightProperty.hasMatchingLonghandProperty):
(WebInspector.CSSStyleDeclarationTextEditor.prototype.highlightProperty):
Determines if a given CSS property is in the CSS style section and if so, select that CSS property, focus on the section containing that CSS property, and return true.
* UserInterface/Views/CSSStyleDeclarationSection.js:
(WebInspector.CSSStyleDeclarationSection.prototype.highlightProperty):
Searches through the properties of the section for a given property and scrolls to it if found.
* UserInterface/Views/CSSStyleDetailsSidebarPanel.js:
(WebInspector.CSSStyleDetailsSidebarPanel.prototype.computedStyleDetailsPanelShowProperty):
Switches to the rulesStyleDetailsPanel and scrolls to and hightlights a given property in that panel.
(WebInspector.CSSStyleDetailsSidebarPanel.prototype._navigationItemSelected):
(WebInspector.CSSStyleDetailsSidebarPanel.prototype._switchPanels):
Moved this function out of _navigationItemSelected for better reusablity.
* UserInterface/Views/ComputedStyleDetailsPanel.js:
(WebInspector.ComputedStyleDetailsPanel.prototype.cssStyleDeclarationTextEditorShowProperty):
Function that calls the delegate (which should be CSSStyleDetailsSidebarPanel) function computedStyleDetailsPanelShowProperty.
* UserInterface/Views/RulesStyleDetailsPanel.js:
(WebInspector.RulesStyleDetailsPanel):
(WebInspector.RulesStyleDetailsPanel.prototype.refresh):
(WebInspector.RulesStyleDetailsPanel.prototype.scrollToSectionAndHighlightProperty):
Searches through all the sections of the RulesStyleDetailsPanel for a given CSS property.
(WebInspector.RulesStyleDetailsPanel.prototype.shown):
(WebInspector.RulesStyleDetailsPanel.prototype.hidden):
(WebInspector.RulesStyleDetailsPanel.prototype.nodeStylesRefreshed):
Added a flag to this function that will highlight and scroll to a given property (_propertyToSelectAndHighlight) if set on load.
* UserInterface/Views/StyleDetailsPanel.js:
(WebInspector.StyleDetailsPanel.prototype.markAsNeedsRefresh):
(WebInspector.StyleDetailsPanel.prototype.nodeStylesRefreshed):
Made into protected function to allow it to be overridden.
(WebInspector.StyleDetailsPanel.prototype._nodeStylesRefreshed): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184977 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoCrash under ICU with ASAN during editing/selection/move-by-word-visually-crash-test...
mmaxfield@apple.com [Fri, 29 May 2015 04:30:09 +0000 (04:30 +0000)]
Crash under ICU with ASAN during editing/selection/move-by-word-visually-crash-test-5.html
https://bugs.webkit.org/show_bug.cgi?id=145429

Unreviewed test gardening.

* TestWebKitAPI/Tests/mac/StringTruncator.mm:
(TestWebKitAPI::TEST): Update test due to the progression.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184976 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUpdate results of WebKit1.StringTruncator after r184965.
ap@apple.com [Fri, 29 May 2015 04:03:13 +0000 (04:03 +0000)]
Update results of WebKit1.StringTruncator after r184965.

* TestWebKitAPI/Tests/mac/StringTruncator.mm: (TestWebKitAPI::TEST):

* TestWebKitAPI/Tests/mac/StringTruncator.mm:
(TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184975 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: Should have a keyboard shortcut to switch between inspector tabs
commit-queue@webkit.org [Fri, 29 May 2015 02:06:10 +0000 (02:06 +0000)]
Web Inspector: Should have a keyboard shortcut to switch between inspector tabs
https://bugs.webkit.org/show_bug.cgi?id=144207

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2015-05-28
Reviewed by Timothy Hatcher.

These keyboard shortcuts work well in an undocked inspector window.
Be careful that some keyboard shortcuts, when used inside a text
editor / text field are just text actions. However, some keyboard
shortcuts will work no matter where you trigger them.

* UserInterface/Models/KeyboardShortcut.js:
* UserInterface/Views/TabBar.js:
(WebInspector.TabBar.prototype.selectPreviousTab):
(WebInspector.TabBar.prototype.selectNextTab):
* UserInterface/Views/TabBrowser.js:
(WebInspector.TabBrowser):
(WebInspector.TabBrowser.prototype._showPreviousTab):
(WebInspector.TabBrowser.prototype._showNextTab):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184974 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed, updated my list of email addresses.
gyuyoung.kim@webkit.org [Fri, 29 May 2015 01:56:28 +0000 (01:56 +0000)]
Unreviewed, updated my list of email addresses.

* Scripts/webkitpy/common/config/contributors.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184973 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoPurge PassRefPtr in StringConcatenate.h
gyuyoung.kim@webkit.org [Fri, 29 May 2015 01:08:06 +0000 (01:08 +0000)]
Purge PassRefPtr in StringConcatenate.h
https://bugs.webkit.org/show_bug.cgi?id=145376

Reviewed by Darin Adler.

Replace PassRefPtr with RefPtr. Additionally use nullptr instead of 0.

* wtf/text/StringConcatenate.h:
(WTF::tryMakeString):
(WTF::makeString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184972 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: NewTabContentView should update button disabled state as other tabs...
commit-queue@webkit.org [Fri, 29 May 2015 00:20:40 +0000 (00:20 +0000)]
Web Inspector: NewTabContentView should update button disabled state as other tabs are added/removed
https://bugs.webkit.org/show_bug.cgi?id=145448

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2015-05-28
Reviewed by Timothy Hatcher.

* UserInterface/Views/NewTabContentView.js:
(WebInspector.NewTabContentView):
Always add a click listener and add the type as a private property on the element.

(WebInspector.NewTabContentView.prototype.shown):
(WebInspector.NewTabContentView.prototype.hidden):
Add / remove event listeners for tab changes. Also update the view when shown.

(WebInspector.NewTabContentView.prototype._createNewTab):
A disabled button an now be clicked. Do nothing if a new tab of this type is not allowed.

(WebInspector.NewTabContentView.prototype._updateTabItems):
Refresh the disabled state for each of the buttons.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184971 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoSubpixel rendering: Pixel crack in text selection of simple text in <textarea>.
zalan@apple.com [Fri, 29 May 2015 00:15:24 +0000 (00:15 +0000)]
Subpixel rendering: Pixel crack in text selection of simple text in <textarea>.
https://bugs.webkit.org/show_bug.cgi?id=145393
rdar://problem/19918941

Reviewed by Darin Adler.

Float to LayoutUnit conversion is lossy. To ensure that selection
painting always lines up (snaps) properly, the calculated width needs to
be adjusted by ceiling the float to the next LayoutUnit value.

Source/WebCore:

Test: fast/text/hidpi-text-selection-gap-between-words.html

* platform/graphics/FontCascade.cpp:
(WebCore::FontCascade::adjustSelectionRectForSimpleText):
* platform/graphics/cocoa/FontCascadeCocoa.mm:
(WebCore::FontCascade::adjustSelectionRectForComplexText):

LayoutTests:

* fast/text/hidpi-text-selection-gap-between-words-expected.html: Added.
* fast/text/hidpi-text-selection-gap-between-words.html: Added.
* platform/mac/platform/mac/editing/input/caret-primary-bidi-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184970 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agomozilla/js1_5/Array/regress-154338.js test causes ARM 32 bit iOS devices to run out...
msaboff@apple.com [Thu, 28 May 2015 23:51:57 +0000 (23:51 +0000)]
mozilla/js1_5/Array/regress-154338.js test causes ARM 32 bit iOS devices to run out of memory
https://bugs.webkit.org/show_bug.cgi?id=145444

Reviewed by Geoffrey Garen.

Disabled mozilla/js1_5/Array/regress-154338.js when run on iOS ARM 32 bit devices and
the --memory-limited option is passed to run-jsc-stress-tests.

* tests/mozilla/mozilla-tests.yaml:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184969 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoGraphicsLayerCA should start out with m_intersectsCoverageRect=false
akling@apple.com [Thu, 28 May 2015 23:48:29 +0000 (23:48 +0000)]
GraphicsLayerCA should start out with m_intersectsCoverageRect=false
<https://webkit.org/b/145443>
<rdar://problem/19685798>

Reviewed by Simon Fraser.

Source/WebCore:

Since the coverage rect of a newly-constructed GraphicsLayerCA is always empty,
there's no way that the layer's own bounding box can intersect with it.
Hence the m_intersectsCoverageRect flag should be false after construction.

This fixes an issue where compositing layers in subframes would be marked as
non-volatile, despite being far outside the viewport and coverage rect.

Test: compositing/visible-rect/iframe-with-layers-outside-viewport.html

* platform/graphics/ca/GraphicsLayerCA.cpp:
(WebCore::GraphicsLayerCA::GraphicsLayerCA):

LayoutTests:

Add a test (based on iframe-and-layers.html) that makes an iframe with compositing
layers inside it and places it far outside the visible viewport.

The layer dump should show that the "intersects coverage rect" flag is 0 for the
iframe and its sublayers.

* compositing/visible-rect/iframe-with-layers-outside-viewport-expected.txt: Added.
* compositing/visible-rect/iframe-with-layers-outside-viewport.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184968 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed, rolling out r184796 and r184832.
commit-queue@webkit.org [Thu, 28 May 2015 23:20:51 +0000 (23:20 +0000)]
Unreviewed, rolling out r184796 and r184832.
https://bugs.webkit.org/show_bug.cgi?id=145449

2% performance regression (Requested by litherum on #webkit).

Reverted changesets:

"[Cocoa] Use CTFontDrawGlyphs() instead of
CGContextShowGlyphsWithAdvances()/CGContextShowGlyphsAtPositions()"
https://bugs.webkit.org/show_bug.cgi?id=145234
http://trac.webkit.org/changeset/184796

"Cleanup after r184796"
https://bugs.webkit.org/show_bug.cgi?id=145333
http://trac.webkit.org/changeset/184832

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184967 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRename HTMLMediaSession to MediaElementSession to reserve "media session" for the...
commit-queue@webkit.org [Thu, 28 May 2015 23:03:34 +0000 (23:03 +0000)]
Rename HTMLMediaSession to MediaElementSession to reserve "media session" for the Media Session API.
https://bugs.webkit.org/show_bug.cgi?id=145440

Patch by Matt Rajca <mrajca@apple.com> on 2015-05-28
Reviewed by Eric Carlson.

* CMakeLists.txt:
* WebCore.vcxproj/WebCore.vcxproj:
* WebCore.vcxproj/WebCore.vcxproj.filters:
* WebCore.xcodeproj/project.pbxproj:
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::HTMLMediaElement):
(WebCore::HTMLMediaElement::loadInternal):
(WebCore::HTMLMediaElement::loadResource):
(WebCore::HTMLMediaElement::suspend):
(WebCore::HTMLMediaElement::resume):
(WebCore::HTMLMediaElement::removeBehaviorsRestrictionsAfterFirstUserGesture):
(WebCore::HTMLMediaElement::mediaState):
* html/HTMLMediaElement.h:
(WebCore::HTMLMediaElement::mediaSession):
* html/MediaElementSession.cpp: Renamed from Source/WebCore/html/HTMLMediaSession.cpp.
(WebCore::restrictionName):
(WebCore::MediaElementSession::MediaElementSession):
(WebCore::MediaElementSession::registerWithDocument):
(WebCore::MediaElementSession::unregisterWithDocument):
(WebCore::MediaElementSession::addBehaviorRestriction):
(WebCore::MediaElementSession::removeBehaviorRestriction):
(WebCore::MediaElementSession::playbackPermitted):
(WebCore::MediaElementSession::dataLoadingPermitted):
(WebCore::MediaElementSession::fullscreenPermitted):
(WebCore::MediaElementSession::pageAllowsDataLoading):
(WebCore::MediaElementSession::pageAllowsPlaybackAfterResuming):
(WebCore::MediaElementSession::showPlaybackTargetPicker):
(WebCore::MediaElementSession::hasWirelessPlaybackTargets):
(WebCore::MediaElementSession::wirelessVideoPlaybackDisabled):
(WebCore::MediaElementSession::setWirelessVideoPlaybackDisabled):
(WebCore::MediaElementSession::setHasPlaybackTargetAvailabilityListeners):
(WebCore::MediaElementSession::setPlaybackTarget):
(WebCore::MediaElementSession::targetAvailabilityChangedTimerFired):
(WebCore::MediaElementSession::externalOutputDeviceAvailableDidChange):
(WebCore::MediaElementSession::canPlayToWirelessPlaybackTarget):
(WebCore::MediaElementSession::isPlayingToWirelessPlaybackTarget):
(WebCore::MediaElementSession::setShouldPlayToPlaybackTarget):
(WebCore::MediaElementSession::mediaStateDidChange):
(WebCore::MediaElementSession::effectivePreloadForElement):
(WebCore::MediaElementSession::requiresFullscreenForVideoPlayback):
(WebCore::MediaElementSession::mediaEngineUpdated):
(WebCore::MediaElementSession::allowsAlternateFullscreen):
(WebCore::MediaElementSession::maximumMediaSourceBufferSize):
* html/MediaElementSession.h: Renamed from Source/WebCore/html/HTMLMediaSession.h.
* testing/Internals.cpp:
(WebCore::Internals::setMediaElementRestrictions):
(WebCore::Internals::setAudioContextRestrictions):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184966 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoCrash under ICU with ASAN during editing/selection/move-by-word-visually-crash-test...
mmaxfield@apple.com [Thu, 28 May 2015 22:41:19 +0000 (22:41 +0000)]
Crash under ICU with ASAN during editing/selection/move-by-word-visually-crash-test-5.html
https://bugs.webkit.org/show_bug.cgi?id=145429
<rdar://problem/20992218>

Reviewed by Alexey Proskuryakov.

WebKit uses some strings which contain the lower 8-bits of UTF-16 (thereby saving space). However,
ICU doesn't understand this encoding. When we want to use ICU functions with strings in this encoding,
we create a UTextProvider which converts our encoded strings to UTF-16 for ICU, one chunk at a time.
This object contains a vtable which we populate to perform the conversion.

The WebKit function which actually returns the UTF-16 chunks has two relevant arguments: an index into
the encoded string which ICU is requesting, and a direction from that index which ICU is interested
in. This function populates a "chunk" which is characterized by a pointer to a buffer, the length of
the populated data in the buffer, and an offset into the chunk which represents the index that the
requested character was put into.

When ICU requests data going backward, we fill in the chunk accordingly, with the requested character
all the way at the end. We then set the offset equal to the length of the buffer. However, this length
value is stale from the previous time the function ran. Therefore, ICU was reading the wrong index in
the chunk when expecting the requested character.

Covered by editing/selection/move-by-word-visually-crash-test-5.html.

* platform/text/icu/UTextProviderLatin1.cpp:
(WebCore::uTextLatin1Access):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184965 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoMake sure that we create the private WebKit.framework symlinks last
andersca@apple.com [Thu, 28 May 2015 22:40:02 +0000 (22:40 +0000)]
Make sure that we create the private WebKit.framework symlinks last
https://bugs.webkit.org/show_bug.cgi?id=145442
<rdar://problem/21142213>

Reviewed by Dan Bernstein.

Move the "Add Symlinks in /System/Library/PrivateFrameworks/WebKit.framework" build phase to
the "All" aggregate target. Also update the script now that INSTALL_PATH is different.

* WebKit2.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184964 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[iOS] Fix controls bug caused by r184794
eric.carlson@apple.com [Thu, 28 May 2015 22:31:17 +0000 (22:31 +0000)]
[iOS] Fix controls bug caused by r184794
https://bugs.webkit.org/show_bug.cgi?id=145445

Reviewed by Dean Jackson.

* Modules/mediacontrols/mediaControlsiOS.js:
(ControllerIOS.prototype.showControls): Check video.controls instead of trying to call
showInlinePlaybackPlaceholderOnly, which doesn't exist.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184963 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoCode clean up for extracting information from the mix of WritingMode and TextDirection
commit-queue@webkit.org [Thu, 28 May 2015 21:55:32 +0000 (21:55 +0000)]
Code clean up for extracting information from the mix of WritingMode and TextDirection
https://bugs.webkit.org/show_bug.cgi?id=143399

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2015-05-28
Reviewed by Darin Adler.
Source/WebCore:

There should be eight TextFlows. A TextFlow consists of two directions.
The first one controls the inline growing direction and the second one
controls the block growing direction. We are going to use the cardinal
directions to define these combinations.

We need to extract the following information easily from the TextFlow

-- isTextFlowReversed() // RTL
-- isTextFlowFlipped()  // the opposite direction to normal
-- isTextFlowVertical() // The line grows vertically

Mixing WritingMode and TextDirection into TextFlow and choosing their values
carefully to ease implementing the above functions can make things clearer.

* Modules/notifications/Notification.h:
* platform/graphics/TextRun.h:
* platform/ios/SelectionRect.h:
* platform/text/BidiResolver.h:
* platform/text/StringWithDirection.h:
* platform/text/TextFlags.h:
(WebCore::isLeftToRightDirection): Deleted.
enum TextDirection is moved from TextFlags.h to WritingMode.h. So include
WritingMode.h instead of including TextFlags.h.

* css/CSSProperty.cpp:
(WebCore::resolveToPhysicalProperty): Instead of handling the eight possible
cases of the text flows one by one, we can map from a LogicalSide to a
PhysicalSide using the TextFlow value.

* platform/LengthBox.h:
(WebCore::BoxExtent::before):
(WebCore::BoxExtent::after):
(WebCore::BoxExtent::start):
(WebCore::BoxExtent::end):
Use mapLogicalSideToPhysicalSide() to convert from a LogicalSide to a PhysicalSide.

* platform/text/WritingMode.h:
(WebCore::isLeftToRightDirection): Usually the default enum is defined
first. And usually the default value is equal to zero. Since LTR is the
default for the text direction, it should be defined before RTL.

(WebCore::makeTextFlow): Returns a TextFlow given a WritingMode and a
TextDirection.

(WebCore::isReversedTextFlow): Returns true if the text direction is RTL.

(WebCore::isFlippedTextFlow):
(WebCore::isFlippedWritingMode): Returns true if the block growing direction
is in the opposite direction of normal.

(WebCore::isVerticalTextFlow):
(WebCore::isVerticalWritingMode): Returns true if the line grows vertically.

(WebCore::isHorizontalWritingMode):
(WebCore::isFlippedLinesWritingMode): Use isVerticalWritingMode() and
isFlippedWritingMode() to answer these questions.

(WebCore::isHorizontalPhysicalSide):
(WebCore::mirrorPhysicalSide): top <-> bottom and left <-> right conversion.
(WebCore::rotatePhysicalSide): top <-> left and right <-> bottom conversion.

(WebCore::mapLogicalSideToPhysicalSide): Maps a LogicalSide to a PhysicalSide
using either the TextFlow or the WritingMode.

(WebCore::isFlippedBlocksWritingMode): Deleted.
* rendering/RenderMultiColumnSet.cpp:
(WebCore::RenderMultiColumnSet::collectLayerFragments):
(WebCore::RenderMultiColumnSet::columnTranslationForOffset):
* rendering/shapes/Shape.cpp:
(WebCore::physicalRectToLogical):
(WebCore::physicalPointToLogical):
* rendering/style/RenderStyle.h:
Rename isFlippedBlocksWritingMode() to be isFlippedWritingMode() so we can
be consistent with other WritingMode query functions.

Source/WebKit2:

* Shared/WebPopupItem.h:
* UIProcess/API/C/efl/WKAPICastEfl.h:
* UIProcess/WebPopupMenuProxy.h:
enum TextDirection is moved from TextFlags.h to WritingMode.h. So include
WritingMode.h instead of including TextFlags.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184962 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoDisable network cache for old clients
antti@apple.com [Thu, 28 May 2015 21:05:26 +0000 (21:05 +0000)]
Disable network cache for old clients
https://bugs.webkit.org/show_bug.cgi?id=145418

Reviewed by Andreas Kling.

Follow-up: Don't do linked-on-or-after checks for iOS Safari.

* UIProcess/Cocoa/VersionChecks.mm:
(WebKit::linkTimeVersion):
(WebKit::linkedOnOrAfter):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184961 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[iOS8][ARMv7(s)] Optimized Object.create in 'use strict' context sometimes breaks.
benjamin@webkit.org [Thu, 28 May 2015 20:57:48 +0000 (20:57 +0000)]
[iOS8][ARMv7(s)] Optimized Object.create in 'use strict' context sometimes breaks.
https://bugs.webkit.org/show_bug.cgi?id=138038

Reviewed by Michael Saboff.

TL;DR: sometimes the baseline JIT could accidentally nuke the tag before calling
       to C++, making put_by_id behave erratically.

The bug was that put_by_id would randomly not work correctly in 32bits. It happened
in the baseline JIT if we were unlucky enough:
-The code get hot enough and the structure is stable so we get a fast path for
 put_by_id.
-We repatch the fast-path branch with a stub generated by
 emitPutTransitionStubAndGetOldStructure().
-In emitPutTransitionStubAndGetOldStructure(), we only preserve the payload of the base
 register, the tag register is ignored.
-emitPutTransitionStubAndGetOldStructure() allocate 2 to 3 registers. Any of those
 could be the one used for the base's tag before the fast path and the value is trashed.
-If we hit one of the failure case, we fallback to the slow path, but we destroyed
 the tag pointer.
-We now have unrelated bits in the tag, the most likely value type is now "double"
 and we fail the put_by_id because we try to set a property on a number.

The most obvious solution would be to change emitPutTransitionStubAndGetOldStructure()
to preserve the tag register in addition to the value register.
I decided against that option because of the added complexity. The DFG does not need
that case, so I would have to add branches everywhere to distinguish the cases
were we need to preserve the tag or not.

Instead, I just load the tag back from memory in the slow path. The function in the slow
path is several order of magnitude slower than a load, it is not worth eliminating it,
especially in baseline JIT.

I also discovered 4 useless loads in the fast path, so even with my extra load, this patch
makes the baseline faster :)

* jit/JITPropertyAccess32_64.cpp:
(JSC::JIT::emitSlow_op_put_by_id):
(JSC::JIT::emit_op_put_by_id): Deleted.
* tests/stress/put-by-id-on-new-object-after-prototype-transition-non-strict.js: Added.
(opaqueNewObject):
(putValueOnNewObject):
* tests/stress/put-by-id-on-new-object-after-prototype-transition-strict.js: Added.
(string_appeared_here.opaqueNewObject):
(putValueOnNewObject):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184960 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[JSC] reduction the iteration count of the DoubleRep stress tests
benjamin@webkit.org [Thu, 28 May 2015 20:46:14 +0000 (20:46 +0000)]
[JSC] reduction the iteration count of the DoubleRep stress tests

Once again, I used big numbers for manual testing and I forgot to fix them before landing.

* tests/stress/double-rep-with-non-cell.js:
* tests/stress/double-rep-with-null.js:
* tests/stress/double-rep-with-undefined.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184959 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAdd debug mode assertions for accessors casting JSC::DFG::Node.m_opInfo
basile_clement@apple.com [Thu, 28 May 2015 19:14:30 +0000 (19:14 +0000)]
Add debug mode assertions for accessors casting JSC::DFG::Node.m_opInfo
https://bugs.webkit.org/show_bug.cgi?id=145441

Reviewed by Filip Pizlo.

Most accessor functions casting m_opInfo in JSC::DFG::Node are
performing debug checks that they are only accessed for node types that
should have them. This patch adds similar checks for the accessors that
were missing them.

* dfg/DFGNode.h:
(JSC::DFG::Node::watchpointSet):
(JSC::DFG::Node::storagePointer):
(JSC::DFG::Node::multiGetByOffsetData):
(JSC::DFG::Node::multiPutByOffsetData):
(JSC::DFG::Node::hasTypeLocation):
(JSC::DFG::Node::typeLocation):
(JSC::DFG::Node::hasBasicBlockLocation):
(JSC::DFG::Node::basicBlockLocation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184958 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAdd ENABLE_MEDIA_SESSION feature flag (which is off by default).
commit-queue@webkit.org [Thu, 28 May 2015 17:40:49 +0000 (17:40 +0000)]
Add ENABLE_MEDIA_SESSION feature flag (which is off by default).
https://bugs.webkit.org/show_bug.cgi?id=145415

Patch by Matt Rajca <mrajca@apple.com> on 2015-05-28
Reviewed by Eric Carlson.

* Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184956 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Streams API] ReadableStreamReader should not be exposed
youenn.fablet@crf.canon.fr [Thu, 28 May 2015 12:16:45 +0000 (12:16 +0000)]
[Streams API] ReadableStreamReader should not be exposed
https://bugs.webkit.org/show_bug.cgi?id=143564

Reviewed by Darin Adler.

Source/WebCore:

Covered by rebased tests.

* Modules/streams/ReadableStreamReader.idl: Adding NoInterfaceObject to ReadableStreamReader.idl.

LayoutTests:

Rebasing tests as ReadableStreamReader is no longer defined globally.

* js/dom/global-constructors-attributes-expected.txt:
* platform/efl/js/dom/global-constructors-attributes-expected.txt:
* platform/gtk/js/dom/global-constructors-attributes-expected.txt:
* platform/mac-mavericks/js/dom/global-constructors-attributes-expected.txt:
* platform/mac/js/dom/global-constructors-attributes-expected.txt:
* platform/win/js/dom/global-constructors-attributes-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184955 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[CMake] Improve detection and usage of GL/GLES/EGL libraries.
clopez@igalia.com [Thu, 28 May 2015 11:30:52 +0000 (11:30 +0000)]
[CMake] Improve detection and usage of GL/GLES/EGL libraries.
https://bugs.webkit.org/show_bug.cgi?id=145408

Reviewed by Carlos Garcia Campos.

.:

* Source/cmake/FindEGL.cmake: Improve detection of EGL libraries.
* Source/cmake/FindGLES.cmake: Removed. It was used by the EGL port.
Remove it and make the EGL port use the improved FindOpenGLES2.cmake
instead.
* Source/cmake/FindOpenGL.cmake: Added. Add module to detect OpenGL
libraries. Detect also GLX libraries.
* Source/cmake/FindOpenGLES2.cmake: Improve detection of OpenGLES-v2
libraries. Use find_path() to get the include path.
* Source/cmake/OptionsEfl.cmake: Use now the improved FindOpenGLES2
module.
* Source/cmake/OptionsGTK.cmake: Set default value for ENABLE_GLES2
depending on the libraries found on the system.
Move the detection of GLX (and the include of CMakePushCheckState)
to FindOpenGL.cmake.
Ensure that we only define USE_GLX when we build with OpenGL
(but not with GLESv2).

Source/WebCore:

No new tests, no behavior changes.

* CMakeLists.txt: Ensure that we include the libraries and includes
for the GL/GLESv2/EGL libraries before including the ANGLE directories.
Define also any CFLAG that the system GL/GLESv2/EGL libraries may need.
* PlatformEfl.cmake: Remove some includes that are now unneeded,
because we are including now the EGL libraries on CMakeLists.txt
* PlatformGTK.cmake: Remove unneeded include (We are including the EGL
libraries now on CMakeLists.txt)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184954 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoBinding generator should support interfaces with CustomConstructor and NoInterfaceObject
youenn.fablet@crf.canon.fr [Thu, 28 May 2015 10:12:46 +0000 (10:12 +0000)]
Binding generator should support interfaces with CustomConstructor and NoInterfaceObject
https://bugs.webkit.org/show_bug.cgi?id=145016

Reviewed by Darin Adler.

Updated code generator to generate the necessary declarations (constructor property in the prototype, the constructor class).
In the case of CustomConstructor with NoInterfaceObject, the implementation of the constructor property related JS function is changed as follow:
It creates a constructor object and shadows the constructor property JS function with this object.
This ensures that only one constructor object is created per prototype.
The constructor is not exposed as it is not added to the global object constructor map.

NoInterfaceObject is asking to not make the interface visible, which is contradictory with having a visible constructor of the interface.
The case for having NoInterfaceObject and CustomConstructor is the following:
We do not want to expose the interface but we still want to have a constructor property on the prototype of objects implementing the interface.
One such case is https://streams.spec.whatwg.org/#globals

Added TestCustomConstructorWithNoInterfaceObject.idl to check that case.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateHeader):
(GenerateAttributesHashTable):
(GenerateImplementation):
(HasConstructorProperty):
* bindings/scripts/test/GObject/WebKitDOMTestCustomConstructorWithNoInterfaceObject.cpp: Added.
(WebKit::kit):
(WebKit::core):
(WebKit::wrapTestCustomConstructorWithNoInterfaceObject):
(webkit_dom_test_custom_constructor_with_no_interface_object_finalize):
(webkit_dom_test_custom_constructor_with_no_interface_object_constructor):
(webkit_dom_test_custom_constructor_with_no_interface_object_class_init):
(webkit_dom_test_custom_constructor_with_no_interface_object_init):
* bindings/scripts/test/GObject/WebKitDOMTestCustomConstructorWithNoInterfaceObject.h: Added.
* bindings/scripts/test/GObject/WebKitDOMTestCustomConstructorWithNoInterfaceObjectPrivate.h: Added.
* bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.cpp: Added.
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectPrototype::create):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectPrototype::createStructure):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectPrototype::JSTestCustomConstructorWithNoInterfaceObjectPrototype):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectConstructor::create):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectConstructor::createStructure):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectConstructor::JSTestCustomConstructorWithNoInterfaceObjectConstructor):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectConstructor::finishCreation):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectConstructor::getConstructData):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectPrototype::finishCreation):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::JSTestCustomConstructorWithNoInterfaceObject):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::createPrototype):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::getPrototype):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::destroy):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::~JSTestCustomConstructorWithNoInterfaceObject):
(WebCore::jsTestCustomConstructorWithNoInterfaceObjectConstructor):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::getConstructor):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectOwner::isReachableFromOpaqueRoots):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectOwner::finalize):
(WebCore::toJS):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::toWrapped):
* bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.h: Added.
(WebCore::wrapperOwner):
(WebCore::toJS):
* bindings/scripts/test/ObjC/DOMTestCustomConstructorWithNoInterfaceObject.h: Added.
* bindings/scripts/test/ObjC/DOMTestCustomConstructorWithNoInterfaceObject.mm: Added.
(-[DOMTestCustomConstructorWithNoInterfaceObject dealloc]):
(-[DOMTestCustomConstructorWithNoInterfaceObject finalize]):
(core):
(kit):
* bindings/scripts/test/ObjC/DOMTestCustomConstructorWithNoInterfaceObjectInternal.h: Added.
* bindings/scripts/test/TestCustomConstructor.idl: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184953 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed, Fix build break since r184930.
gyuyoung.kim@webkit.org [Thu, 28 May 2015 09:40:36 +0000 (09:40 +0000)]
Unreviewed, Fix build break since r184930.

* UIProcess/Storage/StorageManager.cpp:
(WebKit::StorageManager::createTransientLocalStorageMap): Pass a pointer of SecurityOrigin
in isSameSchemeHostPort().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184952 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed. Fixing the build after r184950.
zandobersek@gmail.com [Thu, 28 May 2015 09:23:56 +0000 (09:23 +0000)]
Unreviewed. Fixing the build after r184950.

* UIProcess/Storage/StorageManager.cpp:
(WebKit::StorageManager::createSessionStorageMap): getOrCreateStorageArea()
now expects a Ref<WebCore::SecurityOrigin> object.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184951 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[WK2] Handle WebCore::SecurityOrigin objects through Ref<> where possible
zandobersek@gmail.com [Thu, 28 May 2015 08:18:55 +0000 (08:18 +0000)]
[WK2] Handle WebCore::SecurityOrigin objects through Ref<> where possible
https://bugs.webkit.org/show_bug.cgi?id=145373

Reviewed by Andreas Kling.

WebCore::SecurityOrigin creation functions in most cases return a
Ref<WebCore::SecurityOrigin> object. In the WebKit2 layer, those objects
were still largely converted to RefPtr<>s.

This patch switches to using Ref<> where possible. The remaining uses of
RefPtr<WebCore::SecurityOrigin> are present due to HashTables, to-be-copied
Vectors, WebCore::DatabaseManager::origins() API and code where non-null
WebCore::SecurityOrigin can't be guaranteed.

* DatabaseProcess/DatabaseProcess.cpp:
(WebKit::DatabaseProcess::deleteIndexedDatabaseEntriesForOrigin):
* NetworkProcess/NetworkProcess.cpp:
(WebKit::cfURLCacheOrigins):
* UIProcess/Storage/LocalStorageDatabase.cpp:
(WebKit::LocalStorageDatabase::create):
(WebKit::LocalStorageDatabase::LocalStorageDatabase):
(WebKit::LocalStorageDatabase::openDatabase):
(WebKit::LocalStorageDatabase::close):
* UIProcess/Storage/LocalStorageDatabase.h:
* UIProcess/Storage/LocalStorageDatabaseTracker.cpp:
(WebKit::LocalStorageDatabaseTracker::origins):
* UIProcess/Storage/LocalStorageDatabaseTracker.h:
* UIProcess/Storage/StorageManager.cpp:
(WebKit::StorageManager::StorageArea::securityOrigin):
(WebKit::StorageManager::TransientLocalStorageNamespace::getOrCreateStorageArea):
(WebKit::StorageManager::TransientLocalStorageNamespace::origins):
(WebKit::StorageManager::TransientLocalStorageNamespace::clearStorageAreasMatchingOrigin):
(WebKit::StorageManager::StorageArea::create):
(WebKit::StorageManager::StorageArea::StorageArea):
(WebKit::StorageManager::StorageArea::openDatabaseAndImportItemsIfNeeded):
(WebKit::StorageManager::LocalStorageNamespace::getOrCreateStorageArea):
(WebKit::StorageManager::LocalStorageNamespace::didDestroyStorageArea):
(WebKit::StorageManager::SessionStorageNamespace::origins):
(WebKit::StorageManager::SessionStorageNamespace::clearStorageAreasMatchingOrigin):
(WebKit::StorageManager::SessionStorageNamespace::getOrCreateStorageArea):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::reachedApplicationCacheOriginQuota):
* WebProcess/ApplicationCache/WebApplicationCacheManager.cpp:
(WebKit::WebApplicationCacheManager::getApplicationCacheOrigins):
* WebProcess/InjectedBundle/InjectedBundle.cpp:
(WebKit::InjectedBundle::setApplicationCacheOriginQuota):
(WebKit::InjectedBundle::resetApplicationCacheOriginQuota):
* WebProcess/ResourceCache/WebResourceCacheManager.cpp:
(WebKit::WebResourceCacheManager::returnCacheOrigins):
(WebKit::WebResourceCacheManager::clearCacheForOrigin):
* WebProcess/WebCoreSupport/WebDatabaseManager.cpp:
(WebKit::WebDatabaseManager::deleteDatabaseWithNameForOrigin):
(WebKit::WebDatabaseManager::deleteDatabasesForOrigin):
(WebKit::WebDatabaseManager::setQuotaForOrigin):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184950 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoClean up HashTable constructors
zandobersek@gmail.com [Thu, 28 May 2015 08:13:10 +0000 (08:13 +0000)]
Clean up HashTable constructors
https://bugs.webkit.org/show_bug.cgi?id=145369

Reviewed by Andreas Kling.

Use nullptr to initialize pointer member variables in the HashTable
default constructor. Copy and move constructors can use constructor
delegation instead of replicating all member initializations. Move
constructor should simply call HashTable::swap() instead of replicating
all the swap operations.

* wtf/HashTable.h:
(WTF::KeyTraits>::HashTable):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184949 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoBuild fix after r184946.
jonlee@apple.com [Thu, 28 May 2015 07:44:41 +0000 (07:44 +0000)]
Build fix after r184946.

* WebView/WebPreferences.mm:
(-[WebPreferences allowsAirPlayForMediaPlayback]):
(-[WebPreferences setAllowsAirPlayForMediaPlayback:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184948 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[iOS] When viewing an MJPEG stream as the main resource, only the first
jhoneycutt@apple.com [Thu, 28 May 2015 07:34:21 +0000 (07:34 +0000)]
[iOS] When viewing an MJPEG stream as the main resource, only the first
frame paints

<https://bugs.webkit.org/show_bug.cgi?id=145185>
<rdar://problem/20124694>

Source/WebCore:

This bug is caused by state not being restored between successive loads
of multipart/x-mixed-replace content. Each part of the stream is not
treated as a wholly new load, so if state is cleared as part of the
previous page being destroyed, it may not be restored when loading
the next stream part.

On WK1, tile cache updates are disabled in FrameView::clear(), which is
called as the previous page is destroyed, but were not being
reenabled when loading new replacement content from the stream.

On WK2, the exposed content rect was being reset to an empty rect as
the previous page was destroyed, but it was not being reset by loading
replacement content.

To fix these issues, I added two WebFrameLoaderClient callbacks;
willReplaceMultipartContent() and didReplaceMultipartContent(). These
are used to save and restore state between successive loads.

Reviewed by Darin Adler.

* loader/DocumentLoader.cpp:
Call the new client callbacks.
(WebCore::DocumentLoader::commitLoad):
If replacing content in a multipart/x-mixed-replace stream, call the
WebFrameLoaderClient's didReplaceMultipartContent().

(WebCore::DocumentLoader::setupForReplace):
Call the WebFrameLoaderClient's willReplaceMultipartContent().

* loader/EmptyClients.h:

* loader/FrameLoaderClient.h:

* page/FrameView.cpp:
(WebCore::FrameView::didReplaceMultipartContent):
Re-enable tile cache updates that were disabled in FrameView::clear().
This fixes the issue for WK1.

* page/FrameView.h:
Declare an exported function, didReplaceMultipartContent(), that WebKit
can call.

Source/WebKit/mac:

Reviewed by Darin Adler.

* WebCoreSupport/WebFrameLoaderClient.h:

* WebCoreSupport/WebFrameLoaderClient.mm:
(WebFrameLoaderClient::didReplaceMultipartContent):
Call the FrameView's didReplaceMultipartContent().

Source/WebKit/win:

Reviewed by Darin Adler.

* WebCoreSupport/WebFrameLoaderClient.h:
Stubbed new functions that aren't used on Windows.

Source/WebKit2:

Reviewed by Darin Adler.

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::willReplaceMultipartContent):
Tell the WebPage that we will replace multipart content.

(WebKit::WebFrameLoaderClient::didReplaceMultipartContent):
Tell the WebPage that we did replace multipart content.

* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:
* WebProcess/WebPage/DrawingArea.h:

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::willReplaceMultipartContent):
Save the exposed content rect.
(WebKit::WebPage::didReplaceMultipartContent):
Restore the exposed content rect that was saved before this load. This
fixes the issue for WK2.

* WebProcess/WebPage/WebPage.h:

* WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.h:
* WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.mm:
(WebKit::RemoteLayerTreeDrawingArea::exposedContentRect):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184947 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoFix macros for wireless playback
jonlee@apple.com [Thu, 28 May 2015 07:27:13 +0000 (07:27 +0000)]
Fix macros for wireless playback
https://bugs.webkit.org/show_bug.cgi?id=145239
rdar://problem/21049541

Reviewed by Anders Carlsson.

Encompass the logic within ENABLE(WIRELESS_TARGET_PLAYBACK). Update property name.

Source/WebKit/mac:

* WebView/WebPreferenceKeysPrivate.h:
* WebView/WebPreferences.h:
* WebView/WebPreferences.mm:
(+[WebPreferences initialize]): Wrap in ENABLE.
(-[WebPreferences allowsAirPlayForMediaPlayback]): Ditto.
(-[WebPreferences setAllowsAirPlayForMediaPlayback:]): Ditto.
* WebView/WebView.mm:
(-[WebView _preferencesChanged:]): Ditto.

Source/WebKit2:

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView initWithFrame:configuration:]):
* UIProcess/API/Cocoa/WKWebViewConfiguration.h: Update names. Add a WKDeprecated
category section.
* UIProcess/API/Cocoa/WKWebViewConfiguration.mm:
(-[WKWebViewConfiguration init]): Update name of variables to reflect new property names.
(-[WKWebViewConfiguration copyWithZone:]): Ditto.
(-[WKWebViewConfiguration mediaPlaybackAllowsAirPlay]): Get and set off of the new
variable.
(-[WKWebViewConfiguration setMediaPlaybackAllowsAirPlay:]): Ditto.
(-[WKWebViewConfiguration mediaPlaybackRequiresUserAction]): Ditto.
(-[WKWebViewConfiguration setMediaPlaybackRequiresUserAction:]): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184946 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRemove unused ResourceRequest "user initiated" flag.
beidson@apple.com [Thu, 28 May 2015 06:14:50 +0000 (06:14 +0000)]
Remove unused ResourceRequest "user initiated" flag.
https://bugs.webkit.org/show_bug.cgi?id=145428

Reviewed by Alexey Proskuryakov.

No new tests (No change in behavior).

* platform/ios/WebCoreSystemInterfaceIOS.mm:
* platform/mac/WebCoreSystemInterface.mm:

* platform/network/ResourceRequestBase.h:
(WebCore::ResourceRequestBase::initiatedByUserGesture): Deleted.
(WebCore::ResourceRequestBase::setInitiatedByUserGesture): Deleted.
(WebCore::ResourceRequestBase::encodeWithoutPlatformData): Deleted.
(WebCore::ResourceRequestBase::decodeWithoutPlatformData): Deleted.

* platform/network/cf/ResourceRequestCFNet.cpp:
(WebCore::ResourceRequest::doUpdatePlatformRequest): Deleted.
(WebCore::ResourceRequest::doUpdateResourceRequest): Deleted.

* platform/network/cocoa/ResourceRequestCocoa.mm:
(WebCore::ResourceRequest::doUpdateResourceRequest): Deleted.
(WebCore::ResourceRequest::doUpdatePlatformRequest): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184945 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Mac] Fix clang static analyzer build
ap@apple.com [Thu, 28 May 2015 06:12:46 +0000 (06:12 +0000)]
[Mac] Fix clang static analyzer build
https://bugs.webkit.org/show_bug.cgi?id=145426
rdar://problem/20947408

Reviewed by Dan Bernstein.

* wtf/spi/darwin/XPCSPI.h: Correct declarations of xpc_retain and xpc_release.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184944 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoDelete media/controls-layers.html. It's a bad test that fails and asserts, and there
ap@apple.com [Thu, 28 May 2015 05:52:59 +0000 (05:52 +0000)]
Delete media/controls-layers.html. It's a bad test that fails and asserts, and there
are apparently no plans to fix any of that (see https://bugs.webkit.org/show_bug.cgi?id=145420).

* media/controls-layers.html: Removed.
* platform/mac-mavericks/media/controls-layers-expected.txt: Removed.
* platform/mac/media/controls-layers-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184943 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoArray.of should work with other constructors
commit-queue@webkit.org [Thu, 28 May 2015 05:47:11 +0000 (05:47 +0000)]
Array.of should work with other constructors
https://bugs.webkit.org/show_bug.cgi?id=145365
Source/JavaScriptCore:

Per https://people.mozilla.org/~jorendorff/es6-draft.html#sec-array.of
step 4

Patch by Jordan Harband <ljharb@gmail.com> on 2015-05-27
Reviewed by Yusuke Suzuki.

* builtins/ArrayConstructor.js:
(of):
* runtime/ArrayConstructor.cpp:
(JSC::arrayConstructorOf): Deleted.

LayoutTests:

Patch by Jordan Harband <ljharb@gmail.com> on 2015-05-27
Reviewed by Yusuke Suzuki.

* js/array-of-expected.txt:
* js/script-tests/array-of.js:
(Foo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184942 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAdd a "should open urls externally" flag to DocumentLoader.
beidson@apple.com [Thu, 28 May 2015 04:52:25 +0000 (04:52 +0000)]
Add a "should open urls externally" flag to DocumentLoader.
rdar://problem/21025301 and https://bugs.webkit.org/show_bug.cgi?id=145417

Reviewed by Geoff Garen.

Source/WebCore:

No new tests (No change in behavior, testable at API level with a followup patch)

* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::setTriggeringAction): Apply the current flag value to the
  new NavigationAction so it can be used later.
* loader/DocumentLoader.h:
(WebCore::DocumentLoader::setShouldOpenExternalURLsPolicy):
(WebCore::DocumentLoader::shouldOpenExternalURLsPolicy):
(WebCore::DocumentLoader::setTriggeringAction): Deleted.

FrameLoadRequest should hold the new flag to pass into FrameLoader machinery:
* loader/FrameLoadRequest.h:
(WebCore::FrameLoadRequest::setShouldOpenExternalURLsPolicy):
(WebCore::FrameLoadRequest::shouldOpenExternalURLsPolicy):

* loader/FrameLoader.cpp:
(WebCore::FrameLoader::load): Set the flag on the newly created DocumentLoader but
  only if it is for the main frame.

* loader/FrameLoaderTypes.h: Add a new enum for the flag.

Navigation action should hold the new flag to pass to the policy delegate:
* loader/NavigationAction.cpp:
(WebCore::NavigationAction::NavigationAction):
* loader/NavigationAction.h:
(WebCore::NavigationAction::shouldOpenExternalURLsPolicy):
(WebCore::NavigationAction::setShouldOpenExternalURLsPolicy):

* loader/PolicyChecker.cpp:
(WebCore::PolicyChecker::checkNavigationPolicy): Include the flag from the DocumentLoader
  in the NavigationAction.

Source/WebKit2:

Rename "user initiated" to "should open external URLs":
* Shared/API/Cocoa/_WKNSURLRequestExtras.h:
* Shared/API/Cocoa/_WKNSURLRequestExtras.mm:
(-[NSURLRequest _web_shouldOpenExternalURLs]):
(-[NSMutableURLRequest _web_setShouldOpenExternalURLs:]):
(-[NSURLRequest _web_isUserInitiated]): Deleted.
(-[NSMutableURLRequest _web_setIsUserInitiated:]): Deleted.

Hold the new flag as a bool, and encode/decode it across IPC:
* Shared/NavigationActionData.cpp:
(WebKit::NavigationActionData::encode):
(WebKit::NavigationActionData::decode):
(WebKit::NavigationActionData::NavigationActionData): Deleted.
* Shared/NavigationActionData.h:

* UIProcess/API/APINavigationAction.h: Add accessor for the bool flag.

* UIProcess/API/C/WKPage.cpp:
(WKPageLoadURLWithUserData): Use "ShouldNotAllow" in the legacy C API for now.
(WKPageLoadURLRequestWithUserData): Ditto.

* UIProcess/API/Cocoa/WKBrowsingContextController.mm:
(-[WKBrowsingContextController loadRequest:userData:]): Set the value of the flag
  based on whether or not the initiating NSURLRequest was user initiated.

* UIProcess/API/Cocoa/WKNavigationAction.mm:
(-[WKNavigationAction _shouldOpenExternalURLs]): Add accessor for the bool flag.
* UIProcess/API/Cocoa/WKNavigationActionPrivate.h:

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView loadRequest:]): Set the value of the flag based on whether or not the
  initiating NSURLRequest was user initiated.

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::loadRequest): Pass the value down to the WebProcess to load.
(WebKit::WebPageProxy::loadFile): Ditto, but always with "ShouldNotAllow".
* UIProcess/WebPageProxy.h:

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction): Translate
  the NavigationAction flag to a bool in the NavigationActionData object.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::loadRequest): Set the value of the flag in the FrameLoadRequest that
  will be loaded.
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184941 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoPurge PassRefPtr in WebCore/Modules - 2
gyuyoung.kim@webkit.org [Thu, 28 May 2015 04:22:37 +0000 (04:22 +0000)]
Purge PassRefPtr in WebCore/Modules - 2
https://bugs.webkit.org/show_bug.cgi?id=145304

Reviewed by Darin Adler.

As a step to remove PassRefPtr, this patch replaces PassRefPtr with Ref or RefPtr.

No new tests, no behavior changes

* Modules/mediasource/SourceBufferList.h:
* Modules/mediastream/AllAudioCapabilities.h:
(WebCore::AllAudioCapabilities::create):
* Modules/mediastream/AllVideoCapabilities.h:
(WebCore::AllVideoCapabilities::create):
* Modules/mediastream/CapabilityRange.cpp:
(WebCore::CapabilityRange::create):
* Modules/mediastream/CapabilityRange.h:
* Modules/mediastream/MediaSourceStates.cpp:
(WebCore::MediaSourceStates::create):
* Modules/mediastream/MediaStream.cpp:
(WebCore::MediaStream::create):
* Modules/mediastream/MediaStreamTrack.cpp:
(WebCore::MediaStreamTrack::create):
* Modules/mediastream/MediaStreamTrackSourcesRequest.cpp:
(WebCore::MediaStreamTrackSourcesRequest::create):
* Modules/mediastream/MediaTrackConstraint.cpp:
(WebCore::MediaTrackConstraint::create):
* Modules/mediastream/MediaTrackConstraint.h:
* Modules/mediastream/RTCConfiguration.h:
(WebCore::RTCConfiguration::create):
* Modules/mediastream/RTCIceServer.h:
(WebCore::RTCIceServer::create):
* Modules/mediastream/RTCSessionDescription.cpp:
(WebCore::RTCSessionDescription::create):
* Modules/mediastream/RTCSessionDescription.h:
* Modules/mediastream/RTCSessionDescriptionRequestImpl.cpp:
(WebCore::RTCSessionDescriptionRequestImpl::create):
* Modules/mediastream/RTCStatsReport.cpp:
(WebCore::RTCStatsReport::create):
* Modules/mediastream/RTCStatsReport.h:
* Modules/mediastream/RTCStatsRequestImpl.cpp:
(WebCore::RTCStatsRequestImpl::create):
* Modules/mediastream/RTCStatsRequestImpl.h:
* Modules/mediastream/RTCStatsResponse.cpp:
(WebCore::RTCStatsResponse::create):
* Modules/mediastream/RTCStatsResponse.h:
* Modules/mediastream/RTCVoidRequestImpl.cpp:
(WebCore::RTCVoidRequestImpl::create):
* Modules/mediastream/RTCVoidRequestImpl.h:
* Modules/mediastream/SourceInfo.cpp:
(WebCore::SourceInfo::create):
* Modules/mediastream/SourceInfo.h:
* Modules/mediastream/UserMediaRequest.cpp:
(WebCore::UserMediaRequest::create):
* Modules/mediastream/UserMediaRequest.h:
* Modules/quota/StorageInfo.h:
(WebCore::StorageInfo::create):
* Modules/quota/StorageQuota.h:
(WebCore::StorageQuota::create):
* Modules/speech/SpeechSynthesis.cpp:
(WebCore::SpeechSynthesis::create):
* Modules/speech/SpeechSynthesis.h:
* Modules/speech/SpeechSynthesisUtterance.cpp:
(WebCore::SpeechSynthesisUtterance::create):
* Modules/speech/SpeechSynthesisUtterance.h:
* Modules/speech/SpeechSynthesisVoice.cpp:
(WebCore::SpeechSynthesisVoice::create):
* Modules/speech/SpeechSynthesisVoice.h:
* Modules/webaudio/ChannelMergerNode.cpp:
(WebCore::ChannelMergerNode::create):
* Modules/webaudio/ChannelMergerNode.h:
* Modules/webaudio/ChannelSplitterNode.cpp:
(WebCore::ChannelSplitterNode::create):
* Modules/webaudio/ChannelSplitterNode.h:
* Modules/webaudio/DefaultAudioDestinationNode.h:
(WebCore::DefaultAudioDestinationNode::create):
* Modules/webaudio/GainNode.h:
(WebCore::GainNode::create):
* Modules/webaudio/MediaElementAudioSourceNode.cpp:
(WebCore::MediaElementAudioSourceNode::create):
* Modules/webaudio/MediaElementAudioSourceNode.h:
* Modules/webaudio/MediaStreamAudioDestinationNode.cpp:
(WebCore::MediaStreamAudioDestinationNode::create):
* Modules/webaudio/MediaStreamAudioDestinationNode.h:
* Modules/webaudio/MediaStreamAudioSource.cpp:
(WebCore::MediaStreamAudioSource::create):
* Modules/webaudio/MediaStreamAudioSource.h:
* Modules/webaudio/MediaStreamAudioSourceNode.cpp:
(WebCore::MediaStreamAudioSourceNode::create):
* Modules/webaudio/MediaStreamAudioSourceNode.h:
* Modules/webaudio/OscillatorNode.cpp:
(WebCore::OscillatorNode::setType):
* Modules/webaudio/PannerNode.h:
(WebCore::PannerNode::create):
* Modules/webaudio/PeriodicWave.cpp:
(WebCore::PeriodicWave::create):
(WebCore::PeriodicWave::createSine):
(WebCore::PeriodicWave::createSquare):
(WebCore::PeriodicWave::createSawtooth):
(WebCore::PeriodicWave::createTriangle):
* Modules/webaudio/PeriodicWave.h:
* Modules/webaudio/ScriptProcessorNode.cpp:
(WebCore::ScriptProcessorNode::create):
* Modules/webaudio/ScriptProcessorNode.h:
* Modules/webaudio/WaveShaperNode.h:
(WebCore::WaveShaperNode::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184940 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: Show Page Source does not switch to Resources tab if inspector was...
commit-queue@webkit.org [Thu, 28 May 2015 01:55:25 +0000 (01:55 +0000)]
Web Inspector: Show Page Source does not switch to Resources tab if inspector was closed
https://bugs.webkit.org/show_bug.cgi?id=145423

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2015-05-27
Reviewed by Timothy Hatcher.

* UserInterface/Base/Main.js:
(WebInspector._frameWasAdded.delayedWork):
(WebInspector._frameWasAdded):
In the special case where we are holding to show a frame's source code when
the frame becomes available (_frameIdentifierToShowSourceCodeWhenAvailable)
we should be forcing to show the tab.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184935 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: Toggling an inherited property sometimes stops working
commit-queue@webkit.org [Thu, 28 May 2015 01:51:41 +0000 (01:51 +0000)]
Web Inspector: Toggling an inherited property sometimes stops working
https://bugs.webkit.org/show_bug.cgi?id=145416

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2015-05-27
Reviewed by Timothy Hatcher.

* UserInterface/Models/DOMNodeStyles.js:
(WebInspector.DOMNodeStyles.prototype._parseStyleDeclarationPayload):
Ensure we update the existing model objects with the new state information
before possibly bailing. This may produce changed events that update
parts of the inspector appropriately.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184934 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[JSC] Add undefined->double conversion to DoubleRep
benjamin@webkit.org [Thu, 28 May 2015 01:30:58 +0000 (01:30 +0000)]
[JSC] Add undefined->double conversion to DoubleRep
https://bugs.webkit.org/show_bug.cgi?id=145293

Patch by Benjamin Poulain <bpoulain@apple.com> on 2015-05-27
Reviewed by Filip Pizlo.

Source/JavaScriptCore:

This patch adds undefined to double conversion to the DoubleRep
node for the cases were we speculate "undefined" as part of the types
processed.

The use case is doing math with accidental out-of-bounds access. For example,
something like:
    for (var i = 0; i <= length; ++i)
        ouptput += array[i];

would cause us to OSR exit every time i === length.

When hitting one of those cases, we would already speculate double math,
but the DoubleRep node was unable to convert the undefined and would exit.

With this patch the use kind NotCellUse cover this conversion for DoubleRep.
I have been quite conservative so in general we will not find "undefined"
until a few recompile but being optimistic seems better since this is a corner case.

This patch is a 80% progression on WebXPRT's DNA Sequencing test.

* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::injectTypeConversionsForEdge):
* dfg/DFGNode.h:
(JSC::DFG::Node::sawUndefined):
* dfg/DFGPredictionPropagationPhase.cpp:
(JSC::DFG::PredictionPropagationPhase::propagate):
* dfg/DFGSafeToExecute.h:
(JSC::DFG::SafeToExecuteEdge::operator()):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileDoubleRep):
* dfg/DFGUseKind.cpp:
(WTF::printInternal):
* dfg/DFGUseKind.h:
(JSC::DFG::typeFilterFor):
* ftl/FTLCapabilities.cpp:
(JSC::FTL::canCompile):
* ftl/FTLLowerDFGToLLVM.cpp:
(JSC::FTL::LowerDFGToLLVM::compileDoubleRep):
(JSC::FTL::LowerDFGToLLVM::jsValueToDouble):
* tests/stress/double-rep-with-undefined.js: Added.
(addArgsNumberAndUndefined):
(addArgsInt32AndUndefined):
(testFallbackWithDouble):
(addArgsDoubleAndUndefined):
(testFallbackWithObject.):
(testFallbackWithObject):
(addArgsOnlyUndefined):
(testFallbackWithString):

LayoutTests:

* js/regress/math-with-out-of-bounds-array-values-expected.txt: Added.
* js/regress/math-with-out-of-bounds-array-values.html: Added.
* js/regress/script-tests/math-with-out-of-bounds-array-values.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184933 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoREGRESSION (r183820): webkit.org/blog/ background painting issue on reload, when...
simon.fraser@apple.com [Thu, 28 May 2015 00:39:35 +0000 (00:39 +0000)]
REGRESSION (r183820): webkit.org/blog/ background painting issue on reload, when the page contains videos
https://bugs.webkit.org/show_bug.cgi?id=145420

Reviewed by Dean Jackson.
Source/WebCore:

After r183820, the media controls no longer had a wrapper that created CSS stacking context.
The media controls on Mac use mix-blend-mode, which causes the compositing code to look for
a stacking context ancestor and make it composited. After this change, it would walk up
to a layer outside of the media element (e.g. the document element's layer), and make
that composited. This triggered bugs with root background painting.

Prevent mix-blend-mode affecting content outside the media elements by having the media element's
layer act as a stacking context.

Test: media/controls-layers.html

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::RenderLayer):
* rendering/RenderLayer.h:

LayoutTests:

Test that dumps compositing layers in a document with media controls.

* media/controls-layers.html: Added.
* platform/mac/media/controls-layers-expected.txt: Added.
* platform/mac-mavericks/media/controls-layers-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184932 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoCan't load local files in WKWebView from containerized app
andersca@apple.com [Thu, 28 May 2015 00:34:39 +0000 (00:34 +0000)]
Can't load local files in WKWebView from containerized app
https://bugs.webkit.org/show_bug.cgi?id=145424
rdar://problem/20831176

Reviewed by Dan Bernstein.

Allow the network process to load local files from its bundle as long as it has
the sandbox extensions to do so.

This was originally added to only allow SSO to read the Info.plist inside its own application
bundle, but being able to read the application bundle from within the network process should be OK.

* Resources/SandboxProfiles/ios/com.apple.WebKit.Networking.sb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184931 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[WK2] Local storage areas should get torn down when they have no remaining references.
akling@apple.com [Thu, 28 May 2015 00:30:59 +0000 (00:30 +0000)]
[WK2] Local storage areas should get torn down when they have no remaining references.
<https://webkit.org/b/143339>
<rdar://problem/20156436>

Reviewed by Darin Adler.

Source/WebCore:

Add StorageArea::securityOrigin() implementations.

* loader/EmptyClients.cpp:
* storage/StorageArea.h:

Source/WebKit:

Add StorageArea::securityOrigin() implementation.

* Storage/StorageAreaImpl.h:

Source/WebKit2:

Let StorageNamespaceImpl own its StorageAreaMaps weakly instead of through RefPtr.
Ownership is flipped so that StorageAreaMap refs the StorageNamespaceImpl instead.
This allows the StorageAreaMaps to get destroyed once all of its clients are gone.

Practically speaking, this means that the garbage collector now decides when local
storage databases can be closed, instead of us keeping them open for the lifetime
of the web process.

For session storage, it works a bit differently. In the web process, they get torn
down when their last client disappears, but they stay alive in the UI process.
If/when the web process asks the UI process to open session storage for an origin,
the UI process checks if one already exists, and if so, just updates the ID of the
old storage with the new one provided by the web process.

* UIProcess/Storage/StorageManager.cpp:
(WebKit::StorageManager::StorageArea::isSessionStorage):
(WebKit::StorageManager::createTransientLocalStorageMap):
(WebKit::StorageManager::createSessionStorageMap):
(WebKit::StorageManager::destroyStorageMap):
* WebProcess/Storage/StorageAreaImpl.cpp:
(WebKit::StorageAreaImpl::securityOrigin):
* WebProcess/Storage/StorageAreaImpl.h:
* WebProcess/Storage/StorageAreaMap.cpp:
(WebKit::StorageAreaMap::StorageAreaMap):
(WebKit::StorageAreaMap::~StorageAreaMap):
* WebProcess/Storage/StorageAreaMap.h:
(WebKit::StorageAreaMap::securityOrigin):
* WebProcess/Storage/StorageNamespaceImpl.cpp:
(WebKit::StorageNamespaceImpl::didDestroyStorageAreaMap):
(WebKit::StorageNamespaceImpl::storageArea):
* WebProcess/Storage/StorageNamespaceImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184930 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoDisable network cache for old clients
antti@apple.com [Thu, 28 May 2015 00:25:52 +0000 (00:25 +0000)]
Disable network cache for old clients
https://bugs.webkit.org/show_bug.cgi?id=145418
rdar://problem/21126587

Reviewed by Andy Estes.

Old clients might use NSURLCache API to clear the cache. New cache requires use of new APIs.

* UIProcess/Cocoa/VersionChecks.h: Added.
* UIProcess/Cocoa/VersionChecks.mm: Added.

    Add linked-on-or-after check mechanism similar to WebKit1.

(WebKit::linkTimeVersion):
(WebKit::linkedOnOrAfter):
* UIProcess/Cocoa/WebProcessPoolCocoa.mm:
(WebKit::WebProcessPool::platformInitializeNetworkProcess):

    Use it when deciding whether to enable the cache.

* WebKit2.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184929 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoimg.currentSrc problem in strict mode with old picturefill
dino@apple.com [Thu, 28 May 2015 00:16:47 +0000 (00:16 +0000)]
img.currentSrc problem in strict mode with old picturefill
https://bugs.webkit.org/show_bug.cgi?id=144095
<rdar://problem/21087013>

Reviewed by Simon Fraser.

.:

Add a PICTURE_SIZES flag.

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

Source/JavaScriptCore:

Add a PICTURE_SIZES flag.

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

Add a PICTURE_SIZES flag. This allows us to easily remove the currentSrc
attribute on HTMLImageElement, which is causing some issues with
<picture> polyfills.

* Configurations/FeatureDefines.xcconfig:
* html/HTMLImageElement.cpp:
(WebCore::HTMLImageElement::setBestFitURLAndDPRFromImageCandidate): Add guard around use of m_currentSrc.
* html/HTMLImageElement.h:
* html/HTMLImageElement.idl: Guard the currentSrc attribute.
* html/parser/HTMLPreloadScanner.cpp:
(WebCore::TokenPreloadScanner::StartTagScanner::processAttributes):
(WebCore::TokenPreloadScanner::StartTagScanner::processAttribute):

Source/WebKit/mac:

Add a PICTURE_SIZES flag.

* Configurations/FeatureDefines.xcconfig:

Source/WebKit2:

Add a PICTURE_SIZES flag.

* Configurations/FeatureDefines.xcconfig:

Source/WTF:

Add a PICTURE_SIZES flag.

* wtf/FeatureDefines.h:

Tools:

Add a PICTURE_SIZES flag.

* Scripts/webkitperl/FeatureList.pm:

WebKitLibraries:

Add a PICTURE_SIZES flag.

* win/tools/vsprops/FeatureDefines.props:
* win/tools/vsprops/FeatureDefinesCairo.props:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184928 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoLazyNode comparison can return incorrect results when comparing an empty value
basile_clement@apple.com [Wed, 27 May 2015 23:47:40 +0000 (23:47 +0000)]
LazyNode comparison can return incorrect results when comparing an empty value
https://bugs.webkit.org/show_bug.cgi?id=145421

Reviewed by Geoffrey Garen.

When comparing a LazyNode to another, we compare the value pointers if
we have one, and otherwise compare the nodes.
We should be comparing value pointers if the other LazyNode has one as
well, otherwise we risk an incoherency when we are a empty LazyNode
being compared to a FrozenValue without node.

Note that this is not a problem in any other case because if we don't
have a FrozenValue and we are not an empty LazyNode, we are a
non-constant node, and comparing the node pointers is correct.

* dfg/DFGLazyNode.h:
(JSC::DFG::LazyNode::operator==):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184927 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoREGRESSION: These sorting idioms used by Peacekeeper and Browsermark are ~20X slower
ggaren@apple.com [Wed, 27 May 2015 23:43:22 +0000 (23:43 +0000)]
REGRESSION: These sorting idioms used by Peacekeeper and Browsermark are ~20X slower
https://bugs.webkit.org/show_bug.cgi?id=145412

Reviewed by Benjamin Poulain.

Cache strings when doing a string-converting sort.

This is a 21% speedup.

* builtins/Array.prototype.js:
(sort.stringComparator): Use subtraction instead of branching because
it's slightly faster.

(sort.comparatorSort):
(sort.stringSort):
(sort): Add a special case for string sorting to avoid redundant string
conversion.

* parser/Parser.cpp:
(JSC::Parser<LexerType>::createBindingPattern): Names can be empty if
they are private names.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184926 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: REGRESSION(r179286) Editing Style Metrics Values no longer works
commit-queue@webkit.org [Wed, 27 May 2015 23:21:58 +0000 (23:21 +0000)]
Web Inspector: REGRESSION(r179286) Editing Style Metrics Values no longer works
https://bugs.webkit.org/show_bug.cgi?id=143164

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2015-05-27
Reviewed by Brian Burg.

* UserInterface/Views/BoxModelDetailsSectionRow.js:
(WebInspector.BoxModelDetailsSectionRow.prototype._updateMetrics.createElement):
(WebInspector.BoxModelDetailsSectionRow.prototype._applyUserInput.resolvedNode.toggleInlineStyleProperty):
(WebInspector.BoxModelDetailsSectionRow.prototype._applyUserInput.resolvedNode.didToggle):
(WebInspector.BoxModelDetailsSectionRow.prototype._applyUserInput.resolvedNode):
(WebInspector.BoxModelDetailsSectionRow.prototype._applyUserInput):
Make it so editing in the metrics pane just evaluates elem.style.setProperty
on the inspected page. Use "!important" to try and give the maximum priority
possible, which is a change from older behavior. Finally, refresh the sidebar
to update all values, and update the UI if bad input didn't change styles.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184925 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAttempt to fix internal build following <http://trac.webkit.org/changeset/184760>
dbates@webkit.org [Wed, 27 May 2015 22:11:27 +0000 (22:11 +0000)]
Attempt to fix internal build following <trac.webkit.org/changeset/184760>
(https://bugs.webkit.org/show_bug.cgi?id=145289)

Reviewed by Jer Noble.

* platform/spi/mac/AVFoundationSPI.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184924 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAssertion hit in WebPage::didChangeSelection()
cdumez@apple.com [Wed, 27 May 2015 21:51:22 +0000 (21:51 +0000)]
Assertion hit in WebPage::didChangeSelection()
https://bugs.webkit.org/show_bug.cgi?id=145413
<rdar://problem/21001129>

Reviewed by Ryosuke Niwa.

We sometimes hit the "ASSERT(layoutCount == view->layoutCount())"
assertion in WebPage::didChangeSelection(). We manage to prevent
synchronous layouts in most cases when calling editorState(). However,
it seems it can still happen in some cases. Crashing in this case seems
overkill and can be annoying to other developers, especially
considering the implications are only on performance.

This patch drops the assertion.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::didChangeSelection): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184923 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRefactor WebKit1 specific threading code out of WebVideoFullscreen code that is share...
commit-queue@webkit.org [Wed, 27 May 2015 21:41:02 +0000 (21:41 +0000)]
Refactor WebKit1 specific threading code out of WebVideoFullscreen code that is shared with WebKit2.
https://bugs.webkit.org/show_bug.cgi?id=143954

Patch by Jeremy Jones <jeremyj@apple.com> on 2015-05-27
Reviewed by Darin Adler.

WebVideoFullscreenInterfaceAVKit is used in both WebKit1 and WebKit2. In WebKit1, the model runs in the WebThread, while
the interface is on the main thread. So there is code to dispatch between these two thread when communicating between the
interface and the model. In WebKit2, this is handled automatically by the IPC mechanism. As a result, the threading code
in WebVideoFullscreenbInterfaceAVKit and in WebVideoFullscreenModelVideoElement is redundant in WebKit2 and relies on
WebThreadRun being a no-op in WebKit2.

This change clarifies this and simplifies the WebKit2 path by moving the WebKit1 specific threading code into
WebVideoFullscreenControllerContext, which is the WebKit1 specific controller.

* platform/ios/WebVideoFullscreenControllerAVKit.mm:
(WebVideoFullscreenControllerContext::WebVideoFullscreenControllerContext):
(WebVideoFullscreenControllerContext::setController):
(WebVideoFullscreenControllerContext::didSetupFullscreen):
(WebVideoFullscreenControllerContext::didExitFullscreen):
(WebVideoFullscreenControllerContext::didCleanupFullscreen):
(WebVideoFullscreenControllerContext::fullscreenMayReturnToInline):
(WebVideoFullscreenControllerContext::resetMediaState):
(WebVideoFullscreenControllerContext::setDuration):
(WebVideoFullscreenControllerContext::setCurrentTime):
(WebVideoFullscreenControllerContext::setBufferedTime):
(WebVideoFullscreenControllerContext::setRate):
(WebVideoFullscreenControllerContext::setVideoDimensions):
(WebVideoFullscreenControllerContext::setSeekableRanges):
(WebVideoFullscreenControllerContext::setCanPlayFastReverse):
(WebVideoFullscreenControllerContext::setAudioMediaSelectionOptions):
(WebVideoFullscreenControllerContext::setLegibleMediaSelectionOptions):
(WebVideoFullscreenControllerContext::setExternalPlayback):
(WebVideoFullscreenControllerContext::play):
(WebVideoFullscreenControllerContext::pause):
(WebVideoFullscreenControllerContext::togglePlayState):
(WebVideoFullscreenControllerContext::beginScrubbing):
(WebVideoFullscreenControllerContext::endScrubbing):
(WebVideoFullscreenControllerContext::seekToTime):
(WebVideoFullscreenControllerContext::fastSeek):
(WebVideoFullscreenControllerContext::beginScanningForward):
(WebVideoFullscreenControllerContext::beginScanningBackward):
(WebVideoFullscreenControllerContext::endScanning):
(WebVideoFullscreenControllerContext::requestExitFullscreen):
(WebVideoFullscreenControllerContext::setVideoLayerFrame):
(WebVideoFullscreenControllerContext::setVideoLayerGravity):
(WebVideoFullscreenControllerContext::selectAudioMediaOption):
(WebVideoFullscreenControllerContext::selectLegibleMediaOption):
(WebVideoFullscreenControllerContext::fullscreenModeChanged):
(WebVideoFullscreenControllerContext::setupFullscreen):
(WebVideoFullscreenControllerContext::exitFullscreen):
(WebVideoFullscreenControllerContext::requestHideAndExitFullscreen):
(-[WebVideoFullscreenController init]):
(-[WebVideoFullscreenController dealloc]):
(-[WebVideoFullscreenController enterFullscreen:mode:]):
(-[WebVideoFullscreenController exitFullscreen]):
(-[WebVideoFullscreenController requestHideAndExitFullscreen]):
(-[WebVideoFullscreenController didFinishFullscreen:]):
(WebVideoFullscreenControllerChangeObserver::setTarget): Deleted.
(-[WebVideoFullscreenController didSetupFullscreen]): Deleted.
(-[WebVideoFullscreenController didEnterFullscreen]): Deleted.
(-[WebVideoFullscreenController didExitFullscreen]): Deleted.
(-[WebVideoFullscreenController didCleanupFullscreen]): Deleted.
(-[WebVideoFullscreenController fullscreenMayReturnToInline]): Deleted.
* platform/ios/WebVideoFullscreenInterfaceAVKit.h:
* platform/ios/WebVideoFullscreenInterfaceAVKit.mm:
(-[WebAVVideoLayer setBounds:]):
(-[WebAVVideoLayer resolveBounds]):
(WebVideoFullscreenInterfaceAVKit::resetMediaState):
(WebVideoFullscreenInterfaceAVKit::setDuration):
(WebVideoFullscreenInterfaceAVKit::setCurrentTime):
(WebVideoFullscreenInterfaceAVKit::setBufferedTime):
(WebVideoFullscreenInterfaceAVKit::setRate):
(WebVideoFullscreenInterfaceAVKit::setVideoDimensions):
(WebVideoFullscreenInterfaceAVKit::setSeekableRanges):
(WebVideoFullscreenInterfaceAVKit::setCanPlayFastReverse):
(WebVideoFullscreenInterfaceAVKit::setAudioMediaSelectionOptions):
(WebVideoFullscreenInterfaceAVKit::setLegibleMediaSelectionOptions):
(WebVideoFullscreenInterfaceAVKit::setExternalPlayback):
(WebVideoFullscreenInterfaceAVKit::setupFullscreen):
(WebVideoFullscreenInterfaceAVKit::enterFullscreen):
(WebVideoFullscreenInterfaceAVKit::enterFullscreenStandard):
(WebVideoFullscreenInterfaceAVKit::exitFullscreen):
(WebVideoFullscreenInterfaceAVKit::exitFullscreenInternal):
(WebVideoFullscreenInterfaceAVKit::cleanupFullscreen):
(WebVideoFullscreenInterfaceAVKit::cleanupFullscreenInternal):
(WebVideoFullscreenInterfaceAVKit::requestHideAndExitFullscreen):
(WebVideoFullscreenInterfaceAVKit::preparedToReturnToInline):
(WebVideoFullscreenInterfaceAVKit::willStartOptimizedFullscreen):
(WebVideoFullscreenInterfaceAVKit::didStartOptimizedFullscreen):
(WebVideoFullscreenInterfaceAVKit::willStopOptimizedFullscreen):
(WebVideoFullscreenInterfaceAVKit::didStopOptimizedFullscreen):
(WebVideoFullscreenInterfaceAVKit::willCancelOptimizedFullscreen):
(WebVideoFullscreenInterfaceAVKit::didCancelOptimizedFullscreen):
(WebVideoFullscreenInterfaceAVKit::setVideoLayerFrame):
(WebVideoFullscreenInterfaceAVKit::setupFullscreenInternal): Deleted.
* platform/ios/WebVideoFullscreenModel.h:
* platform/ios/WebVideoFullscreenModelVideoElement.h:
* platform/ios/WebVideoFullscreenModelVideoElement.mm:
(WebVideoFullscreenModelVideoElement::setVideoFullscreenLayer):
(WebVideoFullscreenModelVideoElement::play):
(WebVideoFullscreenModelVideoElement::pause):
(WebVideoFullscreenModelVideoElement::togglePlayState):
(WebVideoFullscreenModelVideoElement::beginScrubbing):
(WebVideoFullscreenModelVideoElement::endScrubbing):
(WebVideoFullscreenModelVideoElement::seekToTime):
(WebVideoFullscreenModelVideoElement::fastSeek):
(WebVideoFullscreenModelVideoElement::beginScanningForward):
(WebVideoFullscreenModelVideoElement::beginScanningBackward):
(WebVideoFullscreenModelVideoElement::endScanning):
(WebVideoFullscreenModelVideoElement::requestExitFullscreen):
(WebVideoFullscreenModelVideoElement::fullscreenModeChanged):
(WebVideoFullscreenModelVideoElement::videoLayerFrame): Deleted.
(WebVideoFullscreenModelVideoElement::videoLayerGravity): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184922 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoJIT-generated store barrier code should assume the buffer pointer and capacity to...
fpizlo@apple.com [Wed, 27 May 2015 21:29:09 +0000 (21:29 +0000)]
JIT-generated store barrier code should assume the buffer pointer and capacity to be compile-time constants
https://bugs.webkit.org/show_bug.cgi?id=145404

Reviewed by Andreas Kling.

We never change the capacity of a write barrier buffer. We never repoint the buffer
pointer. So, the JIT shouldn't load those from memory; it should take advantage of the
fact that these are compile-time constants.

* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::storeToWriteBarrierBuffer):
* ftl/FTLLowerDFGToLLVM.cpp:
(JSC::FTL::LowerDFGToLLVM::emitStoreBarrier):
* heap/WriteBarrierBuffer.h:
(JSC::WriteBarrierBuffer::currentIndexAddress):
(JSC::WriteBarrierBuffer::capacity):
(JSC::WriteBarrierBuffer::buffer):
(JSC::WriteBarrierBuffer::currentIndexOffset): Deleted.
(JSC::WriteBarrierBuffer::capacityOffset): Deleted.
(JSC::WriteBarrierBuffer::bufferOffset): Deleted.
* jit/Repatch.cpp:
(JSC::emitPutTransitionStubAndGetOldStructure):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184920 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUpdate Mac WebKit1 TestExpectations for platform/mac/fast/ruby/ruby-expansion-cjk...
ap@apple.com [Wed, 27 May 2015 21:13:16 +0000 (21:13 +0000)]
Update Mac WebKit1 TestExpectations for platform/mac/fast/ruby/ruby-expansion-cjk-2.html

* platform/mac-wk1/TestExpectations: It's an image failure, so counter-intuitively,
Failure expectation is not the right one.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184918 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoREGRESSION: These sorting idioms used by Peacekeeper and Browsermark are ~20X slower
ggaren@apple.com [Wed, 27 May 2015 20:42:00 +0000 (20:42 +0000)]
REGRESSION: These sorting idioms used by Peacekeeper and Browsermark are ~20X slower
https://bugs.webkit.org/show_bug.cgi?id=145412

Reviewed by Darin Adler.

Use @toString instead of the String constructor because calls to the
String constructor are never optimized. (See
https://bugs.webkit.org/show_bug.cgi?id=144458.)

This is a ~2X speedup.

* builtins/Array.prototype.js:
(sort.stringComparator):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184917 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Mac] short-circuit MIME type lookup when possible
eric.carlson@apple.com [Wed, 27 May 2015 20:33:09 +0000 (20:33 +0000)]
[Mac] short-circuit MIME type lookup when possible
https://bugs.webkit.org/show_bug.cgi?id=145362

Reviewed by Jer Noble.

* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::isUnsupportedMIMEType): Renamed from unsupportedMIMEType. Convert type to lower case
once instead of calling equalIgnoringCase many times.
(WebCore::staticMIMETypeList): Renamed from staticMimeTypeCache. Initialize static set in a lambda,
using an array of C strings in a loop.
(WebCore::avfMIMETypes): Renamed from avfMimeTypeCache. Initialize static set in a lambda.
(WebCore::MediaPlayerPrivateAVFoundationObjC::getSupportedTypes): avfMimeTypeCache -> avfMIMETypes.
(WebCore::keySystemIsSupported): equalIgnoringCase -> equalIgnoringASCIICase.
(WebCore::MediaPlayerPrivateAVFoundationObjC::supportsType): Ditto.
(WebCore::MediaPlayerPrivateAVFoundationObjC::supportsKeySystem): unsupportedMIMEType ->
isUnsupportedMIMEType, equalIgnoringCase -> equalIgnoringASCIICase, staticMimeTypeCache ->
staticMIMETypeList, avfMimeTypeCache -> avfMIMETypes.
(WebCore::unsupportedMIMEType): Deleted.
(WebCore::staticMimeTypeCache): Deleted.
(WebCore::avfMimeTypeCache): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184916 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUpdate Mac WebKit1 TestExpectations for platform/mac/fast/ruby/ruby-expansion-cjk...
dbates@webkit.org [Wed, 27 May 2015 20:14:05 +0000 (20:14 +0000)]
Update Mac WebKit1 TestExpectations for platform/mac/fast/ruby/ruby-expansion-cjk-2.html

* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184915 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoFix Mac Yosemite layout test failure following <http://trac.webkit.org/changeset...
dbates@webkit.org [Wed, 27 May 2015 19:17:07 +0000 (19:17 +0000)]
Fix Mac Yosemite layout test failure following <trac.webkit.org/changeset/184899>
(https://bugs.webkit.org/show_bug.cgi?id=145336)

Mark test platform/mac/fast/ruby/ruby-expansion-cjk-2.html as failing on Yosemite.

Also remove references to fast/ruby/ruby-expansion-cjk*.html from TestExpectations files
of other ports since we consider these tests specific to Mac.

* platform/efl/TestExpectations:
* platform/gtk/TestExpectations:
* platform/ios-simulator/TestExpectations:
* platform/mac-wk1/TestExpectations:
* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184914 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAddress a follow-up review comment from Darin.
andersca@apple.com [Wed, 27 May 2015 19:00:22 +0000 (19:00 +0000)]
Address a follow-up review comment from Darin.

* UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::removeMediaKeys):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184913 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWKWebsiteDataStore should handle media keys
andersca@apple.com [Wed, 27 May 2015 18:53:49 +0000 (18:53 +0000)]
WKWebsiteDataStore should handle media keys
https://bugs.webkit.org/show_bug.cgi?id=145394
rdar://problem/20617794.

Reviewed by Darin Adler.

* Shared/WebsiteData/WebsiteDataTypes.h:
Add WebsiteDataTypeMediaKeys enum.

* UIProcess/API/Cocoa/APIWebsiteDataStoreCocoa.mm:
(API::WebsiteDataStore::defaultDataStoreConfiguration):
Set up mediaKeysStorageDirectory.

* UIProcess/API/Cocoa/WKWebsiteDataRecord.mm:
(dataTypesToString):
* UIProcess/API/Cocoa/WKWebsiteDataRecordInternal.h:
(WebKit::toWebsiteDataTypes):
(WebKit::toWKWebsiteDataTypes):
* UIProcess/API/Cocoa/WKWebsiteDataRecordPrivate.h: Added.
Add _WKWebsiteDataTypeMediaKeys.

* UIProcess/WebProcessPool.cpp:
(WebKit::legacyWebsiteDataStoreConfiguration):
Set up mediaKeysStorageDirectory.

* UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::WebsiteDataStore):
Set m_mediaKeysStorageDirectory.

(WebKit::WebsiteDataStore::fetchData):
Make CallbackAggregator ThreadSafeRefCounted.
Handle WebsiteDataTypeMediaKeys by calling mediaKeyOrigins on our background queue.

(WebKit::WebsiteDataStore::removeData):
Make CallbackAggregator ThreadSafeRefCounted.
Handle WebsiteDataTypeMediaKeys by calling removeMediaKeys on our background queue.

(WebKit::WebsiteDataStore::mediaKeyOrigins):
Compute the origins.

(WebKit::computeMediaKeyFile):
Add helper function that returns the media key given a containing directory.

(WebKit::WebsiteDataStore::removeMediaKeys):
Remove media key files.

* UIProcess/WebsiteData/WebsiteDataStore.h:
Add new members.

* WebKit2.xcodeproj/project.pbxproj:
Add new files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184912 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Mac] short-circuit MIME type lookup when possible
eric.carlson@apple.com [Wed, 27 May 2015 18:37:28 +0000 (18:37 +0000)]
[Mac] short-circuit MIME type lookup when possible
https://bugs.webkit.org/show_bug.cgi?id=145362

Reviewed by Dean Jackson.

* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::unsupportedMIMEType): New, reject types known to not be supported.
(WebCore::staticMimeTypeCache): Accept MIME types known to be supported.
(WebCore::avfMimeTypeCache): Renamed from mimeTypeCache.
(WebCore::MediaPlayerPrivateAVFoundationObjC::getSupportedTypes): mimeTypeCache -> staticMimeTypeCache.
(WebCore::MediaPlayerPrivateAVFoundationObjC::supportsType): Return immediately if
unsupportedMIMEType returns true, don't call AVFoundation if staticMimeTypeCache returns true.
(WebCore::MediaPlayerPrivateAVFoundationObjC::supportsKeySystem): Ditto.
(WebCore::mimeTypeCache): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184910 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[Mac] occasional crash in MediaPlayerPrivateAVFoundationObjC::didStopLoadingRequest
eric.carlson@apple.com [Wed, 27 May 2015 18:31:12 +0000 (18:31 +0000)]
[Mac] occasional crash in MediaPlayerPrivateAVFoundationObjC::didStopLoadingRequest
https://bugs.webkit.org/show_bug.cgi?id=145409

Reviewed by Dean Jackson.

* platform/graphics/avfoundation/objc/WebCoreAVFResourceLoader.mm:
(WebCore::WebCoreAVFResourceLoader::stopLoading): NULL-check m_avRequest.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184909 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoBackdrop filters don't animate
dino@apple.com [Wed, 27 May 2015 18:15:08 +0000 (18:15 +0000)]
Backdrop filters don't animate
https://bugs.webkit.org/show_bug.cgi?id=145386
<rdar://problem/21110037>

Reviewed by Simon Fraser.

Source/WebCore:

Add support for animation of backdrop filters.

Note that, at the moment, we can only animate/transition
backdrop-filter if it is already present on the element. See
https://bugs.webkit.org/show_bug.cgi?id=145107

Test: css3/filters/backdrop/animation.html

* page/animation/AnimationBase.h: Add m_backdropFilterFunctionListsMatch and backdropFilterFunctionListsMatch.
(WebCore::AnimationBase::backdropFilterFunctionListsMatch):

* page/animation/CSSPropertyAnimation.cpp:
(WebCore::blendFunc): backdrop-filter is also supported.
(WebCore::PropertyWrapperAcceleratedBackdropFilter::PropertyWrapperAcceleratedBackdropFilter): Added. Works
similarly to the PropertyWrapperAcceleratedFilter.
(WebCore::PropertyWrapperAcceleratedBackdropFilter::animationIsAccelerated):
(WebCore::PropertyWrapperAcceleratedBackdropFilter::blend):
(WebCore::CSSPropertyAnimationWrapperMap::CSSPropertyAnimationWrapperMap): Construct the
PropertyWrapperAcceleratedBackdropFilter.

* page/animation/ImplicitAnimation.cpp:
(WebCore::ImplicitAnimation::reset): Call checkForMatchingBackdropFilterFunctionLists.
(WebCore::ImplicitAnimation::validateTransformFunctionList): Fix typo.
(WebCore::ImplicitAnimation::checkForMatchingFilterFunctionLists): Remove whitespace.
(WebCore::ImplicitAnimation::checkForMatchingBackdropFilterFunctionLists): New method that
checks if the individual filters in a filter list match.
* page/animation/ImplicitAnimation.h: Add checkForMatchingBackdropFilterFunctionLists.

* page/animation/KeyframeAnimation.cpp:
(WebCore::KeyframeAnimation::KeyframeAnimation): Call checkForMatchingBackdropFilterFunctionLists.
(WebCore::KeyframeAnimation::checkForMatchingBackdropFilterFunctionLists): Copied from
checkForMatchingFilterFunctionLists, but calls backdropFilter() instead of filter().
(WebCore::KeyframeAnimation::checkForMatchingFilterFunctionLists): This was accidentally checking
for backdrop filters as well, but it wouldn't have worked.
* page/animation/KeyframeAnimation.h: Add checkForMatchingBackdropFilterFunctionLists.

* platform/graphics/GraphicsLayer.cpp:
(WebCore::GraphicsLayer::validateFilterOperations): Modify the ASSERT to allow AnimatedPropertyWebkitBackdropFilter.
* platform/graphics/GraphicsLayerClient.h: Add AnimatedPropertyWebkitBackdropFilter to the enum.

* platform/graphics/ca/GraphicsLayerCA.cpp:
(WebCore::propertyIdToString): Support new enum.
(WebCore::GraphicsLayerCA::addAnimation): Support AnimatedPropertyWebkitBackdropFilter.
(WebCore::GraphicsLayerCA::createFilterAnimationsFromKeyframes): Ditto.
(WebCore::GraphicsLayerCA::animatedLayer): Use a switch statement now that we have more than
two options, and handle AnimatedPropertyWebkitBackdropFilter.
(WebCore::GraphicsLayerCA::updateAnimations): Deleted a blank line.

* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::startAnimation): Support animation of backdrop-filter.
(WebCore::RenderLayerBacking::startTransition): Ditto. Copied the code from the filter transition.
(WebCore::RenderLayerBacking::graphicsLayerToCSSProperty):
(WebCore::RenderLayerBacking::cssToGraphicsLayerProperty):
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::requiresCompositingForAnimation):

LayoutTests:

Add a test for animation of backdrop-filter, and do some
minor cleanups in related files.

* animations/resources/animation-test-helpers.js:
(parseFilterImage): Fix a typo.
(getPropertyValue): Support webkitBackdropFilter.
(comparePropertyValue): Ditto.
* css3/filters/backdrop/animation-expected.txt: Added.
* css3/filters/backdrop/animation.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184908 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoHandle case where -startOptimizedFullscreen fails.
commit-queue@webkit.org [Wed, 27 May 2015 17:12:14 +0000 (17:12 +0000)]
Handle case where -startOptimizedFullscreen fails.
https://bugs.webkit.org/show_bug.cgi?id=145340

Patch by Jeremy Jones <jeremyj@apple.com> on 2015-05-27
Reviewed by Eric Carlson.

Add new AVPlayerViewController delegate methods and remove old ones.
-startOptimizedFullscreen can either fail silently because it is not allowed or it will call a delegate.
We prevent silent failure by testing preconditions in enterFullscreenOptimized, and add the
delegate -playerViewControllerFailedToStartOptimizedFullscreen to handle explicitly failure.

* platform/ios/WebVideoFullscreenInterfaceAVKit.h: added/removed method declarations.
* platform/ios/WebVideoFullscreenInterfaceAVKit.mm:
(-[WebAVPlayerController playerViewControllerFailedToStartOptimizedFullscreen:withError:]): Added.
(WebVideoFullscreenInterfaceAVKit::enterFullscreenOptimized): Handle silent failure.
(WebVideoFullscreenInterfaceAVKit::didStartOptimizedFullscreen): Hide window on main thread.
(WebVideoFullscreenInterfaceAVKit::failedToStartOptimizedFullscreen): Added
(-[WebAVPlayerController playerViewControllerWillCancelOptimizedFullscreen:]): Deleted.
(-[WebAVPlayerController playerViewControllerDidCancelOptimizedFullscreen:]): Deleted.
(WebVideoFullscreenInterfaceAVKit::willCancelOptimizedFullscreen): Deleted.
(WebVideoFullscreenInterfaceAVKit::didCancelOptimizedFullscreen): Deleted.
* platform/spi/cocoa/AVKitSPI.h: Remove dead SPI. Add isOptimizedFullscreenPossible.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184907 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoRemove JSC_OBJC_API_AVAILABLE_MAC_OS_X_1080
mitz@apple.com [Wed, 27 May 2015 16:52:51 +0000 (16:52 +0000)]
Remove JSC_OBJC_API_AVAILABLE_MAC_OS_X_1080
https://bugs.webkit.org/show_bug.cgi?id=145403

Reviewed by Anders Carlsson.

JSC_OBJC_API_AVAILABLE_MAC_OS_X_1080 was used to enable the JavaScriptCore Objective-C API
for WebKit and Safari projects building with JavaScriptCore targeting OS X 10.8. We don’t
need it anymore.

* API/JSBase.h:
* API/JSContext.h:
* API/JSManagedValue.h:
* API/JSValue.h:
* API/JSVirtualMachine.h:
* Configurations/Base.xcconfig:
* postprocess-headers.sh:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184906 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[iOS] Rebaseline expected results for tests in LayoutTests/compositing
mattbaker@apple.com [Wed, 27 May 2015 16:08:45 +0000 (16:08 +0000)]
[iOS] Rebaseline expected results for tests in LayoutTests/compositing
https://bugs.webkit.org/show_bug.cgi?id=145401

Reviewed by Simon Fraser.

Rebaselined iOS tests failing due to expected differences in converage rect, scrollbar and contentsScale.

* platform/ios-simulator/compositing/layer-creation/scale-rotation-animation-overlap-expected.txt: Added.
* platform/ios-simulator/compositing/masks/compositing-clip-path-change-no-repaint-expected.txt: Added.
* platform/ios-simulator/compositing/tiling/rotated-tiled-clamped-expected.txt:
* platform/ios-simulator/compositing/tiling/rotated-tiled-preserve3d-clamped-expected.txt:
* platform/ios-simulator/compositing/tiling/transform-origin-tiled-expected.txt: Copied from LayoutTests/platform/ios-simulator/compositing/tiling/rotated-tiled-clamped-expected.txt.
* platform/ios-simulator/compositing/visible-rect/backing-change-height-from-zero-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184905 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed. Update OptionsGTK.cmake and NEWS for 2.9.2 release.
carlosgc@webkit.org [Wed, 27 May 2015 09:50:44 +0000 (09:50 +0000)]
Unreviewed. Update OptionsGTK.cmake and NEWS for 2.9.2 release.

.:

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

Source/WebKit2:

* gtk/NEWS: Add release notes for 2.9.2.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184902 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[SOUP] Network Cache: Handle the case when we fail to create the IO stream
carlosgc@webkit.org [Wed, 27 May 2015 08:37:20 +0000 (08:37 +0000)]
[SOUP] Network Cache: Handle the case when we fail to create the IO stream
https://bugs.webkit.org/show_bug.cgi?id=145406

Reviewed by Sergio Villar Senin.

We were asserting in that case, but it can happen that we
fail to create the stream. It happened to me after r184690, that
changed the NetworkCache::Key hash. Since this was not expected to
happen, the async operation never finished and the completion
handler never called, leaving resources loading forever. We
should make sure we call the completion handler with an error
code, so that the load finishes and the entry is silently removed
from the cache.

* NetworkProcess/cache/NetworkCacheIOChannelSoup.cpp:
(WebKit::NetworkCache::IOChannel::read):
(WebKit::NetworkCache::IOChannel::readSync):
(WebKit::NetworkCache::IOChannel::write):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184901 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[iOS] Arabic ligatures are broken in Google Maps
mmaxfield@apple.com [Wed, 27 May 2015 07:14:23 +0000 (07:14 +0000)]
[iOS] Arabic ligatures are broken in Google Maps
https://bugs.webkit.org/show_bug.cgi?id=145336
<rdar://problem/20689607>

Reviewed by Darin Adler.

Source/WebCore:

When we fall off the end of the font fallback list, we ask the platform for which particular font we should use
to display a particular codepoint. When we do this, we supply the font which is first in the fallback list so
that the platform can try to match it as best it can. However, web fonts' FontPlatformData have a null m_font
property, which means that we were passing nullptr to this platform fallback function.

The solution is something I've wanted to do for a while, which is migrate webfonts from using
CGDataProviderCreateWithCFData() to using CTFontManagerCreateFontDescriptorFromData(). This call creates a
CoreText font for an in-memory buffer (the old call was creating a CoreGraphics front from the same thing).
This means that we can use the FontPlatformData constructor which accepts a CoreText font, thereby making web
fonts not have a null m_font property. Note that not all platforms can use this call, which means that this
new codepath is gated to only certain OSes (which is okay because only certain OSes use the
CTFontCreatePhysicalFontDescriptorForCharactersWithLanguage() call inside FontCache). Because web fonts are the
only user of the constructor which just takes a CG font, we can make the constructor private to the
FontPlatformData class (as soon as all platforms can use this new codepath).

Test: platform/ios-simulator/fast/text/arabic-with-no-supporting-webfont.html

* platform/graphics/FontPlatformData.h: Make the FontPlatformData constructor which takes a CGFontRef private.
* platform/graphics/mac/FontCustomPlatformData.cpp:
(WebCore::FontCustomPlatformData::fontPlatformData): Use the CoreText type instead of the CoreGraphics type.
(WebCore::createFontCustomPlatformData): Migrate to CTFontManagerCreateFontDescriptorFromData().
* platform/graphics/mac/FontCustomPlatformData.h:
(WebCore::FontCustomPlatformData::FontCustomPlatformData): Store a CTFontDescriptorRef instead of a CGFontRef.

LayoutTests:

This patch adds a DRT test to make sure that text gets drawn with the correct font.

It also updates expected results for 4 tests, which I have verified are still correct. However,
because the behavior of Mavericks differs from Yosemite, I copied the old expected results to
platform/mac-mavericks before updating the results in platform/mac.

This patch also moves fast/ruby/ruby-expansion-cjk*.html to platform/mac because they rely on
platform-specific screen-space quantization. This patch also skips these tests on Mavericks
because I can't add platform-specific expected results for reftests.

* platform/ios-simulator/fast/text/arabic-with-no-supporting-webfont-expected.txt:
* platform/ios-simulator/fast/text/arabic-with-no-supporting-webfont.html:

* platform/mac-mavericks/fast/css/font-face-opentype-expected.txt: Copied from platform/mac.
* platform/mac-mavericks/svg/batik/text/xmlSpace-expected.txt: Ditto.
* platform/mac-mavericks/svg/custom/svg-fonts-fallback-expected.txt: Ditto.
* platform/mac-mavericks/svg/custom/svg-fonts-without-missing-glyph-expected.txt: Ditto.
* platform/mac-mavericks/TestExpectations: Skip fast/ruyb/ruby-expansion-cjk*.html
* platform/mac/fast/css/font-face-opentype-expected.txt: Updated.
* platform/mac/svg/batik/text/xmlSpace-expected.txt: Ditto.
* platform/mac/svg/custom/svg-fonts-fallback-expected.txt: Ditto.
* platform/mac/svg/custom/svg-fonts-without-missing-glyph-expected.txt: Ditto.
* LayoutTests/platform/mac/fast/ruby/resources/green.png: Moved from fast/ruby/resources/green.png
* LayoutTests/platform/mac/fast/ruby/resources/ruby-expansion.svg: Moved from fast/ruby/resources/ruby-expansion.svg
* LayoutTests/platform/mac/fast/ruby/ruby-expansion-cjk-2-expected.html: Moved from fast/ruby/ruby-expansion-cjk-2-expected.html
* LayoutTests/platform/mac/fast/ruby/ruby-expansion-cjk-2.html: Moved from fast/ruby/ruby-expansion-cjk-2.html
* LayoutTests/platform/mac/fast/ruby/ruby-expansion-cjk-3-expected.html: Moved from fast/ruby/ruby-expansion-cjk-3-expected.html
* LayoutTests/platform/mac/fast/ruby/ruby-expansion-cjk-3.html: Moved from fast/ruby/ruby-expansion-cjk-3.html
* LayoutTests/platform/mac/fast/ruby/ruby-expansion-cjk-4-expected.html: Moved from fast/ruby/ruby-expansion-cjk-4-expected.html
* LayoutTests/platform/mac/fast/ruby/ruby-expansion-cjk-4.html: Moved from fast/ruby/ruby-expansion-cjk-4.html
* LayoutTests/platform/mac/fast/ruby/ruby-expansion-cjk-5-expected.html: Moved from fast/ruby/ruby-expansion-cjk-5-expected.html
* LayoutTests/platform/mac/fast/ruby/ruby-expansion-cjk-5.html: Moved from fast/ruby/ruby-expansion-cjk-5.html
* LayoutTests/platform/mac/fast/ruby/ruby-expansion-cjk-expected.html: Moved from fast/ruby/ruby-expansion-cjk-expected.html
* LayoutTests/platform/mac/fast/ruby/ruby-expansion-cjk.html: Moved from fast/ruby/ruby-expansion-cjk.html

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184899 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoNetwork Cache: Add cache capacity to the totals of JSON file
carlosgc@webkit.org [Wed, 27 May 2015 06:41:02 +0000 (06:41 +0000)]
Network Cache: Add cache capacity to the totals of JSON file
https://bugs.webkit.org/show_bug.cgi?id=145246

Reviewed by Antti Koivisto.

* NetworkProcess/cache/NetworkCache.cpp:
(WebKit::NetworkCache::Cache::dumpContentsToFile):
* NetworkProcess/cache/NetworkCacheStorage.h:
(WebKit::NetworkCache::Storage::capacity):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184898 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoVersioning.
bshafiei@apple.com [Wed, 27 May 2015 06:19:20 +0000 (06:19 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184896 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAn SVG with no intrinsic size does not draw correct slices when used as a border...
commit-queue@webkit.org [Wed, 27 May 2015 04:18:04 +0000 (04:18 +0000)]
An SVG with no intrinsic size does not draw correct slices when used as a border-image for an HTML element.
https://bugs.webkit.org/show_bug.cgi?id=139405

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2015-05-26
Reviewed by Darin Adler.
LayoutTests/imported/mozilla:

* svg/as-image/border-image-simple-2.html: Add "border: 0px none;" to
style of the <div> although the style has "border-image: url() 0 fill;".
If the border width is not set to zero in the style, the <div> will have
a 3px border. This seems to be a bug in WebKit but should be unrelated to
the non-intrinsic-sized images with border-image.

Source/WebCore:

When using a non-intrinsic-sized image as an image-border, all the source
slices have to be extracted from the top-left corner of the source image.
This is because the right and bottom sides of the image can not be known.
Also all the slices should not be stretched. In other words, the sizes of
source slices for a non-intrinsic-sized image should be equal to the sizes
of the destination container slices.

This is not compliant with the css3 w3c specs. I think the specs makes it
harder to predict what exactly will be drawn for the image-border in this
case. See http://www.w3.org/TR/css3-background/#border-image-slice. This
approach is implemented by FireFox. And I think it gives a more predictable
rendering for image-border in the case of non-intrinsic-sized images.

Test: fast/borders/border-image-fill-no-intrinsic-size.html

* CMakeLists.txt:
* WebCore.vcxproj/WebCore.vcxproj:
* WebCore.vcxproj/WebCore.vcxproj.filters:
* WebCore.xcodeproj/project.pbxproj:
* platform/LengthBox.cpp: Removed.
* platform/graphics/LayoutBoxExtent.cpp: Removed.
* platform/graphics/LayoutBoxExtent.h: Removed.
* platform/graphics/LayoutRect.h:
Delete LengthBox.cpp and move all the functionalities to LengthBox.h.
Delete LayoutBoxExtent.cpp since it is the same class as LengthBox.

* css/CSSProperty.cpp:
* platform/text/WritingMode.h:
Move the enums LogicalBoxSide and PhysicalBoxSide
from CSSProperty.cpp to WritingMode.h so it can be used by other classes.

* css/CSSToStyleMap.cpp:
(WebCore::CSSToStyleMap::mapNinePieceImageSlice):
(WebCore::CSSToStyleMap::mapNinePieceImageQuad):
* css/CSSToStyleMap.h:
* inspector/InspectorOverlay.cpp:
(WebCore::buildRendererHighlight):
* rendering/InlineFlowBox.cpp:
(WebCore::InlineFlowBox::addBorderOutsetVisualOverflow):
* rendering/RenderBox.h:
(WebCore::RenderBox::marginLogicalLeft):
(WebCore::RenderBox::marginLogicalRight):
(WebCore::RenderBox::setMarginBefore):
(WebCore::RenderBox::setMarginAfter):
(WebCore::RenderBox::setMarginStart):
(WebCore::RenderBox::setMarginEnd):
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::setClip):
(WebCore::RenderStyle::noneDashboardRegions):
* rendering/style/RenderStyle.h:
Use the new BoxExtent access methods for getting and settings the sides of
LengthBox and LayoutBoxExtent.

* platform/LengthBox.h:
(WebCore::BoxExtent::BoxExtent):
(WebCore::BoxExtent::at):
(WebCore::BoxExtent::top):
(WebCore::BoxExtent::right):
(WebCore::BoxExtent::bottom):
(WebCore::BoxExtent::left):
(WebCore::BoxExtent::setAt):
(WebCore::BoxExtent::setTop):
(WebCore::BoxExtent::setRight):
(WebCore::BoxExtent::setBottom):
(WebCore::BoxExtent::setLeft):
(WebCore::BoxExtent::before):
(WebCore::BoxExtent::end):
(WebCore::BoxExtent::after):
(WebCore::BoxExtent::start):
(WebCore::BoxExtent::setBefore):
(WebCore::BoxExtent::setEnd):
(WebCore::BoxExtent::setAfter):
(WebCore::BoxExtent::setStart):
(WebCore::BoxExtent::operator==):
(WebCore::BoxExtent::operator!=):
(WebCore::LengthBox::LengthBox):
(WebCore::LengthBox::isZero):
(WebCore::LengthBox::left): Deleted.
(WebCore::LengthBox::right): Deleted.
(WebCore::LengthBox::top): Deleted.
(WebCore::LengthBox::bottom): Deleted.
(WebCore::LengthBox::operator==): Deleted.
(WebCore::LengthBox::operator!=): Deleted.
(WebCore::LengthBox::nonZero): Deleted.
Define a new template class for 'BoxExtent'. A 'BoxExtent' represents the
extent of four sides of a box. Use this class template to define the exiting
classes 'LengthBox' and 'LayoutBoxExtent'. Use it also to and define the
new class FloatBoxExtent

* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::calculateImageIntrinsicDimensions):
* rendering/RenderBoxModelObject.h:
(WebCore::RenderBoxModelObject::calculateFillTileSize):
* rendering/RenderListMarker.cpp:
(WebCore::RenderListMarker::updateContent):
* rendering/shapes/ShapeOutsideInfo.cpp:
(WebCore::ShapeOutsideInfo::createShapeForImage):
Change the return value of calculateImageIntrinsicDimensions() to be a
bool which indicates whether the image has an intrinsic size or not. Add
a new reference argument to this function receive the resolved image size.

(WebCore::RenderBoxModelObject::paintNinePieceImage): Move all the painting
code of this function to NinePieceImage::paint()

(WebCore::computeBorderImageSide): Deleted.
Renamed to be NinePieceImage::computeSlice().

* rendering/style/BorderData.h:
(WebCore::BorderData::borderWidth): Add a new method to return the extents
of the border in a FloatBoxExtent.

* rendering/style/NinePieceImage.cpp:
(WebCore::NinePieceImage::computeSlice): Moved from NinePieceImage::computeSlice().

(WebCore::NinePieceImage::computeSlices): The first version of this function
computes the slices given their lengths. The slices have to be clamped to
the container size. The second version of this function computes the slices
given their lengths, their actual extents and their source slices.

(WebCore::NinePieceImage::scaleSlicesIfNeeded):  Reduce the slices if they
are too large.

(WebCore::NinePieceImage::isEmptyPieceRect): Returns true if an ImagePiece
should not be drawn.

(WebCore::NinePieceImage::horizontalTileRules):
(WebCore::NinePieceImage::verticalTileRules):
Fill vectors of tiling rules to be passed to GraphicsContext::drawTiledImage()

(WebCore::NinePieceImage::computeIntrinsicRects): Computes the nine pieces
rectangles for an intrinsic-sized container.

(WebCore::NinePieceImage::computeNonIntrinsicRects): Computes the nine
pieces rectangles for an non-intrinsic-sized source image. The computed
rectangles have to start at (0, 0) and their sizes should be equal to
the sizes of the nine pieces rectangles of the destination container.

(WebCore::NinePieceImage::computeIntrinsicSideTileScale):
(WebCore::NinePieceImage::computeIntrinsicMiddleTileScale):
(WebCore::NinePieceImage::computeIntrinsicTileScales):
(WebCore::NinePieceImage::computeNonIntrinsicTileScales):
Computes the scaling factors for drawing the tiles. For non-intrinsic source
images, there should not be any scaling factors.

(WebCore::NinePieceImage::paint): Moved from RenderBoxModelObject::paintNinePieceImage()
but simplified and restructured.

(WebCore::NinePieceImageData::NinePieceImageData): Use nullptr instead of 0.

* rendering/style/NinePieceImage.h:
(WebCore::operator++):
(WebCore::isCornerPiece):
(WebCore::isMiddlePiece):
(WebCore::imagePieceHorizontalSide):
(WebCore::imagePieceVerticalSide):
Add helper functions for the enum ImagePiece.

LayoutTests:

* fast/borders/border-image-fill-no-intrinsic-size-expected.html: Added.
* fast/borders/border-image-fill-no-intrinsic-size.html: Added.
* fast/borders/resources/svg-border-100x100-intrinsic.svg: Added.
* fast/borders/resources/svg-border-140x140-intrinsic.svg: Added.
* fast/borders/resources/svg-border-no-intrinsic.svg: Added.
Make sure the slices of the border-image are drawn correctly. When using an
image with no intrinsic size, all the slices has to be gotten from (0, 0)
of the source image.

* TestExpectations: Removed previously-failing test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184895 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago<font face> doesn't support plus character in font family names.
akling@apple.com [Wed, 27 May 2015 04:13:54 +0000 (04:13 +0000)]
<font face> doesn't support plus character in font family names.
<https://webkit.org/b/145400>
<rdar://problem/21097484>

Reviewed by Darin Adler.

Source/WebCore:

Turn CSSParser::parseFontFaceValue() into a fast-path parser for
<font face> attributes.
Going through the full CSS parser was hurting us here, as it would
misunderstand unquoted family names and choke on e.g '+'.

Test: fast/dom/HTMLFontElement/face-attribute.html

* css/CSSParser.cpp:
(WebCore::CSSParser::parseFontFaceValue):

LayoutTests:

Add a little test for <font face> attributes to cover this problem
and some other interesting cases with spaces and commas.

* fast/dom/HTMLFontElement/face-attribute-expected.txt: Added.
* fast/dom/HTMLFontElement/face-attribute.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184894 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAdd a symlink for every file inside /S/L/F/WebKit.framework to the PrivateFrameworks...
andersca@apple.com [Wed, 27 May 2015 02:44:40 +0000 (02:44 +0000)]
Add a symlink for every file inside /S/L/F/WebKit.framework to the PrivateFrameworks mirror.

rdar://problem/21111232.

Reviewed by Dan Bernstein.

* WebKit2.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184893 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[EFL] webview should be updated from (0,0) when using the evasGL.
commit-queue@webkit.org [Wed, 27 May 2015 01:08:29 +0000 (01:08 +0000)]
[EFL] webview should be updated from (0,0) when using the evasGL.
https://bugs.webkit.org/show_bug.cgi?id=137948

Patch by Hunseop Jeong <hs85.jeong@samsung.com> on 2015-05-26
Reviewed by Gyuyoung Kim.

Changed the point of the start position to update the evasObject
correctly when using the evasGL for graphics backend.

* UIProcess/API/efl/EwkView.cpp:
(EwkView::displayTimerFired):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184892 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAX: display:none content exposed to accessibility when aria-hidden is toggled on...
cfleizach@apple.com [Wed, 27 May 2015 00:54:38 +0000 (00:54 +0000)]
AX: display:none content exposed to accessibility when aria-hidden is toggled on ancestor element
https://bugs.webkit.org/show_bug.cgi?id=139142

Reviewed by Darin Adler.

* platform/mac/accessibility/aria-hidden-false-works-in-subtrees-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184891 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAX: display:none content exposed to accessibility when aria-hidden is toggled on...
cfleizach@apple.com [Wed, 27 May 2015 00:53:00 +0000 (00:53 +0000)]
AX: display:none content exposed to accessibility when aria-hidden is toggled on ancestor element
https://bugs.webkit.org/show_bug.cgi?id=139142

Reviewed by Darin Adler.
Source/WebCore:

Amend the code that determines when an invisible, but aria-hidden=false, element is exposed to accessibility.

The new guideline is that you must have aria-hidden=false on every node that is not rendered (except text which inherits)
otherwise the element will not be visible.

Modified existing test: accessibility/aria-hidden-false-works-in-subtrees.html

* accessibility/AXObjectCache.cpp:
(WebCore::isNodeAriaVisible):
(WebCore::AXObjectCache::rootWebArea):

LayoutTests:

* accessibility/aria-hidden-false-works-in-subtrees.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoEnable element unsigned index in WebGL2 again.
roger_fong@apple.com [Wed, 27 May 2015 00:18:58 +0000 (00:18 +0000)]
Enable element unsigned index in WebGL2 again.
https://bugs.webkit.org/show_bug.cgi?id=145392
<rdar://problem/21112585>.

Reviewed by Dean Jackson.

* html/canvas/WebGLRenderingContextBase.cpp:
(WebCore::WebGLRenderingContextBase::validateDrawElements):
Fix some incorrect logic that disables the feature for WebGL2.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAdd assertions to make sure pages in the PageCache are not loading
cdumez@apple.com [Wed, 27 May 2015 00:05:52 +0000 (00:05 +0000)]
Add assertions to make sure pages in the PageCache are not loading
https://bugs.webkit.org/show_bug.cgi?id=145397
<rdar://problem/20613631>

Reviewed by Antti Koivisto.

Add assertions to make sure pages in the PageCache are not loading.
These will hopefully help track down why PacheCache entries sometimes
have pending subresource loads when being destroyed.

* history/CachedFrame.cpp:
(WebCore::CachedFrame::CachedFrame):
* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::stopLoadingSubresources):
(WebCore::DocumentLoader::addSubresourceLoader):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184888 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: Function parameter string parsing improperly handles empty parameter...
commit-queue@webkit.org [Tue, 26 May 2015 23:11:13 +0000 (23:11 +0000)]
Web Inspector: Function parameter string parsing improperly handles empty parameter list
https://bugs.webkit.org/show_bug.cgi?id=145391

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2015-05-26
Reviewed by Darin Adler.

* UserInterface/Views/ObjectTreePropertyTreeElement.js:
(WebInspector.ObjectTreePropertyTreeElement.prototype._functionParameterString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184887 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUnreviewed, rolling out r184872.
commit-queue@webkit.org [Tue, 26 May 2015 23:00:44 +0000 (23:00 +0000)]
Unreviewed, rolling out r184872.
https://bugs.webkit.org/show_bug.cgi?id=145396

Bindings tests broken on the bots. (Requested by dethbakin on
#webkit).

Reverted changeset:

"Binding generator should support interfaces with
CustomConstructor and NoInterfaceObject"
https://bugs.webkit.org/show_bug.cgi?id=145016
http://trac.webkit.org/changeset/184872

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoOverhanging float sets are not cleaned up properly when floating renderer is destroyed.
zalan@apple.com [Tue, 26 May 2015 22:59:40 +0000 (22:59 +0000)]
Overhanging float sets are not cleaned up properly when floating renderer is destroyed.
https://bugs.webkit.org/show_bug.cgi?id=145323
rdar://problem/20980628

Reviewed by Dave Hyatt.

This patch ensures when an overhanging float renderer is destroyed,
all the sibling containers' floating object set(m_floatingObjects) gets properly cleaned up.

When an overhanging float is present, we cache the renderer on the parent and on the affected
sibling containers too. (RenderBlockFlow::m_floatingObjects) These caches(sets) get cleared and repopulated
during ::layout(). In order to have a float renderer removed from a set, a layout needs to be initiated on the container.
This is normally done through RenderBlockFlow::markSiblingsWithFloatsForLayout() and RenderBlockFlow::markAllDescendantsWithFloatsForLayout().
However, when the float container's parent's writing direction changes (and we promote the children containers to new formatting contexts),
the layout propagation through siblings does not work anymore.

The avoidsFloats() check in RenderBlockFlow::markSiblingsWithFloatsForLayout() has very little performance gain, but it prevents us
from propagating layout to siblings when certain properties of the parent container changes.

Source/WebCore:

Test: fast/block/float/crash-when-floating-object-is-removed.xhtml

* rendering/RenderBlockFlow.cpp:
(WebCore::RenderBlockFlow::markSiblingsWithFloatsForLayout):
* rendering/RenderBox.cpp:
(WebCore::outermostBlockContainingFloatingObject):
(WebCore::RenderBox::removeFloatingOrPositionedChildFromBlockLists):
(WebCore::RenderBox::outermostBlockContainingFloatingObject): Deleted.
* rendering/RenderBox.h:

LayoutTests:

* fast/block/float/crash-when-floating-object-is-removed-expected.txt: Added.
* fast/block/float/crash-when-floating-object-is-removed.xhtml: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184885 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoPhoto Booth hangs under JSC::MachineThreads::tryCopyOtherThreadStacks
ggaren@apple.com [Tue, 26 May 2015 22:59:38 +0000 (22:59 +0000)]
Photo Booth hangs under JSC::MachineThreads::tryCopyOtherThreadStacks
https://bugs.webkit.org/show_bug.cgi?id=145395

Reviewed by Mark Hahnenberg.

No test case because we already have --threaded mode, which runs lots of
parallel GC, but it (and the original in-app test case) can't reproduce
this bug.

* heap/MachineStackMarker.cpp:
(JSC::MachineThreads::tryCopyOtherThreadStacks): Use a lock to prevent
two threads from mutually suspending each other.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184884 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoInteger overflow in XLarge allocation (due to unchecked roundUpToMultipleOf)
ggaren@apple.com [Tue, 26 May 2015 22:48:13 +0000 (22:48 +0000)]
Integer overflow in XLarge allocation (due to unchecked roundUpToMultipleOf)
https://bugs.webkit.org/show_bug.cgi?id=145385

Reviewed by Andreas Kling.

Added some checking to verify that round-up operations will not overflow
a size_t.

The simplest way to do this was to introduce a notion of xLargeMax, like
we have for smallMax, mediumMax, and largeMax. It's a bit surprising at
first to think that there is an xLargeMax, since xLarge is what we use
to handle the biggest things. But computers have limits, so it makes sense.

FWIW, TCMalloc used to have an xLargeMax too, which it called kMaxValidPages.

No test because this bug was found by code inspection and I don't know
of a practical way to convince WebKit to make an allocation this large.

* bmalloc/Allocator.cpp:
(bmalloc::Allocator::tryAllocate):
(bmalloc::Allocator::allocate):
(bmalloc::Allocator::reallocate):
(bmalloc::Allocator::allocateSlowCase): Check against xLargeMax to avoid
overflow when rounding up.

* bmalloc/BAssert.h: Added support for explicit crashing.

* bmalloc/Sizes.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184883 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoFixes compatibility issues with recent dashboard cleanup.
dburkart@apple.com [Tue, 26 May 2015 22:31:04 +0000 (22:31 +0000)]
Fixes compatibility issues with recent dashboard cleanup.
https://bugs.webkit.org/show_bug.cgi?id=144814

Reviewed by Alexey Proskuryakov.

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotStaticAnalyzerQueueView.js:
(BuildbotStaticAnalyzerQueueView):
(BuildbotStaticAnalyzerQueueView.prototype.update):
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Main.js:
(documentReady):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184882 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agostorage/indexeddb/deleteIndex-bug110792.html is flaky
bdakin@apple.com [Tue, 26 May 2015 22:15:34 +0000 (22:15 +0000)]
storage/indexeddb/deleteIndex-bug110792.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=145390

Marking as flaky.
* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184881 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoAdd Array.prototype.copyWithin to JSC features.json
utatane.tea@gmail.com [Tue, 26 May 2015 22:06:33 +0000 (22:06 +0000)]
Add Array.prototype.copyWithin to JSC features.json
https://bugs.webkit.org/show_bug.cgi?id=145387

Reviewed by Darin Adler.

* features.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184880 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoBotwatcher's dashboard doesn't show JSC test regressions on Apple bots
ap@apple.com [Tue, 26 May 2015 22:06:05 +0000 (22:06 +0000)]
Botwatcher's dashboard doesn't show JSC test regressions on Apple bots
https://bugs.webkit.org/show_bug.cgi?id=143091
rdar://problem/19330328

Reviewed by Darin Adler and Timothy Hatcher.

We have many of these, which are hard to fit on the dashboard. Added a view that
collapses to a single green bubble when everything is good, and expands when there
are failures (or manually).

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/index.html: Added BuildbotCombinedQueueView.js

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Buildbot.js:
(Buildbot.prototype.set isAuthenticated):
(Buildbot.prototype._normalizeQueueInfo):
(Buildbot.prototype._normalizeQueuesInfo):
(Buildbot.prototype.updateQueues):
Moved queue info normalization here from BuildbitQueue. The latter is a model class
that shouldn't have to know about presentation, and this lets us leep the knowledge
about combined queues out if it. Later, we can refactor existing code, and move out
all knowledge about headings and such.

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotCombinedQueueView.js:
Added the new view. It's not quite universal, and doesn't have as helpful popovers
as other views, but we can extend it when/if we use it for more than JSC.

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotQueue.js:
(BuildbotQueue): Now that info is normalized before creating a queue, don't do that here.

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Main.js:
(documentReady): Added support for combined queues. These are currently always ending up
in Other column, but it's easy to customize in the future if we need to.

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:
(WebKitBuildbot): Added JSC queus.

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Styles/QueueView.css:
(.combined-queue-popover):
(.combined-queue-popover .revision):
Added styles for the combined view.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184879 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: Add Array.prototype.copyWithin parameter list
commit-queue@webkit.org [Tue, 26 May 2015 21:50:39 +0000 (21:50 +0000)]
Web Inspector: Add Array.prototype.copyWithin parameter list
https://bugs.webkit.org/show_bug.cgi?id=145388

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2015-05-26
Reviewed by Darin Adler.

* UserInterface/Models/NativeFunctionParameters.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184878 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: focus outline of a search field should have a radius
nvasilyev@apple.com [Tue, 26 May 2015 21:48:41 +0000 (21:48 +0000)]
Web Inspector: focus outline of a search field should have a radius
https://bugs.webkit.org/show_bug.cgi?id=145383

Add a focus outline animation to roughly mimic the default focus outline of OS X.

Reviewed by Timothy Hatcher.

* UserInterface/Views/Toolbar.css:
(.toolbar .search-bar > input[type="search"]):
-webkit-focus-ring-color doesn't follow element's curvature (e.g. border-radius)
and it cannot be animated. Replace it with box-shadow.

(.toolbar .search-bar > input[type="search"]:focus):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184877 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years ago[EFL][CoordinatedGraphics] Remove CoordinatedTileClient and CoordinatedTileBackend
ryuan.choi@navercorp.com [Tue, 26 May 2015 21:38:57 +0000 (21:38 +0000)]
[EFL][CoordinatedGraphics] Remove CoordinatedTileClient and CoordinatedTileBackend
https://bugs.webkit.org/show_bug.cgi?id=133337

Reviewed by Darin Adler.

This patch simplifies relationship between CoordinatedGraphicsLayer and TiledBackingStore
by removing TiledBackingStoreBackend because TiledBackingStore only support
CoordinatedTileBackend.
And it also moves CoordinatedTileClient interface to TiledBackingStoreClient.

* platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.cpp:
(WebCore::CoordinatedGraphicsLayer::createBackingStore):
* platform/graphics/texmap/coordinated/CoordinatedGraphicsLayer.h:
* platform/graphics/texmap/coordinated/CoordinatedTile.cpp:
(WebCore::CoordinatedTile::create):
(WebCore::CoordinatedTile::CoordinatedTile):
(WebCore::CoordinatedTile::~CoordinatedTile):
(WebCore::CoordinatedTile::updateBackBuffer):
(WebCore::CoordinatedTileBackend::CoordinatedTileBackend): Deleted.
(WebCore::CoordinatedTileBackend::createTile): Deleted.
(WebCore::CoordinatedTileBackend::paintCheckerPattern): Deleted.
* platform/graphics/texmap/coordinated/CoordinatedTile.h:
(WebCore::CoordinatedTile::create): Deleted.
(WebCore::CoordinatedTileClient::~CoordinatedTileClient): Deleted.
* platform/graphics/texmap/coordinated/TiledBackingStore.cpp:
(WebCore::TiledBackingStore::TiledBackingStore):
(WebCore::TiledBackingStore::createTiles):
* platform/graphics/texmap/coordinated/TiledBackingStore.h:
* platform/graphics/texmap/coordinated/TiledBackingStoreBackend.h: Removed.
(WebCore::TiledBackingStoreBackend::~TiledBackingStoreBackend): Deleted.
* platform/graphics/texmap/coordinated/TiledBackingStoreClient.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184876 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoFullscreen video does not display on platforms without accelerated video rendering
commit-queue@webkit.org [Tue, 26 May 2015 21:37:40 +0000 (21:37 +0000)]
Fullscreen video does not display on platforms without accelerated video rendering
https://bugs.webkit.org/show_bug.cgi?id=145118

In order to paint video contents in fullscreen mode, add supportsAcceleratedRendering condition.

Patch by Daegyu Lee <daegyu.lee@navercorp.com> on 2015-05-26
Reviewed by Simon Fraser.

* rendering/RenderVideo.cpp:
(WebCore::RenderVideo::paintReplaced):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184875 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoSVG fragment identifier rendering issue
commit-queue@webkit.org [Tue, 26 May 2015 21:24:05 +0000 (21:24 +0000)]
SVG fragment identifier rendering issue
https://bugs.webkit.org/show_bug.cgi?id=137328

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2015-05-26
Reviewed by Darin Adler.

Source/WebCore:

This is a follow up for http://trac.webkit.org/changeset/164983. In this
changeset, scrolling to the fragment should have been added before the
the paint to guarantee setting the proper display position for the SVG
fragment.

Test: svg/css/svg-resource-fragment-identifier-order.html

* svg/graphics/SVGImage.cpp:
(WebCore::SVGImage::draw): Move view->scrollToFragment() before calling
view->paint().

LayoutTests:

* svg/css/svg-resource-fragment-identifier-order-expected.html: Added.
* svg/css/svg-resource-fragment-identifier-order.html: Added.
Ensure the SVG fragment is drawn correctly when the same SVG image is
referenced multiple times.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184874 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoWeb Inspector: The bottom part "debugger" is clipped in the tab’s title
nvasilyev@apple.com [Tue, 26 May 2015 21:20:50 +0000 (21:20 +0000)]
Web Inspector: The bottom part "debugger" is clipped in the tab’s title
https://bugs.webkit.org/show_bug.cgi?id=145381

Reviewed by Timothy Hatcher.

* UserInterface/Views/TabBar.css:
(.tab-bar > .item):
Set line-height to fit "debugger". Setting line-height to 16px, which is the
size of the tabs icons, moves the text one pixel up, so set it to 15px.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184873 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoBinding generator should support interfaces with CustomConstructor and NoInterfaceObject
youenn.fablet@crf.canon.fr [Tue, 26 May 2015 21:18:37 +0000 (21:18 +0000)]
Binding generator should support interfaces with CustomConstructor and NoInterfaceObject
https://bugs.webkit.org/show_bug.cgi?id=145016

Reviewed by Darin Adler.

Updated code generator to generate the necessary declarations (constructor property in the prototype, the constructor class).
In the case of CustomConstructor with NoInterfaceObject, the implementation of the constructor property related JS function is changed as follow:
It creates a constructor object and shadows the constructor property JS function with this object.
This ensures that only one constructor object is created per prototype.
The constructor is not exposed as it is not added to the global object constructor map.

NoInterfaceObject is asking to not make the interface visible, which is contradictory with having a visible constructor of the interface.
The case for having NoInterfaceObject and CustomConstructor is the following:
We do not want to expose the interface but we still want to have a constructor property on the prototype of objects implementing the interface.
One such case is https://streams.spec.whatwg.org/#globals

Added TestCustomConstructorWithNoInterfaceObject.idl to check that case.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateHeader):
(GenerateAttributesHashTable):
(GenerateImplementation):
(HasConstructorProperty):
* bindings/scripts/test/GObject/WebKitDOMTestCustomConstructorWithNoInterfaceObject.cpp: Added.
(WebKit::kit):
(WebKit::core):
(WebKit::wrapTestCustomConstructorWithNoInterfaceObject):
(webkit_dom_test_custom_constructor_with_no_interface_object_finalize):
(webkit_dom_test_custom_constructor_with_no_interface_object_constructor):
(webkit_dom_test_custom_constructor_with_no_interface_object_class_init):
(webkit_dom_test_custom_constructor_with_no_interface_object_init):
* bindings/scripts/test/GObject/WebKitDOMTestCustomConstructorWithNoInterfaceObject.h: Added.
* bindings/scripts/test/GObject/WebKitDOMTestCustomConstructorWithNoInterfaceObjectPrivate.h: Added.
* bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.cpp: Added.
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectPrototype::create):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectPrototype::createStructure):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectPrototype::JSTestCustomConstructorWithNoInterfaceObjectPrototype):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectConstructor::create):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectConstructor::createStructure):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectConstructor::JSTestCustomConstructorWithNoInterfaceObjectConstructor):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectConstructor::finishCreation):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectConstructor::getConstructData):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectPrototype::finishCreation):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::JSTestCustomConstructorWithNoInterfaceObject):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::createPrototype):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::getPrototype):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::destroy):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::~JSTestCustomConstructorWithNoInterfaceObject):
(WebCore::jsTestCustomConstructorWithNoInterfaceObjectConstructor):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::getConstructor):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectOwner::isReachableFromOpaqueRoots):
(WebCore::JSTestCustomConstructorWithNoInterfaceObjectOwner::finalize):
(WebCore::toJS):
(WebCore::JSTestCustomConstructorWithNoInterfaceObject::toWrapped):
* bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.h: Added.
(WebCore::wrapperOwner):
(WebCore::toJS):
* bindings/scripts/test/ObjC/DOMTestCustomConstructorWithNoInterfaceObject.h: Added.
* bindings/scripts/test/ObjC/DOMTestCustomConstructorWithNoInterfaceObject.mm: Added.
(-[DOMTestCustomConstructorWithNoInterfaceObject dealloc]):
(-[DOMTestCustomConstructorWithNoInterfaceObject finalize]):
(core):
(kit):
* bindings/scripts/test/ObjC/DOMTestCustomConstructorWithNoInterfaceObjectInternal.h: Added.
* bindings/scripts/test/TestCustomConstructor.idl: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184872 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoReflect nits for r184863
utatane.tea@gmail.com [Tue, 26 May 2015 21:17:22 +0000 (21:17 +0000)]
Reflect nits for r184863
https://bugs.webkit.org/show_bug.cgi?id=145107

Reviewed by Darin Adler.

1. Added the copyright line.
2. Added an optional argument (/*, end */). To do so, fixed generate-js-builtins.
3. Dropped the unnecessary variable `thisValue`.
4. Fix the type error messages. This is also found in StringIterator.prototype.js.
5. Added tests for 0 arguments.

* builtins/Array.prototype.js:
(copyWithin):
* builtins/StringIterator.prototype.js:
(next):
* generate-js-builtins:
* tests/stress/array-copywithin.js:
* tests/stress/string-iterators.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184871 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5 years agoUpdate Autofill button icon
dbates@webkit.org [Tue, 26 May 2015 20:40:34 +0000 (20:40 +0000)]
Update Autofill button icon
https://bugs.webkit.org/show_bug.cgi?id=145191
<rdar://problem/20178942>

Update expected results for Windows.

* platform/win/fast/forms/input-auto-fill-button-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@184870 268f45cc-cd09-0410-ab3c-d52691b4dbfc