WebKit-https.git
7 years ago[EFL] Skip new compositing tests for EFL port
commit-queue@webkit.org [Mon, 30 Jul 2012 15:39:04 +0000 (15:39 +0000)]
[EFL] Skip new compositing tests for EFL port
https://bugs.webkit.org/show_bug.cgi?id=92633

Unreviewed EFL gardening.

Skip new compositing layout tests until accelerated compositing is enabled.

Patch by Alexander Shalamov <alexander.shalamov@intel.com> on 2012-07-30

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124028 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoProhibit having AuthorShadowDOM of input or textarea element for a while and having...
shinyak@chromium.org [Mon, 30 Jul 2012 15:16:24 +0000 (15:16 +0000)]
Prohibit having AuthorShadowDOM of input or textarea element for a while and having a flag to enable it in Internals.
https://bugs.webkit.org/show_bug.cgi?id=92611

Reviewed by Hajime Morita.

.:

* Source/autotools/symbols.filter:

Source/WebCore:

Prohibit having AuthorShadowDOM of input elemnet unless a flag in RuntimeEnabledFeatures is not enabled.
The flag can be enabled in Internals.

The elements in UserAgentShadowDOM of input element assume that they have a renderer when the input
element has a renderer. However, this does not hold when AuthorShadowDOM is added to input element.
So until we fix the issue, we should disable to have AuthorShadowDOM for input element, because it may
cause crashes and security problems.

Tests: fast/dom/shadow/input-with-validation-without-shadow.html
       fast/dom/shadow/input-with-validation.html
       fast/dom/shadow/shadow-disable.html

* bindings/generic/RuntimeEnabledFeatures.cpp:
(WebCore):
* bindings/generic/RuntimeEnabledFeatures.h:
(RuntimeEnabledFeatures):
(WebCore::RuntimeEnabledFeatures::authorShadowDOMForAnyElementEnabled):
(WebCore::RuntimeEnabledFeatures::setAuthorShadowDOMForAnyElementEnabled):
* dom/ShadowRoot.cpp:
(WebCore::allowsAuthorShadowRoot): The input element or textarea element does not allow to have a ShadowDOM unless
is a flag in RuntimeEnabledFeatures is not enabled.
* html/HTMLInputElement.h:
(WebCore::isHTMLInputElement):
(WebCore):
* html/HTMLTextAreaElement.h:
(WebCore::isHTMLTextAreaElement):
(WebCore):
* testing/Internals.cpp:
(WebCore::Internals::setAuthorShadowDOMForAnyElementEnabled):
(WebCore):
* testing/Internals.h:
(Internals):
* testing/Internals.idl:

LayoutTests:

* fast/dom/shadow/input-with-validation-without-shadow.html: Make the test runs even if
new WebKitShadowRoot() throws an exception.
* fast/dom/shadow/input-with-validation.html: Sets a flag to make AuthorShadowDOM available.
* fast/dom/shadow/shadow-disable-expected.txt:
* fast/dom/shadow/shadow-disable.html: Makes it unavailable for input and textarea to have an AuthorShadowDOM.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124027 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: support --line-numbers mapping for SASS
pfeldman@chromium.org [Mon, 30 Jul 2012 15:12:23 +0000 (15:12 +0000)]
Web Inspector: support --line-numbers mapping for SASS
https://bugs.webkit.org/show_bug.cgi?id=92400

Reviewed by Vsevolod Vlasov.

There is --debug-info and --line-numbers ways of referencing the source scss file, adding support for --line-numbers here.

* inspector/front-end/SASSSourceMapping.js:
(WebInspector.SASSSourceMapping.prototype._resourceAdded.didRequestContent):
(WebInspector.SASSSourceMapping.prototype._resourceAdded):
(_bindUISourceCode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124026 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoImplement datalist UI for input type color for Chromium
keishi@webkit.org [Mon, 30 Jul 2012 14:32:16 +0000 (14:32 +0000)]
Implement datalist UI for input type color for Chromium
https://bugs.webkit.org/show_bug.cgi?id=92075

Reviewed by Kent Tamura.

Source/Platform:

* chromium/public/WebLocalizedString.h: Added OtherColorLabel.

Source/WebCore:

Test: platform/chromium/fast/forms/color/color-suggestion-picker-appearance.html

* rendering/RenderThemeChromiumCommon.cpp:
(WebCore::RenderThemeChromiumCommon::supportsDataListUI): Added color to types supporting datalist.

Source/WebKit/chromium:

Integrated ColorChooserProxy and WebColorChooserClientImpl into ColorChooserUIController.
ColorChooserUIController can control whether to open the color suggestion picker
popup or the color chooser depending on the responses from ColorChooserClient.

* WebKit.gyp:
* src/ChromeClientImpl.cpp:
(WebKit::ChromeClientImpl::createColorChooser): Returns ColorChooserUIController which is a ColorChooser.
(WebKit::ChromeClientImpl::createWebColorChooser): Returns WebColorChooser that was created by Chromium.
* src/ChromeClientImpl.h:
(WebKit):
(ChromeClientImpl):
* src/ColorChooserProxy.cpp: Removed.
* src/ColorChooserProxy.h: Removed.
* src/ColorChooserUIController.cpp: Added.
(WebKit):
(WebKit::ColorChooserUIController::ColorChooserUIController): Controls the UI for color chooser. Opens the color suggestion picker popup or color chooser depending on the ColorChooserClient.
(WebKit::ColorChooserUIController::~ColorChooserUIController):
(WebKit::ColorChooserUIController::setSelectedColor):
(WebKit::ColorChooserUIController::endChooser): This is called by WebCore so we close both the popup and the color chooser.
(WebKit::ColorChooserUIController::didChooseColor):
(WebKit::ColorChooserUIController::didEndChooser):
(WebKit::ColorChooserUIController::contentSize): The popup will be resized later from JS so we return an arbitrary size.
(WebKit::ColorChooserUIController::writeDocument):
(WebKit::ColorChooserUIController::setValueAndClosePopup): Performs action based on the numValue.
(WebKit::ColorChooserUIController::didClosePopup):
(WebKit::ColorChooserUIController::openPopup): Opens color chooser suggestion popup.
(WebKit::ColorChooserUIController::closePopup):
(WebKit::ColorChooserUIController::openColorChooser): Opens color chooser.
* src/ColorChooserUIController.h: Renamed from Source/WebKit/chromium/src/WebColorChooserClientImpl.h.
(WebCore):
(WebKit):
(ColorChooserUIController):
* src/WebColorChooserClientImpl.cpp: Removed.

LayoutTests:

* platform/chromium-mac/platform/chromium/fast/forms/color/color-suggestion-picker-appearance-expected.png: Added.
* platform/chromium/TestExpectations:
* platform/chromium/fast/forms/color/color-suggestion-picker-appearance-expected.txt: Added.
* platform/chromium/fast/forms/color/color-suggestion-picker-appearance.html: Added. Tests color suggestion picker popup appearance.
* platform/chromium/fast/forms/datalist/input-list-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124025 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWheel events on a page with frames are not handled in fixed layout
allan.jensen@nokia.com [Mon, 30 Jul 2012 14:23:13 +0000 (14:23 +0000)]
Wheel events on a page with frames are not handled in fixed layout
https://bugs.webkit.org/show_bug.cgi?id=90550

Reviewed by Antonio Gomes.

Based on patch by Andras Becsi.

Frame-rect size is not the right value to check against to see if the document
can be contained within the full visible content rect, since it does not handle
fixedVisibleContentRect. Instead visibleContentRect(true) should be used which
returns the full visible content rect disregarding scrollbars, and which is
the same a frame-rect size when fixedVisibleContentRect is not set.

* platform/ScrollView.cpp:
(WebCore::ScrollView::updateScrollbars):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124024 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSource/WebCore: There is no way to tell whether an element can be activated or not
allan.jensen@nokia.com [Mon, 30 Jul 2012 14:11:32 +0000 (14:11 +0000)]
Source/WebCore: There is no way to tell whether an element can be activated or not
https://bugs.webkit.org/show_bug.cgi?id=92093

Reviewed by Antonio Gomes.

This patch adds the willRespondToMouseClickEvents and willRespondToMouseMoveEvents
functions, named after similar functions in the iOS fork of WebKit.

Touch-adjustment will now use these two functions, which improves accuracy when
targeting shadow-dom elements, and also means touch adjustment will prefer enabled
form-elements over disabled ones.

Test: touchadjustment/disabled-formelements.html

* dom/Node.cpp:
(WebCore::Node::willRespondToMouseMoveEvents):
(WebCore::Node::willRespondToMouseClickEvents):
* dom/Node.h:
(Node):
* editing/DeleteButton.h:
* html/HTMLAnchorElement.cpp:
(WebCore::HTMLAnchorElement::willRespondToMouseClickEvents):
* html/HTMLAnchorElement.h:
(HTMLAnchorElement):
* html/HTMLButtonElement.cpp:
(WebCore::HTMLButtonElement::willRespondToMouseClickEvents):
* html/HTMLButtonElement.h:
(HTMLButtonElement):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::willRespondToMouseClickEvents):
* html/HTMLInputElement.h:
(HTMLInputElement):
* html/HTMLLabelElement.cpp:
(WebCore::HTMLLabelElement::willRespondToMouseClickEvents):
* html/HTMLLabelElement.h:
(HTMLLabelElement):
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::willRespondToMouseClickEvents):
* html/HTMLMediaElement.h:
* html/HTMLSummaryElement.cpp:
(WebCore::HTMLSummaryElement::willRespondToMouseClickEvents):
* html/HTMLSummaryElement.h:
(HTMLSummaryElement):
* html/shadow/CalendarPickerElement.cpp:
(WebCore::CalendarPickerElement::willRespondToMouseClickEvents):
* html/shadow/CalendarPickerElement.h:
(CalendarPickerElement):
* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlTimelineElement::willRespondToMouseClickEvents):
(WebCore::MediaControlTimelineElement::setPosition):
(WebCore::MediaControlVolumeSliderElement::willRespondToMouseMoveEvents):
(WebCore::MediaControlVolumeSliderElement::willRespondToMouseClickEvents):
* html/shadow/MediaControlElements.h:
(MediaControlTimelineElement):
(MediaControlVolumeSliderElement):
* html/shadow/MediaControlRootElement.h:
* html/shadow/SliderThumbElement.cpp:
(WebCore::SliderThumbElement::willRespondToMouseMoveEvents):
(WebCore::SliderThumbElement::willRespondToMouseClickEvents):
* html/shadow/SliderThumbElement.h:
(SliderThumbElement):
* html/shadow/TextControlInnerElements.cpp:
(WebCore::SearchFieldResultsButtonElement::willRespondToMouseClickEvents):
(WebCore::SearchFieldCancelButtonElement::willRespondToMouseClickEvents):
(WebCore::SpinButtonElement::willRespondToMouseMoveEvents):
(WebCore::SpinButtonElement::willRespondToMouseClickEvents):
(WebCore::InputFieldSpeechButtonElement::willRespondToMouseClickEvents):
* html/shadow/TextControlInnerElements.h:
(SearchFieldResultsButtonElement):
(SearchFieldCancelButtonElement):
(SpinButtonElement):
* html/shadow/TextFieldDecorationElement.cpp:
(WebCore::TextFieldDecorationElement::willRespondToMouseClickEvents):
* html/shadow/TextFieldDecorationElement.h:
(TextFieldDecorationElement):
* page/TouchAdjustment.cpp:
(WebCore::TouchAdjustment::nodeRespondsToTapGesture):

Source/WebKit/gtk: [Gtk] Add WK1 API for snapshot retrieval
https://bugs.webkit.org/show_bug.cgi?id=92261

Reviewed by Martin Robinson.

Add API to WebKitWebView to retrieve a snapshot of its
visible contents as a cairo_surface_t.

* docs/webkitgtk-sections.txt: Add new symbols.
* webkit/webkitwebview.cpp:
(webkit_web_view_get_snapshot): New
method to paint a webview snapshot.
* webkit/webkitwebview.h: Ditto.

LayoutTests: There is no way to tell whether an element can be activated or not
https://bugs.webkit.org/show_bug.cgi?id=92093

Reviewed by Antonio Gomes.

Tests that enabled formelements are now preferred over disabled formelements.

* touchadjustment/disabled-formelements-expected.txt: Added.
* touchadjustment/disabled-formelements.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124023 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoThere is no way to tell whether an element can be activated or not
allan.jensen@nokia.com [Mon, 30 Jul 2012 13:33:14 +0000 (13:33 +0000)]
There is no way to tell whether an element can be activated or not
https://bugs.webkit.org/show_bug.cgi?id=92093

Reviewed by Antonio Gomes.

Source/WebCore:

This patch adds the willRespondToMouseClickEvents and willRespondToMouseMoveEvents
functions, named after similar functions in the iOS fork of WebKit.

Touch-adjustment will now use these two functions, which improves accuracy when
targeting shadow-dom elements, and also means touch adjustment will prefer enabled
form-elements over disabled ones.

Test: touchadjustment/disabled-formelements.html

* dom/Node.cpp:
(WebCore::Node::willRespondToMouseMoveEvents):
(WebCore::Node::willRespondToMouseClickEvents):
* dom/Node.h:
(Node):
* editing/DeleteButton.h:
* html/HTMLAnchorElement.cpp:
(WebCore::HTMLAnchorElement::willRespondToMouseClickEvents):
* html/HTMLAnchorElement.h:
(HTMLAnchorElement):
* html/HTMLButtonElement.cpp:
(WebCore::HTMLButtonElement::willRespondToMouseClickEvents):
* html/HTMLButtonElement.h:
(HTMLButtonElement):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::willRespondToMouseClickEvents):
* html/HTMLInputElement.h:
(HTMLInputElement):
* html/HTMLLabelElement.cpp:
(WebCore::HTMLLabelElement::willRespondToMouseClickEvents):
* html/HTMLLabelElement.h:
(HTMLLabelElement):
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::willRespondToMouseClickEvents):
* html/HTMLMediaElement.h:
* html/HTMLSummaryElement.cpp:
(WebCore::HTMLSummaryElement::willRespondToMouseClickEvents):
* html/HTMLSummaryElement.h:
(HTMLSummaryElement):
* html/shadow/CalendarPickerElement.cpp:
(WebCore::CalendarPickerElement::willRespondToMouseClickEvents):
* html/shadow/CalendarPickerElement.h:
(CalendarPickerElement):
* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlTimelineElement::willRespondToMouseClickEvents):
(WebCore::MediaControlTimelineElement::setPosition):
(WebCore::MediaControlVolumeSliderElement::willRespondToMouseMoveEvents):
(WebCore::MediaControlVolumeSliderElement::willRespondToMouseClickEvents):
* html/shadow/MediaControlElements.h:
(MediaControlTimelineElement):
(MediaControlVolumeSliderElement):
* html/shadow/MediaControlRootElement.h:
* html/shadow/SliderThumbElement.cpp:
(WebCore::SliderThumbElement::willRespondToMouseMoveEvents):
(WebCore::SliderThumbElement::willRespondToMouseClickEvents):
* html/shadow/SliderThumbElement.h:
(SliderThumbElement):
* html/shadow/TextControlInnerElements.cpp:
(WebCore::SearchFieldResultsButtonElement::willRespondToMouseClickEvents):
(WebCore::SearchFieldCancelButtonElement::willRespondToMouseClickEvents):
(WebCore::SpinButtonElement::willRespondToMouseMoveEvents):
(WebCore::SpinButtonElement::willRespondToMouseClickEvents):
(WebCore::InputFieldSpeechButtonElement::willRespondToMouseClickEvents):
* html/shadow/TextControlInnerElements.h:
(SearchFieldResultsButtonElement):
(SearchFieldCancelButtonElement):
(SpinButtonElement):
* html/shadow/TextFieldDecorationElement.cpp:
(WebCore::TextFieldDecorationElement::willRespondToMouseClickEvents):
* html/shadow/TextFieldDecorationElement.h:
(TextFieldDecorationElement):
* page/TouchAdjustment.cpp:
(WebCore::TouchAdjustment::nodeRespondsToTapGesture):

LayoutTests:

Tests that enabled formelements are now preferred over disabled formelements.

* touchadjustment/disabled-formelements-expected.txt: Added.
* touchadjustment/disabled-formelements.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124022 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: Content size variables in NetworkResourcesData should have size_t...
vsevik@chromium.org [Mon, 30 Jul 2012 13:29:41 +0000 (13:29 +0000)]
Web Inspector: Content size variables in NetworkResourcesData should have size_t type instead of int.
https://bugs.webkit.org/show_bug.cgi?id=92618

Reviewed by Yury Semikhatsky.

Source/WebCore:

Changed content size variables type in NetworkResourcesData from int to size_t.
Drive-by: Switched String length calculation to recently added StringImpl's sizeInBytes.
* inspector/NetworkResourcesData.cpp:
(WebCore::NetworkResourcesData::ResourceData::removeContent):
(WebCore::NetworkResourcesData::ResourceData::dataLength):
(WebCore::NetworkResourcesData::ResourceData::appendData):
(WebCore::NetworkResourcesData::ResourceData::decodeDataToContent):
(WebCore::NetworkResourcesData::setResourceContent):
(WebCore::NetworkResourcesData::maybeAddResourceData):
(WebCore::NetworkResourcesData::maybeDecodeDataToContent):
(WebCore::NetworkResourcesData::setResourcesDataSizeLimits):
(WebCore::NetworkResourcesData::ensureFreeSpace):
* inspector/NetworkResourcesData.h:
(ResourceData):
(NetworkResourcesData):

LayoutTests:

* http/tests/inspector/network/network-content-replacement-embed-expected.txt:
* http/tests/inspector/network/network-content-replacement-embed.html:
* http/tests/inspector/network/network-content-replacement-xhr-expected.txt:
* http/tests/inspector/network/network-content-replacement-xhr.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124021 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt][NRWT] REGRESSION(123729): Forcing pixel tests with -p doesn't work
ossy@webkit.org [Mon, 30 Jul 2012 13:21:24 +0000 (13:21 +0000)]
[Qt][NRWT] REGRESSION(123729): Forcing pixel tests with -p doesn't work
https://bugs.webkit.org/show_bug.cgi?id=92627

Patch by Balazs Kelemen <kbalazs@webkit.org> on 2012-07-30
Reviewed by Noam Rosenthal.

Remove the default behavior I added in r123729 so that we run all tests as pixel
test with -p. It was intentional but it seems like we don't have consensus on
that we want it.

* Scripts/webkitpy/layout_tests/port/qt.py:
(QtPort._supports_switching_pixel_tests_per_test):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124020 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRemove an useless member variable, m_shouldPreventDispatch, from EventDispatcher.
hayato@chromium.org [Mon, 30 Jul 2012 13:20:59 +0000 (13:20 +0000)]
Remove an useless member variable, m_shouldPreventDispatch, from EventDispatcher.
https://bugs.webkit.org/show_bug.cgi?id=92625

Reviewed by Hajime Morita.

EventDispatcher::m_shouldPreventDispatch does not have any effects. We can remove that.

No new tests, no behavior change.

* dom/EventDispatcher.cpp:
(WebCore::EventDispatcher::EventDispatcher):
(WebCore::EventDispatcher::dispatchEvent):
* dom/EventDispatcher.h:
(EventDispatcher):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124019 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt][WK1] Gardening after r123988. It made fast/writing-mode/flipped-blocks-hit-test...
ossy@webkit.org [Mon, 30 Jul 2012 13:17:13 +0000 (13:17 +0000)]
[Qt][WK1] Gardening after r123988. It made fast/writing-mode/flipped-blocks-hit-test-line-edges.html fail
https://bugs.webkit.org/show_bug.cgi?id=92636

Patch by János Badics <jbadics@inf.u-szeged.hu> on 2012-07-30
Reviewed by Csaba Osztrogonác.

* platform/qt-4.8/Skipped:
* platform/qt-5.0-wk1/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124018 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: make MemoryInstrumentation methods private
yurys@chromium.org [Mon, 30 Jul 2012 12:59:12 +0000 (12:59 +0000)]
Web Inspector: make MemoryInstrumentation methods private
https://bugs.webkit.org/show_bug.cgi?id=92379

Reviewed by Vsevolod Vlasov.

MemoryInstrumentation was refactored.
- All reporting methods on MemoryInstrumentation except add*Object are
  private to enforce MemoryClassInfo usage.
- MemoryInstrumentation::Add*Member methods renamed to Add*Object
- String reporting is implemented directly in the MemoryInstrumentation,
  not in its descendants.

* CMakeLists.txt:
* GNUmakefile.list.am:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/v8/ScriptProfiler.cpp:
(WebCore::ScriptProfiler::collectBindingMemoryInfo):
* dom/MemoryInstrumentation.cpp: Added.
(WebCore):
(WebCore::MemoryInstrumentation::addString):
* dom/MemoryInstrumentation.h:
(WebCore::MemoryInstrumentation::addInstrumentedObject):
(WebCore::MemoryInstrumentation::addObject):
(MemoryInstrumentation):
(InstrumentedPointerBase):
(WebCore::MemoryInstrumentation::InstrumentedPointerBase::~InstrumentedPointerBase):
(InstrumentedPointer):
(WebCore::MemoryInstrumentation::InstrumentedPointer::InstrumentedPointer):
(WebCore::MemoryInstrumentation::OwningTraits::addInstrumentedObject):
(WebCore::MemoryInstrumentation::OwningTraits::addObject):
(WebCore::MemoryInstrumentation::addInstrumentedObjectImpl):
(WebCore::MemoryClassInfo::addInstrumentedMember):
(WebCore::MemoryClassInfo::addMember):
(WebCore):
(WebCore::MemoryInstrumentation::addObjectImpl):
(WebCore::MemoryInstrumentation::addInstrumentedCollection):
(WebCore::MemoryInstrumentation::calculateContainerSize):
* inspector/InspectorMemoryAgent.cpp:
(WebCore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124017 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: [Styles] No color swatch for the SVG "fill" and "stroke" property...
apavlov@chromium.org [Mon, 30 Jul 2012 12:51:04 +0000 (12:51 +0000)]
Web Inspector: [Styles] No color swatch for the SVG "fill" and "stroke" property values
https://bugs.webkit.org/show_bug.cgi?id=92638

Reviewed by Vsevolod Vlasov.

Added the "fill" and "stroke" properties to the list of color-aware properties.

* inspector/front-end/CSSKeywordCompletions.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124016 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r124000.
vsevik@chromium.org [Mon, 30 Jul 2012 12:21:00 +0000 (12:21 +0000)]
Unreviewed, rolling out r124000.
http://trac.webkit.org/changeset/124000
https://bugs.webkit.org/show_bug.cgi?id=92632

seems to have broken chromium Range_InstanceSizeUnknown unit
test across many platforms (Requested by tomhudson on
#webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-07-30

.:

* Source/autotools/symbols.filter:

Source/WebCore:

* WebCore.exp.in:
* inspector/InspectorController.cpp:
(WebCore::InspectorController::InspectorController):
(WebCore::InspectorController::inspectedPageDestroyed):
* inspector/InspectorInstrumentation.cpp:
(WebCore):
* inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
* inspector/InspectorResourceAgent.cpp:
(WebCore::InspectorResourceAgent::didReceiveData):
(WebCore::InspectorResourceAgent::getResponseBody):
* inspector/InspectorResourceAgent.h:
(InspectorResourceAgent):
* inspector/NetworkResourcesData.cpp:
(WebCore::NetworkResourcesData::ResourceData::ResourceData):
(WebCore::NetworkResourcesData::ResourceData::setContent):
(WebCore::NetworkResourcesData::setResourceContent):
* inspector/NetworkResourcesData.h:
(ResourceData):
(WebCore::NetworkResourcesData::ResourceData::cachedResource):
(NetworkResourcesData):
* loader/cache/CachedResource.cpp:
(WebCore::CachedResource::~CachedResource):
(WebCore::CachedResource::removeClient):
(WebCore::CachedResource::deleteIfPossible):
(WebCore):
* loader/cache/CachedResource.h:
(CachedResource):
* loader/cache/CachedResourceLoader.cpp:
(WebCore::CachedResourceLoader::garbageCollectDocumentResourcesTimerFired):
(WebCore::CachedResourceLoader::clearPreloads):
* loader/cache/CachedResourceLoader.h:
(CachedResourceLoader):
* loader/cache/MemoryCache.cpp:
(WebCore::MemoryCache::evict):
* testing/InternalSettings.cpp:
* testing/Internals.cpp:
* testing/Internals.h:
(Internals):
* testing/Internals.idl:

Source/WebKit2:

* win/WebKit2.def:
* win/WebKit2CFLite.def:

LayoutTests:

* http/tests/inspector/network/cached-resource-destroyed-moved-to-storage-expected.txt: Removed.
* http/tests/inspector/network/cached-resource-destroyed-moved-to-storage.html: Removed.
* http/tests/inspector/network/cached-resource-destroyed-too-big-discarded-expected.txt: Removed.
* http/tests/inspector/network/cached-resource-destroyed-too-big-discarded.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124015 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRemove an unused member variable, m_originalTarget, from EventDispatcher.
hayato@chromium.org [Mon, 30 Jul 2012 11:38:33 +0000 (11:38 +0000)]
Remove an unused member variable, m_originalTarget, from EventDispatcher.
https://bugs.webkit.org/show_bug.cgi?id=92614

Reviewed by Andreas Kling.

No one uses EventDispatcher::m_originalTarget.

No new tests, no behavior change.

* dom/EventDispatcher.h:
(EventDispatcher):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124014 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Remove accurateVisibleContentsPosition and use a FloatRect for visibleContentsRe...
jocelyn.turcotte@nokia.com [Mon, 30 Jul 2012 11:33:41 +0000 (11:33 +0000)]
[Qt] Remove accurateVisibleContentsPosition and use a FloatRect for visibleContentsRect instead
https://bugs.webkit.org/show_bug.cgi?id=92536

Reviewed by Noam Rosenthal.

AccurateVisibleContentsPosition carries the same information as visibleContentsRect.location() and
is needed for fixed layer position updates.
This patch instead uses a FloatRect for the visible contents rect from the web view down to the
LayerCoordinatorProxy, and converts it to an IntRect at this moment, before sending it to the
web process.
This also moves the check to prevent sending the same rect to the moment of the conversion
in LayerTreeCoordinatorProxy to fix a bug where the fixed layer deltas wouldn't be updated
unless the integer part of position components changed.

* UIProcess/API/qt/qquickwebview.cpp:
(QQuickWebViewLegacyPrivate::updateViewportSize):
* UIProcess/DrawingAreaProxy.h:
(WebKit::DrawingAreaProxy::setVisibleContentsRect):
* UIProcess/DrawingAreaProxyImpl.cpp:
(WebKit::DrawingAreaProxyImpl::setVisibleContentsRect):
* UIProcess/DrawingAreaProxyImpl.h:
(DrawingAreaProxyImpl):
* UIProcess/LayerTreeCoordinatorProxy.cpp:
(WebKit::LayerTreeCoordinatorProxy::LayerTreeCoordinatorProxy):
(WebKit::LayerTreeCoordinatorProxy::setVisibleContentsRect):
* UIProcess/LayerTreeCoordinatorProxy.h:
(LayerTreeCoordinatorProxy):
* UIProcess/WebLayerTreeRenderer.cpp:
(WebKit::WebLayerTreeRenderer::WebLayerTreeRenderer):
(WebKit::WebLayerTreeRenderer::setVisibleContentsRect):
(WebKit::WebLayerTreeRenderer::adjustPositionForFixedLayers):
* UIProcess/WebLayerTreeRenderer.h:
(WebLayerTreeRenderer):
* UIProcess/qt/QtViewportHandler.cpp:
(WebKit::QtViewportHandler::visibleContentsRect):
(WebKit::QtViewportHandler::informVisibleContentChange):
* UIProcess/qt/QtViewportHandler.h:
(QtViewportHandler):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124013 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Fix issues remaining with fixed positioned layers
jocelyn.turcotte@nokia.com [Mon, 30 Jul 2012 11:31:27 +0000 (11:31 +0000)]
[Qt] Fix issues remaining with fixed positioned layers
https://bugs.webkit.org/show_bug.cgi?id=92528

Reviewed by Noam Rosenthal.

Source/WebCore:

* platform/graphics/texmap/TextureMapperLayer.cpp:
(WebCore::TextureMapperLayer::syncCompositingStateSelf):
Make sure that the delta is used when syncCompositingStateSelf is called afterward.
(WebCore::TextureMapperLayer::setScrollPositionDeltaIfNeeded):
* platform/graphics/texmap/TextureMapperLayer.h:
Use a FloatSize instead of an IntPoint for the delta.
(TextureMapperLayer):
(WebCore::TextureMapperLayer::adjustedPosition):

Source/WebKit2:

This partially fixes jittering of fixed layers when zoomed in and the
blurriness of rendered tiles of the non-composited contents layer.

* UIProcess/WebLayerTreeRenderer.cpp:
(WebKit::boundedScrollPosition):
(WebKit::WebLayerTreeRenderer::paintToCurrentGLContext):
Remove the adjustment on the root layer. This cancels setPixelAligned and causes the blurriness.
(WebKit::WebLayerTreeRenderer::adjustPositionForFixedLayers):
Use m_accurateVisibleContentsPosition to calculate the delta.
(WebKit::WebLayerTreeRenderer::didChangeScrollPosition):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124012 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[DRT] LTC:: pageNumberForElementById() could be moved to Internals
kaustubh@motorola.com [Mon, 30 Jul 2012 10:33:43 +0000 (10:33 +0000)]
[DRT] LTC:: pageNumberForElementById() could be moved to Internals
https://bugs.webkit.org/show_bug.cgi?id=92091

Reviewed by Hajime Morita.

Move the pageNumberForElementById from LayoutTestCotroller to Internals and
remove the old platform specific implementations as it exclusively tests WebCore functionality.

* public/WebFrame.h:
(WebFrame):
* src/WebFrameImpl.cpp:
(WebKit):
* src/WebFrameImpl.h:
(WebFrameImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124011 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoReplace UnicodeWinCE with UnicodeWchar https://bugs.webkit.org/show_bug.cgi?id=92539
paroga@webkit.org [Mon, 30 Jul 2012 10:25:41 +0000 (10:25 +0000)]
Replace UnicodeWinCE with UnicodeWchar https://bugs.webkit.org/show_bug.cgi?id=92539

Reviewed by Ryosuke Niwa.

UnicodeWinCE never contained WinCE specific code. UnicodeWchar
is a replacement for it, which is mainly based on the functions
from <wchar.h>. It is ment as a minimal Unicode backend, which
can be used very easy and has no external dependencies.

.:

* Source/cmake/OptionsWinCE.cmake:

Source/WebCore:

* platform/text/TextEncoding.cpp:
(WebCore::TextEncoding::encode):
* platform/text/TextEncodingRegistry.cpp:
(WebCore::extendTextCodecMaps):

Source/WebKit/blackberry:

* WebCoreSupport/AboutDataUseFeatures.in:

Source/WTF:

* WTF.gyp/WTF.gyp:
* WTF.gypi:
* wtf/Platform.h:
* wtf/PlatformWinCE.cmake:
* wtf/unicode/Unicode.h:
* wtf/unicode/wchar/UnicodeWchar.cpp: Added.
(Unicode):
(WTF::Unicode::category):
(WTF::Unicode::combiningClass):
(WTF::Unicode::direction):
(WTF::Unicode::decompositionType):
(WTF::Unicode::hasLineBreakingPropertyComplexContext):
(WTF::Unicode::mirroredChar):
(WTF::Unicode::convertWithFunction):
(WTF::Unicode::foldCase):
(WTF::Unicode::toLower):
(WTF::Unicode::toUpper):
* wtf/unicode/wchar/UnicodeWchar.h: Added.
(Unicode):
(WTF::Unicode::isAlphanumeric):
(WTF::Unicode::isDigit):
(WTF::Unicode::isLetter):
(WTF::Unicode::isLower):
(WTF::Unicode::isPrintableChar):
(WTF::Unicode::isPunct):
(WTF::Unicode::isSpace):
(WTF::Unicode::isUpper):
(WTF::Unicode::isArabicChar):
(WTF::Unicode::isSeparatorSpace):
(WTF::Unicode::foldCase):
(WTF::Unicode::toLower):
(WTF::Unicode::toUpper):
(WTF::Unicode::toTitleCase):
(WTF::Unicode::umemcasecmp):
* wtf/unicode/wince/UnicodeWinCE.cpp: Removed.
* wtf/unicode/wince/UnicodeWinCE.h: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124010 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRemove unnecessary code which set event's target from EventDispatcher::dispatchEvent.
hayato@chromium.org [Mon, 30 Jul 2012 09:45:47 +0000 (09:45 +0000)]
Remove unnecessary code which set event's target from EventDispatcher::dispatchEvent.
https://bugs.webkit.org/show_bug.cgi?id=92613

Reviewed by Hajime Morita.

EventContext::handleLocalEvents does all required jobs so we don't
need to set target in EventDispacher::dispatchEvent() for TARGET phase.

No new tests, no behavior change.

* dom/EventDispatcher.cpp:
(WebCore::EventDispatcher::dispatchEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124009 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r124004.
commit-queue@webkit.org [Mon, 30 Jul 2012 09:29:45 +0000 (09:29 +0000)]
Unreviewed, rolling out r124004.
http://trac.webkit.org/changeset/124004
https://bugs.webkit.org/show_bug.cgi?id=92622

Broke Android build (Requested by keishi on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-07-30

Source/Platform:

* chromium/public/WebLocalizedString.h:

Source/WebCore:

* rendering/RenderThemeChromiumCommon.cpp:
(WebCore::RenderThemeChromiumCommon::supportsDataListUI):

Source/WebKit/chromium:

* WebKit.gyp:
* src/ChromeClientImpl.cpp:
(WebKit::ChromeClientImpl::createColorChooser):
* src/ChromeClientImpl.h:
(WebCore):
(ChromeClientImpl):
* src/ColorChooserProxy.cpp: Copied from Source/WebKit/chromium/src/ColorChooserUIController.h.
(WebKit):
(WebKit::ColorChooserProxy::ColorChooserProxy):
(WebKit::ColorChooserProxy::~ColorChooserProxy):
(WebKit::ColorChooserProxy::setSelectedColor):
(WebKit::ColorChooserProxy::endChooser):
* src/ColorChooserProxy.h: Copied from Source/WebKit/chromium/src/ColorChooserUIController.h.
(WebKit):
(ColorChooserProxy):
* src/ColorChooserUIController.cpp: Removed.
* src/WebColorChooserClientImpl.cpp: Copied from Source/WebKit/chromium/src/ColorChooserUIController.h.
(WebKit):
(WebKit::WebColorChooserClientImpl::WebColorChooserClientImpl):
(WebKit::WebColorChooserClientImpl::~WebColorChooserClientImpl):
(WebKit::WebColorChooserClientImpl::didChooseColor):
(WebKit::WebColorChooserClientImpl::didEndChooser):
* src/WebColorChooserClientImpl.h: Renamed from Source/WebKit/chromium/src/ColorChooserUIController.h.
(WebCore):
(WebKit):
(WebColorChooserClientImpl):

LayoutTests:

* platform/chromium-mac/platform/chromium/fast/forms/color/color-suggestion-picker-appearance-expected.png: Removed.
* platform/chromium/TestExpectations:
* platform/chromium/fast/forms/color/color-suggestion-picker-appearance-expected.txt: Removed.
* platform/chromium/fast/forms/color/color-suggestion-picker-appearance.html: Removed.
* platform/chromium/fast/forms/datalist/input-list-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124008 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: element tag disappears after editing
pfeldman@chromium.org [Mon, 30 Jul 2012 09:25:21 +0000 (09:25 +0000)]
Web Inspector: element tag disappears after editing
https://bugs.webkit.org/show_bug.cgi?id=92385

Reviewed by Vsevolod Vlasov.

This has regressed when we started hiding elements panel upon
batch updates. Narrowing hiding cases, updating selection
in on-screen mode only.

* inspector/front-end/ElementsTreeOutline.js:
(WebInspector.ElementsTreeElement.prototype.updateSelection):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124007 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: move StringImpl size calculation to StringImpl
yurys@chromium.org [Mon, 30 Jul 2012 09:05:11 +0000 (09:05 +0000)]
Web Inspector: move StringImpl size calculation to StringImpl
https://bugs.webkit.org/show_bug.cgi?id=92359

Reviewed by Pavel Feldman.

Moved stringSize(StringImpl*) implementation from InspectorMemoryAgent to
StringImpl::sizeInBytes();

Source/WebCore:

* inspector/InspectorMemoryAgent.cpp:
(WebCore):

Source/WTF:

* wtf/text/StringImpl.cpp:
(WTF::StringImpl::sizeInBytes):
(WTF):
* wtf/text/StringImpl.h:
(StringImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124006 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: workers inspector is broken in the remote debugging mode
pfeldman@chromium.org [Mon, 30 Jul 2012 08:57:44 +0000 (08:57 +0000)]
Web Inspector: workers inspector is broken in the remote debugging mode
https://bugs.webkit.org/show_bug.cgi?id=92540

Reviewed by Yury Semikhatsky.

The problem was that worker manager re-used complete url with query parameters
for spin-off worker front-ends. As a result, in case of remote debugging, newly
created front-ends were connecting to the same web socket (because of the ws parameter).

* inspector/front-end/WorkerManager.js:
(WebInspector.WorkerManager.isDedicatedWorkerFrontend):
* inspector/front-end/inspector.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124005 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoImplement datalist UI for input type color for Chromium
keishi@webkit.org [Mon, 30 Jul 2012 08:33:24 +0000 (08:33 +0000)]
Implement datalist UI for input type color for Chromium
https://bugs.webkit.org/show_bug.cgi?id=92075

Reviewed by Kent Tamura.

Source/Platform:

* chromium/public/WebLocalizedString.h: Added OtherColorLabel.

Source/WebCore:

Test: platform/chromium/fast/forms/color/color-suggestion-picker-appearance.html

* rendering/RenderThemeChromiumCommon.cpp:
(WebCore::RenderThemeChromiumCommon::supportsDataListUI): Added color to types supporting datalist.

Source/WebKit/chromium:

Integrated ColorChooserProxy and WebColorChooserClientImpl into ColorChooserUIController.
ColorChooserUIController can control whether to open the color suggestion picker
popup or the color chooser depending on the responses from ColorChooserClient.

* WebKit.gyp:
* src/ChromeClientImpl.cpp:
(WebKit::ChromeClientImpl::createColorChooser): Returns ColorChooserUIController which is a ColorChooser.
(WebKit::ChromeClientImpl::createWebColorChooser): Returns WebColorChooser that was created by Chromium.
* src/ChromeClientImpl.h:
(WebKit):
(ChromeClientImpl):
* src/ColorChooserProxy.cpp: Removed.
* src/ColorChooserProxy.h: Removed.
* src/ColorChooserUIController.cpp: Added.
(WebKit):
(WebKit::ColorChooserUIController::ColorChooserUIController): Controls the UI for color chooser. Opens the color suggestion picker popup or color chooser depending on the ColorChooserClient.
(WebKit::ColorChooserUIController::~ColorChooserUIController):
(WebKit::ColorChooserUIController::setSelectedColor):
(WebKit::ColorChooserUIController::endChooser): This is called by WebCore so we close both the popup and the color chooser.
(WebKit::ColorChooserUIController::didChooseColor):
(WebKit::ColorChooserUIController::didEndChooser):
(WebKit::ColorChooserUIController::contentSize): The popup will be resized later from JS so we return an arbitrary size.
(WebKit::ColorChooserUIController::writeDocument):
(WebKit::ColorChooserUIController::setValueAndClosePopup): Performs action based on the numValue.
(WebKit::ColorChooserUIController::didClosePopup):
(WebKit::ColorChooserUIController::openPopup): Opens color chooser suggestion popup.
(WebKit::ColorChooserUIController::closePopup):
(WebKit::ColorChooserUIController::openColorChooser): Opens color chooser.
* src/ColorChooserUIController.h: Renamed from Source/WebKit/chromium/src/WebColorChooserClientImpl.h.
(WebCore):
(WebKit):
(ColorChooserUIController):
* src/WebColorChooserClientImpl.cpp: Removed.

LayoutTests:

* platform/chromium-mac/platform/chromium/fast/forms/color/color-suggestion-picker-appearance-expected.png: Added.
* platform/chromium/TestExpectations:
* platform/chromium/fast/forms/color/color-suggestion-picker-appearance-expected.txt: Added.
* platform/chromium/fast/forms/color/color-suggestion-picker-appearance.html: Added. Tests color suggestion picker popup appearance.
* platform/chromium/fast/forms/datalist/input-list-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124004 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION (r123767): platform/gtk/accessibility/object-with-title.html failing on GTK
mario@webkit.org [Mon, 30 Jul 2012 08:28:07 +0000 (08:28 +0000)]
REGRESSION (r123767): platform/gtk/accessibility/object-with-title.html failing on GTK
https://bugs.webkit.org/show_bug.cgi?id=92477

Reviewed by Chris Fleizach.

Source/WebCore:

Consider title and accessibilityDescription together with helpText to
decide whether an object should be exposed in the accessibility tree.

* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::accessibilityIsIgnored):

LayoutTests:

Remove passing test from TestExpectations.

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124003 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed Qt/Mac build fix: Use libWebKitSystemInterfaceMountainLion.a
hausmann@webkit.org [Mon, 30 Jul 2012 08:18:07 +0000 (08:18 +0000)]
Unreviewed Qt/Mac build fix: Use libWebKitSystemInterfaceMountainLion.a
on Mountain Lion.

* WebKit1.pri:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124002 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WTR] Enforce SHIFT modifier to upper case keys in event sender.
commit-queue@webkit.org [Mon, 30 Jul 2012 07:58:25 +0000 (07:58 +0000)]
[EFL][WTR] Enforce SHIFT modifier to upper case keys in event sender.
https://bugs.webkit.org/show_bug.cgi?id=92366

Patch by Mikhail Pozdnyakov <mikhail.pozdnyakov@intel.com> on 2012-07-30
Reviewed by Antonio Gomes.

Add SHIFT modifier for capital letters as VK_
codes do not distinguish between lows and caps and also to be consistent
with other ports which send caps as lowercase letters with the shift key down.

* WebKitTestRunner/efl/EventSenderProxyEfl.cpp:
(WTR::EventSenderProxy::keyDown):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124001 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: Resource agent's reference to cached resources should be weak.
vsevik@chromium.org [Mon, 30 Jul 2012 07:54:54 +0000 (07:54 +0000)]
Web Inspector: Resource agent's reference to cached resources should be weak.
https://bugs.webkit.org/show_bug.cgi?id=92108

Reviewed by Pavel Feldman.

.:

* Source/autotools/symbols.filter:

Source/WebCore:

NetworkResourcesData now stores a raw pointer to CachedResource.
CachedResource now notifies InspectorInstrumentation that it will be destroyed.
InspectorInstrumentation stores a set of InstrumentingAgents and broadcasts willDestroyCachedResource event to all available resourceAgents.
Destroyed resources content is saved to NetworkResourcesData.

Tests: http/tests/inspector/network/cached-resource-destroyed-moved-to-storage.html
       http/tests/inspector/network/cached-resource-destroyed-too-big-discarded.html

* WebCore.exp.in:
* inspector/InspectorController.cpp:
(WebCore::InspectorController::InspectorController):
(WebCore::InspectorController::inspectedPageDestroyed):
* inspector/InspectorInstrumentation.cpp:
(WebCore):
(WebCore::InspectorInstrumentation::willDestroyCachedResourceImpl):
(WebCore::InspectorInstrumentation::registerInstrumentingAgents):
(WebCore::InspectorInstrumentation::unregisterInstrumentingAgents):
* inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
(WebCore::InspectorInstrumentation::willDestroyCachedResource):
(WebCore):
* inspector/InspectorResourceAgent.cpp:
(WebCore::InspectorResourceAgent::didReceiveData):
(WebCore::InspectorResourceAgent::willDestroyCachedResource):
(WebCore):
(WebCore::InspectorResourceAgent::getResponseBody):
* inspector/InspectorResourceAgent.h:
(InspectorResourceAgent):
* inspector/NetworkResourcesData.cpp:
(WebCore::NetworkResourcesData::ResourceData::ResourceData):
(WebCore::NetworkResourcesData::ResourceData::setContent):
(WebCore::NetworkResourcesData::setResourceContent):
(WebCore::NetworkResourcesData::removeCachedResource):
(WebCore):
* inspector/NetworkResourcesData.h:
(ResourceData):
(WebCore::NetworkResourcesData::ResourceData::base64Encoded):
(WebCore::NetworkResourcesData::ResourceData::cachedResource):
(NetworkResourcesData):
* loader/cache/CachedResource.cpp:
(WebCore::CachedResource::~CachedResource):
(WebCore::CachedResource::removeClient):
(WebCore::CachedResource::deleteIfPossible):
(WebCore):
* loader/cache/CachedResource.h:
(CachedResource):
* loader/cache/CachedResourceLoader.cpp:
(WebCore::CachedResourceLoader::garbageCollectDocumentResourcesTimerFired):
(WebCore::CachedResourceLoader::garbageCollectDocumentResources):
(WebCore::CachedResourceLoader::clearPreloads):
* loader/cache/CachedResourceLoader.h:
(CachedResourceLoader):
* loader/cache/MemoryCache.cpp:
(WebCore::MemoryCache::evict):
* testing/InternalSettings.cpp:
* testing/Internals.cpp:
(WebCore::Internals::garbageCollectDocumentResources):
(WebCore):
* testing/Internals.h:
(Internals):
* testing/Internals.idl:

Source/WebKit2:

* win/WebKit2.def:
* win/WebKit2CFLite.def:

LayoutTests:

* http/tests/inspector/network/cached-resource-destroyed-moved-to-storage-expected.txt: Added.
* http/tests/inspector/network/cached-resource-destroyed-moved-to-storage.html: Added.
* http/tests/inspector/network/cached-resource-destroyed-too-big-discarded-expected.txt: Added.
* http/tests/inspector/network/cached-resource-destroyed-too-big-discarded.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@124000 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WK2][EFL] Accelerated compositing support using TextureMapper on Efl WebKit2
commit-queue@webkit.org [Mon, 30 Jul 2012 06:25:19 +0000 (06:25 +0000)]
[WK2][EFL] Accelerated compositing support using TextureMapper on Efl WebKit2
https://bugs.webkit.org/show_bug.cgi?id=91583

Patch by YoungTaeck Song <youngtaeck.song@samsung.com> on 2012-07-29
Reviewed by Noam Rosenthal.

This patch is a subset of Efl's UI_SIDE_COMPOSITING implementation.
This patch adds Texture Mapper related files to PlatformEfl.cmake.
And wrapped TextureMapperLayer and TextureMapperBackingStore with USE(ACCELERATED_COMPOSITING) for fixing building error when disable ACCELERATED_COMPOSITING.

* CMakeLists.txt:
* PlatformEfl.cmake:
* platform/graphics/GraphicsLayer.cpp:
(WebCore):
* platform/graphics/GraphicsLayer.h:
(GraphicsLayer):
* platform/graphics/GraphicsLayerAnimation.cpp:
* platform/graphics/GraphicsLayerAnimation.h:
* platform/graphics/PlatformLayer.h:
(WebCore):
* platform/graphics/texmap/TextureMapperBackingStore.cpp:
* platform/graphics/texmap/TextureMapperBackingStore.h:
* platform/graphics/texmap/TextureMapperLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123999 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WK2][EFL] TILED_BACKING_STORE support on Efl WebKit2
commit-queue@webkit.org [Mon, 30 Jul 2012 06:09:55 +0000 (06:09 +0000)]
[WK2][EFL] TILED_BACKING_STORE support on Efl WebKit2
https://bugs.webkit.org/show_bug.cgi?id=91582

Patch by YoungTaeck Song <youngtaeck.song@samsung.com> on 2012-07-29
Reviewed by Noam Rosenthal.

Source/WebCore:

This patch is a subset of Efl's UI_SIDE_COMPOSITING implementation.
This patch adds TILED_BACKING_STORE related files to CMakeLists.txt and PlatformEfl.cmake

* CMakeLists.txt:
* PlatformEfl.cmake:

Source/WebKit/efl:

Add a dummy function for fixing building error when enabling TILED_BACKING_STORE.

* WebCoreSupport/ChromeClientEfl.cpp:
(WebCore):
(WebCore::ChromeClientEfl::delegatedScrollRequested):
* WebCoreSupport/ChromeClientEfl.h:
(ChromeClientEfl):

Source/WebKit2:

This patch is a subset of Efl's UI_SIDE_COMPOSITING implementation.
Add a dummy function for fixing building error when enabling TILED_BACKING_STORE.

* UIProcess/API/efl/PageClientImpl.cpp:
(WebKit):
(WebKit::PageClientImpl::pageDidRequestScroll):
* UIProcess/API/efl/PageClientImpl.h:
(PageClientImpl):
* WebProcess/WebCoreSupport/WebChromeClient.cpp:
(WebKit::WebChromeClient::contentsSizeChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123998 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix removing invalid values from color input suggestions
keishi@webkit.org [Mon, 30 Jul 2012 05:49:29 +0000 (05:49 +0000)]
Fix removing invalid values from color input suggestions
https://bugs.webkit.org/show_bug.cgi?id=92502

Reviewed by Kent Tamura.

Make ColorInputType::typeMismatchFor() return false for invalid color
values so that they don't appear in the list of suggestions that
ColorInputType::suggestions() returns.

No new tests. Tests to be added in Bug 92075.

* html/ColorInputType.cpp:
(WebCore::ColorInputType::typeMismatchFor):
(WebCore):
* html/ColorInputType.h:
(ColorInputType):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123997 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agogetChannelData should raise exception when index is more than numberOfChannels.
commit-queue@webkit.org [Mon, 30 Jul 2012 05:28:45 +0000 (05:28 +0000)]
getChannelData should raise exception when index is more than numberOfChannels.
https://bugs.webkit.org/show_bug.cgi?id=92223

Patch by Li Yin <li.yin@intel.com> on 2012-07-29
Reviewed by Kentaro Hara.

Source/WebCore:

Spec: https://dvcs.w3.org/hg/audio/raw-file/tip/webaudio/specification.html#AudioBuffer-section
In getChannelData function, the index value MUST be less than numberOfChannels
or an exception will be thrown.

Test: webaudio/audiobuffer.html

* Modules/webaudio/AudioBuffer.cpp:
(WebCore::AudioBuffer::getChannelData):
(WebCore):
* Modules/webaudio/AudioBuffer.h:
(WebCore):
(AudioBuffer):
* Modules/webaudio/AudioBuffer.idl: raise exception when index is not less than numberOfChannels.

LayoutTests:

Spec: https://dvcs.w3.org/hg/audio/raw-file/tip/webaudio/specification.html#AudioBuffer-section
Add test to cover basic attributes of AudioBuffer.

* webaudio/audiobuffer-expected.txt: Added.
* webaudio/audiobuffer.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123996 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoforward-delete in the last cell of a table moves the caret after the table
commit-queue@webkit.org [Mon, 30 Jul 2012 05:22:58 +0000 (05:22 +0000)]
forward-delete in the last cell of a table moves the caret after the table
https://bugs.webkit.org/show_bug.cgi?id=24250

Patch by Sukolsak Sakshuwong <sukolsak@google.com> on 2012-07-29
Reviewed by Ryosuke Niwa.

Source/WebCore:

The ForwardDelete command moved the cursor after the table when it's invoked
in an empty cell that is the last cell of a table. It also removed the break
element when invoked in an empty cell that is not the last cell of a table.
Make it do nothing inside an empty table cell.

Test: editing/deleting/forward-delete-empty-table-cell.html

* editing/TypingCommand.cpp:
(WebCore::TypingCommand::forwardDeleteKeyPressed):

LayoutTests:

* editing/deleting/forward-delete-empty-table-cell-expected.txt: Added.
* editing/deleting/forward-delete-empty-table-cell.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123995 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION(r123993): fast/forms/date/calendar-picker-type-change-onclick.html fails
tkent@chromium.org [Mon, 30 Jul 2012 05:17:30 +0000 (05:17 +0000)]
REGRESSION(r123993): fast/forms/date/calendar-picker-type-change-onclick.html fails

* fast/forms/date/calendar-picker-appearance-expected.txt:
* fast/forms/date/calendar-picker-appearance.html:
Clear webkitTransitionEnd event handler before notifyDone().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123994 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix opacity issue of fast/forms/date/calendar-picker-appearance.html
tkent@chromium.org [Mon, 30 Jul 2012 04:19:10 +0000 (04:19 +0000)]
Fix opacity issue of fast/forms/date/calendar-picker-appearance.html
https://bugs.webkit.org/show_bug.cgi?id=92605

Reviewed by Hajime Morita.

We need to wait for pixel dumping until CSS transition completion for
the 'main' element, which has transition from opacity:0 to opacity:1.

* fast/forms/date/calendar-picker-appearance.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123993 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRemove unused method HTMLConstructionSiteTask::take(HTMLConstructionSiteTask&)
commit-queue@webkit.org [Mon, 30 Jul 2012 01:58:03 +0000 (01:58 +0000)]
Remove unused method HTMLConstructionSiteTask::take(HTMLConstructionSiteTask&)
https://bugs.webkit.org/show_bug.cgi?id=92601

Patch by Kwang Yul Seo <skyul@company100.net> on 2012-07-29
Reviewed by Adam Barth.

This method was introduced in r104000, but has never been used from the beginning.

No functional change.

* html/parser/HTMLConstructionSite.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123992 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] HarfBuzzShaper can't handle segmented text run
bashi@chromium.org [Mon, 30 Jul 2012 01:46:58 +0000 (01:46 +0000)]
[Chromium] HarfBuzzShaper can't handle segmented text run
https://bugs.webkit.org/show_bug.cgi?id=92445

Reviewed by Tony Chang.

Pass the range to be displayed to HarfBuzzShaper. The shaper calculates positions and advances of each glyph
regardless of the range, but only add glyphs which are in the range.

No new tests. platform/chromium-linux/fast/text/international/draw-complex-text-from-to.html should
work as expected once we move to use harfbuzz-ng on linux. Note that the current expectation will need to be rebaselined
because the old hb shaper (ComplexTextController) mishandles the range. |to| should be exclusive. I'll rebaseline
the expectation later.

* platform/graphics/harfbuzz/FontHarfBuzz.cpp:
(WebCore::Font::drawComplexText): Call shaper.setDrawRange().
* platform/graphics/harfbuzz/ng/HarfBuzzShaper.cpp:
(WebCore::HarfBuzzShaper::HarfBuzzShaper):
(WebCore::HarfBuzzShaper::setDrawRange): Added.
(WebCore):
(WebCore::HarfBuzzShaper::shouldDrawCharacterAt): Added.
(WebCore::HarfBuzzShaper::shapeHarfBuzzRuns): Added variables that hold pending advances.
(WebCore::HarfBuzzShaper::setGlyphPositionsForHarfBuzzRun): Add only glyphs which are in the given range to glyphBuffer.
* platform/graphics/harfbuzz/ng/HarfBuzzShaper.h:
(HarfBuzzShaper):
* platform/graphics/mac/FontComplexTextMac.cpp:
(WebCore::Font::drawComplexText): Call shaper.setDrawRange().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123991 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoHit testing in the gap between pages returns incorrect results in flipped blocks...
mitz@apple.com [Mon, 30 Jul 2012 00:44:08 +0000 (00:44 +0000)]
Hit testing in the gap between pages returns incorrect results in flipped blocks writing modes
https://bugs.webkit.org/show_bug.cgi?id=92600

Reviewed by Simon Fraser.

Source/WebCore:

Test: fast/multicol/hit-test-gap-between-pages-flipped.html

* rendering/RenderBlock.h:
(WebCore::RenderBlock): Promoted offsetForContents() from private to protected, and demoted
adjustPointToColumnContents() back to private.
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::hitTest): Changed to flip the point passed to updateHitTestResults().
* rendering/RenderView.cpp:
(WebCore::RenderView::updateHitTestResult): Changed to call offsetForContents(), which, as
of r123977, correctly accounts for flippedness.

LayoutTests:

* fast/multicol/hit-test-gap-between-pages-flipped-expected.txt: Added.
* fast/multicol/hit-test-gap-between-pages-flipped.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123990 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoPropertyNameArray::m_shouldCache is only assigned and never used
fpizlo@apple.com [Sun, 29 Jul 2012 23:15:30 +0000 (23:15 +0000)]
PropertyNameArray::m_shouldCache is only assigned and never used
https://bugs.webkit.org/show_bug.cgi?id=92598

Reviewed by Dan Bernstein.

* runtime/PropertyNameArray.h:
(JSC::PropertyNameArray::PropertyNameArray):
(PropertyNameArray):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123989 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIn flipped blocks, a point on the top edge of a text box is considered outside the...
mitz@apple.com [Sun, 29 Jul 2012 22:55:51 +0000 (22:55 +0000)]
In flipped blocks, a point on the top edge of a text box is considered outside the box (and vice versa)
https://bugs.webkit.org/show_bug.cgi?id=92597

Reviewed by Ryosuke Niwa.

Source/WebCore:

This is similar to <http://webkit.org/b/92593> except for RenderText. Fixing this required
fixing some code that creates a point inside a box by setting its y coordinate to the
logical top of the box. In a flipped block, this would be just outside the box.

Enhanced fast/writing-mode/flipped-blocks-hit-test-line-edges.html to cover this.

* editing/visible_units.cpp:
(WebCore::absoluteLineDirectionPointToLocalPointInBlock): Changed to use
RootInlineBox::blockDirectionPointInLine in the vertical case as well.
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::positionForPointWithInlineChildren): Changed to use
RootInlineBox::blockDirectionPointInLine.
* rendering/RenderText.cpp:
(WebCore::RenderText::positionForPoint): Changed the strictness of y inequalities based on
the flippedness of the block. This fixes the bug.
* rendering/RootInlineBox.cpp:
(WebCore::RootInlineBox::blockDirectionPointInLine): Moved the implementation here from the
header, and made it return the bottom in the flipped block case, so that the returned point
is inside the line in that case as well.
* rendering/RootInlineBox.h:

LayoutTests:

Added hits inside the text to this test.

* fast/writing-mode/flipped-blocks-hit-test-line-edges-expected.txt:
* fast/writing-mode/flipped-blocks-hit-test-line-edges.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123988 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] pass acceleratePainting info to settings
commit-queue@webkit.org [Sun, 29 Jul 2012 22:36:16 +0000 (22:36 +0000)]
[chromium] pass acceleratePainting info to settings
https://bugs.webkit.org/show_bug.cgi?id=92467

Patch by Lu Guanqun <guanqun.lu@intel.com> on 2012-07-29
Reviewed by Eric Seidel.

The acceleratePainting info is not passed into settings correctly so
that GPU composition would not use the code path of accelerated painting.

* src/WebLayerTreeView.cpp:
(WebKit::WebLayerTreeView::Settings::operator CCLayerTreeSettings):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123987 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][jhbuild] Make run-with-jhbuild become the jhbuild process itself.
rakuco@webkit.org [Sun, 29 Jul 2012 22:26:55 +0000 (22:26 +0000)]
[EFL][jhbuild] Make run-with-jhbuild become the jhbuild process itself.
https://bugs.webkit.org/show_bug.cgi?id=92592

Reviewed by Eric Seidel.

Adapt to the changes made to GTK+'s run-with-jhbuild in r123979
and make EFL's run-with-jhbuild call os.execve() instead of
launching a subprocess as well.

This should make it possible to interrupt execution of jhbuild
cleanly.

* efl/run-with-jhbuild:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123986 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agorun-perf-tests should generate a results page
rniwa@webkit.org [Sun, 29 Jul 2012 22:03:35 +0000 (22:03 +0000)]
run-perf-tests should generate a results page
https://bugs.webkit.org/show_bug.cgi?id=92575

Reviewed by Eric Seidel.

PerformanceTests:

Add a results page template. This page includes flot and flot.fillbetween.
jQuery and the json that contains test results are injected by run-perf-tests.

* resources/results-template.html: Added.

Tools:

Added the ability to generate a results page that summarizes performance test results from
multiple runs of run-perf-tests when --output-json-path is specified and --test-results-server
is not specified. We cannot generate a results page when --test-results-server is specified
because perf-o-matic cannot parse the new JSON format.

The new JSON format is simply an array of the old JSON output. This JSON is then merged into
results-template.html along with jQuery and copied as a HTML file of the same filename as
the output JSON file with a .html extension.

We merge all scripts and the JSON output into the results page as opposed to including them
with the src content attribute to make it standalone so that we may post it on Bugzilla or
e-mail it to someone else without breaking it.

* Scripts/webkitpy/performance_tests/perftestsrunner.py:
(PerfTestsRunner.run):
(PerfTestsRunner._generate_json): Merge "contents" with the existing JSON file when generating
a results page, and generate the results page from results-template.html merged with jQuery and
the generated JSON output.
* Scripts/webkitpy/performance_tests/perftestsrunner_unittest.py:
(_test_run_with_json_output): Extracted from test_run_with_upload_json to be shared with
several test cases.
(_test_run_with_json_output.mock_upload_json): Ditto.
(test_run_with_json_output): Refactored to use _test_run_with_json_output.
(test_run_generates_results_page): Added. Tests the new feature.
(test_run_with_json_source): Refactored to use _test_run_with_json_output.
(test_run_with_multiple_repositories): Ditto.
(test_run_with_upload_json): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123985 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago.: Add ENABLE_CSS_COMPOSITING flag
commit-queue@webkit.org [Sun, 29 Jul 2012 21:59:53 +0000 (21:59 +0000)]
.: Add ENABLE_CSS_COMPOSITING flag
https://bugs.webkit.org/show_bug.cgi?id=92553

Patch by Rik Cabanier <cabanier@adobe.com> on 2012-07-29
Reviewed by Dirk Schulze.

Adds compiler flag CSS_COMPOSITING to build systems to enable CSS blending and compositing. See spec https://dvcs.w3.org/hg/FXTF/rawfile/tip/compositing/index.html

* Source/cmake/WebKitFeatures.cmake:
* Source/cmakeconfig.h.cmake:
* configure.ac:

Source/JavaScriptCore: Add ENABLE_CSS_COMPOSITING flag
https://bugs.webkit.org/show_bug.cgi?id=92553

Patch by Rik Cabanier <cabanier@adobe.com> on 2012-07-29
Reviewed by Dirk Schulze.

Adds compiler flag CSS_COMPOSITING to build systems to enable CSS blending and compositing. See spec https://dvcs.w3.org/hg/FXTF/rawfile/tip/compositing/index.html

* Configurations/FeatureDefines.xcconfig:

Source/WebCore: ENABLE_CSS_COMPOSITING flag
https://bugs.webkit.org/show_bug.cgi?id=92553

Patch by Rik Cabanier <cabanier@adobe.com> on 2012-07-29
Reviewed by Dirk Schulze.

Adds compiler flag CSS_COMPOSITING to build systems to enable CSS blending and compositing. See spec https://dvcs.w3.org/hg/FXTF/rawfile/tip/compositing/index.html

* GNUmakefile.am:
* Configurations/FeatureDefines.xcconfig:

Source/WebKit/chromium: Add ENABLE_CSS_COMPOSITING flag
https://bugs.webkit.org/show_bug.cgi?id=92553

Patch by Rik Cabanier <cabanier@adobe.com> on 2012-07-29
Reviewed by Dirk Schulze.

Adds compiler flag CSS_COMPOSITING to build systems to enable CSS blending and compositing. See spec https://dvcs.w3.org/hg/FXTF/rawfile/tip/compositing/index.html

* features.gypi:

Source/WebKit/mac: Add ENABLE_CSS_COMPOSITING flag
https://bugs.webkit.org/show_bug.cgi?id=92553

Patch by Rik Cabanier <cabanier@adobe.com> on 2012-07-29
Reviewed by Dirk Schulze.

Adds compiler flag CSS_COMPOSITING to build systems to enable CSS blending and compositing. See spec https://dvcs.w3.org/hg/FXTF/rawfile/tip/compositing/index.html

* Configurations/FeatureDefines.xcconfig:

Tools: Add ENABLE_CSS_COMPOSITING flag
https://bugs.webkit.org/show_bug.cgi?id=92553

Patch by Rik Cabanier <cabanier@adobe.com> on 2012-07-29
Reviewed by Dirk Schulze.

Adds compiler flag CSS_COMPOSITING to build systems to enable CSS blending and compositing. See spec https://dvcs.w3.org/hg/FXTF/rawfile/tip/compositing/index.html

* Scripts/webkitperl/FeatureList.pm:

WebKitLibraries: Add ENABLE_CSS_COMPOSITING flag
https://bugs.webkit.org/show_bug.cgi?id=92553

Patch by Rik Cabanier <cabanier@adobe.com> on 2012-07-29
Reviewed by Dirk Schulze.

Adds compiler flag CSS_COMPOSITING to build systems to enable CSS blending and compositing. See spec https://dvcs.w3.org/hg/FXTF/rawfile/tip/compositing/index.html

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123984 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CMake] Use WEBKIT_CREATE_FORWARDING_HEADERS to generate WK2 forwarding headers
paroga@webkit.org [Sun, 29 Jul 2012 21:46:26 +0000 (21:46 +0000)]
[CMake] Use WEBKIT_CREATE_FORWARDING_HEADERS to generate WK2 forwarding headers
https://bugs.webkit.org/show_bug.cgi?id=83579

Reviewed by Daniel Bates.

Using a list of directories in combination with the WEBKIT_CREATE_FORWARDING_HEADERS
macro allows us to share the common generation of forwarding headers across ports.

* CMakeLists.txt:
* PlatformEfl.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123983 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WIN] Add missing export macro to friend decleration.
paroga@webkit.org [Sun, 29 Jul 2012 21:02:59 +0000 (21:02 +0000)]
[WIN] Add missing export macro to friend decleration.

Since the __declspec is part of the function signature
on windows, we need it at all friend declerations too.

* wtf/MediaTime.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123982 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CMake] Add missing MediaTime files after r123878.
paroga@webkit.org [Sun, 29 Jul 2012 21:02:18 +0000 (21:02 +0000)]
[CMake] Add missing MediaTime files after r123878.

* wtf/CMakeLists.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123981 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIn flipped blocks, a point on the top edge of a box is considered outside the box...
mitz@apple.com [Sun, 29 Jul 2012 17:27:35 +0000 (17:27 +0000)]
In flipped blocks, a point on the top edge of a box is considered outside the box (and vice versa)
https://bugs.webkit.org/show_bug.cgi?id=92593

Reviewed by Simon Fraser.

Source/WebCore:

With respect to hit testing, boxes should always behave as half-open intervals which include
the physical top and left edges and not the bottom and right edges.
RenderBlock::positionForPoint was not adhering to this, since it was comparing flipped
coordinates.

Tests: fast/writing-mode/flipped-blocks-hit-test-box-edges.html
       fast/writing-mode/flipped-blocks-hit-test-line-edges.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::positionForPointWithInlineChildren): When blocks are flipped, changed
strict inequalities of y coordinates into non-strict ones and non-strict inequalities into
strict ones.
(WebCore::RenderBlock::positionForPoint): Ditto, except for the test for being under the top
of the last candidate box, which was made non-strict in the unflipped case and remained
strict.

LayoutTests:

* fast/writing-mode/flipped-blocks-hit-test-box-edges-expected.txt: Added.
* fast/writing-mode/flipped-blocks-hit-test-box-edges.html: Added.
* fast/writing-mode/flipped-blocks-hit-test-line-edges-expected.txt: Added.
* fast/writing-mode/flipped-blocks-hit-test-line-edges.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123980 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoReviewed by Martin Robinson.
kov@webkit.org [Sun, 29 Jul 2012 14:49:07 +0000 (14:49 +0000)]
Reviewed by Martin Robinson.

Make the jhbuild wrapper scripts become the jhbuild process
instead of starting it as a child. This should make it possible to
interrupt execution of jhbuild cleanly.

* gtk/run-with-jhbuild:
* jhbuild/jhbuild-wrapper:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123979 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoExtend `application/x-webkit-test-netscape` plugins to better support multiple frames.
commit-queue@webkit.org [Sun, 29 Jul 2012 09:49:14 +0000 (09:49 +0000)]
Extend `application/x-webkit-test-netscape` plugins to better support multiple frames.
https://bugs.webkit.org/show_bug.cgi?id=92478

Patch by Mike West <mkwst@chromium.org> on 2012-07-29
Reviewed by Adam Barth.

Tools:

DumpRenderTree currently understands
`<object src="data:application/x-webkit-test-netscape,alertwhenloaded">`.
It would be useful if it understood `logifloaded`, which could give
more context regarding test expectations, and can be written in such
as way as to support plugins loaded into multiple iframes in a single
test.

* DumpRenderTree/TestNetscapePlugIn/main.cpp:
(NPP_New):
    Adding support for `logifloaded`.

LayoutTests:

* http/tests/plugins/resources/mock-plugin-logger.js:
    Exciting JS "framework" for use when logging via `logifloaded`.
* http/tests/security/contentSecurityPolicy/object-src-none-allowed.html:
* http/tests/security/contentSecurityPolicy/object-src-none-allowed-expected.html:
* http/tests/security/contentSecurityPolicy/object-src-none-blocked.html:
* http/tests/security/contentSecurityPolicy/object-src-none-blocked-expected.html:
    Adjusting these tests to use the newly created `logifloaded` plugin mock.
* http/tests/security/contentSecurityPolicy/resources/echo-object-data.pl:
    Loading `mock-plugin-logger.js`, and adding a `log` attribute for `logifloaded`.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123978 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSource/WebCore: RenderBlock::offsetForContents() is wrong in flipped blocks writing...
mitz@apple.com [Sun, 29 Jul 2012 04:59:30 +0000 (04:59 +0000)]
Source/WebCore: RenderBlock::offsetForContents() is wrong in flipped blocks writing modes
https://bugs.webkit.org/show_bug.cgi?id=92588

Reviewed by Sam Weinig.

Tests: fast/multicol/hit-test-block-axis-flipped.html
       fast/writing-mode/flipped-blocks-hit-test-overflow-scroll.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::offsetForContents): Flip the point, which puts it back in unflipped
coordinates, before adding the scroll offset and adjusting for columns, then flip it back.
(WebCore::RenderBlock::adjustPointToColumnContents): Changed the inline-axis cases to
work with an unflipped point.

LayoutTests: Hit test results are wrong in flipped blocks writing modes with block column axis
https://bugs.webkit.org/show_bug.cgi?id=92588

Reviewed by Sam Weinig.

* fast/multicol/hit-test-block-axis-flipped-expected.txt: Added.
* fast/multicol/hit-test-block-axis-flipped.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123977 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoStop copying unnecessary files in WebKit2 build
weinig@apple.com [Sun, 29 Jul 2012 03:38:11 +0000 (03:38 +0000)]
Stop copying unnecessary files in WebKit2 build
https://bugs.webkit.org/show_bug.cgi?id=92587

Reviewed by Dan Bernstein.

* WebKit2.xcodeproj/project.pbxproj:
We were accidentally copying ColorSpaceData.h and EventDispatcher.messages.in into the
build products. Stop this.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123976 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] New baselines and test expectations update.
pkasting@chromium.org [Sun, 29 Jul 2012 01:35:02 +0000 (01:35 +0000)]
[Chromium] New baselines and test expectations update.
https://bugs.webkit.org/show_bug.cgi?id=72039
https://bugs.webkit.org/show_bug.cgi?id=92569
https://bugs.webkit.org/show_bug.cgi?id=92583

Unreviewed, test expectations update.

* platform/chromium-linux-x86/compositing/geometry/bounds-ignores-hidden-composited-descendant-expected.txt: Added.
* platform/chromium-linux-x86/compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.png: Added.
* platform/chromium-linux-x86/compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.txt: Added.
* platform/chromium-linux/compositing/geometry/bounds-ignores-hidden-composited-descendant-expected.txt: Added.
* platform/chromium-linux/compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.png: Added.
* platform/chromium-linux/compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.txt: Added.
* platform/chromium-mac-snowleopard/compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.png: Added.
* platform/chromium-mac/compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.png: Added.
* platform/chromium-win-xp/compositing/geometry/bounds-ignores-hidden-composited-descendant-expected.txt: Added.
* platform/chromium-win-xp/compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.png: Added.
* platform/chromium-win-xp/compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.txt: Added.
* platform/chromium-win/compositing/geometry/bounds-ignores-hidden-composited-descendant-expected.txt: Added.
* platform/chromium-win/compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.png: Added.
* platform/chromium-win/compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.txt: Added.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123975 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WK2][EFL] Add ewk_view_scale_{get|set} to EwkView.
ryuan.choi@samsung.com [Sat, 28 Jul 2012 23:28:17 +0000 (23:28 +0000)]
[WK2][EFL] Add ewk_view_scale_{get|set} to EwkView.
https://bugs.webkit.org/show_bug.cgi?id=92446

Reviewed by Kenneth Rohde Christiansen.

Add methods to Ewk_View to get or set the scale factor of page.

* UIProcess/API/efl/ewk_view.cpp:
(ewk_view_scale_set):
(ewk_view_scale_get):
* UIProcess/API/efl/ewk_view.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123974 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIn flipped lines writing modes, hit testing at the beginning of a column may return...
mitz@apple.com [Sat, 28 Jul 2012 21:40:48 +0000 (21:40 +0000)]
In flipped lines writing modes, hit testing at the beginning of a column may return a result from the previous column
https://bugs.webkit.org/show_bug.cgi?id=92566

Reviewed by Simon Fraser.

Source/WebCore:

Enhanced the fix for <http://webkit.org/b/92524> to work with flipped lines.

Extended fast/multicol/hit-test-end-of-column-with-line-height.html.

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::positionForPointWithInlineChildren): For flipped lines, check if the
hit line is the last one before a page break, and in that case, check if the hit point was
after the break.

LayoutTests:

* fast/multicol/hit-test-end-of-column-with-line-height-expected.txt: Updated.
* fast/multicol/hit-test-end-of-column-with-line-height.html: Extended with tests for
horizontal-bt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123973 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSize changes on a layer with negative z-index children don't repaint correctly
simon.fraser@apple.com [Sat, 28 Jul 2012 21:32:02 +0000 (21:32 +0000)]
Size changes on a layer with negative z-index children don't repaint correctly
https://bugs.webkit.org/show_bug.cgi?id=92583

Reviewed by Dean Jackson.

Source/WebCore:

Negative z-index descendants cause a composited layer to have a separate
layer for foreground content. If something caused that foreground layer
to change size (like toggling visibility on a descendant), we failed to
repaint the foreground layer, resulting in stretched content.

Test: compositing/geometry/bounds-ignores-hidden-dynamic-negzindex.html

* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateGraphicsLayerGeometry): Repaint the foreground
layer if its size changes, like we do for the main layer.

LayoutTests:

Test that causes a composited layer with a foreground layer (by virtue of having
negative z-index children) to resize.

Add missing pixel result for bounds-ignores-hidden-dynamic.html too.

* compositing/geometry/bounds-ignores-hidden-dynamic-expected.png: Added.
* compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.png: Added.
* compositing/geometry/bounds-ignores-hidden-dynamic-negzindex-expected.txt: Added.
* compositing/geometry/bounds-ignores-hidden-dynamic-negzindex.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123972 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIgnore visibility:hidden elements when computing compositing layer bounds
simon.fraser@apple.com [Sat, 28 Jul 2012 19:44:23 +0000 (19:44 +0000)]
Ignore visibility:hidden elements when computing compositing layer bounds
https://bugs.webkit.org/show_bug.cgi?id=92569

Reviewed by Dan Bernstein.

Source/WebCore:

When computing the bounds of compositing layers, we would take into account
layers with no visible content (visibility:hidden, and no non-hidden descendants).
This caused some huge layers in mobile Google maps.

Tests: compositing/geometry/bounds-ignores-hidden-composited-descendant.html
       compositing/geometry/bounds-ignores-hidden-dynamic.html
       compositing/geometry/bounds-ignores-hidden.html

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::calculateLayerBounds): If the ExcludeHiddenDescendants flag is set, return
the empty rect for layers with no visible content or descendants, as long as the layer is not
the root of the subtree whose bounds are being computed.
calculateLayerBounds() currently (I think incorrectly) passes the default flags to the recursive calls.
It should probably just pass 'flags', but to make this change safe, OR in the ExcludeHiddenDescendants
flag.
* rendering/RenderLayer.h: Add an ExcludeHiddenDescendants flag to the CalculateLayerBoundsFlags.
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::calculateCompositedBounds): Pass the ExcludeHiddenDescendants
flag.

LayoutTests:

Tests related to visibility:hidden's affect on composited layer bounds.

* compositing/geometry/bounds-ignores-hidden-composited-expected.txt: Added.
* compositing/geometry/bounds-ignores-hidden-composited-descendant.html: Added.
Test with visibility:visible descendants of a visibility:hidden element in a
compositing layer.

* compositing/geometry/bounds-ignores-hidden-dynamic-expected.txt: Added.
* compositing/geometry/bounds-ignores-hidden-dynamic.html: Added.
Test that changes visibility on descendants of a compositing layer, to test dynamic changes.

* compositing/geometry/bounds-ignores-hidden-expected.txt: Added.
* compositing/geometry/bounds-ignores-hidden.html: Added.
Test with a compositing layer having visibility:hidden descendants, which dumps
layer tree information.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123971 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Widen expectation for crashing test.
pkasting@chromium.org [Sat, 28 Jul 2012 19:35:27 +0000 (19:35 +0000)]
[Chromium] Widen expectation for crashing test.
http://crbug.com/139493

Unreviewed, test expectations update.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123970 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAdd an overlay play button to media controls on android
commit-queue@webkit.org [Sat, 28 Jul 2012 19:23:50 +0000 (19:23 +0000)]
Add an overlay play button to media controls on android
https://bugs.webkit.org/show_bug.cgi?id=92132

Patch by Min Qin <qinmin@chromium.org> on 2012-07-28
Reviewed by Adam Barth.

This change adds an overlay play button to the media controls for android.
No tests for now as this setting is not being used by other port.
Will add a test and adjust all the test expectations when we upstream the layout tests for android.

* WebCore.gypi:
* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
* css/CSSValueKeywords.in:
* css/mediaControls.css:
(audio::-webkit-media-controls-overlay-play-button, video::-webkit-media-controls-overlay-play-button):
* css/mediaControlsChromiumAndroid.css:
(audio::-webkit-media-controls-overlay-enclosure):
(video::-webkit-media-controls-overlay-enclosure):
(audio::-webkit-media-controls-play-button, video::-webkit-media-controls-play-button):
(video::-webkit-media-controls-overlay-play-button):
* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlOverlayPlayButtonElement::MediaControlOverlayPlayButtonElement):
(WebCore):
(WebCore::MediaControlOverlayPlayButtonElement::create):
(WebCore::MediaControlOverlayPlayButtonElement::defaultEventHandler):
(WebCore::MediaControlOverlayPlayButtonElement::updateDisplayType):
(WebCore::MediaControlOverlayPlayButtonElement::shadowPseudoId):
* html/shadow/MediaControlElements.h:
(MediaControlOverlayPlayButtonElement):
(WebCore):
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlPanelEnclosureElement::MediaControlPanelEnclosureElement):
(WebCore::MediaControlPanelEnclosureElement::create):
(WebCore):
(WebCore::MediaControlPanelEnclosureElement::shadowPseudoId):
(WebCore::MediaControlRootElementChromium::create):
(WebCore::MediaControlRootElementChromium::initializeControls):
* html/shadow/MediaControlRootElementChromium.h:
(MediaControlChromiumEnclosureElement):
(WebCore):
(MediaControlPanelEnclosureElement):
(MediaControlRootElementChromium):
* html/shadow/MediaControlRootElementChromiumAndroid.cpp: Added.
(WebCore):
(WebCore::MediaControlOverlayEnclosureElement::MediaControlOverlayEnclosureElement):
(WebCore::MediaControlOverlayEnclosureElement::create):
(WebCore::MediaControlOverlayEnclosureElement::shadowPseudoId):
(WebCore::MediaControlRootElementChromiumAndroid::MediaControlRootElementChromiumAndroid):
(WebCore::MediaControls::create):
(WebCore::MediaControlRootElementChromiumAndroid::create):
(WebCore::MediaControlRootElementChromiumAndroid::setMediaController):
(WebCore::MediaControlRootElementChromiumAndroid::playbackStarted):
(WebCore::MediaControlRootElementChromiumAndroid::playbackStopped):
* html/shadow/MediaControlRootElementChromiumAndroid.h: Added.
(WebCore):
(MediaControlOverlayEnclosureElement):
(MediaControlRootElementChromiumAndroid):
* platform/ThemeTypes.h:
* rendering/RenderMediaControlsChromium.cpp:
(WebCore::paintMediaOverlayPlayButton):
(WebCore):
(WebCore::RenderMediaControlsChromium::paintMediaControlsPart):
* rendering/RenderTheme.cpp:
(WebCore::RenderTheme::paint):
* rendering/RenderTheme.h:
(WebCore::RenderTheme::paintMediaOverlayPlayButton):
* rendering/RenderThemeChromiumAndroid.cpp:
(WebCore::RenderThemeChromiumAndroid::paintMediaOverlayPlayButton):
(WebCore):
* rendering/RenderThemeChromiumAndroid.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123969 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix Chromium/Win compile.
pkasting@chromium.org [Sat, 28 Jul 2012 19:22:46 +0000 (19:22 +0000)]
Fix Chromium/Win compile.
https://bugs.webkit.org/show_bug.cgi?id=88787

Unreviewed, build fix.

Chromium/Win needs _USE_MATH_DEFINES but doesn't define PLATFORM(WIN).
The other files in WebKit that define this macro (both in
WebCore/platform/graphics/cg) do it by unconditionally #defining it to 1
above the #include of "config.h", so copy that pattern here.

* TestWebKitAPI/Tests/WTF/MediaTime.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123968 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WK2][GTK] Implement a new spell checker API for WebKit2GTK+
mario@webkit.org [Sat, 28 Jul 2012 18:45:12 +0000 (18:45 +0000)]
[WK2][GTK] Implement a new spell checker API for WebKit2GTK+
https://bugs.webkit.org/show_bug.cgi?id=90268

Reviewed by Martin Robinson.

Source/WebKit2:

Add a simple spell checking API to WK2, allowing to enable/disable
this feature and to define a list of languages associated to it.

* GNUmakefile.am: Add flags to handle the SPELLCHECK feature.
* GNUmakefile.list.am: Added new files

Added new internal class that will act both as the implementation
of TextCheckerClient and as the object WebKitWebContext will
depend on to implement the newly added API.

* UIProcess/API/gtk/WebKitTextChecker.cpp: Added.
(toTextChecker):
(continuousSpellCheckingEnabledCallback):
(setContinuousSpellCheckingEnabledCallback):
(checkSpellingOfStringCallback):
(guessesForWordCallback):
(learnWordCallback):
(ignoreWordCallback):
(WebKitTextChecker::~WebKitTextChecker):
(WebKitTextChecker::create):
(WebKitTextChecker::WebKitTextChecker):
(WebKitTextChecker::checkSpellingOfString):
(WebKitTextChecker::getGuessesForWord):
(WebKitTextChecker::learnWord):
(WebKitTextChecker::ignoreWord):
(WebKitTextChecker::setSpellCheckingEnabled):
(WebKitTextChecker::setSpellCheckingLanguages):
* UIProcess/API/gtk/WebKitTextChecker.h: Added.
(WebKitTextChecker):
(WebKitTextChecker::isSpellCheckingEnabled):
(WebKitTextChecker::getSpellCheckingLanguages):

Added WKTextChecker to WebKitPrivate.h, needed in WebKitTextChecker.

* UIProcess/API/gtk/WebKitPrivate.h:

Add new API to WebKitWebContext to allow enabling/disabling this
spell checking feature, and to set/get the list of languages used
to decide which dictionaries will be consulted.

* UIProcess/API/gtk/WebKitWebContext.cpp:
(_WebKitWebContextPrivate):
(createDefaultWebContext):
(webkit_web_context_get_spell_checking_enabled): New API.
(webkit_web_context_set_spell_checking_enabled): Ditto.
(webkit_web_context_get_spell_checking_languages): Ditto.
(webkit_web_context_set_spell_checking_languages): Ditto.
* UIProcess/API/gtk/WebKitWebContext.h:

New tests for checking the new API added to WebKitWebContext.

* UIProcess/API/gtk/tests/TestWebKitWebContext.cpp:
(testWebContextSpellChecker):
(beforeAll):

Update documentation with new sections and symbols.

* UIProcess/API/gtk/docs/webkit2gtk-docs.sgml:
* UIProcess/API/gtk/docs/webkit2gtk-sections.txt:

Tools:

Ignore WebKitTextChecker.* private files for gtk-doc.

* gtk/generate-gtkdoc:
(get_webkit2_options):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123967 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[GTK] Add a new and reusable enchant-based spellchecker in WebCore
mario@webkit.org [Sat, 28 Jul 2012 18:28:26 +0000 (18:28 +0000)]
[GTK] Add a new and reusable enchant-based spellchecker in WebCore
https://bugs.webkit.org/show_bug.cgi?id=90269

Reviewed by Martin Robinson.

Source/WebCore:

Move enchant specific code down to WebCore, into a new class
TextCheckerEnchant, that we can use from WK1 and WK2.

* GNUmakefile.am: Add flags to handle the SPELLCHECK feature.
* GNUmakefile.list.am: Added new files.
* platform/text/gtk/TextCheckerEnchant.cpp: Added.
(getAvailableDictionariesCallback):
(TextCheckerEnchant::TextCheckerEnchant):
(TextCheckerEnchant::~TextCheckerEnchant):
(TextCheckerEnchant::ignoreWord):
(TextCheckerEnchant::learnWord):
(TextCheckerEnchant::checkSpellingOfString):
(TextCheckerEnchant::getGuessesForWord):
(TextCheckerEnchant::updateSpellCheckingLanguages):
(TextCheckerEnchant::freeEnchantBrokerDictionaries):
* platform/text/gtk/TextCheckerEnchant.h: Added.
(WebCore):
(TextCheckerEnchant):
(WebCore::TextCheckerEnchant::create):

Source/WebKit/gtk:

Remove enchant specific code from WebKitSpellCheckerEnchant and
implement it relying in the new TextCheckerEnchant class in WebCore.

* webkit/webkitspellcheckerenchant.cpp:
(_WebKitSpellCheckerEnchantPrivate):
(webkit_spell_checker_enchant_finalize):
(webkit_spell_checker_enchant_class_init):
(webkit_spell_checker_enchant_init):
(checkSpellingOfString):
(getGuessesForWord):
(updateSpellCheckingLanguages):
(learnWord):
(ignoreWord):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123966 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSigh. Just disable all the AutoResize tests after the first three.
pkasting@chromium.org [Sat, 28 Jul 2012 18:05:51 +0000 (18:05 +0000)]
Sigh.  Just disable all the AutoResize tests after the first three.
https://bugs.webkit.org/show_bug.cgi?id=92318

Unreviewed, making bots happier.

* tests/WebViewTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123965 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Rebaselining and test expectations update.
pkasting@chromium.org [Sat, 28 Jul 2012 17:36:24 +0000 (17:36 +0000)]
[Chromium] Rebaselining and test expectations update.
https://bugs.webkit.org/show_bug.cgi?id=77188
https://bugs.webkit.org/show_bug.cgi?id=85856
https://bugs.webkit.org/show_bug.cgi?id=87844
https://bugs.webkit.org/show_bug.cgi?id=91756
http://crbug.com/10364
http://crbug.com/138159
http://crbug.com/139162
http://crbug.com/139493

Unreviewed, test expectations update.

* fast/overflow/unreachable-overflow-rtl-bug-expected.txt: Added.
* platform/chromium-linux-x86/platform/chromium: Added.
* platform/chromium-linux-x86/platform/chromium/virtual: Added.
* platform/chromium-linux-x86/platform/chromium/virtual/gpu: Added.
* platform/chromium-linux-x86/platform/chromium/virtual/gpu/fast: Added.
* platform/chromium-linux-x86/platform/chromium/virtual/gpu/fast/canvas: Added.
* platform/chromium-linux-x86/platform/chromium/virtual/gpu/fast/canvas/canvas-text-alignment-expected.png: Added.
* platform/chromium-linux-x86/platform/chromium/virtual/gpu/fast/canvas/gradient-add-second-start-end-stop-expected.png: Added.
* platform/chromium-linux/platform/chromium/virtual/gpu/fast/canvas/canvas-text-alignment-expected.png:
* platform/chromium-linux/platform/chromium/virtual/gpu/fast/canvas/gradient-add-second-start-end-stop-expected.png: Added.
* platform/chromium-linux/svg/W3C-SVG-1.1/filters-example-01-b-expected.png:
* platform/chromium-mac-snowleopard/fast/block/float/026-expected.png:
* platform/chromium-mac-snowleopard/fast/block/float/028-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/datalist: Added.
* platform/chromium-mac-snowleopard/fast/forms/datalist/input-appearance-range-with-datalist-expected.png: Added.
* platform/chromium-mac-snowleopard/fast/forms/datalist/input-appearance-range-with-datalist-zoomed-expected.png: Added.
* platform/chromium-mac-snowleopard/platform/chromium/virtual/gpu/fast/canvas/canvas-text-alignment-expected.png:
* platform/chromium-mac/fast/block/float/026-expected.png:
* platform/chromium-mac/fast/block/float/028-expected.png:
* platform/chromium-mac/fast/overflow/unreachable-overflow-rtl-bug-expected.png:
* platform/chromium-mac/fast/overflow/unreachable-overflow-rtl-bug-expected.txt:
* platform/chromium-mac/platform/chromium/virtual/gpu/fast/canvas/canvas-text-alignment-expected.png:
* platform/chromium-mac/platform/chromium/virtual/gpu/fast/canvas/gradient-add-second-start-end-stop-expected.png: Added.
* platform/chromium-win/fast/forms/box-shadow-override-expected.png:
* platform/chromium-win/platform/chromium/virtual/gpu/fast/canvas/canvas-text-alignment-expected.png:
* platform/chromium-win/platform/chromium/virtual/gpu/fast/canvas/gradient-add-second-start-end-stop-expected.png:
* platform/chromium/TestExpectations:
* platform/efl/fast/overflow/unreachable-overflow-rtl-bug-expected.txt: Removed.
* platform/gtk/fast/overflow/unreachable-overflow-rtl-bug-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123964 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoTry disabling different test in hopes of tracking down the real problem.
pkasting@chromium.org [Sat, 28 Jul 2012 16:19:41 +0000 (16:19 +0000)]
Try disabling different test in hopes of tracking down the real problem.
https://bugs.webkit.org/show_bug.cgi?id=92318

Unreviewed, making bots happier.

* tests/WebViewTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123963 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Clamp] support in binding generator.
commit-queue@webkit.org [Sat, 28 Jul 2012 16:11:42 +0000 (16:11 +0000)]
[Clamp] support in binding generator.
https://bugs.webkit.org/show_bug.cgi?id=77605

Patch by Vineet Chaudhary <rgf748@motorola.com> on 2012-07-28
Reviewed by Kentaro Hara.

Currently [Clamp] extended attribute in IDL files.
Spec: http://www.w3.org/TR/2012/CR-WebIDL-20120419/#Clamp

Tests: TestObj.idl

* bindings/scripts/CodeGeneratorCPP.pm:
(SkipFunction): Skips methods with [Clamp] parameters.
* bindings/scripts/CodeGeneratorGObject.pm:
(SkipFunction): Skips methods with [Clamp] parameters.
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateParametersCheck): Support [Clamp] for JSBindings.
* bindings/scripts/CodeGeneratorObjC.pm:
(SkipFunction): Skips methods with [Clamp] parameters.
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateParametersCheck): Support [Clamp] for V8Bindings.
* bindings/scripts/IDLAttributes.txt: Added [Clamp] to attributes list.
* bindings/scripts/test/JS/JSTestObj.cpp: Test.
(WebCore::jsTestObjPrototypeFunctionClassMethodWithClamp):
* bindings/scripts/test/JS/JSTestObj.h:
* bindings/scripts/test/TestObj.idl:
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::TestObjV8Internal::classMethodWithClampCallback):
(TestObjV8Internal):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123962 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Unreviewed gardening.
ossy@webkit.org [Sat, 28 Jul 2012 15:18:00 +0000 (15:18 +0000)]
[Qt] Unreviewed gardening.

* platform/qt-4.8/Skipped: Unskip now passing tests. (bug85203)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123961 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, GTK rebaseline after r123937.
philn@webkit.org [Sat, 28 Jul 2012 13:54:21 +0000 (13:54 +0000)]
Unreviewed, GTK rebaseline after r123937.

* platform/gtk/fast/dom/navigator-detached-no-crash-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123960 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[GTK] Transparent Media controls timeline and panel
philn@webkit.org [Sat, 28 Jul 2012 13:09:08 +0000 (13:09 +0000)]
[GTK] Transparent Media controls timeline and panel
https://bugs.webkit.org/show_bug.cgi?id=85279

Reviewed by Martin Robinson.

* gtk/jhbuild.modules: Provide gnome-themes-standard build
support, this is necessary so the media controls panel colors are
properly initialized from the GTK+ theme.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123959 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRemove obsolete functions from WTF::Unicode
commit-queue@webkit.org [Sat, 28 Jul 2012 12:34:18 +0000 (12:34 +0000)]
Remove obsolete functions from WTF::Unicode
https://bugs.webkit.org/show_bug.cgi?id=92571

Patch by Patrick Gansterer <paroga@webkit.org> on 2012-07-28
Reviewed by Kentaro Hara.

Remove hasLineBreakingPropertyComplexContextOrIdeographic() and digitValue(),
since they are never used and defined for a few Unicode backends only.

* wtf/unicode/glib/UnicodeGLib.h:
* wtf/unicode/icu/UnicodeIcu.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123958 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoDisable failing test.
pkasting@chromium.org [Sat, 28 Jul 2012 06:55:58 +0000 (06:55 +0000)]
Disable failing test.
https://bugs.webkit.org/show_bug.cgi?id=92318

Unreviewed, making bots happier.

* tests/WebViewTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123957 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WebGL] Align contents of drawbuffer-test.html layout test with the one in the WebGL...
commit-queue@webkit.org [Sat, 28 Jul 2012 06:55:50 +0000 (06:55 +0000)]
[WebGL] Align contents of drawbuffer-test.html layout test with the one in the WebGL conformance suite.
https://bugs.webkit.org/show_bug.cgi?id=92375

Patch by Nayan Kumar K <nayankk@motorola.com> on 2012-07-27
Reviewed by Kenneth Russell.

* fast/canvas/webgl/drawingbuffer-test-expected.txt:
* fast/canvas/webgl/drawingbuffer-test.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123956 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoChromium rebaseline after r123285.
pkasting@chromium.org [Sat, 28 Jul 2012 06:16:47 +0000 (06:16 +0000)]
Chromium rebaseline after r123285.
https://bugs.webkit.org/show_bug.cgi?id=90624

Unreviewed, rebaselining.

* platform/chromium-mac/fast/borders/border-image-rotate-transform-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123955 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agohttps://bugs.webkit.org/show_bug.cgi?id=92573
msaboff@apple.com [Sat, 28 Jul 2012 06:11:15 +0000 (06:11 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=92573
Mac Lion: http/tests/misc/submit-post-keygen.html fails on buildbots

* platform/mac-lion/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123954 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] Support for HTML media capture
commit-queue@webkit.org [Sat, 28 Jul 2012 06:00:39 +0000 (06:00 +0000)]
[EFL] Support for HTML media capture
https://bugs.webkit.org/show_bug.cgi?id=91842

Patch by Kihong Kwon <kihong.kwon@samsung.com> on 2012-07-27
Reviewed by Ryosuke Niwa.

.:

Add HTML media capture feature, and enable that on the Efl port.

* Source/cmake/OptionsEfl.cmake:
* Source/cmake/WebKitFeatures.cmake:
* Source/cmakeconfig.h.cmake:

Tools:

Add HTML media capture feature to the feature list.
In addition, enable that on the Efl port.

* Scripts/webkitperl/FeatureList.pm:

LayoutTests:

Remove file-input-capture.html from skipped list.

* platform/efl/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123953 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r123908.
commit-queue@webkit.org [Sat, 28 Jul 2012 05:48:46 +0000 (05:48 +0000)]
Unreviewed, rolling out r123908.
http://trac.webkit.org/changeset/123908
https://bugs.webkit.org/show_bug.cgi?id=92572

Broke Chromium Linux compile (Requested by pkasting on
#webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-07-27

* WebCore.gypi:
* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
* css/CSSValueKeywords.in:
* css/mediaControls.css:
* css/mediaControlsChromiumAndroid.css:
(audio::-webkit-media-controls-play-button, video::-webkit-media-controls-play-button):
* html/shadow/MediaControlElements.cpp:
(WebCore):
* html/shadow/MediaControlElements.h:
(WebCore):
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlChromiumEnclosureElement::create):
(WebCore):
(WebCore::MediaControlChromiumEnclosureElement::shadowPseudoId):
(WebCore::MediaControls::create):
(WebCore::MediaControlRootElementChromium::create):
* html/shadow/MediaControlRootElementChromium.h:
(MediaControlChromiumEnclosureElement):
(MediaControlRootElementChromium):
* html/shadow/MediaControlRootElementChromiumAndroid.cpp: Removed.
* html/shadow/MediaControlRootElementChromiumAndroid.h: Removed.
* platform/ThemeTypes.h:
* rendering/RenderMediaControlsChromium.cpp:
(WebCore::RenderMediaControlsChromium::paintMediaControlsPart):
* rendering/RenderTheme.cpp:
(WebCore::RenderTheme::paint):
* rendering/RenderTheme.h:
* rendering/RenderThemeChromiumAndroid.cpp:
* rendering/RenderThemeChromiumAndroid.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123952 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoMark three tests as flakily timing out.
pkasting@chromium.org [Sat, 28 Jul 2012 05:12:56 +0000 (05:12 +0000)]
Mark three tests as flakily timing out.
https://bugs.webkit.org/show_bug.cgi?id=92570

Unreviewed, Chromium test expectations update.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123951 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Use new LocaleHandler.
commit-queue@webkit.org [Sat, 28 Jul 2012 04:30:06 +0000 (04:30 +0000)]
[BlackBerry] Use new LocaleHandler.
https://bugs.webkit.org/show_bug.cgi?id=92548

Patch by Eli Fidler <efidler@rim.com> on 2012-07-27
Reviewed by Yong Li.

RIM PR 128381

* platform/blackberry/LocalizedStringsBlackBerry.cpp:
(WebCore::platformLanguage):
* platform/network/blackberry/ResourceRequestBlackBerry.cpp:
(WebCore::ResourceRequest::initializePlatformRequest):
* platform/text/blackberry/TextBreakIteratorInternalICUBlackBerry.cpp:
(WebCore::currentSearchLocaleID):
(WebCore::currentTextBreakLocaleID):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123950 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoFix broken script reference after test move.
pkasting@chromium.org [Sat, 28 Jul 2012 04:29:53 +0000 (04:29 +0000)]
Fix broken script reference after test move.
https://bugs.webkit.org/show_bug.cgi?id=92111

Unreviewed, fixing bad test move.

* platform/win/fast/events/panScroll-click-hyperlink.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123949 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoChromium rebaselines after r123912.
pkasting@chromium.org [Sat, 28 Jul 2012 04:20:15 +0000 (04:20 +0000)]
Chromium rebaselines after r123912.
https://bugs.webkit.org/show_bug.cgi?id=92327

Unreviewed, rebaselining.

* platform/chromium-linux-x86/fast/hidpi: Added.
* platform/chromium-linux-x86/fast/hidpi/gradient-with-scaled-ancestor-expected.png: Added.
* platform/chromium-linux/fast/hidpi/gradient-with-scaled-ancestor-expected.png: Added.
* platform/chromium-mac-snowleopard/fast/hidpi/gradient-with-scaled-ancestor-expected.png: Added.
* platform/chromium-mac/fast/hidpi/gradient-with-scaled-ancestor-expected.png: Added.
* platform/chromium-win-xp/fast/hidpi: Added.
* platform/chromium-win-xp/fast/hidpi/gradient-with-scaled-ancestor-expected.png: Added.
* platform/chromium-win/fast/hidpi/gradient-with-scaled-ancestor-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123948 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoChromium rebaseline after r123706.
pkasting@chromium.org [Sat, 28 Jul 2012 04:12:23 +0000 (04:12 +0000)]
Chromium rebaseline after r123706.
https://bugs.webkit.org/show_bug.cgi?id=92236

Unreviewed, rebaselining.

* platform/chromium-win/fast/dom/HTMLProgressElement/progress-element-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123947 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoChromium rebaselines after r123790.
pkasting@chromium.org [Sat, 28 Jul 2012 03:19:15 +0000 (03:19 +0000)]
Chromium rebaselines after r123790.
​https://bugs.webkit.org/show_bug.cgi?id=92059

Unreviewed, rebaselining.

* platform/chromium-linux-x86/css3/filters/effect-brightness-clamping-expected.txt: Added.
* platform/chromium-linux-x86/css3/filters/effect-brightness-clamping-hw-expected.txt: Added.
* platform/chromium-linux/css3/filters/effect-brightness-clamping-expected.txt: Added.
* platform/chromium-linux/css3/filters/effect-brightness-clamping-hw-expected.txt: Added.
* platform/chromium-win-xp/css3: Added.
* platform/chromium-win-xp/css3/filters: Added.
* platform/chromium-win-xp/css3/filters/effect-brightness-clamping-expected.txt: Added.
* platform/chromium-win-xp/css3/filters/effect-brightness-clamping-hw-expected.txt: Added.
* platform/chromium-win/css3/filters/effect-brightness-clamping-expected.txt: Added.
* platform/chromium-win/css3/filters/effect-brightness-clamping-hw-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123946 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoChange constructor with explicit constructor in WebCore/Modules
commit-queue@webkit.org [Sat, 28 Jul 2012 01:23:58 +0000 (01:23 +0000)]
Change constructor with explicit constructor in WebCore/Modules
https://bugs.webkit.org/show_bug.cgi?id=92529

Patch by Gyuyoung Kim <gyuyoung.kim@samsung.com> on 2012-07-27
Reviewed by Kentaro Hara.

As a step to let *explicit* keyword add to constructors which have a parameter,
*explicit* keyword is added to Source/WebCore/Modules in order to avoid implicit
type conversion.

No new tests. Covered by existing tests.

* Modules/filesystem/LocalFileSystem.h:
(WebCore::LocalFileSystem::LocalFileSystem):
* Modules/filesystem/Metadata.h:
(WebCore::Metadata::Metadata):
* Modules/filesystem/SyncCallbackHelper.h:
(WebCore::SyncCallbackHelper::SuccessCallbackImpl::SuccessCallbackImpl):
(WebCore::SyncCallbackHelper::ErrorCallbackImpl::ErrorCallbackImpl):
* Modules/mediastream/IceCandidate.h:
(IceCandidate):
* Modules/mediastream/NavigatorUserMediaError.h:
(WebCore::NavigatorUserMediaError::NavigatorUserMediaError):
* Modules/mediastream/SessionDescription.h:
(SessionDescription):
* Modules/mediastream/UserMediaController.h:
(UserMediaController):
* Modules/notifications/NotificationController.h:
(NotificationController):
* Modules/speech/SpeechRecognition.h:
(SpeechRecognition):
* Modules/speech/SpeechRecognitionController.h:
(SpeechRecognitionController):
* Modules/speech/SpeechRecognitionEvent.h:
(SpeechRecognitionEvent):
* Modules/speech/SpeechRecognitionResultList.h:
(SpeechRecognitionResultList):
* Modules/vibration/Vibration.h:
(Vibration):
* Modules/webaudio/AudioBuffer.h:
(AudioBuffer):
* Modules/webaudio/AudioContext.h:
(AudioContext):
* Modules/webaudio/AudioNodeInput.h:
(AudioNodeInput):
* Modules/webaudio/AudioSummingJunction.h:
(AudioSummingJunction):
* Modules/webaudio/BiquadDSPKernel.h:
(WebCore::BiquadDSPKernel::BiquadDSPKernel):
* Modules/webaudio/DefaultAudioDestinationNode.h:
(DefaultAudioDestinationNode):
* Modules/webaudio/DelayDSPKernel.h:
(DelayDSPKernel):
* Modules/webaudio/OfflineAudioCompletionEvent.h:
(OfflineAudioCompletionEvent):
* Modules/webaudio/WaveShaperDSPKernel.h:
(WebCore::WaveShaperDSPKernel::WaveShaperDSPKernel):
* Modules/webaudio/WaveShaperNode.h:
(WaveShaperNode):
* Modules/webaudio/WaveTable.h:
(WaveTable):
* Modules/webdatabase/DatabaseAuthorizer.h:
(DatabaseAuthorizer):
* Modules/webdatabase/DatabaseTask.h:
(Database::DatabaseTransactionTask):
* Modules/webdatabase/DatabaseTracker.h:
(DatabaseTracker):
* Modules/webdatabase/SQLCallbackWrapper.h:
(WebCore::SQLCallbackWrapper::SafeReleaseTask::SafeReleaseTask):
* Modules/webdatabase/SQLException.h:
(WebCore::SQLException::SQLException):
* Modules/websockets/WebSocket.h:
* Modules/websockets/WebSocketExtensionProcessor.h:
(WebCore::WebSocketExtensionProcessor::WebSocketExtensionProcessor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123945 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[V8] Optimize Element::getAttributeNS() by replacing String with AtomicString
commit-queue@webkit.org [Sat, 28 Jul 2012 01:21:15 +0000 (01:21 +0000)]
[V8] Optimize Element::getAttributeNS() by replacing String with AtomicString
https://bugs.webkit.org/show_bug.cgi?id=90276

Patch by Kentaro Hara <haraken@chromium.org> on 2012-07-27
Reviewed by Adam Barth.

In r121439, we changed the argument type of Element::getAttribute()
from String to AtomicString for optimization. This patch changes the
argument type of Element::getAttributeNS() from String to AtomicString.

Actually this change affects V8 performance only.

[JSC without the patch]
The binding code converts a given JavaScript string to a String
-> the String is passed to getAttributeNS()
-> the String is converted to an AtomicString when it is passed to
QualifiedName() constructor.

[JSC with the patch]
The binding code converts a given JavaScript string to a String
-> the String is converted to an AtomicString when it is passed to
getAttributeNS()
-> the AtomicString is passed to QualifiedName() constructor.

[V8 without the patch]
The binding code retrieves a cached external String for a given
JavaScript string
-> the String is passed to getAttributeNS()
-> the String is converted to AtomicString when it is passed to
QualifiedName() constructor.

[V8 with the patch]
The binding code retrieves a cached external AtomicString for a given
JavaScript string
-> the AtomicString is passed to getAttributeNS()
-> the AtomicString is passed to QualifiedName() constructor.

In this way, in JSC, the change affects where a String is converted to
an AtomicString. On the other hand, in V8, (since the AtomicString is already
cached,) we can skip the conversion step.

Performance test: https://bugs.webkit.org/attachment.cgi?id=150158

getAttributeNS (Chromium/Linux):
577.80ms => 454.60ms

* dom/Element.cpp:
(WebCore::Element::getAttributeNS):
* dom/Element.h:
(Element):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123944 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, rolling out r123679.
commit-queue@webkit.org [Sat, 28 Jul 2012 01:02:33 +0000 (01:02 +0000)]
Unreviewed, rolling out r123679.
http://trac.webkit.org/changeset/123679
https://bugs.webkit.org/show_bug.cgi?id=92565

Slowed down HTML parsing by 3.6% (Requested by abarth on
#webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-07-27

Source/WebCore:

* html/FTPDirectoryDocument.cpp:
(WebCore::FTPDirectoryDocumentParser::append):
* html/parser/HTMLEntityParser.cpp:
(WebCore::HTMLEntityParser::consumeNamedEntity):
* html/parser/HTMLParserIdioms.cpp:
(WebCore::stripLeadingAndTrailingHTMLSpaces):
* html/parser/HTMLSourceTracker.cpp:
(WebCore::HTMLSourceTracker::sourceForToken):
* loader/cache/CachedScript.cpp:
(WebCore::CachedScript::script):
* platform/text/SegmentedString.cpp:
(WebCore::SegmentedString::SegmentedString):
(WebCore::SegmentedString::operator=):
(WebCore::SegmentedString::clear):
(WebCore::SegmentedString::append):
(WebCore::SegmentedString::prepend):
(WebCore::SegmentedString::advanceSubstring):
(WebCore::SegmentedString::advance):
(WebCore::SegmentedString::advanceSlowCase):
(WebCore::SegmentedString::advanceAndUpdateLineNumberSlowCase):
* platform/text/SegmentedString.h:
(WebCore::SegmentedSubstring::SegmentedSubstring):
(WebCore::SegmentedSubstring::clear):
(WebCore::SegmentedSubstring::appendTo):
(SegmentedSubstring):
(WebCore::SegmentedString::SegmentedString):
(WebCore::SegmentedString::push):
(WebCore::SegmentedString::isEmpty):
(WebCore::SegmentedString::lookAhead):
(WebCore::SegmentedString::lookAheadIgnoringCase):
(WebCore::SegmentedString::advance):
(WebCore::SegmentedString::advanceAndASSERT):
(WebCore::SegmentedString::advanceAndASSERTIgnoringCase):
(WebCore::SegmentedString::advancePastNewlineAndUpdateLineNumber):
(SegmentedString):
(WebCore::SegmentedString::advancePastNonNewline):
(WebCore::SegmentedString::advanceAndUpdateLineNumber):
(WebCore::SegmentedString::operator*):
(WebCore::SegmentedString::operator->):
(WebCore::SegmentedString::current):
(WebCore::SegmentedString::equalsLiterally):
(WebCore::SegmentedString::equalsIgnoringCase):
(WebCore::SegmentedString::lookAheadInline):
(WebCore::SegmentedString::lookAheadSlowCase):
* xml/parser/CharacterReferenceParserInlineMethods.h:
(WebCore::consumeCharacterReference):
* xml/parser/MarkupTokenizerBase.h:
(WebCore::MarkupTokenizerBase::InputStreamPreprocessor::nextInputCharacter):
(WebCore::MarkupTokenizerBase::InputStreamPreprocessor::peek):
(WebCore::MarkupTokenizerBase::InputStreamPreprocessor::advance):

Source/WTF:

* wtf/text/WTFString.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123943 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoReset the set of "seen" plugins when the main frame load is committed.
jer.noble@apple.com [Sat, 28 Jul 2012 00:41:18 +0000 (00:41 +0000)]
Reset the set of "seen" plugins when the main frame load is committed.
https://bugs.webkit.org/show_bug.cgi?id=92564

Reviewed by Anders Carlsson.

Because the Page object is re-used across navigation and reload, reset the list
of seen plugins when the main frame load commits. This gives a good baseline to
compare against the number of pages loaded.

No new tests; the "seen" plugin list is for diagnostic purposes only.

* loader/FrameLoader.cpp:
(WebCore::FrameLoader::dispatchDidCommitLoad):
* page/Page.cpp:
(WebCore::Page::resetSeenPlugins):
* page/Page.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123942 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago"webkit-patch rebaseline <test>" fails with exit_code 1
dpranke@chromium.org [Sat, 28 Jul 2012 00:34:51 +0000 (00:34 +0000)]
"webkit-patch rebaseline <test>" fails with exit_code 1
https://bugs.webkit.org/show_bug.cgi?id=91560

Reviewed by Adam Barth.

webkit-patch rebaseline-json crashes in an svn checkout if it
tries to add files and there are no files to add; arguably
svn.add() should be robust against this, but rebaseline-json
shouldn't be calling it, either. This patch fixes the latter and
adds better debugging info to garden-o-matic so we can see what
caused the crash.

* Scripts/webkitpy/tool/commands/rebaseline.py:
(AbstractParallelRebaselineCommand._rebaseline):
* Scripts/webkitpy/tool/servers/gardeningserver.py:
(GardeningHTTPRequestHandler.rebaselineall):
(GardeningHTTPRequestHandler.rebaselineall.error_handler):
(GardeningHTTPRequestHandler):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123941 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoexecCommand copies the backgroung-color of the enclosing element to the element being...
commit-queue@webkit.org [Sat, 28 Jul 2012 00:32:15 +0000 (00:32 +0000)]
execCommand copies the backgroung-color of the enclosing element to the element being edited.
https://bugs.webkit.org/show_bug.cgi?id=91783

Patch by Sukolsak Sakshuwong <sukolsak@google.com> on 2012-07-27
Reviewed by Ryosuke Niwa.

Source/WebCore:

InsertListCommand and some other edit commands remove redundant
style properties after their DOM modification by using
StylePropertySet::removeEquivalentProperties.  However, this method
does not remove redundant background-color properties such as
"background-color: white" and "background-color: #fff" because
it compares them using cssText and thus treats them as different.
Add code to convert background colors to RGBA and then compare them.

Test: editing/style/redundant-background-color.html

* editing/EditingStyle.cpp:
(WebCore::getRGBABackgroundColor):
(WebCore):
(WebCore::getPropertiesNotIn):

LayoutTests:

This tests to make sure that WebKit removes redundant "background-color"
property after execCommand.

* editing/style/redundant-background-color-expected.txt: Added.
* editing/style/redundant-background-color.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123940 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] Return 0 when network is offline
commit-queue@webkit.org [Sat, 28 Jul 2012 00:30:31 +0000 (00:30 +0000)]
[EFL] Return 0 when network is offline
https://bugs.webkit.org/show_bug.cgi?id=92454

Patch by Gyuyoung Kim <gyuyoung.kim@samsung.com> on 2012-07-27
Reviewed by Kenneth Rohde Christiansen.

Network Info latest spec says 0 is returned when network is offline.
r123769 modified to return infinity value when network is offiline.

* platform/efl/NetworkInfoProviderEfl.cpp:
(WebCore::NetworkInfoProviderEfl::bandwidth):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123939 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed perf-o-matic build fix for dashboard image generation.
rniwa@webkit.org [Sat, 28 Jul 2012 00:08:44 +0000 (00:08 +0000)]
Unreviewed perf-o-matic build fix for dashboard image generation.
Also, only show Parser/html5-full-render on the dashboard by default
so that it won't take forever to load it.

* Websites/webkit-perf.appspot.com/controller.py:
(schedule_runs_update):
(RunsChartHandler.post):
* Websites/webkit-perf.appspot.com/models.py:
(Runs.chart_params):
(DashboardImage.get_image):
* Websites/webkit-perf.appspot.com/models_unittest.py:
(RunsTest.test_chart_params_with_value):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123938 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoChange navigator.webkitGamepads[] to navigator.webkitGetGamepads()
commit-queue@webkit.org [Fri, 27 Jul 2012 23:51:48 +0000 (23:51 +0000)]
Change navigator.webkitGamepads[] to navigator.webkitGetGamepads()
https://bugs.webkit.org/show_bug.cgi?id=92533

Patch by Scott Graham <scottmg@chromium.org> on 2012-07-27
Reviewed by Adam Barth.

Source/WebCore:

Per discussion on public-webapps. Pages that enumerate navigator were
causing background resources to be created to poll hardware
unintentionally. By changing to an explicit function call, this is
avoided (and the API becomes more clear too). This change updates the
implementation to match the current ED spec.

* Modules/gamepad/NavigatorGamepad.cpp:
(WebCore::NavigatorGamepad::webkitGetGamepads):
* Modules/gamepad/NavigatorGamepad.h:
(NavigatorGamepad):
* Modules/gamepad/NavigatorGamepad.idl:
* bindings/generic/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::setWebkitGetGamepadsEnabled):
(WebCore::RuntimeEnabledFeatures::webkitGetGamepadsEnabled):

Source/WebKit/chromium:

Renames for updated access point.

* src/WebRuntimeFeatures.cpp:
(WebKit::WebRuntimeFeatures::enableGamepad):
(WebKit::WebRuntimeFeatures::isGamepadEnabled):

LayoutTests:

Updates for renamed access point.

* gamepad/gamepad-api-expected.txt:
* gamepad/gamepad-api.html:
* gamepad/gamepad-polling-access-expected.txt:
* gamepad/gamepad-polling-access.html:
* platform/chromium/fast/dom/navigator-detached-no-crash-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123937 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoPlugins should not be allowed to override standard properties/attributes in non-stand...
beidson@apple.com [Fri, 27 Jul 2012 23:44:27 +0000 (23:44 +0000)]
Plugins should not be allowed to override standard properties/attributes in non-standard worlds
<rdar://problem/11975252> and https://bugs.webkit.org/show_bug.cgi?id=92519

Reviewed by Anders Carlsson.

Source/WebCore:

Change the 3 plugin-owning element's custom bindings to prefer built-in properties if they exist.
When they do they don't give the plugin a chance to override.

Test: plugins/npruntime/overrides-all-properties.html

Add plugin custom functions to prefer built-in properties over plugin scriptable object properties:
* bindings/js/JSPluginElementFunctions.h:
(WebCore::pluginElementCustomGetOwnPropertySlot):
(WebCore::pluginElementCustomGetOwnPropertyDescriptor):

Use those new custom functions for getting properties:
* bindings/js/JSHTMLAppletElementCustom.cpp:
(WebCore::JSHTMLAppletElement::getOwnPropertySlotDelegate):
(WebCore::JSHTMLAppletElement::getOwnPropertyDescriptorDelegate):
* bindings/js/JSHTMLEmbedElementCustom.cpp:
(WebCore::JSHTMLEmbedElement::getOwnPropertySlotDelegate):
(WebCore::JSHTMLEmbedElement::getOwnPropertyDescriptorDelegate):
* bindings/js/JSHTMLObjectElementCustom.cpp:
(WebCore::JSHTMLObjectElement::getOwnPropertySlotDelegate):
(WebCore::JSHTMLObjectElement::getOwnPropertyDescriptorDelegate):

Tools:

Add a test plugin which overrides all properties it is asked about:
* DumpRenderTree/TestNetscapePlugIn/Tests/PluginScriptableObjectOverridesAllProperties.cpp: Added.
(PluginScriptableObjectOverridesAllProperties):
(PluginScriptableObjectOverridesAllProperties::PluginScriptableObjectOverridesAllProperties):
(PluginObject):
(PluginScriptableObjectOverridesAllProperties::PluginObject::PluginObject):
(PluginScriptableObjectOverridesAllProperties::PluginObject::~PluginObject):
(PluginScriptableObjectOverridesAllProperties::PluginObject::hasProperty):
(PluginScriptableObjectOverridesAllProperties::PluginObject::getProperty):
(PluginScriptableObjectOverridesAllProperties::NPP_GetValue):

Expose NPN_MemAlloc to the test plugin:
* DumpRenderTree/TestNetscapePlugIn/PluginTest.cpp:
(PluginTest::NPN_MemAlloc):
* DumpRenderTree/TestNetscapePlugIn/PluginTest.h:
(PluginTest):

* DumpRenderTree/DumpRenderTree.gypi:
* DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
* DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.vcproj:
* DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:

LayoutTests:

* plugins/npruntime/overrides-all-properties-expected.txt: Added.
* plugins/npruntime/overrides-all-properties.html: Added.
* platform/chromium/TestExpectations: Test behaves different in V8, so skip on chromium

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123936 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agouse createUninitialized when creating TypedArray from another array
commit-queue@webkit.org [Fri, 27 Jul 2012 23:32:19 +0000 (23:32 +0000)]
use createUninitialized when creating TypedArray from another array
https://bugs.webkit.org/show_bug.cgi?id=92518

Patch by Arnaud Renevier <a.renevier@sisa.samsung.com> on 2012-07-27
Reviewed by Kenneth Russell.

PerformanceTests:

* Bindings/typed-array-construct-from-array.html: Added.

Source/WebCore:

When creating a Typed Array from another array, we fill it with values
from the source array. So, we can create it uninitialized. This is
make constructor about 35% faster when constructed with another typed
array.

Test: fast/canvas/webgl/array-constructor.html

* bindings/js/JSArrayBufferViewHelper.h:
(WebCore::constructArrayBufferViewWithTypedArrayArgument):
(WebCore::constructArrayBufferView):

Source/WTF:

Expose a createUninitialized static method on TypedArray classes.

* wtf/Float32Array.h:
(Float32Array):
(WTF::Float32Array::createUninitialized):
(WTF):
* wtf/Float64Array.h:
(Float64Array):
(WTF::Float64Array::createUninitialized):
(WTF):
* wtf/Int16Array.h:
(Int16Array):
(WTF::Int16Array::createUninitialized):
(WTF):
* wtf/Int32Array.h:
(Int32Array):
(WTF::Int32Array::createUninitialized):
(WTF):
* wtf/Int8Array.h:
(Int8Array):
(WTF::Int8Array::createUninitialized):
(WTF):
* wtf/Uint16Array.h:
(Uint16Array):
(WTF::Uint16Array::createUninitialized):
(WTF):
* wtf/Uint32Array.h:
(Uint32Array):
(WTF::Uint32Array::createUninitialized):
(WTF):
* wtf/Uint8Array.h:
(Uint8Array):
(WTF::Uint8Array::createUninitialized):
(WTF):
* wtf/Uint8ClampedArray.h:
(Uint8ClampedArray):

LayoutTests:

Add test to check that constructing a Typed Array with nan values does
not result in random values.

* fast/canvas/webgl/array-constructor-expected.txt: Added.
* fast/canvas/webgl/array-constructor.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123935 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Adapt to change in the FontInfo platform API.
commit-queue@webkit.org [Fri, 27 Jul 2012 23:29:20 +0000 (23:29 +0000)]
[BlackBerry] Adapt to change in the FontInfo platform API.
https://bugs.webkit.org/show_bug.cgi?id=92547

Patch by Eli Fidler <efidler@rim.com> on 2012-07-27
Reviewed by Yong Li

RIM PR 161263

Source/WebCore:

* page/blackberry/SettingsBlackBerry.cpp:
(WebCore::Settings::initializeDefaultFontFamilies):

Source/WebKit/blackberry:

* Api/WebSettings.cpp:
(BlackBerry::WebKit::WebSettings::standardSettings):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123934 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL][WK2] Assertion hit in ewk_cookie_manager.cpp
commit-queue@webkit.org [Fri, 27 Jul 2012 23:12:33 +0000 (23:12 +0000)]
[EFL][WK2] Assertion hit in ewk_cookie_manager.cpp
https://bugs.webkit.org/show_bug.cgi?id=92544

Patch by Christophe Dumez <christophe.dumez@intel.com> on 2012-07-27
Reviewed by Kenneth Rohde Christiansen.

Removed wrong assertion in ewk_cookie_manager.cpp
and replace it by an if check. The assertion was
sometimes hit by our unit tests.

* UIProcess/API/efl/ewk_cookie_manager.cpp:
(cookiesDidChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123933 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Fix color matrix filter computation so that filters that don't need clampi...
commit-queue@webkit.org [Fri, 27 Jul 2012 23:02:03 +0000 (23:02 +0000)]
[chromium] Fix color matrix filter computation so that filters that don't need clamping are correctly detected as such
https://bugs.webkit.org/show_bug.cgi?id=92452

Patch by Antoine Labour <piman@chromium.org> on 2012-07-27
Reviewed by Stephen White.

Because of precision issues, some filters were returning true to
matrixNeedsClamping when they can return false. This fixes it.

No new tests in this CL, a follow-up change refactors the code to add
several tests.

* platform/graphics/chromium/cc/CCRenderSurfaceFilters.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123932 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoSplit functionality of MarkedAllocator::m_currentBlock
mhahnenberg@apple.com [Fri, 27 Jul 2012 22:59:14 +0000 (22:59 +0000)]
Split functionality of MarkedAllocator::m_currentBlock
https://bugs.webkit.org/show_bug.cgi?id=92550

Reviewed by Filip Pizlo.

MarkedAllocator::m_currentBlock serves two purposes right now; it indicates the block that is currently
being used for allocation and the beginning of the list of blocks that need to be swept. We should split
these two functionalities into two separate fields.

* heap/MarkedAllocator.cpp:
(JSC::MarkedAllocator::tryAllocateHelper): Use m_blocksToSweep instead of m_currentBlock as the
initializer/reference of the loop. Only change m_currentBlock when we know what the result will be.
(JSC::MarkedAllocator::addBlock): When we add a new block we know that both m_blocksToSweep and
m_currentBlock are null. In order to preserve the invariant that m_currentBlock <= m_blocksToSweep,
we assign both of them to point to the new block.
(JSC::MarkedAllocator::removeBlock): We need a separate check to see if the block we're removing is
m_blocksToSweep and if so, advance it to the next block in the list.
* heap/MarkedAllocator.h:
(MarkedAllocator): Initialize m_blocksToSweep.
(JSC::MarkedAllocator::MarkedAllocator):
(JSC::MarkedAllocator::reset): We set m_blocksToSweep to be the head of our list. This function is called
at the end of a collection, so all of the blocks in our allocator need to be swept. We need to sweep a
block before we can start allocating, so m_currentBlock is set to null. We also set the freeList to
the empty FreeList to emphasize the fact that we can't start allocating until we do some sweeping.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123931 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAdd diagnostic logging for plugins-per-page.
jer.noble@apple.com [Fri, 27 Jul 2012 22:58:02 +0000 (22:58 +0000)]
Add diagnostic logging for plugins-per-page.
https://bugs.webkit.org/show_bug.cgi?id=92538

Reviewed by Anders Carlsson.

Add some diagnostic logging for whether a page has seen a plugin, and
whether a page has seen a plugin of a specific type.

Move the diagnostic logging out of the elemements themselves:
* html/HTMLEmbedElement.cpp:
(WebCore::HTMLEmbedElement::updateWidget):
* html/HTMLObjectElement.cpp:
(WebCore::HTMLObjectElement::updateWidget):

Instead, log when the plugin is requested, thereby catching plugins which are
rejected because, e.g., Java is disabled or not installed:
* loader/SubframeLoader.cpp:
(WebCore::logPluginRequest):
(WebCore::SubframeLoader::requestObject):
(WebCore::SubframeLoader::createJavaAppletWidget):

Add new diagnostic key values:
* page/DiagnosticLoggingKeys.cpp:
(WebCore::DiagnosticLoggingKeys::pageContainsPluginKey):
(WebCore::DiagnosticLoggingKeys::pageContainsAtLeastOnePluginKey):
* page/DiagnosticLoggingKeys.h:

Add a map of plugin types seen per-page for diagnostic purposes:
* page/Page.cpp:
(WebCore::Page::hasSeenAnyPlugin):
(WebCore::Page::hasSeenPlugin):
(WebCore::Page::sawPlugin):
* page/Page.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123930 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[WINCE] Use macros from ICU instead of defining the same functionality again
commit-queue@webkit.org [Fri, 27 Jul 2012 22:46:12 +0000 (22:46 +0000)]
[WINCE] Use macros from ICU instead of defining the same functionality again
https://bugs.webkit.org/show_bug.cgi?id=92530

Patch by Patrick Gansterer <paroga@webkit.org> on 2012-07-27
Reviewed by Ryosuke Niwa.

Replace isHighSurrogate() with U16_IS_LEAD(), isLowSurrogate() with
U16_IS_TRAIL() and surrogateToUcs4() with U16_GET_SUPPLEMENTARY().

Source/WebCore:

* html/FTPDirectoryDocument.cpp:
(WebCore::processFileDateString):
* loader/archive/mhtml/MHTMLArchive.cpp:
(WebCore::MHTMLArchive::generateMHTMLData):
* platform/graphics/wince/FontWinCE.cpp:
(WebCore::generateComponents):

Source/WTF:

* wtf/unicode/wince/UnicodeWinCE.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@123929 268f45cc-cd09-0410-ab3c-d52691b4dbfc