Google Git
Sign in
webkit / WebKit / 7db48aab1d0d24738cdad52014593f5c89cb2935 / . / Source / WebCore / ChangeLog
blob: 0b075217e22f175cf13999c8e8713c93da61c5aa [file] [log] [blame]
2012-06-17 Yoshifumi Inoue <yosin@chromium.org>
Unreviewed, rolling out r120565.
http://trac.webkit.org/changeset/120565
https://bugs.webkit.org/show_bug.cgi?id=89155
Bad commit message
* rendering/RenderSearchField.cpp: Removed.
* rendering/RenderSearchField.h: Removed.
2012-06-17 Yoshifumi Inoue <yosin@chromium.org>
[Forms] Copy RenderTextControlSingleLine.{cpp,h} to RenderSearchFiled.{cpp,h}
https://bugs.webkit.org/show_bug.cgi?id=89155
This patch copies RenderTextControlSingleLine.cpp and .h into RenderSearchField.cpp
and .h with just "cp" code. New files aren't compiled until bug 88980.
No new tests. This patch doesn't change behavior.
* rendering/RenderSearchField.cpp: Added.
(WebCore):
(WebCore::RenderTextControlInnerBlock::positionForPoint):
(WebCore::RenderTextControlSingleLine::RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::~RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::containerElement):
(WebCore::RenderTextControlSingleLine::innerBlockElement):
(WebCore::RenderTextControlSingleLine::innerSpinButtonElement):
(WebCore::RenderTextControlSingleLine::resultsButtonElement):
(WebCore::RenderTextControlSingleLine::cancelButtonElement):
(WebCore::RenderTextControlSingleLine::textBaseStyle):
(WebCore::RenderTextControlSingleLine::addSearchResult):
(WebCore::RenderTextControlSingleLine::showPopup):
(WebCore::RenderTextControlSingleLine::hidePopup):
(WebCore::RenderTextControlSingleLine::paint):
(WebCore::RenderTextControlSingleLine::layout):
(WebCore::RenderTextControlSingleLine::nodeAtPoint):
(WebCore::RenderTextControlSingleLine::styleDidChange):
(WebCore::RenderTextControlSingleLine::capsLockStateMayHaveChanged):
(WebCore::RenderTextControlSingleLine::hasControlClip):
(WebCore::RenderTextControlSingleLine::controlClipRect):
(WebCore::RenderTextControlSingleLine::getAvgCharWidth):
(WebCore::RenderTextControlSingleLine::preferredContentWidth):
(WebCore::RenderTextControlSingleLine::computeControlHeight):
(WebCore::RenderTextControlSingleLine::updateFromElement):
(WebCore::RenderTextControlSingleLine::createInnerTextStyle):
(WebCore::RenderTextControlSingleLine::createInnerBlockStyle):
(WebCore::RenderTextControlSingleLine::updateCancelButtonVisibility):
(WebCore::RenderTextControlSingleLine::visibilityForCancelButton):
(WebCore::RenderTextControlSingleLine::textShouldBeTruncated):
(WebCore::RenderTextControlSingleLine::autosaveName):
(WebCore::RenderTextControlSingleLine::valueChanged):
(WebCore::RenderTextControlSingleLine::itemText):
(WebCore::RenderTextControlSingleLine::itemLabel):
(WebCore::RenderTextControlSingleLine::itemIcon):
(WebCore::RenderTextControlSingleLine::itemIsEnabled):
(WebCore::RenderTextControlSingleLine::itemStyle):
(WebCore::RenderTextControlSingleLine::menuStyle):
(WebCore::RenderTextControlSingleLine::clientInsetLeft):
(WebCore::RenderTextControlSingleLine::clientInsetRight):
(WebCore::RenderTextControlSingleLine::clientPaddingLeft):
(WebCore::RenderTextControlSingleLine::clientPaddingRight):
(WebCore::RenderTextControlSingleLine::listSize):
(WebCore::RenderTextControlSingleLine::selectedIndex):
(WebCore::RenderTextControlSingleLine::popupDidHide):
(WebCore::RenderTextControlSingleLine::itemIsSeparator):
(WebCore::RenderTextControlSingleLine::itemIsLabel):
(WebCore::RenderTextControlSingleLine::itemIsSelected):
(WebCore::RenderTextControlSingleLine::setTextFromItem):
(WebCore::RenderTextControlSingleLine::fontSelector):
(WebCore::RenderTextControlSingleLine::hostWindow):
(WebCore::RenderTextControlSingleLine::autoscroll):
(WebCore::RenderTextControlSingleLine::scrollWidth):
(WebCore::RenderTextControlSingleLine::scrollHeight):
(WebCore::RenderTextControlSingleLine::scrollLeft):
(WebCore::RenderTextControlSingleLine::scrollTop):
(WebCore::RenderTextControlSingleLine::setScrollLeft):
(WebCore::RenderTextControlSingleLine::setScrollTop):
(WebCore::RenderTextControlSingleLine::scroll):
(WebCore::RenderTextControlSingleLine::logicalScroll):
(WebCore::RenderTextControlSingleLine::createScrollbar):
(WebCore::RenderTextControlSingleLine::inputElement):
* rendering/RenderSearchField.h: Added.
(WebCore):
(RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::popupIsVisible):
(WebCore::RenderTextControlSingleLine::isTextField):
(WebCore::toRenderTextControlSingleLine):
(RenderTextControlInnerBlock):
(WebCore::RenderTextControlInnerBlock::RenderTextControlInnerBlock):
(WebCore::RenderTextControlInnerBlock::hasLineIfEmpty):
2012-06-17 MORITA Hajime <morrita@google.com>
Unreviewed build fix.
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::removePassesWithCachedTextures):
2012-06-17 Philippe Normand <pnormand@igalia.com>
[GStreamer] optimize ::naturalSize()
https://bugs.webkit.org/show_bug.cgi?id=89314
Reviewed by Martin Robinson.
Keep track of the video-sink sinkpad in the MediaPlayerPrivate and
adapt the webkitGstGetPadCaps API accordingly. The benefit of this
change is to reduce the number of calls to
gst_element_get_static_pad() and the corresponding gst_pad_unref().
* platform/graphics/gstreamer/GStreamerVersioning.cpp:
(webkitGstGetPadCaps):
* platform/graphics/gstreamer/GStreamerVersioning.h:
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
(WebCore::MediaPlayerPrivateGStreamer::naturalSize):
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
(MediaPlayerPrivateGStreamer):
2012-06-17 Arvid Nilsson <anilsson@rim.com>
Don't mark main frame layer opaque if the frame view is transparent
https://bugs.webkit.org/show_bug.cgi?id=88234
Reviewed by Simon Fraser.
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::createPrimaryGraphicsLayer):
2012-06-17 Philip Rogers <pdr@google.com>
Prevent crash in SVGDocumentExtensions::removeAllElementReferencesForTarget.
https://bugs.webkit.org/show_bug.cgi?id=88144
Reviewed by Abhishek Arya.
When iterating over referencing elements to rebuild after a reference change in
SVGDocumentExtensions::removeAllElementReferencesForTarget, we can
modify the underlying toBeNotified vector, invalidating it. This change checks
that a vector element is valid before rebuilding, preventing a crash.
Some definitions from SVGDocumentExtensions that may put this patch in context:
An example of a "referenced elements" is a <path>.
An example of a "referencing element" is a <textPath href='some_path_id'>.
m_elementDependencies is a map from referenced elements (e.g., paths) to
a set of referencing elements (e.g., textPaths).
The check that the vector element is valid relies on checking if the referencing
element is in m_elementDependencies. This check is allowed because in the
destructor of SVGTextPathElement (and SVGFeImageElement),
removeAllTargetReferencesForElement() is called, removing the referencing element
from m_elementDependencies.
Simply checking if the referencing element is anywhere in m_elementDependencies
is enough to show it is valid, but that requires iterating over all referenced
elements to see if the given referencing element is present. This change
only checks if the textPath is still in the elements referencing the
path being removed, and only removes the referenced element from
m_elementDependencies after forcing the referencing elements to be rebuilt.
Test: svg/text/textpath-reference-crash.html
* svg/SVGDocumentExtensions.cpp:
(WebCore::SVGDocumentExtensions::removeAllElementReferencesForTarget):
2012-06-17 Yoshifumi Inoue <yosin@chromium.org>
Unreviewed, rolling out r120390.
http://trac.webkit.org/changeset/120390
https://bugs.webkit.org/show_bug.cgi?id=89155
Failed to copy merge history to RenderSerachField.{cpp,h}
* rendering/RenderSearchField.cpp: Removed.
* rendering/RenderSearchField.h: Removed.
2012-06-17 Yoshifumi Inoue <yosin@chromium.org>
Unreviewed, rolling out r120432.
http://trac.webkit.org/changeset/120432
https://bugs.webkit.org/show_bug.cgi?id=88980
Failed to copy merge history to RenderSerachField.{cpp,h}
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::subtreeHasChanged):
(WebCore::HTMLInputElement::searchEventsShouldBeDispatched):
(WebCore):
(WebCore::HTMLInputElement::addSearchResult):
* html/HTMLInputElement.h:
(HTMLInputElement):
* html/InputType.cpp:
* html/InputType.h:
(InputType):
* html/SearchInputType.cpp:
* html/SearchInputType.h:
(SearchInputType):
* html/shadow/TextControlInnerElements.cpp:
(WebCore::SearchFieldResultsButtonElement::defaultEventHandler):
* loader/FormSubmission.cpp:
(WebCore::FormSubmission::create):
* rendering/RenderSearchField.cpp:
(WebCore::RenderTextControlInnerBlock::positionForPoint):
(WebCore):
(WebCore::RenderTextControlSingleLine::RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::~RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::containerElement):
(WebCore::RenderTextControlSingleLine::innerBlockElement):
(WebCore::RenderTextControlSingleLine::innerSpinButtonElement):
(WebCore::RenderTextControlSingleLine::resultsButtonElement):
(WebCore::RenderTextControlSingleLine::cancelButtonElement):
(WebCore::RenderTextControlSingleLine::textBaseStyle):
(WebCore::RenderTextControlSingleLine::addSearchResult):
(WebCore::RenderTextControlSingleLine::showPopup):
(WebCore::RenderTextControlSingleLine::hidePopup):
(WebCore::RenderTextControlSingleLine::paint):
(WebCore::RenderTextControlSingleLine::layout):
(WebCore::RenderTextControlSingleLine::nodeAtPoint):
(WebCore::RenderTextControlSingleLine::styleDidChange):
(WebCore::RenderTextControlSingleLine::capsLockStateMayHaveChanged):
(WebCore::RenderTextControlSingleLine::hasControlClip):
(WebCore::RenderTextControlSingleLine::controlClipRect):
(WebCore::RenderTextControlSingleLine::getAvgCharWidth):
(WebCore::RenderTextControlSingleLine::preferredContentWidth):
(WebCore::RenderTextControlSingleLine::computeControlHeight):
(WebCore::RenderTextControlSingleLine::updateFromElement):
(WebCore::RenderTextControlSingleLine::createInnerTextStyle):
(WebCore::RenderTextControlSingleLine::createInnerBlockStyle):
(WebCore::RenderTextControlSingleLine::updateCancelButtonVisibility):
(WebCore::RenderTextControlSingleLine::visibilityForCancelButton):
(WebCore::RenderTextControlSingleLine::textShouldBeTruncated):
(WebCore::RenderTextControlSingleLine::autosaveName):
(WebCore::RenderTextControlSingleLine::valueChanged):
(WebCore::RenderTextControlSingleLine::itemText):
(WebCore::RenderTextControlSingleLine::itemLabel):
(WebCore::RenderTextControlSingleLine::itemIcon):
(WebCore::RenderTextControlSingleLine::itemIsEnabled):
(WebCore::RenderTextControlSingleLine::itemStyle):
(WebCore::RenderTextControlSingleLine::menuStyle):
(WebCore::RenderTextControlSingleLine::clientInsetLeft):
(WebCore::RenderTextControlSingleLine::clientInsetRight):
(WebCore::RenderTextControlSingleLine::clientPaddingLeft):
(WebCore::RenderTextControlSingleLine::clientPaddingRight):
(WebCore::RenderTextControlSingleLine::listSize):
(WebCore::RenderTextControlSingleLine::selectedIndex):
(WebCore::RenderTextControlSingleLine::popupDidHide):
(WebCore::RenderTextControlSingleLine::itemIsSeparator):
(WebCore::RenderTextControlSingleLine::itemIsLabel):
(WebCore::RenderTextControlSingleLine::itemIsSelected):
(WebCore::RenderTextControlSingleLine::setTextFromItem):
(WebCore::RenderTextControlSingleLine::fontSelector):
(WebCore::RenderTextControlSingleLine::hostWindow):
(WebCore::RenderTextControlSingleLine::autoscroll):
(WebCore::RenderTextControlSingleLine::scrollWidth):
(WebCore::RenderTextControlSingleLine::scrollHeight):
(WebCore::RenderTextControlSingleLine::scrollLeft):
(WebCore::RenderTextControlSingleLine::scrollTop):
(WebCore::RenderTextControlSingleLine::setScrollLeft):
(WebCore::RenderTextControlSingleLine::setScrollTop):
(WebCore::RenderTextControlSingleLine::scroll):
(WebCore::RenderTextControlSingleLine::logicalScroll):
(WebCore::RenderTextControlSingleLine::createScrollbar):
(WebCore::RenderTextControlSingleLine::inputElement):
* rendering/RenderSearchField.h:
(RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::isTextField):
(WebCore::toRenderTextControlSingleLine):
(WebCore):
(RenderTextControlInnerBlock):
(WebCore::RenderTextControlInnerBlock::RenderTextControlInnerBlock):
(WebCore::RenderTextControlInnerBlock::hasLineIfEmpty):
* rendering/RenderTextControlSingleLine.cpp:
(WebCore::RenderTextControlSingleLine::RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::~RenderTextControlSingleLine):
(WebCore):
(WebCore::RenderTextControlSingleLine::containerElement):
(WebCore::RenderTextControlSingleLine::innerBlockElement):
(WebCore::RenderTextControlSingleLine::resultsButtonElement):
(WebCore::RenderTextControlSingleLine::cancelButtonElement):
(WebCore::RenderTextControlSingleLine::addSearchResult):
(WebCore::RenderTextControlSingleLine::showPopup):
(WebCore::RenderTextControlSingleLine::hidePopup):
(WebCore::RenderTextControlSingleLine::layout):
(WebCore::RenderTextControlSingleLine::computeControlHeight):
(WebCore::RenderTextControlSingleLine::updateFromElement):
(WebCore::RenderTextControlSingleLine::updateCancelButtonVisibility):
(WebCore::RenderTextControlSingleLine::visibilityForCancelButton):
(WebCore::RenderTextControlSingleLine::autosaveName):
(WebCore::RenderTextControlSingleLine::valueChanged):
(WebCore::RenderTextControlSingleLine::itemText):
(WebCore::RenderTextControlSingleLine::itemLabel):
(WebCore::RenderTextControlSingleLine::itemIcon):
(WebCore::RenderTextControlSingleLine::itemIsEnabled):
(WebCore::RenderTextControlSingleLine::itemStyle):
(WebCore::RenderTextControlSingleLine::menuStyle):
(WebCore::RenderTextControlSingleLine::clientInsetLeft):
(WebCore::RenderTextControlSingleLine::clientInsetRight):
(WebCore::RenderTextControlSingleLine::clientPaddingLeft):
(WebCore::RenderTextControlSingleLine::clientPaddingRight):
(WebCore::RenderTextControlSingleLine::listSize):
(WebCore::RenderTextControlSingleLine::selectedIndex):
(WebCore::RenderTextControlSingleLine::popupDidHide):
(WebCore::RenderTextControlSingleLine::itemIsSeparator):
(WebCore::RenderTextControlSingleLine::itemIsLabel):
(WebCore::RenderTextControlSingleLine::itemIsSelected):
(WebCore::RenderTextControlSingleLine::setTextFromItem):
(WebCore::RenderTextControlSingleLine::fontSelector):
(WebCore::RenderTextControlSingleLine::hostWindow):
(WebCore::RenderTextControlSingleLine::createScrollbar):
* rendering/RenderTextControlSingleLine.h:
(WebCore):
(RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::popupIsVisible):
* rendering/RenderingAllInOne.cpp:
2012-06-15 Darin Adler <darin@apple.com>
REGRESSION (r111041): Missing element type check in RenderThemeMac::paintMediaFullscreenButton
https://bugs.webkit.org/show_bug.cgi?id=89270
Reviewed by Oliver Hunt.
* rendering/RenderThemeMac.mm:
(WebCore::RenderThemeMac::paintMediaFullscreenButton): Use the proper idiom for getting
a media control element's type.
2012-06-17 Zeev Lieber <zlieber@chromium.org>
[Chromium] Compositor should avoid drawing quads when cached textures are available and contents unchanged
https://bugs.webkit.org/show_bug.cgi?id=88482
Reviewed by Adrienne Walker.
Post-processing CCRenderPassList after it's been generated and
removing all render surface quads for which there are
cached textures, and whose content didn't change. Added a new
flag to CCLayerImpl to differentiate surface property change and
layer property change.
The changes are covered by new unit tests that check that
the removal algorithm functions propertly. No rendering
behaviour change, so no new layout tests.
* platform/graphics/chromium/cc/CCDamageTracker.cpp:
(WebCore::layerNeedsToRedrawOntoItsTargetSurface):
(WebCore):
(WebCore::CCDamageTracker::extendDamageForLayer):
(WebCore::CCDamageTracker::extendDamageForRenderSurface):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::CCLayerImpl):
(WebCore::CCLayerImpl::layerSurfacePropertyChanged):
(WebCore):
(WebCore::CCLayerImpl::resetAllChangeTrackingForSubtree):
(WebCore::CCLayerImpl::setOpacity):
(WebCore::CCLayerImpl::setTransform):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
(WebCore::CCLayerTreeHostImpl::removeRenderPassesRecursive):
(WebCore):
(WebCore::CCLayerTreeHostImpl::removePassesWithCachedTextures):
(WebCore::CCLayerTreeHostImpl::prepareToDraw):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(FrameData):
(CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCRenderPass.h:
(CCRenderPass):
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore::CCRenderSurface::prepareContentsTexture):
(WebCore::CCRenderSurface::hasCachedContentsTexture):
(WebCore):
(WebCore::CCRenderSurface::contentsChanged):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(CCRenderSurface):
2012-06-17 Adam Barth <abarth@webkit.org>
Attempt to fix a large number of tests I broke with http://trac.webkit.org/changeset/120547
We need to reset the device scale facter after each test.
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::InternalSettings):
(WebCore::InternalSettings::restoreTo):
* testing/InternalSettings.h:
(InternalSettings):
2012-06-16 Huang Dongsung <luxtella@company100.net>
[Texmap] Share gaussian formula between shaders in TextureMapperShaderManager.
https://bugs.webkit.org/show_bug.cgi?id=89277
This patch makes blur and drop shadow shader share gaussian formula.
On the other hand, blur and drop shadow filter computed a gaussian weight in
pixel shader. However, a gaussian kernal has always same values, so this patch
computes the gaussian kernel only one time using CPU.
It is more accurate and faster.
Reviewed by Noam Rosenthal.
Covered by existing tests, particularly css3/filters/.
* platform/graphics/texmap/TextureMapperShaderManager.cpp:
(WebCore):
(WebCore::StandardFilterProgram::StandardFilterProgram):
(WebCore::gauss):
(WebCore::gaussianKernel):
(WebCore::StandardFilterProgram::prepare):
* platform/graphics/texmap/TextureMapperShaderManager.h:
2012-06-16 Robert Kroeger <rjkroege@chromium.org>
[chromium] Make the deviceScaleFactor dynamically adjustable.
https://bugs.webkit.org/show_bug.cgi?id=88916
Reviewed by James Robinson.
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp: Removed use of static
deviceScaleFactor from settings with dyanmic m_deviceScaleFactor
(WebCore::CCLayerTreeHost::CCLayerTreeHost):
(WebCore::CCLayerTreeHost::finishCommitOnImplThread):
(WebCore::CCLayerTreeHost::setViewportSize):
(WebCore::CCLayerTreeHost::updateLayers):
(WebCore::CCLayerTreeHost::setDeviceScaleFactor): Added function to
set deviceScaleFactor.
(WebCore):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCSettings::CCSettings): Removed static deviceScaleFactor
(CCSettings):
(CCLayerTreeHost): Added dynamic m_deviceScaleFactor in its place.
(WebCore::CCLayerTreeHost::deviceScaleFactor):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp: Same as above.
(WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::calculateRenderSurfaceLayerList):
(WebCore::CCLayerTreeHostImpl::setViewportSize):
(WebCore::CCLayerTreeHostImpl::setDeviceScaleFactor): Added function to
set deviceScaleFactor.
(WebCore):
(WebCore::CCLayerTreeHostImpl::updateMaxScrollPosition):
(WebCore::CCLayerTreeHostImpl::scrollBegin):
(WebCore::CCLayerTreeHostImpl::computePinchZoomDeltas):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h: Added dynamic
m_deviceScaleFactor instance variable.
(WebCore::CCLayerTreeHostImpl::deviceScaleFactor):
(CCLayerTreeHostImpl):
2012-06-16 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r120536.
http://trac.webkit.org/changeset/120536
https://bugs.webkit.org/show_bug.cgi?id=89296
Does not compile on chromium-linux (Requested by abarth on
#webkit).
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.cpp:
(WebCore):
(WebCore::FontPlatformData::setHinting):
(WebCore::FontPlatformData::setAutoHint):
(WebCore::FontPlatformData::setUseBitmaps):
(WebCore::FontPlatformData::setAntiAlias):
(WebCore::FontPlatformData::setSubpixelRendering):
(WebCore::FontPlatformData::setSubpixelPositioning):
(WebCore::FontPlatformData::setupPaint):
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.h:
(FontPlatformData):
2012-06-16 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r120539.
http://trac.webkit.org/changeset/120539
https://bugs.webkit.org/show_bug.cgi?id=89295
Does not compile on chromium-mac (Requested by abarth on
#webkit).
* platform/graphics/chromium/cc/CCDamageTracker.cpp:
(WebCore::CCDamageTracker::extendDamageForLayer):
(WebCore::CCDamageTracker::extendDamageForRenderSurface):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::CCLayerImpl):
(WebCore::CCLayerImpl::resetAllChangeTrackingForSubtree):
(WebCore::CCLayerImpl::setOpacity):
(WebCore::CCLayerImpl::setTransform):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
(WebCore::CCLayerTreeHostImpl::prepareToDraw):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCRenderPass.h:
(CCRenderPass):
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore::CCRenderSurface::prepareContentsTexture):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(CCRenderSurface):
2012-06-16 Mike West <mkwst@chromium.org>
Ignore paths in Content Security Policy sources rather than failing to parse them.
https://bugs.webkit.org/show_bug.cgi?id=89281
Reviewed by Adam Barth.
In short: `script-src http://example.com/` should allow scripts from
http://example.com. Currently, it allows no scripts at all, as the
terminal `/` isn't accepted as part of a hostname.
This patch adjusts CSPSourceList::parseSource to accept paths (and
discard them). Once this lands, the next step will be to keep the
path, and use it when comparing source origins in the various
allowXXXFromSource methods.
Tests: http/tests/security/contentSecurityPolicy/source-list-parsing-05.html
http/tests/security/contentSecurityPolicy/source-list-parsing-06.html
* page/ContentSecurityPolicy.cpp:
(CSPSourceList):
(WebCore):
(WebCore::CSPSourceList::parseSource):
Reworked this method entirely to support paths.
(WebCore::CSPSourceList::parsePath):
More or less a no-op at the moment.
(WebCore::CSPSourceList::parsePort):
Moved the `:` assertion here from parseSource.
2012-06-16 Zeev Lieber <zlieber@chromium.org>
[Chromium] Compositor should avoid drawing quads when cached textures are available and contents unchanged
https://bugs.webkit.org/show_bug.cgi?id=88482
Reviewed by Adrienne Walker.
Post-processing CCRenderPassList after it's been generated and
removing all render surface quads for which there are
cached textures, and whose content didn't change. Added a new
flag to CCLayerImpl to differentiate surface property change and
layer property change.
The changes are covered by new unit tests that check that
the removal algorithm functions propertly. No rendering
behaviour change, so no new layout tests.
* platform/graphics/chromium/cc/CCDamageTracker.cpp:
(WebCore::layerNeedsToRedrawOntoItsTargetSurface):
(WebCore):
(WebCore::CCDamageTracker::extendDamageForLayer):
(WebCore::CCDamageTracker::extendDamageForRenderSurface):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::CCLayerImpl):
(WebCore::CCLayerImpl::layerSurfacePropertyChanged):
(WebCore):
(WebCore::CCLayerImpl::resetAllChangeTrackingForSubtree):
(WebCore::CCLayerImpl::setOpacity):
(WebCore::CCLayerImpl::setTransform):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
(WebCore::CCLayerTreeHostImpl::removeRenderPassesRecursive):
(WebCore):
(WebCore::CCLayerTreeHostImpl::removePassesWithCachedTextures):
(WebCore::CCLayerTreeHostImpl::prepareToDraw):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(FrameData):
(CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCRenderPass.h:
(CCRenderPass):
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore::CCRenderSurface::prepareContentsTexture):
(WebCore::CCRenderSurface::hasCachedContentsTexture):
(WebCore):
(WebCore::CCRenderSurface::contentsChanged):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(CCRenderSurface):
2012-06-16 Xianzhu Wang <wangxianzhu@chromium.org>
[Chromium] Move chromium/public/linuxish/WebFontRendering.h out of linuxish directory
https://bugs.webkit.org/show_bug.cgi?id=89228
Reverse the dependency originally from WebKit::WebFontRendering to WebCore::FontPlatformDataHarfBuzz
so that WebKit::WebFontRendering can be platform-independent.
Reviewed by Tony Chang.
Refactory only. No new tests.
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.cpp:
(WebCore::FontPlatformData::setupPaint):
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.h:
(FontPlatformData):
2012-06-16 Adam Barth <abarth@webkit.org>
Settings::devicePixelRatio doesn't do anything and is confusing
https://bugs.webkit.org/show_bug.cgi?id=89272
Reviewed by James Robinson.
Settings::devicePixelRatio is yet another piece of state trying to
represent the device scale factor. The canonical place to store this
state is Page::m_deviceScaleFactor. Nothing in WebCore references
Settings::devicePixelRatio anymore, so we can remove it.
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(Settings):
2012-06-16 Huang Dongsung <luxtella@company100.net>
[Texmap] SIGSEV in WebCore::TextureMapperGL::drawTexture.
https://bugs.webkit.org/show_bug.cgi?id=89113
TextureMapperTile::m_texture is created lazilly, so we need null check before
using it.
Reviewed by Noam Rosenthal.
No new tests. This patch doesn't change behavior.
* platform/graphics/texmap/TextureMapperBackingStore.cpp:
(WebCore::TextureMapperTile::paint):
2012-06-16 Huang Dongsung <luxtella@company100.net>
[Texmap] Remove unused code in Texmap.
https://bugs.webkit.org/show_bug.cgi?id=89265
Reviewed by Noam Rosenthal.
* platform/graphics/texmap/GraphicsLayerTextureMapper.cpp:
(WebCore::GraphicsLayerTextureMapper::didSynchronize):
* platform/graphics/texmap/GraphicsLayerTextureMapper.h:
(GraphicsLayerTextureMapper):
* platform/graphics/texmap/TextureMapperLayer.cpp:
(WebCore::TextureMapperLayer::syncCompositingState):
* platform/graphics/texmap/TextureMapperLayer.h:
(TextureMapperLayer):
2012-06-16 Igor Oliveira <igor.o@sisa.samsung.com>
[TexmapGL] Reduce the number of glTexSubImage2D calls
https://bugs.webkit.org/show_bug.cgi?id=83665
Instead of copy the pixels row by row, put the pixels in a buffer and
call glTexSubImage2D just once.
Reviewed by Noam Rosenthal.
* platform/graphics/texmap/TextureMapperGL.cpp:
(WebCore::BitmapTextureGL::updateContents):
2012-06-15 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r120280.
http://trac.webkit.org/changeset/120280
https://bugs.webkit.org/show_bug.cgi?id=89273
Enabling CSS regions broke all Windows tests (Requested by
jhomeycutt on #webkit).
* css/CSSPropertyNames.in:
2012-06-15 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r120511.
http://trac.webkit.org/changeset/120511
https://bugs.webkit.org/show_bug.cgi?id=89255
Breaks at least Android builder (Requested by wangxianzhu on
#webkit).
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.cpp:
(WebCore):
(WebCore::FontPlatformData::setHinting):
(WebCore::FontPlatformData::setAutoHint):
(WebCore::FontPlatformData::setUseBitmaps):
(WebCore::FontPlatformData::setAntiAlias):
(WebCore::FontPlatformData::setSubpixelRendering):
(WebCore::FontPlatformData::setSubpixelPositioning):
(WebCore::FontPlatformData::setupPaint):
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.h:
(FontPlatformData):
2012-06-15 Xianzhu Wang <wangxianzhu@chromium.org>
[Chromium] Move chromium/public/linuxish/WebFontRendering.h out of linuxish directory
https://bugs.webkit.org/show_bug.cgi?id=89228
Reverse the dependency originally from WebKit::WebFontRendering to WebCore::FontPlatformDataHarfBuzz
so that WebKit::WebFontRendering can be platform-independent.
Reviewed by Tony Chang.
Refactory only. No new tests.
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.cpp:
(WebCore::FontPlatformData::setupPaint):
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.h:
(FontPlatformData):
2012-06-15 Adrienne Walker <enne@google.com>
[chromium] Fix composited scrollbars with transparent thumbs
https://bugs.webkit.org/show_bug.cgi?id=89247
Reviewed by James Robinson.
On some platforms, the thumb of a scrollbar can be transparent. Fix by
always drawing the thumb quad with blending.
* platform/graphics/chromium/cc/CCScrollbarLayerImpl.cpp:
(WebCore::CCScrollbarLayerImpl::appendQuads):
* platform/graphics/chromium/cc/CCTextureDrawQuad.cpp:
(WebCore::CCTextureDrawQuad::setNeedsBlending):
(WebCore):
* platform/graphics/chromium/cc/CCTextureDrawQuad.h:
(CCTextureDrawQuad):
2012-06-14 James Robinson <jamesr@chromium.org>
[chromium] Use SkBitmap in ImageLayerChromium
https://bugs.webkit.org/show_bug.cgi?id=89134
Reviewed by Adrienne Walker.
GraphicsLayer::setContentsToImage(Image*) is called whenever an image layer's image is or might have changed.
In Chromium, this used to hang on to a RefPtr<WebCore::Image> until the compositor was ready to upload texture contents.
This is potentially a bit fishy since the Image itself might not be in exactly the same state when we get around
to uploading textures and it also creates a bad dependency from ImageLayerChromium on WebCore::Image.
This patch grabs the underlying SkBitmap in the setContentsTo call and passes that into ImageLayerChromium
instead. I've also removed the venerable but redundant PlatformImage concept since all of chromium's images are
skia bitmaps these days.
Covered by existing tests, particularly compositing/images/ and compositing/color-matching/.
* WebCore.gypi:
* platform/graphics/chromium/GraphicsLayerChromium.cpp:
(WebCore::GraphicsLayerChromium::setContentsToImage):
* platform/graphics/chromium/ImageLayerChromium.cpp:
(WebCore::ImageLayerTextureUpdater::updateTextureRect):
(WebCore::ImageLayerTextureUpdater::setBitmap):
(ImageLayerTextureUpdater):
(WebCore::ImageLayerChromium::ImageLayerChromium):
(WebCore::ImageLayerChromium::setBitmap):
(WebCore::ImageLayerChromium::update):
(WebCore::ImageLayerChromium::contentBounds):
(WebCore::ImageLayerChromium::drawsContent):
* platform/graphics/chromium/ImageLayerChromium.h:
(ImageLayerChromium):
* platform/graphics/chromium/PlatformImage.cpp: Removed.
* platform/graphics/chromium/PlatformImage.h: Removed.
2012-06-15 Eli Fidler <efidler@rim.com>
[BlackBerry] Use platform font settings for the standard settings.
https://bugs.webkit.org/show_bug.cgi?id=89232
Reviewed by Rob Buis.
RIM PR 159708
* page/blackberry/SettingsBlackBerry.cpp:
(WebCore):
(WebCore::Settings::initializeDefaultFontFamilies):
2012-06-15 Joshua Bell <jsbell@chromium.org>
IndexedDB: Raise exceptions when methods are called on deleted objects
https://bugs.webkit.org/show_bug.cgi?id=89243
Reviewed by Tony Chang.
Implement the IDB spec requirement that InvalidStateError exceptions are
thrown when methods are called on objects (i.e. object stores and indexes)
that have been deleted within a version change transaction.
Test: storage/indexeddb/deleted-objects.html
* Modules/indexeddb/IDBDatabase.cpp:
(WebCore::IDBDatabase::deleteObjectStore): Don't relay to transaction if back-end failed.
* Modules/indexeddb/IDBIndex.cpp: Check deleted state in methods, raise if set.
(WebCore::IDBIndex::IDBIndex):
(WebCore::IDBIndex::openCursor):
(WebCore::IDBIndex::count):
(WebCore::IDBIndex::openKeyCursor):
(WebCore::IDBIndex::get):
(WebCore::IDBIndex::getKey):
* Modules/indexeddb/IDBIndex.h: Add flag to track deleted state, method to mark it.
(WebCore::IDBIndex::markDeleted):
(IDBIndex):
* Modules/indexeddb/IDBObjectStore.cpp: Check deleted state in methods, raise if set.
(WebCore::IDBObjectStore::IDBObjectStore):
(WebCore::IDBObjectStore::get):
(WebCore::IDBObjectStore::add):
(WebCore::IDBObjectStore::put):
(WebCore::IDBObjectStore::deleteFunction):
(WebCore::IDBObjectStore::clear):
(WebCore::IDBObjectStore::createIndex):
(WebCore::IDBObjectStore::index):
(WebCore::IDBObjectStore::deleteIndex): If the index being deleted has been instantiated,
mark it as deleted.
(WebCore::IDBObjectStore::openCursor):
(WebCore::IDBObjectStore::count):
* Modules/indexeddb/IDBObjectStore.h: Add flag to track deleted state, method to mark it.
(WebCore::IDBObjectStore::markDeleted):
(IDBObjectStore):
* Modules/indexeddb/IDBTransaction.cpp:
(WebCore::IDBTransaction::objectStore):
(WebCore::IDBTransaction::objectStoreDeleted): If the store being deleted has been instantiated,
mark it as deleted.
2012-06-15 James Robinson <jamesr@chromium.org>
[chromium] Fix LayoutTests/platform/chromium/compositing/accelerated-drawing/svg-filters.html
https://bugs.webkit.org/show_bug.cgi?id=89126
Reviewed by Stephen White.
When drawing an image buffer into a deferred device context, we have to make a deep copy of the source (Skia
can't for some reason). Stephen White fixed this in r101325 by adding a bit on PlatformContextSkia indicating if
a given context is deferred, but I broke this again in r120346 by refactoring the SkCanvas creation to happen in
a different location from the PlatformContextSkia construction.
This moves the check into ImageBufferSkia and directly queries the state of the destination SkDevice to be more
robust and work without needing special PlatformContextSkia setup.
Covered by LayoutTests/platform/chromium/compositing/accelerated-drawing/svg-filters.html
* platform/graphics/skia/ImageBufferSkia.cpp:
(WebCore::drawNeedsCopy):
* platform/graphics/skia/PlatformContextSkia.cpp:
(WebCore::PlatformContextSkia::PlatformContextSkia):
* platform/graphics/skia/PlatformContextSkia.h:
(PlatformContextSkia):
2012-06-15 Ami Fischman <fischman@chromium.org>
[chromium] Compositor should be aware of |flipped| status of video textures per-platform
https://bugs.webkit.org/show_bug.cgi?id=89189
Reviewed by James Robinson.
No new tests (sadly HW video decode is still only being tested manually for orientation).
* platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
(WebCore::CCVideoLayerImpl::appendQuads):
2012-06-14 Ryosuke Niwa <rniwa@webkit.org>
The initial value of text-align should be start instead of -webkit-auto
https://bugs.webkit.org/show_bug.cgi?id=79914
Reviewed by Tony Chang.
Change the initial value of text-align CSS property from -webkit-auto to start as specified in
http://www.w3.org/TR/css3-text/#text-align
In editing, we have to coerce start and end to left and right for now because match-parent,
which is the text-align value of li in UA stylesheet is going to result in the computed values of
left as supposed to start by default, and this causes editing code not being able to remove
text-align: left when computing typing styles and preserving styles.
In the long term, we should detect this specific case and ignore match-parent but that seemed like
too much work to be done in a single patch also because the test result improved because of this change.
This behavior change is covered by existing regression tests.
* accessibility/gtk/WebKitAccessibleInterfaceText.cpp:
(WebCore::getAttributeSetForAccessibilityObject):
* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue): No longer supports TAAUTO since it's identical to TASTART.
(WebCore::CSSPrimitiveValue::operator ETextAlign): Ditto. Parse -webkit-auto as start.
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList): Replace TAAUTO by TASTART.
* editing/EditingStyle.cpp:
(WebCore):
(WebCore::textAlignResolvingStartAndEnd): Coerce start and end into left and right based on the directionality
for editing. Otherwise, we end up adding lots of text-align: left due to li's UA style rule having match-parent
as the text alignment and its descendent inherits this style. We need to handle it better in the future
since start and left or end and right are semantically different.
(WebCore::EditingStyle::prepareToApplyAt):
(WebCore::getPropertiesNotIn):
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::textAlignmentForLine): Justified text is aligned at start by default.
(WebCore::RenderBlock::updateLogicalWidthForAlignment):
(WebCore::RenderBlock::startAlignedOffsetForLine):
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::localCaretRectForEmptyElement):
* rendering/RenderListBox.cpp:
(WebCore::itemOffsetForAlignment):
* rendering/RenderMarquee.cpp:
(WebCore::RenderMarquee::updateMarqueeStyle):
* rendering/RenderRubyText.cpp:
(WebCore::RenderRubyText::textAlignmentForLine):
(WebCore::RenderRubyText::adjustInlineDirectionLineBounds):
* rendering/RenderText.cpp:
(WebCore::RenderText::localCaretRect):
* rendering/style/RenderStyle.h:
* rendering/style/RenderStyleConstants.h:
2012-06-15 Jian Li <jianli@chromium.org>
[File API] FileReader should work in sandbox iframe
https://bugs.webkit.org/show_bug.cgi?id=89242
Reviewed by Adam Barth.
Test: fast/files/file-reader-sandbox-iframe.html
* fileapi/ThreadableBlobRegistry.cpp: Should remove from map for all threads.
(WebCore::ThreadableBlobRegistry::unregisterBlobURL):
* page/SecurityOrigin.cpp:
(WebCore::SecurityOrigin::canRequest): Check and use cached origin for blob URL first.
2012-06-15 David Barton <dbarton@mathscribe.com>
Inherit style changes in MathML anonymous renderers
https://bugs.webkit.org/show_bug.cgi?id=88476
Reviewed by Julien Chaffraix.
A RenderObject that is not the main renderer() for a DOM node is called "anonymous".
Standard WebCore practice is to mark such a renderer as isAnonymous(). For example,
RenderBlock::styleDidChange calls propagateStyleToAnonymousChildren to flow style
changes to such children, by re-initializing their styles and then inheriting from
this->style(). A derived class' styleDidChange() can then set non-default style
properties as needed.
This patch implements this standard practice for RenderMathMLBlock classes, except
RenderMathMLOperator which currently uses a RenderLayer, which reportedly means it
should not be isAnonymous(). We also follow common practice and change
isAnonymousBlock() to return false for RenderMathMLBlock classes, since
isAnonymousBlock() is really used by RenderBlock to detect its own anonymous blocks for
wrapping inline children, which RenderBlock then combines or deletes assuming this.
Test: mathml/presentation/style-changed.html, also added to mathml/presentation/over.xhtml
* rendering/RenderObject.h:
(WebCore::RenderObject::isAnonymousBlock):
* rendering/RenderTreeAsText.cpp:
(WebCore::RenderTreeAsText::writeRenderObject):
* rendering/mathml/RenderMathMLBlock.cpp:
(WebCore::RenderMathMLBlock::createAnonymousMathMLBlock):
(WebCore::RenderMathMLBlock::renderName):
* rendering/mathml/RenderMathMLBlock.h:
* rendering/mathml/RenderMathMLFenced.cpp:
(WebCore::RenderMathMLFenced::createMathMLOperator):
(WebCore::RenderMathMLFenced::makeFences):
(WebCore::RenderMathMLFenced::addChild):
(WebCore::RenderMathMLFenced::styleDidChange):
* rendering/mathml/RenderMathMLFenced.h:
* rendering/mathml/RenderMathMLFraction.cpp:
(WebCore::RenderMathMLFraction::RenderMathMLFraction):
(WebCore::RenderMathMLFraction::fixChildStyle):
(WebCore::RenderMathMLFraction::addChild):
(WebCore::RenderMathMLFraction::styleDidChange):
* rendering/mathml/RenderMathMLFraction.h:
* rendering/mathml/RenderMathMLOperator.cpp:
(WebCore::RenderMathMLOperator::styleDidChange):
* rendering/mathml/RenderMathMLOperator.h:
* rendering/mathml/RenderMathMLSquareRoot.h:
* rendering/mathml/RenderMathMLSubSup.cpp:
(WebCore::RenderMathMLSubSup::fixScriptsStyle):
(WebCore::RenderMathMLSubSup::addChild):
(WebCore::RenderMathMLSubSup::styleDidChange):
* rendering/mathml/RenderMathMLSubSup.h:
* rendering/mathml/RenderMathMLUnderOver.cpp:
(WebCore::RenderMathMLUnderOver::addChild):
(WebCore::RenderMathMLUnderOver::styleDidChange):
* rendering/mathml/RenderMathMLUnderOver.h:
2012-06-13 Vincent Scheib <scheib@chromium.org>
Add new Pointer Lock spec webkitRequestPointerLock and webkitExitPointerLock methods.
https://bugs.webkit.org/show_bug.cgi?id=88891
Reviewed by Dimitri Glazkov.
Part of a series of refactoring changes to update pointer lock API to
the fullscreen locking style. https://bugs.webkit.org/show_bug.cgi?id=84402
Entering and exiting mouse lock is provided in the new webkitRequestPointerLock and
webkitExitPointerLock methods.
Existing pointer-lock tests updated to use the new methods.
* dom/Document.cpp:
(WebCore::Document::webkitExitPointerLock):
(WebCore):
* dom/Document.h:
(Document):
* dom/Document.idl:
* dom/Element.cpp:
(WebCore):
(WebCore::Element::webkitRequestPointerLock):
* dom/Element.h:
* dom/Element.idl:
2012-06-15 Tony Payne <tpayne@chromium.org>
[chromium] Add iccjpeg and qcms to chromium port
https://bugs.webkit.org/show_bug.cgi?id=81974
Reviewed by Adam Barth.
Covered by existing layout tests which will be rebaselined.
* WebCore.gyp/WebCore.gyp: Add qcms to the build.
* platform/image-decoders/ImageDecoder.h:
(WebCore::ImageDecoder::qcmsOutputDeviceProfile): Return an sRGB profile.
On OSX, return the default RGB profile. Add FIXME to use the user's
monitor profile and verify that profile for other platforms.
* platform/image-decoders/jpeg/JPEGImageDecoder.cpp:
(turboSwizzled): For libjpeg-turbo, JCS_EXT_BGRA and JCS_EXT_RGBA are the
two known output color spaces for which the decoder uses a data swizzle.
(colorSpaceHasAlpha): JPEG's have no alpha in the output color space. For
libjpeg-turbo, alpha may be present in the swizzled output color space.
(WebCore::JPEGImageReader::JPEGImageReader):
(WebCore::JPEGImageReader::close):
(WebCore::JPEGImageReader::decode): For QCMSLIB, create the color transform
to use during decoding, and ensure we switch to inputing RGBA data to qcms
even if the desired output data is BGRA: outputScanlines() sends BGRA data
to the frame buffer following color correction if needed.
(JPEGImageReader):
(WebCore::JPEGImageReader::colorTransform): qcms color transform getter.
(WebCore::JPEGImageReader::createColorTransform): Create color transform.
Release the existing transform (if any) and assign to the color transform
created from the color profile data.
(WebCore::JPEGImageDecoder::outputScanlines): Minor style fix. Apply color
transform to each decoded image row.
* platform/image-decoders/png/PNGImageDecoder.cpp:
(WebCore::PNGImageReader::PNGImageReader):
(WebCore::PNGImageReader::close):
(WebCore::PNGImageReader::currentBufferSize): Move this adjacent to other
setters and getters.
(WebCore::PNGImageReader::decodingSizeOnly): Ditto.
(WebCore::PNGImageReader::setHasAlpha): Ditto.
(WebCore::PNGImageReader::hasAlpha): Ditto.
(WebCore::PNGImageReader::interlaceBuffer): Ditto.
(WebCore::PNGImageReader::createRowBuffer): Creates a temporary row buffer,
used when a color transform is applied to the decoded image pixels.
(WebCore::PNGImageReader::rowBuffer): Return the temporary row buffer.
(WebCore::PNGImageReader::colorTransform): qcms color transform getter.
(WebCore::PNGImageReader::createColorTransform): Create color transform.
Release the existing transform (if any) and assign to the color transform
created from the color profile data.
(WebCore::PNGImageDecoder::headerAvailable): For QCMSLIB, create the color
transform to use for decoding. Clear m_colorProfile (not used anymore).
(WebCore::PNGImageDecoder::rowAvailable): Create temporary row buffer if
a color transform is needed for decoding. Apply color transform to each
decoded image row.
* platform/image-decoders/skia/ImageDecoderSkia.cpp:
(WebCore::ImageFrame::setColorProfile): Old method of colorProfiles is no
longer used. Add a FIXME to remove the old implementation.
(WebCore::ImageFrame::setStatus): Remove old color correction code.
2012-06-15 Silvia Pfeiffer <silviapf@chromium.org>
Remove volume thumb for videos without audio track.
https://bugs.webkit.org/show_bug.cgi?id=89093
Reviewed by Eric Carlson.
No new tests, since this was already tested in media/video-no-audio.html.
* rendering/RenderMediaControlsChromium.cpp:
(WebCore::paintMediaMuteButton):
Change mute button when there is no audio or no source file.
(WebCore::paintMediaVolumeSlider):
Set volume slider to 0 when there is no audio or no source file.
(WebCore::paintMediaVolumeSliderThumb):
Don't paint the volume slider thumb when there is no audio or no source file.
2012-06-15 Abhishek Arya <inferno@chromium.org>
Cleanup empty anonymous block continuation.
https://bugs.webkit.org/show_bug.cgi?id=74976
Reviewed by Julien Chaffraix.
Fix rendering on http://docs.google.com/demo.
Test: fast/inline/inline-empty-block-continuation-remove.html
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::removeChild): If we are removing our last child,
and are an anonymous block which is part of a continuation chain, then
we need to first fix our continuation chain by setting our previous
continuation renderer to point to the next continuation renderer. Then,
we reset our continuation pointer and destroy ourselves. Since we no
longer are part of continuation chain, we will be removed and previous
and next anonymous block will be merged automatically.
* rendering/RenderObject.cpp:
(WebCore::RenderObject::previousInPreOrder): add an argument to previousInOrder
- stayWithin.
(WebCore):
* rendering/RenderObject.h:
(RenderObject):
2012-06-15 Taiju Tsuiki <tzik@chromium.org>
Web Inspector: Move FileSystem frame management from frontend to backend
https://bugs.webkit.org/show_bug.cgi?id=89190
Reviewed by Vsevolod Vlasov.
Test: http/tests/inspector/filesystem/read-directory.html
* inspector/Inspector.json:
* inspector/InspectorFileSystemAgent.cpp:
(WebCore::InspectorFileSystemAgent::readDirectory):
(WebCore::InspectorFileSystemAgent::InspectorFileSystemAgent):
(WebCore::InspectorFileSystemAgent::scriptExecutionContextForOrigin):
(WebCore):
* inspector/InspectorFileSystemAgent.h:
(WebCore):
(InspectorFileSystemAgent):
2012-06-15 Alexander Pavlov <apavlov@chromium.org>
Unreviewed, build fix after r120469.
* css/CSSPropertySourceData.h:
(WebCore):
2012-06-15 Taiju Tsuiki <tzik@chromium.org>
Web Inspector: Add FileSystemRequestManager and FileSystemDispatcher
https://bugs.webkit.org/show_bug.cgi?id=89191
Reviewed by Vsevolod Vlasov.
* inspector/front-end/FileSystemModel.js:
(WebInspector.FileSystemModel):
(WebInspector.FileSystemRequestManager):
(WebInspector.FileSystemRequestManager.prototoype._requestId):
(WebInspector.FileSystemRequestManager.prototoype.readDirectory):
(WebInspector.FileSystemRequestManager.prototoype._didReadDirectory):
(WebInspector.FileSystemDispatcher):
(WebInspector.FileSystemDispatcher.prototype.gotFileSystemRoot):
(WebInspector.FileSystemDispatcher.prototype.didReadDirectory):
2012-06-15 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: extension's Resource.getContent always returns original revision.
https://bugs.webkit.org/show_bug.cgi?id=89219
Reviewed by Vsevolod Vlasov.
Fetching proper content from the UISourceCode now.
* inspector/front-end/DebuggerResourceBinding.js:
(WebInspector.DebuggerResourceBinding.prototype.canSetContent):
(WebInspector.DebuggerResourceBinding.prototype.setContent):
* inspector/front-end/DebuggerScriptMapping.js:
(WebInspector.DebuggerScriptMapping.prototype._debuggerReset):
* inspector/front-end/ExtensionServer.js:
(WebInspector.ExtensionServer.prototype._onGetResourceContent):
* inspector/front-end/JavaScriptSource.js:
(WebInspector.JavaScriptSource):
* inspector/front-end/Resource.js:
(WebInspector.Resource.prototype.isHidden):
(WebInspector.Resource.prototype.uiSourceCode):
(WebInspector.Resource.prototype.setUISourceCode):
* inspector/front-end/UISourceCode.js:
(WebInspector.UISourceCode):
2012-06-15 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: CRASH when DebuggerAgent.setBreakpoint was called twice for the same location.
https://bugs.webkit.org/show_bug.cgi?id=89225
Reviewed by Pavel Feldman.
* inspector/InspectorDebuggerAgent.cpp:
(WebCore::InspectorDebuggerAgent::setBreakpoint):
2012-06-08 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: CSSParser::parseSheet() should provide ready-to-use source data
https://bugs.webkit.org/show_bug.cgi?id=88646
Reviewed by Antti Koivisto.
This change moves the post-processing step from InspectorStyleSheet into CSSParser, so that
CSSParser::parseSheet() will return a ready-to-use list with style rule source code data.
Also, universal data structures are introduced, which allow for the full rule source data tree building.
No new tests, as this is a refactoring.
* css/CSSParser.cpp: Use universal data structures, which can be used for building the full rule tree.
(WebCore::CSSParser::CSSParser):
(WebCore::CSSParser::setupParser):
(WebCore::CSSParser::parseSheet): Return ready-to-use source code data entries rather than an intermediate structure.
(WebCore::CSSParser::parseDeclaration):
(WebCore::CSSParser::addNewRuleToSourceTree):
(WebCore):
(WebCore::CSSParser::popRuleData):
(WebCore::CSSParser::createStyleRule):
(WebCore::CSSParser::fixUnparsedPropertyRanges): Moved in from InspectorStyleSheet.
(WebCore::CSSParser::markSelectorListStart):
(WebCore::CSSParser::markSelectorListEnd):
(WebCore::CSSParser::markRuleBodyStart):
(WebCore::CSSParser::markRuleBodyEnd):
(WebCore::CSSParser::markPropertyEnd):
* css/CSSParser.h:
(CSSParser):
(WebCore::CSSParser::resetPropertyRange): Renamed.
(WebCore::CSSParser::isExtractingSourceData): A convenience check.
* css/CSSPropertySourceData.h: Introduce the RuleSourceDataList typedef.
(WebCore):
* inspector/InspectorStyleSheet.cpp: Make use of RuleSourceDataList and follow the CSSParser::parse*() API changes.
(ParsedStyleSheet::sourceData):
(ParsedStyleSheet):
(ParsedStyleSheet::setSourceData):
(WebCore::InspectorStyleSheet::ensureSourceData): Remove source data postprocessing, follow the new parseSheet() API.
* inspector/InspectorStyleSheet.h:
(WebCore::InspectorCSSId::InspectorCSSId): Drive-by: uninitialized field fix.
(WebCore::InspectorStyleProperty::InspectorStyleProperty): Ditto.
(InspectorStyleSheet):
2012-06-15 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: Long frame urls make all/errors/warnings/logs buttons inaccessible.
https://bugs.webkit.org/show_bug.cgi?id=88907
Reviewed by Vsevolod Vlasov.
* inspector/front-end/inspector.css:
(#console-context):
2012-06-15 Florin Malita <fmalita@chromium.org>
Specular light filters produce dark results
https://bugs.webkit.org/show_bug.cgi?id=89116
Reviewed by Dirk Schulze.
Tests: svg/filters/feSpecularLight-premultiplied-expected.svg
svg/filters/feSpecularLight-premultiplied.svg
Per spec (http://www.w3.org/TR/SVG/filters.html#feSpecularLightingElement),
light filters should operate on pre-multiplied RGBA. Currently, the results
are stored in the unmultiplied buffer but the alpha channel is calculated
for premultiplied values (which causes a darkening of the result upon the
subsequent unmultiplied->premultiplied conversion). This patch updates
FELighting to generate premultiplied results.
* platform/graphics/filters/FELighting.cpp:
(WebCore::FELighting::platformApplySoftware):
2012-06-15 Andrey Kosyakov <caseq@chromium.org>
Web Inspector: incorrect filtering of async timeline events
https://bugs.webkit.org/show_bug.cgi?id=89214
Reviewed by Vsevolod Vlasov.
* inspector/front-end/TimelineOverviewPane.js:
(WebInspector.TimelineOverviewPane.prototype.accept):
2012-06-15 Stephen Chenney <schenney@chromium.org>
SVG Composite of Offset filters incorrectly clips
https://bugs.webkit.org/show_bug.cgi?id=77660
Reviewed by Dirk Schulze.
Prior to this patch, when a group of filtered objects was used as input to another filter,
the filter only operated on the stroke boundary of the group, and hence excluded the results
of filtering elements within the group, or extraneously included regions clipped from the
elements in the group.
This patch modifies the strokeBoundingBox of SVG container elements to
be the union of the repaint rects for the children. This modifes the
results returned for sizing filters and for absoluteRects, which will cause
inline layout around the group to factor in the resources applied to
the group's children.
The relevant spec entry is this, in Section 3.7 of the SVG 1.1 spec: "...the result must be
as though the paint operations had been applied to an intermediate canvas initialized to
transparent black, of a size determined by the rules given in Filter Effects then filtered
by the processes defined in Filter Effects." In this case the "paint operations" is implied
to include the result of applying "paint" but no resources to the group, which in turn would
have resources applied to the children of the group. This makes the most sense, as the current,
incorrect behavior makes it extremely diffucult to understand the actions of filters on
groups of filtered content.
Tests: svg/filters/container-with-filters-expected.svg
svg/filters/container-with-filters.svg
* rendering/svg/RenderSVGContainer.cpp:
(WebCore::RenderSVGContainer::updateCachedBoundaries):
* rendering/svg/RenderSVGRoot.cpp:
(WebCore::RenderSVGRoot::updateCachedBoundaries):
* rendering/svg/SVGRenderSupport.cpp:
(WebCore::SVGRenderSupport::computeContainerBoundingBoxes):
2012-06-15 David Kilzer <ddkilzer@apple.com>
Sort ENABLE(INSPECTOR) section of WebCore.exp.in
* WebCore.exp.in: Sort ENABLE(INSPECTOR) symbols.
2012-06-15 David Kilzer <ddkilzer@apple.com>
Remove duplicate symbol from WebCore.exp.in
* WebCore.exp.in: Remove duplicate symbol for:
WebCore::Range::textQuads(WTF::Vector<WebCore::FloatQuad, 0ul>&, bool, WebCore::Range::RangeInFixedPosition*) const
2012-06-15 Florin Malita <fmalita@chromium.org>
[Chromium] Unreviewed debug build fix for r120457
* platform/graphics/chromium/cc/CCDrawQuad.cpp:
(WebCore::CCDrawQuad::toYUVVideoDrawQuad):
2012-06-15 'Pavel Feldman' <pfeldman@chromium.org>
Not reviewed: remove redundant Mac-specific style rule from inspector front-end.
* inspector/front-end/helpScreen.css:
2012-06-15 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: fix help close button appearance on Mac
https://bugs.webkit.org/show_bug.cgi?id=89208
Reviewed by Vsevolod Vlasov.
* inspector/front-end/helpScreen.css:
(.help-close-button):
(body.platform-mac .help-close-button):
(body.platform-mac .help-window-main .tabbed-pane-header-contents):
2012-06-11 Dana Jansens <danakj@chromium.org>
[chromium] Create a CCYUVVideoDrawQuad and remove the now-unused generic CCVideoDrawQuad
https://bugs.webkit.org/show_bug.cgi?id=88828
Reviewed by Adrienne Walker.
The CCVideoDrawQuad is now only used for YUV video, so we remove the
class and replace it with CCYUVVideoDrawQuad. This class holds what is
needed to draw a YUV video frame.
No new tests, no change in behaviour.
* WebCore.gypi:
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::drawQuad):
(WebCore::LayerRendererChromium::drawYUVVideoQuad):
* platform/graphics/chromium/LayerRendererChromium.h:
(WebCore):
(LayerRendererChromium):
* platform/graphics/chromium/cc/CCDrawQuad.cpp:
(WebCore::CCDrawQuad::toYUVVideoDrawQuad):
* platform/graphics/chromium/cc/CCDrawQuad.h:
(WebCore):
(CCDrawQuad):
* platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
(WebCore::CCVideoLayerImpl::appendQuads):
* platform/graphics/chromium/cc/CCYUVVideoDrawQuad.cpp: Renamed from Source/WebCore/platform/graphics/chromium/cc/CCVideoDrawQuad.cpp.
(WebCore):
(WebCore::CCYUVVideoDrawQuad::create):
(WebCore::CCYUVVideoDrawQuad::CCYUVVideoDrawQuad):
* platform/graphics/chromium/cc/CCYUVVideoDrawQuad.h: Renamed from Source/WebCore/platform/graphics/chromium/cc/CCVideoDrawQuad.h.
(WebCore):
(CCYUVVideoDrawQuad):
(WebCore::CCYUVVideoDrawQuad::yPlane):
(WebCore::CCYUVVideoDrawQuad::uPlane):
(WebCore::CCYUVVideoDrawQuad::vPlane):
2012-06-15 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: CRASH: getProfile is crashing for unknown profiles.
https://bugs.webkit.org/show_bug.cgi?id=89202
agents' functions have to set a value to errorString if it can't assign values to the mandatory out arguments.
Reviewed by Pavel Feldman.
Test: inspector/profiler/heap-snapshot-get-profile-crash.html
* inspector/InspectorProfilerAgent.cpp:
(WebCore::InspectorProfilerAgent::getProfile):
2012-06-15 Max Feil <mfeil@rim.com>
[BlackBerry] media volume slider in wrong position (master_38 regression)
https://bugs.webkit.org/show_bug.cgi?id=89165
Reviewed by Antonio Gomes.
Due to upstream simplification of the volume slider (webkit.org
bug 82150) it now appears in the wrong location below the
media controls. The programmatic controls sizing we do in
RenderThemeBlackBerry.cpp relies on absolute positioning of the
volume slider container, so the best way to fix this is to set
the bottom offset.
No new tests since this is a regression fix. BlackBerry media
controls are manually tested.
* platform/blackberry/RenderThemeBlackBerry.cpp:
(WebCore::RenderThemeBlackBerry::adjustMediaControlStyle):
2012-06-15 'Pavel Feldman' <pfeldman@chromium.org>
Not reviewed: remove garbage line from inspector view in docked-to-right mode.
* inspector/front-end/inspectorCommon.css:
(body.docked.dock-to-right.docked):
(body.docked.dock-to-right.inactive):
2012-06-15 Zoltan Arvai <zarvai@inf.u-szeged.hu>
Buildfix for !ENABLE(BLOB) platforms after r120433.
https://bugs.webkit.org/show_bug.cgi?id=78648
Reviewed by Csaba Osztrogonác.
* fileapi/ThreadableBlobRegistry.cpp:
(WebCore::ThreadableBlobRegistry::registerBlobURL):
2012-06-15 Sami Kyostila <skyostil@chromium.org>
[chromium] Allow scrolling non-root layers in the compositor thread
https://bugs.webkit.org/show_bug.cgi?id=73350
Reviewed by James Robinson.
This patch enables scrolling child layers in the compositor thread.
Scroll deltas are accumulated for each scrolled CCLayerImpl and
synchronized to the main thread.
If a layer has no room to scroll in a given direction, one of its
ancestor layers is scrolled instead if possible.
Layer hit testing code by Shawn Singh.
Added new unit tests to verify layer scrolling behavior:
CCLayerTreeHostCommonTest.verifySubtreeSearch
CCLayerTreeHostImplTest.clearRootRenderSurfaceAndScroll
CCLayerTreeHostImplTest.inhibitScrollAndPageScaleUpdatesWhileAnimatingPageScale
CCLayerTreeHostImplTest.inhibitScrollAndPageScaleUpdatesWhilePinchZooming
CCLayerTreeHostImplTest.replaceTreeWhileScrolling
CCLayerTreeHostImplTest.scrollBeforeRedraw
CCLayerTreeHostImplTest.scrollBlockedByContentLayer
CCLayerTreeHostImplTest.scrollChildAndChangePageScaleOnMainThread
CCLayerTreeHostImplTest.scrollChildBeyondLimit
CCLayerTreeHostImplTest.scrollChildCallsCommitAndRedraw
CCLayerTreeHostImplTest.scrollEventBubbling
CCLayerTreeHostImplTest.scrollMissesBackfacingChild
CCLayerTreeHostImplTest.scrollMissesChild
CCLayerTreeHostImplTest.scrollNonCompositedRoot
CCLayerTreeHostImplTest.scrollRootAndChangePageScaleOnImplThread
CCLayerTreeHostImplTest.scrollRootAndChangePageScaleOnMainThread
CCLayerTreeHostImplTest.scrollRootIgnored
CCLayerTreeHostImplTest.scrollWithoutRootLayer
CCLayerTreeHostTestScrollChildLayer
WebCompositorInputHandlerImplTest.gestureScrollOnMainThread
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::LayerChromium):
(WebCore::LayerChromium::setMaxScrollPosition):
(WebCore):
(WebCore::LayerChromium::scrollBy):
(WebCore::LayerChromium::pushPropertiesTo):
* platform/graphics/chromium/LayerChromium.h:
(WebCore):
(LayerChromiumScrollDelegate):
(WebCore::LayerChromiumScrollDelegate::~LayerChromiumScrollDelegate):
(LayerChromium):
(WebCore::LayerChromium::maxScrollPosition):
(WebCore::LayerChromium::scrollable):
(WebCore::LayerChromium::setLayerScrollDelegate):
* platform/graphics/chromium/cc/CCInputHandler.h:
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::tryScroll):
(WebCore):
(WebCore::sortLayers):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::finishCommitOnImplThread):
(WebCore::findFirstScrollableLayer):
(WebCore):
(WebCore::CCLayerTreeHost::applyScrollAndScale):
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.h:
(CCLayerTreeHostCommon):
(WebCore):
(WebCore::CCLayerTreeHostCommon::findLayerInSubtree):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::~CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::startPageScaleAnimation):
(WebCore::CCLayerTreeHostImpl::calculateRenderSurfaceLayerList):
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
(WebCore::CCLayerTreeHostImpl::contentSize):
(WebCore::CCLayerTreeHostImpl::prepareToDraw):
(WebCore::CCLayerTreeHostImpl::drawLayers):
(WebCore::findRootScrollLayer):
(WebCore):
(WebCore::findScrollLayerForContentLayer):
(WebCore::CCLayerTreeHostImpl::setRootLayer):
(WebCore::CCLayerTreeHostImpl::detachLayerTree):
(WebCore::adjustScrollsForPageScaleChange):
(WebCore::applyPageScaleDeltaToScrollLayers):
(WebCore::CCLayerTreeHostImpl::setPageScaleFactorAndLimits):
(WebCore::CCLayerTreeHostImpl::setPageScaleDelta):
(WebCore::CCLayerTreeHostImpl::updateMaxScrollPosition):
(WebCore::CCLayerTreeHostImpl::ensureRenderSurfaceLayerList):
(WebCore::CCLayerTreeHostImpl::clearCurrentlyScrollingLayer):
(WebCore::CCLayerTreeHostImpl::scrollBegin):
(WebCore::CCLayerTreeHostImpl::scrollBy):
(WebCore::CCLayerTreeHostImpl::scrollEnd):
(WebCore::CCLayerTreeHostImpl::pinchGestureUpdate):
(WebCore::CCLayerTreeHostImpl::computePinchZoomDeltas):
(WebCore::CCLayerTreeHostImpl::makeScrollAndScaleSet):
(WebCore::collectScrollDeltas):
(WebCore::CCLayerTreeHostImpl::processScrollDeltas):
(WebCore::CCLayerTreeHostImpl::animatePageScale):
(WebCore::CCLayerTreeHostImpl::animateLayers):
(WebCore::CCLayerTreeHostImpl::clearRenderSurfaces):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(FrameData):
(CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::rootScrollLayer):
2012-06-15 Jian Li <jianli@chromium.org>
FileReader is dysfunctional in documents with "null" origin string
https://bugs.webkit.org/show_bug.cgi?id=78648
Reviewed by Adam Barth.
The fix is to keep in-memory map from blob URL to SecurityOrigin for the
unique origin case.
Test: fast/files/file-reader-file-url.html
* fileapi/Blob.cpp:
(WebCore::Blob::Blob):
* fileapi/BlobURL.cpp:
(WebCore::BlobURL::getOrigin): Return the origin string embeded in the blob URL.
(WebCore):
(WebCore::BlobURL::createBlobURL): Remove the check for null origin string since it is handled now.
* fileapi/BlobURL.h:
(BlobURL):
* fileapi/FileReaderLoader.cpp:
(WebCore::FileReaderLoader::start):
* fileapi/ThreadableBlobRegistry.cpp:
(WebCore):
(WebCore::originMap): Thread-specific in-memory map from the blob URL to the origin.
(WebCore::ThreadableBlobRegistry::registerBlobURL): Add the map from the blob URL to the origin.
(WebCore::ThreadableBlobRegistry::unregisterBlobURL): Remove the map for the unregistered blob URL.
(WebCore::ThreadableBlobRegistry::getCachedOrigin): Retrieve the origin associated with the blob URL.
* fileapi/ThreadableBlobRegistry.h:
(WebCore):
(ThreadableBlobRegistry):
* html/DOMURL.cpp:
(WebCore::DOMURL::createObjectURL):
* page/SecurityOrigin.cpp:
(WebCore::getCachedOrigin): Return the cached origin for the blob URL if it exists.
(WebCore):
(WebCore::SecurityOrigin::create): Call getCachedOrigin to get the cached origin first.
2012-06-15 Yoshifumi Inoue <yosin@chromium.org>
[Forms] Move search field related code to RenderSearchField from RenderTextControlSingleLine
https://bugs.webkit.org/show_bug.cgi?id=88980
Reviewed by Kent Tamura.
This patch moves search field related methods in RenderTextControlSingleLine
to new class RenderSearchField and changes related classes to use
RenderSearchField.
No new tests. This patch doesn't change behavior.
* CMakeLists.txt: Added new file RenderSearchField.cpp
* GNUmakefile.list.am: Added new file RenderSearchField.cpp and RenderSearchField.h
* Target.pri: ditto
* WebCore.gypi: ditto
* WebCore.vcproj/WebCore.vcproj: ditto
* WebCore.xcodeproj/project.pbxproj: ditto
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::subtreeHasChanged): Call RenderSearchField::updateCancelButtonVisibility
(WebCore::HTMLInputElement::addSearchResult): Changed to call InputType::addSearchResult.
* html/InputType.cpp:
(WebCore::InputType::addSearchResult): Added.
* html/InputType.h:
* html/SearchInputType.cpp:
(WebCore::SearchInputType::addSearchResult): Moved from HTMLInputElement.
(WebCore::SearchInputType::createRenderer): Added to create RenderSearchField.
* html/SearchInputType.h:
* html/shadow/TextControlInnerElements.cpp:
(WebCore::SearchFieldResultsButtonElement::defaultEventHandler): Use RenderSearchField instead of RenderTextControlSingleLine.
* loader/FormSubmission.cpp:
(WebCore::FormSubmission::create): Got rid isSearchField() guard for addSearchResult().
* rendering/RenderSearchField.cpp:
(WebCore::RenderSearchField::RenderSearchField): Renamed to RenderSearchField.
(WebCore::RenderSearchField::~RenderSearchField): ditto
(WebCore::RenderSearchField::resultsButtonElement): ditto
(WebCore::RenderSearchField::cancelButtonElement): ditto
(WebCore::RenderSearchField::addSearchResult): ditto
(WebCore::RenderSearchField::showPopup): ditto
(WebCore::RenderSearchField::hidePopup): ditto
(WebCore::RenderSearchField::computeControlHeight): Added.
(WebCore::RenderSearchField::updateFromElement): Moved search field related code from RenderTextControlSingleLine.
(WebCore::RenderSearchField::updateCancelButtonVisibility): Renamed to RenderSearchField.
(WebCore::RenderSearchField::visibilityForCancelButton): ditto
(WebCore::RenderSearchField::autosaveName): ditto
(WebCore::RenderSearchField::valueChanged): ditto
(WebCore::RenderSearchField::itemText): ditto
(WebCore::RenderSearchField::itemLabel): ditto
(WebCore::RenderSearchField::itemIcon): ditto
(WebCore::RenderSearchField::itemIsEnabled): ditto
(WebCore::RenderSearchField::itemStyle): ditto
(WebCore::RenderSearchField::menuStyle): ditto
(WebCore::RenderSearchField::clientInsetLeft): ditto
(WebCore::RenderSearchField::clientInsetRight): ditto
(WebCore::RenderSearchField::clientPaddingLeft): ditto
(WebCore::RenderSearchField::clientPaddingRight): ditto
(WebCore::RenderSearchField::listSize): ditto
(WebCore::RenderSearchField::selectedIndex): ditto
(WebCore::RenderSearchField::popupDidHide): ditto
(WebCore::RenderSearchField::itemIsSeparator): ditto
(WebCore::RenderSearchField::itemIsLabel): ditto
(WebCore::RenderSearchField::itemIsSelected): ditto
(WebCore::RenderSearchField::setTextFromItem): ditto
(WebCore::RenderSearchField::fontSelector): ditto
(WebCore::RenderSearchField::hostWindow): ditto
(WebCore::RenderSearchField::createScrollbar): ditto
(WebCore::RenderSearchField::computeHeightLimit): Added.
(WebCore::RenderSearchField::centerContainerIfNeeded): Added.
* rendering/RenderSearchField.h:
(WebCore::toRenderSearchField): Added.
* rendering/RenderTextControlSingleLine.cpp:
(WebCore::RenderTextControlSingleLine::RenderTextControlSingleLine): Remove search field related code.
(WebCore::RenderTextControlSingleLine::~RenderTextControlSingleLine): ditto.
(WebCore::RenderTextControlSingleLine::computeHeightLimit): Added.
(WebCore::RenderTextControlSingleLine::layout): Use new methods for removing isSearchField.
(WebCore::RenderTextControlSingleLine::computeControlHeight): Added.
(WebCore::RenderTextControlSingleLine::updateFromElement): Remove search field related code.
* rendering/RenderTextControlSingleLine.h:
(RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::centerContainerIfNeeded): Added.
(WebCore::RenderTextControlSingleLine::containerElement): Exposed for RenderSearchField.
(WebCore::RenderTextControlSingleLine::innerBlockElement): Exposed for RenderSearchField.
* rendering/RenderingAllInOne.cpp: Added RenderSearchField.cpp
2012-06-15 Hironori Bono <hbono@chromium.org>
Allow platforms to choose whether to remove markers on editing
https://bugs.webkit.org/show_bug.cgi?id=88838
Reviewed by Hajime Morita.
This change allows platforms to choose whether to remove markers on a word being
edited. WebKit does not remove markers when we move a selection to a markered
word on platforms that shouldEraseMarkersAfterChangeSelection returns false.
On such platforms, WebKit expects to set WTF_USE_MARKER_REMOVAL_UPON_EDITING to
1 so Editor::updateMarkersForWordsAffectedByEditing can remove markers. This
change also checks the return value of shouldEraseMarkersAfterChangeSelection so
platform can choose it. This change also adds grammar markers so it can also
remove grammar markers.
Test: editing/spelling/grammar-edit-word.html
* editing/Editor.cpp:
(WebCore::Editor::updateMarkersForWordsAffectedByEditing):
2012-06-15 Andrey Adaikin <aandrey@chromium.org>
Web Inspector: [WebGL] Simple implementation of the InjectedWebGLScriptSource to support capturing WebGL calls for a frame
https://bugs.webkit.org/show_bug.cgi?id=89088
Simple experimental implementation of the InjectedWebGLScriptSource.js that allows to wrap
a WebGL context and capture names (for now) of the WebGL function calls for a frame being captured.
Reviewed by Vsevolod Vlasov.
* inspector/InjectedWebGLScriptSource.js:
(.):
2012-06-15 Silvia Pfeiffer <silviapf@chromium.org>
Add fullscreen button to Chrome video controls for video.
https://bugs.webkit.org/show_bug.cgi?id=88818
Reviewed by Eric Carlson.
No new tests, final patch will contain the rebaselined tests.
The Chrome video controls are receiving a visual update.
This patch includes a fullscreen button for video elements and the rendering of the controls
in fullscreen including hiding them after 2 seconds when the mouse is out of the controls
and not moved.
* css/fullscreen.css:
(video:-webkit-full-screen, audio:-webkit-full-screen):
Add audio to the default fullscreen styling rules.
* css/mediaControlsChromium.css:
(video:-webkit-full-page-media::-webkit-media-controls-panel):
Align controls to the bottom of the fullscreen page.
(audio::-webkit-media-controls-fullscreen-button, video::-webkit-media-controls-fullscreen-button):
Include styling for the fullscreen button.
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore):
Add a constant for when to hide the controls in fullscreen.
(WebCore::MediaControlRootElementChromium::MediaControlRootElementChromium):
Add member fields for fullscreen button, hiding timer and tracking of whether we are in fullscreen.
(WebCore::MediaControlRootElementChromium::create):
Add fullscreen button element to the visual layout.
(WebCore::MediaControlRootElementChromium::setMediaController):
Add fullscreen button element to the media controller.
(WebCore::MediaControlRootElementChromium::reset):
Show the fullscreen button if the controller supports fullscreen.
(WebCore::MediaControlRootElementChromium::playbackStarted):
Start the timer to hide the controls in fullscreen.
(WebCore::MediaControlRootElementChromium::playbackStopped):
Don't hide the controls when the video is paused in fullscreen.
(WebCore::MediaControlRootElementChromium::reportedError):
Hide the fullscreen button when we hit an error.
(WebCore::MediaControlRootElementChromium::defaultEventHandler):
Add logic for mouse events in fullscreen to start/stop the hiding timer.
(WebCore::MediaControlRootElementChromium::startHideFullscreenControlsTimer):
Start the fullscreen hiding timer.
(WebCore::MediaControlRootElementChromium::hideFullscreenControlsTimerFired):
Timer fired: hide the video controls in fullscreen.
(WebCore::MediaControlRootElementChromium::stopHideFullscreenControlsTimer):
Reset the fullscreen hiding timer.
(WebCore::MediaControlRootElementChromium::enteredFullscreen):
Add logic to enter fullscreen.
(WebCore::MediaControlRootElementChromium::exitedFullscreen):
Add logic to exit fullscreen.
* html/shadow/MediaControlRootElementChromium.h:
(MediaControlRootElementChromium):
Add declaration of member functions and fields for fullscreen.
* rendering/RenderMediaControlsChromium.cpp:
(WebCore::paintMediaFullscreenButton):
Use the new image for the fullscreen button.
* rendering/RenderThemeChromiumMac.h:
(RenderThemeChromiumMac):
Declare the fullscreen painting function for Chrome Mac.
* rendering/RenderThemeChromiumMac.mm:
(WebCore::RenderThemeChromiumMac::paintMediaFullscreenButton):
Hook up the fullscreen paining function for Chrome Mac.
* rendering/RenderThemeChromiumSkia.cpp:
(WebCore::RenderThemeChromiumSkia::paintMediaFullscreenButton):
Hook up the fullscreen paining function for Chrome Skia.
* rendering/RenderThemeChromiumSkia.h:
(RenderThemeChromiumSkia):
Declare the fullscreen painting function for Chrome Skia.
2012-06-14 Kent Tamura <tkent@chromium.org>
Unreviewed, rolling out r110340.
http://trac.webkit.org/changeset/110340
https://bugs.webkit.org/show_bug.cgi?id=88749
r110340 added a broken implementation of
HTMLSelectElement::selectedOptions. This roll out removes
HTMLSelectElement::selectedOptions.
* html/CollectionType.h:
* html/HTMLCollection.cpp:
(WebCore::HTMLCollection::shouldIncludeChildren):
(WebCore::HTMLCollection::isAcceptableElement):
* html/HTMLSelectElement.cpp:
* html/HTMLSelectElement.h:
(HTMLSelectElement):
* html/HTMLSelectElement.idl:
2012-06-14 Keishi Hattori <keishi@webkit.org>
Add color property to input type=range
https://bugs.webkit.org/show_bug.cgi?id=89067
Reviewed by Kent Tamura.
Adding color property to input type=range in preparation for
supporting tick marks for datalist for input type=range.
* css/html.css:
(input[type="range"]): Setting color property for input type=range.
This will set the color of tick mark when we implement <datalist>.
* css/themeChromiumLinux.css:
(input[type=range]):
* css/themeWin.css:
(input[type="range"]):
2012-06-14 Kent Tamura <tkent@chromium.org>
Support file extensions in HTMLInputElement::accept
https://bugs.webkit.org/show_bug.cgi?id=88298
Reviewed by Hajime Morita.
The WHATWG specification added file extensions supoprt for
HTMLInputElement::accept recently.
In the WebCore FileChooser interface, we rejected invalid MIME type
strings in HTMLInputElement::acceptMIMETypes(), and platform file
choosers don't expect that it containts file extensions. So, this patch
introduces additional member "acceptFileExtensions" to
FileChooserSettings, and it contains only file extensions specified by
an accept attribute.
* html/FileInputType.cpp:
Sets HTMLInputElement::acceptFileExtensions() to
FileChooserSettings::acceptFileExtensions.
(WebCore::FileInputType::handleDOMActivateEvent):
(WebCore::FileInputType::receiveDropForDirectoryUpload):
* html/HTMLInputElement.cpp:
(WebCore::isValidFileExtension): Added.
(WebCore::parseAcceptAttribute):
Common part for acceptMIMETyps() and acceptFileExtensions().
(WebCore::HTMLInputElement::acceptMIMETypes):
Uses parseAcceptAttribute() with isValidMIMEType().
(WebCore::HTMLInputElement::acceptFileExtensions):
Added. Uses parseAcceptAttribute() with isValidFileExtension().
* html/HTMLInputElement.h:
(HTMLInputElement): Add acceptFileExtensions().
* platform/FileChooser.h:
(FileChooserSettings): Add acceptFileExtensions and acceptTypes().
* platform/FileChooser.cpp:
(WebCore::FileChooserSettings::acceptTypes): Added.
2012-06-14 James Robinson <jamesr@chromium.org>
[chromium] Remove unused CanvasLayerChromium.h/cpp from the tree
https://bugs.webkit.org/show_bug.cgi?id=89152
Reviewed by Adrienne Walker.
These files have been unused and not in the build system since r119769
* platform/graphics/chromium/CanvasLayerChromium.cpp: Removed.
* platform/graphics/chromium/CanvasLayerChromium.h: Removed.
2012-06-14 Emil A Eklund <eae@chromium.org>
Cast paddings to int in RenderTableCell
https://bugs.webkit.org/show_bug.cgi?id=88918
Reviewed by Levi Weintraub.
Table layout uses integers throughout yet the TableCell paddingLeft/
Right/Top/Bottom methods returns LayoutUnits. This causes inconsistent
rounding as some call sites cast the numbers to ints before doing
computation and others do computation before casting.
By changing the methods to always cast the padding values to int we
ensure consistent padding calculations.
Ideally we'd change the type of the return value for the methods but as
they are overriden that would likely cause more confusion.
Test: fast/sub-pixel/table-cells-with-padding-do-not-wrap.html
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::paddingTop):
(WebCore::RenderTableCell::paddingBottom):
(WebCore::RenderTableCell::paddingLeft):
(WebCore::RenderTableCell::paddingRight):
(WebCore::RenderTableCell::paddingBefore):
(WebCore::RenderTableCell::paddingAfter):
2012-06-14 Kenichi Ishibashi <bashi@chromium.org>
[Chromium] Check the result of FontCache::getCachedFontPlatformData()
https://bugs.webkit.org/show_bug.cgi?id=89141
Reviewed by Kent Tamura.
The value of FontCache::getCachedFontPlatformData() could be invalid,
so we should check the value before use it.
No new tests. No new functionality.
* platform/graphics/skia/FontCacheSkia.cpp:
(WebCore::FontCache::getFontDataForCharacters):
2012-06-14 Gregg Tavares <gman@google.com>
Fix framebuffer completeness test
https://bugs.webkit.org/show_bug.cgi?id=89127
Reviewed by Kenneth Russell.
No new tests just fixing failing tests.
* html/canvas/WebGLFramebuffer.cpp:
(WebCore::WebGLFramebuffer::initializeAttachments):
* html/canvas/WebGLRenderingContext.cpp:
(WebCore):
(WebCore::WebGLRenderingContext::isTexInternalFormatColorBufferCombinationValid):
* platform/graphics/GraphicsContext3D.cpp:
(WebCore::GraphicsContext3D::getClearBitsByAttachmentType):
(WebCore):
(WebCore::GraphicsContext3D::getClearBitsByFormat):
(WebCore::GraphicsContext3D::getChannelBitsByFormat):
* platform/graphics/GraphicsContext3D.h:
2012-06-14 Adrienne Walker <enne@google.com>
[chromium] Make TiledLayerChromium robust to unexpected null tiles
https://bugs.webkit.org/show_bug.cgi?id=89143
Reviewed by James Robinson.
Although there shouldn't ever be null tiles in the map, it appears to
be occurring frequently enough to show up in crash reports. In the
short term, be robust to this so the tiler doesn't crash. This should
eventually be reverted.
* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::TiledLayerChromium::pushPropertiesTo):
(WebCore::TiledLayerChromium::setLayerTreeHost):
(WebCore::TiledLayerChromium::invalidateRect):
(WebCore::TiledLayerChromium::updateTiles):
(WebCore::TiledLayerChromium::resetUpdateState):
2012-06-14 Kent Tamura <tkent@chromium.org>
Validate form state strings in FormController::setStateForNewFormElements()
https://bugs.webkit.org/show_bug.cgi?id=88768
Reviewed by Hajime Morita.
Reject invalid form state vectors. This state vectors are generated by
WebKit itself. However it can be invalid because
- Serialized state vectors can be corrupted
- A future version of WebKit might change the format
So we had better reject unexpected state vectors as possible.
Test: fast/forms/state-restore-broken-state.html
* WebCore.exp.in: Expose some symbols used by Internals.cpp.
* html/FormController.cpp:
(WebCore::isNotFormControlTypeCharacter): A helper for state validation.
(WebCore::FormController::setStateForNewFormElements):
Reject state vectors of which size is not a multiple of 3, or a type name is invalid.
* testing/Internals.cpp:
(WebCore::Internals::formControlStateOfPreviousHistoryItem):
Returns a form state vector of the previous document.
(WebCore::Internals::setFormControlStateOfPreviousHistoryItem):
Sets a form state vector for the previous document.
* testing/Internals.h: Declare new functions.
* testing/Internals.idl: ditto.
2012-06-14 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r120393.
http://trac.webkit.org/changeset/120393
https://bugs.webkit.org/show_bug.cgi?id=89163
breaks cr-mac build (Requested by morrita on #webkit).
* WebCore.gyp/WebCore.gyp:
* platform/image-decoders/ImageDecoder.h:
(ImageFrame):
(ImageDecoder):
* platform/image-decoders/jpeg/JPEGImageDecoder.cpp:
(turboSwizzled):
(WebCore::JPEGImageReader::JPEGImageReader):
(WebCore::JPEGImageReader::close):
(WebCore::JPEGImageReader::decode):
(JPEGImageReader):
(WebCore::JPEGImageDecoder::outputScanlines):
* platform/image-decoders/png/PNGImageDecoder.cpp:
(WebCore::PNGImageReader::PNGImageReader):
(WebCore::PNGImageReader::close):
(WebCore::PNGImageReader::currentBufferSize):
(PNGImageReader):
(WebCore::PNGImageReader::decodingSizeOnly):
(WebCore::PNGImageReader::interlaceBuffer):
(WebCore::PNGImageReader::hasAlpha):
(WebCore::PNGImageReader::setHasAlpha):
(WebCore::PNGImageDecoder::headerAvailable):
(WebCore::PNGImageDecoder::rowAvailable):
* platform/image-decoders/skia/ImageDecoderSkia.cpp:
(WebCore):
(WebCore::resolveColorSpace):
(WebCore::createColorSpace):
(WebCore::ImageFrame::setColorProfile):
(WebCore::ImageFrame::setStatus):
2012-06-14 Julien Chaffraix <jchaffraix@webkit.org>
RenderLayer subtrees without any self-painting layer shouldn't be walked during painting
https://bugs.webkit.org/show_bug.cgi?id=88888
Reviewed by Simon Fraser.
Performance optimization, covered by existing tests.
The gist of this change is to add a has-self-painting-layer-descendant flag (including an
invalidation logic) that is used to avoid walking subtrees without any self-painting layer.
On http://dglazkov.github.com/performance-tests/biggrid.html with a 100,000 rows
by 100 columns table, it brings the paint time during scrolling from ~45ms to ~6ms
on my machine. The test case is a pathologic example here but the optimization should
apply in other cases.
The new update logic piggy-backs on top of the existing updateVisibilityStatus() one that
got repurposed and renamed as part of this change.
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::RenderLayer):
(WebCore::RenderLayer::addChild):
(WebCore::RenderLayer::removeChild):
(WebCore::RenderLayer::styleChanged):
These functions were updated to dirty / set the new flag.
(WebCore::RenderLayer::dirtyAncestorChainHasSelfPaintingLayerDescendantStatus):
(WebCore::RenderLayer::setAncestorChainHasSelfPaintingLayerDescendant):
Added those functions to handle setting / invalidating the new flag.
(WebCore::RenderLayer::updateSelfPaintingLayerAfterStyleChange):
Added this function to handle style update.
(WebCore::RenderLayer::paintLayer):
(WebCore::RenderLayer::paintLayerContentsAndReflection):
(WebCore::RenderLayer::paintLayerContents):
(WebCore::RenderLayer::paintList):
Changed this logic to bail out if we have no self-painting descendants. This is what
is giving the performance improvement. Also added some performance ASSERTs to ensure
the methods are not called when they shouldn't.
(WebCore::RenderLayer::updateDescendantDependentFlags):
Renamed from updateVisibilityStatus to account for the new usage.
(WebCore::RenderLayer::updateLayerPositions):
(WebCore::RenderLayer::updateLayerPositionsAfterScroll):
(WebCore::RenderLayer::collectLayers):
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateGraphicsLayerGeometry):
Updated after updateVisibilityStatus rename.
* rendering/RenderLayer.h:
(WebCore::RenderLayer::hasSelfPaintingLayerDescendant):
Added the declaration of the new functions as well as the new flag and dirty bit.
2012-06-14 Tony Payne <tpayne@chromium.org>
[chromium] Add iccjpeg and qcms to chromium port
https://bugs.webkit.org/show_bug.cgi?id=81974
Reviewed by Adam Barth.
Covered by existing layout tests which will be rebaselined.
* WebCore.gyp/WebCore.gyp: Add qcms to the build.
* platform/image-decoders/ImageDecoder.h:
(WebCore::ImageDecoder::qcmsOutputDeviceProfile): Return an sRGB profile.
On OSX, return the default RGB profile. Add FIXME to use the user's
monitor profile and verify that profile for other platforms.
* platform/image-decoders/jpeg/JPEGImageDecoder.cpp:
(turboSwizzled): For libjpeg-turbo, JCS_EXT_BGRA and JCS_EXT_RGBA are the
two known output color spaces for which the decoder uses a data swizzle.
(colorSpaceHasAlpha): JPEG's have no alpha in the output color space. For
libjpeg-turbo, alpha may be present in the swizzled output color space.
(WebCore::JPEGImageReader::JPEGImageReader):
(WebCore::JPEGImageReader::close):
(WebCore::JPEGImageReader::decode): For QCMSLIB, create the color transform
to use during decoding, and ensure we switch to inputing RGBA data to qcms
even if the desired output data is BGRA: outputScanlines() sends BGRA data
to the frame buffer following color correction if needed.
(JPEGImageReader):
(WebCore::JPEGImageReader::colorTransform): qcms color transform getter.
(WebCore::JPEGImageReader::createColorTransform): Create color transform.
Release the existing transform (if any) and assign to the color transform
created from the color profile data.
(WebCore::JPEGImageDecoder::outputScanlines): Minor style fix. Apply color
transform to each decoded image row.
* platform/image-decoders/png/PNGImageDecoder.cpp:
(WebCore::PNGImageReader::PNGImageReader):
(WebCore::PNGImageReader::close):
(WebCore::PNGImageReader::currentBufferSize): Move this adjacent to other
setters and getters.
(WebCore::PNGImageReader::decodingSizeOnly): Ditto.
(WebCore::PNGImageReader::setHasAlpha): Ditto.
(WebCore::PNGImageReader::hasAlpha): Ditto.
(WebCore::PNGImageReader::interlaceBuffer): Ditto.
(WebCore::PNGImageReader::createRowBuffer): Creates a temporary row buffer,
used when a color transform is applied to the decoded image pixels.
(WebCore::PNGImageReader::rowBuffer): Return the temporary row buffer.
(WebCore::PNGImageReader::colorTransform): qcms color transform getter.
(WebCore::PNGImageReader::createColorTransform): Create color transform.
Release the existing transform (if any) and assign to the color transform
created from the color profile data.
(WebCore::PNGImageDecoder::headerAvailable): For QCMSLIB, create the color
transform to use for decoding. Clear m_colorProfile (not used anymore).
(WebCore::PNGImageDecoder::rowAvailable): Create temporary row buffer if
a color transform is needed for decoding. Apply color transform to each
decoded image row.
* platform/image-decoders/skia/ImageDecoderSkia.cpp:
(WebCore::ImageFrame::setColorProfile): Old method of colorProfiles is no
longer used. Add a FIXME to remove the old implementation.
(WebCore::ImageFrame::setStatus): Remove old color correction code.
2012-06-14 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r120384.
http://trac.webkit.org/changeset/120384
https://bugs.webkit.org/show_bug.cgi?id=89157
breaks create-blob-url-from-data-url.html (Requested by
morrita on #webkit).
* fileapi/Blob.cpp:
(WebCore::Blob::Blob):
* fileapi/BlobURL.cpp:
(WebCore::BlobURL::createBlobURL):
* fileapi/BlobURL.h:
(BlobURL):
* fileapi/FileReaderLoader.cpp:
(WebCore::FileReaderLoader::start):
* fileapi/ThreadableBlobRegistry.cpp:
(WebCore):
(WebCore::ThreadableBlobRegistry::registerBlobURL):
(WebCore::ThreadableBlobRegistry::unregisterBlobURL):
* fileapi/ThreadableBlobRegistry.h:
(WebCore):
(ThreadableBlobRegistry):
* html/DOMURL.cpp:
(WebCore::DOMURL::createObjectURL):
* page/SecurityOrigin.cpp:
(WebCore::SecurityOrigin::create):
2012-06-14 Yoshifumi Inoue <yosin@chromium.org>
[Forms] Copy RenderTextControlSingleLine.{cpp,h} to RenderSearchFiled.{cpp,h}
https://bugs.webkit.org/show_bug.cgi?id=89155
Reviewed by Kent Tamura.
This patch copies RenderTextControlSingleLine.cpp and .h into RenderSearchField.cpp
and .h with just "cp" code. New files aren't compiled until bug 88980.
No new tests. This patch doesn't change behavior.
* rendering/RenderSearchField.cpp: Added.
(WebCore::RenderTextControlInnerBlock::positionForPoint):
(WebCore::RenderTextControlSingleLine::RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::~RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::containerElement):
(WebCore::RenderTextControlSingleLine::innerBlockElement):
(WebCore::RenderTextControlSingleLine::innerSpinButtonElement):
(WebCore::RenderTextControlSingleLine::resultsButtonElement):
(WebCore::RenderTextControlSingleLine::cancelButtonElement):
(WebCore::RenderTextControlSingleLine::textBaseStyle):
(WebCore::RenderTextControlSingleLine::addSearchResult):
(WebCore::RenderTextControlSingleLine::showPopup):
(WebCore::RenderTextControlSingleLine::hidePopup):
(WebCore::RenderTextControlSingleLine::paint):
(WebCore::RenderTextControlSingleLine::layout):
(WebCore::RenderTextControlSingleLine::nodeAtPoint):
(WebCore::RenderTextControlSingleLine::styleDidChange):
(WebCore::RenderTextControlSingleLine::capsLockStateMayHaveChanged):
(WebCore::RenderTextControlSingleLine::hasControlClip):
(WebCore::RenderTextControlSingleLine::controlClipRect):
(WebCore::RenderTextControlSingleLine::getAvgCharWidth):
(WebCore::RenderTextControlSingleLine::preferredContentWidth):
(WebCore::RenderTextControlSingleLine::computeControlHeight):
(WebCore::RenderTextControlSingleLine::updateFromElement):
(WebCore::RenderTextControlSingleLine::createInnerTextStyle):
(WebCore::RenderTextControlSingleLine::createInnerBlockStyle):
(WebCore::RenderTextControlSingleLine::updateCancelButtonVisibility):
(WebCore::RenderTextControlSingleLine::visibilityForCancelButton):
(WebCore::RenderTextControlSingleLine::textShouldBeTruncated):
(WebCore::RenderTextControlSingleLine::autosaveName):
(WebCore::RenderTextControlSingleLine::valueChanged):
(WebCore::RenderTextControlSingleLine::itemText):
(WebCore::RenderTextControlSingleLine::itemLabel):
(WebCore::RenderTextControlSingleLine::itemIcon):
(WebCore::RenderTextControlSingleLine::itemIsEnabled):
(WebCore::RenderTextControlSingleLine::itemStyle):
(WebCore::RenderTextControlSingleLine::menuStyle):
(WebCore::RenderTextControlSingleLine::clientInsetLeft):
(WebCore::RenderTextControlSingleLine::clientInsetRight):
(WebCore::RenderTextControlSingleLine::clientPaddingLeft):
(WebCore::RenderTextControlSingleLine::clientPaddingRight):
(WebCore::RenderTextControlSingleLine::listSize):
(WebCore::RenderTextControlSingleLine::selectedIndex):
(WebCore::RenderTextControlSingleLine::popupDidHide):
(WebCore::RenderTextControlSingleLine::itemIsSeparator):
(WebCore::RenderTextControlSingleLine::itemIsLabel):
(WebCore::RenderTextControlSingleLine::itemIsSelected):
(WebCore::RenderTextControlSingleLine::setTextFromItem):
(WebCore::RenderTextControlSingleLine::fontSelector):
(WebCore::RenderTextControlSingleLine::hostWindow):
(WebCore::RenderTextControlSingleLine::autoscroll):
(WebCore::RenderTextControlSingleLine::scrollWidth):
(WebCore::RenderTextControlSingleLine::scrollHeight):
(WebCore::RenderTextControlSingleLine::scrollLeft):
(WebCore::RenderTextControlSingleLine::scrollTop):
(WebCore::RenderTextControlSingleLine::setScrollLeft):
(WebCore::RenderTextControlSingleLine::setScrollTop):
(WebCore::RenderTextControlSingleLine::scroll):
(WebCore::RenderTextControlSingleLine::logicalScroll):
(WebCore::RenderTextControlSingleLine::createScrollbar):
(WebCore::RenderTextControlSingleLine::inputElement):
* rendering/RenderSearchField.h: Added.
(RenderTextControlSingleLine):
(WebCore::RenderTextControlSingleLine::popupIsVisible):
(WebCore::RenderTextControlSingleLine::isTextField):
(WebCore::toRenderTextControlSingleLine):
(RenderTextControlInnerBlock):
(WebCore::RenderTextControlInnerBlock::RenderTextControlInnerBlock):
(WebCore::RenderTextControlInnerBlock::hasLineIfEmpty):
2012-06-14 Kent Tamura <tkent@chromium.org>
[JSC/V8] "DOMString[]" for function return values should not be null
https://bugs.webkit.org/show_bug.cgi?id=89151
Reviewed by Kentaro Hara.
jsArray and v8Array are used for "DOMString[]", not "DOMString[]?". So
we should not return JavaScript null object.
No new tests. We have no plan to use this code in production code for
now. It will be used for window.internals functions.
* bindings/js/JSDOMBinding.cpp:
(WebCore::jsArray): For null input, returns an empty array, not null.
* bindings/v8/V8Binding.cpp:
(WebCore::v8Array): Implement DOMStringList -> JavaScript array conversion.
Like jsArray(), we don't return null object.
* bindings/v8/V8Binding.h:
(WebCore): Declare v8Array().
2012-06-14 Dale Curtis <dalecurtis@chromium.org>
Disable click/dbl-click on MediaDocument for Chromium. Fix layout test for all platforms.
https://bugs.webkit.org/show_bug.cgi?id=89129
As titled, click-to-pause and double-click-to-play behavior are unwanted on Chromium. A
long standing issue (~3 years) has been to disable this feature.
Reviewed by Eric Carlson.
Uses existing click/double-click test.
* html/MediaDocument.cpp:
(WebCore::MediaDocument::defaultEventHandler):
2012-06-14 Jian Li <jianli@chromium.org>
FileReader is dysfunctional in documents with "null" origin string
https://bugs.webkit.org/show_bug.cgi?id=78648
Reviewed by Adam Barth.
The fix is to keep in-memory map from blob URL to SecurityOrigin for the
unique origin case.
Test: fast/files/file-reader-file-url.html
* fileapi/Blob.cpp:
(WebCore::Blob::Blob):
* fileapi/BlobURL.cpp:
(WebCore::BlobURL::getOrigin): Return the origin string embeded in the blob URL.
(WebCore):
(WebCore::BlobURL::createBlobURL): Remove the check for null origin string since it is handled now.
* fileapi/BlobURL.h:
(BlobURL):
* fileapi/FileReaderLoader.cpp:
(WebCore::FileReaderLoader::start):
* fileapi/ThreadableBlobRegistry.cpp:
(WebCore):
(WebCore::originMap): Thread-specific in-memory map from the blob URL to the origin.
(WebCore::ThreadableBlobRegistry::registerBlobURL): Add the map from the blob URL to the origin.
(WebCore::ThreadableBlobRegistry::unregisterBlobURL): Remove the map for the unregistered blob URL.
(WebCore::ThreadableBlobRegistry::cachedOrigin): Retrieve the origin associated with the blob URL.
* fileapi/ThreadableBlobRegistry.h:
(WebCore):
(ThreadableBlobRegistry):
* html/DOMURL.cpp:
(WebCore::DOMURL::createObjectURL):
* page/SecurityOrigin.cpp:
(WebCore::cachedOrigin): Return the cached origin for the blob URL if it exists.
(WebCore):
(WebCore::SecurityOrigin::create): Call cachedOrigin to get the cached origin first.
2012-06-14 Sudarsana Nagineni <sudarsana.nagineni@linux.intel.com>
[CMAKE] Move JSDeprecatedPeerConnectionCustom.cpp from CMakeLists.txt to UseJSC.cmake
https://bugs.webkit.org/show_bug.cgi?id=89015
Reviewed by Antonio Gomes.
Include JSDeprecatedPeerConnectionCustom.cpp in UseJSC.cmake
instead of CMakeLists.txt.
No change in functionality so no new tests.
* CMakeLists.txt:
* UseJSC.cmake:
2012-06-14 Xianzhu Wang <wangxianzhu@chromium.org>
[Chromium] Add setAutoHint() and setUseBitmaps() in WebFontRendering
https://bugs.webkit.org/show_bug.cgi?id=89014
Reviewed by James Robinson.
No new tests because of no change of functionality.
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.cpp:
(WebCore):
(WebCore::FontPlatformData::setAutoHint):
(WebCore::FontPlatformData::setUseBitmaps):
(WebCore::FontPlatformData::setAntiAlias):
(WebCore::FontPlatformData::setupPaint):
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.h:
(FontPlatformData):
2012-06-14 Ryuan Choi <ryuan.choi@samsung.com>
[EFL] Move cursor related code from WidgetEfl to ewk_view and EflScreenUtilities.
https://bugs.webkit.org/show_bug.cgi?id=88803
Reviewed by Chang Shu.
Because WebProcess can not control cursor,
setCursor should be passed to ChromeClient like other ports.
* platform/Widget.h:
(Widget):
* platform/efl/EflScreenUtilities.cpp:
(WebCore::applyFallbackCursor): Moved from WidgetEfl.cpp.
(WebCore):
(WebCore::isUsingEcoreX): Moved from WidgetEfl.cpp.
* platform/efl/EflScreenUtilities.h:
(WebCore):
* platform/efl/WidgetEfl.cpp: Removes cursor related codes.
(WebCore::WidgetPrivate::WidgetPrivate):
(WebCore::Widget::~Widget):
(WebCore::Widget::setCursor): Redirect to HostWindow.
(WebCore::Widget::setEvasObject):
2012-06-14 Ryosuke Niwa <rniwa@webkit.org>
Radio node lists do not get updated when inserted back into a document
https://bugs.webkit.org/show_bug.cgi?id=88825
Reviewed by Darin Adler.
The bug was caused by RadioNodeList's m_node not being adjusted back to a document after
it was removed from the document and inserted back. While we could be doing this adjustment again
when the HTMLFormElement is inserted back into the document, this whole approach is fragile
because adds a maintenance burden on root elements of RadioNodeList.
Change the approach to the problem instead and detect when m_node is detached from the document.
Now RadioNodeList and LabelsNodeList will always use its root element as m_node and
DynamicSubtreeNodeList's node() will then determine whether it should use its document or not based
on the flag (rootedAtDocument) passed by the constructor of the node list and m_node->inDocument().
Also invalidate node lists rooted at the document level regardless where the DOM mutation took place.
* dom/ClassNodeList.cpp:
(WebCore::ClassNodeList::~ClassNodeList):
* dom/Document.cpp:
(WebCore::Document::registerDynamicSubtreeNodeList): Moved from Node.
(WebCore::Document::unregisterDynamicSubtreeNodeList): Ditto.
* dom/Document.h:
(Document):
* dom/DynamicNodeList.cpp:
(WebCore::DynamicSubtreeNodeList::DynamicSubtreeNodeList): Takes DynamicNodeListRootType which tell us
whether node() should return m_node's document by default or not.
(WebCore::DynamicSubtreeNodeList::length): Refactor code to avoid node() multiple times.
(WebCore::DynamicSubtreeNodeList::itemForwardsFromCurrent): Ditto.
(WebCore::DynamicSubtreeNodeList::itemBackwardsFromCurrent): Ditto.
(WebCore::DynamicNodeList::itemWithName): Ditto.
* dom/DynamicNodeList.h:
(WebCore::DynamicNodeList::DynamicNodeList):
(WebCore::DynamicNodeList::node): Returns m_node->document() if m_node is still in the document and
the node list was rooted at document.
(WebCore::DynamicNodeList::Caches::Caches):
(Caches):
(DynamicNodeList):
(DynamicSubtreeNodeList):
* dom/NameNodeList.cpp:
(WebCore::NameNodeList::~NameNodeList):
* dom/Node.cpp:
(WebCore):
(WebCore::Node::invalidateNodeListsCacheAfterAttributeChanged): Invalidate m_listsInvalidatedAtDocument
at document instead of trying to invalidate it on every ancestor we traverse.
(WebCore::Node::invalidateNodeListsCacheAfterChildrenChanged): Ditto.
(WebCore::NodeListsNodeData::invalidateCachesForDocument): Extracted from
invalidateCachesThatDependOnAttributes.
(WebCore::NodeListsNodeData::invalidateCachesThatDependOnAttributes):
(WebCore::Node::radioNodeList):
* dom/Node.h:
(Node):
* dom/NodeRareData.h:
(NodeListsNodeData):
* dom/TagNodeList.cpp:
(WebCore::TagNodeList::~TagNodeList):
* html/HTMLFormElement.cpp:
(WebCore::HTMLFormElement::removedFrom):
* html/LabelsNodeList.cpp:
(WebCore::LabelsNodeList::LabelsNodeList):
* html/RadioNodeList.cpp:
(WebCore::RadioNodeList::RadioNodeList): Takes Element* first since that's the order we use elsewhere.
* html/RadioNodeList.h:
(WebCore::RadioNodeList::create):
(RadioNodeList):
2012-06-14 Tony Chang <tony@chromium.org>
Fix bit packing of StyleRareInheritedData on Windows
https://bugs.webkit.org/show_bug.cgi?id=89112
Reviewed by Ryosuke Niwa.
On my machine (64 bit Chromium Linux), StyleRareInheritedData is 168 bytes.
No new tests, added a compile assert to ensure correctness going forward.
* rendering/style/StyleRareInheritedData.cpp:
(SameSizeAsStyleRareInheritedData):
(WebCore):
* rendering/style/StyleRareInheritedData.h:
(StyleRareInheritedData):
2012-06-14 Andreas Kling <kling@webkit.org>
Crashes below IconDatabase::performPendingRetainAndReleaseOperations().
<http://webkit.org/b/88846>
<rdar://problem/11629106>
Reviewed by Brady Eidson.
Put isolatedCopy() strings in the retain/release operation queues to make sure it's safe
for secondary threads to ref/deref them in performPendingRetainAndReleaseOperations().
Also added assertions as appropriate.
* loader/icon/IconDatabase.cpp:
(WebCore::IconDatabase::retainIconForPageURL):
(WebCore::IconDatabase::releaseIconForPageURL):
(WebCore::IconDatabase::performPendingRetainAndReleaseOperations):
2012-06-14 Shawn Singh <shawnsingh@chromium.org>
[chromium] For hit testing in CCLayerTreeHostCommon, need to check that the transform is invertible before inverting it.
https://bugs.webkit.org/show_bug.cgi?id=89049
Reviewed by Adrienne Walker.
Unit test added to CCLayerTreeHostCommonTest.cpp:
CCLayerTreeHostCommonTest.verifyHitTestingForUninvertibleTransform
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::pointHitsRect):
2012-06-14 Maciej Stachowiak <mjs@apple.com>
The whole world rebuilds when you touch any IDL file
https://bugs.webkit.org/show_bug.cgi?id=89125
Reviewed by Sam Weinig.
No new tests; this only changes the build system and its dependencies.
* bindings/scripts/preprocess-idls.pl: Optionally generate the
dependencies for supplemental IDLs in Makefile format.
* DerivedSources.make: Use the new feature of preprocess-idls.pl to generate
the correct IDL dependencies created by Supplemental, then include it. This way,
we avoid the situation where every bindings file implicitly depends on every single
IDL file in the project.
2012-06-14 Ian Vollick <vollick@chromium.org>
[chromium] Certain settings in CCSettings could be global
https://bugs.webkit.org/show_bug.cgi?id=88384
Reviewed by James Robinson.
No new tests. No new functionality.
* WebCore.gypi:
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(Settings):
* platform/graphics/chromium/ContentLayerChromium.cpp:
(WebCore::ContentLayerChromium::createTextureUpdaterIfNeeded):
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::addAnimation):
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::initialize):
* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::TiledLayerChromium::updateTileSizeAndTilingOption):
* platform/graphics/chromium/cc/CCDebugRectHistory.cpp:
(WebCore::CCDebugRectHistory::enabled):
(WebCore::CCDebugRectHistory::saveDebugRectsForCurrentFrame):
* platform/graphics/chromium/cc/CCDebugRectHistory.h:
(WebCore):
(CCDebugRectHistory):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
(WebCore::CCHeadsUpDisplay::enabled):
(WebCore::CCHeadsUpDisplay::showPlatformLayerTree):
(WebCore::CCHeadsUpDisplay::showDebugRects):
(WebCore::CCHeadsUpDisplay::draw):
(WebCore::CCHeadsUpDisplay::drawHudContents):
(WebCore::CCHeadsUpDisplay::drawDebugRects):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.h:
(WebCore):
(CCHeadsUpDisplay):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::create):
(WebCore::CCLayerTreeHost::CCLayerTreeHost):
(WebCore::CCLayerTreeHost::animateLayers):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCLayerTreeSettings::CCLayerTreeSettings):
(CCLayerTreeSettings):
(CCLayerTreeHost):
(WebCore::CCLayerTreeHost::settings):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::create):
(WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::calculateRenderSurfaceLayerList):
(WebCore::CCLayerTreeHostImpl::animateLayers):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCRenderer.h:
(CCRendererClient):
(WebCore::CCRenderer::settings):
* platform/graphics/chromium/cc/CCSettings.cpp: Added.
(WebCore):
(WebCore::CCSettings::perTilePaintingEnabled):
(WebCore::CCSettings::setPerTilePaintingEnabled):
(WebCore::CCSettings::partialSwapEnabled):
(WebCore::CCSettings::setPartialSwapEnabled):
(WebCore::CCSettings::acceleratedAnimationEnabled):
(WebCore::CCSettings::setAcceleratedAnimationEnabled):
(WebCore::CCSettings::reset):
* platform/graphics/chromium/cc/CCSettings.h: Added.
(WebCore):
(CCSettings):
* testing/InternalSettings.cpp:
* testing/InternalSettings.h:
(InternalSettings):
2012-06-14 Alejandro G. Castro <alex@igalia.com>
[GTK] Add TextureMapper ImageBuffer support as a fallback from the hardware accelerated path
https://bugs.webkit.org/show_bug.cgi?id=73634
Remove the TextureMapperCairo classes, now we use
TextureMapperImageBuffer to implement the software fallback. Add
implementation for the 3D transformations in the 2D cairo graphics
context.
Reviewed by Martin Robinson.
This patch adds a new implementation of already tested cases.
* GNUmakefile.list.am:
* platform/graphics/cairo/GraphicsContext3DPrivate.cpp:
(WebCore):
(WebCore::GraphicsContext3DPrivate::paintToTextureMapper):
Required for compilation when WebGL is enabled.
* platform/graphics/cairo/GraphicsContextCairo.cpp:
(WebCore::GraphicsContext::get3DTransform):
(WebCore::GraphicsContext::concat3DTransform):
(WebCore::GraphicsContext::set3DTransform):
* platform/graphics/cairo/TextureMapperCairo.cpp: Removed.
* platform/graphics/cairo/TextureMapperCairo.h: Removed.
* platform/graphics/texmap/TextureMapperImageBuffer.cpp:
(WebCore::BitmapTextureImageBuffer::updateContents):
2012-06-14 Jia Pu <jpu@apple.com>
Mark text with text alternative with blue underline.
https://bugs.webkit.org/show_bug.cgi?id=83047
Reviewed by Enrica Casucci.
Tests: platform/mac/editing/input/edit-dictated-text-with-alternative.html
platform/mac/editing/input/insert-dictated-text.html
This patch implements visual indication on dictated text with alternatives, and provides UI
to show alternative text on OS X. Majority of the changes is for generalizing existing AlternativeTextController
class to handle dictation alternatives. The two new classes, AlternativeTextUIController and
TextAlternativeWithRange, are used by both WebKit and WK2. So WebCore seems to be the natural place
for them.
* WebCore.exp.in:
* WebCore.xcodeproj/project.pbxproj:
* editing/AlternativeTextController.cpp: Expanded exising class interface to support dictation alternatives.
(DictationAlternativeDetails): Marker detail class for dictation alternative mark.
(WebCore::DictationAlternativeDetails::create):
(WebCore::DictationAlternativeDetails::dictationContext):
(WebCore::DictationAlternativeDetails::DictationAlternativeDetails):
(WebCore::markerTypesForAppliedDictationAlternative):
(WebCore::AlternativeTextController::applyAlternativeTextToRange): Generalized existing applyAlternativeTextToRange() to handle dictation alternatives.
(WebCore::AlternativeTextController::timerFired): Expanded existing code to handle dictation alternatives.
(WebCore::AlternativeTextController::handleAlternativeTextUIResult): Expanded existing code to handle dictation alternatives.
(WebCore::AlternativeTextController::respondToChangedSelection): Moved part of the function into respondToMarkerAtEndOfWord() to improve readability.
(WebCore::AlternativeTextController::shouldStartTimerFor):
(WebCore::AlternativeTextController::respondToMarkerAtEndOfWord):
(WebCore::AlternativeTextController::markerDescriptionForAppliedAlternativeText):
(WebCore::AlternativeTextController::removeDictationAlternativesForMarker):
(WebCore::AlternativeTextController::dictationAlternativesForMarker):
(WebCore::AlternativeTextController::applyDictationAlternative):
* editing/AlternativeTextController.h:
* editing/Editor.cpp:
(WebCore::Editor::notifyComponentsOnChangedSelection): Renamed existing respondToChangedSelection() function to avoid naming collision.
(WebCore::Editor::appliedEditing):
(WebCore::Editor::unappliedEditing):
(WebCore::Editor::reappliedEditing):
(WebCore::Editor::updateMarkersForWordsAffectedByEditing):
(WebCore::Editor::changeSelectionAfterCommand):
(WebCore::Editor::respondToChangedSelection):
(WebCore::Editor::dictationAlternativesForMarker):
(WebCore::Editor::applyDictationAlternativelternative):
* editing/Editor.h:
* editing/FrameSelection.h:
* editing/mac/AlternativeTextUIController.h: Added. WK1 and WK2 use this class to keep track of text alternatives objects.
(AlternativeTextUIController):
(WebCore::AlternativeTextUIController::AlternativeTextUIController):
(AlernativeTextContextController):
(WebCore::AlternativeTextUIController::AlernativeTextContextController::AlernativeTextContextController):
* editing/mac/AlternativeTextUIController.mm: Added.
(WebCore::AlternativeTextUIController::AlernativeTextContextController::addAlternatives):
(WebCore::AlternativeTextUIController::AlernativeTextContextController::alternativesForContext):
(WebCore::AlternativeTextUIController::AlernativeTextContextController::removeAlternativesForContext):
(WebCore::AlternativeTextUIController::AlernativeTextContextController::clear):
(WebCore::AlternativeTextUIController::addAlternatives):
(WebCore::AlternativeTextUIController::alternativesForContext):
(WebCore::AlternativeTextUIController::clear):
(WebCore::AlternativeTextUIController::showAlternatives):
(WebCore::AlternativeTextUIController::handleAcceptedAlternative):
(WebCore::AlternativeTextUIController::dismissAlternatives):
(WebCore::AlternativeTextUIController::removeAlternatives):
* editing/mac/TextAlternativeWithRange.h: Added. A simple struct to make it easier to pass around a pair of text alternatives object and range.
* editing/mac/TextAlternativeWithRange.mm: Added.
(WebCore::TextAlternativeWithRange::TextAlternativeWithRange):
(WebCore::collectDictationTextAlternatives):
* page/AlternativeTextClient.h:
* page/ContextMenuController.cpp: Added code to show alternative dictated text in context menu.
(WebCore::ContextMenuController::contextMenuItemSelected):
(WebCore::ContextMenuController::populate):
(WebCore::ContextMenuController::checkOrEnableIfNeeded):
* platform/ContextMenuItem.h:
* rendering/HitTestResult.cpp:
(WebCore::HitTestResult::dictationAlternatives):
* rendering/HitTestResult.h:
* rendering/InlineTextBox.cpp:
(WebCore::InlineTextBox::paintDocumentMarker):
2012-06-14 Emil A Eklund <eae@chromium.org>
Change RenderTableSection::calcRowLogicalHeight to round rather than floor height
https://bugs.webkit.org/show_bug.cgi?id=88813
Reviewed by Julien Chaffraix.
Change RenderTableSection::calcRowLogicalHeight to round the
logicalHeight instead of flooring it. This matches our rounding logic
elsewhere and results in table rows better matching the expected height.
Test: fast/sub-pixel/table-rows-have-stable-height.html
* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::calcRowLogicalHeight):
2012-06-14 Andrey Adaikin <aandrey@chromium.org>
Web Inspector: Remove devtoolsInjectedScript hidden property
https://bugs.webkit.org/show_bug.cgi?id=89087
We use the devtoolsInjectedScript hidden property to reference a InjectedScript object
for a given ScriptState, but we also have the same reference in the InjectedScriptManager's
hash maps. Just use the hash maps and remove the hidden property from the global object.
Reviewed by Pavel Feldman.
* bindings/js/JSDOMGlobalObject.cpp:
(WebCore::JSDOMGlobalObject::visitChildren):
* bindings/js/JSDOMGlobalObject.h:
(JSDOMGlobalObject):
* bindings/js/JSInjectedScriptManager.cpp:
(WebCore::InjectedScriptManager::createInjectedScript):
* bindings/v8/V8HiddenPropertyName.h:
(WebCore):
* bindings/v8/custom/V8InjectedScriptManager.cpp:
* inspector/InjectedScriptManager.cpp:
(WebCore::InjectedScriptManager::discardInjectedScripts):
(WebCore::InjectedScriptManager::discardInjectedScriptsFor):
(WebCore::InjectedScriptManager::injectedScriptFor):
(WebCore):
(WebCore::InjectedScriptManager::wrapWebGLRenderingContextForInstrumentation):
* inspector/InjectedScriptManager.h:
(InjectedScriptManager):
2012-06-14 James Robinson <jamesr@chromium.org>
[chromium] Move opaque rectangle tracking logic out of compositor core
https://bugs.webkit.org/show_bug.cgi?id=89031
Reviewed by Adrienne Walker.
This moves the opaque paint tracking logic from the compositor (specifically CanvasLayerTextureUpdater) out to a
dedicated class used by GraphicsLayerChromium for GraphicsContext-painted content. This changes the interface
exposed by the compositor to pass a SkCanvas and let the user fill in an opaque rect if needed. Callers that are
using GraphicsContext to paint can instantiate a PlatformContextSkia / GraphicsContext around this SkCanvas if
they choose to, or use some other strategy, or not fill it in at all. This avoids creating extra wrapper classes
if the eventual caller is going to use the underlying SkCanvas anyway (such as WebContentLayer) and breaks the
dependency from the compositor code on GraphicsContext / PlatformContextSkia.
OpaqueRectTrackingContentLayerDelegate is a helper class that wraps a ContentLayerDelegate implementation with
PlatformContextSkia's opaque region tracking logic.
* WebCore.gypi:
* platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.cpp:
(WebCore::BitmapCanvasLayerTextureUpdater::BitmapCanvasLayerTextureUpdater):
(WebCore::BitmapCanvasLayerTextureUpdater::prepareToUpdate):
(WebCore::BitmapCanvasLayerTextureUpdater::updateTextureRect):
(WebCore::BitmapCanvasLayerTextureUpdater::setOpaque):
* platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.h:
(WebCore):
(BitmapCanvasLayerTextureUpdater):
* platform/graphics/chromium/CanvasLayerTextureUpdater.cpp:
(WebCore::CanvasLayerTextureUpdater::paintContents):
* platform/graphics/chromium/CanvasLayerTextureUpdater.h:
(WebCore):
(CanvasLayerTextureUpdater):
* platform/graphics/chromium/ContentLayerChromium.cpp:
(WebCore::ContentLayerPainter::paint):
* platform/graphics/chromium/ContentLayerChromium.h:
(ContentLayerDelegate):
(WebCore):
* platform/graphics/chromium/GraphicsLayerChromium.cpp:
(WebCore::GraphicsLayerChromium::GraphicsLayerChromium):
(WebCore::GraphicsLayerChromium::paint):
* platform/graphics/chromium/GraphicsLayerChromium.h:
(GraphicsLayerChromium):
* platform/graphics/chromium/LayerPainterChromium.h:
(LayerPainterChromium):
* platform/graphics/chromium/LinkHighlight.cpp:
(WebCore::LinkHighlight::paintContents):
* platform/graphics/chromium/LinkHighlight.h:
(LinkHighlight):
* platform/graphics/chromium/OpaqueRectTrackingContentLayerDelegate.cpp: Added.
(WebCore):
(WebCore::OpaqueRectTrackingContentLayerDelegate::OpaqueRectTrackingContentLayerDelegate):
(WebCore::OpaqueRectTrackingContentLayerDelegate::~OpaqueRectTrackingContentLayerDelegate):
(WebCore::OpaqueRectTrackingContentLayerDelegate::paintContents):
* platform/graphics/chromium/OpaqueRectTrackingContentLayerDelegate.h: Added.
(WebCore):
(GraphicsContextPainter):
(WebCore::GraphicsContextPainter::~GraphicsContextPainter):
(OpaqueRectTrackingContentLayerDelegate):
(WebCore::OpaqueRectTrackingContentLayerDelegate::setOpaque):
* platform/graphics/chromium/PlatformCanvas.cpp:
(WebCore::PlatformCanvas::Painter::Painter):
* platform/graphics/chromium/SkPictureCanvasLayerTextureUpdater.cpp:
(WebCore::SkPictureCanvasLayerTextureUpdater::prepareToUpdate):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore):
2012-06-14 Justin Novosad <junov@chromium.org>
[Chromium] webkitImageSmoothingEnabled canvas property does not work on redraw
https://bugs.webkit.org/show_bug.cgi?id=89018
Reviewed by Stephen White.
Test: fast/canvas/canvas-imageSmoothingEnabled-repaint.html
When an accelerated canvas layer prepares its texture for the
compositor, it must send a notification to skia to invalidate texture
proprties that are cached by skia, since the compositor may modify them.
The use case this fixes is when a canvas to canvas copy is performed
with webkitImageSmoothingEnabled=false on the destination canvas.
The backing texture of the source canvas will be set to "nearest"
filtering by skia in order to perform the copy. Then, the compositor
sets filtering back to "linear" when the source canvas is drawn.
Skia is designed to only update GL attributes when required, so errors
occur when the skia-side GL state cache is out of sync.
* platform/graphics/chromium/Canvas2DLayerBridge.cpp:
(WebCore::Canvas2DLayerBridge::prepareTexture):
2012-06-14 Jon Honeycutt <jhoneycutt@apple.com>
Fix some failing tests on Windows by resetting the page scale factor
between tests.
https://bugs.webkit.org/show_bug.cgi?id=89078
Reviewed by Dirk Pranke.
* testing/Internals.cpp:
(WebCore::Internals::reset):
Set the scale factor to 1.0.
2012-06-14 Sami Kyostila <skyostil@chromium.org>
Register fixed position layers with ScrollingCoordinator
https://bugs.webkit.org/show_bug.cgi?id=78864
Reviewed by James Robinson.
This patch makes ScrollingCoordinator aware of page elements that are
either fixed positioned directly or through their ancestors and that
also have their own composited backing layers. This allows the
ScrollingCoordinator implementation to potentially enable fast
scrolling even if there are fixed elements on the page, because the
compositor can automatically reposition the fixed element layers after
scrolling.
Since fixed position layers need to be explicitly supported by the
ScrollingCoordinator implementation, we add a new
supportsFixedPositionLayers() query for checking this.
* page/FrameView.cpp:
(WebCore::FrameView::FrameView):
(WebCore::FrameView::useSlowRepaints):
(WebCore::FrameView::addFixedObject):
(WebCore::FrameView::removeFixedObject):
* page/FrameView.h:
(FrameView):
(WebCore::FrameView::fixedObjects):
(WebCore::FrameView::hasFixedObjects):
* page/scrolling/ScrollingCoordinator.cpp:
(WebCore::ScrollingCoordinator::frameViewFixedObjectsDidChange):
(WebCore::ScrollingCoordinator::hasNonLayerFixedObjects):
(WebCore):
(WebCore::ScrollingCoordinator::updateShouldUpdateScrollLayerPositionOnMainThread):
(WebCore::ScrollingCoordinator::setLayerIsContainerForFixedPosition):
(WebCore::ScrollingCoordinator::setLayerIsFixedToContainerLayer):
* page/scrolling/ScrollingCoordinator.h:
(ScrollingCoordinator):
* page/scrolling/chromium/ScrollingCoordinatorChromium.cpp:
(WebCore::ScrollingCoordinator::setLayerIsContainerForFixedPosition):
(WebCore):
(WebCore::ScrollingCoordinator::setLayerIsFixedToContainerLayer):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::willBeDestroyed):
(WebCore::RenderBox::styleWillChange):
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateGraphicsLayerGeometry):
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::updateBacking):
(WebCore):
(WebCore::RenderLayerCompositor::fixedPositionedByAncestor):
(WebCore::RenderLayerCompositor::ensureRootLayer):
* rendering/RenderLayerCompositor.h:
2012-06-14 Alexey Proskuryakov <ap@apple.com>
[CFNetwork] XMLHttpRequest incorrectly returns cached responses even when there is a Vary header field
https://bugs.webkit.org/show_bug.cgi?id=88925
Reviewed by David Kilzer.
Test: http/tests/cache/xhr-vary-header.html
* platform/network/cf/ResourceHandleCFNet.cpp: (WebCore::willCacheResponse): Don't
cache the response at all if there is a Vary header field. CFNetwork won't look at
it if we allowed the response cached.
* platform/network/mac/ResourceHandleMac.mm:
(WebCore::ResourceHandle::createNSURLConnection): [NSURLResponse copy] is mutable
in practice, but we should be nice, and use -mutableCopy.
(-[WebCoreResourceHandleAsDelegate connection:willCacheResponse:]): Same fix as in
CFNet code.
2012-06-13 Andrey Kosyakov <caseq@chromium.org>
Web Inspector: use CSS gradients instead of images for timeline bars, checkboxes & bullets
https://bugs.webkit.org/show_bug.cgi?id=89024
Reviewed by Pavel Feldman.
This brings a bunch of UI changes to the way Timeline bars and related controls are rendered:
- introduce new color palette for categories;
- replace images used to render timeline bars, list bullets and categories checkboxes with CSS gradients;
- "flatten" timeline bars
- make square list bullets, sharpen corners for lower pane bars;
- make links on the list sidebar grey
- add more space between categories checkboxes, make labels lighter.
* WebCore.gypi: Remove old images, add tickMark.png.
* inspector/front-end/Images/timelineBarBlue.png: Removed.
* inspector/front-end/Images/timelineBarDarkGreen.png: Removed.
* inspector/front-end/Images/timelineBarGray.png: Removed.
* inspector/front-end/Images/timelineBarGreen.png: Removed.
* inspector/front-end/Images/timelineBarLightPurple.png: Removed.
* inspector/front-end/Images/timelineBarOrange.png: Removed.
* inspector/front-end/Images/timelineBarPurple.png: Removed.
* inspector/front-end/Images/timelineBarRed.png: Removed.
* inspector/front-end/Images/timelineBarYellow.png: Removed.
* inspector/front-end/Images/timelineCheckmarks.png: Removed.
* inspector/front-end/Images/timelineDots.png: Removed.
* inspector/front-end/TimelinePanel.js:
(WebInspector.TimelinePanel.prototype.wasShown):
(WebInspector.TimelinePanel.prototype._closeRecordDetails): Fixed trailing spaces;
(WebInspector.TimelinePanel.prototype._injectCategoryStyles): Inject generated styles for categories, if necessary.
* inspector/front-end/TimelinePresentationModel.js:
(WebInspector.TimelinePresentationModel.categories):
(WebInspector.TimelinePresentationModel.createFillStyle):
(WebInspector.TimelinePresentationModel.createStyleRuleForCategory):
* inspector/front-end/WebKit.qrc: Remove old images, add tickMark.png
* inspector/front-end/inspector.css:
(.sidebar-tree-section):
* inspector/front-end/timelinePanel.css:
(#timeline-container .webkit-html-resource-link):
(.timeline-category-statusbar-item):
(.timeline-category-statusbar-item .timeline-category-checkbox):
(.timeline-category-statusbar-item .timeline-category-checkbox:checked::before):
(.timeline-tree-item):
(.timeline-tree-item .timeline-tree-icon):
(.timeline-graph-bar):
(.timeline-graph-bar.with-children):
(.timeline-graph-bar.cpu):
(.timeline-aggregated-category):
* inspector/front-end/utilities.js:
(Object.values): Return an array of object field's values.
2012-06-14 Silvia Pfeiffer <silviapf@chromium.org>
Change mute button and volume slider behaviour for new Chromium video controls.
https://bugs.webkit.org/show_bug.cgi?id=88743
Reviewed by Eric Carlson.
No new tests, final patch will contain the rebaselined tests.
The Chrome video controls are receiving a visual update. This patch makes the mute button and
the volume slider interact with each other. When the user mutes the video through clicking the
mute button, the volume slider animates to 0% volume. Upon unmute, the previous volume level
is restored. Further, when the user drags the volume slider thumb or clicks on the slider and
the video is muted, the video unmutes and sets the volume to the user's chosen level.
* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlVolumeSliderElement::MediaControlVolumeSliderElement):
Add a member field to the volume slider to allow user interaction to unmute the video.
(WebCore::MediaControlVolumeSliderElement::defaultEventHandler):
Add the logic to unmute the video if the m_clearMutedOnUserInteraction member field is true.
(WebCore::MediaControlVolumeSliderElement::setClearMutedOnUserInteraction):
Add a member function to change the value of the m_clearMutedOnUserInteraction member field.
* html/shadow/MediaControlElements.h:
(MediaControlVolumeSliderElement):
Add the setClearMutedOnUserInteraction member function and the m_clearMutedOnUserInteraction member field.
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlRootElementChromium::create):
Activate unmute on user interaction on the Chromium video volume slider.
(WebCore::MediaControlRootElementChromium::changedMute):
Add logic to set volume slider to 0 when muted and reset to volume when unmuted.
* rendering/RenderMediaControlsChromium.cpp:
(WebCore::paintMediaVolumeSlider):
Add logic to paint the volume slider active area as 0 when muted.
2012-06-14 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: Breakpoints should not be set when UISourceCode content is diverged from VM version of the script.
https://bugs.webkit.org/show_bug.cgi?id=89109
Reviewed by Pavel Feldman.
* inspector/front-end/BreakpointManager.js:
(WebInspector.BreakpointManager.Breakpoint.prototype._updateBreakpoint):
* inspector/front-end/JavaScriptSource.js:
(WebInspector.JavaScriptSource.prototype.isDivergedFromVM):
(WebInspector.JavaScriptSource.prototype.workingCopyCommitted):
2012-06-14 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: DebuggerModel should support setting breakpoints without script by sourceURL only.
https://bugs.webkit.org/show_bug.cgi?id=88988
Reviewed by Pavel Feldman.
This change allows setting all saved breakpoints in debugger model by url before the site
with corresponding scripts was loaded (allows breaking in onload handler on first load).
This change also simplifies implementation of snippets debugging.
* inspector/front-end/BreakpointManager.js:
* inspector/front-end/DebuggerModel.js:
(WebInspector.DebuggerModel.prototype.setBreakpointByScriptLocation):
2012-06-14 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: refactor settings screen to move user agent and experiments out of general tab.
https://bugs.webkit.org/show_bug.cgi?id=89101
Reviewed by Vsevolod Vlasov.
See image attached to the bug.
* English.lproj/localizedStrings.js:
* inspector/front-end/SettingsScreen.js:
(WebInspector.SettingsScreen.prototype.selectTab):
(WebInspector.SettingsScreen.prototype._tabSelected):
(WebInspector.SettingsScreen.prototype.set wasShown):
(WebInspector.SettingsScreen.prototype.willHide):
(WebInspector.SettingsTab):
(WebInspector.SettingsTab.prototype._appendSection):
(WebInspector.SettingsTab.prototype._createCustomSetting):
(WebInspector.GenericSettingsTab):
(WebInspector.GenericSettingsTab.prototype._showPaintRectsChanged):
(WebInspector.GenericSettingsTab.prototype.get _updateScriptDisabledCheckbox):
(WebInspector.GenericSettingsTab.prototype._javaScriptDisabledChanged):
(WebInspector.UserAgentSettingsTab):
(WebInspector.UserAgentSettingsTab.prototype._createUserAgentSelectRowElement.get const):
(WebInspector.UserAgentSettingsTab.prototype._createUserAgentSelectRowElement.textDoubleClicked):
(WebInspector.UserAgentSettingsTab.prototype._createUserAgentSelectRowElement.textChanged):
(WebInspector.UserAgentSettingsTab.prototype._createUserAgentSelectRowElement.set checkboxClicked):
(WebInspector.UserAgentSettingsTab.prototype._createDeviceMetricsElement):
(WebInspector.ExperimentsSettingsTab):
(WebInspector.ExperimentsSettingsTab.prototype._createExperimentsWarningSubsection):
(WebInspector.ExperimentsSettingsTab.prototype._createExperimentCheckbox.listener):
(WebInspector.ExperimentsSettingsTab.prototype._createExperimentCheckbox):
* inspector/front-end/ShortcutsScreen.js:
* inspector/front-end/TabbedPane.js:
(WebInspector.TabbedPane.prototype.get headerContentsElement):
* inspector/front-end/helpScreen.css:
(.settings-tab.help-container):
(.settings-tab .help-block):
(.help-content fieldset):
(#settings-screen .tabbed-pane-header-tabs):
(#settings-screen .tabbed-pane-header-tab):
2012-06-13 Mark Hahnenberg <mhahnenberg@apple.com>
Worker tear-down can re-enter JSC during GC finalization pt. 2
https://bugs.webkit.org/show_bug.cgi?id=88601
Reviewed by David Levin.
No new tests. Current regression tests are sufficient.
* workers/WorkerMessagingProxy.cpp:
(WebCore::WorkerMessagingProxy::WorkerMessagingProxy):
(WebCore::WorkerMessagingProxy::workerObjectDestroyed): We clear the m_workerObject here because
we don't want anybody else trying to send messages to the Worker now that it has been destroyed.
We also queue the asynchronous task for the various other cleanup that still needs to be done.
This allows us to avoid the problem of re-entrant JS code execution during GC.
(WebCore):
(WebCore::WorkerMessagingProxy::workerObjectDestroyedInternal): Here we set m_mayBeDestroyed to true.
This is the point after which deleting the WorkerMessagingProxy in workerContextDestroyedInternal()
is okay. It could happen during this function call if the worker thread has been shutdown already, or
it could be called later after we shut down the worker thread.
(WebCore::WorkerMessagingProxy::workerContextDestroyedInternal): We check m_mayBeDestroyed here
instead of checking m_workerObject. This change effectively orthogonalizes the roles that m_workerObject
was filling. Since we were eagerly clearing m_workerObject, but we wanted to asynchronously call
workerObjectDestroyed(), we needed to make sure we didn't accidentally try to delete the WorkerMessagingProxy
twice (once from destroying the Worker and once from destroying the WorkerContext). This boolean field
should fix that issue--we set it lazily like we wanted to do without being in danger of causing use-after-free
issues with m_workerObject.
* workers/WorkerMessagingProxy.h: Added the new field and function.
(WorkerMessagingProxy):
2012-06-14 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: Selector list start position is not extracted for style rules inside @media rules
https://bugs.webkit.org/show_bug.cgi?id=89094
Reviewed by Pavel Feldman.
Selector list start is now marked inside the "ruleset" rule rather than separately, before its invocation(s).
* css/CSSGrammar.y:
2012-06-14 Eric Penner <epenner@google.com>
[chromium] Fix race condition where animations start, finish and are deleted on the composite thread, all before the start even arrives on the main thread.
https://bugs.webkit.org/show_bug.cgi?id=88439
Reviewed by James Robinson.
Scheduling behavior covered by existing tests.
* platform/graphics/chromium/cc/CCActiveAnimation.cpp:
(WebCore::CCActiveAnimation::isFinishedAt):
* platform/graphics/chromium/cc/CCActiveAnimation.h:
(WebCore::CCActiveAnimation::isFinished):
* platform/graphics/chromium/cc/CCLayerAnimationController.cpp:
(WebCore::CCLayerAnimationController::suspendAnimations):
(WebCore::CCLayerAnimationController::pushAnimationUpdatesTo):
(WebCore::CCLayerAnimationController::animate):
(WebCore::CCLayerAnimationController::hasActiveAnimation):
(WebCore):
(WebCore::CCLayerAnimationController::markAnimationsForDeletion):
(WebCore::CCLayerAnimationController::purgeAnimationsMarkedForDeletion):
* platform/graphics/chromium/cc/CCLayerAnimationController.h:
(CCLayerAnimationController):
2012-06-14 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: ConsoleView.evaluateUsingTextPrompt should evaluate without command line API.
https://bugs.webkit.org/show_bug.cgi?id=89096
Reviewed by Pavel Feldman.
* inspector/front-end/ConsoleView.js:
2012-06-14 Silvia Pfeiffer <silviapf@chromium.org>
Update the time display of the new Chromium media controls.
https://bugs.webkit.org/show_bug.cgi?id=88724
Reviewed by Eric Carlson.
No new tests, final patch will contain the rebaselined tests.
The Chrome video controls are receiving a visual update. This patch changes
the format of the time digits that are displayed. They are different for media
files of different duration. For non-autoplaying videos, the duration is shown
before playback is started.
* css/mediaControlsChromium.css:
(audio::-webkit-media-controls-time-remaining-display, video::-webkit-media-controls-time-remaining-display):
Format the new duration display the same as the current time display.
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlRootElementChromium::MediaControlRootElementChromium):
Add a member field for the duration display.
(WebCore::MediaControlRootElementChromium::create):
Add an element for the duration display and initially hide the current time display.
(WebCore::MediaControlRootElementChromium::setMediaController):
Add the duration display to the controller.
(WebCore::MediaControlRootElementChromium::reset):
Add reset functionality for the duration display.
(WebCore::MediaControlRootElementChromium::playbackStarted):
Show the current time display and hide the duration display upon playback start.
(WebCore::MediaControlRootElementChromium::updateTimeDisplay):
Show the current time display and hide the duration display when current time is non-zero.
* html/shadow/MediaControlRootElementChromium.h:
(WebCore):
Add the MediaControlTimeRemainingDisplayElement which provides use with the duration display.
(MediaControlRootElementChromium):
Add a duration member field, reusing the existing MediaControlTimeRemainingDisplayElement.
* rendering/RenderMediaControlsChromium.cpp:
(WebCore::formatChromiumMediaControlsTime):
Formats a given time under the condition of a given duration.
(WebCore::RenderMediaControlsChromium::formatMediaControlsTime):
When formatting a single time, assume it to also provide the duration conditions.
(WebCore::RenderMediaControlsChromium::formatMediaControlsCurrentTime):
Format the current time given the duration.
(WebCore::RenderMediaControlsChromium::formatMediaControlsRemainingTime):
Format the remaining time from the current time given the duration.
* rendering/RenderMediaControlsChromium.h:
(RenderMediaControlsChromium):
Declare the time formatting member functions.
* rendering/RenderThemeChromiumMac.h:
(RenderThemeChromiumMac):
Declare the time formatting member functions for Chromium on the Mac.
* rendering/RenderThemeChromiumMac.mm:
(WebCore::RenderThemeChromiumMac::formatMediaControlsTime):
(WebCore::RenderThemeChromiumMac::formatMediaControlsCurrentTime):
(WebCore::RenderThemeChromiumMac::formatMediaControlsRemainingTime):
Call the time formatting member functions of RenderMediaControlsChromium on the Mac.
* rendering/RenderThemeChromiumSkia.cpp:
(WebCore::formatMediaControlsTime):
(WebCore::formatMediaControlsCurrentTime):
(WebCore::formatMediaControlsRemainingTime):
Call the time formatting member functions of RenderMediaControlsChromium on Skia.
* rendering/RenderThemeChromiumSkia.h:
(RenderThemeChromiumSkia):
Declare the time formatting member functions for Chromium Skia.
2012-06-14 Eugene Klyuchnikov <eustas.bug@gmail.com>
Web Inspector: Refactor message loop instrumentation.
https://bugs.webkit.org/show_bug.cgi?id=88978
Reviewed by Pavel Feldman.
1) Remove "messageLoop" term
2) Reuse WebThread::TaskObserver interface
3) Move implementation (from embedder) to platform code.
* inspector/InspectorClient.h:
(WebCore::InspectorClient::startMainThreadMonitoring): Renamed method.
(WebCore::InspectorClient::stopMainThreadMonitoring): Ditto.
* inspector/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::start):
(WebCore::InspectorTimelineAgent::stop):
2012-06-14 Dominik Röttsches <dominik.rottsches@intel.com>
[svg] SVGResources applied to Text with Incorrect Transformations in non-CG Implementations
https://bugs.webkit.org/show_bug.cgi?id=64966
Reviewed by Nikolas Zimmermann.
SVGInlineTextBox::paintTextWithShadows for non-CG implementations resets the scale aspect
of the currently applied CTM. This leads to patterns and gradients being incorrectly downscaled.
This way of resetting the CTM and drawing fonts "natively" scaled was introduced
in r77485 to fix a number of issues with SVG font scaling.
Unfortunately, this breaks scaling of patterns and gradients. To fix it,
we need to push the transformation back into the pattern/gradient space transformation.
Tests: svg/transforms/transformed-text-fill-gradient.html
svg/transforms/transformed-text-fill-pattern.html
* rendering/svg/RenderSVGInlineText.cpp:
(WebCore::RenderSVGInlineText::computeNewScaledFontForStyle): Moving the scaling factor calculation out into SVGRenderingContext.
* rendering/svg/RenderSVGResourceContainer.cpp:
(WebCore::RenderSVGResourceContainer::shouldTransformOnTextPainting): Reusing the scaling factor calculation from SVGRenderingContext.
(WebCore):
* rendering/svg/RenderSVGResourceContainer.h:
(RenderSVGResourceContainer):
* rendering/svg/RenderSVGResourceGradient.cpp:
(WebCore::RenderSVGResourceGradient::applyResource): If needed, push down transformation into gradient space.
* rendering/svg/RenderSVGResourcePattern.cpp:
(WebCore::RenderSVGResourcePattern::applyResource): If needed, push transformation into pattern space.
* rendering/svg/SVGRenderingContext.cpp:
(WebCore::SVGRenderingContext::calculateScreenFontSizeScalingFactor): Common scaling factor calculation for RenderSVGInlineText and RenderSVGResourceContainer.
* rendering/svg/SVGRenderingContext.h:
(SVGRenderingContext):
2012-06-14 Yoshifumi Inoue <yosin@chromium.org>
[Form] Replace InputNumber type to Decimal type and drop InputNumber
https://bugs.webkit.org/show_bug.cgi?id=88746
Reviewed by Kent Tamura.
This patch removes InputNumber introduced by r119540 by replacing
InputNumber type to Decimal type and changing convertDoubleToDecimal
and convertDecimalToDouble to Decimal::fromDouble and
Decimal::toDouble, respectively.
No new tests. This patch doesn't change behavior.
* html/BaseDateAndTimeInputType.cpp:
(WebCore::BaseDateAndTimeInputType::setValueAsDecimal):
(WebCore::BaseDateAndTimeInputType::defaultValueForStepUp):
(WebCore::BaseDateAndTimeInputType::parseToNumber):
(WebCore::BaseDateAndTimeInputType::serialize):
(WebCore::BaseDateAndTimeInputType::serializeWithComponents):
(WebCore::BaseDateAndTimeInputType::serializeWithMilliseconds):
* html/BaseDateAndTimeInputType.h:
(BaseDateAndTimeInputType):
* html/DateInputType.cpp:
(WebCore::DateInputType::createStepRange):
* html/DateTimeInputType.cpp:
(WebCore::DateTimeInputType::defaultValueForStepUp):
(WebCore::DateTimeInputType::createStepRange):
* html/DateTimeInputType.h:
(DateTimeInputType):
* html/DateTimeLocalInputType.cpp:
(WebCore::DateTimeLocalInputType::createStepRange):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::getAllowedValueStep):
* html/HTMLInputElement.h:
(HTMLInputElement):
* html/InputType.cpp:
(WebCore::InputType::setValueAsDouble):
(WebCore::InputType::setValueAsDecimal):
(WebCore::InputType::rangeUnderflow):
(WebCore::InputType::rangeOverflow):
(WebCore::InputType::defaultValueForStepUp):
(WebCore::InputType::minimum):
(WebCore::InputType::maximum):
(WebCore::InputType::isInRange):
(WebCore::InputType::isOutOfRange):
(WebCore::InputType::stepMismatch):
(WebCore::InputType::validationMessage):
(WebCore::InputType::parseToNumber):
(WebCore::InputType::parseToNumberOrNaN):
(WebCore::InputType::serialize):
(WebCore::InputType::applyStep):
(WebCore::InputType::getAllowedValueStep):
(WebCore::InputType::stepUpFromRenderer):
* html/InputType.h:
(InputType):
* html/MonthInputType.cpp:
(WebCore::MonthInputType::defaultValueForStepUp):
(WebCore::MonthInputType::createStepRange):
(WebCore::MonthInputType::parseToNumber):
* html/MonthInputType.h:
(MonthInputType):
* html/NumberInputType.cpp:
(WebCore::NumberInputType::setValueAsDecimal):
(WebCore::NumberInputType::createStepRange):
(WebCore::NumberInputType::parseToNumber):
(WebCore::NumberInputType::serialize):
* html/NumberInputType.h:
(NumberInputType):
* html/RangeInputType.cpp:
(WebCore::ensureMaximum):
(WebCore::RangeInputType::setValueAsDecimal):
(WebCore::RangeInputType::createStepRange):
(WebCore::RangeInputType::handleKeydownEvent):
(WebCore::RangeInputType::parseToNumber):
(WebCore::RangeInputType::serialize):
(WebCore::RangeInputType::sanitizeValue):
* html/RangeInputType.h:
(RangeInputType):
* html/StepRange.cpp:
(WebCore::StepRange::StepRange):
(WebCore::StepRange::acceptableError):
(WebCore::StepRange::alignValueForStep):
(WebCore::StepRange::clampValue):
(WebCore::StepRange::parseStep):
(WebCore::StepRange::roundByStep):
(WebCore::StepRange::stepMismatch):
* html/StepRange.h:
(WebCore::StepRange::StepDescription::defaultValue):
(StepRange):
(WebCore::StepRange::maximum):
(WebCore::StepRange::minimum):
(WebCore::StepRange::step):
(WebCore::StepRange::stepBase):
(WebCore::StepRange::defaultValue):
(WebCore::StepRange::proportionFromValue):
(WebCore::StepRange::valueFromProportion):
* html/TimeInputType.cpp:
(WebCore::TimeInputType::defaultValueForStepUp):
(WebCore::TimeInputType::createStepRange):
* html/TimeInputType.h:
(TimeInputType):
* html/WeekInputType.cpp:
(WebCore::WeekInputType::createStepRange):
* html/shadow/CalendarPickerElement.cpp:
(WebCore::CalendarPickerElement::writeDocument):
* html/shadow/SliderThumbElement.cpp:
(WebCore::sliderPosition):
(WebCore::RenderSliderThumb::layout):
(WebCore::SliderThumbElement::setPositionFromPoint):
2012-06-14 Jan Keromnes <janx@linux.com>
Web Inspector: Implement ExtensionPanel.show() method
https://bugs.webkit.org/show_bug.cgi?id=88473
Implemented ExtensionPanel.show() API method to allow extension
developers to show their panel when needed.
Reviewed by Pavel Feldman.
Tests for this method are in:
LayoutTests/inspector/extensions/extensions-panel.html
* inspector/front-end/ExtensionAPI.js:
(defineCommonExtensionSymbols):
(injectedExtensionAPI.ExtensionPanelImpl.prototype.createStatusBarButton):
(injectedExtensionAPI.ExtensionPanelImpl.prototype.show):
* inspector/front-end/ExtensionServer.js:
(WebInspector.ExtensionServer):
(WebInspector.ExtensionServer.prototype._onShowPanel):
2012-06-14 Kent Tamura <tkent@chromium.org>
Fix a typo in r120304.
https://bugs.webkit.org/show_bug.cgi?id=89070
* bindings/scripts/CodeGeneratorV8.pm:
(NativeToJSValue):
* bindings/scripts/test/V8/V8TestObj.cpp:
2012-06-14 David Barr <davidbarr@chromium.org>
Add dpcm to css3-images image-resolution
https://bugs.webkit.org/show_bug.cgi?id=85442
Reviewed by Tony Chang.
The css3-images module is at candidate recommendation.
http://www.w3.org/TR/2012/CR-css3-images-20120417/#the-image-resolution
No new tests; extended fast/css/image-resolution/image-resolution.html
* css/CSSGrammar.y: Add DPCM token and unary term.
* css/CSSParser.cpp: Extend range of units accepted for resolution and tokenize "dpcm".
(WebCore::CSSParser::validUnit): Accept CSSPrimitiveValue::CSS_DPCM.
(WebCore::CSSParser::createPrimitiveNumericValue): Accept CSSPrimitiveValue::CSS_DPCM.
(WebCore::unitFromString): Map "dpcm" to CSSPrimitiveValue::CSS_DPCM.
(WebCore::CSSParser::parseValidPrimitive): Accept CSSPrimitiveValue::CSS_DPCM.
(WebCore::CSSParser::detectNumberToken): Map "dpcm" to CSSPrimitiveValue::CSS_DPCM.
* css/CSSPrimitiveValue.cpp: Add CSSPrimitiveValue::CSS_DPCM as a numeric unit.
(WebCore::isValidCSSUnitTypeForDoubleConversion): Accept CSSPrimitiveValue::CSS_DPCM.
(WebCore::unitCategory): Map CSSPrimitiveValue::CSS_DPCM to CSSPrimitiveValue::UResolution.
(WebCore::conversionToCanonicalUnitsScaleFactor): Scale CSSPrimitiveValue::CSS_DPCM to CSSPrimitiveValue::CSS_DPPX.
(WebCore::CSSPrimitiveValue::customCssText): Format CSSPrimitiveValue::CSS_DPCM as "<num>dpcm".
(WebCore::CSSPrimitiveValue::cloneForCSSOM): Accept CSSPrimitiveValue::CSS_DPCM as numeric unit.
* css/CSSPrimitiveValue.h: Add CSSPrimitiveValue::CSS_DPCM.
2012-06-14 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: REGRESSION HeapProfiler can't remove heap snapshot from the list if Dominators view has been opened.
https://bugs.webkit.org/show_bug.cgi?id=89080
There was no dispose method on artificial ShowMoreDataGrid node.
Reviewed by Pavel Feldman.
* inspector/front-end/ShowMoreDataGridNode.js:
(WebInspector.ShowMoreDataGridNode.prototype.nodeHeight):
(WebInspector.ShowMoreDataGridNode.prototype.dispose):
2012-06-14 Kent Tamura <tkent@chromium.org>
[JSC/V8] Improve DOMString[] support
https://bugs.webkit.org/show_bug.cgi?id=89070
Reviewed by Kentaro Hara.
Tests: new test case in TestObj.idl
* bindings/js/JSDOMBinding.cpp:
(WebCore::jsArray): Implement jsArray() for DOMStringList.
* bindings/js/JSDOMBinding.h: Declare jsArray() for DOMStringList.
* bindings/scripts/CodeGeneratorJS.pm:
(NativeToJSValue): Avoid to include JSDOMString.h and DOMString.h.
* bindings/scripts/CodeGeneratorV8.pm:
(GetNativeType): Move a check for DOMString[] because it's a RefPtr type
and we'd like to avoid to generate RefPtr<DOMString[]>
(%non_wrapper_types): DOMString[] is not a wrapper type. This is needed
to avoid to generate unnecessary custom function template.
(NativeToJSValue): Avoid to include V8DOMString.h and DOMString.h.
* bindings/scripts/test/TestObj.idl: Add non-overload function using DOMString[].
* bindings/scripts/test/JS/JSTestObj.cpp: Update expectation.
* bindings/scripts/test/JS/JSTestObj.h: ditto.
* bindings/scripts/test/V8/V8TestObj.cpp: ditto.
2012-06-14 Matt Falkenhagen <falken@chromium.org>
LocaleToScriptMappingDefault.cpp should also do what scriptCodeForFontSelection does in LocaleToScriptMappingICU.cpp
https://bugs.webkit.org/show_bug.cgi?id=88845
Reviewed by Kent Tamura.
Test: fast/text/international/jpan-script-equivalence.html
* platform/text/LocaleToScriptMappingDefault.cpp:
(WebCore::scriptNameToCode):
2012-06-14 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: render dock-to-right icon in status bar.
https://bugs.webkit.org/show_bug.cgi?id=89040
Reviewed by Vsevolod Vlasov.
Today we render dock to bottom icon for both actions (dock to bottom and dock to right).
* inspector/front-end/Images/statusbarButtonGlyphs.png:
* inspector/front-end/inspector.css:
(button.dock-status-bar-item.status-bar-item.toggled-undock .glyph):
(button.dock-status-bar-item.status-bar-item.toggled-bottom .glyph):
(button.dock-status-bar-item.status-bar-item.toggled-right .glyph):
* inspector/front-end/inspector.js:
(WebInspector._createGlobalStatusBarItems):
(WebInspector._installDockToRight.listener):
(windowLoaded):
(WebInspector.setDockingUnavailable):
2012-06-14 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: clear execution live before the text edit.
https://bugs.webkit.org/show_bug.cgi?id=89016
Reviewed by Vsevolod Vlasov.
Otherwise we can't remove them after committing the edit.
* inspector/front-end/JavaScriptSourceFrame.js:
(WebInspector.JavaScriptSourceFrame.prototype.beforeTextChanged):
(WebInspector.JavaScriptSourceFrame.prototype.clearExecutionLine):
2012-06-14 Min Qin <qinmin@google.com>
Remove the const keyword from MediaPlayerPrivate::enterFullscreen()
https://bugs.webkit.org/show_bug.cgi?id=89044
Reviewed by Adam Barth.
enterFullscreen() and exitFullscreen() could both modify the player state.
This change reverts the change we made in r120213.
Just function declaration change, no test needed.
* platform/graphics/MediaPlayerPrivate.h:
(WebCore::MediaPlayerPrivateInterface::enterFullscreen):
2012-06-14 Edaena Salinas Jasso <edaena@apple.com>
el.getAttribute('type') returns 'null' when setting el.type to null
https://bugs.webkit.org/show_bug.cgi?id=89030
Reviewed by Alexey Proskuryakov.
Test: fast/dom/HTMLButtonElement/change-type.html
* html/HTMLButtonElement.cpp:
(WebCore::HTMLButtonElement::setType): Changed parameter to be AtomicString to
address an unrelated review comment from bug 14439 for slightly better performance.
* html/HTMLButtonElement.h:
(HTMLButtonElement): Changed parameter to be AtomicString to address an unrelated
review comment from bug 14439 for slightly better performance.
* html/HTMLButtonElement.idl: Specified TreatNullAs NullString for the type attribute.
2012-06-14 Alec Flett <alecflett@chromium.org>
IndexedDB: Creating a transaction from within a transaction callback should fail
https://bugs.webkit.org/show_bug.cgi?id=80547
Reviewed by Tony Chang.
Prevent creation of a transaction, via IDBDatabase.transaction()
or IDBDatabase.setVersion(), from inside a setVersion()
transaction, as per spec. Update many tests to deal with this.
Test restructuring uncovered some inconsistencies in transaction
mode handling. So also do some minor cleanup in the way the
transaction mode is passed between the frontend and backend
transaction objects, and add additional asserts.
* Modules/indexeddb/IDBCursor.cpp:
(WebCore::IDBCursor::update):
* Modules/indexeddb/IDBCursorBackendImpl.cpp:
(WebCore::IDBCursorBackendImpl::update):
* Modules/indexeddb/IDBDatabase.cpp:
(WebCore::IDBDatabase::setVersion):
(WebCore::IDBDatabase::transaction):
* Modules/indexeddb/IDBObjectStoreBackendImpl.cpp:
(WebCore::IDBObjectStoreBackendImpl::putInternal):
* Modules/indexeddb/IDBRequest.cpp:
(WebCore::IDBRequest::onSuccess):
* Modules/indexeddb/IDBTransaction.cpp:
(WebCore::IDBTransaction::create):
(WebCore::IDBTransaction::IDBTransaction):
(WebCore::IDBTransaction::stringToMode):
(WebCore::IDBTransaction::modeToString):
* Modules/indexeddb/IDBTransaction.h:
(IDBTransaction):
2012-06-14 Li Yin <li.yin@intel.com>
[WebSocket] WebSocket object in Worker Context should fire error event when it fails connection.
https://bugs.webkit.org/show_bug.cgi?id=88744
Reviewed by Kent Tamura.
Spec:http://dev.w3.org/html5/websockets/#feedback-from-the-protocol
The behavior of firing error event in Document has been implemented in Committed r117944.
WebSocket in WorkerContext is also required to fire error event.
Test: http/tests/websocket/tests/hybi/workers/close.html
* Modules/websockets/ThreadableWebSocketChannelClientWrapper.cpp:
(WebCore::ThreadableWebSocketChannelClientWrapper::didReceiveMessageError):
(WebCore):
(WebCore::ThreadableWebSocketChannelClientWrapper::didReceiveMessageErrorCallback):
* Modules/websockets/ThreadableWebSocketChannelClientWrapper.h:
(ThreadableWebSocketChannelClientWrapper):
* Modules/websockets/WorkerThreadableWebSocketChannel.cpp:
(WebCore::workerContextDidReceiveMessageError):
(WebCore):
(WebCore::WorkerThreadableWebSocketChannel::Peer::didReceiveMessageError):
* Modules/websockets/WorkerThreadableWebSocketChannel.h:
(Peer):
2012-06-13 Kinuko Yasuda <kinuko@chromium.org>
Unreviewed build fix attempt for r120285
* Modules/filesystem/DOMFileSystemBase.cpp:
(WebCore):
2012-06-13 MORITA Hajime <morrita@google.com>
Unreviewed expectation update.
* bindings/scripts/test/V8/V8Float64Array.cpp:
* bindings/scripts/test/V8/V8TestActiveDOMObject.cpp:
* bindings/scripts/test/V8/V8TestCustomNamedGetter.cpp:
* bindings/scripts/test/V8/V8TestEventConstructor.cpp:
* bindings/scripts/test/V8/V8TestEventTarget.cpp:
* bindings/scripts/test/V8/V8TestException.cpp:
* bindings/scripts/test/V8/V8TestInterface.cpp:
* bindings/scripts/test/V8/V8TestMediaQueryListListener.cpp:
* bindings/scripts/test/V8/V8TestNamedConstructor.cpp:
* bindings/scripts/test/V8/V8TestNode.cpp:
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::V8TestObj::installPerContextProperties):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.cpp:
2012-06-13 Kinuko Yasuda <kinuko@chromium.org>
Move isolated file system code out of chromium-specific directory
https://bugs.webkit.org/show_bug.cgi?id=88997
Reviewed by Adam Barth.
The code does not need to be chromium-specific. Any platforms that
have file system implementation and manage mappings from file system
id to a set of files should share the code.
No new tests as this has no functional/behavioral changes.
* Modules/filesystem/DOMFileSystem.cpp:
(WebCore::DOMFileSystem::createIsolatedFileSystem):
(WebCore):
* Modules/filesystem/DOMFileSystem.h:
(DOMFileSystem):
* Modules/filesystem/DOMFileSystemBase.h:
(DOMFileSystemBase):
* Modules/filesystem/FileSystemType.h:
* Modules/filesystem/chromium/DOMFileSystemChromium.cpp:
* Modules/filesystem/chromium/DOMFileSystemChromium.h: Removed.
* Modules/filesystem/chromium/DraggedIsolatedFileSystem.cpp:
(WebCore::DraggedIsolatedFileSystem::getDOMFileSystem):
* WebCore.gypi:
2012-06-07 Kinuko Yasuda <kinuko@chromium.org>
File::lastModifiedDate should use NaN or separate boolean flag for null Date value
https://bugs.webkit.org/show_bug.cgi?id=87826
Reviewed by Kent Tamura.
Test: http/tests/local/fileapi/file-last-modified-after-delete.html
* fileapi/File.cpp:
(WebCore::File::File):
(WebCore::File::captureSnapshot):
(WebCore::File::lastModifiedDate):
(WebCore::File::lastModifiedDateForBinding): Removed.
* fileapi/File.h:
(File):
* fileapi/File.idl:
* platform/FileMetadata.h:
(WebCore::FileMetadata::FileMetadata):
* platform/FileSystem.h:
* platform/FileStream.cpp:
(WebCore::FileStream::getSize):
* platform/chromium/support/WebHTTPBody.cpp:
(WebKit::WebHTTPBody::elementAt):
* platform/network/BlobData.h:
(WebCore::BlobDataItem::doNotCheckFileChange): Removed.
* platform/network/BlobData.cpp:
* platform/network/cf/FormDataStreamCFNet.cpp:
(WebCore::advanceCurrentStream):
* platform/network/soup/ResourceHandleSoup.cpp:
(WebCore::blobIsOutOfDate):
2012-06-13 Eli Fidler <efidler@rim.com>
Fix CMake build with SMOOTH_SCROLLING enabled
https://bugs.webkit.org/show_bug.cgi?id=89032
Reviewed by Antonio Gomes.
SMOOTH_SCROLLING pulls in ScrollAnimatorNone.cpp, which has
dependencies on platform/ActivePlatformGestureAnimation.cpp and
platform/TouchpadFlingPlatformGestureCurve.cpp
* CMakeLists.txt:
2012-06-13 Mihai Balan <mibalan@adobe.com>
Touched CSSPropertyNames.in to trigger a proper rebuild
[CSS Regions] Adding feature defines for CSS Regions & exclusions for Windows
https://bugs.webkit.org/show_bug.cgi?id=88645
Reviewed by Sam Weinig.
Previous work on #87442 added feature defines to allow proper disabling of CSS Regions and
exclusions (previously even with the feature disabled, parsing still worked). However, the
feature defines were enabled on Mac only - they were broken on Windows. This
patch adds the feature defines to (re)enable regions and exclusions on Windows.
* css/CSSPropertyNames.in: Touched to trigger a proper rebuild. Incremental builds were broken on Windows.
2012-06-13 Arnaud Renevier <arno@renevier.net>
make sure headers are included only once per file
https://bugs.webkit.org/show_bug.cgi?id=88931
Reviewed by Darin Adler.
* Modules/indexeddb/IDBDatabase.cpp:
* WebCorePrefix.h:
* bindings/js/JSHTMLAllCollectionCustom.cpp:
* bindings/v8/V8Utilities.cpp:
* bindings/v8/custom/V8LocationCustom.cpp:
* bridge/NP_jsobject.cpp:
* editing/Editor.cpp:
* html/HTMLMediaElement.cpp:
* platform/graphics/cairo/PlatformContextCairo.cpp:
* rendering/InlineFlowBox.cpp:
2012-06-13 MORITA Hajime <morrita@google.com>
Unreviewed typo fix.
* platform/RefCountedSupplement.h:
2012-06-13 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r120268.
http://trac.webkit.org/changeset/120268
https://bugs.webkit.org/show_bug.cgi?id=89060
WebCompositor::setPerTilePaintingEnabled hits an assertion in
DEBUG (Requested by dominicc|work on #webkit).
* WebCore.gypi:
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(WebCore::Settings::setPerTileDrawingEnabled):
(WebCore::Settings::perTileDrawingEnabled):
(Settings):
(WebCore::Settings::setPartialSwapEnabled):
(WebCore::Settings::partialSwapEnabled):
(WebCore::Settings::setThreadedAnimationEnabled):
(WebCore::Settings::threadedAnimationEnabled):
* platform/graphics/chromium/ContentLayerChromium.cpp:
(WebCore::ContentLayerChromium::createTextureUpdaterIfNeeded):
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::addAnimation):
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::initialize):
* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::TiledLayerChromium::updateTileSizeAndTilingOption):
* platform/graphics/chromium/cc/CCDebugRectHistory.cpp:
(WebCore::CCDebugRectHistory::enabled):
(WebCore::CCDebugRectHistory::saveDebugRectsForCurrentFrame):
* platform/graphics/chromium/cc/CCDebugRectHistory.h:
(WebCore):
(CCDebugRectHistory):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
(WebCore::CCHeadsUpDisplay::enabled):
(WebCore::CCHeadsUpDisplay::showPlatformLayerTree):
(WebCore::CCHeadsUpDisplay::showDebugRects):
(WebCore::CCHeadsUpDisplay::draw):
(WebCore::CCHeadsUpDisplay::drawHudContents):
(WebCore::CCHeadsUpDisplay::drawDebugRects):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.h:
(WebCore):
(CCHeadsUpDisplay):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::create):
(WebCore::CCLayerTreeHost::CCLayerTreeHost):
(WebCore::CCLayerTreeHost::animateLayers):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCSettings::CCSettings):
(CCSettings):
(CCLayerTreeHost):
(WebCore::CCLayerTreeHost::settings):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::create):
(WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::calculateRenderSurfaceLayerList):
(WebCore::CCLayerTreeHostImpl::animateLayers):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCRenderer.h:
(CCRendererClient):
(WebCore::CCRenderer::settings):
* platform/graphics/chromium/cc/CCSettings.cpp: Removed.
* platform/graphics/chromium/cc/CCSettings.h: Removed.
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::setPerTileDrawingEnabled):
(WebCore):
* testing/InternalSettings.h:
(InternalSettings):
2012-06-13 Ian Vollick <vollick@chromium.org>
[chromium] Certain settings in CCSettings could be global
https://bugs.webkit.org/show_bug.cgi?id=88384
Reviewed by James Robinson.
No new tests. No new functionality.
* WebCore.gypi:
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(Settings):
* platform/graphics/chromium/ContentLayerChromium.cpp:
(WebCore::ContentLayerChromium::createTextureUpdaterIfNeeded):
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::addAnimation):
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::initialize):
* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::TiledLayerChromium::updateTileSizeAndTilingOption):
* platform/graphics/chromium/cc/CCDebugRectHistory.cpp:
(WebCore::CCDebugRectHistory::enabled):
(WebCore::CCDebugRectHistory::saveDebugRectsForCurrentFrame):
* platform/graphics/chromium/cc/CCDebugRectHistory.h:
(WebCore):
(CCDebugRectHistory):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
(WebCore::CCHeadsUpDisplay::enabled):
(WebCore::CCHeadsUpDisplay::showPlatformLayerTree):
(WebCore::CCHeadsUpDisplay::showDebugRects):
(WebCore::CCHeadsUpDisplay::draw):
(WebCore::CCHeadsUpDisplay::drawHudContents):
(WebCore::CCHeadsUpDisplay::drawDebugRects):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.h:
(WebCore):
(CCHeadsUpDisplay):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::create):
(WebCore::CCLayerTreeHost::CCLayerTreeHost):
(WebCore::CCLayerTreeHost::animateLayers):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCLayerTreeSettings::CCLayerTreeSettings):
(CCLayerTreeSettings):
(CCLayerTreeHost):
(WebCore::CCLayerTreeHost::settings):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::create):
(WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::calculateRenderSurfaceLayerList):
(WebCore::CCLayerTreeHostImpl::animateLayers):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCRenderer.h:
(CCRendererClient):
(WebCore::CCRenderer::settings):
* platform/graphics/chromium/cc/CCSettings.cpp: Added.
(WebCore):
(WebCore::CCSettings::perTilePaintingEnabled):
(WebCore::CCSettings::setPerTilePaintingEnabled):
(WebCore::CCSettings::partialSwapEnabled):
(WebCore::CCSettings::setPartialSwapEnabled):
(WebCore::CCSettings::acceleratedAnimationEnabled):
(WebCore::CCSettings::setAcceleratedAnimationEnabled):
(WebCore::CCSettings::reset):
* platform/graphics/chromium/cc/CCSettings.h: Added.
(WebCore):
(CCSettings):
* testing/InternalSettings.cpp:
* testing/InternalSettings.h:
(InternalSettings):
2012-06-13 Crystal Zhang <haizhang@rim.com>
[BlackBerry] Apply New Style and JS for Select Popup
https://bugs.webkit.org/show_bug.cgi?id=89020
Reviewed by Rob Buis.
PR 162854
Use new python script for data generating, as the old one is too simple and problematic.
* PlatformBlackBerry.cmake:
2012-06-13 Dana Jansens <danakj@chromium.org>
[chromium] Assert if iterating an invalid RenderSurfaceLayerList, where a layer in the list has no RenderSurface
https://bugs.webkit.org/show_bug.cgi?id=89004
Reviewed by Adrienne Walker.
A RenderSurfaceLayerList expects that all layers in the list own a
RenderSurface. If an invalid RSLL is iterated over, the
CCLayerIterator class will now ASSERT in debug mode, as well as
considering the list empty in release mode.
We will be adding code to CCLayerTreeHostImpl to save a RSLL across
frames, so adding a clearRenderSurfaces() method with a FIXME comment
to make it clear that we should clear the RSLL when we remove
RenderSurfaces from the layers in the saved RSLL.
* platform/graphics/chromium/cc/CCLayerIterator.h:
(WebCore::CCLayerIterator::CCLayerIterator):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::~CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::initializeLayerRenderer):
(WebCore::clearRenderSurfacesOnCCLayerImplRecursive):
(WebCore::CCLayerTreeHostImpl::clearRenderSurfaces):
(WebCore):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(CCLayerTreeHostImpl):
2012-06-13 Yael Aharon <yael.aharon@nokia.com>
Remove redundant code from RenderView and RenderBlock
https://bugs.webkit.org/show_bug.cgi?id=89019
Reviewed by Abhishek Arya.
As pointed out in https://bugs.webkit.org/show_bug.cgi?id=83981#c7 , RenderView::insertFixedPositionedObject()
is redundant and should be removed.
No new tests.
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::insertPositionedObject):
(WebCore::RenderBlock::removePositionedObject):
* rendering/RenderView.cpp:
(WebCore):
* rendering/RenderView.h:
(RenderView):
2012-06-13 Dana Jansens <danakj@chromium.org>
[chromium] Create a CCStreamVideoDrawQuad used for StreamTexture video output
https://bugs.webkit.org/show_bug.cgi?id=88826
Reviewed by Adrienne Walker.
Remove the matrix from the CCVideoDrawQuad, and create a new quad
specifically for StreamTexture video decoder output. This quad
contains only the fields needed for display - textureId and the
matrix to be applied to texture coords.
Removes the (0, 0, 1, 1) "transform" from the shader used for these
quads, as it was a no-op.
No new tests, no change in behaviour.
* WebCore.gypi:
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::drawQuad):
(WebCore::LayerRendererChromium::drawStreamVideoQuad):
(WebCore::LayerRendererChromium::drawVideoQuad):
* platform/graphics/chromium/LayerRendererChromium.h:
(WebCore):
(LayerRendererChromium):
* platform/graphics/chromium/ShaderChromium.cpp:
(WebCore::VertexShaderVideoTransform::VertexShaderVideoTransform):
(WebCore::VertexShaderVideoTransform::init):
(WebCore::VertexShaderVideoTransform::getShaderString):
* platform/graphics/chromium/ShaderChromium.h:
(VertexShaderVideoTransform):
* platform/graphics/chromium/cc/CCDrawQuad.cpp:
(WebCore::CCDrawQuad::toStreamVideoDrawQuad):
(WebCore):
* platform/graphics/chromium/cc/CCDrawQuad.h:
(WebCore):
(CCDrawQuad):
* platform/graphics/chromium/cc/CCStreamVideoDrawQuad.cpp: Added.
(WebCore):
(WebCore::CCStreamVideoDrawQuad::create):
(WebCore::CCStreamVideoDrawQuad::CCStreamVideoDrawQuad):
* platform/graphics/chromium/cc/CCStreamVideoDrawQuad.h: Added.
(WebCore):
(CCStreamVideoDrawQuad):
(WebCore::CCStreamVideoDrawQuad::textureId):
(WebCore::CCStreamVideoDrawQuad::matrix):
* platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
(WebCore::CCVideoLayerImpl::appendQuads):
2012-06-13 Martin Robinson <mrobinson@igalia.com>
[GTK] Add an accelerated compositing implementation for WebKit2
https://bugs.webkit.org/show_bug.cgi?id=86037
Reviewed by Alejandro G. Castro.
No new tests. These changes are covered by the compositing tests, which
we will gradually try to turn on for the bots.
Add a new GLContext method for getting to current default framebuffer size, which
is quite useful for preventing composited content quivering while resizing.
* platform/graphics/cairo/GLContext.h:
* platform/graphics/glx/GLContextGLX.cpp:
(WebCore::GLContextGLX::defaultFrameBufferSize): Added this implementation for X11
which asks the X server for the window size.
* platform/graphics/glx/GLContextGLX.h:
2012-06-13 Shawn Singh <shawnsingh@chromium.org>
[chromium] Implement hit-testing for impl-side input handling in accelerated compositor
https://bugs.webkit.org/show_bug.cgi?id=88972
Reviewed by Adrienne Walker.
The original hit testing code by Sami Kyostila.
This patch implements a simple impl-side hit-testing helper
function. At this time, the main motivation for this is so that
composited scrollable layers can be scrolled by the impl thread --
and a prerequisite for that is the ability to hit-test on the impl
thread, to determine where the scrolling action should be applied.
To implement the actual hit test, it's possible to compute the hit
test in two equivalent ways: (a) inverse-project the point into
local space, or (b) forward-transform the rect to screen space
It's not obvious which one would perform better, or which one
would be more robust to w < 0 homogeneous clipping issues. For
this patch, we simply opt for the inverse-projection approach.
10 unit tests added to CCLayerTreeHostCommonTest.cpp:
CCLayerTreeHostCommonTest.verifyHitTestingForEmptyLayerList
CCLayerTreeHostCommonTest.verifyHitTestingForSingleLayer
CCLayerTreeHostCommonTest.verifyHitTestingForSinglePositionedLayer
CCLayerTreeHostCommonTest.verifyHitTestingForSingleRotatedLayer
CCLayerTreeHostCommonTest.verifyHitTestingForSinglePerspectiveLayer
CCLayerTreeHostCommonTest.verifyHitTestingForSingleLayerWithScaledContents
CCLayerTreeHostCommonTest.verifyHitTestingForSimpleClippedLayer
CCLayerTreeHostCommonTest.verifyHitTestingForMultiClippedRotatedLayer
CCLayerTreeHostCommonTest.verifyHitTestingForMultipleLayers
CCLayerTreeHostCommonTest.verifyHitTestingForMultipleLayerLists
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::pointHitsRect):
(WebCore):
(WebCore::pointIsClippedBySurfaceOrClipRect):
(WebCore::CCLayerTreeHostCommon::findLayerThatIsHitByPoint):
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.h:
(CCLayerTreeHostCommon):
* platform/graphics/chromium/cc/CCMathUtil.cpp:
(WebCore::projectHomogeneousPoint):
(WebCore::mapHomogeneousPoint):
(WebCore::CCMathUtil::mapClippedRect):
(WebCore::CCMathUtil::projectClippedRect):
(WebCore::CCMathUtil::mapClippedQuad):
(WebCore::CCMathUtil::mapQuad):
(WebCore::CCMathUtil::projectPoint):
(WebCore):
* platform/graphics/chromium/cc/CCMathUtil.h:
(WebCore):
(CCMathUtil):
2012-06-13 Charles Wei <charles.wei@torchmobile.com.cn>
Need customized toDOMStringList for IndexedDB JSC binding.
https://bugs.webkit.org/show_bug.cgi?id=87965
Reviewed by Kentaro Hara.
IndexedDB database interface uses DOMStringList as the input parameter.
The generated toDOMStringList() can't convert an JSArray of Strings to
native DOMStringList. We need a custom function to do this.
Because this patch adds one new file (JSDOMStringList.cpp), all build systems
are updated accordingly to pick up this file.
Tests: Corresponding bindings/scripts/test/JS files have been updated accordingly.
* GNUmakefile.list.am:
* Target.pri:
* UseJSC.cmake:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSBindingsAllInOne.cpp:
* bindings/js/JSDOMStringListCustom.cpp: Added.
(WebCore):
(WebCore::toDOMStringList):
* bindings/scripts/CodeGeneratorJS.pm:
(IndexGetterReturnsStrings):
(GenerateHeader):
(GetNativeType):
(GetNativeTypeForCallbacks):
(JSValueToNative):
* bindings/scripts/test/JS/JSTestCallback.cpp:
(WebCore::JSTestCallback::callbackWithStringList):
* bindings/scripts/test/JS/JSTestCallback.h:
(JSTestCallback):
* bindings/scripts/test/JS/JSTestObj.cpp:
(WebCore::jsTestObjPrototypeFunctionOverloadedMethod6):
(WebCore::jsTestObjPrototypeFunctionOverloadedMethod7):
* dom/DOMStringList.idl:
2012-06-13 Jeffrey Pfau <jpfau@apple.com>
Padding and borders can cause integer overflow in block layouts
https://bugs.webkit.org/show_bug.cgi?id=88820
<rdar://problem/11328762>
Reviewed by Tony Chang.
Tests: fast/block/block-size-integer-overflow.html
fast/flexbox/box-size-integer-overflow.html
fast/table/table-size-integer-overflow.html
* rendering/AutoTableLayout.cpp: Decreased max int.
(WebCore::AutoTableLayout::computePreferredLogicalWidths):
* rendering/FixedTableLayout.cpp: Use shared constant.
(WebCore::FixedTableLayout::computePreferredLogicalWidths):
* rendering/RenderBlock.cpp: Removed unused constant.
* rendering/TableLayout.h: Add shared constant.
(TableLayout):
2012-06-13 Gregg Tavares <gman@google.com>
Refactor WebGLFramebuffer to handle texture attachments
https://bugs.webkit.org/show_bug.cgi?id=88697
Reviewed by Adrienne Walker.
In order to add support for depth textures WebGLFramebuffer
needed to be refactored to handle textures as attachments.
No new tests as no new functionality.
* html/canvas/WebGLFramebuffer.cpp:
(WebCore::WebGLFramebuffer::WebGLAttachment::WebGLAttachment):
(WebCore):
(WebCore::WebGLFramebuffer::WebGLAttachment::~WebGLAttachment):
(WebCore::WebGLFramebuffer::WebGLFramebuffer):
(WebCore::WebGLFramebuffer::setAttachmentForBoundFramebuffer):
(WebCore::WebGLFramebuffer::attach):
(WebCore::WebGLFramebuffer::getAttachmentObject):
(WebCore::WebGLFramebuffer::getAttachment):
(WebCore::WebGLFramebuffer::removeAttachmentFromBoundFramebuffer):
(WebCore::WebGLFramebuffer::getColorBufferWidth):
(WebCore::WebGLFramebuffer::getColorBufferHeight):
(WebCore::WebGLFramebuffer::getColorBufferFormat):
(WebCore::WebGLFramebuffer::checkStatus):
(WebCore::WebGLFramebuffer::onAccess):
(WebCore::WebGLFramebuffer::hasStencilBuffer):
(WebCore::WebGLFramebuffer::deleteObjectImpl):
(WebCore::WebGLFramebuffer::initializeAttachments):
* html/canvas/WebGLFramebuffer.h:
(WebGLAttachment):
(WebGLFramebuffer):
* html/canvas/WebGLRenderingContext.cpp:
(WebCore):
(WebCore::WebGLRenderingContext::framebufferRenderbuffer):
(WebCore::WebGLRenderingContext::reattachDepthStencilAttachments):
(WebCore::WebGLRenderingContext::framebufferTexture2D):
(WebCore::WebGLRenderingContext::getFramebufferAttachmentParameter):
* html/canvas/WebGLRenderingContext.h:
(WebGLRenderingContext):
2012-06-13 Konrad Piascik <kpiascik@rim.com>
Remove return statement in void method
https://bugs.webkit.org/show_bug.cgi?id=88998
Reviewed by Andreas Kling
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::collectStyleForAttribute):
2012-06-13 Benjamin Poulain <benjamin@webkit.org>
Implement the simple constructors WTFURL's KURL
https://bugs.webkit.org/show_bug.cgi?id=85724
Reviewed by Adam Barth.
This patch implement the correct initialization of KURL from an absolute URL.
The URL correctness is also relaxed in order to have KURLWTFURL works similarily to KURL.
* platform/KURLWTFURL.cpp:
(WebCore::KURL::KURL):
(WebCore::KURL::isValid):
(WebCore::KURL::string):
(WebCore::KURL::protocol):
(WebCore::KURL::host):
(WebCore::KURL::port): Fix a bug where invalidPortNumber was returned if no port was defined.
(WebCore::KURL::hasFragmentIdentifier):
(WebCore::KURL::fragmentIdentifier):
(WebCore::KURL::protocolIs):
(WebCore::KURL::removeFragmentIdentifier):
(WebCore):
(WebCore::KURL::print):
* platform/cf/KURLCFNet.cpp:
(WebCore::KURL::KURL):
(WebCore::KURL::createCFURL):
(WebCore):
* platform/mac/KURLMac.mm:
(WebCore::KURL::KURL):
(WebCore::KURL::createCFURL):
(WebCore):
Use the code of KURL to make the conversion from WTFURL to NSURL.
2012-06-12 Vincent Scheib <scheib@chromium.org>
Remove duplicate webkitpointerlockchange events when unlocking.
https://bugs.webkit.org/show_bug.cgi?id=88917
Reviewed by Dimitri Glazkov.
Subsequent patches adding the new API for entering and exiting
pointer lock would be burdened by handling multiple change events
when the lock target is changed. This change introduces
temporary code to suppress duplicate change events.
Existing tests simplified by removing the duplicate change events.
* page/PointerLockController.cpp:
(WebCore::PointerLockController::requestPointerLock):
(WebCore::PointerLockController::didLosePointerLock):
* page/PointerLockController.h:
(PointerLockController):
2012-06-13 Silvia Pfeiffer <silviapf@chromium.org>
Update range sliders rendering for volume and playback position of new Chrome video controls.
https://bugs.webkit.org/show_bug.cgi?id=88623
Reviewed by Eric Carlson.
No new tests, final patch will contain the rebaselined tests.
The Chrome video controls are receiving a visual update. This patch contains updates to the
rendering of the range sliders for the playback position and the volume. This includes the
creation of a shadowPseudoId to be able to style the range sliders via CSS, the rendering
of the slider background and the highlighted ranges.
* css/mediaControlsChromium.css:
(input[type="range"]::-webkit-media-slider-container):
Adjust the styling of the slider container backgrounds.
(input[type="range"]::-webkit-media-slider-thumb):
Adjust the styling of the slider thumbs.
* html/shadow/SliderThumbElement.cpp:
(WebCore::sliderThumbShadowPseudoId):
Add an accessor function for the slider thumb shadowPseudoId.
(WebCore::mediaSliderThumbShadowPseudoId):
Add an accessor function for the media slider thumb shadowPseudoId.
(WebCore::SliderThumbElement::shadowPseudoId):
Add a different CSS pseudo-class name to the media slider thumbs.
(WebCore::TrackLimiterElement::shadowPseudoId):
Add a different CSS pseudo-class name to the media slider thumbs.
(WebCore::SliderContainerElement::shadowPseudoId):
Add a different CSS pseudo-class name to the media slider background container.
* rendering/RenderMediaControlsChromium.cpp:
(WebCore::paintRoundedSliderBackground):
Add function to draw rounded background of sliders in a dark color.
(WebCore::paintSliderRangeHighlight):
Add function to draw a highlighted grey region on the sliders, which is rounded only at
its ends.
(WebCore::paintMediaSlider):
Update the background and buffered range renderings on the playback position slider by making
use of the new helper functions.
(WebCore::paintMediaVolumeSlider):
Update the background and buffered range renderings on the volume slider by making
use of the new helper functions.
2012-06-13 Christopher Hutten-Czapski <chutten@rim.com>
2012-06-13 Christopher Hutten-Czapski <chutten@rim.com>
[BlackBerry] Update Select Popup's styling
https://bugs.webkit.org/show_bug.cgi?id=89005
Reviewed by Rob Buis.
Add in some nicely-styled CSS and JS for the BlackBerry port's <select>
* Resources/blackberry/popupControlBlackBerry.css:
(body):
(.popup-area):
(.popup-area>:first-child):
(.popup-buttons):
(.popup-button-divider):
(.popup-button):
(.popup-button:active):
(.popup-button:first-child:active):
(.popup-button:last-child:active):
(.popup-button:only-child:active):
* Resources/blackberry/selectControlBlackBerry.css: Added.
(.select-area):
(.select-area:last-child):
(.option):
(.optgroup:last-child):
(.selected):
(.optgroup):
(.optgroup .option):
* Resources/blackberry/selectControlBlackBerry.js: Added.
2012-06-13 Alec Flett <alecflett@chromium.org>
[V8] IndexedDB: Cursor value modifications should be preserved until cursor iterates
https://bugs.webkit.org/show_bug.cgi?id=83526
Reviewed by Kentaro Hara.
Cache the 'value' attribute of IDBCursorWithValue with policy
determined by IDBCursor.cpp, to follow spec behavior of keeping a
consistent script object until the cursor advances. See
http://www.w3.org/TR/IndexedDB/#widl-IDBCursorWithValueSync-value
for details.
Test: storage/indexeddb/cursor-value.html
* Modules/indexeddb/IDBCursor.cpp:
(WebCore::IDBCursor::IDBCursor):
(WebCore::IDBCursor::value):
(WebCore::IDBCursor::setValueReady):
* Modules/indexeddb/IDBCursor.h:
(IDBCursor):
(WebCore::IDBCursor::valueIsDirty):
* Modules/indexeddb/IDBCursorWithValue.idl:
* WebCore.gypi:
* bindings/v8/IDBCustomBindings.cpp: Added.
(WebCore):
(WebCore::V8IDBCursorWithValue::valueAccessorGetter):
2012-06-13 Silvia Pfeiffer <silviapf@chromium.org>
Code cleanup from bug 88881 to share the SliderVerticalPart code.
https://bugs.webkit.org/show_bug.cgi?id=88930
Reviewed by Eric Carlson.
No new tests - this is just code cleanup.
* html/shadow/SliderThumbElement.cpp:
(WebCore::hasVerticalAppearance):
Avoid having two ifdef-ed returns in the code depending on ENABLE(VIDEO).
2012-06-13 Mark Mentovai <mark@chromium.org>
[chromium mac] Don't include things in subframeworks of
ApplicationServices.framework.
A CoreText #include (with quotes instead of brackets, even) sprung up
in r120506, bug 69826.
https://bugs.webkit.org/show_bug.cgi?id=88569
Reviewed by Stephen White.
No new tests.
* platform/graphics/harfbuzz/ng/HarfBuzzFaceCoreText.cpp:
2012-06-13 Eriq Augustine <eaugusti@chromium.org>
Removing an extra HandleScope in V8Proxy::evaluateInIsolatedWorld
https://bugs.webkit.org/show_bug.cgi?id=88451
Reviewed by Adam Barth.
Any values retuned by the evaluation will get destroyed with the extra
HandleScope.
I was unable to cause a test failure, however this feature is tested by:
platform/chromium/http/tests/misc/execute-and-return-value.html
* bindings/v8/V8Proxy.cpp:
(WebCore::V8Proxy::evaluateInIsolatedWorld):
2012-06-13 Zan Dobersek <zandobersek@gmail.com>
[Gtk] Enable link prefetch support in the developer builds
https://bugs.webkit.org/show_bug.cgi?id=89011
Reviewed by Martin Robinson.
Handle the ENABLE_LINK_PREFETCH conditional in standard way.
No new tests - covered by existing link prefetch tests.
* GNUmakefile.am:
2012-06-13 Amy Ousterhout <aousterh@chromium.org>
Rename currentDeviceMotion to lastMotion in DeviceMotionClient
https://bugs.webkit.org/show_bug.cgi?id=88854
Reviewed by Adam Barth.
Rename the function currentDeviceMotion to lastMotion in DeviceMotionClient.
This makes it consistent with the similar function lastOrientation in DeviceOrientationClient.
Just a rename, so does not require any new tests.
* dom/DeviceMotionClient.h:
(DeviceMotionClient):
* dom/DeviceMotionController.cpp:
(WebCore::DeviceMotionController::timerFired):
(WebCore::DeviceMotionController::addListener):
* loader/EmptyClients.h:
(WebCore::EmptyDeviceMotionClient::lastMotion):
* platform/qt/DeviceMotionClientQt.cpp:
(WebCore::DeviceMotionClientQt::lastMotion):
* platform/qt/DeviceMotionClientQt.h:
(DeviceMotionClientQt):
* platform/qt/DeviceMotionProviderQt.h:
(WebCore::DeviceMotionProviderQt::lastMotion):
2012-06-13 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: Breakpoint's _uiLocationRemoved() method is never called.
https://bugs.webkit.org/show_bug.cgi?id=89003
Reviewed by Pavel Feldman.
* inspector/front-end/BreakpointManager.js:
(WebInspector.BreakpointManager.prototype.breakpointLocationsForUISourceCode):
(WebInspector.BreakpointManager.Breakpoint):
(WebInspector.BreakpointManager.Breakpoint.prototype._locationUpdated):
(WebInspector.BreakpointManager.Breakpoint.prototype._resetLocations):
(WebInspector.BreakpointManager.Breakpoint.prototype._fakeBreakpointAtPrimaryLocation):
* inspector/front-end/FileSystemModel.js: Drive by compilation fix.
(WebInspector.FileSystemModel):
2012-06-13 David Grogan <dgrogan@chromium.org>
IndexedDB: Add tests, an assert, and some comments around open/close/setVersion call sequencing
https://bugs.webkit.org/show_bug.cgi?id=88829
Reviewed by Tony Chang.
Test: storage/indexeddb/three-setversion-calls.html
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::processPendingCalls):
2012-06-09 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: Errors when pending console messages are attached to JavaScriptSourceFrame.
https://bugs.webkit.org/show_bug.cgi?id=88712
Reviewed by Pavel Feldman.
* inspector/front-end/PresentationConsoleMessageHelper.js:
(WebInspector.PresentationConsoleMessageHelper):
(WebInspector.PresentationConsoleMessageHelper.prototype._addPendingConsoleMessage):
(WebInspector.PresentationConsoleMessageHelper.prototype._parsedScriptSource):
2012-06-13 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: Support user attributes in DOMNode
https://bugs.webkit.org/show_bug.cgi?id=88706
Reviewed by Pavel Feldman.
This change allows clients to store arbitrary properties on DOMNodes and know if there are descendants
(and how many of them) having a certain property set.
Test: inspector/elements/user-attributes.html
* inspector/front-end/DOMAgent.js:
(WebInspector.DOMNode.prototype._removeChild):
(WebInspector.DOMNode.prototype._updateChildUserPropertyCountsOnRemoval):
(WebInspector.DOMNode.prototype._updateDescendantUserPropertyCount):
(WebInspector.DOMNode.prototype.setUserProperty):
(WebInspector.DOMNode.prototype.removeUserProperty):
(WebInspector.DOMNode.prototype.getUserProperty):
(WebInspector.DOMNode.prototype.descendantUserPropertyCount):
2012-06-13 Vineet Chaudhary <vineet.chaudhary@motorola.com>
REGRESSION:Bindings sequence<T> in Console.idl, Internals.idl and ScriptProfileNode.idl should be T[]
https://bugs.webkit.org/show_bug.cgi?id=84863
Reviewed by Kentaro Hara.
With reference to http://www.w3.org/TR/WebIDL/#idl-sequence
"Sequences must not be used as the type of an attribute, constant or exception field."
So we should use T[] instead of sequence<T>.
Tests: bindings/scripts/test/TestObj.idl
fast/harness/user-preferred-language.html
inspector/profiler/cpu-profiler-profiling-without-inspector.html
media/track/track-language-preference.html
* bindings/scripts/CodeGenerator.pm:
(GetSequenceType): Getter for getting sequence type.
(GetArrayType): Getter for getting array type.
(AssertNotSequenceType): Check sequences must not be used as the type of an attribute,
constant or exception field.
* bindings/scripts/CodeGeneratorCPP.pm:
(SkipFunction): Exclude bindings code for type sequence.
(SkipAttribute): Check for sequences must not be used as the type of an attribute,
constant or exception field.
(AddIncludesForType): Exclude header inclusion code for type array/sequence.
* bindings/scripts/CodeGeneratorGObject.pm:
(SkipAttribute): Check for sequences must not be used as the type of an attribute,
constant or exception field.
(SkipFunction): Exclude bindings code for type sequence.
* bindings/scripts/CodeGeneratorJS.pm:
(AddIncludesForType): Exclude header inclusion code for type array/sequence.
(GenerateImplementation): Check for sequences must not be used as the type of an attribute,
constant or exception field.
(GetNativeType):
(JSValueToNative):
(NativeToJSValue):
* bindings/scripts/CodeGeneratorObjC.pm:
(SkipFunction): Exclude bindings code for type sequence.
(SkipAttribute): Exclude bindings code for type array.
(AddForwardDeclarationsForType):
(AddIncludesForType):
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateNormalAttrGetter):
(GetNativeType):
(JSValueToNative):
(CreateCustomSignature):
(NativeToJSValue):
* bindings/scripts/test/JS/JSTestObj.cpp: Updated code from run-bindings-tests.
(WebCore):
(WebCore::jsTestObjPrototypeFunctionMethodWithSequenceArg):
* bindings/scripts/test/JS/JSTestObj.h: Updated code from run-bindings-tests.
(WebCore):
* bindings/scripts/test/TestObj.idl: Tests.
* bindings/scripts/test/V8/V8TestObj.cpp: Updated code from run-bindings-tests.
(TestObjV8Internal):
(WebCore::TestObjV8Internal::methodWithSequenceArgCallback):
(WebCore):
* inspector/ScriptProfileNode.idl: Using sequence<T> for children().
* page/Console.idl: Using Array[T] for profiles attribute.
* testing/Internals.idl: Using sequence<T> for userPreferredLanguages().
2012-06-13 Simon Hausmann <simon.hausmann@nokia.com>
[Qt] Remove dependency to QtScript for the Qt 5 build
https://bugs.webkit.org/show_bug.cgi?id=88993
Reviewed by Kenneth Rohde Christiansen.
Replace the use of QScriptEngine::ValueOwnership with an enum local
to the class where it is used (QtInstance).
* Target.pri:
* bindings/js/ScriptControllerQt.cpp:
(WebCore::ScriptController::createScriptInstanceForWidget):
* bridge/qt/qt_instance.cpp:
(JSC::Bindings::QtInstance::QtInstance):
(JSC::Bindings::QtInstance::~QtInstance):
(JSC::Bindings::QtInstance::getQtInstance):
* bridge/qt/qt_instance.h:
(QtInstance):
(JSC::Bindings::QtInstance::create):
* bridge/qt/qt_runtime.cpp:
(JSC::Bindings::convertQVariantToValue):
* bridge/qt/qt_runtime_qt4.cpp:
(JSC::Bindings::convertQVariantToValue):
2012-06-13 Caio Marcelo de Oliveira Filho <caio.oliveira@openbossa.org>
[Qt] Save a JSGlobalContextRef instead of a JSContextRef in QtConnectionObject
https://bugs.webkit.org/show_bug.cgi?id=88905
Make QtConnectionObject use JSContextGetGlobalContext and store the
JSGlobalContextRef. Previous code was working but saving JSContextRef was wrong in terms of
API.
There's an assumption that JSGlobalContextRef will outlive the connection objects, and
because of it no JSGlobalContextRetain/Release calls were added.
Reviewed by Noam Rosenthal.
* bridge/qt/qt_runtime.cpp:
(JSC::Bindings::QtRuntimeConnectionMethod::call):
(JSC::Bindings::QtConnectionObject::QtConnectionObject):
* bridge/qt/qt_runtime.h:
(QtConnectionObject):
* bridge/qt/qt_runtime_qt4.cpp:
(JSC::Bindings::QtRuntimeConnectionMethod::call):
(JSC::Bindings::QtConnectionObject::QtConnectionObject):
2012-06-13 David Barr <davidbarr@chromium.org>
Add dpi to css3-images image-resolution
https://bugs.webkit.org/show_bug.cgi?id=85439
Reviewed by Tony Chang.
No new tests; extended fast/css/image-resolution/image-resolution.html
* css/CSSGrammar.y:
* css/CSSParser.cpp:
(WebCore::CSSParser::validUnit):
(WebCore::CSSParser::createPrimitiveNumericValue):
(WebCore::unitFromString):
(WebCore::CSSParser::parseValidPrimitive):
(WebCore::CSSParser::detectNumberToken):
* css/CSSPrimitiveValue.cpp:
(WebCore::isValidCSSUnitTypeForDoubleConversion):
(WebCore::unitCategory):
(WebCore::conversionToCanonicalUnitsScaleFactor):
(WebCore::CSSPrimitiveValue::customCssText):
(WebCore::CSSPrimitiveValue::cloneForCSSOM):
* css/CSSPrimitiveValue.h:
2012-06-13 Li Yin <li.yin@intel.com>
FileAPI: type should be converted to be lower case in Blob constructor.
https://bugs.webkit.org/show_bug.cgi?id=88696
Reviewed by Kentaro Hara.
Spec: http://dev.w3.org/2006/webapi/FileAPI/#dfn-BlobPropertyBag
Spec: http://dev.w3.org/2006/webapi/FileAPI/#attributes-blob
The ASCII-encoded string in lower case representing the media type of the Blob.
Test: fast/files/blob-constructor.html
* bindings/js/JSBlobCustom.cpp:
(WebCore::JSBlobConstructor::constructJSBlob):
* bindings/v8/custom/V8BlobCustom.cpp:
(WebCore::V8Blob::constructorCallback):
2012-06-13 Robin Cao <robin.cao@torchmobile.com.cn>
[BlackBerry] Enable MEDIA_STREAM by default
https://bugs.webkit.org/show_bug.cgi?id=88849
Reviewed by Antonio Gomes.
Covered by existing test cases.
* CMakeLists.txt:
* PlatformBlackBerry.cmake:
* platform/mediastream/blackberry/DeprecatedPeerConnectionHandler.cpp: Added.
(WebCore):
(WebCore::DeprecatedPeerConnectionHandler::create):
(WebCore::DeprecatedPeerConnectionHandler::DeprecatedPeerConnectionHandler):
(WebCore::DeprecatedPeerConnectionHandler::~DeprecatedPeerConnectionHandler):
(WebCore::DeprecatedPeerConnectionHandler::produceInitialOffer):
(WebCore::DeprecatedPeerConnectionHandler::handleInitialOffer):
(WebCore::DeprecatedPeerConnectionHandler::processSDP):
(WebCore::DeprecatedPeerConnectionHandler::processPendingStreams):
(WebCore::DeprecatedPeerConnectionHandler::sendDataStreamMessage):
(WebCore::DeprecatedPeerConnectionHandler::stop):
* platform/mediastream/blackberry/MediaStreamCenterBlackBerry.cpp: Added.
(WebCore):
(WebCore::MediaStreamCenter::instance):
(WebCore::MediaStreamCenterBlackBerry::MediaStreamCenterBlackBerry):
(WebCore::MediaStreamCenterBlackBerry::~MediaStreamCenterBlackBerry):
(WebCore::MediaStreamCenterBlackBerry::queryMediaStreamSources):
(WebCore::MediaStreamCenterBlackBerry::didSetMediaStreamTrackEnabled):
(WebCore::MediaStreamCenterBlackBerry::didStopLocalMediaStream):
(WebCore::MediaStreamCenterBlackBerry::didCreateMediaStream):
(WebCore::MediaStreamCenterBlackBerry::constructSDP):
* platform/mediastream/blackberry/MediaStreamCenterBlackBerry.h: Added.
(WebCore):
(MediaStreamCenterBlackBerry):
2012-06-13 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: working copy should be committed even if domain binding returns error.
https://bugs.webkit.org/show_bug.cgi?id=88884
Reviewed by Vsevolod Vlasov.
Committing working copy is about saving file. It should always succeed.
* inspector/front-end/UISourceCode.js:
(WebInspector.UISourceCode.prototype.commitWorkingCopy):
2012-06-12 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: show revisions of all modified files in local modifications view
https://bugs.webkit.org/show_bug.cgi?id=88864
Reviewed by Vsevolod Vlasov.
This change allows all local modifications to be seen in a single view. View is now
singleton, listening to all revisions being added. Revert captions were reworked
(see the screenshot).
* English.lproj/localizedStrings.js:
* inspector/front-end/Resource.js:
* inspector/front-end/RevisionHistoryView.js:
(WebInspector.RevisionHistoryView.populateRevisions):
(WebInspector.RevisionHistoryView):
(WebInspector.RevisionHistoryView.showHistory):
(WebInspector.RevisionHistoryView.reset):
(WebInspector.RevisionHistoryView.prototype._createResourceItem.clearHistory):
(WebInspector.RevisionHistoryView.prototype._createResourceItem):
(WebInspector.RevisionHistoryView.prototype._revisionAdded.get if):
(WebInspector.RevisionHistoryView.prototype._revisionAdded):
(WebInspector.RevisionHistoryView.prototype._revealResource.get if):
(WebInspector.RevisionHistoryView.prototype._revealResource):
(WebInspector.RevisionHistoryView.prototype._reset):
(WebInspector.RevisionHistoryTreeElement):
(WebInspector.RevisionHistoryTreeElement.prototype.onattach):
(WebInspector.RevisionHistoryTreeElement.prototype.oncollapse):
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype._reset):
(WebInspector.ScriptsPanel.prototype._editorSelected):
* inspector/front-end/revisionHistory.css:
(.revision-history-link-row):
(.revision-history-drawer li.revision-history-revision):
2012-06-13 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: Make "Go to source and line" possible with "go to file" dialog
https://bugs.webkit.org/show_bug.cgi?id=88740
Reviewed by Vsevolod Vlasov.
Selection dialog now allows search query rewrite and uses line number
suffix to go to a particular line.
* inspector/front-end/FilteredItemSelectionDialog.js:
(WebInspector.FilteredItemSelectionDialog.prototype.onEnter):
(WebInspector.FilteredItemSelectionDialog.prototype._createSearchRegExp):
(WebInspector.SelectionDialogContentProvider.prototype.selectItem):
(WebInspector.SelectionDialogContentProvider.prototype.rewriteQuery):
(WebInspector.JavaScriptOutlineDialog.prototype.selectItem):
(WebInspector.JavaScriptOutlineDialog.prototype.rewriteQuery):
(WebInspector.OpenResourceDialog.prototype.selectItem):
(WebInspector.OpenResourceDialog.prototype.rewriteQuery):
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype.showUISourceCode):
* inspector/front-end/StylesPanel.js:
(WebInspector.StyleSheetOutlineDialog.prototype.selectItem):
(WebInspector.StyleSheetOutlineDialog.prototype.rewriteQuery):
2012-06-12 Hans Wennborg <hans@chromium.org>
Speech JavaScript API: Add test for constructing SpeechRecognitionError events
https://bugs.webkit.org/show_bug.cgi?id=88868
Reviewed by Adam Barth.
This adds a test for constructing SpeechRecognitionError events, and
code to make it work.
Test: fast/events/constructors/speech-recognition-error-constructor.html
* Modules/speech/SpeechRecognitionError.cpp:
(WebCore::SpeechRecognitionError::create):
(WebCore):
(WebCore::SpeechRecognitionError::SpeechRecognitionError):
(WebCore::SpeechRecognitionErrorInit::SpeechRecognitionErrorInit):
* Modules/speech/SpeechRecognitionError.h:
(SpeechRecognitionErrorInit):
(WebCore):
(SpeechRecognitionError):
(WebCore::SpeechRecognitionError::code):
* Modules/speech/SpeechRecognitionError.idl:
2012-06-13 Taiju Tsuiki <tzik@chromium.org>
Implement InspectorFileSystemAgent::readDirectory for FileSystem support.
https://bugs.webkit.org/show_bug.cgi?id=87724
Reviewed by Vsevolod Vlasov.
Test: http/tests/inspector/filesystem/read-directory.html
* inspector/Inspector.json:
* inspector/InspectorController.cpp:
(WebCore::InspectorController::InspectorController):
* inspector/InspectorFileSystemAgent.cpp:
(InspectorFileSystemAgent::FrontendProvider):
(WebCore):
(WebCore::InspectorFileSystemAgent::create):
(WebCore::InspectorFileSystemAgent::readDirectory):
(WebCore::InspectorFileSystemAgent::InspectorFileSystemAgent):
* inspector/InspectorFileSystemAgent.h:
(WebCore):
(InspectorFileSystemAgent):
2012-06-13 Silvia Pfeiffer <silviapf@chromium.org>
Change the styling of the Chromium video controls.
https://bugs.webkit.org/show_bug.cgi?id=88297
Reviewed by Eric Carlson.
No new tests, final patch will contain the rebaselined tests.
The Chrome video controls are receiving a visual update. This patch includes CSS changes,
the removal of the now no longer needed controls background div, and the introduction of
new images for the buttons.
* css/mediaControlsChromium.css:
(audio):
Adjust height of audio controls.
(::-webkit-media-controls):
Use flexbox on media controls shadow DOM.
(audio::-webkit-media-controls-enclosure, video::-webkit-media-controls-enclosure):
Adjust height, max-width, text-indent and box-sizing, and reset padding to 0.
(video::-webkit-media-controls-enclosure):
Use 5px padding only for video controls and adjust height accordingly.
(audio::-webkit-media-controls-panel, video::-webkit-media-controls-panel):
Use flexbox on panel, adjust bottom alignment, height, background color and border radius.
(audio::-webkit-media-controls-mute-button, video::-webkit-media-controls-mute-button):
Display buttons inline with fixed width and height, vertically centered, no padding and
6px space to next controls element (this will be 15px incl the space in the image).
(audio::-webkit-media-controls-play-button, video::-webkit-media-controls-play-button):
Display buttons inline with fixed width and height, vertically centered, no padding and
9px space to next controls element (this will be 15px incl the space in the image).
(audio::-webkit-media-controls-current-time-display, video::-webkit-media-controls-current-time-display):
Display text as block, without broder, adjusted height, font-family, font-size and 9px space
to next controls element (this will be 15px incl the space in the image).
(audio::-webkit-media-controls-timeline, video::-webkit-media-controls-timeline):
Adjust height, padding, background-color, min-wdith and 15px space to next controls element.
(audio::-webkit-media-controls-volume-slider, video::-webkit-media-controls-volume-slider):
Adjust height, max- and min-width, background-color and 15px space to next controls element.
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlRootElementChromium::MediaControlRootElementChromium):
Remove timeline container, which is not necessary any longer because of the use of flexbox.
(WebCore::MediaControlRootElementChromium::create):
Remove timeline container, which is not necessary any longer because of the use of flexbox.
(WebCore::MediaControlRootElementChromium::setMediaController):
Remove timeline container, which is not necessary any longer because of the use of flexbox.
(WebCore::MediaControlRootElementChromium::reset):
Show the timeline instead of the timeline container when resetting.
(WebCore::MediaControlRootElementChromium::reportedError):
Hide the timeline instead of the timeline container when running into an error.
* rendering/RenderMediaControlsChromium.cpp:
(WebCore::paintMediaMuteButton):
Introduce new image resources for the mute button and adjust it based on volume levels.
(WebCore::paintMediaPlayButton):
Introduce new image resources for the play button.
(WebCore::getMediaSliderThumb):
Introduce new image resources for the position slider thumb.
(WebCore::paintMediaVolumeSliderThumb):
Introduce new image resources for the volume slider thumb.
Remove timeline container rendering.
(WebCore::RenderMediaControlsChromium::paintMediaControlsPart):
Add timeline container painting to the "not reached" elements.
(WebCore):
Provide size settings for the position and volume sliders.
(WebCore::RenderMediaControlsChromium::adjustMediaSliderThumbSize):
Calculate correct width and height settings for the new image resources of the position and volume
slider thumbs.
* rendering/RenderThemeChromiumMac.h:
(RenderThemeChromiumMac):
Remove timeline container rendering (i.e. the media controls background).
* rendering/RenderThemeChromiumMac.mm:
Remove timeline container rendering (i.e. the media controls background).
* rendering/RenderThemeChromiumSkia.cpp:
Remove timeline container rendering (i.e. the media controls background).
* rendering/RenderThemeChromiumSkia.h:
(RenderThemeChromiumSkia):
Remove timeline container rendering (i.e. the media controls background).
2012-06-13 Kinuko Yasuda <kinuko@chromium.org>
Unreviewed, fixing English nits suggested by review for r120165.
* fileapi/Blob.cpp:
(WebCore::Blob::webkitSlice):
* fileapi/Blob.h:
(Blob):
2012-06-13 MORITA Hajime <morrita@google.com>
REGRESSION(r118098): <content> element does not render distributed children when cloned from another document
https://bugs.webkit.org/show_bug.cgi?id=88148
Reviewed by Dimitri Glazkov.
The shadow DOM availability was decided through FrameLoaderClient.
But there are documents which don't have any associated frames, in
which any shadow DOM related elements cannot be constructed. That
resulsted the regression.
This change introduces a Page-based client called
ContextFeaturesClient which takes the role from FrameLoaderClient
to decide the feature availability, and ContextFeatures which is a
proxy of ContextFeaturesClient. ContextEnabledFeatures is is
replaced with ContextFeatures.
Each ContextFeatures object is owned by a Page, and is attached to
each Document in the page even if the page itself has no referenct
to Frames or Pages. With ContextFeatures, each Document can decide
the feature availability even if it cannot reach Page or Frame.
On RefCountedSupplement: Document instance can live longer than a
Page where it is shwon. This means that ContextFeatures instance
needs to survive after owner Page destruction because it's
referenced from possibly surviving Documents. RefCountedSupplement
is introduced to cover this scenario: It allows supplement classes
to live after Page's destruction.
RefCountedSupplement::hostDestroyed() is notified when the hosting
page is gone. ContextFeatures clears its reference to the client
using this notification.
Test: fast/dom/shadow/elements-in-frameless-document.html
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateImplementation):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::determineScope):
* dom/ContextFeatures.cpp: Added.
(WebCore):
(WebCore::ContextFeaturesClient::empty):
(WebCore::ContextFeatures::supplementName):
(WebCore::ContextFeatures::defaultSwitch):
(WebCore::ContextFeatures::shadowDOMEnabled):
(WebCore::ContextFeatures::styleScopedEnabled):
(WebCore::ContextFeatures::pagePopupEnabled):
(WebCore::provideContextFeaturesTo):
(WebCore::provideContextFeaturesToDocumentFrom):
* dom/ContextFeatures.h: Added.
(WebCore):
(ContextFeatures):
(WebCore::ContextFeatures::ContextFeatures):
(WebCore::ContextFeatures::hostDestroyed):
(ContextFeaturesClient):
(WebCore::ContextFeaturesClient::~ContextFeaturesClient):
(WebCore::ContextFeaturesClient::isEnabled):
(WebCore::ContextFeatures::create):
(WebCore::ContextFeatures::isEnabled):
* dom/DOMAllInOne.cpp:
* dom/DOMImplementation.cpp:
(WebCore::DOMImplementation::createDocument):
(WebCore::DOMImplementation::createHTMLDocument):
* dom/Document.h:
(WebCore::Document::contextFeatures):
* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::setContextFeatures):
* dom/Position.cpp:
(WebCore::Position::Position):
(WebCore::Position::findParent):
* dom/TreeScope.cpp:
(WebCore::TreeScope::getSelection):
* dom/make_names.pl:
(printConstructorInterior):
(printFactoryCppFile):
(printWrapperFunctions):
(printWrapperFactoryCppFile):
* editing/markup.cpp:
(WebCore::createFragmentFromMarkupWithContext):
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::registerWithScopingNode):
(WebCore::HTMLStyleElement::unregisterWithScopingNode):
* html/shadow/HTMLContentElement.cpp:
(WebCore::contentTagName):
* inspector/DOMPatchSupport.cpp:
(WebCore::DOMPatchSupport::patchDocument):
* loader/FrameLoaderClient.h:
* platform/RefCountedSupplement.h: Copied from Source/WebCore/platform/Supplementable.h.
(WebCore):
(RefCountedSupplement):
(WebCore::RefCountedSupplement::hostDestroyed):
(Wrapper):
(WebCore::RefCountedSupplement::Wrapper::Wrapper):
(WebCore::RefCountedSupplement::Wrapper::~Wrapper):
(WebCore::RefCountedSupplement::Wrapper::wrapped):
(WebCore::RefCountedSupplement::provideTo):
(WebCore::RefCountedSupplement::from):
* platform/Supplementable.h:
(WebCore::Supplement::isRefCountedWrapper):
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::responseXML):
2012-06-11 Kinuko Yasuda <kinuko@chromium.org>
Unprefix Blob.webkitSlice
https://bugs.webkit.org/show_bug.cgi?id=78111
Reviewed by Adam Barth.
Add unprefixed Blob.slice and start showing a deprecation message when webkitSlice is called.
No new tests: existing tests that call webkitSlice() are converted to slice(). They should keep running.
* bindings/gobject/WebKitDOMCustom.cpp:
(webkit_dom_blob_slice): Removed.
(webkit_dom_blob_webkit_slice): Added.
* fileapi/Blob.cpp:
(WebCore::Blob::slice):
(WebCore):
(WebCore::Blob::webkitSlice):
(WebCore::Blob::sliceInternal): Added.
* fileapi/Blob.h:
(WebCore):
(Blob):
* fileapi/Blob.idl:
2012-06-12 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r120036.
http://trac.webkit.org/changeset/120036
https://bugs.webkit.org/show_bug.cgi?id=88965
"assert failed in debug build" (Requested by maryw on
#webkit).
* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::handleNotifyDataReceived):
2012-06-12 Wei James <james.wei@intel.com>
webcore should depend on webcore_arm_neon for only arm target_arch
https://bugs.webkit.org/show_bug.cgi?id=88839
Reviewed by Adam Barth.
* WebCore.gyp/WebCore.gyp:
2012-06-12 Edaena Salinas Jasso <edaena@apple.com>
Can't set el.type on a <button> element
https://bugs.webkit.org/show_bug.cgi?id=14439
Reviewed by Darin Adler.
Test: fast/dom/HTMLButtonElement/change-type.html
* bindings/objc/PublicDOMInterfaces.h:
* html/HTMLButtonElement.cpp:
(WebCore::HTMLButtonElement::setType): Added implementation of setType method.
(WebCore):
* html/HTMLButtonElement.h:
(HTMLButtonElement): Added setType method.
* html/HTMLButtonElement.idl: Removed readonly restriction to the type variable.
2012-06-12 Kenichi Ishibashi <bashi@chromium.org>
[Chromium] Implement font shaping with font-feature-settings on Mac
https://bugs.webkit.org/show_bug.cgi?id=69826
Reviewed by Tony Chang.
Add HarfBuzz-ng text shaper.
Chromium mac port uses it as secondary text shaper to support OpenType features.
HarfBuzz-ng is only used when -webkit-font-feature-settings is specified and
corresponding font is not an AAT font.
No new tests. css3/font-feature-settings-rendering.html should pass on Chromium mac port.
* WebCore.gyp/WebCore.gyp: Added harfbuzz-ng dependencies.
* WebCore.gypi: Added harfbuzz-ng files.
* platform/graphics/FontPlatformData.h:
(FontPlatformData): Added m_harfbuzzFace.
* platform/graphics/cocoa/FontPlatformDataCocoa.mm:
(WebCore::FontPlatformData::platformDataInit): Copy m_harfbuzzFace.
(WebCore::FontPlatformData::platformDataAssign): Ditto.
(WebCore):
(WebCore::isAATFont):
(WebCore::FontPlatformData::harfbuzzFace):
* platform/graphics/harfbuzz/ng/HarfBuzzFace.cpp: Added.
(WebCore):
(WebCore::harfbuzzFaceCache):
(WebCore::HarfBuzzFace::HarfBuzzFace):
(WebCore::HarfBuzzFace::~HarfBuzzFace):
* platform/graphics/harfbuzz/ng/HarfBuzzFace.h: Added.
(WebCore):
(HarfBuzzFace):
(WebCore::HarfBuzzFace::create):
* platform/graphics/harfbuzz/ng/HarfBuzzFaceCoreText.cpp: Added.
(WebCore):
(WebCore::floatToHarfBuzzPosition):
(WebCore::getGlyph):
(WebCore::getGlyphHorizontalAdvance):
(WebCore::getGlyphHorizontalOrigin):
(WebCore::getGlyphExtents):
(WebCore::harfbuzzCoreTextGetFontFuncs):
(WebCore::releaseTableData):
(WebCore::harfbuzzCoreTextGetTable):
(WebCore::HarfBuzzFace::createFace):
(WebCore::HarfBuzzFace::createFont):
(WebCore::HarfBuzzShaper::createGlyphBufferAdvance):
* platform/graphics/harfbuzz/ng/HarfBuzzShaper.cpp: Added.
(WebCore):
(WebCore::harfbuzzPositionToFloat):
(WebCore::HarfBuzzShaper::HarfBuzzRun::HarfBuzzRun):
(WebCore::HarfBuzzShaper::HarfBuzzRun::setGlyphAndPositions):
(WebCore::HarfBuzzShaper::HarfBuzzRun::characterIndexForXPosition):
(WebCore::HarfBuzzShaper::HarfBuzzRun::xPositionForOffset):
(WebCore::HarfBuzzShaper::HarfBuzzShaper):
(WebCore::HarfBuzzShaper::~HarfBuzzShaper):
(WebCore::HarfBuzzShaper::setFontFeatures):
(WebCore::HarfBuzzShaper::shape):
(WebCore::HarfBuzzShaper::setupHarfBuzzRun):
(WebCore::HarfBuzzShaper::shapeHarfBuzzRun):
(WebCore::HarfBuzzShaper::setGlyphPositionsForHarfBuzzRun):
(WebCore::HarfBuzzShaper::offsetForPosition):
(WebCore::HarfBuzzShaper::selectionRect):
* platform/graphics/harfbuzz/ng/HarfBuzzShaper.h: Added.
(WebCore):
(HarfBuzzShaper):
(WebCore::HarfBuzzShaper::totalWidth):
(HarfBuzzRun):
(WebCore::HarfBuzzShaper::HarfBuzzRun::create):
(WebCore::HarfBuzzShaper::HarfBuzzRun::setWidth):
(WebCore::HarfBuzzShaper::HarfBuzzRun::numCharacters):
(WebCore::HarfBuzzShaper::HarfBuzzRun::numGlyphs):
(WebCore::HarfBuzzShaper::HarfBuzzRun::width):
(WebCore::HarfBuzzShaper::HarfBuzzRun::rtl):
* platform/graphics/mac/FontComplexTextMac.cpp:
(WebCore):
(WebCore::preferHarfBuzz): Added.
(WebCore::Font::selectionRectForComplexText): Use HarfBuzzShaper if font-feature-settings exists.
(WebCore::Font::drawComplexText): Ditto.
(WebCore::Font::floatWidthForComplexText): Ditto.
(WebCore::Font::offsetForPositionForComplexText): Ditto.
2012-06-12 Luke Macpherson <macpherson@chromium.org>
WIP: Implement CSS Variables Standard.
https://bugs.webkit.org/show_bug.cgi?id=85580
Reviewed by Ojan Vafai.
Implement CSS Variables Module Level 1. (See http://www.w3.org/TR/css-variables/)
A HashMap of properties is referenced by RenderStyle, and used to store variables as strings.
That HashMap is copy-on-write, and unless new variables are defined is simply a pointer to the
parent's definitions. At usage time the variable's value is parsed according to the CSS property
where it is used.
Variables can:
Define lists of values (ie. entire shorthand values).
Define individual values.
Refer to other variables. (Cycle detection is implemented).
Please see the supplied test cases for example usage.
Missing features:
CSS variables cannot yet be used inside some functions such as -webkit-calc.
Covered by existing test suite under fast/css/variables.
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSGrammar.y:
* css/CSSParser.cpp:
(WebCore):
(WebCore::filterProperties):
(WebCore::CSSParser::createStylePropertySet):
(WebCore::CSSParser::validUnit):
(WebCore::CSSParser::createPrimitiveNumericValue):
(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::storeVariableDeclaration):
(WebCore::CSSParser::detectDashToken):
(WebCore::CSSParser::lex):
* css/CSSParser.h:
(CSSParser):
* css/CSSParserValues.cpp:
(WebCore::CSSParserValue::createCSSValue):
* css/CSSPrimitiveValue.cpp:
(WebCore::isValidCSSUnitTypeForDoubleConversion):
(WebCore::CSSPrimitiveValue::getStringValue):
(WebCore::CSSPrimitiveValue::customSerializeResolvingVariables):
(WebCore):
* css/CSSPrimitiveValue.h:
(CSSPrimitiveValue):
(WebCore::CSSPrimitiveValue::isVariableName):
* css/CSSProperty.cpp:
(WebCore::CSSProperty::isInheritedProperty):
* css/CSSValue.cpp:
(WebCore::CSSValue::serializeResolvingVariables):
(WebCore):
(WebCore::CSSValue::destroy):
* css/CSSValue.h:
(CSSValue):
(WebCore::CSSValue::isVariableValue):
* css/CSSValueList.cpp:
(WebCore):
(WebCore::CSSValueList::customSerializeResolvingVariables):
* css/CSSValueList.h:
(CSSValueList):
* css/CSSVariableValue.h: Added.
(WebCore):
(CSSVariableValue):
(WebCore::CSSVariableValue::create):
(WebCore::CSSVariableValue::name):
(WebCore::CSSVariableValue::value):
(WebCore::CSSVariableValue::CSSVariableValue):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* css/StyleResolver.h:
* css/WebKitCSSTransformValue.cpp:
(WebCore):
(WebCore::WebKitCSSTransformValue::customSerializeResolvingVariables):
* css/WebKitCSSTransformValue.h:
(WebKitCSSTransformValue):
* css/makeprop.pl:
* rendering/style/RenderStyle.h:
* rendering/style/StyleVariableData.h: Added.
(WebCore):
(StyleVariableData):
(WebCore::StyleVariableData::create):
(WebCore::StyleVariableData::copy):
(WebCore::StyleVariableData::operator==):
(WebCore::StyleVariableData::operator!=):
(WebCore::StyleVariableData::setVariable):
(WebCore::StyleVariableData::StyleVariableData):
2012-06-12 MORITA Hajime <morrita@google.com>
Shadow Pseudo ID should be able to nest to point nested shadow DOM.
https://bugs.webkit.org/show_bug.cgi?id=62218
Reviewed by Dimitri Glazkov.
- updateSpecifiersWithElementName() didn't take nesting into account.
tag history can contain selector entries which isn't marked as ShadowDescendant yet.
such entry can be found by investigating isUnknownPseudoElement().
- SelectorChecker::checkSelector() was too strict. Unknown pseudo elements are essentially a kind of class or id
in implementation perspective. For such, rejecting by missing elementStyle doesn't make sense:
It isn't a pseudo but a real element for WebCore after all.
- Added Internals::setShadowPseudoId() to create a test harness whose DOM tree contains pseudo shadow ids.
Before this change, following selecdtor chain is created for "p::-shadow-child::-nested-shadow-child":
[both tag name and -shadow-id are set] <-(ShadowDescendant)- [-nested-shadow-id]
What we want, which is created by this change is:
[tag] <-(ShadowDescendant)- [-shadow-id is set] <-(ShadowdescenDant)- [-nested-shadow-id]
Test: fast/dom/shadow/shadow-nested-pseudo-id.html
* WebCore.exp.in:
* css/CSSParser.cpp:
(WebCore::CSSParser::updateSpecifiersWithElementName):
* css/SelectorChecker.cpp:
(WebCore::SelectorChecker::checkOneSelector):
* testing/Internals.cpp:
(WebCore::Internals::setShadowPseudoId):
(WebCore):
* testing/Internals.h:
(Internals):
* testing/Internals.idl:
2012-06-12 Christophe Dumez <christophe.dumez@intel.com>
[soup] Prevent setting or editing httpOnly cookies from JavaScript
https://bugs.webkit.org/show_bug.cgi?id=88760
Reviewed by Gustavo Noronha Silva.
Prevent setting or overwriting httpOnly cookies from JavaScript.
Fix setCookies() so that it parses all the cookies and not just
the first one.
Test: http/tests/cookies/js-get-and-set-http-only-cookie.html
* platform/network/soup/CookieJarSoup.cpp:
(WebCore::httpOnlyCookieExists):
(WebCore):
(WebCore::setCookies):
2012-06-12 Christophe Dumez <christophe.dumez@intel.com>
[EFL] Enable SHADOW_DOM flag
https://bugs.webkit.org/show_bug.cgi?id=87732
Reviewed by Kentaro Hara.
Test: fast/dom/shadow/shadow-root-innerHTML.html
* dom/ShadowRoot.idl: Add JSGenerateToJSObject extended attribute so
that the ShadowRoot properties are accessible on JS side.
* testing/Internals.idl: Fix SHADOW_DOM ifdef check.
2012-06-12 Leo Yang <leo.yang@torchmobile.com.cn>
Dynamic hash table in DOMObjectHashTableMap is wrong in multiple threads
https://bugs.webkit.org/show_bug.cgi?id=87334
Reviewed by Geoffrey Garen.
Adapt to JSC::HashTable::copy to avoid copy dynamic table member of a HashTable.
The dynamic table may be allocated on other thread and contains thread specific
identifiers. For example, a hash table of JSEntryArray was first initialized on a
worker thread, and then the user reloaded the page, another worker thread is
created due to reload, the dynamic allocated table in *staticTable* is specific
to the first worker thread which has died. If the user reload the page again,
the dynamic table will be freed and memory corruption will occur.
No functionalities changed, no new tests.
* bindings/js/DOMObjectHashTableMap.h:
(WebCore::DOMObjectHashTableMap::get):
2012-06-12 James Robinson <jamesr@chromium.org>
[chromium] Port Canvas2DLayerBridge over to WebExternalTextureLayer
https://bugs.webkit.org/show_bug.cgi?id=88597
Reviewed by Adrienne Walker.
This converts Canvas2DLayerBridge to using public WebLayer API types instead of LayerChromium types.
* platform/graphics/chromium/Canvas2DLayerBridge.cpp:
(WebCore::AcceleratedDeviceContext::AcceleratedDeviceContext):
(WebCore::AcceleratedDeviceContext::prepareForDraw):
(AcceleratedDeviceContext):
(WebCore::Canvas2DLayerBridge::Canvas2DLayerBridge):
(WebCore::Canvas2DLayerBridge::~Canvas2DLayerBridge):
(WebCore::Canvas2DLayerBridge::skCanvas):
(WebCore::Canvas2DLayerBridge::prepareTexture):
(WebCore::Canvas2DLayerBridge::context):
(WebCore::Canvas2DLayerBridge::layer):
(WebCore::Canvas2DLayerBridge::contextAcquired):
* platform/graphics/chromium/Canvas2DLayerBridge.h:
(WebCore):
(Canvas2DLayerBridge):
2012-06-12 Alexis Menard <alexis.menard@openbossa.org>
Fix packing in StyleBoxData for Windows.
https://bugs.webkit.org/show_bug.cgi?id=88858
Reviewed by Tony Chang.
MSVC doesn't pack bitfields if the types are mixed.
We can safely change the bool type to an unsigned so
the packing will happen. I have added a compile assert
for future regressions.
No new tests : we should not see any regressions.
* rendering/style/StyleBoxData.h:
* rendering/style/StyleBoxData.cpp:
(StyleBoxData):
2012-06-12 Adrienne Walker <enne@google.com>
[chromium] Paint scrollbars on WebKit thread and composite those textures
https://bugs.webkit.org/show_bug.cgi?id=88145
Reviewed by James Robinson.
Scrollbars were previously painted and uploaded on the compositor
thread. This isn't possible to do for many scrollbar themes. This
patch changes ScrollbarLayerChromium to paint the scrollbar into two
parts: the thumb, and everything else. These are uploaded into
textures and synced over to the CCScrollbarImpl where they are
composited and drawn.
Mac and overlay scrollbars are still not enabled to have compositor
thread-updated scrollbars.
As a bonus, fix LayerRendererChromium::drawTextureQuad to draw quads
that do not fill the entire layer bounds.
Tested by existing layout and unit tests by removing the condition
that these scrollbar layers are only created when the threaded proxy
exists.
* page/scrolling/chromium/ScrollingCoordinatorChromium.cpp:
(WebCore::scrollbarLayerDidChange):
* platform/ScrollbarThemeComposite.cpp:
(WebCore::ScrollbarThemeComposite::thumbRect):
(WebCore):
* platform/ScrollbarThemeComposite.h:
(ScrollbarThemeComposite):
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::drawTextureQuad):
* platform/graphics/chromium/ScrollbarLayerChromium.cpp:
(WebCore::ScrollbarLayerChromium::ScrollbarLayerChromium):
(WebCore::ScrollbarLayerChromium::theme):
(WebCore):
(WebCore::ScrollbarLayerChromium::pushPropertiesTo):
(ScrollbarBackgroundPainter):
(WebCore::ScrollbarBackgroundPainter::create):
(WebCore::ScrollbarBackgroundPainter::paint):
(WebCore::ScrollbarBackgroundPainter::ScrollbarBackgroundPainter):
(ScrollbarThumbPainter):
(WebCore::ScrollbarThumbPainter::create):
(WebCore::ScrollbarThumbPainter::paint):
(WebCore::ScrollbarThumbPainter::ScrollbarThumbPainter):
(WebCore::ScrollbarLayerChromium::setLayerTreeHost):
(WebCore::ScrollbarLayerChromium::createTextureUpdaterIfNeeded):
(WebCore::ScrollbarLayerChromium::updatePart):
(WebCore::ScrollbarLayerChromium::update):
* platform/graphics/chromium/ScrollbarLayerChromium.h:
(WebCore):
(ScrollbarLayerChromium):
* platform/graphics/chromium/cc/CCScrollbarLayerImpl.cpp:
(WebCore::CCScrollbarLayerImpl::CCScrollbarLayerImpl):
(WebCore::CCScrollbarLayerImpl::appendQuads):
* platform/graphics/chromium/cc/CCScrollbarLayerImpl.h:
(WebCore::CCScrollbarLayerImpl::setBackgroundTextureId):
(WebCore::CCScrollbarLayerImpl::setThumbTextureId):
(CCScrollbarLayerImpl):
2012-06-12 Tony Chang <tony@chromium.org>
Replaced items in a flexbox should be coerced to display:block
https://bugs.webkit.org/show_bug.cgi?id=87068
Reviewed by Ojan Vafai.
The flexbox spec lists HTML tags that should automatically be converted
to display:block when a flex child. It also says that atomic inline-level
children should become block (e.g., inline-table should be treated as table).
http://dev.w3.org/csswg/css3-flexbox/#flex-items
Test: css3/flexbox/flexitem.html
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
2012-06-12 Joshua Bell <jsbell@chromium.org>
IndexedDB: ObjectStore/Index shouldn't hold reference to backing store
https://bugs.webkit.org/show_bug.cgi?id=83074
Reviewed by Tony Chang.
We should be able to collect and close the leveldb backing store as soon as the database
connection is closed, but the IDBObjectStoreBackendImpl and IDBIndexBackendImpl were
holding RefPtrs, and those objects are kept alive by script references.
Replaced RefPtrs to the IDBBackingStore with pointers to the IDBDatabase. On the back end,
IDBDatabaseBackendImpl maintains a RefPtr to the IDBObjectStoreBackendImpl object, so
a raw pointer back is safe. On the front end, the IDBObjectStore maintains a RefPtr to
the IDBDatabase so script can navigate upwards. Ditto on both ends for the ObjectStore/Index
relationship. The frontend objects maintain RefPtrs to the backend objects, so the backend
objects and their owners are maintained as long as there's a script reference. To ensure
that all LevelDB objects are destructed in the right order, transactions indirectly track all
open cursors and ensure they are closed when the transaction finishes, and database closing
is explicitly prevented from occurring until transactions are complete.
Also made IDBDatabaseBackendImpl handle a null IDBFactoryBackendImpl pointer, for testing.
Tests: storage/indexeddb/database-close.html
storage/indexeddb/persistence.html
[Chromium] webkit_unit_tests --gtest_filter=IDBDatabaseBackendTest.BackingStoreRetention
* Modules/indexeddb/IDBCursorBackendImpl.cpp:
(WebCore::IDBCursorBackendImpl::close): Release LevelDB resources when closed to allow backing store to close.
* Modules/indexeddb/IDBCursorBackendImpl.h:
(WebCore):
* Modules/indexeddb/IDBDatabase.cpp: Ensure connection close deferred until all transactions are finished.
(WebCore::IDBDatabase::transactionCreated): Accounting.
(WebCore::IDBDatabase::transactionFinished): Accounting.
(WebCore::IDBDatabase::close):
(WebCore):
(WebCore::IDBDatabase::closeConnection):
* Modules/indexeddb/IDBDatabase.h:
(IDBDatabase):
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp: Track outstanding transactions for forced-close scenarios.
(WebCore::IDBDatabaseBackendImpl::~IDBDatabaseBackendImpl):
(WebCore::IDBDatabaseBackendImpl::createObjectStore): Don't pass in backing store reference.
(WebCore::IDBDatabaseBackendImpl::setVersion): Use transaction() for consistent accounting.
(WebCore::IDBDatabaseBackendImpl::transactionFinished): Accounting.
(WebCore::IDBDatabaseBackendImpl::transaction): Accounting.
(WebCore::IDBDatabaseBackendImpl::close): Unregister from factory when last connection is closed.
(WebCore::IDBDatabaseBackendImpl::loadObjectStores): Don't pass in backing store reference.
(WebCore::IDBDatabaseBackendImpl::removeObjectStoreFromMap):
* Modules/indexeddb/IDBDatabaseBackendImpl.h:
(WebCore):
(IDBDatabaseBackendImpl):
* Modules/indexeddb/IDBFactoryBackendImpl.cpp:
(WebCore::IDBFactoryBackendImpl::deleteDatabase): Don't hang on to deleted databases eternally.
* Modules/indexeddb/IDBFactoryBackendImpl.h: Backend map controls IDBDatabaseBackend lifetimes.
(IDBFactoryBackendImpl):
* Modules/indexeddb/IDBIndexBackendImpl.cpp: Don't hold backing store references.
(WebCore::IDBIndexBackendImpl::IDBIndexBackendImpl):
* Modules/indexeddb/IDBIndexBackendImpl.h:
(WebCore::IDBIndexBackendImpl::create):
(IDBIndexBackendImpl):
(WebCore::IDBIndexBackendImpl::backingStore):
(WebCore::IDBIndexBackendImpl::databaseId):
* Modules/indexeddb/IDBObjectStoreBackendImpl.cpp: Don't hold backing store references.
(WebCore::IDBObjectStoreBackendImpl::IDBObjectStoreBackendImpl):
(WebCore::IDBObjectStoreBackendImpl::createIndex):
(WebCore::IDBObjectStoreBackendImpl::loadIndexes):
* Modules/indexeddb/IDBObjectStoreBackendImpl.h:
(WebCore::IDBObjectStoreBackendImpl::create):
(IDBObjectStoreBackendImpl):
(WebCore::IDBObjectStoreBackendImpl::backingStore):
(WebCore::IDBObjectStoreBackendImpl::databaseId):
* Modules/indexeddb/IDBRequest.cpp:
(WebCore::IDBRequest::onSuccess):
* Modules/indexeddb/IDBTransaction.cpp:
(WebCore::IDBTransaction::IDBTransaction): For IDBDatabase accounting, allow ref during ctor.
(WebCore::IDBTransaction::onAbort): Accounting.
(WebCore::IDBTransaction::onComplete): Accounting.
* Modules/indexeddb/IDBTransaction.h:
(WebCore::IDBTransaction::isVersionChange): Convenience function.
(IDBTransaction):
* Modules/indexeddb/IDBTransactionBackendImpl.cpp:
(WebCore::IDBTransactionBackendImpl::abort): Cleanup ordering to allow backing store to close.
(WebCore::IDBTransactionBackendImpl::commit): Cleanup ordering to allow backing store to close.
2012-06-12 Adrienne Walker <enne@google.com>
[chromium] Don't set visible layer rect in CCLayerTreeHost paint iteration
https://bugs.webkit.org/show_bug.cgi?id=88895
Reviewed by James Robinson.
We should only set visible layer rects during the recursion in
CClayerTreeHostCommon and not while we're painting. This is one last
straggling piece of code that never got moved.
Tested by existing mask-related layout tests.
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::paintMasksForRenderSurface):
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::calculateVisibleAndScissorRectsInternal):
2012-06-12 Dana Jansens <danakj@chromium.org>
[chromium] Return empty visibleLayerRect for layers with empty content bounds
https://bugs.webkit.org/show_bug.cgi?id=88901
Reviewed by Adrienne Walker.
This change should only affect tests. Currently if you set up a
CCLayerImpl with some bounds, but don't set the contentBounds, then
they are empty by default. In this case the visibleLayerRect gets
set inappropriately to the layer's target surface contentRect which
in a completely different coordinate space.
Fixed up tests that were passing bogusly and failed after this change.
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::calculateVisibleLayerRect):
2012-06-12 Adrienne Walker <enne@google.com>
[chromium] ScrollingCoordinator::setScrollLayer should update scroll layer ids
https://bugs.webkit.org/show_bug.cgi?id=88882
Reviewed by James Robinson.
Tested manually, as scrollbar layers are currently only created when
threaded compositing is enabled and none of those layout tests trigger
this bug.
* page/scrolling/chromium/ScrollingCoordinatorChromium.cpp:
(WebCore::ScrollingCoordinatorPrivate::setScrollLayer):
(ScrollingCoordinatorPrivate):
(WebCore::ScrollingCoordinatorPrivate::setHorizontalScrollbarLayer):
(WebCore::ScrollingCoordinatorPrivate::setVerticalScrollbarLayer):
(WebCore::createScrollbarLayer):
(WebCore::ScrollingCoordinator::frameViewHorizontalScrollbarLayerDidChange):
(WebCore::ScrollingCoordinator::frameViewVerticalScrollbarLayerDidChange):
* platform/graphics/chromium/ScrollbarLayerChromium.h:
(WebCore::ScrollbarLayerChromium::setScrollLayerId):
(ScrollbarLayerChromium):
* platform/graphics/chromium/TreeSynchronizer.cpp:
(WebCore::TreeSynchronizer::updateScrollbarLayerPointersRecursive):
2012-06-12 Mark Mentovai <mark@chromium.org>
[chromium mac] Don't include things in subframeworks of
ApplicationServices.framework.
Subframeworks move around from SDK to SDK and OS release to OS
release, but just using the top-level umbrella framework guarantees
forward compatibility. When building for Chromium Mac, just #include
<ApplicationServices/ApplicationServices.h> instead of reaching in to
subframeworks.
https://bugs.webkit.org/show_bug.cgi?id=88569
Reviewed by Stephen White.
No new tests.
* platform/graphics/GlyphBuffer.h:
* platform/graphics/mac/ComplexTextControllerCoreText.mm:
* platform/graphics/skia/GraphicsContextSkia.cpp:
2012-06-12 Igor Oliveira <igor.o@sisa.samsung.com>
Apply animations and transitions for first-letter element
https://bugs.webkit.org/show_bug.cgi?id=85253
Add animations and transitions support for the first-letter
pseudo element.
Instead of calling RenderOject::node() in the animations code,
now we need to call RenderObject::styledGeneratingNode() because
pseudo elements does not have a Node associated with the
RenderObject.
Initial patch by Simon Fraser
Reviewed by Simon Fraser.
Tests: animations/first-letter-animation.html
animations/first-letter-play-state.html
transitions/first-letter-color-transition.html
transitions/first-letter-transition.html
* page/animation/AnimationBase.cpp:
(WebCore::AnimationBase::updateStateMachine):
* page/animation/AnimationController.cpp:
(WebCore::AnimationControllerPrivate::updateAnimations):
(WebCore::AnimationControllerPrivate::pauseAnimationAtTime):
(WebCore::AnimationControllerPrivate::pauseTransitionAtTime):
(WebCore::AnimationController::cancelAnimations):
(WebCore::AnimationController::updateAnimations):
* page/animation/ImplicitAnimation.cpp:
(WebCore::ImplicitAnimation::pauseAnimation):
(WebCore::ImplicitAnimation::sendTransitionEvent):
* page/animation/KeyframeAnimation.cpp:
(WebCore::KeyframeAnimation::KeyframeAnimation):
(WebCore::KeyframeAnimation::pauseAnimation):
(WebCore::KeyframeAnimation::endAnimation):
(WebCore::KeyframeAnimation::sendAnimationEvent):
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::updateFirstLetterStyle):
(WebCore::RenderBlock::createFirstLetterRenderer):
* rendering/RenderInline.cpp:
(WebCore::RenderInline::clippedOverflowRectForRepaint):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::setAnimatableStyle):
(WebCore::RenderObject::styledGeneratingNode):
(WebCore):
* rendering/RenderObject.h:
(RenderObject):
2012-06-12 Kent Tamura <tkent@chromium.org>
Checking a radio button doesn't uncheck other buttons in the same group in some cases.
https://bugs.webkit.org/show_bug.cgi?id=88835
Reviewed by Ryosuke Niwa.
This change fixes a bug that checking a radio button in a radio button
group in a form detached from a document tree doesn't uncheck another
checked radio button in the radio button group.
A radio button participates in a radio button group in the following
conditions:
- If it is owned by a form element regardless of the form is in a
document tree or not, or
- If it is not owned by any form elements and it is in a document tree.
A radio button group for the radio button is owned by the document.
For HTMLInputElement::removedFrom():
The old code always unregistered the radio button if it was removed from
the document tree. It was incorrect because we don't need to unregister
it if it has an owner form and the owner form is not changed by
removedFrom().
If the owner form is cleared by removedFrom(), willChangeForm()
unregisters the radio button. So what we should do in removedFrom() is
to unregister the radio button only if the radio button group is owned
by the document.
For HTMLInputElement::insertedInto():
The old code always registered the radio button if it is inserted into
the document tree. It was incorrect because we don't need to register it
if it has an owner form and the owner form is not changed by
insertedInto().
If the owner form is changed by insertedInto(), didChangeForm()
registers the radio button. So We should register the radio button only
if its radio button group will be owned by the document.
Test: Add test cases to fast/forms/radio/radio-group.html
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::insertedInto):
Register this to CheckedRadioButtons only if new group owner is Document.
(WebCore::HTMLInputElement::removedFrom):
Unregister this from CheckedRadioButtons only if old group owner was Document.
2012-06-12 James Robinson <jamesr@chromium.org>
[chromium] REGRESSION(119769): Canvas2DLayerBridge may go away before its TextureLayerChromium
https://bugs.webkit.org/show_bug.cgi?id=88910
Reviewed by Adrienne Walker.
Unregister the bridge as its TextureLayerChromium client when going away.
Covered by new unit test in Canvas2DLayerBridgeTest.cpp.
* platform/graphics/chromium/Canvas2DLayerBridge.cpp:
(WebCore::Canvas2DLayerBridge::~Canvas2DLayerBridge):
2012-06-12 Alec Flett <alecflett@chromium.org>
IndexedDB: Error codes, phase two
https://bugs.webkit.org/show_bug.cgi?id=88690
Reviewed by Tony Chang.
Update the next phase of IndexedDB error codes to match the spec.
This update is mostly focused on DATA_ERR vs CONSTRAINT_ERR.
Also moves some of the error checking for the transaction mode
into the frontend and replaces the backend checks with asserts, so that
we can consolidate error checking in one place.
This also removes a few FIXME's that are no longer true (about
I/O errors - the spec was updated and our implementation was
already correct)
No new tests, existing tests cover changed behavior.
* Modules/indexeddb/IDBCursor.cpp:
(WebCore::IDBCursor::deleteFunction):
* Modules/indexeddb/IDBCursorBackendImpl.cpp:
(WebCore::IDBCursorBackendImpl::deleteFunction):
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::setVersionInternal):
* Modules/indexeddb/IDBObjectStore.cpp:
(WebCore::IDBObjectStore::add):
(WebCore::IDBObjectStore::put):
(WebCore::IDBObjectStore::deleteFunction):
(WebCore::IDBObjectStore::clear):
* Modules/indexeddb/IDBObjectStoreBackendImpl.cpp:
(WebCore::IDBObjectStoreBackendImpl::put):
(WebCore::IDBObjectStoreBackendImpl::putInternal):
(WebCore::IDBObjectStoreBackendImpl::deleteFunction):
(WebCore::IDBObjectStoreBackendImpl::clear):
* Modules/indexeddb/IDBTransaction.h:
(WebCore::IDBTransaction::enumMode):
* Modules/indexeddb/IDBTransactionBackendImpl.h:
2012-06-12 Min Qin <qinmin@google.com>
Disable autoplay when user gesture is required for media playback
https://bugs.webkit.org/show_bug.cgi?id=88897
Reviewed by Adam Barth.
When user gesture is required for media playback, we should block autoplay.
Otherwise, people will use autoplay instead of video.play() to bypass the gesture requirement.
Test: media/no-autoplay-with-user-gesture-requirement.html
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::setReadyState):
2012-06-12 Shrey Banga <banga@chromium.org>
HTML parser should yield more to improve perceived page load time
https://bugs.webkit.org/show_bug.cgi?id=86165
Reviewed by Tony Gentilcore.
Test: fast/events/event-yield-timing.html
We want the parser to yield at least every 500ms to keep the page somewhat responsive and allow painting.
Since it would be too expensive to check the time after each token, the previous heuristic was to check every 4,096 tokens.
That works fine for most tokens, but a script may spend an arbitrary amount of time executing.
This patch fixes the issue by causing the parser to check the elapsed time immediately after executing a script.
* html/parser/HTMLParserScheduler.cpp:
(WebCore::HTMLParserScheduler::checkForYieldBeforeScript):
* html/parser/HTMLParserScheduler.h:
(WebCore::PumpSession::PumpSession):
(PumpSession):
(WebCore::HTMLParserScheduler::checkForYieldBeforeToken):
2012-06-12 Sami Kyostila <skyostil@chromium.org>
[chromium] Don't crash in CCLayerIterator if the root layer doesn't have a render surface
https://bugs.webkit.org/show_bug.cgi?id=88886
Reviewed by James Robinson.
If we are iterating over a render render surface layer list where the
root layer does not have a render surface, fail gracefully instead of
crashing.
Tests: CCLayerIteratorTest.{emptyTree,rootLayerWithoutRenderSurface}
* platform/graphics/chromium/cc/CCLayerIterator.h:
(WebCore::CCLayerIterator::CCLayerIterator):
2012-06-06 Ojan Vafai <ojan@chromium.org>
Change default for flex-grow back to 0
https://bugs.webkit.org/show_bug.cgi?id=88443
Reviewed by Tony Chang.
* rendering/style/RenderStyle.h:
2012-06-12 Brady Eidson <beidson@apple.com>
<rdar://problem/11593686> and https://bugs.webkit.org/show_bug.cgi?id=88683
Garbage collection of an <img> element can cause reentrant event dispatch.
Reviewed by Darin Adler.
The most straightforward solution is for ImageLoader to keep its Element alive
with ref/deref any time the Image is actually loading.
ImageLoader should always do this for all Elements, and if those Elements want/need
different behavior for when they are detached then they need to manually stop their
loads.
Tests: http/tests/loading/embed-image-load-outlives-gc-without-crashing.html
http/tests/loading/image-input-type-outlives-gc-without-crashing.html
http/tests/loading/image-load-outlives-gc-without-crashing.html
http/tests/loading/object-image-load-outlives-gc-without-crashing.html
http/tests/loading/svg-image-load-outlives-gc-without-crashing.html
http/tests/loading/video-poster-image-load-outlives-gc-without-crashing.html
* loader/ImageLoader.cpp:
(WebCore::ImageLoader::ImageLoader):
(WebCore::ImageLoader::~ImageLoader):
(WebCore::ImageLoader::setImage):
(WebCore::ImageLoader::updateFromElement):
(WebCore::ImageLoader::notifyFinished):
(WebCore::ImageLoader::updatedHasPendingLoadEvent):
(WebCore::ImageLoader::dispatchPendingBeforeLoadEvent):
(WebCore::ImageLoader::dispatchPendingLoadEvent):
* loader/ImageLoader.h:
(ImageLoader):
2012-06-12 Shawn Singh <shawnsingh@chromium.org>
[chromium] Make damage tracking more robust to early exits
https://bugs.webkit.org/show_bug.cgi?id=84803
Reviewed by James Robinson.
All CCDamageTrackerTest unit tests updated;
One new unit test added:
CCDamageTrackerTest.verifyDamageAccumulatesUntilReset
Damage tracking inherently needs to be aware of all changes that
occur for every frame. Before this patch, anyone modifying details
of CCLayerTreeHostImpl and the drawing phases of the compositor
needed to be careful to avoid early exiting once draw is
initiated, otherwise running the risk of damage tracking's state
getting out-of-sync from actual changes. This patch makes the
damage tracker API more asynchronous-like, so that damage will
continue to accumulate until it is explicitly notified that
damaged areas have been drawn. This will allow more freedom in
designing and restructuing drawing code in the compositor, without
worrying as much about damage tracking getting out-of-sync.
* platform/graphics/chromium/cc/CCDamageTracker.h:
(WebCore::CCDamageTracker::didDrawDamagedArea):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::drawLayers):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(CCLayerTreeHostImpl):
2012-06-12 Mike West <mkwst@chromium.org>
Make document.documentURI readonly from JavaScript
https://bugs.webkit.org/show_bug.cgi?id=65187
Reviewed by Alexey Proskuryakov.
The DOM4 working draft marks the documentURI attribute as read only
(http://www.w3.org/TR/dom/#document). Firefox has shipped with this
behavior since https://hg.mozilla.org/mozilla-central/rev/3bc751906409
landed in October 2011, IE9 doesn't support the property, and
Opera throws a NO_MODIFICATION_ALLOWED_ERR. This patch changes WebKit
to silently fail (matching Firefox's behavior) by setting the property
to readonly in the IDL.
Document::setDocumentURI and the m_documentURI property are retained
for compatibility with ObjC clients, and the readonly attribute is
ifdeffed out for ObjC.
This patch adds a single test to verify the behavior, and removes a
variety of tests that depended on the writable behavior. In particular,
potential security issues involving the document.baseURL property are
avoided completely as long as this property can't be changed.
Test: fast/dom/documenturi-readonly.html
* dom/Document.cpp:
(WebCore::Document::setDocumentURI):
Adds a comment explaining that the fallback is necessary only to
support ObjC, not for JS calls.
* dom/Document.cpp:
(WebCore::Document::updateBaseURL):
Ditto.
(Document):
* dom/Document.idl:
Add readonly to the attribute and drop null value when not in
LANGUAGE_OBJECTIVE_C.
2012-06-12 Silvia Pfeiffer <silviapf@chromium.org>
Support !ENABLE(VIDEO) builds with horizontally layed out video controls.
https://bugs.webkit.org/show_bug.cgi?id=88881
Reviewed by Csaba Osztrogonác.
No new tests as this is a regression from WK87835.
* html/shadow/SliderThumbElement.cpp:
(WebCore::hasVerticalAppearance):
Ifdef out video related functionality, since usesVerticalVolumeSlider
does not exist in that case.
2012-06-12 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: show revisions of all modified files in local modifications view
https://bugs.webkit.org/show_bug.cgi?id=88864
Reviewed by Vsevolod Vlasov.
This change allows all local modifications to be seen in a single view. View is now
singleton, listening to all revisions being added. Revert captions were reworked
(see the screenshot).
* English.lproj/localizedStrings.js:
* inspector/front-end/Resource.js:
* inspector/front-end/RevisionHistoryView.js:
(WebInspector.RevisionHistoryView.populateRevisions):
(WebInspector.RevisionHistoryView):
(WebInspector.RevisionHistoryView.showHistory):
(WebInspector.RevisionHistoryView.reset):
(WebInspector.RevisionHistoryView.prototype._createResourceItem.clearHistory):
(WebInspector.RevisionHistoryView.prototype._createResourceItem):
(WebInspector.RevisionHistoryView.prototype._revisionAdded.get if):
(WebInspector.RevisionHistoryView.prototype._revisionAdded):
(WebInspector.RevisionHistoryView.prototype._revealResource.get if):
(WebInspector.RevisionHistoryView.prototype._revealResource):
(WebInspector.RevisionHistoryView.prototype._reset):
(WebInspector.RevisionHistoryTreeElement):
(WebInspector.RevisionHistoryTreeElement.prototype.onattach):
(WebInspector.RevisionHistoryTreeElement.prototype.oncollapse):
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype._reset):
(WebInspector.ScriptsPanel.prototype._editorSelected):
* inspector/front-end/revisionHistory.css:
(.revision-history-link-row):
(.revision-history-drawer li.revision-history-revision):
2012-06-12 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: allow clearing revision history.
https://bugs.webkit.org/show_bug.cgi?id=88847
Reviewed by Vsevolod Vlasov.
This change also renames "Revision history" to "Local modifications".
I intend to list modifications to all files in that view in the subsequent change.
* English.lproj/localizedStrings.js:
* inspector/front-end/Resource.js:
(WebInspector.Resource._clearResourceHistory):
* inspector/front-end/RevisionHistoryView.js:
(WebInspector.RevisionHistoryView.else.clearHistory):
(WebInspector.RevisionHistoryView):
(WebInspector.RevisionHistoryView.prototype._revisionAdded):
(WebInspector.RevisionHistoryTreeElement):
(WebInspector.RevisionHistoryTreeElement.prototype._createLine):
(WebInspector.RevisionHistoryTreeElement.prototype.allowRevert):
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype.appendApplicableItems):
2012-06-12 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: use jsdifflib for revision diff calculation.
https://bugs.webkit.org/show_bug.cgi?id=88791
Reviewed by Vsevolod Vlasov.
Present diff calculation is incomplete, it does not properly handle the following case:
[a, b, c, d, e, f, g, h] => [c, d, e, f, g, h, a, b]
We should not be implementing diff by ourselves. I imported jsdifflib from
https://github.com/cemerick/jsdifflib. It has BSD license that is compatible with WebKit.
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* inspector/front-end/RevisionHistoryView.js:
(WebInspector.RevisionHistoryTreeElement.prototype.onexpand.step2):
(WebInspector.RevisionHistoryTreeElement.prototype.onexpand):
(WebInspector.RevisionHistoryTreeElement.prototype._createLine.appendLineNumber):
(WebInspector.RevisionHistoryTreeElement.prototype._createLine):
* inspector/front-end/WebKit.qrc:
* inspector/front-end/externs.js:
(difflib.stringAsLines):
(difflib.SequenceMatcher):
(difflib.SequenceMatcher.prototype.get_opcodes):
* inspector/front-end/inspector.html:
* inspector/front-end/jsdifflib.js: Added.
(difflib.defaultJunkFunction):
(difflib.stripLinebreaks):
(difflib.stringAsLines):
(difflib.__reduce):
(difflib.__ntuplecomp):
(difflib.__calculate_ratio):
(difflib.__dictget):
(difflib.SequenceMatcher.this.set_seqs):
(difflib.SequenceMatcher.this.set_seq1):
(difflib.SequenceMatcher.this.set_seq2):
(difflib.SequenceMatcher.this.__chain_b):
(difflib.SequenceMatcher.this.find_longest_match):
(difflib.SequenceMatcher.this.get_matching_blocks):
(difflib.SequenceMatcher.this.get_opcodes):
(difflib.SequenceMatcher.this.get_grouped_opcodes):
(difflib.SequenceMatcher.this.quick_ratio):
(difflib.SequenceMatcher.this.real_quick_ratio):
* inspector/front-end/revisionHistory.css:
(.revision-history-line-separator .webkit-line-number):
2012-06-12 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: use jsdifflib for revision diff calculation.
https://bugs.webkit.org/show_bug.cgi?id=88791
Reviewed by Vsevolod Vlasov.
Present diff calculation is incomplete, it does not properly handle the following case:
[a, b, c, d, e, f, g, h] => [c, d, e, f, g, h, a, b]
We should not be implementing diff by ourselves. I imported jsdifflib from
https://github.com/cemerick/jsdifflib. It has BSD license that is compatible with WebKit.
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* inspector/front-end/RevisionHistoryView.js:
(WebInspector.RevisionHistoryTreeElement.prototype.onexpand.step2):
(WebInspector.RevisionHistoryTreeElement.prototype.onexpand):
(WebInspector.RevisionHistoryTreeElement.prototype._createLine.appendLineNumber):
(WebInspector.RevisionHistoryTreeElement.prototype._createLine):
* inspector/front-end/WebKit.qrc:
* inspector/front-end/externs.js:
(difflib.stringAsLines):
(difflib.SequenceMatcher):
(difflib.SequenceMatcher.prototype.get_opcodes):
* inspector/front-end/inspector.html:
* inspector/front-end/jsdifflib.js: Added.
(difflib.defaultJunkFunction):
(difflib.stripLinebreaks):
(difflib.stringAsLines):
(difflib.__reduce):
(difflib.__ntuplecomp):
(difflib.__calculate_ratio):
(difflib.__dictget):
(difflib.SequenceMatcher.this.set_seqs):
(difflib.SequenceMatcher.this.set_seq1):
(difflib.SequenceMatcher.this.set_seq2):
(difflib.SequenceMatcher.this.__chain_b):
(difflib.SequenceMatcher.this.find_longest_match):
(difflib.SequenceMatcher.this.get_matching_blocks):
(difflib.SequenceMatcher.this.get_opcodes):
(difflib.SequenceMatcher.this.get_grouped_opcodes):
(difflib.SequenceMatcher.this.quick_ratio):
(difflib.SequenceMatcher.this.real_quick_ratio):
* inspector/front-end/revisionHistory.css:
(.revision-history-line-separator .webkit-line-number):
2012-06-12 Christophe Dumez <christophe.dumez@intel.com>
[GStreamer] http/tests/media/video-buffered-range-contains-currentTime.html is failing
https://bugs.webkit.org/show_bug.cgi?id=87575
Reviewed by Philippe Normand.
When EOS is reached in forward playback, the position does not necessarily
match the duration. This confuses HTMLMediaElement and causes the "ended"
event not to be fired. To avoid this, we synchronize position and duration
values in this case.
Test: http/tests/media/video-buffered-range-contains-currentTime.html
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
2012-06-12 Kinuko Yasuda <kinuko@chromium.org>
Refactor InputType::receiveDroppedFiles to take DragData
https://bugs.webkit.org/show_bug.cgi?id=88860
Reviewed by Kent Tamura.
So that we can centralize the file paths related code into FileInputType
and makes it easier to extract more data from DragData for
<input type=file> (like bug 88293 for directory drag-and-drop).
No new tests as this has no behavioral changes.
* html/FileInputType.cpp:
(WebCore::FileInputType::receiveDroppedFiles):
* html/FileInputType.h:
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::receiveDroppedFiles):
* html/HTMLInputElement.h:
* html/InputType.cpp:
(WebCore::InputType::receiveDroppedFiles):
* html/InputType.h:
* page/DragController.cpp:
(WebCore::DragController::concludeEditDrag):
2012-06-12 Thiago Marcos P. Santos <thiago.santos@intel.com>
Generate -webkit-box-decoration-break property only when the feature is enabled
https://bugs.webkit.org/show_bug.cgi?id=88861
Reviewed by Alexis Menard.
Otherwise it will hit an assertion if the property is present on the
style but the feature disabled. Also fixes a compilation warning when
the feature is disabled.
* css/CSSPropertyNames.in:
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
2012-06-12 Darin Adler <darin@apple.com>
Tweaked the Mac export file a bit more, adding more-correct iOS cases.
Reviewed by David Kilzer.
* WebCore.exp.in: Rearrangement.
2012-06-11 Jocelyn Turcotte <jocelyn.turcotte@nokia.com>
Using extrernal ICU library on case unsensitive drives will not work
https://bugs.webkit.org/show_bug.cgi?id=70913
Reviewed by Csaba Osztrogonác.
* platform/graphics/surfaces/GraphicsSurface.h:
2012-06-11 Dana Jansens <danakj@chromium.org>
[chromium] Use TextureDrawQuad for video layer output when it produces a single GL texture
https://bugs.webkit.org/show_bug.cgi?id=88814
Reviewed by Adrienne Walker.
When a CCVideLayerImpl produces quads for a frame that is contained
in a single GL texture, then use the TextureDrawQuad instead of the
VideoDrawQuad. Moves the computation for the RGBA(software) and
NativeTexture (hardware) cases from LayerRendererChromium into
CCVideoLayerImpl, so it can embed these values in the TextureDrawQuad.
No new tests, behaviour should be the same.
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::drawStreamTexture):
(WebCore::LayerRendererChromium::drawVideoQuad):
* platform/graphics/chromium/LayerRendererChromium.h:
(LayerRendererChromium):
* platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
(WebCore::CCVideoLayerImpl::appendQuads):
2012-06-12 Silvia Pfeiffer <silviapf@chromium.org>
Change the volume slider to horizontal rendering for Chrome video controls.
https://bugs.webkit.org/show_bug.cgi?id=87835
Reviewed by Eric Carlson.
No new tests - final patch in the Chrome controls update series will contain rebaselined tests.
The Chrome video controls are receiving a visual update. The volume slider is moved into
the controls with horizontal rendering, the volume slider container is removed. The visual
update itself is in a separate patch.
* css/mediaControlsChromium.css:
(audio::-webkit-media-controls-mute-button, video::-webkit-media-controls-mute-button):
Removed relative positioning to render all controls elements equally in DOM order.
(audio::-webkit-media-controls-play-button, video::-webkit-media-controls-play-button):
Removed absolute positioning to render all controls elements equally in DOM order.
(audio::-webkit-media-controls-timeline-container, video::-webkit-media-controls-timeline-container):
Removed absolute positioning to render all controls elements equally in DOM order.
(audio::-webkit-media-controls-volume-slider-container, video::-webkit-media-controls-volume-slider-container):
Removed volume slider container - it's not necessary any more.
(audio::-webkit-media-controls-current-time-display, video::-webkit-media-controls-current-time-display):
Remove flexbox and introduce inline display to always display this field.
(audio::-webkit-media-controls-volume-slider, video::-webkit-media-controls-volume-slider):
Removed absolute positioning to render all controls elements equally in DOM order.
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlRootElementChromium::MediaControlRootElementChromium):
Removed volume slider container - it's not necessary any more.
(WebCore::MediaControlRootElementChromium::create):
Appended the volume slider and the mute button directly to the panel, removed volume slider container.
(WebCore::MediaControlRootElementChromium::setMediaController):
Removed volume slider container.
(WebCore::MediaControlRootElementChromium::reportedError):
Removed volume slider container.
(WebCore::MediaControlRootElementChromium::showVolumeSlider):
Removed volume slider container.
* html/shadow/MediaControlRootElementChromium.h:
(WebCore):
(MediaControlRootElementChromium):
Removed volume slider container.
* html/shadow/SliderThumbElement.cpp:
(WebCore::hasVerticalAppearance):
Allow use of horizontal media volume slider.
(WebCore::RenderSliderThumb::layout):
Reuse hasVerticalAppearance function.
* rendering/RenderMediaControlsChromium.cpp:
(WebCore::RenderMediaControlsChromium::paintMediaVolumeSlider):
Change line drawing from vertical to horizontal.
* rendering/RenderTheme.h:
(WebCore::RenderTheme::usesVerticalVolumeSlider):
Add usesVerticalVolumeSlider function to allow horizontal volume sliders.
* rendering/RenderThemeChromiumMac.h:
(WebCore::RenderThemeChromiumMac::usesVerticalVolumeSlider):
Add usesVerticalVolumeSlider function to allow horizontal volume sliders.
Add usesVerticalVolumeSlider function to allow horizontal volume sliders.
* rendering/RenderThemeChromiumSkia.h:
(WebCore::RenderThemeChromiumSkia::usesVerticalVolumeSlider):
Add usesVerticalVolumeSlider function to allow horizontal volume sliders.
2012-06-12 Andrei Bucur <abucur@adobe.com>
[CSSRegions]NamedFlow::overset should return true when there's no region chain attached
https://bugs.webkit.org/show_bug.cgi?id=88515
The CSS Regions specifies that the "overset" attribute needs to return true if the content overflows, there's no region attached to the flow
or if the NamedFlow is in the NULL state:
http://www.w3.org/TR/css3-regions/#dom-named-flow-overset
This patch updates WebKit's behavior in the last two situations to match the one in the spec.
Reviewed by Andreas Kling.
Test: fast/regions/webkit-named-flow-overset.html was updated
* rendering/RenderFlowThread.cpp:
(WebCore::RenderFlowThread::RenderFlowThread):
(WebCore::RenderFlowThread::computeOverflowStateForRegions): If the last valid region is null, the flow has not region chain attached or it's in the NULL state.
2012-06-12 Rahul Tiwari <rahultiwari.cse.iitr@gmail.com>
Web Inspector: Provide context menu for deleting recorded profile on Profiles Panel
https://bugs.webkit.org/show_bug.cgi?id=88286
Added the context menu to delete the selected profile.
Reviewed by Pavel Feldman.
No new tests required as UI related change
* English.lproj/localizedStrings.js: Removed property svn:mime-type.
* inspector/front-end/ProfilesPanel.js:
(WebInspector.ProfileSidebarTreeElement.prototype.handleContextMenuEvent):
2012-06-12 Pavel Feldman <pfeldman@chromium.org>
Not reviewed: removed svn:mime-type property from English.lproj/localizedStrings.js
* English.lproj/localizedStrings.js
2012-06-12 Dan Bernstein <mitz@apple.com>
Tried to fix the notification-enabled build after r120044.
* notifications/NotificationController.cpp:
(WebCore::NotificationController::NotificationController):
(WebCore::NotificationController::create):
(WebCore::provideNotification):
* notifications/NotificationController.h:
(NotificationController):
2012-06-12 Huang Dongsung <luxtella@company100.net>
[Qt][Texmap] Falling leaves demo missing clipping.
https://bugs.webkit.org/show_bug.cgi?id=88704
The bug originated from begining native painting mode of QPainter after enabling
scissor test, because begining native painting mode disables scissor test.
Covered by existing compositing tests.
Reviewed by Noam Rosenthal.
* platform/graphics/texmap/TextureMapperGL.cpp:
(WebCore::TextureMapperGL::beginPainting):
2012-06-12 Csaba Osztrogonác <ossy@webkit.org>
[Qt][Win] Buildfix related to fromWinHBITMAP
https://bugs.webkit.org/show_bug.cgi?id=88863
Reviewed by Jocelyn Turcotte.
* platform/graphics/qt/GraphicsContextQt.cpp: Add qt_pixmapFromWinHBITMAP and HBitmapFormat enum.
(WebCore):
(WebCore::GraphicsContext::releaseWindowsContext):
* platform/graphics/qt/ImageQt.cpp:
(WebCore): Move qt_pixmapFromWinHBITMAP to outside of WebCore namespace. (Fix after r119924.)
2012-06-12 Takashi Sakamoto <tasak@google.com>
[Shadow DOM] <style> inside Shadow subtree should be scoped inside the subtree.
https://bugs.webkit.org/show_bug.cgi?id=87805
Modified HTMLStyleElement to be treated as a scoped style independent
of its "scoped" attribute's value if HTMLStyleElement is in shadow
subtree. HTMLStyleElement has the following four cases talking about
"scoped" attribute and whether is in shadow subtree or not:
1, HTMLStyleElement is "scoped" and is in document tree,
2, HTMLStyleElement is "scoped" and is in shadow subtree,
3, HTMLStyleElement is not "scoped" and is in document tree, and
4, HTMLStyleElement is not "scoped" and is in shadow subtree.
The case 1 and 2 are not changed. This patch affected the case 4.
At the case 4, the HTMLStyleElement's scoping node is the shadow root.
This change is the first step for fixing style scope bug,
i.e. HTMLStyleElements in shadow subtree are applied to any element
in document tree.
Reviewed by Dimitri Glazkov.
Test: fast/css/style-scoped/style-scoped-change-scoped-in-shadow.html
* css/StyleResolver.cpp:
(WebCore::StyleResolver::determineScope):
Modified to return a shadow root if a targeted stylesheet owner node is
not scoped but in some shadow DOM subtree.
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::HTMLStyleElement):
(WebCore::HTMLStyleElement::parseAttribute):
Modified to use the below scopedAttributeChanged when a changed
attribute's name is "scoped".
(WebCore::HTMLStyleElement::scopedAttributeChanged):
Newly added. According to new scoped value and isInShadowTree,
add or remove style rules to/from scopedAuthorRules via
registerWithScopingNode or unregisterWithScopingNode.
(WebCore::HTMLStyleElement::registerWithScopingNode):
Added one boolean argument to determine whether HTMLStyleElement is
scoped or not. The reason why not using scoped() is the comment:
"We cannot rely on the scoped element already being present when this
method is invoked. Therefore we cannot rely on scoped()"
(WebCore::HTMLStyleElement::unregisterWithScopingNode):
Changed the code for updating m_isRegisteredWithScopingNode. Now
set m_scopedStyleRegistrationState to be NotRegistered.
(WebCore::HTMLStyleElement::insertedInto):
Modified to invoke registerWithScopingNode when an element is not
scoped but in some shadow subtree.
(WebCore::HTMLStyleElement::removedFrom):
Modified to invoke unregisterWithScopingNode when an element is not
scoped but in some shadow subtree.
* html/HTMLStyleElement.h:
(HTMLStyleElement):
Modified the type of m_isRegisteredWithScopingNode from bool to
enum and renamed to m_scopedStyleRegistrationState. Now the member
variable keeps what HTMLStyleElement's scoping node is, i.e.
none (this means, not in shadow subtree and not scoped), shadow root or
parent node.
Added one new method scopedAttributeChanged's declaration and
modified the declaration of registerWithScopingNode.
2012-06-12 Amy Ousterhout <aousterh@chromium.org>
Removing unused parameter Page
https://bugs.webkit.org/show_bug.cgi?id=88848
Reviewed by Kentaro Hara.
Removed the unused parameter Page* supplied to the create function
and constructor in DeviceOrientationController.
The parameter Page* was previously stored in an instance variable
m_page, but this was removed (see https://bugs.webkit.org/show_bug.cgi?id=88812).
No new tests because the removed parameter was unused and untested.
* dom/DeviceOrientationController.cpp:
(WebCore::DeviceOrientationController::DeviceOrientationController):
(WebCore::DeviceOrientationController::create):
(WebCore::provideDeviceOrientationTo):
* dom/DeviceOrientationController.h:
(DeviceOrientationController):
2012-06-12 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r120051.
http://trac.webkit.org/changeset/120051
https://bugs.webkit.org/show_bug.cgi?id=88852
some tests are crashing (Requested by morrita on #webkit).
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/generic/ContextEnabledFeatures.cpp: Renamed from Source/WebKit/chromium/src/ContextFeaturesClientImpl.cpp.
(WebCore):
(WebCore::ContextEnabledFeatures::shadowDOMEnabled):
(WebCore::ContextEnabledFeatures::styleScopedEnabled):
(WebCore::ContextEnabledFeatures::pagePopupEnabled):
* bindings/generic/ContextEnabledFeatures.h: Renamed from Source/WebKit/chromium/src/ContextFeaturesClientImpl.h.
(WebCore):
(ContextEnabledFeatures):
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateImplementation):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::determineScope):
* dom/ContextFeatures.cpp: Removed.
* dom/ContextFeatures.h: Removed.
* dom/DOMAllInOne.cpp:
* dom/DOMImplementation.cpp:
(WebCore::DOMImplementation::createDocument):
(WebCore::DOMImplementation::createHTMLDocument):
* dom/Document.cpp:
(WebCore::Document::Document):
* dom/Document.h:
(WebCore):
(Document):
* dom/Position.cpp:
(WebCore::Position::Position):
(WebCore::Position::findParent):
* dom/TreeScope.cpp:
(WebCore::TreeScope::getSelection):
* dom/make_names.pl:
(printConstructorInterior):
(printFactoryCppFile):
(printWrapperFunctions):
(printWrapperFactoryCppFile):
* editing/markup.cpp:
(WebCore::createFragmentFromMarkupWithContext):
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::registerWithScopingNode):
(WebCore::HTMLStyleElement::unregisterWithScopingNode):
* html/shadow/HTMLContentElement.cpp:
(WebCore::contentTagName):
* inspector/DOMPatchSupport.cpp:
(WebCore::DOMPatchSupport::patchDocument):
* loader/FrameLoaderClient.h:
(WebCore::FrameLoaderClient::allowShadowDOM):
(WebCore::FrameLoaderClient::allowStyleScoped):
(FrameLoaderClient):
(WebCore::FrameLoaderClient::allowPagePopup):
* platform/RefCountedSupplement.h: Removed.
* platform/Supplementable.h:
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::responseXML):
2012-06-11 Kaustubh Atrawalkar <kaustubh@motorola.com>
[DRT] LTC:: counterValueForElementById() could be moved to Internals.
https://bugs.webkit.org/show_bug.cgi?id=84406
Reviewed by Hajime Morita.
Move the counterValueForElementById from LayoutTestCotroller to Internals and
remove the old platform specific implementations as it exclusively tests WebCore functionality.
Covered by existing test cases.
* testing/Internals.cpp:
(WebCore::Internals::counterValue):
(WebCore):
* testing/Internals.h:
(Internals):
* testing/Internals.idl:
2012-06-11 Hans Wennborg <hans@chromium.org>
Speech JavaScript API: Make SpeechRecognitionError an Event
https://bugs.webkit.org/show_bug.cgi?id=88784
Reviewed by Adam Barth.
Make SpeechRecognitionError an Event. The spec was updated to make it
an event in its own right, rather than an attribute of
SpeechRecognitionEvent.
Test: fast/speech/scripted/speechrecognition-errors.html
* Modules/speech/SpeechRecognition.cpp:
(WebCore::SpeechRecognition::didReceiveError):
* Modules/speech/SpeechRecognitionError.cpp:
(WebCore::SpeechRecognitionError::SpeechRecognitionError):
(WebCore::SpeechRecognitionError::interfaceName):
(WebCore):
* Modules/speech/SpeechRecognitionError.h:
(WebCore::SpeechRecognitionError::create):
(SpeechRecognitionError):
* Modules/speech/SpeechRecognitionError.idl:
* Modules/speech/SpeechRecognitionEvent.cpp:
(WebCore::SpeechRecognitionEvent::SpeechRecognitionEvent):
* Modules/speech/SpeechRecognitionEvent.h:
(SpeechRecognitionEventInit):
(SpeechRecognitionEvent):
* Modules/speech/SpeechRecognitionEvent.idl:
* dom/EventNames.in:
2012-06-12 MORITA Hajime <morrita@google.com>
REGRESSION(r118098): <content> element does not render distributed children when cloned from another document
https://bugs.webkit.org/show_bug.cgi?id=88148
Reviewed by Dimitri Glazkov.
The shadow DOM availability was decided through FrameLoaderClient.
But there are documents which don't have any associated frames, in
which any shadow DOM related elements cannot be constructed. That
resulsted the regression.
This change introduces a Page-based client called
ContextFeaturesClient which takes the role from FrameLoaderClient
to decide the feature availability, and ContextFeatures which is a
proxy of ContextFeaturesClient. ContextEnabledFeatures is is
replaced with ContextFeatures.
Each ContextFeatures object is owned by a Page, and is attached to
each Document in the page even if the page itself has no referenct
to Frames or Pages. With ContextFeatures, each Document can decide
the feature availability even if it cannot reach Page or Frame.
On RefCountedSupplement: Document instance can live longer than a
Page where it is shwon. This means that ContextFeatures instance
needs to survive after owner Page destruction because it's
referenced from possibly surviving Documents. RefCountedSupplement
is introduced to cover this scenario: It allows supplement classes
to live after Page's destruction.
RefCountedSupplement::hostDestroyed() is notified when the hosting
page is gone. ContextFeatures clears its reference to the client
using this notification.
Test: fast/dom/shadow/elements-in-frameless-document.html
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateImplementation):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::determineScope):
* dom/ContextFeatures.cpp: Added.
(WebCore):
(WebCore::ContextFeaturesClient::empty):
(WebCore::ContextFeatures::supplementName):
(WebCore::ContextFeatures::defaultSwitch):
(WebCore::ContextFeatures::shadowDOMEnabled):
(WebCore::ContextFeatures::styleScopedEnabled):
(WebCore::ContextFeatures::pagePopupEnabled):
(WebCore::provideContextFeaturesTo):
(WebCore::provideContextFeaturesToDocumentFrom):
* dom/ContextFeatures.h: Added.
(WebCore):
(ContextFeatures):
(WebCore::ContextFeatures::ContextFeatures):
(WebCore::ContextFeatures::hostDestroyed):
(ContextFeaturesClient):
(WebCore::ContextFeaturesClient::~ContextFeaturesClient):
(WebCore::ContextFeaturesClient::isEnabled):
(WebCore::ContextFeatures::create):
(WebCore::ContextFeatures::isEnabled):
* dom/DOMAllInOne.cpp:
* dom/DOMImplementation.cpp:
(WebCore::DOMImplementation::createDocument):
(WebCore::DOMImplementation::createHTMLDocument):
* dom/Document.h:
(WebCore::Document::contextFeatures):
* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::setContextFeatures):
* dom/Position.cpp:
(WebCore::Position::Position):
(WebCore::Position::findParent):
* dom/TreeScope.cpp:
(WebCore::TreeScope::getSelection):
* dom/make_names.pl:
(printConstructorInterior):
(printFactoryCppFile):
(printWrapperFunctions):
(printWrapperFactoryCppFile):
* editing/markup.cpp:
(WebCore::createFragmentFromMarkupWithContext):
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::registerWithScopingNode):
(WebCore::HTMLStyleElement::unregisterWithScopingNode):
* html/shadow/HTMLContentElement.cpp:
(WebCore::contentTagName):
* inspector/DOMPatchSupport.cpp:
(WebCore::DOMPatchSupport::patchDocument):
* loader/FrameLoaderClient.h:
* platform/RefCountedSupplement.h: Copied from Source/WebCore/platform/Supplementable.h.
(WebCore):
(RefCountedSupplement):
(WebCore::RefCountedSupplement::hostDestroyed):
(Wrapper):
(WebCore::RefCountedSupplement::Wrapper::Wrapper):
(WebCore::RefCountedSupplement::Wrapper::~Wrapper):
(WebCore::RefCountedSupplement::Wrapper::wrapped):
(WebCore::RefCountedSupplement::provideTo):
(WebCore::RefCountedSupplement::from):
* platform/Supplementable.h:
(WebCore::Supplement::isRefCountedWrapper):
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::responseXML):
2012-06-12 Kent Tamura <tkent@chromium.org>
Remove unnecessary functions: setName() and formControlName()
https://bugs.webkit.org/show_bug.cgi?id=88392
Reviewed by Kentaro Hara.
'name' IDL attributes of form-related elements should be [Reflected],
and we don't need to have setName(). We used formControlName() for
name() implementation, and formControlName() is not needed because it
just converts a null attribute to an empty string. Our binding code does it.
We don't remove FormAssociatedElement::name() because many C++ code use it.
FormAssociatedElement::name() is virtual, and HTMLInputElement overrides
it so that it returns a cache of the name attribtue value because
CheckedRadioButtons class needs to know an old name attribute value when
the name attribtue is changed.
References:
http://www.whatwg.org/specs/web-apps/current-work/multipage/the-iframe-element.html#dom-object-name
http://www.whatwg.org/specs/web-apps/current-work/multipage/forms.html#dom-form-name
http://www.whatwg.org/specs/web-apps/current-work/multipage/association-of-controls-and-forms.html#dom-fe-name
Test: fast/forms/name-attribute.html
* dom/Element.h: Remove formControlName().
* html/FormAssociatedElement.cpp:
(WebCore::FormAssociatedElement::name):
Copied from formControlName() implementation of HTMLFormControlElement.
* html/FormAssociatedElement.h:
(FormAssociatedElement): Make name() virtual, remove formControlName().
* html/FormController.cpp:
(WebCore::FormController::formElementsState):
Use name() instead of formControlName().
* html/HTMLFormControlElement.cpp:
Remove formControlName() and setName().
* html/HTMLFormControlElement.h: ditto.
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::name): Renamed from formControlName().
* html/HTMLInputElement.h:
(HTMLInputElement): Renamed formControlName() to name(), and make it public.
* html/HTMLKeygenElement.idl: Make 'name' [Reflected].
* html/HTMLObjectElement.cpp: Remove formControlName().
* html/HTMLObjectElement.h: ditto.
* html/HTMLSelectElement.cpp:
(WebCore::HTMLSelectElement::appendFormData):
Use name() instead of formControlName().
* html/HTMLSelectElement.idl: Make 'name' [Reflected].
* html/HTMLTextAreaElement.idl: Make 'name' [Reflected].
2012-06-11 Nico Weber <thakis@chromium.org>
Remove unused member variables found by clang's -Wunused-private-field
https://bugs.webkit.org/show_bug.cgi?id=88812
Reviewed by Anders Carlsson.
Change a few |class|s that were used only to pin down the size of
existing classes to |struct|, so that clang doesn't warn about these
memvars.
No intended behavior change.
* Modules/geolocation/GeolocationController.cpp:
(WebCore::GeolocationController::GeolocationController):
* Modules/geolocation/GeolocationController.h:
(GeolocationController):
* Modules/webaudio/AudioContext.h:
(AudioContext):
* Modules/webaudio/AudioParamTimeline.h:
(AudioParamTimeline):
* accessibility/AccessibilityScrollbar.h:
(AccessibilityScrollbar):
* bindings/v8/ScriptScope.cpp:
(WebCore::ScriptScope::ScriptScope):
* bindings/v8/ScriptScope.h:
(ScriptScope):
* css/CSSValue.cpp:
* css/StylePropertySet.cpp:
* dom/DeviceOrientationController.cpp:
(WebCore::DeviceOrientationController::DeviceOrientationController):
* dom/DeviceOrientationController.h:
(DeviceOrientationController):
* editing/ReplaceSelectionCommand.cpp:
(ReplacementFragment):
(WebCore::ReplacementFragment::ReplacementFragment):
* editing/TextIterator.cpp:
(WebCore::SimplifiedBackwardsTextIterator::SimplifiedBackwardsTextIterator):
* editing/TextIterator.h:
(SimplifiedBackwardsTextIterator):
(BackwardsCharacterIterator):
* html/HTMLFormCollection.cpp:
(WebCore::HTMLFormCollection::HTMLFormCollection):
* html/HTMLFormCollection.h:
(HTMLFormCollection):
* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlFullscreenButtonElement::MediaControlFullscreenButtonElement):
* html/shadow/MediaControlElements.h:
(MediaControlFullscreenButtonElement):
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlRootElementChromium::MediaControlRootElementChromium):
* html/shadow/MediaControlRootElementChromium.h:
(MediaControlRootElementChromium):
* html/track/LoadableTextTrack.cpp:
(WebCore::LoadableTextTrack::LoadableTextTrack):
* html/track/LoadableTextTrack.h:
(LoadableTextTrack):
* inspector/CodeGeneratorInspector.py:
(Generator.go):
* inspector/InspectorCSSAgent.cpp:
(WebCore::InspectorCSSAgent::InspectorCSSAgent):
* inspector/InspectorCSSAgent.h:
(InspectorCSSAgent):
* inspector/InspectorDOMDebuggerAgent.cpp:
(WebCore::InspectorDOMDebuggerAgent::InspectorDOMDebuggerAgent):
* inspector/InspectorDOMDebuggerAgent.h:
(InspectorDOMDebuggerAgent):
* inspector/InspectorMemoryAgent.cpp:
(WebCore::InspectorMemoryAgent::InspectorMemoryAgent):
* inspector/InspectorMemoryAgent.h:
(InspectorMemoryAgent):
* notifications/NotificationController.cpp:
(WebCore::NotificationController::NotificationController):
* notifications/NotificationController.h:
(NotificationController):
* page/animation/CompositeAnimation.h:
(WebCore::CompositeAnimation::CompositeAnimation):
(CompositeAnimation):
* platform/Length.cpp:
* platform/audio/ReverbConvolver.cpp:
(WebCore::ReverbConvolver::ReverbConvolver):
* platform/audio/ReverbConvolver.h:
(ReverbConvolver):
* platform/audio/ReverbConvolverStage.cpp:
(WebCore::ReverbConvolverStage::ReverbConvolverStage):
* platform/audio/ReverbConvolverStage.h:
(ReverbConvolverStage):
* platform/graphics/TextRun.cpp:
(ExpectedTextRunSize):
* platform/graphics/chromium/cc/CCThreadProxy.h:
(CCThreadProxy):
* platform/graphics/filters/FECustomFilter.cpp:
(WebCore::FECustomFilter::FECustomFilter):
* platform/graphics/filters/FECustomFilter.h:
(FECustomFilter):
* platform/image-decoders/jpeg/JPEGImageDecoder.cpp:
(JPEGImageReader):
* platform/leveldb/LevelDBTransaction.cpp:
(WebCore::LevelDBTransaction::TreeIterator::next):
* platform/text/TextCodecICU.cpp:
(WebCore::TextCodecICU::TextCodecICU):
* platform/text/TextCodecICU.h:
(TextCodecICU):
* rendering/InlineBox.cpp:
* rendering/InlineFlowBox.cpp:
* rendering/RenderText.cpp:
* rendering/style/KeyframeList.h:
(WebCore::KeyframeList::KeyframeList):
(KeyframeList):
* rendering/svg/RenderSVGTextPath.cpp:
(WebCore::RenderSVGTextPath::RenderSVGTextPath):
* rendering/svg/RenderSVGTextPath.h:
* rendering/svg/SVGInlineTextBox.cpp:
* svg/SVGPathByteStreamSource.cpp:
* svg/SVGPathByteStreamSource.h:
* svg/SVGPathTraversalStateBuilder.cpp:
(WebCore::SVGPathTraversalStateBuilder::SVGPathTraversalStateBuilder):
* svg/SVGPathTraversalStateBuilder.h:
(SVGPathTraversalStateBuilder):
* xml/XSLStyleSheet.h:
2012-06-11 Kinuko Yasuda <kinuko@chromium.org>
XHR returns size==0 Blob
https://bugs.webkit.org/show_bug.cgi?id=88750
Reviewed by Alexey Proskuryakov.
Response Blob's .size field must have the correct response size.
Test: http/tests/xmlhttprequest/response-blob-size.html
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::responseBlob):
2012-06-11 Silvia Pfeiffer <silviapf@chromium.org>
Introduce an Enclosure Element for Chromium video controls.
https://bugs.webkit.org/show_bug.cgi?id=87683
Reviewed by Eric Carlson.
Updated tests.
The Chrome video controls are receiving a visual update. A new enclosure div is required
to provide for a offset space from the video's boundaries. The visual update itself is in
a separate patch.
* css/mediaControlsChromium.css:
(video::-webkit-media-controls-enclosure):
Introduce CSS for enclosure div so it doesn't show yet.
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlChromiumEnclosureElement::MediaControlChromiumEnclosureElement):
Definition of the new enclosure element.
(WebCore::MediaControlChromiumEnclosureElement::create):
Definition of a create operator for the new enclosure element.
(WebCore::MediaControlChromiumEnclosureElement::shadowPseudoId):
Definition of the shadowPseudoId of -webkit-media-controls-enclosure for the enclosure div.
(WebCore::MediaControlRootElementChromium::MediaControlRootElementChromium):
Initialize member element for enclosure div.
(WebCore::MediaControlRootElementChromium::create):
Instantiate enclosure div and add into DOM between controls and panel elements.
(WebCore::MediaControlRootElementChromium::setMediaController):
Set media controller for enclosure div.
(WebCore::MediaControlRootElementChromium::createTextTrackDisplay):
Insert text track display container to enclosure rather than panel.
* html/shadow/MediaControlRootElementChromium.h:
(WebCore):
(MediaControlChromiumEnclosureElement):
(WebCore::MediaControlChromiumEnclosureElement::setMediaController):
(WebCore::MediaControlChromiumEnclosureElement::mediaController):
(WebCore::MediaControlChromiumEnclosureElement::isMediaControlElement):
Add declaration of the new enclosure element and its member functions.
(MediaControlRootElementChromium):
Add member variable to controls root element to hold the enclosure element.
2012-06-11 Elliott Sprehn <esprehn@gmail.com>
WebKit doesn't allow replacing the document element with a DocumentFragment containing one element
https://bugs.webkit.org/show_bug.cgi?id=88681
Reviewed by Ojan Vafai.
Fix bug where replacing an immediate child of a Document with a DocumentFragment
would throw a HIERARCHY_REQUEST_ERR instead of replacing the node as required by DOM4.
http://www.w3.org/TR/domcore/#mutation-algorithms
* dom/Document.cpp:
(WebCore::Document::canReplaceChild):
When the newChild was a DocumentFragment we incorrectly
iterated over the document's children twice instead of the
fragment when counting the total number of elements and
doctypes to enfoce that there's only one of each.
2012-06-11 Mary Wu <mary.wu@torchmobile.com.cn>
[BlackBerry] Add handling of notifyDataReceived in NetworkJob even there's no data inside
https://bugs.webkit.org/show_bug.cgi?id=88773
Reviewed by Rob Buis.
Handling notifyDataReceived event even if there is no data inside.
This event was critical because it gives downloadFilterStream a chance
to follow up NetworkStream close event, and it also make pluginView in
FrameLoaderBlackBerry return normal logic.
RIM PR# 164563
Reviewed internally by Leo Yang.
* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::handleNotifyDataReceived):
2012-06-11 Mary Wu <mary.wu@torchmobile.com.cn>
[BlackBerry] Wrong mimetype for empty file link
https://bugs.webkit.org/show_bug.cgi?id=88642
Reviewed by Rob Buis.
RIM PR# 164164
For empty file link, we shouldn't give "application/octet-stream"
which will lead to download.
Reviewed internally by Charles Wei.
* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::sendResponseIfNeeded):
2012-06-11 Jin Yang <jin.a.yang@intel.com>
GeneratorGeneratedImage should cache images for the non-tiled case
https://bugs.webkit.org/show_bug.cgi?id=87094
Reviewed by Simon Fraser.
We cache the generated image if generator is not changed. By compared
with generating image on the fly, it will lose a little pixel precision
and several layout tests should be rebaselined. With this patch, the IE test
drive benchmark "Bayou" can gain about 50%.
No new tests. Performance optimization.
* platform/graphics/GeneratorGeneratedImage.cpp:
(WebCore::GeneratorGeneratedImage::draw):
2012-06-11 Vincent Scheib <scheib@chromium.org>
Add new Pointer Lock spec attribute webkitPointerLockElement.
https://bugs.webkit.org/show_bug.cgi?id=88799
Reviewed by Dimitri Glazkov.
Part of a series of refactoring changes to update pointer lock API to
the fullscreen locking style. https://bugs.webkit.org/show_bug.cgi?id=84402
New attribute webkitPointerLockElement added. Follow up patches
will remove the previous isLocked attribute. Tests updated to use
the new attribute.
* bindings/generic/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::webkitPointerLockElementEnabled):
* dom/Document.cpp:
(WebCore):
(WebCore::Document::webkitPointerLockElement):
* dom/Document.h:
(Document):
* dom/Document.idl:
* page/PointerLockController.h:
(WebCore::PointerLockController::element):
2012-06-11 Vincent Scheib <scheib@chromium.org>
Consolidate Pointer Lock runtime enabled flags to just one.
https://bugs.webkit.org/show_bug.cgi?id=88810
Reviewed by Dimitri Glazkov.
No new tests.
* bindings/generic/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::pointerLockEnabled):
(WebCore::RuntimeEnabledFeatures::setPointerLockEnabled):
* dom/MouseEvent.idl:
* page/Navigator.idl:
2012-06-11 Alexis Menard <alexis.menard@openbossa.org>
[CSS3 Backgrounds and Borders] Protect box-decoration-break behind a feature flag.
https://bugs.webkit.org/show_bug.cgi?id=88804
Reviewed by Tony Chang.
Protect box-decoration-break behind a feature flag enabled by default.
No new tests : no behavior change here.
* Configurations/FeatureDefines.xcconfig:
* GNUmakefile.am:
* css/CSSComputedStyleDeclaration.cpp:
(WebCore):
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
(WebCore::isKeywordPropertyID):
(WebCore::CSSParser::parseValue):
* css/CSSPrimitiveValueMappings.h:
(WebCore):
* css/CSSProperty.cpp:
(WebCore::CSSProperty::isInheritedProperty):
* css/CSSValueKeywords.in:
* css/StyleBuilder.cpp:
(WebCore::StyleBuilder::StyleBuilder):
* rendering/style/RenderStyle.h:
* rendering/style/StyleBoxData.cpp:
(WebCore::StyleBoxData::StyleBoxData):
(WebCore::StyleBoxData::operator==):
* rendering/style/StyleBoxData.h:
(StyleBoxData):
2012-06-11 James Robinson <jamesr@chromium.org>
[chromium] Port DrawingBufferChromium from TextureLayerChromium over to WebExternalTextureLayer
https://bugs.webkit.org/show_bug.cgi?id=86273
Reviewed by Adrienne Walker.
This converts more WebCore code over to using public APIs instead of internal compositor layer types.
Refactor only, no change in behavior thus no new tests.
* platform/graphics/chromium/DrawingBufferChromium.cpp:
(WebCore::DrawingBufferPrivate::DrawingBufferPrivate):
(WebCore::DrawingBufferPrivate::~DrawingBufferPrivate):
(WebCore::DrawingBufferPrivate::layer):
(DrawingBufferPrivate):
* platform/graphics/chromium/TextureLayerChromium.cpp:
(WebCore::TextureLayerChromium::~TextureLayerChromium):
(WebCore::TextureLayerChromium::setRateLimitContext):
(WebCore::TextureLayerChromium::setNeedsDisplayRect):
(WebCore::TextureLayerChromium::update):
* platform/graphics/chromium/TextureLayerChromium.h:
(WebKit):
(TextureLayerChromiumClient):
2012-06-11 Max Feil <mfeil@rim.com>
[BlackBerry] Seek calls are being unnecessarily delayed
https://bugs.webkit.org/show_bug.cgi?id=88732
Reviewed by Antonio Gomes.
There is a problem with the way the m_userDrivenSeekTimer is
implemented. When MediaPlayerPrivate::seek() is called, there
is always a 100ms delay even if the timer is not running. The
timer is supposed to space out (i.e. throttle) repeated seeks
that come in too soon after a previous seek, but currently it
is slowing down even single seeks and seeks that come in with
adequate delay after a previous seek. I fixed this in my patch
by improving the way the timer fired function is called.
A note on the new m_lastSeekTimePending flag: This flag is
needed so that userDrivenSeekTimerFired() knows whether or not
to perform the seek. The only case where this flag will not be
set is if no MediaPlayerPrivate::seek() call came in while the
timer was active, in which case it's important to do nothing.
I could encode this flag's information into the m_lastSeekTime
float, for example by initializing it and resetting it to NAN
and using isnan(). But I feel that using a separate bool is a
more portable approach.
No new tests. I would like to propose not including a layout test
with this fix. Doing timing tests for 100ms delays is tricky
from Javascript, and I don't think the benefit of such a test
outweighs the extra time it would take to develop one. The test
would also be a problem to maintain as it may give different
results over different runs and across different target hardware.
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp:
(WebCore::MediaPlayerPrivate::MediaPlayerPrivate):
(WebCore::MediaPlayerPrivate::seek):
(WebCore::MediaPlayerPrivate::userDrivenSeekTimerFired):
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.h:
(MediaPlayerPrivate):
2012-06-11 Max Feil <mfeil@rim.com>
[BlackBerry] Unexpected repeats of short media
https://bugs.webkit.org/show_bug.cgi?id=88733
Reviewed by Antonio Gomes.
The m_userDrivenSeekTimer is causing unwanted repeats of short
media such as sound effects because it is causing the current
time to not reflect that the media has finished playing.
This problem only affects media whose duration is close to
or less than the SeekSubmissionDelay, which is currently
set to 100ms. My fix is to ignore the userDrivenSeekTimer in
MediaPlayerPrivate::currentTime() if the duration of the media
is within twice the SeekSubmissionDelay. Seek drag smoothness
is a non-issue for such short media.
Test: platform/blackberry/media/short-media-repeats-correctly.html
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp:
(WebCore::MediaPlayerPrivate::MediaPlayerPrivate):
(WebCore):
(WebCore::MediaPlayerPrivate::currentTime):
(WebCore::MediaPlayerPrivate::seek):
(WebCore::MediaPlayerPrivate::userDrivenSeekTimerFired):
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.h:
(MediaPlayerPrivate):
2012-06-11 Julien Chaffraix <jchaffraix@webkit.org>
Account for margin after when laying out <legend> element
https://bugs.webkit.org/show_bug.cgi?id=35981
Reviewed by Abhishek Arya.
Tests: fast/forms/legend-after-margin-horizontal-writing-mode.html
fast/forms/legend-after-margin-vertical-writing-mode.html
fast/forms/legend-after-margin-with-before-border-horizontal-mode.html
fast/forms/legend-small-after-margin-before-border-horizontal-mode.html
The existing code would ignore margin after when layouting out the <legend>. This
change only adds the code to handle the margin after, the margin before is still
ignored as it's not obvious how it should be working.
* rendering/RenderFieldset.cpp:
(WebCore::RenderFieldset::layoutSpecialExcludedChild):
Split the code in 2 code paths to reflect how we position and size. Those are covered by the
tests above.
2012-06-11 James Robinson <jamesr@chromium.org>
[chromium] Use WebGraphicsContext3D in rate limiting logic inside compositor
https://bugs.webkit.org/show_bug.cgi?id=86259
Reviewed by Adrienne Walker.
This refactors the compositor's rate limiting implementation to use the Platform API's WebGraphicsContext3D
directly instead of WebCore::GraphicsContext3D to cut down on the number of spurious WebCore dependencies in the
compositor. The one change in contract is that the caller to CCLayerTreeHost::startRateLimit() now has to call
CCLayerTreeHost::stopRateLimit() before allowing the referenced context to go away since CCLayerTreeHost no
longer retains a reference, but this was always happening already.
* platform/graphics/chromium/Canvas2DLayerChromium.cpp:
(WebCore::Canvas2DLayerChromium::~Canvas2DLayerChromium):
(WebCore::Canvas2DLayerChromium::setNeedsDisplayRect):
* platform/graphics/chromium/RateLimiter.cpp:
(WebCore::RateLimiter::create):
(WebCore::RateLimiter::RateLimiter):
(WebCore::RateLimiter::start):
(WebCore::RateLimiter::rateLimitContext):
* platform/graphics/chromium/RateLimiter.h:
(WebKit):
(RateLimiter):
* platform/graphics/chromium/TextureLayerChromium.cpp:
(WebCore::TextureLayerChromium::~TextureLayerChromium):
(WebCore::TextureLayerChromium::setRateLimitContext):
(WebCore::TextureLayerChromium::setNeedsDisplayRect):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::startRateLimiter):
(WebCore::CCLayerTreeHost::stopRateLimiter):
(WebCore::CCLayerTreeHost::rateLimit):
(WebCore):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(CCLayerTreeHost):
2012-06-11 Dana Jansens <danakj@chromium.org>
[chromium] Separate CCVideoDrawQuad and from the layer tree and video provider by removing ManagedTexture and WebVideoFrame pointers from the quad
https://bugs.webkit.org/show_bug.cgi?id=88363
Reviewed by Adrienne Walker.
CCVideoDrawQuad should not contain any pointers to data in the layer
tree so we are able to serialize it across process boundaries. This
patch removes the ManagedTexture pointers from the quad class,
replacing them with texture ids. It removes the WebVideoFrame* from
the quad, replacing it with the frame provider's texture id included
in the WebVideoFrame structure. And it uses a WebTransformationMatrix
instead of a pointer to an array of floats.
Texture allocation is done in CCVideoLayerImpl via the
contentsTextureAllocator, so that the memory usage can be tracked.
We move the copyPlaneToTextures() method back from LayerRendererChromium
to CCVideoLayerImpl, as this method uses the texture data pointer in the
WebVideoFrame, and we do not want to give this pointer to the quad
class. Instead, this method makes use of the LayerTextureSubImage class
to copy the pixel data into the texture.
LayerTextureSubImage is updated to allow non-4byte texture formats.
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::drawYUV):
(WebCore::LayerRendererChromium::drawRGBA):
(WebCore::LayerRendererChromium::drawNativeTexture2D):
(WebCore::LayerRendererChromium::drawStreamTexture):
(WebCore::LayerRendererChromium::drawVideoQuad):
* platform/graphics/chromium/LayerTextureSubImage.cpp:
(WebCore::LayerTextureSubImage::uploadWithTexSubImage):
(WebCore::LayerTextureSubImage::uploadWithMapTexSubImage):
* platform/graphics/chromium/TextureCopier.cpp:
* platform/graphics/chromium/cc/CCVideoDrawQuad.cpp:
(WebCore::CCVideoDrawQuad::create):
(WebCore::CCVideoDrawQuad::CCVideoDrawQuad):
* platform/graphics/chromium/cc/CCVideoDrawQuad.h:
(CCVideoDrawQuad):
(WebCore::CCVideoDrawQuad::planes):
(WebCore::CCVideoDrawQuad::frameProviderTextureId):
* platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
(WebCore::CCVideoLayerImpl::~CCVideoLayerImpl):
(WebCore::CCVideoLayerImpl::willDraw):
(WebCore::CCVideoLayerImpl::willDrawInternal):
(WebCore::CCVideoLayerImpl::appendQuads):
(WebCore::CCVideoLayerImpl::didDraw):
(WebCore::CCVideoLayerImpl::FramePlane::allocateData):
(WebCore):
(WebCore::CCVideoLayerImpl::FramePlane::freeData):
(WebCore::CCVideoLayerImpl::allocatePlaneData):
(WebCore::CCVideoLayerImpl::copyPlaneData):
(WebCore::CCVideoLayerImpl::freePlaneData):
(WebCore::CCVideoLayerImpl::freeUnusedPlaneData):
(WebCore::CCVideoLayerImpl::didLoseContext):
* platform/graphics/chromium/cc/CCVideoLayerImpl.h:
(FramePlane):
(WebCore::CCVideoLayerImpl::FramePlane::FramePlane):
2012-06-11 Joshua Bell <jsbell@chromium.org>
IndexedDB: Object stores are not successfully deleted
https://bugs.webkit.org/show_bug.cgi?id=88788
Reviewed by Tony Chang.
Discovered while working on http://webkit.org/b/83074 - object stores are not
being deleted from the backing store. Deletion would succeed in the in-memory
data structures, but after closing/re-opening the backing store the object store
(but not its indexes/data) would reappear. Due to 83074, this isn't detectable
by DRT without a full restart.
Test: [chromium] webkit_unit_tests --gtest_filter='IDBLevelDBCodingTest.ComparisonTest'
* Modules/indexeddb/IDBLevelDBCoding.cpp:
(WebCore::IDBLevelDBCoding::ObjectStoreMetaDataKey::compare):
2012-06-05 Eric Uhrhane <ericu@chromium.org>
Crash in fast/files/read tests during Garbage Collection
https://bugs.webkit.org/show_bug.cgi?id=87165
Reviewed by Michael Saboff
Fix previous fix for hasPendingActivity, and fix a bug in a complex
abort case as well--abort during the final progress event of a write
would hang the writer.
* Modules/filesystem/FileWriter.cpp:
(WebCore::FileWriter::stop):
(WebCore::FileWriter::write):
(WebCore::FileWriter::truncate):
(WebCore::FileWriter::didWrite):
(WebCore::FileWriter::didTruncate):
(WebCore::FileWriter::didFail):
(WebCore::FileWriter::completeAbort):
(WebCore::FileWriter::doOperation):
(WebCore::FileWriter::signalCompletion):
2012-06-11 Shawn Singh <shawnsingh@chromium.org>
[chromium] Implement position:fixed in compositor thread
https://bugs.webkit.org/show_bug.cgi?id=70103
Reviewed by Adrienne Walker.
Significant contributions to this patch by Alpha Lam and Sami Kyostila.
This patch is the compositor-side change that adds support for
fixed-position layers to be composited layers. Before this patch,
fixed-position elements were positioned only by WebCore (i.e. main
thread when painting), and the compositor did not have enough
knowledge on its own to position it properly. This patch adds the
necessary math and plumbs the necessary layer information so that
the impl thread can properly position fixed-position elements on
its own. This support is necessary for correctness of composited
fixed-position elements, which can greatly help to avoid
repainting container layers when scrolling with a fixed-position
layer.
There also needs to be WebCore-side support for this, which is
addressed in https://bugs.webkit.org/show_bug.cgi?id=78864
Unit tests added to CCLayerTreeHostCommonTest:
CCLayerTreeHostCommonTest.verifyScrollCompensationForFixedPositionLayerWithDirectContainer
CCLayerTreeHostCommonTest.verifyScrollCompensationForFixedPositionLayerWithTransformedDirectContainer
CCLayerTreeHostCommonTest.verifyScrollCompensationForFixedPositionLayerWithDistantContainer
CCLayerTreeHostCommonTest.verifyScrollCompensationForFixedPositionLayerWithDistantContainerAndTransforms
CCLayerTreeHostCommonTest.verifyScrollCompensationForFixedPositionLayerWithMultipleScrollDeltas
CCLayerTreeHostCommonTest.verifyScrollCompensationForFixedPositionLayerWithIntermediateSurfaceAndTransforms
CCLayerTreeHostCommonTest.verifyScrollCompensationForFixedPositionLayerWithMultipleIntermediateSurfaces
CCLayerTreeHostCommonTest.verifyScrollCompensationForFixedPositionLayerWithContainerLayerThatHasSurface
CCLayerTreeHostCommonTest.verifyScrollCompensationForFixedPositionLayerThatIsAlsoFixedPositionContainer
CCLayerTreeHostCommonTest.verifyScrollCompensationForFixedPositionLayerThatHasNoContainer
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::LayerChromium):
(WebCore::LayerChromium::pushPropertiesTo):
* platform/graphics/chromium/LayerChromium.h:
(WebCore::LayerChromium::setIsContainerForFixedPositionLayers):
(WebCore::LayerChromium::isContainerForFixedPositionLayers):
(LayerChromium):
(WebCore::LayerChromium::setFixedToContainerLayerVisibleRect):
(WebCore::LayerChromium::fixedToContainerLayerVisibleRect):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(WebCore::CCLayerImpl::setIsContainerForFixedPositionLayers):
(WebCore::CCLayerImpl::isContainerForFixedPositionLayers):
(CCLayerImpl):
(WebCore::CCLayerImpl::setFixedToContainerLayerVisibleRect):
(WebCore::CCLayerImpl::fixedToContainerLayerVisibleRect):
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::computeScrollCompensationForThisLayer):
(WebCore):
(WebCore::computeScrollCompensationMatrixForChildren):
(WebCore::calculateDrawTransformsInternal):
(WebCore::CCLayerTreeHostCommon::calculateDrawTransforms):
2012-06-11 Sam Weinig <sam@webkit.org>
Remove support for disconnected/excluded from search frames, they are not used by Safari anymore
https://bugs.webkit.org/show_bug.cgi?id=88723
Reviewed by Dan Bernstein.
* WebCore.exp.in:
Update export.
* editing/Editor.cpp:
(WebCore::Editor::rangeOfString):
(WebCore::Editor::countMatchesForText):
* editing/Editor.h:
(Editor):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::findFrameForNavigation):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::parent):
(WebCore::DOMWindow::top):
* page/EventHandler.cpp:
(WebCore::EventHandler::hitTestResultAtPoint):
* page/Frame.cpp:
(WebCore::Frame::Frame):
* page/Frame.h:
(Frame):
* page/FrameTree.cpp:
(WebCore::FrameTree::parent):
(WebCore::FrameTree::top):
* page/FrameTree.h:
(FrameTree):
* page/Location.cpp:
(WebCore::Location::ancestorOrigins):
* xml/XMLTreeViewer.cpp:
(WebCore::XMLTreeViewer::hasNoStyleInformation):
Update for the removal of disconnected frames and text search exclusions concepts.
2012-06-11 Xianzhu Wang <wangxianzhu@chromium.org>
SVGImageCache leaks image data
https://bugs.webkit.org/show_bug.cgi?id=87792
There are two functions to remove a client from a CachedImage:
- CachedResource::removeClient()
- CachedImage::removeClientForRenderer().
It's easy to make error to call the former which will leak the cached
image buffers in SVGImageCache.
This change combined the two by adding the virtual
CachedResource::didRemoveClient(). CachedImage will do SVGImageCache
cleanup in the function.
Reviewed by Nikolas Zimmermann.
Test: svg/as-image/svg-image-leak-cached-data.html
* loader/cache/CachedFont.h:
(WebCore::CachedFontClient::resourceClientType): Added 'const'.
* loader/cache/CachedImage.cpp:
(WebCore):
(WebCore::CachedImage::didRemoveClient): Removes the client from SVGImageCache.
(WebCore::CachedImage::lookupOrCreateImageForRenderer):
* loader/cache/CachedImage.h:
(CachedImage):
(WebCore::CachedImageClient::resourceClientType): Added 'const'.
* loader/cache/CachedRawResource.h:
(WebCore::CachedRawResourceClient::resourceClientType): Added 'const'.
* loader/cache/CachedResource.cpp:
(WebCore::CachedResource::removeClient): Added invocation of didRemoveClient().
* loader/cache/CachedResource.h:
(WebCore::CachedResource::didRemoveClient): Added for subclasses to do additional works.
* loader/cache/CachedResourceClient.h:
(WebCore::CachedResourceClient::resourceClientType): Added 'const'.
* loader/cache/CachedSVGDocument.h:
(WebCore::CachedSVGDocumentClient::resourceClientType): Added 'const'.
* loader/cache/CachedStyleSheetClient.h:
(WebCore::CachedStyleSheetClient::resourceClientType): Added 'const'.
* rendering/style/StyleCachedImage.cpp:
(WebCore::StyleCachedImage::removeClient):
* rendering/style/StyleCachedImageSet.cpp:
(WebCore::StyleCachedImageSet::removeClient):
* svg/graphics/SVGImageCache.cpp:
(WebCore::SVGImageCache::~SVGImageCache): Added checking for leaks.
(WebCore::SVGImageCache::removeClientFromCache):
(WebCore::SVGImageCache::setRequestedSizeAndScales):
(WebCore::SVGImageCache::requestedSizeAndScales):
(WebCore::SVGImageCache::lookupOrCreateBitmapImageForClient):
* svg/graphics/SVGImageCache.h:
(WebCore):
(SVGImageCache):
2012-06-11 Mark Pilgrim <pilgrim@chromium.org>
[Chromium] Call shared timer functions directly
https://bugs.webkit.org/show_bug.cgi?id=88781
Reviewed by Adam Barth.
Part of a refactoring series. See tracking bug 82948.
* platform/chromium/PlatformSupport.h:
(PlatformSupport):
* platform/chromium/SharedTimerChromium.cpp:
(WebCore::setSharedTimerFiredFunction):
(WebCore::setSharedTimerFireInterval):
2012-06-11 Pravin D <pravind.2k4@gmail.com>
Relative pos. input fields in columns vanish when you start typing in them
https://bugs.webkit.org/show_bug.cgi?id=76834
Reviewed by Julien Chaffraix.
Test: fast/multicol/multicol-with-child-renderLayer-for-input.html
* page/FrameView.cpp:
(WebCore::updateLayerPositionFlags):
Helper function to prepare the UpdateLayerPositionsFlags based on the input parameters.
If didFullRepaint flag is set, then CheckForRepaints flag is removed from the default flags.
Also during a relayout of a subtree, if the RenderLayer of the subtree root is paginated then updatePagination flag is
included in the flags.
(WebCore):
(WebCore::FrameView::layout):
Now uses the helper function updateLayerPositionFlags() to get the final set of UpdateLayerPositionsFlags.
* rendering/RenderLayer.h:
(WebCore::RenderLayer::isPaginated):
(RenderLayer):
isPaginated function is made public so that FrameView class can use it prepare UpdateLayerPositionsFlags.
2012-06-11 Mark Pilgrim <pilgrim@chromium.org>
[Chromium] Remove some dead code in PasteboardChromium
https://bugs.webkit.org/show_bug.cgi?id=88782
Reviewed by Adam Barth.
Part of a refactoring series. See tracking bug 82948.
Leftover detritus from bug 88038.
* platform/chromium/PasteboardChromium.cpp:
(WebCore::Pasteboard::documentFragment):
2012-06-11 Peter Beverloo <peter@chromium.org>
[Chromium] Theme updates for Android in menu list rendering and selection backgrounds
https://bugs.webkit.org/show_bug.cgi?id=88775
Reviewed by Adam Barth.
For Chrome on Android, use the width of the scrollbar down arrow instead
of the scrollbar's width for determining the arrow padding to apply for
menu list rendering, as WebKit isn't drawing a scrollbar (thus width=0).
Furthermore, change the default active selection background color to be
equal to the tap highlighting color.
These changes should be covered by existing layout tests.
* rendering/RenderThemeChromiumAndroid.cpp:
(WebCore::RenderThemeChromiumAndroid::menuListArrowPadding):
(WebCore): Retrieve the scrollbar down arrow's size via PlatformSupport.
* rendering/RenderThemeChromiumAndroid.h: Override the menuListArrowPadding
and platformActiveSelectionBackgroundColor methods, and add a static
static RGBA32 color for the default active selection bg color.
* rendering/RenderThemeChromiumSkia.cpp:
(WebCore::RenderThemeChromiumSkia::menuListArrowPadding): Default to
the scrollbar thickness, not changing behavior for non-Android.
(WebCore::RenderThemeChromiumSkia::menuListInternalPadding): Instead
of polling the scrollbar thickness, call menuListArrowPadding().
* rendering/RenderThemeChromiumSkia.h:
(RenderThemeChromiumSkia): Add the menuListArrowPadding() as a protected method.
2012-06-11 David Barr <davidbarr@chromium.org>
Add css3-images image-resolution (dppx only)
https://bugs.webkit.org/show_bug.cgi?id=85332
Reviewed by Tony Chang.
The css3-images module is at candidate recommendation.
http://www.w3.org/TR/2012/CR-css3-images-20120417/#image-resolution
Test: fast/css/image-resolution.html
* css/CSSComputedStyleDeclaration.cpp:
(WebCore):
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSGrammar.y:
* css/CSSParser.cpp:
(WebCore::CSSParser::validUnit):
(WebCore::CSSParser::createPrimitiveNumericValue):
(WebCore::unitFromString):
(WebCore::CSSParser::parseValidPrimitive):
(WebCore::CSSParser::parseValue):
(WebCore):
(WebCore::CSSParser::parseImageResolution):
(WebCore::CSSParser::detectNumberToken):
* css/CSSParser.h:
* css/CSSPrimitiveValue.cpp:
(WebCore::isValidCSSUnitTypeForDoubleConversion):
(WebCore::unitCategory):
(WebCore::CSSPrimitiveValue::canonicalUnitTypeForCategory):
(WebCore::CSSPrimitiveValue::customCssText):
(WebCore::CSSPrimitiveValue::cloneForCSSOM):
* css/CSSPrimitiveValue.h:
* css/CSSProperty.cpp:
(WebCore::CSSProperty::isInheritedProperty):
* css/CSSPropertyNames.in:
* css/StyleBuilder.cpp:
(WebCore):
(ApplyPropertyImageResolution):
(WebCore::ApplyPropertyImageResolution::applyInheritValue):
(WebCore::ApplyPropertyImageResolution::applyInitialValue):
(WebCore::ApplyPropertyImageResolution::applyValue):
(WebCore::ApplyPropertyImageResolution::createHandler):
(WebCore::StyleBuilder::StyleBuilder):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* rendering/RenderImage.cpp:
(WebCore::RenderImage::styleDidChange):
(WebCore::RenderImage::imageDimensionsChanged):
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::diff):
* rendering/style/RenderStyle.h:
* rendering/style/StyleRareInheritedData.cpp:
(WebCore::StyleRareInheritedData::StyleRareInheritedData):
(WebCore::StyleRareInheritedData::operator==):
* rendering/style/StyleRareInheritedData.h:
(StyleRareInheritedData):
2012-06-05 Dana Jansens <danakj@chromium.org>
[chromium] Free texture from CCIOSurfaceLayerImpl when it is destroyed
https://bugs.webkit.org/show_bug.cgi?id=88371
Reviewed by James Robinson.
Unit test: CCLayerTreeHostImplTest.layersFreeTextures
* platform/graphics/chromium/cc/CCIOSurfaceLayerImpl.cpp:
(WebCore::CCIOSurfaceLayerImpl::~CCIOSurfaceLayerImpl):
(WebCore::CCIOSurfaceLayerImpl::willDraw):
2012-06-11 Carlos Garcia Campos <cgarcia@igalia.com>
Unreviewed. Fix make distcheck issues.
* GNUmakefile.am: Add idl files in editing dir to EXTRA_DIST.
* GNUmakefile.list.am: Add missing header file.
2012-06-11 Dan Bernstein <mitz@apple.com>
Reverted r119940 because it caused multiple media tests to fail on Lion.
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::prepareForLoad):
(WebCore):
(WebCore::HTMLMediaElement::userCancelledLoad):
* html/HTMLMediaElement.h:
(HTMLMediaElement):
* platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
(WebCore::MediaPlayerPrivateQTKit::commonMovieAttributes):
2012-06-11 Zoltan Horvath <zoltan@webkit.org>
[Qt] Add config tests for WEBP imagedecoder library, modify HAVE(decoderlibrary) to USE(...)$
https://bugs.webkit.org/show_bug.cgi?id=87841
Reviewed by Simon Hausmann.
No new tests were needed.
* Target.pri:
* WebCore.pri:
* platform/image-decoders/ImageDecoder.cpp:
(WebCore::ImageDecoder::create):
2012-06-11 Kenneth Rohde Christiansen <kenneth@webkit.org>
Add QML api for setting device width and height
https://bugs.webkit.org/show_bug.cgi?id=88777
Reviewed by Simon Hausmann.
Change the default values for deviceWidth/Height to be 0 (unset). The
original values were introduced by Qt and 0 makes it possible to
detect if a value has never been set and therefore fall back.
Only Qt depended on the former values.
* page/Settings.cpp:
(WebCore::Settings::Settings):
2012-06-11 Csaba Osztrogonác <ossy@webkit.org>
[Qt][Win] Fix UString related build problem in Source/WebCore/bridge/qt/qt_instance.cpp
https://bugs.webkit.org/show_bug.cgi?id=88310
Reviewed by Simon Hausmann.
* bridge/qt/qt_instance.cpp:
(JSC::Bindings::QtInstance::getPropertyNames):
2012-06-11 Michael Brüning <michael.bruning@nokia.com>
[Qt][WK2] Compute and set cache capacities using the current CacheModel
https://bugs.webkit.org/show_bug.cgi?id=73918
Unreviewed build fix.
No new tests added as this is a build fix.
* platform/qt/FileSystemQt.cpp:
2012-06-11 Allan Sandfeld Jensen <allan.jensen@nokia.com>
Rect-based hittesting doesn't work in tables.
https://bugs.webkit.org/show_bug.cgi?id=86605
Reviewed by Julien Chaffraix.
The existing code only performed a single binary lookup for the central
hit-test point. This meant area-based hit-testing did not work across
table-cell borders. All cells spanned by a hit-test area must be hit
tested.
This patch introduces three auxilary functions to help calculate rows
and columns spanned by an rect. These are intended to also be used by
repaint logic in a later patch.
For point-based hit-testing we maintain unchanged behavior and still
only hit test a single column in a single row.
Test: fast/dom/nodesFromRect-table.html
* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::logicalRectForWritingModeAndDirection):
(WebCore::RenderTableSection::spannedRows):
(WebCore::RenderTableSection::spannedColumns):
(WebCore::RenderTableSection::nodeAtPoint):
* rendering/RenderTableSection.h:
(RenderTableSection):
2012-06-11 Min Qin <qinmin@google.com>
Adding a flag to show fullscreen media controls in chromium
https://bugs.webkit.org/show_bug.cgi?id=88266
Reviewed by Adam Barth.
The default chromium shadow DOM for media element does not contain fullscreen button.
This change adds a flag to display the fullscreen button.
It will be used by the android port.
No new tests needed. We will use the same layout tests as desktop chromium.
However, new test result expections will be added later.
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlRootElementChromium::MediaControlRootElementChromium):
(WebCore::MediaControlRootElementChromium::create):
(WebCore::MediaControlRootElementChromium::setMediaController):
(WebCore::MediaControlRootElementChromium::reset):
(WebCore::MediaControlRootElementChromium::reportedError):
* html/shadow/MediaControlRootElementChromium.h:
(MediaControlRootElementChromium):
2012-06-11 Michael Brüning <michael.bruning@nokia.com>
[Qt][WK2] Compute and set cache capacities using the current CacheModel
https://bugs.webkit.org/show_bug.cgi?id=73918
Reviewed by Kenneth Rohde Christiansen.
No new tests needed / applicable (using operating system functionality).
Added method to read free file system space for a path to support cache model
implementation in Qt WK 2.
* platform/FileSystem.h:
(WebCore):
* platform/qt/FileSystemQt.cpp:
(WebCore::getVolumeFreeSizeForPath): Added for Qt ports.
(WebCore):
2012-06-11 Peter Wang <peter.wang@torchmobile.com.cn>
[JSC] Web Inspector: implement breaking from native callback
https://bugs.webkit.org/show_bug.cgi?id=43332
Reviewed by Pavel Feldman.
These test cases were enabled:
LayoutTests/inspector/debugger/dom-breakpoints.html
LayoutTests/inspector/debugger/event-listener-breakpoints.html
LayoutTests/inspector/debugger/step-through-event-listeners.html
LayoutTests/inspector/debugger/xhr-breakpoints.html
* bindings/js/ScriptDebugServer.cpp:
(WebCore::ScriptDebugServer::breakProgram):
* bindings/js/ScriptDebugServer.h:
(WebCore::supportsNativeBreakpoints):
2012-06-11 MORITA Hajime <morrita@google.com>
Spellchecker crash in async scenario.
https://bugs.webkit.org/show_bug.cgi?id=88617
Reviewed by Tony Chang.
SpellCheckRequest::invokeRequest() can be reached after the frame is detached.
SpellCheckRequest should ignore such a request since TextCheckerClient is not available
for detached Frames.
No new tests. It's hard to reproduce this.
* editing/SpellChecker.cpp:
(WebCore::SpellChecker::invokeRequest):
2012-06-11 Ryuan Choi <ryuan.choi@samsung.com>
[EFL] Extract CursorMap from WidgetEfl.cpp.
https://bugs.webkit.org/show_bug.cgi?id=88633
Reviewed by Ryosuke Niwa.
In order to share CursorMap in WebKit/efl and UIProcess of WebKit2/Efl,
extract it and expose getEcoreCursor.
No new tests, refactoring only.
* platform/efl/EflScreenUtilities.cpp:
(WebCore):
(CursorMap):
(WebCore::CursorMap::cursor):
(WebCore::CursorMap::CursorMap):
(WebCore::getEcoreCursor):
* platform/efl/EflScreenUtilities.h:
(WebCore):
* platform/efl/WidgetEfl.cpp:
(WebCore::Widget::applyFallbackCursor):
2012-06-10 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119955.
http://trac.webkit.org/changeset/119955
https://bugs.webkit.org/show_bug.cgi?id=88758
Breaks Chromium compile - Chromium depends on removed API
(Requested by dominicc on #webkit).
* WebCore.exp.in:
* editing/Editor.cpp:
(WebCore::Editor::insideVisibleArea):
(WebCore):
(WebCore::Editor::firstVisibleRange):
(WebCore::Editor::lastVisibleRange):
(WebCore::Editor::nextVisibleRange):
(WebCore::Editor::rangeOfString):
(WebCore::Editor::countMatchesForText):
* editing/Editor.h:
(Editor):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::findFrameForNavigation):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::parent):
(WebCore::DOMWindow::top):
* page/EventHandler.cpp:
(WebCore::EventHandler::hitTestResultAtPoint):
* page/Frame.cpp:
(WebCore::Frame::Frame):
* page/Frame.h:
(Frame):
(WebCore::Frame::isDisconnected):
(WebCore):
(WebCore::Frame::setIsDisconnected):
(WebCore::Frame::excludeFromTextSearch):
(WebCore::Frame::setExcludeFromTextSearch):
* page/FrameTree.cpp:
(WebCore::FrameTree::parent):
(WebCore::FrameTree::top):
* page/FrameTree.h:
(FrameTree):
* page/Location.cpp:
(WebCore::Location::ancestorOrigins):
* xml/XMLTreeViewer.cpp:
(WebCore::XMLTreeViewer::hasNoStyleInformation):
2012-06-08 Kinuko Yasuda <kinuko@chromium.org>
Using BlobBuilder should show a deprecation warning message (if it is enabled)
https://bugs.webkit.org/show_bug.cgi?id=88644
Reviewed by Jian Li.
Mac has already disabled BlobBuilder, but Chromium still has it.
We should start showing a warning to let people migrate to using Blob constructor.
Tests: fast/files/blob-builder-crash.html
* fileapi/WebKitBlobBuilder.cpp:
(WebCore):
(WebCore::WebKitBlobBuilder::create):
* fileapi/WebKitBlobBuilder.h:
(WebCore):
(WebKitBlobBuilder):
* fileapi/WebKitBlobBuilder.idl:
2012-06-10 Sam Weinig <sam@webkit.org>
Remove support for disconnected/excluded from search frames, they are not used by Safari anymore
https://bugs.webkit.org/show_bug.cgi?id=88723
Reviewed by Dan Bernstein.
* WebCore.exp.in:
Update export.
* editing/Editor.cpp:
(WebCore::Editor::rangeOfString):
(WebCore::Editor::countMatchesForText):
* editing/Editor.h:
(Editor):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::findFrameForNavigation):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::parent):
(WebCore::DOMWindow::top):
* page/EventHandler.cpp:
(WebCore::EventHandler::hitTestResultAtPoint):
* page/Frame.cpp:
(WebCore::Frame::Frame):
* page/Frame.h:
(Frame):
* page/FrameTree.cpp:
(WebCore::FrameTree::parent):
(WebCore::FrameTree::top):
* page/FrameTree.h:
(FrameTree):
* page/Location.cpp:
(WebCore::Location::ancestorOrigins):
* xml/XMLTreeViewer.cpp:
(WebCore::XMLTreeViewer::hasNoStyleInformation):
Update for the removal of disconnected frames and text search exclusions concepts.
2012-06-07 Kinuko Yasuda <kinuko@google.com>
Move Quota related code out of DOMWindow and into the quota/ folder
https://bugs.webkit.org/show_bug.cgi?id=88512
Reviewed by Adam Barth.
No new tests: no visible changes.
* CMakeLists.txt:
* DerivedSources.cpp:
* DerivedSources.make:
* DerivedSources.pri:
* Modules/quota/DOMWindowQuota.cpp: Added.
* Modules/quota/DOMWindowQuota.h: Added.
* Modules/quota/DOMWindowQuota.idl: Added.
* Modules/quota/StorageInfo.cpp: Renamed from Source/WebCore/storage/StorageInfo.cpp.
* Modules/quota/StorageInfo.h: Renamed from Source/WebCore/storage/StorageInfo.h.
* Modules/quota/StorageInfo.idl: Renamed from Source/WebCore/storage/StorageInfo.idl.
* Modules/quota/StorageInfoErrorCallback.h: Renamed from Source/WebCore/storage/StorageInfoErrorCallback.h.
* Modules/quota/StorageInfoErrorCallback.idl: Renamed from Source/WebCore/storage/StorageInfoErrorCallback.idl.
* Modules/quota/StorageInfoQuotaCallback.h: Renamed from Source/WebCore/storage/StorageInfoQuotaCallback.h.
* Modules/quota/StorageInfoQuotaCallback.idl: Renamed from Source/WebCore/storage/StorageInfoQuotaCallback.idl.
* Modules/quota/StorageInfoUsageCallback.h: Renamed from Source/WebCore/storage/StorageInfoUsageCallback.h.
* Modules/quota/StorageInfoUsageCallback.idl: Renamed from Source/WebCore/storage/StorageInfoUsageCallback.idl.
* Target.pri:
* WebCore.gyp/WebCore.gyp:
* WebCore.gypi:
* WebCore.pri:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.vcproj/WebCoreCommon.vsprops:
* WebCore.vcproj/copyForwardingHeaders.cmd:
* WebCore.xcodeproj/project.pbxproj:
* page/DOMWindow.cpp:
* page/DOMWindow.h:
* page/DOMWindow.idl:
2012-06-10 Hayato Ito <hayato@chromium.org>
An inheritance of '-webkit-user-modify' does not stop at shadow boundary.
https://bugs.webkit.org/show_bug.cgi?id=88514
Reviewed by Ryosuke Niwa.
In StyleResolver::styleForElement(), we reset '-webkit-user-modify'
CSS property after inheriting a parent style, but that is not
enough. We also have to reset '-webkit-user-modify' when we use a
cached result in applying matched properties.
Test: fast/dom/shadow/user-modify-inheritance.html
* css/StyleResolver.cpp:
(WebCore::StyleResolver::applyMatchedProperties):
(WebCore::StyleResolver::styleForElement):
* css/StyleResolver.h:
(StyleResolver):
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::inheritFrom):
* rendering/style/RenderStyle.h:
2012-06-10 Yoshifumi Inoue <yosin@chromium.org>
[Forms] Introduce Decimal behind the InputNumber type
https://bugs.webkit.org/show_bug.cgi?id=88383
Reviewed by Kent Tamura.
This patch introduces decimal arithmetic for steppable input types,
e.g. date, datetime, number, range, and so on, to avoid rounding error
caused by base 2 floating point representation, e.g. C/C++ double type.
Most of decimal arithmetic calculations are implemented in StepRange
class, replacing "double" with "Decimal", InputType::applyStep, and
InputType::stepFromRenderer.
Changes introduced by this patch are still intermediate state.
Following patch will replace InputNumber type to Decimal type for
completion of introducing decimal arithmetic.
Test: fast/forms/range/range-value-rounding.html
* html/BaseDateAndTimeInputType.cpp:
(WebCore::BaseDateAndTimeInputType::serialize): Changed for Decimal type.
(WebCore::BaseDateAndTimeInputType::serializeWithComponents): ditto.
* html/DateInputType.cpp:
(WebCore::DateInputType::createStepRange): Replaced NumberWithDecimalPlaces to InputNumber.
* html/DateTimeInputType.cpp:
(WebCore::DateTimeInputType::createStepRange): Replaced NumberWithDecimalPlaces to InputNumber.
* html/DateTimeLocalInputType.cpp:
(WebCore::DateTimeLocalInputType::createStepRange): Replaced NumberWithDecimalPlaces to InputNumber.
* html/InputType.cpp:
(WebCore::InputType::rangeUnderflow): Changed for Decimal type.
(WebCore::InputType::rangeOverflow): Changed for Decimal type.
(WebCore::InputType::minimum): Changed for Decimal type.
(WebCore::InputType::maximum): ditto.
(WebCore::InputType::isInRange): ditto.
(WebCore::InputType::isOutOfRange): ditto.
(WebCore::InputType::stepMismatch): ditto.
(WebCore::InputType::validationMessage): ditto.
(WebCore::InputType::parseToNumberOrNaN): ditto.
(WebCore::InputType::applyStep): ditto.
(WebCore::InputType::stepUpFromRenderer): ditto.
* html/InputType.h:
(InputType): Removed parseToNumberWIthDecimaplPlaces.
* html/MonthInputType.cpp:
(WebCore::MonthInputType::createStepRange): Changed for Decimal type.
* html/NumberInputType.cpp:
(WebCore::RealNumberRenderSize): Added for calculateRenderSize.
(WebCore::calculateRenderSize): Added. This function replacess lengthBeforeDecimalPoint.
(WebCore::NumberInputType::createStepRange): Changed for Decimal type.
(WebCore::NumberInputType::sizeShouldIncludeDecoration): Changed to use calculateRenderSize.
(WebCore::NumberInputType::parseToNumber): Changed for Decimal type.
(WebCore::NumberInputType::serialize): ditto.
* html/NumberInputType.h:
(NumberInputType): Removed parseToNumberWIthDecimaplPlaces.
* html/RangeInputType.cpp:
(WebCore::RangeInputType::createStepRange): Changed for Decimal type.
(WebCore::RangeInputType::handleKeydownEvent): ditto.
(WebCore::RangeInputType::parseToNumber): ditto.
(WebCore::RangeInputType::serialize): ditto.
* html/StepRange.cpp:
(WebCore::StepRange::StepRange): Removed decimal places and changed for Decimal type.
(WebCore::StepRange::acceptableError): Changed for Decimal type.
(WebCore::StepRange::alignValueForStep): ditto.
(WebCore::StepRange::clampValue): Changed for Decimal type and std::min/max.
(WebCore::StepRange::parseStep): Changed for Decimal type and removed NumberWithDecimalPlaces.
(WebCore::StepRange::stepMismatch): Changed for Decimal type.
(WebCore::convertDoubleToInputNumber): Changed to real implementation.
(WebCore::convertInputNumberToDouble): Changed to real implementation.
* html/StepRange.h:
(InputNumber): Replacement of NumberWithDecimalPlaces.
(WebCore::StepRange::InputNumber::InputNumber):
* html/TimeInputType.cpp:
(WebCore::TimeInputType::createStepRange): Changed for Decimal type.
* html/WeekInputType.cpp:
(WebCore::WeekInputType::createStepRange): Changed for Decimal type.
* html/parser/HTMLParserIdioms.cpp:
(WebCore::serializeForNumberType): Added Decimal version.
(WebCore::parseToDecimalForNumberType): Added.
* html/parser/HTMLParserIdioms.h: Updated comments for parseToDoubleForNumberType.
* html/shadow/SliderThumbElement.cpp:
(WebCore::sliderPosition): Changed for Decimal type.
2012-06-10 Jason Liu <jason.liu@torchmobile.com.cn>
[BlackBerry] Possible to clobber httponly cookie.
https://bugs.webkit.org/show_bug.cgi?id=86067
Reviewed by Rob Buis.
If a cookie is set by javaScript and there is already a same httpOnly cookie in cookieManager,
we should reject it. If it has a httpOnly property, we reject it, too.
Test: http/tests/cookies/js-get-and-set-http-only-cookie.html
* platform/blackberry/CookieJarBlackBerry.cpp:
(WebCore::setCookies):
* platform/blackberry/CookieManager.cpp:
(WebCore::CookieManager::setCookies):
(WebCore::CookieManager::checkAndTreatCookie):
(WebCore::CookieManager::addCookieToMap):
(WebCore::CookieManager::setPrivateMode):
* platform/blackberry/CookieManager.h:
* platform/blackberry/CookieMap.cpp:
(WebCore::CookieMap::addOrReplaceCookie):
(WebCore::CookieMap::removeCookie):
* platform/blackberry/CookieMap.h:
(CookieMap):
2012-06-10 Pablo Flouret <pablof@motorola.com>
Access control allow lists starting with a comma are parsed incorrectly (CORS)
https://bugs.webkit.org/show_bug.cgi?id=88461
Reviewed by Adam Barth.
The parsing algorithm would stop parsing at the first comma (being an
empty entry, of sorts) and bail out, effectively ignoring the whole
header.
Test: http/tests/xmlhttprequest/access-control-allow-lists-starting-with-comma.html
* loader/CrossOriginPreflightResultCache.cpp:
(WebCore::parseAccessControlAllowList):
2012-06-10 Darin Adler <darin@apple.com>
Some additional tweaks to keep TreeShared code simple/clean
https://bugs.webkit.org/show_bug.cgi?id=88738
Reviewed by Adam Barth.
* platform/TreeShared.h: Make constructor and destructor both protected.
It's slightly dangerous to have a public non-virtual destructor, and there
is no reason it needs to be public.
(WebCore::TreeShared::~TreeShared): Make debug-only destructor non-virtual;
we were getting no value from it being virtual. Now that it's non-virtual,
there is no need to use an #if on it since it's an empty inline destructor
in non-debug builds.
2012-06-06 Jer Noble <jer.noble@apple.com>
REGRESSION: Setting invalid media "src" does not cause "error" event
https://bugs.webkit.org/show_bug.cgi?id=88423
Reviewed by Eric Carlson.
Test: http/tests/media/video-src-invalid-error.html
Two problems here. When the loadTimer is scheduled, the m_pendingLoadFlags
variable is set to specify what type of load is scheduled. It is cleared
when the loadTimer fires. But, when prepareForLoad() stops the loadTimer,
it does not clear the m_pendingLoadFlags variable, so the next time
scheduleLoad() is called (when the src is changed to an invalid URL)
prepareForLoad() is not called again.
Second problem: Due to a bug in QTKit (<rdar://problem/11606415>), the
QTMovieLoadStateChangedNotification is never fired for an invalid http
URL if QTMovieOpenAsyncRequiredAttribute:YES is not passed when creating
the QTMovie.
Add a new utility method which both stops the m_loadTimer and clears the
m_pendingLoadFlags, and use it in all the places where m_loadTimer was
stopped explicitly:
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::prepareForLoad):
(WebCore::HTMLMediaElement::stopLoadTimer): Added
(WebCore::HTMLMediaElement::userCancelledLoad):
* html/HTMLMediaElement.h:
And pass QTMovieOpenAsyncRequiredAttribute:YES when creating the QTMovie:
* platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
(WebCore::MediaPlayerPrivateQTKit::commonMovieAttributes):
2012-06-06 Jer Noble <jer.noble@apple.com>
Add logging functions to MediaPlayerPrivateQTKit.
https://bugs.webkit.org/show_bug.cgi?id=88425
Reviewed by Eric Carlson.
No new tests; adds logging functions to aide debugging.
MediaPlayerPrivateQTKit should emit the same kind of logging messages as
other MediaPlayerPrivate engines to aide in debugging.
Distinguish between loadedRangesChanged() and loadStateChanged() for logging
purposes:
* platform/graphics/mac/MediaPlayerPrivateQTKit.h:
* platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
(WebCore::MediaPlayerPrivateQTKit::loadedRangesChanged):
(-[WebCoreMovieObserver loadedRangesChanged:]):
Add LOG messages for each of the following functions:
* platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
(WebCore::MediaPlayerPrivateQTKit::~MediaPlayerPrivateQTKit):
(WebCore::MediaPlayerPrivateQTKit::createQTMovie):
(WebCore::MediaPlayerPrivateQTKit::createQTMovieView):
(WebCore::MediaPlayerPrivateQTKit::detachQTMovieView):
(WebCore::MediaPlayerPrivateQTKit::createQTVideoRenderer):
(WebCore::MediaPlayerPrivateQTKit::destroyQTVideoRenderer):
(WebCore::MediaPlayerPrivateQTKit::createQTMovieLayer):
(WebCore::MediaPlayerPrivateQTKit::destroyQTMovieLayer):
(WebCore::MediaPlayerPrivateQTKit::setUpVideoRendering):
(WebCore::MediaPlayerPrivateQTKit::tearDownVideoRendering):
(WebCore::MediaPlayerPrivateQTKit::load):
(WebCore::MediaPlayerPrivateQTKit::prepareToPlay):
(WebCore::MediaPlayerPrivateQTKit::play):
(WebCore::MediaPlayerPrivateQTKit::pause):
(WebCore::MediaPlayerPrivateQTKit::seek):
(WebCore::MediaPlayerPrivateQTKit::cancelSeek):
(WebCore::MediaPlayerPrivateQTKit::setVolume):
(WebCore::MediaPlayerPrivateQTKit::setRate):
(WebCore::MediaPlayerPrivateQTKit::setPreservesPitch):
(WebCore::MediaPlayerPrivateQTKit::cancelLoad):
(WebCore::MediaPlayerPrivateQTKit::prepareForRendering):
(WebCore::MediaPlayerPrivateQTKit::updateStates):
(WebCore::MediaPlayerPrivateQTKit::loadStateChanged):
(WebCore::MediaPlayerPrivateQTKit::rateChanged):
(WebCore::MediaPlayerPrivateQTKit::sizeChanged):
(WebCore::MediaPlayerPrivateQTKit::timeChanged):
(WebCore::MediaPlayerPrivateQTKit::didEnd):
(WebCore::MediaPlayerPrivateQTKit::clearMediaCache):
(WebCore::MediaPlayerPrivateQTKit::clearMediaCacheForSite):
(WebCore::MediaPlayerPrivateQTKit::disableUnsupportedTracks):
2012-06-07 Jer Noble <jer.noble@apple.com>
WebCore should use a single definition of an invalid media time.
https://bugs.webkit.org/show_bug.cgi?id=88572
Reviewed by Eric Carlson.
No new tests; no change in functionality, so covered by existing tests.
Add a new, universal definition for an invalid media time:
* platform/graphics/MediaPlayer.h:
(WebCore::MediaPlayer::invalidTime):
Use this new definition instead of the hodge podge of (identical) local
definitions for an invalid media time:
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::HTMLMediaElement):
(WebCore::HTMLMediaElement::invalidateCachedTime):
(WebCore::HTMLMediaElement::currentTime):
(WebCore::HTMLMediaElement::initialTime):
(WebCore::HTMLMediaElement::playbackProgressTimerFired):
(WebCore::HTMLMediaElement::prepareMediaFragmentURI):
(WebCore::HTMLMediaElement::applyMediaFragmentURI):
* html/MediaFragmentURIParser.cpp:
(WebCore::MediaFragmentURIParser::invalidTimeValue):
(WebCore::MediaFragmentURIParser::MediaFragmentURIParser):
(WebCore::MediaFragmentURIParser::startTime):
(WebCore::MediaFragmentURIParser::endTime):
(WebCore::MediaFragmentURIParser::parseTimeFragment):
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
(WebCore::MediaPlayerPrivateAVFoundation::MediaPlayerPrivateAVFoundation):
(WebCore::MediaPlayerPrivateAVFoundation::duration):
(WebCore::MediaPlayerPrivateAVFoundation::seeking):
(WebCore::MediaPlayerPrivateAVFoundation::seekCompleted):
(WebCore::MediaPlayerPrivateAVFoundation::invalidateCachedDuration):
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:
(MediaPlayerPrivateAVFoundation):
* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::platformDuration):
* platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
(WebCore::MediaPlayerPrivateQTKit::duration):
(WebCore::MediaPlayerPrivateQTKit::updateStates):
(WebCore::MediaPlayerPrivateQTKit::timeChanged):
2012-06-10 Darin Adler <darin@apple.com>
Remove unneeded callRemovedLastRef function from TreeShared refactoring
https://bugs.webkit.org/show_bug.cgi?id=88653
Reviewed by Sam Weinig.
* WebCore.exp.in: Updated to remove callRemovedLastRef and add removedLastRef.
* dom/ContainerNode.cpp: Removed callRemovedLastRef, which is no longer
needed and also had a bad cast in it.
* dom/Document.h: Made removedLastRef private and non-virtual and made Node
a friend so it can call the private function.
* dom/Node.cpp:
(WebCore::Node::removedLastRef): Moved here. This should not be an inline
function. Added comments explaining why it should not be inlined. Also added
explicit code to call to Document::removedLastRef, and added a comment why
that's the right thing to do here.
* dom/Node.h: Inherit from TreeShared<Node, ContainerNode>. Made the
removedLastRef function non-virtual and removed it from the header file.
Updated TreeShared using statements. Made the TreeShared template class a
friend so it can call the private removedLastRef.
* html/HTMLFormControlElement.h: Updated TreeShared using statements to specify
Node instead, which accomplishes the same thing.
* html/HTMLObjectElement.h: Ditto.
* platform/TreeShared.h: Made TreeShared take two template arguments for the
node type and the parent node type. Removed the callRemovedLastRef functions and
the extra forward declarations it required. Removed the private unused
removedLastRef function; not sure why it was left in here at all.
(WebCore::TreeShared::deref): Call removedLastRef instead of callRemovedLastRef.
(WebCore::TreeShared::setParent): Use ParentNodeType.
(WebCore::TreeShared::parent): Ditto.
(WebCore::adopted): Updated to have both template arguments.
* svg/SVGElementInstance.cpp:
(WebCore::SVGElementInstance::removedLastRef): Replaced callRemovedLastRef with
this. Moved the body out of the header and added a comment explaining why the
function should not be inlined.
* svg/SVGElementInstance.h: Made removedLastRef non-virtual, non-inline, and
private. Made TreeShared a friend so it can call the private removedLastRef.
Updated TreeShared using statements.
2012-06-10 Darin Adler <darin@apple.com>
* WebCore.exp.in: Re-sorted and alphabetized the file.
2012-06-10 Balazs Kelemen <kbalazs@webkit.org>
[Qt][Win] Fix building Source/WebCore/plugins/win/PluginViewWin.cpp
https://bugs.webkit.org/show_bug.cgi?id=88308
Reviewed by Simon Hausmann.
No new tests, just a build fix.
Use ownerWindow() from page client. It has been
introduced for x11 plugins to solve exactly the
same problem.
* plugins/win/PluginViewWin.cpp:
(windowHandleForPageClient):
2012-06-10 Balazs Kelemen <kbalazs@webkit.org>
[Qt][Win] Fix building ImageQt.cpp
https://bugs.webkit.org/show_bug.cgi?id=88306
Reviewed by Simon Hausmann.
No new tests, this is just a build fix.
* platform/graphics/qt/ImageQt.cpp:
(WebCore):
(WebCore::BitmapImage::create):
Adapt to the "API" changes in Qt.
2012-06-09 Gregg Tavares <gman@google.com>
Make WebGL mark draws for compositing even if the draw count is zero
https://bugs.webkit.org/show_bug.cgi?id=88718
Reviewed by Kenneth Russell.
No new tests as no new functionality.
* html/canvas/WebGLRenderingContext.cpp:
(WebCore):
(WebCore::WebGLRenderingContext::drawArrays):
(WebCore::WebGLRenderingContext::drawElements):
2012-06-09 Dominic Cooney <dominicc@chromium.org>
[Chromium] Remove JavaScriptCore dependencies from gyp
https://bugs.webkit.org/show_bug.cgi?id=88510
Reviewed by Adam Barth.
Chromium doesn't support JSC any more and there doesn't seem to be
a strong interest in using GYP as the common build system in other
ports.
No new functionality => No new tests.
* gyp/WebCore.gyp: Removed.
* gyp/copy-forwarding-and-icu-headers.sh: Removed.
* gyp/copy-inspector-resources.sh: Removed.
* gyp/generate-derived-sources.sh: Removed.
* gyp/generate-webcore-export-file-generator.sh: Removed.
* gyp/gtk.gyp: Removed.
* gyp/run-if-exists.sh: Removed.
* gyp/streamline-inspector-source.sh: Removed.
* gyp/update-info-plist.sh: Removed.
2012-06-07 Darin Adler <darin@apple.com>
Refactor and improve code style in TextResourceDecoder to prepare for buffering improvement
https://bugs.webkit.org/show_bug.cgi?id=88566
Reviewed by Alexey Proskuryakov.
* loader/TextResourceDecoder.cpp:
(WebCore::bytesEqual): Added helper function to make comparing against sequences of bytes
easier to read at the call site.
(WebCore::KanjiCode::judge): Changed to use bytesEqual in a few places, merged multiple if
statements, got rid of else after goto.
(WebCore::TextResourceDecoder::checkForCSSCharset): Changed to use early return instead of
nesting the whole function. Changed to use bytesEqual.
(WebCore::skipComment): Changed to use early return for more cases. Changed to use bytesEqual.
(WebCore::TextResourceDecoder::checkForHeadCharset): Changed to use bytesEqual.
(WebCore::TextResourceDecoder::decode): Changed to use emptyString() instead of "", since
the former is more efficient.
2012-06-09 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118618 and r119353.
http://trac.webkit.org/changeset/118618
http://trac.webkit.org/changeset/119353
https://bugs.webkit.org/show_bug.cgi?id=88720
Caused at least 30 different crashes on ClusterFuzz (Requested
by inferno-sec on #webkit).
* loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::checkForHTTPStatusCodeError):
* loader/cache/CachedCSSStyleSheet.cpp:
(WebCore::CachedCSSStyleSheet::allClientsRemoved):
* loader/cache/CachedFont.cpp:
(WebCore::CachedFont::allClientsRemoved):
* loader/cache/CachedFont.h:
(WebCore::CachedFontClient::resourceClientType):
* loader/cache/CachedImage.cpp:
(WebCore::CachedImage::removeClientForRenderer):
(WebCore):
(WebCore::CachedImage::allClientsRemoved):
(WebCore::CachedImage::lookupOrCreateImageForRenderer):
* loader/cache/CachedImage.h:
(CachedImage):
(WebCore::CachedImageClient::resourceClientType):
* loader/cache/CachedRawResource.cpp:
(WebCore::CachedRawResource::allClientsRemoved):
(WebCore):
* loader/cache/CachedRawResource.h:
(CachedRawResource):
(WebCore::CachedRawResourceClient::resourceClientType):
* loader/cache/CachedResource.cpp:
(WebCore::CachedResource::removeClient):
* loader/cache/CachedResource.h:
(WebCore::CachedResource::allClientsRemoved):
* loader/cache/CachedResourceClient.h:
(WebCore::CachedResourceClient::resourceClientType):
* loader/cache/CachedSVGDocument.h:
(WebCore::CachedSVGDocumentClient::resourceClientType):
* loader/cache/CachedScript.cpp:
(WebCore::CachedScript::allClientsRemoved):
* loader/cache/CachedStyleSheetClient.h:
(WebCore::CachedStyleSheetClient::resourceClientType):
* rendering/style/StyleCachedImage.cpp:
(WebCore::StyleCachedImage::removeClient):
* rendering/style/StyleCachedImageSet.cpp:
(WebCore::StyleCachedImageSet::removeClient):
* svg/graphics/SVGImageCache.cpp:
(WebCore::SVGImageCache::~SVGImageCache):
(WebCore::SVGImageCache::removeRendererFromCache):
(WebCore::SVGImageCache::setRequestedSizeAndScales):
(WebCore::SVGImageCache::requestedSizeAndScales):
(WebCore::SVGImageCache::lookupOrCreateBitmapImageForRenderer):
* svg/graphics/SVGImageCache.h:
(WebCore):
(SVGImageCache):
2012-06-09 Florin Malita <fmalita@chromium.org>
Fixed-position foreignObject descendants should be relative to the foreignObject viewport
https://bugs.webkit.org/show_bug.cgi?id=88547
Reviewed by Abhishek Arya.
Tests: svg/foreignObject/fO-fixed-position-crash.html
svg/foreignObject/fixed-position-expected.svg
svg/foreignObject/fixed-position.svg
Fixed position elements are currently registered with the top level
RenderView even when embedded within an SVG foreignOject. This patch
changes containingBlock() & container() to return the containing
foreignObject renderer instead.
The new foreignObject fixed position behavior matches that of current
FireFox and Opera versions and is consistent with the spec:
http://www.w3.org/TR/CSS2/visuren.html#fixed-positioning
http://www.w3.org/TR/SVG/coords.html#EstablishingANewViewport
* rendering/RenderObject.cpp:
(WebCore::RenderObject::containingBlock):
(WebCore::RenderObject::container):
2012-06-09 Adam Barth <abarth@webkit.org>
Can't use eval in iframes sanbdoxed via CSP header
https://bugs.webkit.org/show_bug.cgi?id=88450
Reviewed by Mihai Parparita.
The initial empty document in a frame inherits the security context of
its parent (including the CSP policy). When we load the real document,
in some cases we'll do a "secure transition" to the new document. That
means that we leave the global object in place in case the parent
document has created any properties that it expects will be visible to
the new document.
If the parent document has a CSP policy that blocks eval, the "no eval"
bit will be set on the global object of the initial document. When we
perform a "secure transition" to the new document, we'll keep the bit,
which is wrong. In this patch, we reset the bit by always enabling
eval when clearing the context, regardless of whether we're performing
a "secure transition".
Test: http/tests/security/contentSecurityPolicy/iframe-inside-csp.html
* bindings/js/ScriptController.cpp:
(WebCore::ScriptController::enableEval):
(WebCore):
* bindings/js/ScriptController.h:
(ScriptController):
* bindings/v8/ScriptController.cpp:
(WebCore::ScriptController::enableEval):
(WebCore):
(WebCore::ScriptController::disableEval):
* bindings/v8/ScriptController.h:
(ScriptController):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::clear):
2012-06-09 Pablo Flouret <pablof@motorola.com>
The value in Access-Control-Allow-Origin is not being matched correctly for CORS-enabled requests
https://bugs.webkit.org/show_bug.cgi?id=88139
Reviewed by Adam Barth.
Compare a request's origin with the value given in any
Access-Control-Allow-Origin headers in an exact, case-sensitive manner,
instead of using SecurityOrigin::isSameSchemeHostPort(). Per step 3 of
the resource sharing check algorithm in
http://dvcs.w3.org/hg/cors/raw-file/tip/Overview.html#resource-sharing-check
Test: http/tests/xmlhttprequest/origin-exact-matching.html
* loader/CrossOriginAccessControl.cpp:
(WebCore::passesAccessControlCheck):
2012-06-09 Huang Dongsung <luxtella@company100.net>
[Qt][Texmap] All layers with backingStore are opaque when using TextureMapperGL.
https://bugs.webkit.org/show_bug.cgi?id=88703
The bug originated from StillImage that is not override
Image::currentFrameHasAlpha(), so it always returns false.
StillImage is used by ImageBuffer, and if Texmap draws contents on ImageBuffer,
TextureMapperBackingStore thinks contents does not have alpha.
Covered by existing compositing tests.
Reviewed by Noam Rosenthal.
* platform/graphics/qt/StillImageQt.cpp:
(WebCore::StillImage::currentFrameHasAlpha):
(WebCore):
* platform/graphics/qt/StillImageQt.h:
(StillImage):
2012-06-09 Victor Carbune <victor@rosedu.org>
Basic support for timestamps within a TextTrackCue
https://bugs.webkit.org/show_bug.cgi?id=88187
Implemented support for timestamps within a TextTrackCue.
This enables rendering functionality for Karaoke and Paint-on captions.
Reviewed by Eric Carlson.
Test: media/track/track-cue-rendering-inner-timestamps.html
* css/mediaControls.css: Remove the background shadow pseudo-id and
added two others, which can be further used for styling past / future
contents of the cue.
(video::-webkit-media-text-track-past-nodes): Used to style elements
within the cue that are before the current movie time.
(video::-webkit-media-text-track-future-nodes): Used to style elements
within the cue that are after the current movie time.
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::updateActiveTextTrackCues): Regardless of whether
the active set changed or not, the current cues need to be informed of the current
movie time.
* html/track/TextTrackCue.cpp:
(WebCore::TextTrackCue::TextTrackCue): Initiliazed past and future containers.
(WebCore::TextTrackCue::getCueAsHTML): Checked whether the current cue text has
inner timestamps or not, in order to render faster if no timestamps are there.
(WebCore::TextTrackCue::updateDisplayTree): Added method that keeps the past and future
containers up to date, given a current movie time.
(WebCore):
(WebCore::TextTrackCue::getDisplayTree): Replaced the cue background container with the
past cues container. The *internal* display tree of the cue now has two containers,
representing past and future elements.
* html/track/TextTrackCue.h: Added several variables to support the new functionality.
(TextTrackCue):
* html/track/WebVTTParser.cpp:
(WebCore::WebVTTParser::constructTreeFromToken): Updated representation for this patch.
* html/track/WebVTTParser.h:
(WebVTTParser): Made public the collectTimeStamp method, as this is required for
parsing timestamps within a cue as well.
2012-06-09 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: Open links in Sources panel by default and fallback to Resources and Network panel otherwise.
https://bugs.webkit.org/show_bug.cgi?id=88710
Reviewed by Pavel Feldman.
* inspector/front-end/inspector.js:
(WebInspector.documentClick.followLink):
(WebInspector.documentClick):
(WebInspector._showAnchorLocation):
2012-06-09 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: Only expand root domain in Sources/Scripts panel
https://bugs.webkit.org/show_bug.cgi?id=88135
Reviewed by Pavel Feldman.
* inspector/front-end/NavigatorView.js:
(WebInspector.NavigatorTreeOutline._treeElementsCompare.compare.typeWeight):
(WebInspector.NavigatorTreeOutline._treeElementsCompare):
(WebInspector.NavigatorFolderTreeElement.prototype.onattach):
2012-06-09 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: Pretty print should respect text editor indent setting.
https://bugs.webkit.org/show_bug.cgi?id=88713
Reviewed by Pavel Feldman.
* inspector/front-end/ScriptFormatter.js:
2012-06-09 Dan Bernstein <mitz@apple.com>
Fixed the build after r119895.
* WebCore.xcodeproj/project.pbxproj:
2012-06-09 Andrey Kosyakov <caseq@chromium.org>
Web Inspector: [refactoring] rename TimelineVerticalOverview into TimelineFrameOverview
https://bugs.webkit.org/show_bug.cgi?id=88708
Reviewed by Vsevolod Vlasov.
- rename TimelineVerticalOverview to TimelineFrameOverview;
- rename associated fields, parameters, CSS classes etc;
- rename overview mode designators to match the UI;
- drop unused CSS rules for old-style mode selector;
* inspector/front-end/TimelineOverviewPane.js:
(WebInspector.TimelineOverviewPane):
(WebInspector.TimelineOverviewPane.prototype._showEvents):
(WebInspector.TimelineOverviewPane.prototype._showFrames):
(WebInspector.TimelineOverviewPane.prototype._showMemoryGraph):
(WebInspector.TimelineOverviewPane.prototype._setFrameMode):
(WebInspector.TimelineOverviewPane.prototype._onCategoryVisibilityChanged):
(WebInspector.TimelineOverviewPane.prototype._update):
(WebInspector.TimelineOverviewPane.prototype.addFrame):
(WebInspector.TimelineOverviewPane.prototype.zoomToFrame):
(WebInspector.TimelineOverviewPane.prototype._reset):
(WebInspector.TimelineOverviewPane.prototype._onWindowChanged):
(WebInspector.TimelineFrameOverview):
* inspector/front-end/TimelinePanel.js:
(WebInspector.TimelinePanel):
(WebInspector.TimelinePanel.prototype._shouldShowFrames):
(WebInspector.TimelinePanel.prototype.revealRecordAt):
* inspector/front-end/timelinePanel.css:
(.timeline-frame-overview #timeline-overview-grid):
(.timeline-frame-overview .timeline-overview-window):
(.timeline-frame-overview .timeline-overview-dividers-background):
(.timeline-frame-overview #timeline-overview-memory):
(.timeline-frame-overview-status-bar-item.toggled-on .glyph):
(.timeline-frame-overview-bars):
(.timeline.timeline-frame-overview .resources-divider):
(.sidebar-tree-item .timeline-frame-overview-status-bar-item):
2012-06-08 Vsevolod Vlasov <vsevik@chromium.org>
IndexedDB: Inspector should handle null, string, and array keyPaths
https://bugs.webkit.org/show_bug.cgi?id=84303
Reviewed by Pavel Feldman.
Supported different key path types and updated tests to cover each case.
Key path strings are now highlighted in the DataGrid column headers.
* inspector/Inspector.json:
* inspector/InspectorIndexedDBAgent.cpp:
(WebCore):
* inspector/front-end/IndexedDBModel.js:
(WebInspector.IndexedDBModel.idbKeyPathFromKeyPath):
(WebInspector.IndexedDBModel.keyPathStringFromIDBKeyPath):
(WebInspector.IndexedDBModel.prototype._loadDatabase.callback):
(WebInspector.IndexedDBModel.prototype._loadDatabase):
(WebInspector.IndexedDBModel.ObjectStore.prototype.get keyPathString):
(WebInspector.IndexedDBModel.Index.prototype.get keyPathString):
* inspector/front-end/IndexedDBViews.js:
(WebInspector.IDBDataView.prototype._createDataGrid):
(WebInspector.IDBDataView.prototype._keyColumnHeaderFragment):
(WebInspector.IDBDataView.prototype._keyPathStringFragment):
(WebInspector.IDBDataGridNode.prototype.createCell):
* inspector/front-end/ResourcesPanel.js:
(WebInspector.IDBObjectStoreTreeElement.prototype._updateTooltip):
(WebInspector.IDBIndexTreeElement.prototype._updateTooltip):
2012-06-08 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: Increase size limits for resource content cached in InspectorResourceAgent.
https://bugs.webkit.org/show_bug.cgi?id=88674
Reviewed by Pavel Feldman.
* inspector/NetworkResourcesData.cpp:
2012-05-25 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: Limit DebuggerAgent.Location use to DebuggerModel only, introduce WebInspector.RawLocation to be used elsewhere.
https://bugs.webkit.org/show_bug.cgi?id=87270
Reviewed by Pavel Feldman.
Introduced WebInspector.RawLocation interface, a generic raw location independent from concrete domain (Styles, Debugger).
Made WebInspector.DebuggerModel.Location implement it and switched all debugger related methods to it.
* inspector/front-end/BreakpointManager.js:
(WebInspector.BreakpointManager.prototype._breakpointResolved):
* inspector/front-end/CompilerScriptMapping.js:
(WebInspector.CompilerScriptMapping.prototype.rawLocationToUILocation):
* inspector/front-end/DebuggerModel.js:
(WebInspector.DebuggerModel.Location):
(WebInspector.DebuggerModel.prototype.continueToLocation):
(WebInspector.DebuggerModel.prototype.setBreakpointByScriptLocation):
(WebInspector.DebuggerModel.prototype.setBreakpoint.didSetBreakpoint):
(WebInspector.DebuggerModel.prototype.setBreakpoint):
(WebInspector.DebuggerModel.prototype.setBreakpointBySourceId):
(WebInspector.DebuggerModel.prototype._breakpointResolved):
(WebInspector.DebuggerModel.prototype.createRawLocation):
(WebInspector.DebuggerModel.prototype.createRawLocationByURL):
(WebInspector.DebuggerModel.prototype.setSelectedCallFrame):
(WebInspector.DebuggerModel.prototype.createLiveLocation):
(WebInspector.DebuggerModel.CallFrame.prototype.get location):
(WebInspector.DebuggerModel.CallFrame.prototype.createLiveLocation):
* inspector/front-end/DebuggerResourceBinding.js:
(WebInspector.DebuggerResourceBinding.setScriptSource.didEditScriptSource):
(WebInspector.DebuggerResourceBinding.setScriptSource):
(WebInspector.DebuggerResourceBinding.prototype.setContent.get if):
(WebInspector.DebuggerResourceBinding.prototype.setContent):
(WebInspector.DebuggerResourceBinding.prototype._setContentWithInitialContent):
* inspector/front-end/JavaScriptSource.js:
(WebInspector.JavaScriptSource.prototype.uiLocationToRawLocation):
* inspector/front-end/Linkifier.js:
* inspector/front-end/ObjectPopoverHelper.js:
(WebInspector.ObjectPopoverHelper.prototype._showObjectPopover.showObjectPopover.):
(WebInspector.ObjectPopoverHelper.prototype._showObjectPopover):
* inspector/front-end/PresentationConsoleMessageHelper.js:
* inspector/front-end/RawSourceCode.js:
* inspector/front-end/ResourceScriptMapping.js:
(WebInspector.ResourceScriptMapping.prototype.rawLocationToUILocation):
* inspector/front-end/Script.js:
(WebInspector.Script.prototype.rawLocationToUILocation):
* inspector/front-end/ScriptSnippetModel.js:
(WebInspector.SnippetScriptMapping.prototype.rawLocationToUILocation):
* inspector/front-end/SourceMapping.js:
* inspector/front-end/UISourceCode.js:
(WebInspector.RawLocation):
2012-06-09 Eugene Klyuchnikov <eustas.bug@gmail.com>
Web Inspector: Add message loop instrumentation to public API and timeline agent
https://bugs.webkit.org/show_bug.cgi?id=88639
Reviewed by Vsevolod Vlasov.
Message loop instrumentation will show when the render thread is busy.
That way developer can discover if a render thread business causes low fps, or not.
* inspector/InspectorClient.h:
(WebCore::InspectorClient::startMessageLoopMonitoring): Starts observer.
(WebCore::InspectorClient::stopMessageLoopMonitoring): Stops observer.
* inspector/InspectorController.cpp:
(WebCore::InspectorController::InspectorController):
Changed timeline agent constructor parameters.
* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::willProcessTaskImpl):
Forwards observer notification.
(WebCore):
(WebCore::InspectorInstrumentation::didProcessTaskImpl):
Ditto.
* inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
(WebCore::InspectorInstrumentation::willProcessTask):
Ditto.
(WebCore):
(WebCore::InspectorInstrumentation::didProcessTask):
Ditto.
* inspector/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::start):
Starts observer.
(WebCore::InspectorTimelineAgent::stop):
Stops observer.
(WebCore::InspectorTimelineAgent::supportsFrameInstrumentation):
Changed result origin.
(WebCore::InspectorTimelineAgent::willProcessTask):
Noop implementqation.
(WebCore):
(WebCore::InspectorTimelineAgent::didProcessTask):
Noop implementqation.
(WebCore::InspectorTimelineAgent::InspectorTimelineAgent):
Changed constructor parameters.
* inspector/InspectorTimelineAgent.h:
(WebCore::InspectorTimelineAgent::create):
Ditto.
(InspectorTimelineAgent):
* inspector/WorkerInspectorController.cpp:
(WebCore::WorkerInspectorController::WorkerInspectorController):
Changed timeline agent constructor parameters.
2012-06-08 Andrey Kosyakov <caseq@chromium.org>
Web Inspector: optimize timeline's frame mode overview appearance for 30fps
https://bugs.webkit.org/show_bug.cgi?id=88556
Reviewed by Vsevolod Vlasov.
- adjust frame overview scale to 30fps == full height by default;
- provide a fall-back auto-scale mode when median frame is longer than 1/30s;
- add horizontal lines for 30 and 60fps;
- drive-by fix: always use same order of categories in frame bars.
* inspector/front-end/TimelineOverviewPane.js:
(WebInspector.TimelineVerticalOverview.prototype.update):
(WebInspector.TimelineVerticalOverview.prototype._aggregateFrames):
(WebInspector.TimelineVerticalOverview.prototype._renderBars):
(WebInspector.TimelineVerticalOverview.prototype._drawFPSMarks):
(WebInspector.TimelineVerticalOverview.prototype._renderBar):
2012-06-09 Sukolsak Sakshuwong <sukolsak@google.com>
Add UNDO_MANAGER flag
https://bugs.webkit.org/show_bug.cgi?id=87908
Reviewed by Tony Chang.
Add undoManager property to document behind the UNDO_MANAGER flag and
add placeholders for some methods of undoManager.
Test: editing/undomanager/document-has-undomanager.html
* CMakeLists.txt:
* Configurations/FeatureDefines.xcconfig:
* DerivedSources.cpp:
* DerivedSources.make:
* DerivedSources.pri:
* GNUmakefile.am:
* GNUmakefile.list.am:
* WebCore.gypi:
* WebCore.xcodeproj/project.pbxproj:
* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore):
(WebCore::Document::undoManager):
* dom/Document.h:
(WebCore):
(Document):
* dom/Document.idl:
* editing/UndoManager.cpp: Added.
(WebCore):
(WebCore::UndoManager::create):
(WebCore::UndoManager::UndoManager):
(WebCore::UndoManager::undo):
(WebCore::UndoManager::redo):
(WebCore::UndoManager::clearUndo):
(WebCore::UndoManager::clearRedo):
* editing/UndoManager.h: Added.
(WebCore):
(UndoManager):
* editing/UndoManager.idl: Added.
2012-06-08 Levi Weintraub <leviw@chromium.org>
RenderLayer::paintChildLayerIntoColumns doesn't properly pixel snap
https://bugs.webkit.org/show_bug.cgi?id=88554
Reviewed by Darin Adler.
Column handling code in RenderLayer and RenderBlock was improperly clipping and translating
using floats derived from FractionalLayoutUnits without pixel snapping. RenderLayer also
improperly stored render tree offsets in ints instead of LayoutUnits. This corrects this
issue.
Test: fast/sub-pixel/column-clipping.html
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::paintColumnContents):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::paintChildLayerIntoColumns):
2012-06-08 Rakesh KN <rakesh.kn@motorola.com>
REGRESSION(r116487?): HTMLFormElement::elements['name'] is empty if the form is detached from the document tree
https://bugs.webkit.org/show_bug.cgi?id=88632
Reviewed by Ryosuke Niwa.
Update root node of RadioNodeList when the form element is detached from dom tree.
Test: fast/forms/radionodelist-whose-form-element-detached-from-domtree.html
* dom/Node.cpp:
(WebCore::Node::resetCachedRadioNodeListRootNode):
New function to update the root node of RadioNodeLists to form element from document.
* dom/Node.h: Ditto.
* html/HTMLFormElement.cpp:
(WebCore::HTMLFormElement::removedFrom):
On detach, we update the root node of all RadioNodeLists of this form element.
* html/RadioNodeList.cpp:
(WebCore::RadioNodeList::setRootElement):
Setter for updating root node.
* html/RadioNodeList.h:
(RadioNodeList): Ditto.
2012-06-08 David Reveman <reveman@chromium.org>
[Chromium] Compositor doesn't support translucent root layers.
https://bugs.webkit.org/show_bug.cgi?id=87821
Reviewed by James Robinson.
Forward the isTransparent WebView setting to WebLayerTreeView,
CCLayerTreeHost and CCLayerTreeHostImpl as hasTransparentBackground.
Use hasTransparentBackground setting to determine how to clear the
root render pass and draw the background. Set opaque flag correctly
on the NonCompositedContentHost's graphics layer to make sure
sub-pixel rendering is not used with a transparent WebView.
Unit tests: CCLayerTreeHostImplTest.hasTransparentBackground
LayerRendererChromiumTest2.opaqueRenderPass
LayerRendererChromiumTest2.transparentRenderPass
* platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.cpp:
(WebCore::BitmapCanvasLayerTextureUpdater::prepareToUpdate):
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::clearRenderPass):
(WebCore::LayerRendererChromium::drawRenderPass):
(WebCore::LayerRendererChromium::drawBackgroundFilters):
* platform/graphics/chromium/LayerRendererChromium.h:
(LayerRendererChromium):
* platform/graphics/chromium/SkPictureCanvasLayerTextureUpdater.cpp:
(WebCore::SkPictureCanvasLayerTextureUpdater::prepareToUpdate):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::CCLayerTreeHost):
(WebCore::CCLayerTreeHost::finishCommitOnImplThread):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCLayerTreeHost::setHasTransparentBackground):
(CCLayerTreeHost):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(WebCore::CCLayerTreeHostImpl::hasTransparentBackground):
(WebCore::CCLayerTreeHostImpl::setHasTransparentBackground):
(CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCRenderPass.cpp:
(WebCore::CCRenderPass::CCRenderPass):
* platform/graphics/chromium/cc/CCRenderPass.h:
(WebCore::CCRenderPass::hasTransparentBackground):
(WebCore::CCRenderPass::setHasTransparentBackground):
(CCRenderPass):
2012-06-08 Ian Vollick <vollick@chromium.org>
[chromium] Single thread proxy's animation timer should short circuit if the layer renderer has not been initialized
https://bugs.webkit.org/show_bug.cgi?id=88668
Reviewed by James Robinson.
Unit test: CCSingleThreadProxyTestDidAddAnimationBeforeInitializingLayerRenderer.runSingleThreaded
* platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
* platform/graphics/chromium/cc/CCSingleThreadProxy.h:
2012-06-08 Shezan Baig <shezbaig.wk@gmail.com>
Caret is not rendered in empty inline contenteditable elements
https://bugs.webkit.org/show_bug.cgi?id=85793
Reviewed by Ryosuke Niwa.
Override localCaretRect in RenderInline. The implementation was almost
identical to localCaretRect in RenderBlock for empty block elements, so
I refactored RenderBlock::localCaretRect and moved the logic to a new
method 'localCaretRectForEmptyElement' in RenderBoxModelObject. The
implementation of 'localCaretRect' in RenderBlock and RenderInline both
use this helper method in RenderBoxModelObject.
Tests: editing/selection/caret-in-empty-inline-1.html
editing/selection/caret-in-empty-inline-2.html
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::localCaretRect):
Modified to use RenderBoxModelObject::localCaretRectForEmptyElement.
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::localCaretRectForEmptyElement):
(WebCore):
* rendering/RenderBoxModelObject.h:
(RenderBoxModelObject):
Add localCaretRectForEmptyElement helper method.
* rendering/RenderInline.cpp:
(WebCore::RenderInline::localCaretRect):
(WebCore):
* rendering/RenderInline.h:
(RenderInline):
Override localCaretRect using localCaretRectForEmptyElement.
2012-06-08 Mike West <mkwst@chromium.org>
Treat blob: and filesystem: URLs generated via secure origins as secure.
https://bugs.webkit.org/show_bug.cgi?id=84054
Loading a blob: or filesystem: URL into an iframe or image that's
contained on an HTTPS page shouldn't generate a mixed content warning.
This change adds a SecurityOrigin::isSecure to check both against a
URLs protocol, and the protocol of it's so-called "inner URL" if it's
the type of URL that has such a thing. These sorts of URLs which are
generated from secure sources will themselves be treated as secure.
Reviewed by Adam Barth.
Tests: http/tests/security/mixedContent/blob-url-in-iframe.html
http/tests/security/mixedContent/filesystem-url-in-iframe.html
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::isMixedContent):
* page/SecurityOrigin.cpp:
(WebCore):
(WebCore::SecurityOrigin::isSecure):
* page/SecurityOrigin.h:
(SecurityOrigin):
2012-06-08 Martin Robinson <mrobinson@igalia.com>
[Cairo] [TextureMapperGL] Nvidia card does not update WebGL texture when interacting with canvas
https://bugs.webkit.org/show_bug.cgi?id=87797
Reviewed by Philippe Normand.
No new tests. WebGL is already covered by WebGL tests and this
bug is only present on certain hardware.
After resolving multisampling before drawing the texture via the
TextureMapper, first rebind the multisampling framebuffer. This ensures
that the texture isn't active in two framebuffers at a time.
* platform/graphics/cairo/GraphicsContext3DPrivate.cpp:
(WebCore::GraphicsContext3DPrivate::paintToTextureMapper): Rebind the multisampling
framebuffer.
2012-06-08 Alexey Proskuryakov <ap@apple.com>
<rdar://problem/11412294> Add appropriate iOS default font settings
Reviewed by Dan Bernstein.
* page/mac/SettingsMac.mm: (WebCore::Settings::initializeDefaultFontFamilies):
2012-06-08 Charles Wei <charles.wei@torchmobile.com.cn>
IndexedDB: Transactions without any request scheduled should abort itself.
https://bugs.webkit.org/show_bug.cgi?id=88052
Reviewed by Kentaro Hara.
This is needed for indexedDB to make it work for JSC. We need to clean up all the
pending transactions(transactions without any request scheduled) when leaving Javascript context.
Manually tested using the existing test LayoutTests/storage/indexeddb/tutorial.html
* bindings/js/JSMainThreadExecState.cpp:
(WebCore):
(WebCore::JSMainThreadExecState::didLeaveScriptContext):
* bindings/js/JSMainThreadExecState.h:
(WebCore::JSMainThreadExecState::~JSMainThreadExecState):
(JSMainThreadExecState):
2012-06-08 Ryosuke Niwa <rniwa@webkit.org>
Crash in WebCore::CompositeEditCommand::breakOutOfEmptyListItem
https://bugs.webkit.org/show_bug.cgi?id=88361
Reviewed by Levi Weintraub.
Use RefPtr instead of raw pointers across DOM mutations.
No tests are added for the lack of a reliable reduction.
* editing/CompositeEditCommand.cpp:
(WebCore::CompositeEditCommand::breakOutOfEmptyListItem):
2012-06-08 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119514.
http://trac.webkit.org/changeset/119514
https://bugs.webkit.org/show_bug.cgi?id=88664
Broke JSFiddle.net (Requested by arv on #webkit).
* bindings/v8/V8DOMWindowShell.cpp:
(WebCore::V8DOMWindowShell::initContextIfNeeded):
* bindings/v8/WorkerContextExecutionProxy.cpp:
(WebCore::WorkerContextExecutionProxy::initIsolate):
2012-06-08 Ryosuke Niwa <rniwa@webkit.org>
Crash in WebCore::InsertParagraphSeparatorCommand::doApply
https://bugs.webkit.org/show_bug.cgi?id=88108
Reviewed by Levi Weintraub.
Use NodeVector instead of walking through siblings as we mutate the DOM.
No new tests are added since there is no reliable reduction.
* editing/BreakBlockquoteCommand.cpp:
(WebCore::BreakBlockquoteCommand::doApply):
* editing/CompositeEditCommand.cpp:
(WebCore::CompositeEditCommand::moveRemainingSiblingsToNewParent):
(WebCore):
* editing/CompositeEditCommand.h:
(CompositeEditCommand):
* editing/InsertParagraphSeparatorCommand.cpp:
(WebCore::InsertParagraphSeparatorCommand::doApply):
2012-06-08 David Grogan <dgrogan@chromium.org>
IndexedDB: rename some instances of open to registerFrontendCallbacks
https://bugs.webkit.org/show_bug.cgi?id=88611
We should still make the change in the WebKit API.
Reviewed by Tony Chang.
No new tests, just a method rename.
* Modules/indexeddb/IDBDatabase.cpp:
(WebCore::IDBDatabase::registerFrontendCallbacks):
* Modules/indexeddb/IDBDatabase.h:
(IDBDatabase):
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::registerFrontendCallbacks):
* Modules/indexeddb/IDBDatabaseBackendImpl.h:
(IDBDatabaseBackendImpl):
* Modules/indexeddb/IDBDatabaseBackendInterface.h:
(IDBDatabaseBackendInterface):
* Modules/indexeddb/IDBRequest.cpp:
(WebCore::IDBRequest::onSuccess):
2012-06-08 Dana Jansens <danakj@chromium.org>
[chromium] Skip willDraw() and didDraw() on fully occluded layers
https://bugs.webkit.org/show_bug.cgi?id=88435
Reviewed by Adrienne Walker.
Current willDraw() is called on all layers with non-empty
visibleLayerRect and non-empty scissorRect. This excludes
layers outside the viewport, but does not exclude occluded
layers. We add a check for occlusion to calculateRenderPasses
in order to avoid willDraw() when it will be culled anyway.
We prevent didDraw() from being called for occluded layers, for
which we did not call didDraw() by holding a vector of layers
for which we did call willDraw(). This lets us avoid storing a
flag on the layers, or computing occlusion again in
didDrawAllLayers.
Unit test: CCLayerTreeHostImplTest.willDrawNotCalledOnOccludedLayer
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
(WebCore::CCLayerTreeHostImpl::prepareToDraw):
(WebCore::CCLayerTreeHostImpl::didDrawAllLayers):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(FrameData):
(CCLayerTreeHostImpl):
2012-06-08 Antonio Gomes <agomes@rim.com>
EventHandler shouldn't schedule the fake mousemove event timer when scrolling on devices that don't have a mouse
https://bugs.webkit.org/show_bug.cgi?id=88379
Reviewed by James Robinson.
Paraphrasing Andy Estes:
"In <http://trac.webkit.org/changeset/119465> we stopped dispatching fake
mousemove events when scrolling on devices that don't support mice. This event
is dispatched on a timer, so one better would be to not even schedule the
timer (rather than scheduling the timer but making it a no-op).
No new tests, as no functionality has changed. We basically do not
start a timer under a given circumstance instead of fire it, it times out,
and we make the callback no-op when under the given circumstance.
It is still possible though that we have the following scenario:
- Setting::deviceSupportsMouse set as true;
- we fire the timer;
- before its callback gets executed (timed out) the setting gets toggled.
So for sakeness, lets keep the check in both places.
* page/EventHandler.cpp:
(WebCore::EventHandler::dispatchFakeMouseMoveEventSoon):
(WebCore::EventHandler::fakeMouseMoveEventTimerFired):
2012-06-08 James Robinson <jamesr@chromium.org>
[chromium] Clean up some unnecessary LayerChromium.h includes
https://bugs.webkit.org/show_bug.cgi?id=88599
Reviewed by Adam Barth.
Reworks test code slightly to use WebCore interfaces.
* testing/Internals.cpp:
(WebCore::Internals::setBackgroundBlurOnNode):
2012-06-08 Erik Arvidsson <arv@chromium.org>
[V8] Keep TextTrackList alive as long as its owner is alive
https://bugs.webkit.org/show_bug.cgi?id=88541
Reviewed by Adam Barth.
Add a hidden reference from the owner to the text track list.
Covered by existing tests.
* Target.pri: Add V8TextTrackListCustom.cpp.
* UseV8.cmake: Ditto.
* WebCore.gypi: Ditto.
* bindings/v8/V8HiddenPropertyName.h:
(WebCore):
* bindings/v8/custom/V8TextTrackListCustom.cpp: Copied from Source/WebCore/bindings/v8/V8HiddenPropertyName.h.
(WebCore):
(WebCore::toV8): When the wrapper is created add a reference to the owner of the TextTrackList to the TextTrackList.
* html/track/TextTrackList.idl:
2012-06-08 Ian Vollick <vollick@chromium.org>
[chromium] Accelerated animations should use WebTransformOperations
https://bugs.webkit.org/show_bug.cgi?id=87686
Reviewed by James Robinson.
CCTransformKeyframe new owns a WebTransformOperations rather than a
TransformOperations. LayerChromium's API has been changed so that
LayerChromium::addAnimation should take only a CCActiveAnimation.
GraphicsLayerChromium is new responsible for translating to
WebTransformOperations and creating CCActiveAnimations. Tests that use
the public API (that is, they call addAnimation with KeyframeValueList
and Animation arguments) have been moved to GraphicsLayerChromiumTest.
Unit tests:
GraphicsLayerChromiumTest.createOpacityAnimation
GraphicsLayerChromiumTest.createTransformAnimation
GraphicsLayerChromiumTest.createTransformAnimationWithBigRotation
GraphicsLayerChromiumTest.createTransformAnimationWithSingularMatrix
GraphicsLayerChromiumTest.createReversedAnimation
GraphicsLayerChromiumTest.createAlternatingAnimation
GraphicsLayerChromiumTest.createReversedAlternatingAnimation
* WebCore.gypi:
* platform/graphics/chromium/AnimationTranslationUtil.cpp: Added.
(WebCore):
(WebCore::toWebTransformOperations):
(WebCore::appendKeyframe):
(WebCore::CCKeyframedTransformAnimationCurve):
(WebCore::createActiveAnimation):
* platform/graphics/chromium/AnimationTranslationUtil.h: Added.
(WebCore):
* platform/graphics/chromium/GraphicsLayerChromium.cpp:
(WebCore::GraphicsLayerChromium::addAnimation):
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::addAnimation):
* platform/graphics/chromium/LayerChromium.h:
(WebCore):
(WebCore::LayerChromium::bounds):
(LayerChromium):
* platform/graphics/chromium/LinkHighlight.cpp:
(WebCore::LinkHighlight::LinkHighlight):
* platform/graphics/chromium/cc/CCAnimationCurve.h:
(CCTransformAnimationCurve):
* platform/graphics/chromium/cc/CCKeyframedAnimationCurve.cpp:
(WebCore::CCTransformKeyframe::create):
(WebCore::CCTransformKeyframe::CCTransformKeyframe):
(WebCore::CCTransformKeyframe::value):
(WebCore::CCTransformKeyframe::clone):
(WebCore::CCKeyframedTransformAnimationCurve::getValue):
* platform/graphics/chromium/cc/CCKeyframedAnimationCurve.h:
(CCTransformKeyframe):
(CCKeyframedTransformAnimationCurve):
* platform/graphics/chromium/cc/CCLayerAnimationController.cpp:
(WebCore::CCLayerAnimationController::removeAnimation):
(WebCore):
(WebCore::CCLayerAnimationController::addAnimation):
(WebCore::CCLayerAnimationController::getActiveAnimation):
(WebCore::CCLayerAnimationController::pushNewAnimationsToImplThread):
(WebCore::CCLayerAnimationController::replaceImplThreadAnimations):
(WebCore::CCLayerAnimationController::tickAnimations):
* platform/graphics/chromium/cc/CCLayerAnimationController.h:
(CCLayerAnimationControllerClient):
(CCLayerAnimationController):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(WebCore::CCLayerImpl::bounds):
2012-06-08 No'am Rosenthal <noam.rosenthal@nokia.com>
[Texmap] Accelerated versions of drop-shadow and blur filters
https://bugs.webkit.org/show_bug.cgi?id=87695
Reviewed by Kenneth Rohde Christiansen.
Added shaders for blur and drop-shadow effects.
The shaders use sampling of multiple coordinates and averaging them with normal-
distribution to create a fast gaussian blur effect.
Covered by existing tests in css3/filters.
* platform/graphics/texmap/TextureMapperGL.cpp:
(WebCore::TextureMapperGL::drawFiltered):
Call the prepare function explicitly, with a size argument that's used to calculate
some of the uniform values.
* platform/graphics/texmap/TextureMapperLayer.cpp:
(WebCore::TextureMapperLayer::intermediateSurfaceRect):
The outsets of the effect have to be considered when calculating the intermediate rect.
Otherwise the resulting image is scaled instead of outsetted.
(WebCore::applyFilters):
* platform/graphics/texmap/TextureMapperShaderManager.cpp:
(WebCore::StandardFilterProgram::StandardFilterProgram):
Added the new shaders.
(WebCore::StandardFilterProgram::prepare):
(WebCore::TextureMapperShaderManager::getShaderForFilter):
* platform/graphics/texmap/TextureMapperShaderManager.h:
(StandardFilterProgram):
2012-06-08 Dan Bernstein <mitz@apple.com>
Tried to fix the build after r119844.
* bindings/js/GCController.cpp:
(WebCore::GCController::discardAllCompiledCode):
2012-06-08 Mike West <mkwst@chromium.org>
Add COMPILE_ASSERT to StyledElement to ensure it doesn't accidentally grow larger.
https://bugs.webkit.org/show_bug.cgi?id=88627
Reviewed by Adam Barth.
No change in user-facing behavior.
* dom/StyledElement.cpp:
(WebCore):
2012-06-08 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: Fix several compiler warnings and errors.
https://bugs.webkit.org/show_bug.cgi?id=88660
Reviewed by Pavel Feldman.
Fixed several compiler warnings and errors.
Removed isEmpty method from Object.prototype and made it static method on Object.
* inspector/InjectedScriptExterns.js:
* inspector/front-end/FileSystemModel.js:
(WebInspector.FileSystemModel.prototype._detachFrame):
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshot.prototype._buildDominatorTree):
* inspector/front-end/utilities.js:
(Object.isEmpty):
2012-06-08 Max Feil <mfeil@rim.com>
[BlackBerry] Fix assertion failure introduced by bug 87551
https://bugs.webkit.org/show_bug.cgi?id=88659
Reviewed by Antonio Gomes.
The determineRenderSlider() convenience function added in bug
87551 made an assumption that wasn't always true. This didn't
cause a functional error but it caused an assertion failure,
which indicates a problem with the code. PR164142.
No new tests because this case is already covered by ASSERT's.
* platform/blackberry/RenderThemeBlackBerry.cpp:
(WebCore::determineRenderSlider):
2012-06-08 Ion Rosca <rosca@adobe.com>
Some overlay scrollbar API calls in ScrollAnimatorMac can lead to an assertion in RenderBox::mapAbsoluteToLocalPoint
https://bugs.webkit.org/show_bug.cgi?id=74111
Reviewed by Simon Fraser.
Added zero-delay timer for AppKit scroll notification that can be called during layout.
Manual test: ManualTests/scrollbar-crash-on-hide-scrolled-area.html
This assertion does not fire when running layout tests. It can be easly reproduced using a debug build by loading the manual test page.
* platform/mac/ScrollAnimatorMac.h:
(ScrollAnimatorMac):
* platform/mac/ScrollAnimatorMac.mm:
(WebCore::ScrollAnimatorMac::ScrollAnimatorMac):
(WebCore::ScrollAnimatorMac::notifyContentAreaScrolled):
(WebCore::ScrollAnimatorMac::sendContentAreaScrolledSoon):
(WebCore):
(WebCore::ScrollAnimatorMac::sendContentAreaScrolledTimerFired):
2012-06-08 Renata Hodovan <reni@webkit.org>
Adding few already supported features to the FeatureSet in DOMImplementation
https://bugs.webkit.org/show_bug.cgi?id=86482
Text, Filter and View features are already implemented but they
are not added to the SVG11 feature set. This is corrected in this patch.
Reviewed by Nikolas Zimmermann.
No new testcase is needed because we already have one what checks the svg features via
DOMImplementation (svg/custom/svg-features.html).
* dom/DOMImplementation.cpp:
(WebCore::isSVG11Feature):
2012-06-08 Stephen Chenney <schenney@chromium.org>
WebCore::ImageBuffer.cpp has broken color table code
https://bugs.webkit.org/show_bug.cgi?id=80321
Reviewed by Dirk Schulze.
The code in ImageBuffer::transformColorSpace had two problems that are
fixed by this patch.
First, it was using member variables for data
that is static constant. This is incredibly wasteful, as every
ImageBuffer that is created (and we make a lot of them) gets its own
copy of the LUT and the table is initialized again and again. The
patch replaces this with a static local, with lazy evaluation, in
the method that uses the tables.
Second, the code for filling the table was mapping 255 to 254, thus
reducing the intensity and alpha of every image it pushed through a
color transform.
Test: svg/filters/color-space-conversion.svg
* platform/graphics/ImageBuffer.cpp:
(WebCore::ImageBuffer::transformColorSpace):
* platform/graphics/ImageBuffer.h:
(ImageBuffer):
2012-06-08 Alexei Filippov <alexeif@chromium.org>
Web Inspector: Better labels positioning on the memory pie chart
https://bugs.webkit.org/show_bug.cgi?id=88652
Reviewed by Vsevolod Vlasov.
* inspector/front-end/NativeMemorySnapshotView.js:
(WebInspector.NativeMemoryPieChart.prototype._paint.paintPercentAndLabel):
(WebInspector.NativeMemoryPieChart.prototype._paint):
2012-06-08 Bryan McQuade <bmcquade@google.com>
Web Inspector: Annotate timeline records with a frame identifier
https://bugs.webkit.org/show_bug.cgi?id=86406
Reviewed by Vsevolod Vlasov.
Tests: Updated existing layout tests.
* inspector/InspectorController.cpp:
(WebCore::InspectorController::InspectorController):
* inspector/InspectorInstrumentation.cpp:
(WebCore::frameForScriptExecutionContext):
(WebCore):
(WebCore::InspectorInstrumentation::didScheduleResourceRequestImpl):
(WebCore::InspectorInstrumentation::didInstallTimerImpl):
(WebCore::InspectorInstrumentation::didRemoveTimerImpl):
(WebCore::InspectorInstrumentation::willCallFunctionImpl):
(WebCore::InspectorInstrumentation::willChangeXHRReadyStateImpl):
(WebCore::InspectorInstrumentation::willDispatchEventImpl):
(WebCore::InspectorInstrumentation::willDispatchEventOnWindowImpl):
(WebCore::InspectorInstrumentation::willEvaluateScriptImpl):
(WebCore::InspectorInstrumentation::willFireTimerImpl):
(WebCore::InspectorInstrumentation::willLayoutImpl):
(WebCore::InspectorInstrumentation::willLoadXHRImpl):
(WebCore::InspectorInstrumentation::willPaintImpl):
(WebCore::InspectorInstrumentation::willRecalculateStyleImpl):
(WebCore::InspectorInstrumentation::willSendRequestImpl):
(WebCore::InspectorInstrumentation::willReceiveResourceDataImpl):
(WebCore::InspectorInstrumentation::willReceiveResourceResponseImpl):
(WebCore::InspectorInstrumentation::didFinishLoadingImpl):
(WebCore::InspectorInstrumentation::didFailLoadingImpl):
(WebCore::InspectorInstrumentation::domContentLoadedEventFiredImpl):
(WebCore::InspectorInstrumentation::loadEventFiredImpl):
(WebCore::InspectorInstrumentation::willWriteHTMLImpl):
(WebCore::InspectorInstrumentation::didRequestAnimationFrameImpl):
(WebCore::InspectorInstrumentation::didCancelAnimationFrameImpl):
(WebCore::InspectorInstrumentation::willFireAnimationFrameImpl):
* inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
(WebCore::InspectorInstrumentation::didScheduleResourceRequest):
(WebCore::InspectorInstrumentation::didInstallTimer):
(WebCore::InspectorInstrumentation::didRemoveTimer):
(WebCore::InspectorInstrumentation::willCallFunction):
(WebCore::InspectorInstrumentation::willChangeXHRReadyState):
(WebCore::InspectorInstrumentation::willDispatchEvent):
(WebCore::InspectorInstrumentation::willEvaluateScript):
(WebCore::InspectorInstrumentation::willFireTimer):
(WebCore::InspectorInstrumentation::willLayout):
(WebCore::InspectorInstrumentation::willLoadXHR):
(WebCore::InspectorInstrumentation::willPaint):
(WebCore::InspectorInstrumentation::willRecalculateStyle):
(WebCore::InspectorInstrumentation::willReceiveResourceData):
(WebCore::InspectorInstrumentation::willReceiveResourceResponse):
(WebCore::InspectorInstrumentation::willWriteHTML):
(WebCore::InspectorInstrumentation::didRequestAnimationFrame):
(WebCore::InspectorInstrumentation::didCancelAnimationFrame):
(WebCore::InspectorInstrumentation::willFireAnimationFrame):
* inspector/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::pushGCEventRecords):
(WebCore::InspectorTimelineAgent::didBeginFrame):
(WebCore::InspectorTimelineAgent::willCallFunction):
(WebCore::InspectorTimelineAgent::willDispatchEvent):
(WebCore::InspectorTimelineAgent::willLayout):
(WebCore::InspectorTimelineAgent::willRecalculateStyle):
(WebCore::InspectorTimelineAgent::willPaint):
(WebCore::InspectorTimelineAgent::willComposite):
(WebCore::InspectorTimelineAgent::willWriteHTML):
(WebCore::InspectorTimelineAgent::didInstallTimer):
(WebCore::InspectorTimelineAgent::didRemoveTimer):
(WebCore::InspectorTimelineAgent::willFireTimer):
(WebCore::InspectorTimelineAgent::willChangeXHRReadyState):
(WebCore::InspectorTimelineAgent::willLoadXHR):
(WebCore::InspectorTimelineAgent::willEvaluateScript):
(WebCore::InspectorTimelineAgent::didScheduleResourceRequest):
(WebCore::InspectorTimelineAgent::willSendResourceRequest):
(WebCore::InspectorTimelineAgent::willReceiveResourceData):
(WebCore::InspectorTimelineAgent::willReceiveResourceResponse):
(WebCore::InspectorTimelineAgent::didFinishLoadingResource):
(WebCore::InspectorTimelineAgent::didTimeStamp):
(WebCore::InspectorTimelineAgent::didMarkDOMContentEvent):
(WebCore::InspectorTimelineAgent::didMarkLoadEvent):
(WebCore::InspectorTimelineAgent::didRequestAnimationFrame):
(WebCore::InspectorTimelineAgent::didCancelAnimationFrame):
(WebCore::InspectorTimelineAgent::willFireAnimationFrame):
(WebCore::InspectorTimelineAgent::addRecordToTimeline):
(WebCore::InspectorTimelineAgent::innerAddRecordToTimeline):
(WebCore::InspectorTimelineAgent::didCompleteCurrentRecord):
(WebCore::InspectorTimelineAgent::InspectorTimelineAgent):
(WebCore::InspectorTimelineAgent::appendRecord):
(WebCore::InspectorTimelineAgent::pushCurrentRecord):
(WebCore::InspectorTimelineAgent::pushCancelableRecord):
(WebCore::InspectorTimelineAgent::commitCancelableRecords):
* inspector/InspectorTimelineAgent.h:
(WebCore):
(WebCore::InspectorTimelineAgent::create):
(InspectorTimelineAgent):
(WebCore::InspectorTimelineAgent::TimelineRecordEntry::TimelineRecordEntry):
(TimelineRecordEntry):
* inspector/WorkerInspectorController.cpp:
(WebCore::WorkerInspectorController::WorkerInspectorController):
* inspector/front-end/TimelinePresentationModel.js:
(WebInspector.TimelinePresentationModel.isEventDivider):
(WebInspector.TimelinePresentationModel.Record):
2012-06-08 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119821.
http://trac.webkit.org/changeset/119821
https://bugs.webkit.org/show_bug.cgi?id=88648
Seems to have regression on Mac (Requested by kinuko on
#webkit).
* fileapi/File.cpp:
(WebCore::File::File):
(WebCore::File::lastModifiedDate):
(WebCore::File::lastModifiedDateForBinding):
(WebCore):
(WebCore::File::captureSnapshot):
* fileapi/File.h:
(File):
* fileapi/File.idl:
* platform/FileMetadata.h:
(WebCore::FileMetadata::FileMetadata):
* platform/FileSystem.h:
* platform/chromium/support/WebHTTPBody.cpp:
(WebKit::WebHTTPBody::elementAt):
* platform/network/BlobData.cpp:
(WebCore):
* platform/network/BlobData.h:
(BlobDataItem):
(WebCore::BlobDataItem::BlobDataItem):
* platform/network/FormData.cpp:
(WebCore::FormData::appendFile):
* platform/network/cf/FormDataStreamCFNet.cpp:
(WebCore::advanceCurrentStream):
* platform/network/soup/ResourceHandleSoup.cpp:
(WebCore::blobIsOutOfDate):
2012-06-08 Kinuko Yasuda <kinuko@chromium.org>
Unreviewed, removing duplicated entries in gyp.
* WebCore.gypi:
2012-06-08 Arko Saha <arko@motorola.com>
Microdata: An element with an empty itemprop='' should not get listed in the .properties list.
https://bugs.webkit.org/show_bug.cgi?id=88235
Reviewed by Ryosuke Niwa.
As per the specification change http://html5.org/r/6854, an element should be listed in .properties
if the elelement has an attribute itemprop specified and the element has one or more property names.
Test: fast/dom/MicroData/element-with-empty-itemprop.html
* html/HTMLPropertiesCollection.cpp:
(WebCore::HTMLPropertiesCollection::itemAfter):
2012-06-07 Kinuko Yasuda <kinuko@chromium.org>
File::lastModifiedDate should use NaN or separate boolean flag for null Date value
https://bugs.webkit.org/show_bug.cgi?id=87826
Reviewed by Kent Tamura.
Test: http/tests/local/fileapi/file-last-modified-after-delete.html
* fileapi/File.cpp:
(WebCore::File::File):
(WebCore::File::captureSnapshot):
(WebCore::File::lastModifiedDate):
(WebCore::File::lastModifiedDateForBinding): Removed.
* fileapi/File.h:
(File):
* fileapi/File.idl:
* platform/FileMetadata.h:
(WebCore::FileMetadata::FileMetadata):
* platform/FileSystem.h:
* platform/chromium/support/WebHTTPBody.cpp:
(WebKit::WebHTTPBody::elementAt):
* platform/network/BlobData.h:
(WebCore::BlobDataItem::doNotCheckFileChange): Removed.
* platform/network/BlobData.cpp:
* platform/network/cf/FormDataStreamCFNet.cpp:
(WebCore::advanceCurrentStream):
* platform/network/soup/ResourceHandleSoup.cpp:
(WebCore::blobIsOutOfDate):
2012-06-08 Kent Tamura <tkent@chromium.org>
Move form-related functions of Document to FormController
https://bugs.webkit.org/show_bug.cgi?id=88497
Reviewed by Hajime Morita.
A lot of files depend on Document.h. We had to build many files when we
touched form-related features in Document.h or
CheckedRadioButtons.h. This patch reduces such pain.
No new tests. Just a refactoring.
* CMakeLists.txt: Add FormController.cpp and/or FormController.h.
* GNUmakefile.list.am: ditto.
* Target.pri: ditto.
* WebCore.gypi: ditto.
* WebCore.vcproj/WebCore.vcproj: ditto.
* WebCore.xcodeproj/project.pbxproj: ditto.
* dom/Document.cpp: Removed form-releated features except the followings.
(WebCore::Document::formController): Accessor for a FormController.
(WebCore::Document::formElementsState):
Just returns FormController::formElementsState().
We don't create new FormController.
(WebCore::Document::setStateForNewFormElements):
Just returns FormController::setStateForNewFormElements().
We don't create new FormController if the specified vector is empty.
* dom/Document.h:
Moved form-related functions to FormController.
* html/FormController.cpp: Added. Moved from Document.cpp.
* html/FormController.h: Added. Moved from Document.h
* html/FormAssociatedElement.cpp: Use Document::formController().
(WebCore::FormAssociatedElement::didMoveToNewDocument):
(WebCore::FormAssociatedElement::insertedInto):
(WebCore::FormAssociatedElement::removedFrom):
(WebCore::FormAssociatedElement::formAttributeChanged):
* html/HTMLFormControlElementWithState.cpp: Use Document::formController().
(WebCore::HTMLFormControlElementWithState::HTMLFormControlElementWithState):
(WebCore::HTMLFormControlElementWithState::~HTMLFormControlElementWithState):
(WebCore::HTMLFormControlElementWithState::didMoveToNewDocument):
(WebCore::HTMLFormControlElementWithState::finishParsingChildren):
* html/HTMLFormElement.cpp: Use Document::formController().
(WebCore::HTMLFormElement::didNotifyDescendantInsertions):
(WebCore::HTMLFormElement::removedFrom):
* html/HTMLInputElement.cpp: Use Document::formController().
(WebCore::HTMLInputElement::~HTMLInputElement):
(WebCore::HTMLInputElement::didMoveToNewDocument):
(WebCore::HTMLInputElement::checkedRadioButtons):
* html/HTMLInputElement.h:
Declare CheckedRadioButtons. It was provided by Document.h.
* html/parser/HTMLConstructionSite.h:
Declare HTMLFormElement. It was provided by Document.h
2012-06-08 Taiju Tsuiki <tzik@chromium.org>
Add FileSystem item to storage tree.
https://bugs.webkit.org/show_bug.cgi?id=72691
Reviewed by Vsevolod Vlasov.
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* inspector/compile-front-end.py:
* inspector/front-end/FileSystemModel.js: Added.
* inspector/front-end/Images/fileSystem.png: Added.
* inspector/front-end/ResourcesPanel.js:
(WebInspector.FileSystemListTreeElement):
(WebInspector.FileSystemListTreeElement.prototype.onexpand):
(WebInspector.FileSystemListTreeElement.prototype.onattach):
(WebInspector.FileSystemListTreeElement.prototype._handleContextMenuEvent):
(WebInspector.FileSystemListTreeElement.prototype._fileSystemAdded):
(WebInspector.FileSystemListTreeElement.prototype._fileSystemRemoved):
(WebInspector.FileSystemListTreeElement.prototype._fileSystemTreeElementByName):
(WebInspector.FileSystemListTreeElement.prototype._refreshFileSystem):
(WebInspector.FileSystemTreeElement):
(WebInspector.FileSystemTreeElement.prototype.get fileSystemName):
(WebInspector.FileSystemTreeElement.prototype.get itemURL):
* inspector/front-end/Settings.js:
(WebInspector.ExperimentsSettings):
* inspector/front-end/WebKit.qrc:
* inspector/front-end/inspector.html:
* inspector/front-end/resourcesPanel.css:
(.file-system-storage-tree-item .icon):
* inspector/front-end/utilities.js:
2012-06-08 Kentaro Hara <haraken@chromium.org>
REGRESSION r119802: Remove unnecessary ASSERT(!m_deletionHasBegun) from Document.h
https://bugs.webkit.org/show_bug.cgi?id=88528
Reviewed by Ryosuke Niwa.
m_deletionHasBegun should become true just before 'delete this'.
No tests. No change in behavior.
* dom/Document.h:
(WebCore::Document::guardRef):
(WebCore::Document::guardDeref):
2012-06-08 Adam Barth <abarth@webkit.org>
HTTP 204 No Content responses shouldn't trigger MIME type warnings in the web inspector
https://bugs.webkit.org/show_bug.cgi?id=88603
Reviewed by Pavel Feldman.
A 204 response doesn't have an entity body, so its MIME type is kind of
meaningless. We shouldn't spam the console with a warning about MIME
types.
* inspector/front-end/NetworkManager.js:
(WebInspector.NetworkDispatcher.prototype._mimeTypeIsConsistentWithType):
2012-06-08 Yoshifumi Inoue <yosin@chromium.org>
[Forms] Drop StepRange::NumberWithDecimalPlacesOrMissing
https://bugs.webkit.org/show_bug.cgi?id=88621
Reviewed by Kent Tamura.
This patch changes NumberWithDecimalPlacesOrMissing to NumberWithDecimalPlaces.
We use NaN as indication of having value or not.
No new tests, because this patch doesn't change behavior.
* html/DateInputType.cpp:
(WebCore::DateInputType::createStepRange): Replace NumberWithDecimalPlacesOrMissing to NumberWithDecimalPlaces.
* html/DateTimeInputType.cpp:
(WebCore::DateTimeInputType::createStepRange): ditto.
* html/DateTimeLocalInputType.cpp:
(WebCore::DateTimeLocalInputType::createStepRange): ditto.
* html/MonthInputType.cpp:
(WebCore::MonthInputType::createStepRange): ditto.
* html/NumberInputType.cpp:
(WebCore::NumberInputType::createStepRange): ditto.
* html/RangeInputType.cpp:
(WebCore::RangeInputType::createStepRange): ditto.
* html/StepRange.cpp:
(WebCore::StepRange::StepRange): Checks value is finite or not.
(WebCore::StepRange::parseStep): Replace NumberWithDecimalPlacesOrMissing to NumberWithDecimalPlaces.
* html/StepRange.h:
(StepRange): Removed NumberWithDecimalPlacesOrMissing.
* html/TimeInputType.cpp:
(WebCore::TimeInputType::createStepRange): NumberWithDecimalPlacesOrMissing to NumberWithDecimalPlaces.
* html/WeekInputType.cpp:
(WebCore::WeekInputType::createStepRange): ditto.
2012-06-08 Kentaro Hara <haraken@chromium.org>
Unreviewed. Fixed a comment, pointed out by ap@.
* platform/TreeShared.h:
(WebCore):
2012-06-08 Noel Gordon <noel.gordon@gmail.com>
[chromium] WEBP image blue and red color channels reversed on Android
https://bugs.webkit.org/show_bug.cgi?id=88613
Reviewed by Kent Tamura.
No new tests. Covered by test fast/images/webp-image-decoding.html though I'm
not sure where/how the Androids run their layout tests.
* platform/image-decoders/webp/WEBPImageDecoder.cpp:
(outputMode): Detect little-endian Skia with a blue pixel shift (aka Android)
which seems to be the way for them per bug 75861.
2012-06-08 Ryosuke Niwa <rniwa@webkit.org>
Remove the assertion that has been moved to TreeScope after r119802.
* dom/Document.cpp:
(WebCore::Document::removedLastRef):
2012-06-07 Kentaro Hara <haraken@chromium.org>
Reduce Node object size from 72 byte to 64 byte
https://bugs.webkit.org/show_bug.cgi?id=88528
Reviewed by Ryosuke Niwa.
This patch removes all virtual methods from TreeShared.h,
by which we can remove a virtual method table pointer (8 byte
in a 64bit architechture) from each Node object. Consequently,
this patch reduces the Node object size from 72 byte to 64 byte.
e.g. The HTML spec (http://www.whatwg.org/specs/web-apps/current-work/)
contains 325640 Node objects. Thus this patch saves 325640 * 8 byte = 2.6 MB.
e.g. sizeof(Element) is reduced from 104 byte to 96 byte.
- In multiple inheritance, a virtual method table pointer is allocated
for each base class that has virtual methods. For example, for
'class A : public B, C {};' where B and C have virtual methods,
two virtual method table pointers are allocated for each A object.
In this patch, A = Node, B = EventTarget, and C = TreeShared.
By removing the virtual methods from TreeShared, we can save the virtual
method table pointer for TreeShared. 8 byte saving.
- Node and SVGElementInstance are the only classes that inherit TreeShared.
- This patch removes virtual TreeShared::removeLastRef() and implements
virtual Node::removeLastRef() and virtual SVGElementInstance::removeLastRef().
Node::removeLastRef() calls 'delete this' for 'this' of type Node*,
and SVGElementInstance::removeLastRef() calls 'delete this' for 'this' of type
SVGElementInstance*.
- This patch removes the virtual destructor of TreeShared. This removal is safe
because (1) no one calls 'delete this' for 'this' of type TreeShared*,
and (2) both Node and SVGElementInstance have virtual destructor.
No change in behavior. Confirm no regression in existing tests.
* dom/ContainerNode.cpp:
(WebCore::callRemovedLastRef):
(WebCore):
* dom/Node.h:
(WebCore::Node::removedLastRef):
* platform/TreeShared.h:
(WebCore):
(TreeShared):
(WebCore::TreeShared::deref):
* svg/SVGElementInstance.cpp:
(WebCore::callRemovedLastRef):
(WebCore):
* svg/SVGElementInstance.h:
(WebCore::SVGElementInstance::removedLastRef):
* WebCore.order: Added a symbol for callRemovedLastRef().
* WebCore.exp.in: Ditto.
2012-06-07 Takashi Sakamoto <tasak@google.com>
ShadowRoot needs resetStyleInheritance
https://bugs.webkit.org/show_bug.cgi?id=84048
Added resetStyleInheritance API to ShadowRoot. The spec URL is
http://dvcs.w3.org/hg/webcomponents/raw-file/tip/spec/shadow/index.html#shadow-root-attributes
To implement the API, modified StyleResolver to use empty style
instead of parent style if resetStyleInheritance is true,
and a node and its parent node are placed in different shadow trees.
Reviewed by Hajime Morita.
Test: fast/dom/shadow/shadow-root-resetStyleInheritance.html
fast/dom/shadow/shadow-root-resetStyleInheritance-expected.html
* css/StyleResolver.cpp:
(WebCore::StyleResolver::initForStyleResolve):
Set m_parentStyle to NULL when resetStyleInheritance.
(WebCore::StyleResolver::documentSettings):
Added a new method to obtain current document's settings.
(WebCore::StyleResolver::initializeFontStyle):
Initialize m_style by using document settings. Its code was move
from the code initializing font style in applyProperty.
(WebCore::StyleResolver::styleForElement):
Initialize font style if m_parentStyle is NULL. The old code just
used style() and don't touch font style. The new code uses the above
initializeFontStyle.
(WebCore::StyleResolver::applyProperty):
Moved a code for initializing font style as "initial" to
a new method: initializeFontStyle and modified to use the method.
Modified to use documentSettings().
(WebCore::StyleResolver::checkForGenericFamilyChange):
Modified to use documentSettings().
* css/StyleResolver.h:
(WebCore::StyleResolver::documentSettings):
(WebCore::StyleResolver::initializeFontStyle):
Added two new methods to class StyleResolver.
* dom/NodeRenderingContext.cpp:
(WebCore::NodeRenderingContext::NodeRenderingContext):
Modified to calculate m_resetStyleInheritance in its constructor.
* dom/NodeRenderingContext.h:
(WebCore::NodeRenderingContext::resetStyleInheritance):
Implemented a new method, just returning m_resetStyleInheritance.
* dom/ShadowRoot.cpp:
(WebCore::ShadowRoot::setResetStyleInheritance):
Modified to force to recalculate children's styles if
resetStyleInheritance is changed.
* dom/ShadowRoot.h:
(ShadowRoot):
Added resetStyleInheritance getter and setter.
* dom/ShadowRoot.idl:
Added an attribute, resetStyleInheritance.
* dom/TreeScope.cpp:
(WebCore::TreeScope::resetStyleInheritance):
* dom/TreeScope.h:
(TreeScope):
Added resetStyleInheritance's default getter.
The getter will be overrided by class ShadowRoot's getter.
2012-06-07 Victor Carbune <victor@rosedu.org>
addTextTrack should set track mode to HIDDEN
https://bugs.webkit.org/show_bug.cgi?id=88317
Reviewed by Eric Carlson.
Updated existing test.
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::addTextTrack): Set the default parameters
for the newly created TextTrack.
* html/track/TextTrack.cpp:
(WebCore::TextTrack::TextTrack):
2012-06-07 Li Yin <li.yin@intel.com>
FileAPI: Blob should support ArrayBufferView instead of ArrayBuffer for Constructor Parameters
https://bugs.webkit.org/show_bug.cgi?id=88294
Reviewed by Jian Li.
From Spec: http://dev.w3.org/2006/webapi/FileAPI/#dfn-Blob
Currently we add the support for ArrayBufferView, while still keeping ArrayBuffer for
backward compatibility. We will remove it in the near future.
Test: fast/files/blob-constructor.html
* bindings/js/JSBlobCustom.cpp:
(WebCore::JSBlobConstructor::constructJSBlob):
* bindings/v8/custom/V8BlobCustom.cpp:
(WebCore::V8Blob::constructorCallback):
* fileapi/WebKitBlobBuilder.cpp:
(WebCore::WebKitBlobBuilder::append):
(WebCore):
* fileapi/WebKitBlobBuilder.h:
(WebCore):
(WebKitBlobBuilder):
* fileapi/WebKitBlobBuilder.idl: Add support for ArrayBufferView in append method
2012-06-07 MORITA Hajime <morrita@google.com>
A style in an older shadow subtree causes assert when composing with <shadow>
https://bugs.webkit.org/show_bug.cgi?id=88299
Reviewed by Dimitri Glazkov.
InsertionPoint::attach() assumes its distributed content not being attach()-ed.
But this assumption can break. This change added a guard for that.
This can happen for shadow boundaries in general. But ShadowRoot already handles that case.
Test: fast/dom/shadow/insertion-point-shadow-crash.html
* html/shadow/InsertionPoint.cpp:
(WebCore::InsertionPoint::attach):
2012-06-07 Max Feil <mfeil@rim.com>
[BlackBerry] Show correct fullscreen button image (media controls)
https://bugs.webkit.org/show_bug.cgi?id=88563
Reviewed by Antonio Gomes.
PR159306: Show "exit" fullscreen button image instead of "enter"
fullscreen button image when in fullscreen mode (HTML5 media
controls).
I am not providing a test due to the simplicity of this patch
and the relative difficulty and maintainability issues associated
with testing how an image is rendered.
* platform/blackberry/RenderThemeBlackBerry.cpp:
(WebCore::RenderThemeBlackBerry::paintMediaFullscreenButton):
2012-06-07 Noel Gordon <noel.gordon@gmail.com>
[chromium] Use WEBPImportPictureRGBX|BGRX to import picture data
https://bugs.webkit.org/show_bug.cgi?id=88218
Reviewed by Kent Tamura.
Stop using the RGBA and RGBA variants of the picture import routines (since
these will import alpha channel data in a future libwebp). Use the RGBX and
BGRX variants to import picture data for encoding to make it clear that the
alpha channel is (and must be) ignored.
Test: fast/canvas/canvas-toDataURL-webp.html
* platform/image-encoders/skia/WEBPImageEncoder.cpp:
(WebCore::rgbPictureImport):
(WebCore::importPictureBGRX): Use WEBPImportPictureBGRX for picture imports
that must ignore the alpha channel. Change the template parameter to be the
premultiplied alpha state of the input image data.
(WebCore::importPictureRGBX): Use WEBPImportPictureRGBX, ditto.
(WebCore::encodePixels): Invert the sense of the template parameter to make
it now indicate the premultiplied alpha state of the input image data.
2012-06-07 Takashi Sakamoto <tasak@google.com>
Improve the performance of pushScope in StyleResolver
https://bugs.webkit.org/show_bug.cgi?id=88222
Reviewed by Hajime Morita.
As setupScopeStack always sets m_scopeStackParent to be NULL,
m_scopeStack is never reused. m_scopeStackParent should be the last
element of m_scopeStack.
No new tests, because fast/css/style-scoped/ uses pushScope,
popScope and setupScopeStack and checks whether there exists
any crash bug or not. And this patch doesn't change any behavior
of scoped author syltes.
* css/StyleResolver.cpp:
(WebCore::StyleResolver::setupScopeStack):
Modified to set m_scopeSackParent to be the last element of
m_scopeStack.
(WebCore::StyleResolver::popScope):
Modified to remove the last element if m_scopeStack is not empty and
the last element of m_scopeStack has the same scope as the scoping
element given by the argument.
2012-06-07 Mark Pilgrim <pilgrim@chromium.org>
[Chromium] Move didStartWorkerRunLoop to Platform.h
https://bugs.webkit.org/show_bug.cgi?id=88562
Reviewed by Adam Barth.
Part of a refactoring series. See tracking bug 82948.
* WebCore.gyp/WebCore.gyp:
* bindings/v8/WorkerScriptController.cpp:
(WebCore::WorkerScriptController::~WorkerScriptController):
* platform/chromium/PlatformSupport.h:
(WebCore):
* workers/WorkerThread.cpp:
(WebCore::WorkerThread::workerThread):
2012-06-07 Mike West <mkwst@chromium.org>
Dropping m_startingLineNumber property from StyledElement.
https://bugs.webkit.org/show_bug.cgi?id=86848
Reviewed by Adam Barth.
We can have the same effect by moving the line-number calculation
directly into StyledElement::styleAttributeChanged. This means that
subsequent changes to the style attribute (via script) won't generate
error messages with line numbers, but that's an acceptable tradeoff.
Behavior is covered by existing CSP tests.
* dom/StyledElement.cpp:
(WebCore::StyledElement::StyledElement):
(WebCore::StyledElement::styleAttributeChanged):
* dom/StyledElement.h:
(WebCore::StyledElement::destroyInlineStyle):
2012-06-07 Peter Beverloo <peter@chromium.org>
Initialize m_document in AXObjectCache's constructor for !HAVE(ACCESSIBILITY) builds
https://bugs.webkit.org/show_bug.cgi?id=88527
Reviewed by Adam Barth.
Initialize the m_document member with the passed on document in the stubbed
AXObjectCache constructor. No behavioral changes.
* accessibility/AXObjectCache.h:
(WebCore::AXObjectCache::AXObjectCache):
2012-06-05 James Robinson <jamesr@chromium.org>
[chromium] Move deferral-related logic out of Canvas2DLayerChromium
https://bugs.webkit.org/show_bug.cgi?id=86050
Reviewed by Stephen White.
This removes Canvas2DLayerChromium and puts the deferral and double-buffering related logic in a new class,
Canvas2DLayerBridge, that uses TextureLayerChromium for compositing. The bridge is responsible for managing the
front texture when double buffering is active, inserting the correct makeContextCurrent() and flush() calls on
the GraphicsContext3D and SkCanvas when appropriate, and owning the lifetime of the layer.
* WebCore.gypi:
* platform/graphics/chromium/Canvas2DLayerBridge.cpp: Added.
(WebCore):
(AcceleratedDeviceContext):
(WebCore::AcceleratedDeviceContext::AcceleratedDeviceContext):
(WebCore::AcceleratedDeviceContext::prepareForDraw):
(WebCore::Canvas2DLayerBridge::Canvas2DLayerBridge):
(WebCore::Canvas2DLayerBridge::~Canvas2DLayerBridge):
(WebCore::Canvas2DLayerBridge::skCanvas):
(WebCore::Canvas2DLayerBridge::prepareTexture):
(WebCore::Canvas2DLayerBridge::context):
(WebCore::Canvas2DLayerBridge::layer):
(WebCore::Canvas2DLayerBridge::contextAcquired):
* platform/graphics/chromium/Canvas2DLayerBridge.h: Added.
(WebCore):
(Canvas2DLayerBridge):
(WebCore::Canvas2DLayerBridge::create):
* platform/graphics/chromium/Canvas2DLayerChromium.cpp: Removed.
* platform/graphics/chromium/Canvas2DLayerChromium.h: Removed.
* platform/graphics/chromium/GraphicsLayerChromium.cpp:
* platform/graphics/chromium/ImageBufferDataSkia.h:
(WebCore):
(ImageBufferData):
* platform/graphics/chromium/TextureLayerChromium.cpp:
(WebCore::TextureLayerChromium::willModifyTexture):
willModifyTexture() is called whenever the texture currently owned by the TextureLayerChromium will be
modified by a system out of the compositor's control. It makes sure that the compositor does not attempt to
draw with the potentially-modified texture until the next commit. For canvas, this is used when threading is
enabled and a non-deferrable drawing command is invoked from JS.
(WebCore):
* platform/graphics/chromium/TextureLayerChromium.h:
(TextureLayerChromium):
* platform/graphics/chromium/cc/CCTextureUpdater.cpp:
(WebCore::CCTextureUpdater::hasMoreUpdates):
(WebCore::CCTextureUpdater::update):
(WebCore::CCTextureUpdater::clear):
* platform/graphics/chromium/cc/CCTextureUpdater.h:
(CCTextureUpdater):
* platform/graphics/skia/ImageBufferSkia.cpp:
(WebCore::createAcceleratedCanvas):
(WebCore::ImageBuffer::~ImageBuffer):
(WebCore::ImageBuffer::context):
(WebCore::ImageBuffer::platformLayer):
2012-06-07 Patrick Gansterer <paroga@webkit.org>
Remove HAVE(PATH_BASED_BORDER_RADIUS_DRAWING)
https://bugs.webkit.org/show_bug.cgi?id=87931
Reviewed by Beth Dakin.
r62035 added HAVE(PATH_BASED_BORDER_RADIUS_DRAWING) until all ports implement
GraphicsContext::clipConvexPolygon(). All major ports define HAVE_PATH_BASED_BORDER_RADIUS_DRAWING in the
meantime and the alternative path has unnecessary maintainace costs and even don't compile at the moment.
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::drawBoxSideFromPath):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::paintFocusRing):
* rendering/RenderObject.h:
(RenderObject):
2012-06-06 Vincent Scheib <scheib@chromium.org>
Add new Pointer Lock spec events webkitpointerlockchange and webkitpointerlockerror
https://bugs.webkit.org/show_bug.cgi?id=88377
Reviewed by Dimitri Glazkov.
Part of a series of refactoring changes to update pointer lock API to
the fullscreen locking style. https://bugs.webkit.org/show_bug.cgi?id=84402
New change and error events added. Follow up patches will remove
the previous callbacks and pointerlocklost event. Tests updated
to include the new events, and a new test added specifically for
the new events, pointerlockchange-pointerlockerror-events.html.
Test: pointer-lock/pointerlockchange-pointerlockerror-events.html
* dom/Document.h:
(Document):
* dom/Document.idl:
* dom/EventNames.h:
(WebCore):
* page/PointerLockController.cpp:
(WebCore::PointerLockController::requestPointerLock):
(WebCore::PointerLockController::didAcquirePointerLock):
(WebCore::PointerLockController::didNotAcquirePointerLock):
(WebCore::PointerLockController::didLosePointerLock):
(WebCore::PointerLockController::enqueueEvent):
(WebCore):
* page/PointerLockController.h:
(PointerLockController):
2012-06-07 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119744.
http://trac.webkit.org/changeset/119744
https://bugs.webkit.org/show_bug.cgi?id=88584
Fails assertions in debug builds (Requested by jamesr_ on
#webkit).
* platform/graphics/chromium/cc/CCIOSurfaceLayerImpl.cpp:
(WebCore::CCIOSurfaceLayerImpl::~CCIOSurfaceLayerImpl):
(WebCore::CCIOSurfaceLayerImpl::willDraw):
2012-06-07 Julien Chaffraix <jchaffraix@webkit.org>
Cache RenderLayer::isRootLayer for better performance
https://bugs.webkit.org/show_bug.cgi?id=88570
Reviewed by Ojan Vafai.
Caching covered by existing tests.
RenderLayer::isRootLayer()'s answer will be the same during the RenderLayer's
lifetime as a RenderLayer is tied to a RenderBoxModelObject. This change caches
isRootLayer value in the constructor.
On http://dglazkov.github.com/performance-tests/biggrid.html with a 100,000 rows
by 100 columns table, it saves about 10% on the paint time during scrolling on
my machine (going from 50ms to 45ms). It would expect all code paths to see some
improvement as checking renderer()->isRenderView() was pretty common.
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::RenderLayer):
Added code to cache the result of renderer()->isRenderView() here.
(WebCore::RenderLayer::stackingContext):
(WebCore::isPositionedContainer):
(WebCore::isFixedPositionedContainer):
(WebCore::RenderLayer::enclosingTransformedAncestor):
(WebCore::RenderLayer::clippingRootForPainting):
(WebCore::shouldSuppressPaintingLayer):
(WebCore::RenderLayer::hitTest):
(WebCore::RenderLayer::intersectsDamageRect):
* rendering/RenderLayer.h:
(WebCore::RenderLayer::isStackingContext):
Updated all the call sites to use isRootLayer() instead of renderer()->isRenderView()
as it is faster.
(WebCore::RenderLayer::isRootLayer):
Added our caching here. Also made m_canSkipRepaintRectsUpdateOnScroll to follow
m_isRootLayer example.
2012-06-07 Takashi Sakamoto <tasak@google.com>
Incorrect border rendering when border radius is above 2px.
https://bugs.webkit.org/show_bug.cgi?id=88046
Fixing findInnerVertex's logic to decide which is the better line
to find an inner vertex, x=center point's x of a rendering box or
y=center point's y of a rendering box.
Reviewed by Simon Fraser.
Test: fast/borders/border-radius-valid-border-clipping.html
* rendering/RenderBoxModelObject.cpp:
(WebCore::findInnerVertex):
2012-06-07 Edaena Salinas Jasso <edaena@apple.com>
Should fire error event for empty 404 script
https://bugs.webkit.org/show_bug.cgi?id=50589
Reviewed by Brady Eidson.
Tests:
http/tests/loading/fire-error-event-empty-404-script.html
http/tests/loading/fire-error-script-no-content-type.html
* loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::didReceiveResponse): We should check for http status
codes here to stop depending on the response being non-empty.
(WebCore::SubresourceLoader::didReceiveData): Moved the http status check but
avoid messing with multipart responses.
(WebCore::SubresourceLoader::checkForHTTPStatusCodeError): Renamed from
errorLoadingResource for clarity.
* loader/SubresourceLoader.h:
* loader/cf/SubresourceLoaderCF.cpp:
(WebCore::SubresourceLoader::didReceiveDataArray):
Updated for errorLoadingResource renaming.
2012-06-07 Konrad Piascik <kpiascik@rim.com>
[BlackBerry] Update PlayBook's User Agent String for Web Inspector
https://bugs.webkit.org/show_bug.cgi?id=88543
Reviewed by Antonio Gomes.
We expect users to upgrade their devices so developers only need the latest UA String.
No tests needed.
* inspector/front-end/SettingsScreen.js:
(WebInspector.SettingsScreen.prototype._createUserAgentSelectRowElement.get const):
2012-06-07 Shawn Singh <shawnsingh@chromium.org>
[chromium] Remove obsolete FIXME comment about a layer sorting bug.
https://bugs.webkit.org/show_bug.cgi?id=88546
Reviewed by James Robinson.
No tests needed, only comment changed. The comment referred to a
potential bug that actually will not occur. The W3C spec on CSS 3d
transforms indicates that layers that do not share the same "3d
rendering context" do not sort together, and our existing code
sorts this way already.
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::subtreeShouldRenderToSeparateSurface):
2012-06-07 Adam Barth <abarth@webkit.org>
Settings::defaultDeviceScaleFactor is redundant with Page::deviceScaleFactor
https://bugs.webkit.org/show_bug.cgi?id=88375
Reviewed by James Robinson.
This patch removes Settings::defaultDeviceScaleFactor because it is
redundant with Page::deviceScaleFactor now that we no longer support
target-densitydpi. Page::deviceScaleFactor is the canonical place to
store the device scale factor.
* WebCore.exp.in:
* page/Settings.cpp:
(WebCore::Settings::Settings):
(WebCore):
* page/Settings.h:
(Settings):
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::setDeviceScaleFactor):
2012-06-07 Alec Flett <alecflett@chromium.org>
IndexedDB: Error codes, phase one
https://bugs.webkit.org/show_bug.cgi?id=88459
Reviewed by Tony Chang.
This phase gets rid of the now-deprecated NOT_ALLOWED_ERR and
NON_TRANSIENT_ERR from earlier versions of the spec, and
normalizes the place where TypeError should be thrown so that it's
easier to update when the bindings change.
No new tests: existing tests cover these changes and have been updated.
* Modules/indexeddb/IDBCursor.cpp:
(WebCore::IDBCursor::advance):
(WebCore::IDBCursor::continueFunction):
(WebCore::IDBCursor::stringToDirection):
(WebCore::IDBCursor::directionToString):
* Modules/indexeddb/IDBCursorBackendImpl.cpp:
(WebCore::IDBCursorBackendImpl::update):
(WebCore::IDBCursorBackendImpl::deleteFunction):
* Modules/indexeddb/IDBDatabase.cpp:
(WebCore::IDBDatabase::createObjectStore):
(WebCore::IDBDatabase::deleteObjectStore):
(WebCore::IDBDatabase::setVersion):
(WebCore::IDBDatabase::transaction):
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::createObjectStore):
(WebCore::IDBDatabaseBackendImpl::deleteObjectStore):
(WebCore::IDBDatabaseBackendImpl::setVersion):
* Modules/indexeddb/IDBDatabaseException.cpp:
(WebCore):
* Modules/indexeddb/IDBDatabaseException.h:
* Modules/indexeddb/IDBDatabaseException.idl:
* Modules/indexeddb/IDBFactory.cpp:
(WebCore::IDBFactory::open):
(WebCore::IDBFactory::deleteDatabase):
* Modules/indexeddb/IDBObjectStore.cpp:
(WebCore::IDBObjectStore::index):
* Modules/indexeddb/IDBObjectStoreBackendImpl.cpp:
(WebCore::IDBObjectStoreBackendImpl::createIndex):
(WebCore::IDBObjectStoreBackendImpl::deleteIndex):
* Modules/indexeddb/IDBRequest.cpp:
(WebCore::IDBRequest::result):
(WebCore::IDBRequest::errorCode):
(WebCore::IDBRequest::webkitErrorMessage):
* Modules/indexeddb/IDBTransaction.cpp:
(WebCore::IDBTransaction::objectStore):
(WebCore::IDBTransaction::stringToMode):
(WebCore::IDBTransaction::modeToString):
* Modules/indexeddb/IDBTransactionBackendImpl.cpp:
(WebCore::IDBTransactionBackendImpl::objectStore):
2012-06-07 Dana Jansens <danakj@chromium.org>
[chromium] Free texture from CCIOSurfaceLayerImpl when it is destroyed
https://bugs.webkit.org/show_bug.cgi?id=88371
Reviewed by James Robinson.
Unit test: CCLayerTreeHostImplTest.layersFreeTextures
* platform/graphics/chromium/cc/CCIOSurfaceLayerImpl.cpp:
(WebCore::CCIOSurfaceLayerImpl::~CCIOSurfaceLayerImpl):
(WebCore::CCIOSurfaceLayerImpl::willDraw):
2012-06-07 Peter Beverloo <peter@chromium.org>
Buildfix for ENABLE_OVERFLOW_SCROLLING=0 (though defined)
https://bugs.webkit.org/show_bug.cgi?id=88564
Reviewed by Adam Barth.
Where ENABLE(X) mandates that X both has to be defined and evaluates to
true as a lone preprocessor statement, there were two places in the CSS
property and value input files where only was checked whether is has
been defined. This also goes against other #ifdefs in those files.
* css/CSSPropertyNames.in:
* css/CSSValueKeywords.in:
2012-06-07 Ami Fischman <fischman@chromium.org>
Plumb CORS attribute information from HTMLMediaElement to media players so it can be used
https://bugs.webkit.org/show_bug.cgi?id=88349
Reviewed by Adam Barth.
Test: http/tests/security/video-cross-origin-readback.html
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::mediaPlayerCORSMode):
(WebCore):
* html/HTMLMediaElement.h:
(HTMLMediaElement):
* html/canvas/CanvasRenderingContext.cpp:
(WebCore::CanvasRenderingContext::wouldTaintOrigin):
* platform/graphics/MediaPlayer.cpp:
(WebCore::MediaPlayer::didPassCORSAccessCheck):
(WebCore):
* platform/graphics/MediaPlayer.h:
(WebCore::MediaPlayerClient::mediaPlayerCORSMode):
(MediaPlayer):
* platform/graphics/MediaPlayerPrivate.h:
(WebCore::MediaPlayerPrivateInterface::didPassCORSAccessCheck):
(MediaPlayerPrivateInterface):
2012-06-07 Mark Hahnenberg <mhahnenberg@apple.com>
Worker tear-down can re-enter JSC during GC finalization
https://bugs.webkit.org/show_bug.cgi?id=88449
Reviewed by Geoffrey Garen.
No new tests.
This is the first of two patches to fix this issue with Workers.
* workers/AbstractWorker.cpp:
(WebCore::AbstractWorker::~AbstractWorker): We don't need to call onDestroyWorker() here, it
will be called elsewhere in contextDestroyed().
2012-06-07 Jer Noble <jer.noble@apple.com>
sometimes all slaved videos don't start playing
https://bugs.webkit.org/show_bug.cgi?id=88553
Reviewed by Darin Adler.
Test: media/media-controller-time-clamp.html
Some PlatformClock classes will occasionally return times < 0 and will
always return times slightly > duration() when playback has ended. Clamp
the value of currentTime() to the specified [0..duration] range.
* html/MediaController.cpp:
(MediaController::currentTime):
2012-06-07 Simon Fraser <simon.fraser@apple.com>
Optimize FrameView::scrollXForFixedPosition() / scrollYForFixedPosition()
https://bugs.webkit.org/show_bug.cgi?id=88475
Reviewed by Sam Weinig.
FrameView's scrollXForFixedPosition() and scrollYForFixedPosition()
methods were often called together, but they do duplicate work,
including calling into platform widget code which might be slow.
Fix by converting scrollOffsetForFixedPosition() from being a wrapper
that just calls scrollXForFixedPosition() and scrollYForFixedPosition()
to the method that does all the work, calling just once into platform
widget code.
Changed callers to use scrollOffsetForFixedPosition() rather than make
two separate method calls.
Added ScrollView::layoutSize() and visibleSize() methods for
convenience.
Removed FrameView::scrollXForFixedPosition and FrameView::scrollYForFixedPosition
to avoid inefficient callers in future.
No new tests; refactoring only.
* page/FrameView.cpp:
(WebCore::fixedPositionScrollOffset):
(WebCore::FrameView::scrollOffsetForFixedPosition):
* page/FrameView.h: Removed scrollXForFixedPosition and scrollYForFixedPosition.
* platform/ScrollView.cpp:
(WebCore::ScrollView::layoutSize):
* platform/ScrollView.h:
(WebCore::ScrollView::visibleSize):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::backgroundClipRect):
* rendering/RenderLayer.h:
(WebCore::ClipRect::move):
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::requiresCompositingForPosition):
* rendering/RenderView.cpp:
(WebCore::RenderView::computeRectForRepaint):
2012-06-07 Raymes Khoury <raymes@chromium.org>
Incorrect rect-based hit-test result for culled-inline elements
https://bugs.webkit.org/show_bug.cgi?id=85849
Reviewed by Levi Weintraub.
Modified code which blindly adds culled inlines to rect-based hit-test
results so that it only does so if the child node does not fully cover
the hit-test region.
Test: fast/dom/nodesFromRect-culled-inline.html
* rendering/HitTestResult.cpp:
(WebCore::HitTestResult::addNodeToRectBasedTestResult):
2012-06-07 Daniel Erat <derat@chromium.org>
Make Skia backend honor requests for subpixel-positioned text.
https://bugs.webkit.org/show_bug.cgi?id=88263
Reviewed by Tony Chang.
Add subpixel positioning field to FontRenderStyle and rename
subpixel rendering field.
Pass setting to SkPaint in FontPlatformDataHarfBuzz and remove
round() call when computing glyph widths in SimpleFontDataSkia.
Test: platform/chromium-linux/fast/text/chromium-linux-text-subpixel-positioning.html
* platform/graphics/blackberry/skia/PlatformSupport.cpp:
(WebCore::setFontRenderStyleDefaults):
(WebCore::PlatformSupport::getRenderStyleForStrike):
* platform/graphics/chromium/FontRenderStyle.h:
(WebCore::FontRenderStyle::FontRenderStyle):
(WebCore::FontRenderStyle::operator==):
(FontRenderStyle):
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.cpp:
(WebCore):
(WebCore::FontPlatformData::setSubpixelRendering):
(WebCore::FontPlatformData::setSubpixelPositioning):
(WebCore::FontPlatformData::setupPaint):
(WebCore::FontPlatformData::querySystemForRenderStyle):
* platform/graphics/harfbuzz/FontPlatformDataHarfBuzz.h:
(FontPlatformData):
* platform/graphics/skia/SimpleFontDataSkia.cpp:
(WebCore::SimpleFontData::platformWidthForGlyph):
2012-06-06 Dana Jansens <danakj@chromium.org>
[chromium] In each composited frame, didDraw() should only be called on layers for which willDraw() was called
https://bugs.webkit.org/show_bug.cgi?id=88469
Reviewed by James Robinson.
Unit test: CCLayerTreeHostImplTest.didDrawNotCalledOnScissoredLayer
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::shouldDrawLayer):
(WebCore):
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
(WebCore::CCLayerTreeHostImpl::didDrawAllLayers):
2012-06-07 Alexei Filippov <alexeif@chromium.org>
Web Inspector: sorting of object fields is broken in heap profiler
https://bugs.webkit.org/show_bug.cgi?id=88532
A recent change to heap profiler has replaced getters with functions.
Function calls were missing in couple places after that change.
Reviewed by Vsevolod Vlasov.
Tests: inspector/profiler/heap-snapshot-summary-sorting-fields.html
inspector/profiler/heap-snapshot-summary-sorting-instances.html
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshotEdgesProvider.prototype.sort.compareNodeField):
2012-06-07 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: Consider Ctrl+Shift+key as valid zoom change combinations
https://bugs.webkit.org/show_bug.cgi?id=88520
Reviewed by Vsevolod Vlasov.
* inspector/front-end/inspector.js:
(WebInspector.documentKeyDown):
2012-06-07 Arvid Nilsson <anilsson@rim.com>
[BlackBerry] Allow WebPageCompositor to blend a transparent web page
https://bugs.webkit.org/show_bug.cgi?id=88233
Reviewed by Rob Buis.
RIM PR #159998
Add a parameter to LayerRenderer::compositeBuffer() indicating whether
the buffer contents are opaque, and set the GL blend mode accordingly.
This is not currently testable using BlackBerry testing infrastructure.
Reviewed internally by Jakob Petsovits.
* platform/graphics/blackberry/LayerRenderer.cpp:
(WebCore::LayerRenderer::compositeBuffer):
* platform/graphics/blackberry/LayerRenderer.h:
(LayerRenderer):
2012-06-07 Alexei Filippov <alexeif@chromium.org>
Web Inspector: reuse edge_count field of heap snapshot to store retained size
https://bugs.webkit.org/show_bug.cgi?id=88416
The edge_count field is unused after node first edge indexes have been built.
Store node retained size in there instead of allocating an extra array for it.
Reviewed by Vsevolod Vlasov.
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshotNode.prototype.retainedSize):
(WebInspector.HeapSnapshot.prototype._calculateRetainedSizes):
2012-06-07 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119694.
http://trac.webkit.org/changeset/119694
https://bugs.webkit.org/show_bug.cgi?id=88529
it broke PrerenderBrowserTest.PrerenderHTML5VideoNetwork
(Requested by loislo on #webkit).
* html/HTMLMediaElement.cpp:
* html/HTMLMediaElement.h:
(HTMLMediaElement):
* html/canvas/CanvasRenderingContext.cpp:
(WebCore::CanvasRenderingContext::wouldTaintOrigin):
* platform/graphics/MediaPlayer.cpp:
* platform/graphics/MediaPlayer.h:
(MediaPlayer):
* platform/graphics/MediaPlayerPrivate.h:
2012-06-07 Shinya Kawanaka <shinyak@chromium.org>
Delete a contenteditable element having a Shadow DOM causes a crash.
https://bugs.webkit.org/show_bug.cgi?id=86345
Reviewed by Ryosuke Niwa.
In VisibleSelection::validate(), m_start.downstream() or m_end.upstream() might cross
editing boundaries. So we adjust selection after calling them again.
Test: editing/shadow/select-contenteditable-shadowhost.html
* editing/VisibleSelection.cpp:
(WebCore::VisibleSelection::validate):
2012-06-07 Vivek Galatage <vivekgalatage@gmail.com>
No new line at the end for SVGCSSPropertyNames.in causes compilation warning
https://bugs.webkit.org/show_bug.cgi?id=88503
Reviewed by Ryosuke Niwa.
Added new line at the end of the file.
No new tests required.
* css/SVGCSSPropertyNames.in:
2012-06-07 Csaba Osztrogonác <ossy@webkit.org>
[Qt][Win] Use OS(WINDOWS) instead of Q_WS_WIN in WebCore/platform/graphics/GraphicsContext.h
https://bugs.webkit.org/show_bug.cgi?id=88305
Reviewed by Simon Hausmann.
* platform/graphics/GraphicsContext.h: Buildfix.
(GraphicsContext):
2012-06-07 Shinya Kawanaka <shinyak@chromium.org>
Incorrect data retrieved in calls to get selection data in client->textWillBeReplaced from CharacterData::setDataAndUpdate
https://bugs.webkit.org/show_bug.cgi?id=66120
Reviewed by Ryosuke Niwa.
Since FrameSelection::textWillBeReplaced was called before replacing text, updated Selection became
temporarily invalid. The invalid selection was set to updateSelection(), so client will have
invalid selection.
This patch makes calling updateSelection() after updating selection.
No new tests, it's too hard to create a correct test cases in DRT.
* dom/CharacterData.cpp:
(WebCore::CharacterData::setDataAndUpdate):
* editing/FrameSelection.cpp:
(WebCore::updatePositionAfterAdoptingTextReplacement):
(WebCore::FrameSelection::textWasReplaced):
* editing/FrameSelection.h:
(FrameSelection):
2012-06-07 Li Yin <li.yin@intel.com>
FileAPI: If type consists of non-ASCII characters in Blob constructor, it should throw a SyntaxError.
https://bugs.webkit.org/show_bug.cgi?id=88411
Reviewed by Kentaro Hara.
From spec: http://dev.w3.org/2006/webapi/FileAPI/#constructorBlob
If type consists of any non-ASCII characters, throw a SyntaxError and
return from this algorithm.
This patch checks the String is ASCii or not, if not, throw SyntaxError.
Test: fast/files/blob-constructor.html
* bindings/js/JSBlobCustom.cpp:
(WebCore::JSBlobConstructor::constructJSBlob):
* bindings/v8/custom/V8BlobCustom.cpp:
(WebCore::V8Blob::constructorCallback):
2012-06-07 Arpita Bahuguna <arpitabahuguna@gmail.com>
InsertUnorderedList and InsertOrderedList (execCommand) do not remove bullets
https://bugs.webkit.org/show_bug.cgi?id=84597
Reviewed by Ryosuke Niwa.
Comparison between start and end VisiblePositions will fail when a position inside
a paragraph is compared with one at the start of the paragraph.
This fix thus converts the end VisiblePosition (which could be inside the paragraph)
to the start of the paragraph thereby enabling a proper comparison of start and end
positions.
Test: editing/execCommand/remove-list-from-multi-list-items.html
* editing/InsertListCommand.cpp:
(WebCore::InsertListCommand::selectionHasListOfType):
2012-06-07 Li Yin <li.yin@intel.com>
[FileAPI] FileReader should throw an InvalidStateError exception when readyState is LOADING
https://bugs.webkit.org/show_bug.cgi?id=88212
Reviewed by Jian Li.
From Spec: http://www.w3.org/TR/FileAPI/#readAsBinaryString
FileReader should throw an InvalidStateError exception instead of NOT_ALLOWED_ERR
when the readyState is LOADING.
Also delete OperationNotAllowedException related code, because it is not longer needed.
Tests: fast/files/read-file-async.html
fast/files/workers/worker-read-file-async.html
* CMakeLists.txt:
* DerivedSources.cpp:
* DerivedSources.make:
* DerivedSources.pri:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* dom/DOMExceptions.in:
* fileapi/FileReader.cpp:
(WebCore::FileReader::readInternal): Changed it to throw INVALID_STATE_ERR.
* fileapi/FileReader.idl: Changed it to raise DOMException for all read methods.
* fileapi/OperationNotAllowedException.cpp: Removed.
* fileapi/OperationNotAllowedException.h: Removed.
* fileapi/OperationNotAllowedException.idl: Removed.
2012-06-07 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119689.
http://trac.webkit.org/changeset/119689
https://bugs.webkit.org/show_bug.cgi?id=88516
it broke didDrawNotCalledOnScissoredLayer webkit_unit_test
(Requested by loislo on #webkit).
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
(WebCore::CCLayerTreeHostImpl::didDrawAllLayers):
2012-06-07 Ami Fischman <fischman@chromium.org>
Plumb CORS attribute information from HTMLMediaElement to media players so it can be used
https://bugs.webkit.org/show_bug.cgi?id=88349
Reviewed by Adam Barth.
Test: http/tests/security/video-cross-origin-readback.html
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::mediaPlayerCORSMode):
(WebCore):
* html/HTMLMediaElement.h:
(HTMLMediaElement):
* platform/graphics/MediaPlayer.h:
(WebCore::MediaPlayerClient::mediaPlayerCORSMode):
2012-06-07 Hironori Bono <hbono@chromium.org>
Use light gray for grammar markers on Windows and Linux
https://bugs.webkit.org/show_bug.cgi?id=88398
Reviewed by Ryosuke Niwa.
This change changes the color of grammar markers from dark gray to light gray
and rebaselines a couple of pixel tests affected by this change.
No new tests because it changes the marker color of an existing test 'grammar-markers.html'.
* platform/graphics/skia/GraphicsContextSkia.cpp: Change dark gray to light gray.
2012-06-07 Yoshifumi Inoue <yosin@chromium.org>
[Platform] Introduce conversion from/to Deciaml to/from double
https://bugs.webkit.org/show_bug.cgi?id=88480
Reviewed by Kent Tamura.
This patch added functions for Decimal/Double conversion and isInfinity
for reducing patch size of introducing Decimal arithmetic, bug 88383.
Tests: WebKit/chromium/tests/DecimalTest.cpp
* platform/Decimal.cpp:
(WebCore::Decimal::fromDouble): Added.
(WebCore::Decimal::toDouble): Added.
* platform/Decimal.h:
(WebCore::Decimal::EncodedData::isInfinity): Added for isinf()
(WebCore::Decimal::isInfinity): Added.
2012-06-06 Xianzhu Wang <wangxianzhu@chromium.org>
Element.getBoundingClientRect() and Element.getClientRects() return incorrect values in frames in a scaled page
https://bugs.webkit.org/show_bug.cgi?id=88430
Element.getBoudingClientRect() and Element.getClientRects() should not scale back the rects
if the element is in a frame because page scale only apply to elements in the main frame.
Reviewed by James Robinson.
Tests: fast/dom/Element/scale-page-bounding-client-rect-in-frame.html
fast/dom/Element/scale-page-client-rects-in-frame.html
* dom/Document.cpp:
(WebCore):
(WebCore::Document::adjustFloatQuadsForScrollAndAbsoluteZoomAndFrameScale): Extracted from Element.cpp and Range.cpp and fixed issue of pageScale in frames.
(WebCore::Document::adjustFloatRectForScrollAndAbsoluteZoomAndFrameScale): Extracted from Element.cpp and fixed issue of pageScale in frames.
* dom/Document.h:
(WebCore):
(Document):
* dom/Element.cpp:
(WebCore::Element::getClientRects):
(WebCore::Element::getBoundingClientRect):
* dom/Range.cpp:
(WebCore):
(WebCore::adjustFloatQuadsForScrollAndAbsoluteZoomAndPageScale): Moved into Document and fixed issue of pageScale in frames.
(WebCore::Range::getBorderAndTextQuads):
* rendering/RenderObject.h: Removed adjustFloatQuadForPageScale and adjustFloatRectForPageScale because they seem unrelated to RenderObject.
(WebCore):
2012-06-06 Dana Jansens <danakj@chromium.org>
[chromium] In each composited frame, didDraw() should only be called on layers for which willDraw() was called
https://bugs.webkit.org/show_bug.cgi?id=88469
Reviewed by James Robinson.
Unit test: CCLayerTreeHostImplTest.didDrawNotCalledOnScissoredLayer
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::shouldDrawLayer):
(WebCore):
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
(WebCore::CCLayerTreeHostImpl::didDrawAllLayers):
2012-06-06 Andy Estes <aestes@apple.com>
Wrap uppercaseWord, lowercaseWord, and capitalizeWord with USE(APPKIT)
https://bugs.webkit.org/show_bug.cgi?id=88504
Reviewed by Dan Bernstein.
uppercaseWord, lowercaseWord, and capitalizeWord are methods declared
by NSResponder that subclasses are expected to implement. WebKit should
only support these implementations on platforms that use AppKit.
* WebCore.exp.in:
* editing/Editor.cpp:
* editing/Editor.h:
* loader/EmptyClients.h:
* page/ContextMenuController.cpp:
* page/EditorClient.h:
2012-06-06 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119683.
http://trac.webkit.org/changeset/119683
https://bugs.webkit.org/show_bug.cgi?id=88505
it broke webkit-unit-test
WebLayerTreeViewThreadedTest.InstrumentationCallbacks on mac
(Requested by loislo on #webkit).
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::initializeLayerRenderer):
(WebCore::CCLayerTreeHost::setVisible):
2012-06-06 Robert Kroeger <rjkroege@chromium.org>
[Chromium] Re-enable handling of smooth scrolling on Chromium Linux/Windows
https://bugs.webkit.org/show_bug.cgi?id=87535 by adding support for
m_hasPreciseScrollingDelta on all Chromium platforms and adjusting ScrollAnimator
and ScrollAnimatorNone accordingly to let m_hasPreciseScrollingDelta == true
force exact (non-animated) scrolling.
Reviewed by James Robinson.
Updated WebKit unit tests to validate change.
* platform/PlatformWheelEvent.h:
(WebCore::PlatformWheelEvent::PlatformWheelEvent): Adjusted #ifdefs to
make m_hasPreciseScrollingDelta available on all Chromium platforms.
(PlatformWheelEvent):
* platform/ScrollAnimator.cpp:
(WebCore::ScrollAnimator::handleWheelEvent): Default scroll type is set based on
m_hasPreciseScrollingDelta value.
* platform/ScrollAnimatorNone.cpp:
(WebCore::ScrollAnimatorNone::scroll): Scroll precisely when necessary.
* platform/ScrollTypes.h: Added additional pixel scrolling subtype.
* platform/ScrollableArea.cpp:
(WebCore::ScrollableArea::scroll): Updated for additional pixel scroll subtype.
2012-06-06 Michal Mocny <mmocny@google.com>
[chromium] Stop dropping texture limits when the layer tree host becomes invisible, and initialize with 0 allocation.
https://bugs.webkit.org/show_bug.cgi?id=87747
Reviewed by Adrienne Walker.
GpuMemoryManager manages texture memory allocation limits. It will send a 0 allocation when a renderer becomes
invisible, and an appropriate allocation after a renderer is initialized. These numbers will change
depending on the state of memory across devices and other renderers, so we would like to remove all explicit
management by the renderers themselves.
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::initializeLayerRenderer):
(WebCore::CCLayerTreeHost::setVisible):
2012-06-06 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119680.
http://trac.webkit.org/changeset/119680
https://bugs.webkit.org/show_bug.cgi?id=88498
It broke mac build (Requested by kinuko on #webkit).
* fileapi/File.cpp:
(WebCore::File::File):
(WebCore::File::lastModifiedDate):
(WebCore::File::lastModifiedDateForBinding):
(WebCore):
(WebCore::File::captureSnapshot):
* fileapi/File.h:
(File):
* fileapi/File.idl:
* platform/FileMetadata.h:
(WebCore::FileMetadata::FileMetadata):
* platform/FileSystem.h:
* platform/chromium/support/WebHTTPBody.cpp:
(WebKit::WebHTTPBody::elementAt):
* platform/network/BlobData.cpp:
(WebCore):
2012-06-06 Kinuko Yasuda <kinuko@chromium.org>
File::lastModifiedDate should use NaN or separate boolean flag for null Date value
https://bugs.webkit.org/show_bug.cgi?id=87826
Reviewed by Kent Tamura.
Test: http/tests/local/fileapi/file-last-modified-after-delete.html
* fileapi/File.cpp:
(WebCore::File::File):
(WebCore::File::captureSnapshot):
(WebCore::File::lastModifiedDate):
(WebCore::File::lastModifiedDateForBinding): Removed.
* fileapi/File.h:
(File):
* fileapi/File.idl:
* platform/FileMetadata.h:
(WebCore::FileMetadata::FileMetadata):
* platform/FileSystem.h:
* platform/chromium/support/WebHTTPBody.cpp:
(WebKit::WebHTTPBody::elementAt):
* platform/network/BlobData.cpp:
(WebCore):
2012-06-06 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119668.
http://trac.webkit.org/changeset/119668
https://bugs.webkit.org/show_bug.cgi?id=88493
Hitting assertions in debug builds (Requested by rniwa on
#webkit).
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::localCaretRect):
* rendering/RenderBoxModelObject.cpp:
* rendering/RenderBoxModelObject.h:
(RenderBoxModelObject):
* rendering/RenderInline.cpp:
* rendering/RenderInline.h:
(RenderInline):
2012-06-06 Julien Chaffraix <jchaffraix@webkit.org>
Cache isSelfPaintingLayer() for better performance
https://bugs.webkit.org/show_bug.cgi?id=88464
Reviewed by Simon Fraser.
Covered by existing tests (repaint tests among them).
isSelfPaintingLayer() has shown up several times on some scrolling benchmarks due
to the function being called several time per paint phase. This change caches the
boolean at style change time.
On http://dglazkov.github.com/performance-tests/biggrid.html, this nearly speed-up
painting by 2 (lowering the time taken to paint the newly exposed area when scrolling
on a 10,000 * 100 table from 95ms to 50ms).
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::RenderLayer):
Initialized m_isSelfPaintingLayer here.
(WebCore::RenderLayer::paintLayer):
Changed the order of the checks to get the more likely check first.
(WebCore::RenderLayer::shouldBeSelfPaintingLayer):
Renamed from isSelfPaintingLayer to make m_isNormalFlowOnly.
(WebCore::RenderLayer::styleChanged):
Added code to update our cached m_isSelfPaintingLayer.
* rendering/RenderLayer.h:
(WebCore::RenderLayer::isSelfPaintingLayer):
Changed to return m_isSelfPaintingLayer.
2012-06-06 Yoshifumi Inoue <yosin@chromium.org>
REGRESSION(r109729) [Form] Rendering of select/optgroup/option combination is too slow.
https://bugs.webkit.org/show_bug.cgi?id=88059
Reviewed by Kent Tamura.
This patch changes not to share RenderStyle object for option and
optgroup element regardless attributes.
When we tried to shared RenderStyle object, rendering performance was
good but we had issue in selected option and text transform, see
https://bugs.webkit.org/show_bug.cgi?id=88405 for details.
No new tests. This patch doesn't change behavior but rendering performance.
* css/StyleResolver.cpp:
(WebCore::StyleResolver::canShareStyleWithElement): Changed to always
return false for option and optgroup.
2012-06-06 Alec Flett <alecflett@chromium.org>
IndexedDB: Optimize single-key get()
https://bugs.webkit.org/show_bug.cgi?id=85288
Reviewed by Tony Chang.
No new tests, this is just an optimization.
After a recent refactoring, we started creating
an internal cursor with every call to get(). The
most common use of get() is with a single key,
so provide a fast-path to avoid creating the cursor.
* Modules/indexeddb/IDBIndexBackendImpl.cpp:
(WebCore::IDBIndexBackendImpl::getByRangeInternal):
* Modules/indexeddb/IDBKeyRange.h:
(WebCore::IDBKeyRange::isOnlyKey):
(IDBKeyRange):
* Modules/indexeddb/IDBObjectStoreBackendImpl.cpp:
(WebCore::IDBObjectStoreBackendImpl::getByRangeInternal):
2012-06-06 Shezan Baig <shezbaig.wk@gmail.com>
Caret is not rendered in empty inline contenteditable elements
https://bugs.webkit.org/show_bug.cgi?id=85793
Reviewed by Ryosuke Niwa.
Override localCaretRect in RenderInline. The implementation was almost
identical to localCaretRect in RenderBlock for empty block elements, so
I refactored RenderBlock::localCaretRect and moved the logic to a new
method 'localCaretRectForEmptyElement' in RenderBoxModelObject. The
implementation of 'localCaretRect' in RenderBlock and RenderInline both
use this helper method in RenderBoxModelObject.
Tests: editing/selection/caret-in-empty-inline-1.html
editing/selection/caret-in-empty-inline-2.html
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::localCaretRect):
Modified to use RenderBoxModelObject::localCaretRectForEmptyElement.
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::localCaretRectForEmptyElement):
(WebCore):
* rendering/RenderBoxModelObject.h:
(RenderBoxModelObject):
Add localCaretRectForEmptyElement helper method.
* rendering/RenderInline.cpp:
(WebCore::RenderInline::localCaretRect):
(WebCore):
* rendering/RenderInline.h:
(RenderInline):
Override localCaretRect using localCaretRectForEmptyElement.
2012-06-06 Alexis Menard <alexis.menard@openbossa.org>
Fix border-image-slice failures on http://css3test.com.
https://bugs.webkit.org/show_bug.cgi?id=88328
Reviewed by Tony Chang.
The old parsing code was assuming that the fill keyword comes after the
numbers or the percentage values. http://www.w3.org/TR/css3-background/#the-border-image-slice
defines the property like this : [<number> | <percentage>]{1,4} && fill? where the fill
keyword is not mandatory. But according to http://www.w3.org/TR/CSS2/about.html "&&"
separates two or more components, all of which must occur, in any order. So the previous
code was wrong the fill can come first following by the percentages or the numbers. http://css3test.com
was covering this particular case. I took the opportunity to extend our coverage
by adding tests on many different values the boder-image-slice property could take. I
also did a small cleanup by removing unused functions in BorderImageSliceParseContext.
Test: fast/css/getComputedStyle/getComputedStyle-border-image-slice.html
* css/CSSParser.cpp:
(WebCore::BorderImageSliceParseContext::BorderImageSliceParseContext):
(WebCore::BorderImageSliceParseContext::commitNumber):
(WebCore::BorderImageSliceParseContext::commitFill):
2012-06-06 Levi Weintraub <leviw@chromium.org>
Use enclosingIntRect instead of pixelSnappedIntRect in RenderFileUploadControl
https://bugs.webkit.org/show_bug.cgi?id=88454
Reviewed by Kent Tamura.
Properly clipping the file upload control to avoid clipping out the top of the file upload button
when the control is positioned at a sub-pixel offset that rounds up.
Test: fast/sub-pixel/file-upload-control-at-fractional-offset.html
* rendering/RenderFileUploadControl.cpp:
(WebCore::RenderFileUploadControl::paintObject):
2012-06-06 Filip Pizlo <fpizlo@apple.com>
Global object variable accesses should not require an extra load
https://bugs.webkit.org/show_bug.cgi?id=88385
Reviewed by Gavin Barraclough and Geoffrey Garen.
Updated JSDOMWindowBase.cpp to use the new symbol table API. this->symbolTableFoo(...)
becomes symbolTableFoo(this, ...).
No new tests because no change in behavior.
* bindings/js/JSDOMWindowBase.cpp:
(WebCore::JSDOMWindowBase::updateDocument):
2012-06-06 James Robinson <jamesr@chromium.org>
[chromium] Move implementation of WebCore::GraphicsContext3D and related from WebKit/chromium/src to WebCore/platform/chromium/support
https://bugs.webkit.org/show_bug.cgi?id=86257
Reviewed by Kenneth Russell.
The WebCore platform interfaces GraphicsContext3D and Extensions3DChromium are implemented in chromium
on top of the Platform interface WebGraphicsContext3D. This moves the implementation support code from
WebKit/chromium/src to WebCore/platform/chromium/support, which avoids having code in WebKit/ implementing WebCore
interfaces and allows code in WebCore/platform to use this support code directly where appropriate.
Refactor only, no new functionality or tests.
* WebCore.gypi:
* platform/chromium/support/Extensions3DChromium.cpp: Renamed from Source/WebKit/chromium/src/Extensions3DChromium.cpp.
(WebCore):
(WebCore::Extensions3DChromium::Extensions3DChromium):
(WebCore::Extensions3DChromium::~Extensions3DChromium):
(WebCore::Extensions3DChromium::supports):
(WebCore::Extensions3DChromium::ensureEnabled):
(WebCore::Extensions3DChromium::isEnabled):
(WebCore::Extensions3DChromium::getGraphicsResetStatusARB):
(WebCore::Extensions3DChromium::blitFramebuffer):
(WebCore::Extensions3DChromium::renderbufferStorageMultisample):
(WebCore::Extensions3DChromium::postSubBufferCHROMIUM):
(WebCore::Extensions3DChromium::mapBufferSubDataCHROMIUM):
(WebCore::Extensions3DChromium::unmapBufferSubDataCHROMIUM):
(WebCore::Extensions3DChromium::mapTexSubImage2DCHROMIUM):
(WebCore::Extensions3DChromium::unmapTexSubImage2DCHROMIUM):
(WebCore::Extensions3DChromium::setVisibilityCHROMIUM):
(WebCore::Extensions3DChromium::discardFramebufferEXT):
(WebCore::Extensions3DChromium::ensureFramebufferCHROMIUM):
(WebCore::Extensions3DChromium::setGpuMemoryAllocationChangedCallbackCHROMIUM):
(WebCore::Extensions3DChromium::createVertexArrayOES):
(WebCore::Extensions3DChromium::deleteVertexArrayOES):
(WebCore::Extensions3DChromium::isVertexArrayOES):
(WebCore::Extensions3DChromium::bindVertexArrayOES):
(WebCore::Extensions3DChromium::getTranslatedShaderSourceANGLE):
(WebCore::Extensions3DChromium::setSwapBuffersCompleteCallbackCHROMIUM):
(WebCore::Extensions3DChromium::rateLimitOffscreenContextCHROMIUM):
(WebCore::Extensions3DChromium::paintFramebufferToCanvas):
(WebCore::Extensions3DChromium::texImageIOSurface2DCHROMIUM):
(WebCore::Extensions3DChromium::texStorage2DEXT):
(WebCore::Extensions3DChromium::createQueryEXT):
(WebCore::Extensions3DChromium::deleteQueryEXT):
(WebCore::Extensions3DChromium::isQueryEXT):
(WebCore::Extensions3DChromium::beginQueryEXT):
(WebCore::Extensions3DChromium::endQueryEXT):
(WebCore::Extensions3DChromium::getQueryivEXT):
(WebCore::Extensions3DChromium::getQueryObjectuivEXT):
* platform/chromium/support/GraphicsContext3DChromium.cpp: Renamed from Source/WebKit/chromium/src/GraphicsContext3DChromium.cpp.
(WebCore):
(WebCore::GraphicsContext3D::GraphicsContext3D):
(WebCore::GraphicsContext3D::~GraphicsContext3D):
(WebCore::GraphicsContext3D::setContextLostCallback):
(WebCore::GraphicsContext3D::setErrorMessageCallback):
(WebCore::GraphicsContext3D::create):
(WebCore::GraphicsContext3D::platformGraphicsContext3D):
(WebCore::GraphicsContext3D::platformTexture):
(WebCore::GraphicsContext3D::grContext):
(WebCore::GraphicsContext3D::platformLayer):
(WebCore::GraphicsContext3D::isGLES2Compliant):
(WebCore::GraphicsContext3D::isResourceSafe):
(WebCore::GraphicsContext3D::bindAttribLocation):
(WebCore::GraphicsContext3D::bufferData):
(WebCore::GraphicsContext3D::getActiveAttrib):
(WebCore::GraphicsContext3D::getActiveUniform):
(WebCore::GraphicsContext3D::getAttribLocation):
(WebCore::GraphicsContext3D::getContextAttributes):
(WebCore::GraphicsContext3D::getProgramInfoLog):
(WebCore::GraphicsContext3D::getShaderInfoLog):
(WebCore::GraphicsContext3D::getShaderSource):
(WebCore::GraphicsContext3D::getString):
(WebCore::GraphicsContext3D::getUniformLocation):
(WebCore::GraphicsContext3D::shaderSource):
(WebCore::GraphicsContext3D::texImage2D):
(WebCore::GraphicsContext3D::texSubImage2D):
(WebCore::GraphicsContext3D::reshape):
(WebCore::GraphicsContext3D::markContextChanged):
(WebCore::GraphicsContext3D::layerComposited):
(WebCore::GraphicsContext3D::markLayerComposited):
(WebCore::GraphicsContext3D::paintRenderingResultsToCanvas):
(WebCore::GraphicsContext3D::paintRenderingResultsToImageData):
(WebCore::GraphicsContext3D::paintCompositedResultsToCanvas):
(WebCore::GraphicsContext3D::getExtensions):
(WebCore::GraphicsContext3D::getInternalFramebufferSize):
* platform/chromium/support/GraphicsContext3DPrivate.cpp: Renamed from Source/WebKit/chromium/src/GraphicsContext3DPrivate.cpp.
(WebCore):
(WebCore::GraphicsContext3DPrivate::GraphicsContext3DPrivate):
(WebCore::GraphicsContext3DPrivate::~GraphicsContext3DPrivate):
(WebCore::GraphicsContext3DPrivate::createGraphicsContextFromWebContext):
(WebCore::GraphicsContext3DPrivate::extractWebGraphicsContext3D):
(GrMemoryAllocationChangedCallback):
(WebCore::GrMemoryAllocationChangedCallback::GrMemoryAllocationChangedCallback):
(WebCore::GrMemoryAllocationChangedCallback::onGpuMemoryAllocationChanged):
(WebCore::GraphicsContext3DPrivate::grContext):
(WebCore::GraphicsContext3DPrivate::markContextChanged):
(WebCore::GraphicsContext3DPrivate::layerComposited):
(WebCore::GraphicsContext3DPrivate::markLayerComposited):
(WebCore::GraphicsContext3DPrivate::paintFramebufferToCanvas):
(GraphicsContext3DContextLostCallbackAdapter):
(WebCore::GraphicsContext3DContextLostCallbackAdapter::GraphicsContext3DContextLostCallbackAdapter):
(WebCore::GraphicsContext3DContextLostCallbackAdapter::~GraphicsContext3DContextLostCallbackAdapter):
(WebCore::GraphicsContext3DContextLostCallbackAdapter::onContextLost):
(WebCore::GraphicsContext3DPrivate::setContextLostCallback):
(GraphicsContext3DErrorMessageCallbackAdapter):
(WebCore::GraphicsContext3DErrorMessageCallbackAdapter::GraphicsContext3DErrorMessageCallbackAdapter):
(WebCore::GraphicsContext3DErrorMessageCallbackAdapter::~GraphicsContext3DErrorMessageCallbackAdapter):
(WebCore::GraphicsContext3DErrorMessageCallbackAdapter::onErrorMessage):
(WebCore::GraphicsContext3DPrivate::setErrorMessageCallback):
(WebCore::GraphicsContext3DPrivate::getExtensions):
(WebCore::GraphicsContext3DPrivate::initializeExtensions):
(WebCore::GraphicsContext3DPrivate::supportsExtension):
(WebCore::GraphicsContext3DPrivate::ensureExtensionEnabled):
(WebCore::GraphicsContext3DPrivate::isExtensionEnabled):
(WebCore::GraphicsContext3DPrivate::isResourceSafe):
(GraphicsContext3DMemoryAllocationChangedCallbackAdapter):
(WebCore::GraphicsContext3DMemoryAllocationChangedCallbackAdapter::GraphicsContext3DMemoryAllocationChangedCallbackAdapter):
(WebCore::GraphicsContext3DMemoryAllocationChangedCallbackAdapter::~GraphicsContext3DMemoryAllocationChangedCallbackAdapter):
(WebCore::GraphicsContext3DMemoryAllocationChangedCallbackAdapter::onMemoryAllocationChanged):
(WebCore::GraphicsContext3DPrivate::setGpuMemoryAllocationChangedCallbackCHROMIUM):
(GraphicsContext3DSwapBuffersCompleteCallbackAdapter):
(WebCore::GraphicsContext3DSwapBuffersCompleteCallbackAdapter::GraphicsContext3DSwapBuffersCompleteCallbackAdapter):
(WebCore::GraphicsContext3DSwapBuffersCompleteCallbackAdapter::~GraphicsContext3DSwapBuffersCompleteCallbackAdapter):
(WebCore::GraphicsContext3DSwapBuffersCompleteCallbackAdapter::onSwapBuffersComplete):
(WebCore::GraphicsContext3DPrivate::setSwapBuffersCompleteCallbackCHROMIUM):
* platform/chromium/support/GraphicsContext3DPrivate.h: Renamed from Source/WebKit/chromium/src/GraphicsContext3DPrivate.h.
(WebKit):
(WebCore):
(GraphicsContext3DPrivate):
(WebCore::GraphicsContext3DPrivate::webContext):
(WebCore::GraphicsContext3DPrivate::preserveDrawingBuffer):
2012-06-06 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119624.
http://trac.webkit.org/changeset/119624
https://bugs.webkit.org/show_bug.cgi?id=88472
Caused worker tests to become flaky (Requested by mhahnenberg
on #webkit).
* workers/AbstractWorker.cpp:
(WebCore::AbstractWorker::~AbstractWorker):
* workers/WorkerMessagingProxy.cpp:
(WebCore::WorkerMessagingProxy::workerObjectDestroyed):
2012-06-06 Brady Eidson <beidson@apple.com>
<rdar://problem/11575112> and https://bugs.webkit.org/show_bug.cgi?id=88428
REGRESSION (r115654): Opening many non-English WebArchives shows obvious encoding issues
Reviewed by Nate Chapin.
Test: fast/loader/webarchive-encoding-respected.html
* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::commitData): Properly set the main resource encoding from the webarchive.
2012-06-06 Mark Hahnenberg <mhahnenberg@apple.com>
Worker tear-down can re-enter JSC during GC finalization
https://bugs.webkit.org/show_bug.cgi?id=88449
Reviewed by Geoffrey Garen.
No new tests. Current regression tests cover these changes.
* workers/AbstractWorker.cpp:
(WebCore::AbstractWorker::~AbstractWorker): We don't need to call onDestroyWorker() here, it
will be called elsewhere in contextDestroyed().
* workers/WorkerMessagingProxy.cpp:
(WebCore::WorkerMessagingProxy::workerObjectDestroyed): We make the call to terminateWorkerContext()
and workerContextDestroyed() asynchronous to avoid entering JSC during GC finalization.
2012-06-06 Tony Chang <tony@chromium.org>
webkit crashes on debug builds when running IETC flexbox-ordinal-group-001.htm
https://bugs.webkit.org/show_bug.cgi?id=85832
Reviewed by Ojan Vafai.
We were using a HashSet<unsigned>, which can't hold the value of 0.
According to the spec, 0 isn't a valid value, so at CSS parse time,
reject the value 0. This lets us pass the test, although that's kind
of by accident.
We could allow 0 for box-ordinal-group, but there are some perf implications
to having allowing a value below the initial value (1 in this case).
We still allow 0 for box-flex-group, because that's what the existing code
does and there are tests expecting this behavior.
No new tests, covered by ietestcenter/css3/flexbox/flexbox-ordinal-group-001.htm
which was crashing.
* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue): Reject 0 for -webkit-box-ordinal-group.
2012-06-06 Ojan Vafai <ojan@chromium.org>
Remove some magic from the -webkit-flex shorthand parsing to match the changes in the spec
https://bugs.webkit.org/show_bug.cgi?id=88444
Reviewed by Tony Chang.
-Unset positive flex defaults to 0.
-Unset negative flex defaults to 1. No magic of doing something different if positive flex was set.
* css/CSSParser.cpp:
(WebCore::CSSParser::parseFlex):
2012-06-06 Ryosuke Niwa <rniwa@webkit.org>
Revert r119440 now that I have a time to rebaseline them all!
* platform/ScrollbarThemeComposite.cpp:
(WebCore::ScrollbarThemeComposite::thumbLength):
2012-06-06 Joshua Bell <jsbell@chromium.org>
IndexedDB: Infinite recursion in IDBObjectStore.openCursor(key, legacy_constant)
https://bugs.webkit.org/show_bug.cgi?id=88431
Reviewed by Tony Chang.
Function was calling itself, rather than delegating to the correct (IDBKeyRange) overload.
Test: storage/indexeddb/legacy-constants.html
* Modules/indexeddb/IDBObjectStore.cpp:
(WebCore::IDBObjectStore::openCursor):
2012-06-06 Joshua Bell <jsbell@chromium.org>
IndexedDB: Rename IDBFactory.getDatabaseNames() to webkitGetDatabaseNames()
https://bugs.webkit.org/show_bug.cgi?id=86874
Reviewed by Ojan Vafai.
The function is non-standard, and should be prefixed.
Test: storage/indexeddb/factory-basics.html
* Modules/indexeddb/IDBFactory.idl:
2012-06-06 Julien Chaffraix <jchaffraix@webkit.org>
Add support for direction on table row group with collapsing borders
https://bugs.webkit.org/show_bug.cgi?id=87900
Reviewed by Ojan Vafai.
Tests: fast/table/border-collapsing/first-cell-left-border-hidden-table-ltr-section-rtl.html
fast/table/border-collapsing/last-cell-left-border-hidden-table-ltr-section-rtl.html
fast/table/border-collapsing/left-border-table-ltr-section-rtl.html
fast/table/border-collapsing/left-border-table-rtl-section-ltr.html
fast/table/border-collapsing/left-border-table-rtl-section-rtl.html
fast/table/border-collapsing/left-border-vertical-lr-table-ltr-section-rtl.html
fast/table/border-collapsing/left-border-vertical-lr-table-rtl-section-ltr.html
fast/table/border-collapsing/left-border-vertical-lr-table-rtl-section-rtl.html
fast/table/border-collapsing/rtl-table-left-border-hidden.html
fast/table/border-collapsing/top-border-vertical-rl-table-ltr-section-rtl.html
fast/table/border-collapsing/top-border-vertical-rl-table-rtl-section-ltr.html
fast/table/border-collapsing/top-border-vertical-rl-table-rtl-section-rtl.html
fast/table/table-ltr-section-rtl.html
fast/table/table-rtl-section-ltr.html
fast/table/table-rtl-section-rtl.html
This change enables proper support for direction on table row group.
The current code would allow people to set direction on the row group but would still
use the table's code for directionality checks (with surprising consequences).
The main change involve swapping end / start borders in case of mixed directionality
as those 2 are not in sync.
Example: <table dir="ltr"><tbody dir="rtl"><td id="cell0"></td><td id="cell1"></td></tbody></table>
Visually:
Table: Start ---------------------> End
Tbody: End <---------------------- Start
Cell: | #cell0 | #cell1 |
In this example, the end table border should be compared with the tbody's start border
and the last cell's (in DOM order) start border.
* rendering/RenderTable.cpp:
(WebCore::RenderTable::tableStartBorderAdjoiningCell):
(WebCore::RenderTable::tableEndBorderAdjoiningCell):
Same as the other adjoining functions.
* rendering/RenderTable.h:
(WebCore::RenderTable::lastColumnIndex):
Helper function to get the last column index.
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::computeCollapsedStartBorder):
(WebCore::RenderTableCell::computeCollapsedEndBorder):
Updated to call the table's adjoining border helpers.
* rendering/RenderTableCell.h:
(WebCore::RenderTableCell::styleForCellFlow):
Updated to use the table row group's style now that we properly support it.
(WebCore::RenderTableCell::isFirstOrLastCellInRow):
Debug only helper to make sure we don't call the border adjoining function
on non-terminal cells.
(WebCore::RenderTableCell::borderAdjoiningTableStart):
(WebCore::RenderTableCell::borderAdjoiningTableEnd):
* rendering/RenderTableRow.h:
(WebCore::RenderTableRow::borderAdjoiningTableStart):
(WebCore::RenderTableRow::borderAdjoiningTableEnd):
* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::firstRowCellAdjoiningTableStart):
(WebCore::RenderTableSection::firstRowCellAdjoiningTableEnd):
Updated the previous function to account for mixed directionality.
(WebCore::RenderTableSection::layoutRows):
Move some of the code to setLogicalPositionForCell to match RenderBlock.
(WebCore::RenderTableSection::setLogicalPositionForCell):
Switched direction checks to styleForCellFlow (this doesn't change anything as we
were already properly flipping already but ensure better).
* rendering/RenderTableSection.h:
(WebCore::RenderTableSection::hasSameDirectionAsTable):
Added this helper to know if we have a mixed direction.
(WebCore::RenderTableSection::borderAdjoiningTableStart):
(WebCore::RenderTableSection::borderAdjoiningTableEnd):
Updated to account for mixed directionality.
2012-06-06 Keyar Hood <keyar@chromium.org>
[Chromium] imageSmoothingEnabled should apply to patterns
https://bugs.webkit.org/show_bug.cgi?id=88103
Reviewed by Stephen White.
Test: fast/canvas/canvas-imageSmoothingEnabled-patterns.html
When drawing patterns on a Canvas, the imageSmoothingEnabled flag is
respected.
* platform/graphics/skia/PlatformContextSkia.cpp:
(WebCore::PlatformContextSkia::setupShader):
2012-06-06 Andrey Adaikin <aandrey@chromium.org>
Web Inspector: [JSC] Add WebGL instrumentation support
https://bugs.webkit.org/show_bug.cgi?id=87975
Reviewed by Vsevolod Vlasov.
* bindings/js/JSInjectedScriptManager.cpp:
(WebCore::injectAndExecuteFunction):
(WebCore::InjectedScriptManager::createInjectedScript):
(WebCore):
(WebCore::InjectedScriptManager::injectWebGLScript):
* bindings/v8/custom/V8InjectedScriptManager.cpp:
(WebCore::InjectedScriptManager::injectWebGLScript):
* inspector/InjectedScriptManager.cpp:
(WebCore::InjectedScriptManager::wrapWebGLRenderingContextForInstrumentation):
* inspector/InjectedScriptManager.h:
(InjectedScriptManager):
* inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
* inspector/InspectorWebGLAgent.cpp:
(WebCore::InspectorWebGLAgent::wrapWebGLRenderingContextForInstrumentation):
* inspector/InspectorWebGLAgent.h:
(InspectorWebGLAgent):
* inspector/InspectorWebGLInstrumentation.h:
(WebCore::InspectorInstrumentation::wrapWebGLRenderingContextForInstrumentation):
2012-06-06 Tor Arne Vestbø <tor.arne.vestbo@nokia.com>
Fix a few spelling mistakes in IconDatabase logging
Makes debugging harder if the logging is confusing.
Reviewed by Jocelyn Turcotte.
* loader/icon/IconDatabase.cpp:
(WebCore::IconDatabase::notifyPendingLoadDecisions):
(WebCore::IconDatabase::readFromDatabase):
2012-05-25 Vsevolod Vlasov <vsevik@chromium.org>
Web Inspector: [Compiler] Remove compile-front-end.sh and fix InjectedScriptSource errors.
https://bugs.webkit.org/show_bug.cgi?id=87512
Reviewed by Yury Semikhatsky.
Removed compile-front-end.sh, compile-front-end.py should be used instead now.
Made generate_protocol_externs a python module to call it from compile-front-end.py.
Temporary copy of InjectedScriptSource.js is created and modified to
avoid "This code lacks side effects" warning.
* inspector/compile-front-end.py:
* inspector/compile-front-end.sh: Removed.
* inspector/generate-protocol-externs: Removed.
* inspector/generate_protocol_externs.py: Added.
(full_qualified_type_id):
(param_type):
(generate_protocol_externs):
2012-06-06 Zoltan Herczeg <zherczeg@webkit.org>
[Qt] REGRESSION(r118616): It made all tests crash in debug mode
https://bugs.webkit.org/show_bug.cgi?id=87701
Reviewed by Csaba Osztrogonác.
Replace WriteBarrier<JSObject> to Weak<JSObject> by introducing a new
QtWeakObjectReference sub class. Rewrite all affected type definitions.
All existing test in debug mode cover this issue.
* bridge/qt/qt_class.cpp:
(JSC::Bindings::QtClass::fallbackObject):
* bridge/qt/qt_instance.cpp:
(QtRuntimeObject):
(JSC::Bindings::QtInstance::removeUnusedMethods):
* bridge/qt/qt_instance.h:
(QtInstance):
(QtWeakObjectReference):
(JSC::Bindings::QtInstance::QtWeakObjectReference::QtWeakObjectReference):
(JSC::Bindings::QtInstance::QtWeakObjectReference::operator=):
(JSC::Bindings::QtInstance::QtWeakObjectReference::get):
* bridge/qt/qt_runtime.cpp:
(JSC::Bindings::QtRuntimeMethodData::finalize):
* bridge/qt/qt_runtime_qt4.cpp:
(JSC::Bindings::QtRuntimeMethodData::finalize):
2012-06-06 Eugene Klyuchnikov <eustas.bug@gmail.com>
Web Inspector: help/settings screen does not overlay console in split screen mode
https://bugs.webkit.org/show_bug.cgi?id=88352
Reviewed by Pavel Feldman.
* inspector/front-end/HelpScreen.js:
(WebInspector.HelpScreen): Mark view as root.
(WebInspector.HelpScreen.prototype.showModal): Change view parent.
* inspector/front-end/helpScreen.css:
(.help-window-outer): Add panel and status bar indent.
(body.compact .help-window-outer): Ditto.
2012-06-06 Yoshifumi Inoue <yosin@chromium.org>
[Form] Changes for InputNumber type
https://bugs.webkit.org/show_bug.cgi?id=88394
Reviewed by Kent Tamura.
This patch changes codes for consitency of InputNumber usage for
reducing size of patch when we change InputNumber type alias from
double to Decimal.
No new tests. This patch doesn't change behavior.
* html/BaseDateAndTimeInputType.cpp:
(WebCore::BaseDateAndTimeInputType::serializeWithComponents): Fixed parameter type for getAllowedValueStep.
* html/InputType.cpp:
(WebCore::InputType::minimum): Fixed return value type to InputNumber.
(WebCore::InputType::maximum): ditto.
* html/MonthInputType.cpp:
(WebCore::MonthInputType::defaultValueForStepUp): Fix return value type to InputNumber.
(WebCore::MonthInputType::parseToNumber): ditto.
* html/NumberInputType.cpp:
(WebCore::NumberInputType::setValueAsDouble): Removed extra convertDoubleToInputNumber.
* html/StepRange.cpp:
(WebCore::StepRange::alignValueForStep): Use roundByStep for code sharing.
(WebCore::StepRange::clampValue): ditto.
(WebCore::StepRange::roundByStep): Added.
* html/StepRange.h:
(StepRange): Added a declaration of roundByStep.
2012-06-06 Michael Brüning <michael.bruning@nokia.com>
[Qt] Fix harmattan builds broken by r119247.
https://bugs.webkit.org/show_bug.cgi?id=88330
Reviewed by Jocelyn Turcotte.
Fix Harmattan build break caused by undefined constant
GL_TEXTURE_RECTANGLE_ARB by adding defined(GL_ARB_texture_rectangle)
guards around TextureMapperGL::drawTextureRectangleARB.
* platform/graphics/texmap/TextureMapperGL.cpp: Added defined(GL_ARB_texture_rectangle) guards around drawTextureRectangleARB.
(WebCore):
* platform/graphics/texmap/TextureMapperGL.h: Added definded(GL_ARB_texture_rectangle) guards around drawTextureRectangleARB.
2012-06-06 Shinya Kawanaka <shinyak@chromium.org>
PositionIterator:;setOffsetInLeafNode() is not used anywhere.
https://bugs.webkit.org/show_bug.cgi?id=88397
Reviewed by Ryosuke Niwa.
It's not used anywhere. We can remove it safely.
No new tests, no change in behavior.
* WebCore.order:
* dom/PositionIterator.cpp:
* dom/PositionIterator.h:
2012-06-06 Andrey Adaikin <aandrey@chromium.org>
Web Inspector: [WebGL] Add WebGL instrumentation support on the backend
https://bugs.webkit.org/show_bug.cgi?id=87960
Adds a WebGLAgent, injects a WebGL script upon calling the getContext
method of a HTMLCanvasElement that will wrap the WebGL rendering context
for instrumentation purposes.
Reviewed by Yury Semikhatsky.
* CMakeLists.txt:
* DerivedSources.make:
* DerivedSources.pri:
* GNUmakefile.am:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gyp/WebCore.gyp:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSHTMLCanvasElementCustom.cpp:
(WebCore::JSHTMLCanvasElement::getContext):
(WebCore::JSHTMLCanvasElement::toDataURL):
* bindings/js/JSInjectedScriptManager.cpp:
(WebCore):
(WebCore::InjectedScriptManager::injectWebGLScript):
* bindings/v8/ScriptState.cpp:
(WebCore::ScriptState::forContext):
* bindings/v8/custom/V8HTMLCanvasElementCustom.cpp:
(WebCore::V8HTMLCanvasElement::getContextCallback):
* bindings/v8/custom/V8InjectedScriptManager.cpp:
(WebCore::InjectedScriptManager::createInjectedScript):
(WebCore):
(WebCore::InjectedScriptManager::injectWebGLScript):
(WebCore::InjectedScriptManager::discardInjectedScript):
(WebCore::InjectedScriptManager::injectedScriptFor):
* inspector/CodeGeneratorInspector.py:
* inspector/InjectedScriptManager.cpp:
(WebCore):
(WebCore::InjectedScriptManager::wrapWebGLRenderingContextForInstrumentation):
(WebCore::InjectedScriptManager::injectedWebGLScriptSource):
* inspector/InjectedScriptManager.h:
(InjectedScriptManager):
* inspector/InjectedWebGLScriptSource.js: Added.
* inspector/Inspector.json:
* inspector/InspectorAllInOne.cpp:
* inspector/InspectorController.cpp:
(WebCore::InspectorController::InspectorController):
* inspector/InspectorInstrumentation.h:
(WebCore):
(InspectorInstrumentation):
(WebCore::InspectorInstrumentation::markResourceAsCached):
* inspector/InspectorWebGLAgent.cpp: Added.
(WebCore):
(WebGLAgentState):
(WebCore::InspectorWebGLAgent::InspectorWebGLAgent):
(WebCore::InspectorWebGLAgent::~InspectorWebGLAgent):
(WebCore::InspectorWebGLAgent::setFrontend):
(WebCore::InspectorWebGLAgent::clearFrontend):
(WebCore::InspectorWebGLAgent::restore):
(WebCore::InspectorWebGLAgent::enable):
(WebCore::InspectorWebGLAgent::disable):
(WebCore::InspectorWebGLAgent::wrapWebGLRenderingContextForInstrumentation):
* inspector/InspectorWebGLAgent.h: Added.
(WebCore):
(InspectorWebGLAgent):
(WebCore::InspectorWebGLAgent::create):
(WebCore::InspectorWebGLAgent::enabled):
* inspector/InspectorWebGLInstrumentation.h: Added.
(WebCore):
(WebCore::InspectorInstrumentation::wrapWebGLRenderingContextForInstrumentation):
* inspector/InstrumentingAgents.h:
(WebCore):
(WebCore::InstrumentingAgents::InstrumentingAgents):
(InstrumentingAgents):
(WebCore::InstrumentingAgents::inspectorWebGLAgent):
(WebCore::InstrumentingAgents::setInspectorWebGLAgent):
2012-06-06 Joone Hur <joone.hur@intel.com>
[GTK] Build fix for Accelerated Compositing with OpenGL
https://bugs.webkit.org/show_bug.cgi?id=88292
Reviewed by Martin Robinson.
Build fix after r119247.
Do not include GraphicsSurface.h when not building
with GRAPHICS_SURFACE.
No new tests. This is only a build fix.
* platform/graphics/texmap/TextureMapperBackingStore.cpp:
* platform/graphics/texmap/TextureMapperBackingStore.h:
* platform/graphics/texmap/TextureMapperGL.cpp:
2012-06-05 Joe Thomas <joethomas@motorola.com>
ASSERTION FAILED: ASSERT(!isPercentageIntrinsicSize) in RenderReplaced::computeIntrinsicRatioInformationForRenderBox
https://bugs.webkit.org/show_bug.cgi?id=88197
Reviewed by Daniel Bates.
RenderSVGRoot extends RenderReplaced and it overrides the computeIntrinsicRatioInformation()
method of RenderReplaced. RenderSVGRoot::computeIntrinsicRatioInformation
sets isPercentageIntrinsicSize to true while it handles width and height of percentage types.
So this assertion should be fired only if intrinsicRatio is non-zero.
Test: svg/in-html/svg-assert-failure-percentage.html
* rendering/RenderReplaced.cpp:
(WebCore::RenderReplaced::computeIntrinsicRatioInformationForRenderBox):
2012-06-05 Tony Chang <tony@chromium.org>
Changing flexbox justify/alignment doesn't update their positioning
https://bugs.webkit.org/show_bug.cgi?id=88366
Reviewed by Ojan Vafai.
Test: css3/flexbox/style-change.html
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::diff): Since these member variables were moved
out of StyleFlexibleBoxData, we have to compare them in RenderStyle::diff.
2012-06-05 Charles Wei <charles.wei@torchmobile.com.cn>
JSC implementation of SerializedScriptValue for IndexedDB
https://bugs.webkit.org/show_bug.cgi?id=88048
Reviewed by George Staikos.
No new tests. All the current indexeddb test cases apply when indexeddb works for JSC.
* bindings/js/SerializedScriptValue.cpp:
(WebCore::CloneSerializer::serializeNumber):
(CloneSerializer):
(WebCore::CloneDeserializer::toWireString):
(CloneDeserializer):
(WebCore::SerializedScriptValue::create):
(WebCore::SerializedScriptValue::toWireString):
(WebCore):
(WebCore::SerializedScriptValue::createFromWire):
(WebCore::SerializedScriptValue::numberValue):
(WebCore::SerializedScriptValue::deserialize):
2012-06-05 Ojan Vafai <ojan@chromium.org>
Nested CSS flexbox renders incorrectly
https://bugs.webkit.org/show_bug.cgi?id=87679
Reviewed by Julien Chaffraix.
If the containing block with a perpendicular writing mode has an overrideHeight,
then its logical height should be based off that instead of its RenderStyle.
Test: css3/flexbox/perpendicular-writing-modes-inside-flex-item.html
* rendering/RenderBox.cpp:
(WebCore::RenderBox::perpendicularContainingBlockLogicalHeight):
2012-06-05 Charles Wei <charles.wei@torchmobile.com.cn>
Need special-case toJS(IDBAny) for IDBCursorWithValue
https://bugs.webkit.org/show_bug.cgi?id=87963
Reviewed by George Staikos.
No new tests, this is for indexedDB, which is not working for JSC binding.
When indexedDB working for JSC, all the indexdb test cases should apply.
* bindings/js/JSIDBAnyCustom.cpp:
(WebCore::toJS):
2012-06-05 Michal Mocny <mmocny@google.com>
[chromium] Set default memory allocation limit bytes when GL_CHROMIUM_gpu_memory_manager is not supported by graphics context.
https://bugs.webkit.org/show_bug.cgi?id=87996
Reviewed by James Robinson.
Soon we hope to have renderers not manage there own memory allocation limits, and rely exclusively on the
allocation given by GpuMemoryManager. However, when the graphics context does not support the
GL_CHROMIUM_gpu_memory_manager extension, such as the one used by DumpRenderTree, we must default the
memory allocation limit to some non-0 value.
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::initialize):
* platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
(WebCore::CCSingleThreadProxy::postSetContentsMemoryAllocationLimitBytesToMainThreadOnImplThread):
* platform/graphics/chromium/cc/CCSingleThreadProxy.h:
(WebCore):
(DebugScopedSetMainThreadFromEither):
(WebCore::DebugScopedSetMainThreadFromEither::DebugScopedSetMainThreadFromEither):
(WebCore::DebugScopedSetMainThreadFromEither::~DebugScopedSetMainThreadFromEither):
2012-06-05 Roland Takacs <takacs.roland@stud.u-szeged.hu>
[Qt] Use GraphicsContext3DOpenGLES.cpp when using OpenGL ES
https://bugs.webkit.org/show_bug.cgi?id=78720
Reviewed by Noam Rosenthal.
Functions are removed from GraphicsContext3DQt that already implemented
in GraphicsContext3DCommon. Macros that not supported at GLES have been
changed to their appropriate pair that found in GraphicsContext3D
(e.g. GL_FRAMEBUFFER_EXT to GraphicsContext3D::FRAMEBUFFER).
It was needed because OpenGLES does not support these macros.
Extensions3DQt is removed beceause it has never been used.
Some functions have been moved fron GraphicsContext3DCommon to *OpenGL
and *OpenGLES to provide specialization.
The "m_depthStencilBuffer" member is added to Qt port because it uses
"m_depthStencilBuffer" like the other platforms.
* Target.pri:
* platform/graphics/GraphicsContext3D.h:
* platform/graphics/opengl/GraphicsContext3DOpenGL.cpp:
(WebCore::GraphicsContext3D::depthRange):
(WebCore):
(WebCore::GraphicsContext3D::clearDepth):
* platform/graphics/opengl/GraphicsContext3DOpenGLCommon.cpp:
(WebCore::GraphicsContext3D::prepareTexture):
(WebCore::GraphicsContext3D::readRenderingResults):
(WebCore::GraphicsContext3D::reshape):
(WebCore::GraphicsContext3D::copyTexImage2D):
(WebCore::GraphicsContext3D::copyTexSubImage2D):
(WebCore::GraphicsContext3D::readPixels):
* platform/graphics/opengl/GraphicsContext3DOpenGLES.cpp:
(WebCore::GraphicsContext3D::reshapeFBOs):
(WebCore::GraphicsContext3D::depthRange):
(WebCore):
(WebCore::GraphicsContext3D::clearDepth):
* platform/graphics/qt/GraphicsContext3DQt.cpp:
(WebCore::GraphicsContext3D::isGLES2Compliant):
(WebCore::GraphicsContext3DPrivate::blitMultisampleFramebuffer):
(WebCore::GraphicsContext3D::GraphicsContext3D):
(WebCore::GraphicsContext3D::~GraphicsContext3D):
(WebCore):
2012-06-05 Vitaly Buka <vitalybuka@chromium.org>
Special layout handler should be done on top frame being printed.
https://bugs.webkit.org/show_bug.cgi?id=88201
Reviewed by Brady Eidson.
No new tests. Root case is already covered by tests.
Case described in the issue can be reproduced only by direct call
to Frame::setPrinting of subframe. Probably it's not possible with
layout tests.
* page/Frame.cpp:
(WebCore::Frame::setPrinting):
Use shouldUsePrintingLayout to choose proper version of forceLayout().
(WebCore::Frame::shouldUsePrintingLayout):
Checks if current frame is the top frame being printed.
* rendering/RenderView.cpp:
(WebCore::RenderView::shouldUsePrintingLayout): Forward to Frame.
2012-06-05 Max Feil <mfeil@rim.com>
[BlackBerry] Conditionally enlarge HTML5 video controls in fullscreen mode
https://bugs.webkit.org/show_bug.cgi?id=87551
Reviewed by Antonio Gomes.
Changes to make media controls larger in fullscreen mode. The
BlackBerry media controls used to be sized by a combination
of CSS styles specified in mediaControlsBlackBerry.css and
hard coded constants in RenderThemeBlackBerry.cpp. This patch
moves all the dimensional sizes to a common location, i.e.
RenderThemeBlackBerry.cpp. Having all the logic in C++ code allows
more complex decision making for when and how much to enlarge the
media controls. Currently the controls are enlarged by a factor
of 2 if the document is in fullscreen mode, the video element
is the current fullscreen element, and the deviceScaleFactor of
the page is less than 2.0. This avoids overly large controls
on pages which are already at least doubling the size of the
controls via viewport meta tag device-width settings. In other
words, if the CSS pixels are already enlarged compared to device
pixels by a factor of at least 2 in length and width.
In order to accomplish this, I had to add a new Element* parameter
to RenderTheme::adjustSliderThumbSize(). This is needed so
that fullscreen mode can be checked and the deviceScaleFactor
retrieved. I feel that this does not have a big impact on the
code and the other ports, for which I will simply be adding the
additional parameter in their platform code.
Tests:
video-controls-enlarged-fullscreen-meta.html
video-controls-enlarged-fullscreen-nometa.html
* css/mediaControlsBlackBerry.css:
(audio::-webkit-media-controls-panel, video::-webkit-media-controls-panel):
(audio::-webkit-media-controls-play-button, video::-webkit-media-controls-play-button):
(audio::-webkit-media-controls-timeline-container, video::-webkit-media-controls-timeline-container):
(audio::-webkit-media-controls-current-time-display, video::-webkit-media-controls-current-time-display):
(audio::-webkit-media-controls-time-remaining-display, video::-webkit-media-controls-time-remaining-display):
(audio::-webkit-media-controls-timeline, video::-webkit-media-controls-timeline):
(audio::-webkit-media-controls-fullscreen-button, video::-webkit-media-controls-fullscreen-button):
(audio::-webkit-media-controls-mute-button, video::-webkit-media-controls-mute-button):
(audio::-webkit-media-controls-volume-slider-container, video::-webkit-media-controls-volume-slider-container):
(audio::-webkit-media-controls-volume-slider, video::-webkit-media-controls-volume-slider):
* html/shadow/SliderThumbElement.cpp:
(WebCore::RenderSliderThumb::updateAppearance):
* platform/blackberry/RenderThemeBlackBerry.cpp:
(WebCore):
(WebCore::determineRenderSlider):
(WebCore::determineFullScreenMultiplier):
(WebCore::RenderThemeBlackBerry::adjustSliderThumbSize):
(WebCore::RenderThemeBlackBerry::adjustMediaControlStyle):
(WebCore::RenderThemeBlackBerry::adjustSliderTrackStyle):
(WebCore::RenderThemeBlackBerry::paintMediaSliderTrack):
(WebCore::RenderThemeBlackBerry::paintMediaSliderThumb):
* platform/blackberry/RenderThemeBlackBerry.h:
(RenderThemeBlackBerry):
* platform/efl/RenderThemeEfl.cpp:
(WebCore::RenderThemeEfl::adjustSliderThumbSize):
* platform/efl/RenderThemeEfl.h:
(RenderThemeEfl):
* platform/gtk/RenderThemeGtk.h:
* platform/gtk/RenderThemeGtk2.cpp:
(WebCore::RenderThemeGtk::adjustSliderThumbSize):
* platform/gtk/RenderThemeGtk3.cpp:
(WebCore::RenderThemeGtk::adjustSliderThumbSize):
* platform/qt/RenderThemeQt.cpp:
(WebCore::RenderThemeQt::adjustSliderThumbSize):
* platform/qt/RenderThemeQt.h:
(RenderThemeQt):
* platform/qt/RenderThemeQtMobile.cpp:
(WebCore::RenderThemeQtMobile::adjustSliderThumbSize):
* platform/qt/RenderThemeQtMobile.h:
(RenderThemeQtMobile):
* rendering/RenderThemeChromiumLinux.cpp:
(WebCore::RenderThemeChromiumLinux::adjustSliderThumbSize):
* rendering/RenderThemeChromiumLinux.h:
(RenderThemeChromiumLinux):
* rendering/RenderThemeChromiumSkia.cpp:
(WebCore::RenderThemeChromiumSkia::adjustSliderThumbSize):
* rendering/RenderThemeChromiumSkia.h:
(RenderThemeChromiumSkia):
* rendering/RenderThemeChromiumWin.cpp:
(WebCore):
(WebCore::RenderThemeChromiumWin::adjustSliderThumbSize):
* rendering/RenderThemeChromiumWin.h:
(RenderThemeChromiumWin):
* rendering/RenderThemeMac.h:
(RenderThemeMac):
* rendering/RenderThemeSafari.cpp:
(WebCore::RenderThemeSafari::adjustSliderThumbSize):
* rendering/RenderThemeSafari.h:
(RenderThemeSafari):
* rendering/RenderThemeWin.cpp:
(WebCore::RenderThemeWin::adjustSliderThumbSize):
* rendering/RenderThemeWin.h:
(RenderThemeWin):
* rendering/RenderThemeWinCE.cpp:
(WebCore::RenderThemeWinCE::adjustSliderThumbSize):
* rendering/RenderThemeWinCE.h:
(RenderThemeWinCE):
* rendering/RenderTheme.cpp:
(WebCore::RenderTheme::adjustStyle):
(WebCore::RenderTheme::adjustMediaControlStyle):
(WebCore):
(WebCore::RenderTheme::adjustSliderThumbStyle):
(WebCore::RenderTheme::adjustSliderThumbSize):
* rendering/RenderTheme.h:
(RenderTheme):
2012-06-05 Stephanie Lewis <slewis@apple.com>
https://bugs.webkit.org/show_bug.cgi?id=88370
Memory sampler should trigger low memory signal
Reviewed by Geoff Garen.
No new tests. Verify by running stress test which crashes
in a few minutes without the fix.
Fix assumption in block code. We could get in a state where timer_event_source
had already been released before the block ran.
* platform/mac/MemoryPressureHandlerMac.mm:
(WebCore::MemoryPressureHandler::holdOff):
2012-06-05 Yoshifumi Inoue <yosin@chromium.org>
[Forms] Introduce InputNumber type as an alias of double for replacing it to Decimal
https://bugs.webkit.org/show_bug.cgi?id=88275
Reviewed by Kent Tamura.
This patch introduced InputNumber type as alias of double to be
replaced to Decimal and convertDoubleToInputNumber and
convertToInputNumberToDouble for reducing size of patch when we
introduce Decimal type.
This patch added new vairables to use const variable to reduce
copy operation when introducing Decimal type.
In the implementation, we use InputNumber, will be Decimal, however some methods in IDL
use double. To clarify internal usage of value and external usage, thit patch renamed
following functions:
- parseToDouble => parseNumber
- setValueAsNumber => setValueAsDouble and setValueAsInputNumber
- valueAsNumber => valueAsDouble
This patch is a part of introducing Decimal, https://bugs.webkit.org/show_bug.cgi?id=80009
No new tests. This patch doesn't change behavior.
* html/BaseDateAndTimeInputType.cpp:
(WebCore::BaseDateAndTimeInputType::valueAsDate): Replaced parseToNumber() to valueAsDouble()
(WebCore::BaseDateAndTimeInputType::setValueAsDate): Inserted convertDoubleToInputNumber.
(WebCore::BaseDateAndTimeInputType::valueAsDouble): Use new function parseToDouble().
(WebCore::BaseDateAndTimeInputType::setValueAsInputNumber): Added.
(WebCore::BaseDateAndTimeInputType::defaultValueForStepUp): Changed return type to InputNumber.
(WebCore::BaseDateAndTimeInputType::parseToNumber): This is replacement of parseToDouble().
(WebCore::BaseDateAndTimeInputType::parseToDouble): Added for valueAsDate and valueAsDouble.
(WebCore::BaseDateAndTimeInputType::serialize): Changed parameter type to InputNumber.
(WebCore::BaseDateAndTimeInputType::serializeWithComponents): Replace double to InputNumber for getAllowdStep.
* html/BaseDateAndTimeInputType.h:
(BaseDateAndTimeInputType):
* html/DateInputType.cpp:
(WebCore::DateInputType::createStepRange): Replaced double to InputNumber.
* html/DateInputType.h:
(DateInputType):
* html/DateTimeInputType.cpp:
(WebCore::DateTimeInputType::defaultValueForStepUp): Changed return type to InputNumber.
(WebCore::DateTimeInputType::createStepRange): Replaced double to InputNumber.
* html/DateTimeInputType.h:
(DateTimeInputType):
* html/DateTimeLocalInputType.cpp:
(WebCore::DateTimeLocalInputType::createStepRange): Replaced double to InputNumber.
* html/DateTimeLocalInputType.h:
(DateTimeLocalInputType):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::getAllowedValueStep): Changed parameter type to InputNumber.
(WebCore::HTMLInputElement::valueAsNumber): Use InputType::valueAsDouble.
(WebCore::HTMLInputElement::setValueAsNumber): Use InputType::valueAsDouble.
* html/HTMLInputElement.h:
(HTMLInputElement):
* html/InputType.cpp:
(WebCore::InputType::valueAsDouble): Replacement of valueAsNumber.
(WebCore::InputType::setValueAsDouble): Replacement of valueAsNumber.
(WebCore::InputType::setValueAsInputNumber): Replacement of valueAsNumber.
(WebCore::InputType::rangeUnderflow): Replaced double to InputNumber.
(WebCore::InputType::rangeOverflow): Replaced double to InputNumber.
(WebCore::InputType::defaultValueForStepUp): Changed return type to InputNumber.
(WebCore::InputType::isInRange): Replaced double to InputNumber.
(WebCore::InputType::isOutOfRange): Replaced double to InputNumber.
(WebCore::InputType::stepMismatch): Replaced double to InputNumber.
(WebCore::InputType::validationMessage): Replaced double to InputNumber.
(WebCore::InputType::parseToNumber): Replacement of parseToDouble.
(WebCore::InputType::parseToNumberOrNaN): Added for reducing number of std::numeric_limits<double>::quiet_NaN().
(WebCore::InputType::parseToNumberWithDecimalPlaces): Replaced double to InputNumber.
(WebCore::InputType::serialize): Changed parameter type to InputNumber.
(WebCore::InputType::applyStep): Replaced double to InputNumber.
(WebCore::InputType::getAllowedValueStep): Changed parameter type to InputNumber.
(WebCore::InputType::stepUpFromRenderer): Replaced double to InputNumber.
* html/InputType.h:
(InputType):
* html/MonthInputType.cpp:
(WebCore::MonthInputType::defaultValueForStepUp): Changed return type to InputNumber.
(WebCore::MonthInputType::createStepRange): Replaced double to InputNumber.
(WebCore::MonthInputType::parseToNumber): Replacement of parseToDouble.
* html/MonthInputType.h:
(MonthInputType):
* html/NumberInputType.cpp:
(WebCore::NumberInputType::valueAsDouble): Replacement of valueAsNumber. Changed to use parseToDoubleForNumberType.
(WebCore::NumberInputType::setValueAsDouble): Replacement of setValueAsNumber. Changed to use serializeForNumberType.
(WebCore::NumberInputType::setValueAsInputNumber): Replacement of setValueAsNumber.
(WebCore::NumberInputType::createStepRange): Replaced double to InputNumber.
(WebCore::NumberInputType::parseToNumber): Replacement of parseToDouble.
(WebCore::NumberInputType::parseToNumberWithDecimalPlaces): Replaced double to InputNumber.
(WebCore::NumberInputType::serialize):
* html/NumberInputType.h:
(NumberInputType):
* html/RangeInputType.cpp:
(WebCore::ensureMaximum): Replaced double to InputNumber.
(WebCore::RangeInputType::valueAsDouble): Replacement of valueAsNumber.
(WebCore::RangeInputType::setValueAsInputNumber): Replacement of setValueAsNumber.
(WebCore::RangeInputType::createStepRange): Replaced double to InputNumber.
(WebCore::RangeInputType::handleKeydownEvent): Replaced double to InputNumber.
(WebCore::RangeInputType::parseToNumber): Replacement of parseToDouble.
(WebCore::RangeInputType::serialize): Changed parameter type to InputNumber.
(WebCore::RangeInputType::sanitizeValue): Replaced double to InputNumber.
* html/RangeInputType.h:
(RangeInputType):
* html/StepRange.cpp:
(WebCore::StepRange::StepRange): Replaced double to InputNumber.
(WebCore::StepRange::acceptableError): Replaced double to InputNumber.
(WebCore::StepRange::alignValueForStep): Replaced double to InputNumber.
(WebCore):
(WebCore::StepRange::clampValue): Replaced double to InputNumber.
(WebCore::StepRange::stepMismatch): Replaced double to InputNumber.
* html/StepRange.h:
(WebCore::convertDoubleToInputNumber): Added.
(WebCore::convertInputNumberToDouble): Added.
(WebCore::StepRange::NumberWithDecimalPlaces::NumberWithDecimalPlaces): Replaced double to InputNumber.
(WebCore::StepRange::StepDescription::defaultValue): Replaced double to InputNumber.
(StepRange):
(WebCore::StepRange::maximum): Replaced double to InputNumber.
(WebCore::StepRange::minimum): Replaced double to InputNumber.
(WebCore::StepRange::step): Replaced double to InputNumber.
(WebCore::StepRange::stepBase): Replaced double to InputNumber.
(WebCore::StepRange::defaultValue): Replaced double to InputNumber.
(WebCore::StepRange::proportionFromValue): Replaced double to InputNumber.
(WebCore::StepRange::valueFromProportion): Replaced double to InputNumber.
* html/TimeInputType.cpp:
(WebCore::TimeInputType::defaultValueForStepUp): Replaced double to InputNumber.
(WebCore::TimeInputType::createStepRange): Replaced double to InputNumber.
* html/TimeInputType.h:
(TimeInputType):
* html/WeekInputType.cpp:
(WebCore::WeekInputType::createStepRange): Replaced double to InputNumber.
* html/WeekInputType.h:
(WeekInputType):
* html/shadow/CalendarPickerElement.cpp:
(WebCore::CalendarPickerElement::writeDocument): Replaced double to InputNumber.
* html/shadow/SliderThumbElement.cpp:
(WebCore::sliderPosition): Replaced double to InputNumber.
(WebCore::RenderSliderThumb::layout): Replaced double to InputNumber.
(WebCore::SliderThumbElement::setPositionFromPoint): Replaced double to InputNumber.
2012-06-05 Ryosuke Niwa <rniwa@webkit.org>
Build fix after r119527.
* dom/ViewportArguments.cpp:
(WebCore::computeViewportAttributes):
2012-06-05 Martin Robinson <mrobinson@igalia.com>
[Cairo] [OpenGL] Enable GL_OES_standard_derivatives
https://bugs.webkit.org/show_bug.cgi?id=87583
Reviewed by Alejandro G. Castro.
No new tests. This is covered by oes-standard-derivatives.html
in the Khronos WebGL conformance tests, which may one day be part
of the WebKit test suite.
Activate the GL_OES_standard_derivatives on Cairo along with
Qt and Mac.
* platform/graphics/opengl/Extensions3DOpenGL.cpp:
(WebCore::Extensions3DOpenGL::ensureEnabled):
(WebCore::Extensions3DOpenGL::isEnabled):
2012-06-05 Simon Fraser <simon.fraser@apple.com>
Avoid creating compositing layers for preserve-3d without transformed descendants
https://bugs.webkit.org/show_bug.cgi?id=88115
Reviewed by Antti Koivisto.
Avoid creating compositing layers, and therefore using excess backing store,
for elements that have -webkit-transform-style: preserve-3d, but no 3D-transformed
descendants that would be affected by that preserve-3d.
Test: compositing/layer-creation/no-compositing-for-preserve-3d.html
* rendering/RenderLayer.h: Replace the "mustOverlap" flag with a enum
that describes the different reasons for indirect compositing, so that
we can use that information to decide whether to allocate backing store.
(WebCore::RenderLayer::setIndirectCompositingReason):
(WebCore::RenderLayer::indirectCompositingReason):
(WebCore::RenderLayer::mustCompositeForIndirectReasons):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::RenderLayer): Initialize m_indirectCompositingReason
* rendering/RenderLayerCompositor.h: New out param for computeCompositingRequirements()
that is uses to indicate that a 3d-transformed descendant has been encountered.
Rename requiresCompositingWhenDescendantsAreCompositing() to requiresCompositingForIndirectReason(),
and return the reason as an out param.
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::updateCompositingLayers): Pass in the saw3DTransform
param to computeCompositingRequirements().
(WebCore::RenderLayerCompositor::computeCompositingRequirements): Return a flag
from computeCompositingRequirements() that is set to true if we've seen descendants
that have 3d transforms. This is later used to decide whether to composite for
perspective or preserve-3d.
Change the "mustOverlapCompositedLayers" code to use the new "indirect compositing"
enum flags.
After enumerating children, call requiresCompositingForIndirectReason() and
record the reason in the layer.
(WebCore::RenderLayerCompositor::needsToBeComposited): Use mustCompositeForIndirectReasons() now.
(WebCore::RenderLayerCompositor::requiresOwnBackingStore): Consult the indirect compositing
reason rather than just looking for the overlap flag.
(WebCore::RenderLayerCompositor::reasonForCompositing): Now that we have more information
about indirect compositing reasons, the logging can be more detailed.
(WebCore::RenderLayerCompositor::requiresCompositingForTransform): This now only looks
for 3d transforms. We now treat perspective and perserve-3d as "indirect" reasons, because
whether they composite depends on descendants having non-affine transforms.
(WebCore::RenderLayerCompositor::requiresCompositingForIndirectReason): Includes the logic
previously in requiresCompositingWhenDescendantsAreCompositing(), and now determines
whether to composite for preserve-3d and perspective, based on whether we have transformed descendants.
2012-06-05 Levi Weintraub <leviw@chromium.org>
Block selection gaps painted not properly pixel snapped
https://bugs.webkit.org/show_bug.cgi?id=88000
Reviewed by Ryosuke Niwa.
Fixes for selection painting in both inline line heights and block gaps. We were unintentionally
flooring inline and block offsets in RenderBlock, but actually need to intentionally floor the
logical left and right values in logical(Left|Right)SelectionGap to properly align our block
selection gaps. This is to properly mirror the behavior in pixelSnappedLogicalRightOffsetForLine.
We were also unintentionally upcasting LayoutUnits to floats in InlineTextBox's selection painting
routine. Now we're properly rounding.
Test: fast/sub-pixel/selection/selection-gaps-at-fractional-offsets.html
* rendering/InlineTextBox.cpp:
(WebCore::InlineTextBox::paintSelection): Move paint values to integers and do proper rounding.
* rendering/RenderBlock.cpp:
(WebCore::blockDirectionOffset): Fix incorrect flooring in return value.
(WebCore::inlineDirectionOffset): Ditto.
(WebCore): Removing previous attempt to properly mirror inline box selection pixel alignment. This
implementation rounded the final x value instead of properly rounding our block offset then flooring
the offset that corresponds to the behavior in InlineTextBox. It also didn't properly pixel snap
the y offset and height.
(WebCore::RenderBlock::blockSelectionGap): Properly pixel snapping the gap rect before painting.
(WebCore::RenderBlock::logicalLeftSelectionGap): Correctly match InlineTextBox's selection pixel
alignment by flooring the logicalLeft and pixel snapping the subsequent rect. We do this in order
to maintain proper render tree pixel snapping while simultaneously mirroring the flooring in the
inline direction that occurs in InlineTextBox.cpp's alignSelectionRectToDevicePixels method.
(WebCore::RenderBlock::logicalRightSelectionGap): Ditto with the right.
2012-06-05 Adam Barth <abarth@webkit.org>
Remove support for target-densitydpi in the viewport meta tag
https://bugs.webkit.org/show_bug.cgi?id=88047
Reviewed by Kenneth Rohde Christiansen.
The target-densitydpi parameter was originally implemented on Android,
but it does not appear to be widely used and introduces significant
implementation complexity because it forces us to track three scale
factors (defaultDeviceScaleFactor, deviceScaleFactor, and
pageScaleFactor) rather than just two (deviceScaleFactor and
pageScaleFactor).
There don't appear to be many web sites that use target-densitydpi
because it is not supported on iOS. There are also concerns from the
standards community about the design of the feature. It seems our best
course of action is to remove target-densitydpi and address these use
cases via other mechanisms, such as responsive images and device units
in CSS because those approaches are likely to be implemented broadly.
* dom/ViewportArguments.cpp:
(WebCore::computeViewportAttributes):
(WebCore):
(WebCore::setViewportFeature):
(WebCore::viewportErrorMessageTemplate):
(WebCore::viewportErrorMessageLevel):
* dom/ViewportArguments.h:
(WebCore::ViewportArguments::ViewportArguments):
(ViewportArguments):
(WebCore::ViewportArguments::operator==):
2012-06-05 Erik Arvidsson <arv@chromium.org>
[V8] Improve variable resolution order on window
https://bugs.webkit.org/show_bug.cgi?id=84247
Reviewed by Ojan Vafai.
This changes the V8 flag to turn on es52_globals and updates the layout tests to reflect the fixed behavior.
Test: fast/dom/Window/window-property-shadowing-onclick.html
* bindings/v8/V8DOMWindowShell.cpp:
(WebCore::V8DOMWindowShell::initContextIfNeeded): Set the flag as we initialize V8.
* bindings/v8/WorkerContextExecutionProxy.cpp:
(WebCore::WorkerContextExecutionProxy::initIsolate): Ditto.
2012-06-05 Ojan Vafai <ojan@chromium.org>
Delete dead code from RenderBox::computeLogicalWidthInRegion
https://bugs.webkit.org/show_bug.cgi?id=88353
Reviewed by Tony Chang.
No new tests obviously since this is dead code.
* rendering/RenderBox.cpp:
(WebCore::RenderBox::computeLogicalWidthInRegion):
2012-06-05 Greg Billock <gbillock@google.com>
New constructor for WebIntent to be used for delivery
https://bugs.webkit.org/show_bug.cgi?id=87143
Reviewed by Darin Fisher.
* Modules/intents/Intent.cpp:
(WebCore::Intent::setExtras):
(WebCore):
* Modules/intents/Intent.h:
(Intent):
2012-06-01 Ojan Vafai <ojan@chromium.org>
Change overrideSizes to be content-box instead of border-box
https://bugs.webkit.org/show_bug.cgi?id=88116
Reviewed by Tony Chang.
Previously, the code used border-box everywhere except for one place
where it used content-box when setting the override size. Conversely,
it subtracted the borderAndPaddingSize everywhere it grabbed the override
size except for one.
Instead, make the override size override the content-box size.
Converging on using the border-box size would be tricky given RenderTableCell's
tricky management of intrinsic padding.
There is no behavior change except in RenderBox::availableLogicalHeightUsing,
which fixes a table test.
* rendering/RenderBox.cpp:
(WebCore::RenderBox::setOverrideLogicalContentHeight):
(WebCore::RenderBox::setOverrideLogicalContentWidth):
(WebCore::RenderBox::overrideLogicalContentWidth):
(WebCore::RenderBox::overrideLogicalContentHeight):
(WebCore::RenderBox::computeLogicalWidthInRegion):
(WebCore::RenderBox::computeLogicalHeight):
(WebCore::RenderBox::computePercentageLogicalHeight):
(WebCore::RenderBox::availableLogicalHeightUsing):
We were incorrectly subtracting the borderAndPaddingLogicalWidth instead
of the borderAndPaddingLogicalHeight.
* rendering/RenderBox.h:
(RenderBox):
* rendering/RenderDeprecatedFlexibleBox.cpp:
(WebCore::RenderDeprecatedFlexibleBox::layoutHorizontalBox):
(WebCore::RenderDeprecatedFlexibleBox::layoutVerticalBox):
(WebCore::RenderDeprecatedFlexibleBox::applyLineClamp):
(WebCore::RenderDeprecatedFlexibleBox::allowedChildFlex):
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::computeAvailableFreeSpace):
(WebCore::RenderFlexibleBox::setLogicalOverrideSize):
(WebCore::RenderFlexibleBox::applyStretchAlignmentToChild):
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::setOverrideHeightFromRowHeight):
2012-06-05 Andy Estes <aestes@apple.com>
REGRESSION (r118197): PluginStrategy should be used even if ENABLE(NETSCAPE_PLUGIN_API) is false
https://bugs.webkit.org/show_bug.cgi?id=88296
Reviewed by Dan Bernstein.
refresh() and initPlugins() should call into PluginStrategy even if
ENABLE(NETSCAPE_PLUGIN_API) is false. Some ports support non-Netscape
plug-ins, so PluginStrategy should be consulted even if Netscape
plug-ins specifically are disabled.
Ports that support platform strategies but not Netscape plug-ins should
handle this inside their PluginStrategy rather than modifying generic
plug-in code.
* plugins/PluginData.cpp:
(WebCore::PluginData::refresh):
(WebCore::PluginData::initPlugins):
2012-06-05 Douglas Stockwell <dstockwell@chromium.org>
Crash when modifying fixed-position elements in a detached frame
https://bugs.webkit.org/show_bug.cgi?id=88288
Reviewed by James Robinson.
Test: fast/block/positioning/fixed-position-detached-frame.html
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
2012-06-05 Andrey Kosyakov <caseq@chromium.org>
Web Inspector: render timeline paint category dark green
https://bugs.webkit.org/show_bug.cgi?id=88333
Reviewed by Pavel Feldman.
- Added a new, dark-green color for timeline events.
* WebCore.gypi: Added timelineBarDarkGreen.png
* inspector/front-end/Images/timelineBarDarkGreen.png: Added.
* inspector/front-end/Images/timelineCheckmarks.png: Added a dark-green checkbox.
* inspector/front-end/Images/timelineDots.png: Added a dark-green dot.
* inspector/front-end/TimelinePresentationModel.js:
(WebInspector.TimelinePresentationModel.categories):
* inspector/front-end/WebKit.qrc: Added timelineBarDarkGreen.png.
* inspector/front-end/timelinePanel.css:
(.timeline-category-statusbar-item.timeline-category-painting .timeline-category-checkbox):
(.timeline-category-painting .timeline-graph-bar):
(.popover .timeline-painting):
(.timeline-category-painting .timeline-tree-icon):
2012-06-05 Alexei Filippov <alexeif@chromium.org>
Web Inspector: serialize edge counts instead of indexes in heap snapshot
https://bugs.webkit.org/show_bug.cgi?id=88324
The serialized node structure currently holds an index
of its first containment edge in the edges array.
The index can be quite big (up to 7 digits for large snapshots).
The patch changes the serialization format to pass
node containment edge count instead. For most nodes the count
is just a single digit number.
This reduces serialized snapshot size and therefore its transfer time.
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshotNode.prototype._edgeIndexesStart):
(WebInspector.HeapSnapshotNode.prototype._edgeIndexesEnd):
(WebInspector.HeapSnapshotNode.prototype._ordinal):
(WebInspector.HeapSnapshotNodeIterator):
(WebInspector.HeapSnapshot.prototype._init):
(WebInspector.HeapSnapshot.prototype._buildEdgeIndexes):
(WebInspector.HeapSnapshot.prototype._buildRetainers):
(WebInspector.HeapSnapshot.prototype._bfs):
(WebInspector.HeapSnapshot.prototype._buildAggregates):
(WebInspector.HeapSnapshot.prototype._buildPostOrderIndex):
(WebInspector.HeapSnapshot.prototype._buildDominatorTree):
(WebInspector.HeapSnapshot.prototype._markPageOwnedNodes):
(WebInspector.HeapSnapshot.prototype._markQueriableHeapObjects):
2012-06-05 Arpita Bahuguna <arpitabahuguna@gmail.com>
CSS 2.1 failure: border-conflict-element-021a
https://bugs.webkit.org/show_bug.cgi?id=86885
Reviewed by Julien Chaffraix.
When two adjacent table row groups (thead, tbody, tfoot) have the same border-width
and the same border-style in a 'border-collapse: collapse' table the color of the border
from the top-most table row wins.
Tests: fast/table/border-collapsing/adjacent-row-groups-multi.html
fast/table/border-collapsing/adjacent-row-groups.html
* rendering/RenderTable.cpp:
(WebCore::RenderTable::paintObject):
Currently each row group is sent for paint from top (firstChild) to bottom thereby causing
the borders of the bottom-most row group (which is painted last) to overlap over the
previous row-group. This breaks the precedence for collapsed borders between adjacent cells.
This fix thus reverses the order in which the row-groups/sections are passed for painting.
Additional change has been done to iterate over the RenderTableSections directly.
(WebCore):
(WebCore::RenderTable::bottomSection):
* rendering/RenderTable.h:
New function has been added to obtain the last section of the table.
(RenderTable):
* rendering/RenderTableSection.h:
(RenderTableSection):
RenderTableSection's paint() method has now been made public so as to make it accessible
from RenderTable.
2012-06-05 Csaba Osztrogonác <ossy@webkit.org>
[Qt][Win] Fix IDL dependency generation
https://bugs.webkit.org/show_bug.cgi?id=88304
Reviewed by Tor Arne Vestbø.
* DerivedSources.pri:
2012-06-05 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119467, r119471, and r119476.
http://trac.webkit.org/changeset/119467
http://trac.webkit.org/changeset/119471
http://trac.webkit.org/changeset/119476
https://bugs.webkit.org/show_bug.cgi?id=88332
it broke chromium-mac static initializers test.\ (Requested by
loislo on #webkit).
* WebCore.gyp/WebCore.gyp:
* WebCore.gypi:
* platform/graphics/FontPlatformData.h:
(FontPlatformData):
* platform/graphics/cocoa/FontPlatformDataCocoa.mm:
(WebCore::FontPlatformData::platformDataInit):
(WebCore::FontPlatformData::platformDataAssign):
* platform/graphics/harfbuzz/ng/HarfBuzzFace.cpp: Removed.
* platform/graphics/harfbuzz/ng/HarfBuzzFace.h: Removed.
* platform/graphics/harfbuzz/ng/HarfBuzzFaceCoreText.cpp: Removed.
* platform/graphics/harfbuzz/ng/HarfBuzzShaper.cpp: Removed.
* platform/graphics/harfbuzz/ng/HarfBuzzShaper.h: Removed.
* platform/graphics/mac/FontComplexTextMac.cpp:
(WebCore::Font::selectionRectForComplexText):
(WebCore::Font::drawComplexText):
(WebCore::Font::floatWidthForComplexText):
(WebCore::Font::offsetForPositionForComplexText):
2012-06-05 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: SHIFT key as helper to color adjustment in color picker
https://bugs.webkit.org/show_bug.cgi?id=88319
Reviewed by Pavel Feldman.
With Shift depressed, the dragger in the color picker will snap to the horizontal
or vertical axis while dragging, depending on which axis offset is greater.
* inspector/front-end/Spectrum.js:
(WebInspector.Spectrum.colorDragStart):
(WebInspector.Spectrum.colorDrag):
(WebInspector.Spectrum.draggable.move):
(WebInspector.Spectrum.draggable.start):
2012-06-05 Andrey Kosyakov <caseq@chromium.org>
Web Inspector: aggregate all events before first frame into a fake frame
https://bugs.webkit.org/show_bug.cgi?id=88229
- in Timeline's frame mode, start aggregating events by frame even before we get first frame marker.
* inspector/front-end/TimelineFrameController.js:
(WebInspector.TimelineFrameController.prototype._addRecord):
(WebInspector.TimelineFrameController.prototype._flushFrame):
(WebInspector.TimelineFrameController.prototype._createFrame):
2012-06-05 Charles Wei <charles.wei@torchmobile.com.cn>
JSC:need to implement Dictionary::getWithUndefinedOrNullCheck for IDB
https://bugs.webkit.org/show_bug.cgi?id=88283
Reviewed by Kentaro Hara.
No new tests, idb not working for JSC yet.
* bindings/js/Dictionary.cpp:
(WebCore::Dictionary::getWithUndefinedOrNullCheck):
* bindings/js/Dictionary.h:
(Dictionary):
* bindings/js/JSDictionary.cpp:
(WebCore::JSDictionary::getWithUndefinedOrNullCheck):
(WebCore):
* bindings/js/JSDictionary.h:
(JSDictionary):
2012-06-05 Dongwoo Im <dw.im@samsung.com>
Add 'isProtocolHandlerRegistered' and 'unregisterProtocolHandler'.
https://bugs.webkit.org/show_bug.cgi?id=73176
Reviewed by Adam Barth.
Two more APIs are added in Custom Scheme Handler specification.
http://dev.w3.org/html5/spec/Overview.html#custom-handlers
One is 'isProtocolHandlerRegistered' to query whether the specific URL
is registered or not.
The other is 'unregisterProtocolHandler' to remove the registered URL.
Test: fast/dom/unregister-protocol-handler.html
* Configurations/FeatureDefines.xcconfig: Add a macro 'ENABLE_CUSTOM_SCHEME_HANDLER'.
* loader/EmptyClients.h: Add prototypes.
(EmptyChromeClient):
(WebCore::EmptyChromeClient::isProtocolHandlerRegistered):
(WebCore::EmptyChromeClient::unregisterProtocolHandler):
* page/ChromeClient.h:
(ChromeClient):
* page/NavigatorRegisterProtocolHandler.cpp:
(WebCore):
(WebCore::customHandlersStateString):
(WebCore::NavigatorRegisterProtocolHandler::isProtocolHandlerRegistered): Query if the handler is registered.
(WebCore::NavigatorRegisterProtocolHandler::unregisterProtocolHandler): Unregister the registered handler.
* page/NavigatorRegisterProtocolHandler.h:
(NavigatorRegisterProtocolHandler):
* page/NavigatorRegisterProtocolHandler.idl:
2012-06-04 Kinuko Yasuda <kinuko@chromium.org>
Chromium tests: "Add support for the Blob constructor" [r115582] regressed blob layout tests
https://bugs.webkit.org/show_bug.cgi?id=85174
Reviewed by Kentaro Hara.
This fixes a crash problem which could happen when the constructor is
given an array which contains String-type item(s).
There're still some Text mismatches between v8 results and
JSC results, most of them are v8 not throwing exception
when it is given an object whose toString() method throws
exception. (The issue will be addresse in a separate patch.)
Test: fast/files/blob-constructor.html
* bindings/v8/custom/V8BlobCustom.cpp:
(WebCore::V8Blob::constructorCallback):
2012-06-05 Kentaro Hara <haraken@chromium.org>
Remove unnecessary constructor of SelectorQueryCache
https://bugs.webkit.org/show_bug.cgi?id=87942
Reviewed by Ryosuke Niwa.
Per the comment from darin@ (https://bugs.webkit.org/show_bug.cgi?id=87942#c27),
this patch removes an inline constructor that has no arguments.
To have the compiler generate the constructor,
this patch also removes WTF_MAKE_NONCOPYABLE().
No tests. No change in behavior.
* dom/SelectorQuery.h:
(SelectorQueryCache):
2012-06-05 Kentaro Hara <haraken@chromium.org>
Remove unused methods from SelectorQuery.h
https://bugs.webkit.org/show_bug.cgi?id=88289
Reviewed by Antti Koivisto.
This patch removes SelectorDataList::size() and
SelectorDataList::SelectorDataList().
SelectorDataList::size() is not used by anybody.
SelectorDataList::SelectorDataList() can be auto-generated
by a compiler.
No tests. No change in behavior.
* dom/SelectorQuery.cpp:
* dom/SelectorQuery.h:
(SelectorDataList):
2012-06-05 Kent Tamura <tkent@chromium.org>
Build fix for r119470.
* loader/EmptyClients.cpp: Add #include "IntentRequest.h".
2012-06-05 Kent Tamura <tkent@chromium.org>
Remove unused function: Document::formElements
https://bugs.webkit.org/show_bug.cgi?id=88272
Reviewed by Kentaro Hara.
No new tests. No behavior change.
* dom/Document.h:
(Document): remove formElements(). It was used for radio group
validation, but the current code doesn't use it any more.
2012-06-05 Kenichi Ishibashi <bashi@chromium.org>
Unreviewed build fix on Chromium mac
* platform/graphics/cocoa/FontPlatformDataCocoa.mm:
(WebCore::isAATFont):
2012-06-05 Kent Tamura <tkent@chromium.org>
Move some function definitions in EmptyClients.h to EmptyClients.cpp
https://bugs.webkit.org/show_bug.cgi?id=88285
Reviewed by Ryosuke Niwa.
If we have definitions of functions with RefPtr<> in EmptyClients.h like
virtual void runOpenPanel(Frame*, PassRefPtr<FileChooser>) { }
and a source file includes EmptyClient.h, the source file needs to
include FileChooser.h though the source file doesn't use FileChooser
class explicitly because the definition of runOpenPanel() needs to call
FileChooser::deref().
To avoid this, we had better move such functions to EmptyClients.cpp.
No new tests. No behavior changes.
* loader/EmptyClients.cpp: Move such functions from EmptyClients.h to
here, and move EmptyPopupMenu and EmptySearchPopupMenu too because
they're used only by EmptyChromeClient.
(WebCore):
(EmptyPopupMenu):
(WebCore::EmptyPopupMenu::show):
(WebCore::EmptyPopupMenu::hide):
(WebCore::EmptyPopupMenu::updateFromElement):
(WebCore::EmptyPopupMenu::disconnectClient):
(EmptySearchPopupMenu):
(WebCore::EmptySearchPopupMenu::popupMenu):
(WebCore::EmptySearchPopupMenu::saveRecentSearches):
(WebCore::EmptySearchPopupMenu::loadRecentSearches):
(WebCore::EmptySearchPopupMenu::enabled):
(WebCore::EmptyChromeClient::createPopupMenu):
(WebCore::EmptyChromeClient::createSearchPopupMenu):
(WebCore::EmptyChromeClient::createColorChooser):
(WebCore::EmptyChromeClient::runOpenPanel):
(WebCore::EmptyFrameLoaderClient::dispatchDecidePolicyForNewWindowAction):
(WebCore::EmptyFrameLoaderClient::dispatchDecidePolicyForNavigationAction):
(WebCore::EmptyFrameLoaderClient::dispatchWillSendSubmitEvent):
(WebCore::EmptyFrameLoaderClient::dispatchWillSubmitForm):
(WebCore::EmptyFrameLoaderClient::createDocumentLoader):
(WebCore::EmptyFrameLoaderClient::createFrame):
(WebCore::EmptyFrameLoaderClient::createPlugin):
(WebCore::EmptyFrameLoaderClient::createJavaAppletWidget):
(WebCore::EmptyFrameLoaderClient::createMediaPlayerProxyPlugin):
(WebCore::EmptyFrameLoaderClient::createNetworkingContext):
(WebCore::EmptyFrameLoaderClient::dispatchIntent):
(WebCore::EmptyTextCheckerClient::requestCheckingOfString):
(WebCore::EmptyEditorClient::registerUndoStep):
(WebCore::EmptyEditorClient::registerRedoStep):
(WebCore::EmptyContextMenuClient::customizeMenu):
* loader/EmptyClients.h: Move such function definitions to
EmptyClient.cpp, and append OVERRIDE.
(EmptyChromeClient):
(EmptyFrameLoaderClient):
(EmptyTextCheckerClient):
(EmptyEditorClient):
(EmptyContextMenuClient):
* svg/graphics/SVGImage.cpp: Remove unnecessary #includes.
2012-06-05 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: "-webkit"-prefixed properties are considered inherited in the Computed Style pane
https://bugs.webkit.org/show_bug.cgi?id=88257
Reviewed by Pavel Feldman.
Canonical names for properties should be used when checking if they are inherited (since only canonical names
are put into the usedProperties set).
* inspector/front-end/StylesSidebarPane.js:
(WebInspector.ComputedStylePropertiesSection.prototype._isPropertyInherited):
2012-06-05 Kenichi Ishibashi <bashi@chromium.org>
[Chromium] Implement font shaping with font-feature-settings on Mac
https://bugs.webkit.org/show_bug.cgi?id=69826
Reviewed by Tony Chang.
Add HarfBuzz-ng text shaper.
Chromium mac port uses it as secondary text shaper to support OpenType features.
HarfBuzz-ng is only used when -webkit-font-feature-settings is specified and
corresponding font is not an AAT font.
No new tests. css3/font-feature-settings-rendering.html should pass on Chromium mac port.
* WebCore.gyp/WebCore.gyp: Added harfbuzz-ng dependencies.
* WebCore.gypi: Added harfbuzz-ng files.
* platform/graphics/FontPlatformData.h:
(FontPlatformData): Added m_harfbuzzFace.
* platform/graphics/cocoa/FontPlatformDataCocoa.mm:
(WebCore::FontPlatformData::platformDataInit): Copy m_harfbuzzFace.
(WebCore::FontPlatformData::platformDataAssign): Ditto.
(WebCore):
(WebCore::isAATFont):
(WebCore::FontPlatformData::harfbuzzFace):
* platform/graphics/harfbuzz/ng/HarfBuzzFace.cpp: Added.
(WebCore):
(WebCore::harfbuzzFaceCache):
(WebCore::HarfBuzzFace::HarfBuzzFace):
(WebCore::HarfBuzzFace::~HarfBuzzFace):
* platform/graphics/harfbuzz/ng/HarfBuzzFace.h: Added.
(WebCore):
(HarfBuzzFace):
(WebCore::HarfBuzzFace::create):
* platform/graphics/harfbuzz/ng/HarfBuzzFaceCoreText.cpp: Added.
(WebCore):
(WebCore::floatToHarfBuzzPosition):
(WebCore::getGlyph):
(WebCore::getGlyphHorizontalAdvance):
(WebCore::getGlyphHorizontalOrigin):
(WebCore::getGlyphExtents):
(WebCore::harfbuzzCoreTextGetFontFuncs):
(WebCore::releaseTableData):
(WebCore::harfbuzzCoreTextGetTable):
(WebCore::HarfBuzzFace::createFace):
(WebCore::HarfBuzzFace::createFont):
(WebCore::HarfBuzzShaper::createGlyphBufferAdvance):
* platform/graphics/harfbuzz/ng/HarfBuzzShaper.cpp: Added.
(WebCore):
(WebCore::harfbuzzPositionToFloat):
(WebCore::HarfBuzzShaper::HarfBuzzRun::HarfBuzzRun):
(WebCore::HarfBuzzShaper::HarfBuzzRun::setGlyphAndPositions):
(WebCore::HarfBuzzShaper::HarfBuzzRun::characterIndexForXPosition):
(WebCore::HarfBuzzShaper::HarfBuzzRun::xPositionForOffset):
(WebCore::HarfBuzzShaper::HarfBuzzShaper):
(WebCore::HarfBuzzShaper::~HarfBuzzShaper):
(WebCore::HarfBuzzShaper::setFontFeatures):
(WebCore::HarfBuzzShaper::shape):
(WebCore::HarfBuzzShaper::setupHarfBuzzRun):
(WebCore::HarfBuzzShaper::shapeHarfBuzzRun):
(WebCore::HarfBuzzShaper::setGlyphPositionsForHarfBuzzRun):
(WebCore::HarfBuzzShaper::offsetForPosition):
(WebCore::HarfBuzzShaper::selectionRect):
* platform/graphics/harfbuzz/ng/HarfBuzzShaper.h: Added.
(WebCore):
(HarfBuzzShaper):
(WebCore::HarfBuzzShaper::totalWidth):
(HarfBuzzRun):
(WebCore::HarfBuzzShaper::HarfBuzzRun::create):
(WebCore::HarfBuzzShaper::HarfBuzzRun::setWidth):
(WebCore::HarfBuzzShaper::HarfBuzzRun::numCharacters):
(WebCore::HarfBuzzShaper::HarfBuzzRun::numGlyphs):
(WebCore::HarfBuzzShaper::HarfBuzzRun::width):
(WebCore::HarfBuzzShaper::HarfBuzzRun::rtl):
* platform/graphics/mac/FontComplexTextMac.cpp:
(WebCore):
(WebCore::preferHarfBuzz): Added.
(WebCore::Font::selectionRectForComplexText): Use HarfBuzzShaper if font-feature-settings exists.
(WebCore::Font::drawComplexText): Ditto.
(WebCore::Font::floatWidthForComplexText): Ditto.
(WebCore::Font::offsetForPositionForComplexText): Ditto.
2012-06-05 Li Yin <li.yin@intel.com>
IETC: FileList.item(-1) should return null instead of raising
https://bugs.webkit.org/show_bug.cgi?id=77899
Reviewed by Kentaro Hara.
IETC: http://samples.msdn.microsoft.com/ietestcenter/fileapi/filelist.htm
Accroding to http://es5.github.com/#x9.6
When the index is negative, it should not raise exception.
Firefox and IE returns null when there is not indexth File object.
Test: fast/files/file-list-test.html
* fileapi/FileList.idl:
2012-06-05 Adam Barth <abarth@webkit.org>
EventHandler shouldn't dispatch fake mousemove events when scrolling on devices that don't have a mouse
https://bugs.webkit.org/show_bug.cgi?id=88270
Reviewed by James Robinson.
This patch adds a setting analogous to deviceSupportsTouch to determine
whether the device supports mouse. We then don't dispatch fake mouse
events on devices that don't have a mouse in the first place.
Test: fast/events/touch/scroll-without-mouse-lacks-mousemove-events.html
* page/EventHandler.cpp:
(WebCore::EventHandler::fakeMouseMoveEventTimerFired):
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(WebCore::Settings::setDeviceSupportsMouse):
(WebCore::Settings::deviceSupportsMouse):
(Settings):
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::InternalSettings):
(WebCore::InternalSettings::restoreTo):
(WebCore::InternalSettings::setDeviceSupportsMouse):
(WebCore):
* testing/InternalSettings.h:
(InternalSettings):
* testing/InternalSettings.idl:
2012-06-04 Simon Fraser <simon.fraser@apple.com>
Leaking ClipRects
https://bugs.webkit.org/show_bug.cgi?id=88282
Reviewed by Dan Bernstein.
In r118562 I made the ClipRectsCache use RefPtr<ClipRects>. However, ClipRects
was initialized with m_refCnt=0, not 1 as adoptRef() and friends expect. Also,
there was a manual ref() in RenderLayer::updateClipRects() which this patch removes.
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::updateClipRects):
* rendering/RenderLayer.h:
(WebCore::ClipRects::ClipRects):
2012-06-04 Emil A Eklund <eae@chromium.org> and Levi Weintraub <leviw@chromium.org>
Add missing FractionalLayoutUnit += operator and move LineWidth to use all floats
https://bugs.webkit.org/show_bug.cgi?id=88259
Reviewed by Ryosuke Niwa.
Fix two rounding bugs in LineLayout and RenderBlock that caused inlines
and floats to wrap incorrectly.
Tests: fast/sub-pixel/float-containing-block-with-margin.html
fast/sub-pixel/float-with-right-margin-zoom.html
fast/sub-pixel/inline-block-with-padding.html
* platform/FractionalLayoutUnit.h:
(WebCore::operator+=):
Add missing float version of += operator thus avoiding a silent cast to
int loosing precision.
* rendering/RenderBlockLineLayout.cpp:
(WebCore::LineWidth::computeAvailableWidthFromLeftAndRight):
(WebCore::LineWidth::updateAvailableWidth):
Change LineWidth left/right to floating point and instead of rounding the
left and right edge separately.
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::resolveFlexibleLengths):
Use explicit cast to avoid compiler ambiguity warning.
2012-06-04 Tony Chang <tony@chromium.org>
CSS property 'order' should be a number rather than an int
https://bugs.webkit.org/show_bug.cgi?id=88111
Reviewed by Ojan Vafai.
This changed in the spec:
http://dev.w3.org/csswg/css3-flexbox/#order
I also fixed the compile with CSS3_FLEXBOX disabled.
No new tests, covered by existing tests.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore):
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
* css/CSSProperty.cpp: Parse as float, remove clamping for HashMap<int>.
(WebCore::CSSProperty::isInheritedProperty):
* css/StyleBuilder.cpp:
(WebCore::StyleBuilder::StyleBuilder):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* rendering/RenderFlexibleBox.cpp: Remove traits for HashMap<int>.
(RenderFlexibleBox::OrderIterator):
* rendering/RenderFlexibleBox.h:
* rendering/style/RenderStyle.h:
* rendering/style/StyleRareNonInheritedData.h:
(StyleRareNonInheritedData):
2012-06-04 Hugo Parente Lima <hugo.lima@openbossa.org>
Turn LEGACY_VIEWPORT_ADAPTION USE flag into an ENABLE flag.
https://bugs.webkit.org/show_bug.cgi?id=88243
Reviewed by Adam Barth.
No new tests needed, this just rename a compiler flag.
* dom/Document.cpp:
(WebCore::Document::setDocType):
* dom/ViewportArguments.h:
* html/HTMLMetaElement.cpp:
(WebCore::HTMLMetaElement::process):
2012-06-04 Dana Jansens <danakj@chromium.org>
[chromium] Remove redundant setNeedsCommit when prepareToDraw fails
https://bugs.webkit.org/show_bug.cgi?id=88246
Reviewed by James Robinson.
When prepareToDraw fails it explicitly calls setNeedsCommit, but the
scheduler already takes care of this for it. When a frame is not
successfully drawn, the scheduler state machine sets the needsCommit
bit itself.
This behaviour is covered by CCSchedulerStateMachineTest.TestFailedDrawSetsNeedsCommitAndDoesNotDrawAgain
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::prepareToDraw):
2012-06-04 Takashi Toyoshima <toyoshim@chromium.org>
[WebSocket] Send requires super linear time against data size
https://bugs.webkit.org/show_bug.cgi?id=87383
Reviewed by Kent Tamura.
WebSocket send operation requires super linear time to send data
against data size. This is because removing sent data in Vector<char>
always requires data copy.
Introduced wtf/StreamBuffer realizes O(n) buffer operations.
This patch is only for optimization. No new tests are needed.
* platform/network/SocketStreamHandleBase.cpp:
(WebCore::SocketStreamHandleBase::sendPendingData):
* platform/network/SocketStreamHandleBase.h:
(SocketStreamHandleBase):
2012-06-04 Tim Horton <timothy_horton@apple.com>
Mock scrollbars differ by 1px in DRT vs. WKTR
https://bugs.webkit.org/show_bug.cgi?id=80879
Rubber-stamped by Ryosuke Niwa.
Disable r119431 except for Apple Mac port.
* platform/ScrollbarThemeComposite.cpp:
(WebCore::ScrollbarThemeComposite::thumbLength):
2012-06-04 Jeffrey Pfau <jpfau@apple.com>
Document cleanup can get confused if parser still exists
https://bugs.webkit.org/show_bug.cgi?id=88250
Reviewed by Geoffrey Garen.
No new tests; no behavior changes.
* dom/Document.cpp:
(WebCore::Document::removedLastRef): Detach parser earlier
2012-06-04 Andrew Lo <anlo@rim.com>
[BlackBerry] Split AnimationFrameRateController into its own file
https://bugs.webkit.org/show_bug.cgi?id=88242
Reviewed by Antonio Gomes.
Include new header file to reflect re-organization of platform animation
header files.
* platform/graphics/DisplayRefreshMonitor.h:
2012-06-04 Dana Jansens <danakj@chromium.org>
[chromium] Allow CCLayerImpl to find its layer tree host, and use this for CCVideoLayerImpl instead of always-null pointer.
https://bugs.webkit.org/show_bug.cgi?id=88252
Reviewed by James Robinson.
CCVideoLayerImpl has a pointer to its CCLayerTreeHostImpl which is never
set to be non-null currently. Instead of making this a special case, we
move the CCLayerTreeHostImpl* up to the CCLayerImpl base class, and set
the pointer on every layer during commit via the TreeSynchronizer's walk
through the layer tree.
Unit tests: Tested by TreeSynchronizer tests.
* platform/graphics/chromium/TreeSynchronizer.cpp:
(WebCore::TreeSynchronizer::synchronizeTrees):
(WebCore::TreeSynchronizer::synchronizeTreeRecursive):
* platform/graphics/chromium/TreeSynchronizer.h:
(WebCore):
(TreeSynchronizer):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(WebCore):
(WebCore::CCLayerImpl::layerTreeHostImpl):
(WebCore::CCLayerImpl::setLayerTreeHostImpl):
(CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::finishCommitOnImplThread):
* platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
(WebCore::CCVideoLayerImpl::CCVideoLayerImpl):
(WebCore::CCVideoLayerImpl::setNeedsRedraw):
* platform/graphics/chromium/cc/CCVideoLayerImpl.h:
2012-06-04 Tim Horton <timothy_horton@apple.com>
Mock scrollbars differ by 1px in DRT vs. WKTR
https://bugs.webkit.org/show_bug.cgi?id=80879
Reviewed by Simon Fraser.
Round the length of the scrollbar thumb instead of truncating it.
No new tests, fixes a mismatch between pixel results from DRT and WKTR.
Change is covered by all existing tests with scrollbars run with low tolerance.
* platform/ScrollbarThemeComposite.cpp:
(WebCore::ScrollbarThemeComposite::thumbLength):
2012-06-04 Anders Carlsson <andersca@apple.com>
Fix an unused result warning when building with clang.
* platform/cf/BinaryPropertyList.cpp:
(WebCore::BinaryPropertyListWriter::writePropertyList):
2012-06-04 Ami Fischman <fischman@chromium.org>
Don't crash when re-parenting a removed HTMLMediaElement
https://bugs.webkit.org/show_bug.cgi?id=88058
Reviewed by Eric Carlson.
Test: media/media-reparent.html
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::userCancelledLoad): Clear the pending load flags because any pending load has been cancelled.
2012-06-04 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119421 and r119425.
http://trac.webkit.org/changeset/119421
http://trac.webkit.org/changeset/119425
https://bugs.webkit.org/show_bug.cgi?id=88256
not reliable, let's have a do-over (Requested by kling on
#webkit).
* dom/Element.cpp:
(WebCore::Element::detachAttribute):
(WebCore::Element::removeAttribute):
(WebCore::Element::attributes):
(WebCore::Element::setAttributeInternal):
(WebCore::Element::parserSetAttributes):
(WebCore::Element::hasEquivalentAttributes):
(WebCore::Element::createAttributeData):
(WebCore):
(WebCore::Element::setAttributeNode):
(WebCore::Element::removeAttributeNode):
(WebCore::Element::getAttributeNode):
(WebCore::Element::getAttributeNodeNS):
(WebCore::Element::hasAttribute):
(WebCore::Element::hasAttributeNS):
(WebCore::Element::normalizeAttributes):
(WebCore::Element::cloneAttributesFromElement):
* dom/Element.h:
(WebCore::Element::attributeData):
(Element):
(WebCore::Element::ensureAttributeData):
(WebCore::Element::updatedAttributeData):
(WebCore::Element::ensureUpdatedAttributeData):
* dom/ElementAttributeData.cpp:
(WebCore::ElementAttributeData::attrIfExists):
(WebCore::ElementAttributeData::ensureAttr):
(WebCore::ElementAttributeData::setAttr):
(WebCore::ElementAttributeData::removeAttr):
(WebCore::ElementAttributeData::setClass):
(WebCore):
(WebCore::ElementAttributeData::ensureInlineStyle):
(WebCore::ElementAttributeData::ensureMutableInlineStyle):
(WebCore::ElementAttributeData::destroyInlineStyle):
(WebCore::ElementAttributeData::addAttribute):
(WebCore::ElementAttributeData::removeAttribute):
(WebCore::ElementAttributeData::isEquivalent):
(WebCore::ElementAttributeData::detachAttrObjectsFromElement):
(WebCore::ElementAttributeData::getAttributeItemIndexSlowCase):
(WebCore::ElementAttributeData::cloneDataFrom):
(WebCore::ElementAttributeData::clearAttributes):
(WebCore::ElementAttributeData::replaceAttribute):
(WebCore::ElementAttributeData::getAttributeNode):
* dom/ElementAttributeData.h:
(WebCore::ElementAttributeData::create):
(ElementAttributeData):
(WebCore::ElementAttributeData::setIdForStyleResolution):
(WebCore::ElementAttributeData::setAttributeStyle):
(WebCore::ElementAttributeData::length):
(WebCore::ElementAttributeData::isEmpty):
(WebCore::ElementAttributeData::attributeItem):
(WebCore::ElementAttributeData::getAttributeItem):
(WebCore::ElementAttributeData::ElementAttributeData):
(WebCore::ElementAttributeData::attributeVector):
(WebCore::ElementAttributeData::clonedAttributeVector):
(WebCore::ElementAttributeData::getAttributeItemIndex):
* dom/StyledElement.cpp:
(WebCore::StyledElement::style):
(WebCore::StyledElement::classAttributeChanged):
(WebCore::StyledElement::removeInlineStyleProperty):
(WebCore::StyledElement::addSubresourceAttributeURLs):
* html/parser/HTMLConstructionSite.cpp:
(WebCore::HTMLConstructionSite::createHTMLElementFromSavedElement):
* xml/parser/XMLDocumentParserQt.cpp:
(WebCore::XMLDocumentParser::XMLDocumentParser):
2012-06-04 Andreas Kling <kling@webkit.org>
Unreviewed build fixes for cr-android and Qt.
* dom/ElementAttributeData.h:
(WebCore::ElementAttributeData::setClass):
Change ElementAttributeData::setClass() to take an AtomicString instead of a String.
The only call site has an AtomicString anyway.
* xml/parser/XMLDocumentParserQt.cpp:
(WebCore::XMLDocumentParser::XMLDocumentParser): Add missing const.
2012-06-04 Sadrul Habib Chowdhury <sadrul@chromium.org>
[chromium] Fix software rendering for device-scale-factor > 1
https://bugs.webkit.org/show_bug.cgi?id=88136
Reviewed by Darin Fisher.
Test: fast/hidpi/device-scale-factor-paint.html
* WebCore.exp.in:
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::setDeviceScaleFactor):
(WebCore):
* testing/InternalSettings.h:
(InternalSettings):
* testing/InternalSettings.idl:
2012-06-04 Andreas Kling <kling@webkit.org>
Make ElementAttributeData a variable-sized object to reduce memory use.
<http://webkit.org/b/88240>
Reviewed by Antti Koivisto.
Take advantage of the fact that we know at ElementAttributeData construction time how many attributes
it needs to accomodate and allocate exactly as much space as needed instead of using a Vector.
For elements that never have their attribute list mutated (the vast majority), this saves a lot of
memory and removes the indirection to Vector<Attribute>'s heap-allocated storage.
Introduced a mutability flag to ElementAttributeData and sprinkled assertions all over to make sure
that nothing tries to mutate an element with a raw attribute array.
When an Element's attribute(s) are mutated, we reconstruct the ElementAttributeData, this time using
a Vector as backing instead. This is triggered when calling Element::mutableAttributeData().
This reduces memory consumption by 3.2MB when viewing the full HTML5 spec at <http://whatwg.org/c/>.
This is a 35% reduction in DOM attribute memory use.
Furthermore, that page ends up promoting 27% of the elements to mutable attribute storage due to dynamic
adding of "class" attributes. For more static pages, savings are even greater.
* dom/Element.cpp:
(WebCore::Element::detachAttribute):
(WebCore::Element::removeAttribute):
(WebCore::Element::attributes):
(WebCore::Element::setAttributeInternal):
(WebCore::Element::parserSetAttributes):
(WebCore::Element::hasEquivalentAttributes):
(WebCore::Element::setAttributeNode):
(WebCore::Element::removeAttributeNode):
(WebCore::Element::getAttributeNode):
(WebCore::Element::getAttributeNodeNS):
(WebCore::Element::hasAttribute):
(WebCore::Element::hasAttributeNS):
(WebCore::Element::normalizeAttributes):
(WebCore::Element::cloneAttributesFromElement):
(WebCore::Element::createMutableAttributeData):
* dom/Element.h:
(WebCore::Element::attributeData):
(Element):
(WebCore::Element::updatedAttributeData):
(WebCore::Element::ensureAttributeData):
(WebCore::Element::ensureUpdatedAttributeData):
(WebCore::Element::mutableAttributeData):
* dom/ElementAttributeData.cpp:
(WebCore::ElementAttributeData::createImmutable):
(WebCore::ElementAttributeData::ElementAttributeData):
(WebCore::ElementAttributeData::~ElementAttributeData):
(WebCore::ElementAttributeData::attrIfExists):
(WebCore::ElementAttributeData::ensureAttr):
(WebCore::ElementAttributeData::setAttr):
(WebCore::ElementAttributeData::removeAttr):
(WebCore::ElementAttributeData::ensureInlineStyle):
(WebCore::ElementAttributeData::ensureMutableInlineStyle):
(WebCore::ElementAttributeData::destroyInlineStyle):
(WebCore::ElementAttributeData::addAttribute):
(WebCore::ElementAttributeData::removeAttribute):
(WebCore::ElementAttributeData::isEquivalent):
(WebCore::ElementAttributeData::detachAttrObjectsFromElement):
(WebCore::ElementAttributeData::getAttributeItemIndexSlowCase):
(WebCore::ElementAttributeData::cloneDataFrom):
(WebCore::ElementAttributeData::clearAttributes):
(WebCore::ElementAttributeData::replaceAttribute):
(WebCore::ElementAttributeData::getAttributeNode):
* dom/ElementAttributeData.h:
(ElementAttributeData):
(WebCore::ElementAttributeData::setClass):
(WebCore::ElementAttributeData::setIdForStyleResolution):
(WebCore::ElementAttributeData::inlineStyle):
(WebCore::ElementAttributeData::setAttributeStyle):
(WebCore::ElementAttributeData::isMutable):
(WebCore::ElementAttributeData::makeMutable):
(WebCore::ElementAttributeData::length):
(WebCore::ElementAttributeData::isEmpty):
(WebCore::ElementAttributeData::array):
(WebCore::ElementAttributeData::attributeItem):
(WebCore::ElementAttributeData::getAttributeItem):
(WebCore::ElementAttributeData::getAttributeItemIndex):
* dom/StyledElement.cpp:
(WebCore::StyledElement::style):
(WebCore::StyledElement::classAttributeChanged):
(WebCore::StyledElement::removeInlineStyleProperty):
(WebCore::StyledElement::addSubresourceAttributeURLs):
* html/parser/HTMLConstructionSite.cpp:
(WebCore::HTMLConstructionSite::createHTMLElementFromSavedElement):
* xml/parser/XMLDocumentParserQt.cpp:
(WebCore::XMLDocumentParser::XMLDocumentParser):
2012-06-04 Sadrul Habib Chowdhury <sadrul@chromium.org>
[chromium] Combobox options and autofill options should not be scaled for device-scale factor
https://bugs.webkit.org/show_bug.cgi?id=87921
Reviewed by Darin Fisher.
Chrome takes care of scaling the combobox/autofill lists in the browser side. So it is not
necessary to scale here. This change reverts parts of changes in r111539.
Manually tested with different scale factors.
* platform/chromium/PopupListBox.cpp:
(WebCore::PopupListBox::paint):
(WebCore::PopupListBox::paintRow):
(WebCore::PopupListBox::getRowHeight):
* platform/chromium/PopupListBox.h:
(PopupContainerSettings):
* platform/chromium/PopupMenuChromium.cpp:
(WebCore::PopupMenuChromium::show):
* rendering/RenderMenuList.cpp:
(WebCore::RenderMenuList::showPopup):
2012-06-04 Dana Jansens <danakj@chromium.org>
[chromium] Make LayerRendererChromium use RenderPasses instead of RenderSurfaces
https://bugs.webkit.org/show_bug.cgi?id=88132
Reviewed by Adrienne Walker.
This patch is partly renaming, partly moving data into CCRenderPass
and CCRenderPassDrawQuad. LayerRendererChromium should accept as input
a set of RenderPasses and this should be sufficient for drawing. It
should have no notion of RenderSurfaces at all.
First, what isn't done: RenderSurfaces still hold the textures
where RenderPasses are drawn to/from, so that these textures can
persist across frames. This will be addressed separately.
Otherwise, this completely removes the concept of RenderSurface from
LayerRenderChromium and the CCDrawQuad, replacing it with RenderPass.
The "renderSurfaceTextureManager" is renamed to the "implTextureManager"
and like-wise for the TextureAllocator, as these are not used
exclusively for render surfaces (passes), and a texture manager for
render passes that does not require a RenderSurface* will be the focus
of future changes.
Covered by existing tests.
* WebCore.gypi:
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::LayerRendererChromium):
(WebCore::LayerRendererChromium::setVisible):
(WebCore::LayerRendererChromium::releaseRenderPassTextures):
(WebCore::LayerRendererChromium::viewportChanged):
(WebCore::LayerRendererChromium::clearRenderPass):
(WebCore::LayerRendererChromium::beginDrawingFrame):
(WebCore::LayerRendererChromium::drawRenderPass):
(WebCore::LayerRendererChromium::drawQuad):
(WebCore::LayerRendererChromium::drawBackgroundFilters):
(WebCore::LayerRendererChromium::drawRenderPassQuad):
(WebCore::LayerRendererChromium::copyPlaneToTexture):
(WebCore::LayerRendererChromium::drawHeadsUpDisplay):
(WebCore::LayerRendererChromium::finishDrawingFrame):
(WebCore::LayerRendererChromium::copyTextureToFramebuffer):
(WebCore::LayerRendererChromium::getFramebufferTexture):
(WebCore::LayerRendererChromium::isCurrentRenderPass):
(WebCore::LayerRendererChromium::useRenderPass):
(WebCore::LayerRendererChromium::useManagedTexture):
(WebCore::LayerRendererChromium::bindFramebufferToTexture):
(WebCore::LayerRendererChromium::setScissorToRect):
(WebCore::LayerRendererChromium::setDrawFramebufferRect):
(WebCore::LayerRendererChromium::initializeSharedObjects):
(WebCore::LayerRendererChromium::renderPassProgram):
(WebCore::LayerRendererChromium::renderPassProgramAA):
(WebCore::LayerRendererChromium::renderPassMaskProgram):
(WebCore::LayerRendererChromium::renderPassMaskProgramAA):
(WebCore::LayerRendererChromium::cleanupSharedObjects):
* platform/graphics/chromium/LayerRendererChromium.h:
(WebCore):
(LayerRendererChromium):
* platform/graphics/chromium/cc/CCDrawQuad.cpp:
(WebCore::CCDrawQuad::toRenderPassDrawQuad):
* platform/graphics/chromium/cc/CCDrawQuad.h:
(WebCore):
(CCDrawQuad):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
(WebCore::CCHeadsUpDisplay::draw):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
(WebCore::CCLayerTreeHostImpl::drawLayers):
* platform/graphics/chromium/cc/CCQuadCuller.cpp:
* platform/graphics/chromium/cc/CCRenderPass.cpp:
(WebCore::CCRenderPass::CCRenderPass):
(WebCore::CCRenderPass::appendQuadsForRenderSurfaceLayer):
* platform/graphics/chromium/cc/CCRenderPass.h:
(CCRenderPass):
(WebCore::CCRenderPass::framebufferOutputRect):
* platform/graphics/chromium/cc/CCRenderPassDrawQuad.cpp: Renamed from Source/WebCore/platform/graphics/chromium/cc/CCRenderSurfaceDrawQuad.cpp.
(WebCore):
(WebCore::CCRenderPassDrawQuad::create):
(WebCore::CCRenderPassDrawQuad::CCRenderPassDrawQuad):
* platform/graphics/chromium/cc/CCRenderPassDrawQuad.h: Renamed from Source/WebCore/platform/graphics/chromium/cc/CCRenderSurfaceDrawQuad.h.
(WebCore):
(CCRenderPassDrawQuad):
(WebCore::CCRenderPassDrawQuad::renderPass):
(WebCore::CCRenderPassDrawQuad::isReplica):
(WebCore::CCRenderPassDrawQuad::maskTextureId):
(WebCore::CCRenderPassDrawQuad::filters):
(WebCore::CCRenderPassDrawQuad::backgroundFilters):
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore::CCRenderSurface::prepareContentsTexture):
(WebCore::CCRenderSurface::prepareBackgroundTexture):
(WebCore::CCRenderSurface::appendQuads):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(WebCore):
(CCRenderSurface):
* platform/graphics/chromium/cc/CCRenderer.h:
(CCRenderer):
* platform/graphics/chromium/cc/CCScrollbarLayerImpl.cpp:
(WebCore::CCScrollbarLayerImpl::willDraw):
* platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
(WebCore::CCVideoLayerImpl::reserveTextures):
2012-06-04 Abhishek Arya <inferno@chromium.org>
Crash in multi-column layout.
https://bugs.webkit.org/show_bug.cgi?id=88022
Reviewed by Ojan Vafai.
The patch addresses two problems::
1. |this| in RenderBlock::splitBlocks can get destroyed when we
move its children to the clone and later call updateBeforeAfterContent
on the parent. So, we stop accessing its member variables and cache
it in a local.
2. Positioned objects were not getting cleared from our grand parents.
This will happen if our immediate children got moved to a clone tree,
however at our parent nothing was moved. So, we make sure to remove
the positioned objects at every level while we are doing the cloning.
Tests: fast/multicol/span/empty-anonymous-block-split-crash.html
fast/multicol/span/positioned-objects-not-removed-crash.html
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::splitBlocks):
2012-06-04 Tim Horton <timothy_horton@apple.com>
Mac build fails with SVG disabled
https://bugs.webkit.org/show_bug.cgi?id=88165
Reviewed by Dirk Schulze.
Wrap all of SVGCSSPropertyNames in #if ENABLE(SVG),
and mark an unused argument in collectMatchingRulesForList.
* css/SVGCSSPropertyNames.in:
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
2012-06-04 David Kilzer <ddkilzer@apple.com>
BUILD FIX: Unused variable with ENABLE(CONTEXT_MENUS) off
This broke in r118374.
* bindings/js/JSInspectorFrontendHostCustom.cpp:
(WebCore::JSInspectorFrontendHost::showContextMenu): Add
UNUSED_PARAM(exec) when ENABLE(CONTEXT_MENUS) is off.
2012-06-04 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119393.
http://trac.webkit.org/changeset/119393
https://bugs.webkit.org/show_bug.cgi?id=88236
Test is incorrect (duplicate content) and is causing Chromium
failures (Requested by schenney on #webkit).
* html/canvas/CanvasRenderingContext2D.cpp:
(WebCore::CanvasRenderingContext2D::drawImage):
2012-06-04 Zeev Lieber <zlieber@chromium.org>
[chromium] Cleanup scissor rect computation/use with damage
https://bugs.webkit.org/show_bug.cgi?id=87167
Reviewed by Adrienne Walker.
Performing scissorRect computation during
calculateRenderPasses. Storing scissorRect in shared quad state
instead of computing it during drawQuad. Added scissorRect fields
into layers and render surfaces.
Covered by existing layout tests. Introduced more unit tests to
test end-to-end drawing using mock graphic context, and added more
test cases to CCLayerTreeHostCommon to verify scissorRect computation.
* platform/graphics/chromium/LayerChromium.h:
(WebCore::LayerChromium::scissorRect):
(WebCore::LayerChromium::setScissorRect):
(LayerChromium):
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::clearRenderSurface):
(WebCore::LayerRendererChromium::drawRenderPass):
(WebCore::LayerRendererChromium::drawQuad):
(WebCore::LayerRendererChromium::drawRenderSurfaceQuad):
* platform/graphics/chromium/LayerRendererChromium.h:
(LayerRendererChromium):
* platform/graphics/chromium/RenderSurfaceChromium.cpp:
(WebCore::RenderSurfaceChromium::computeRootScissorRectInCurrentSurface):
(WebCore):
* platform/graphics/chromium/RenderSurfaceChromium.h:
(WebCore::RenderSurfaceChromium::scissorRect):
(WebCore::RenderSurfaceChromium::setScissorRect):
(RenderSurfaceChromium):
* platform/graphics/chromium/cc/CCDrawQuad.h:
(WebCore::CCDrawQuad::scissorRect):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::createSharedQuadState):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(CCLayerImpl):
(WebCore::CCLayerImpl::scissorRect):
(WebCore::CCLayerImpl::setScissorRect):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::updateLayers):
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore):
(WebCore::calculateLayerScissorRect):
(WebCore::calculateSurfaceScissorRect):
(WebCore::calculateDrawTransformsInternal):
(WebCore::calculateVisibleAndScissorRectsInternal):
(WebCore::CCLayerTreeHostCommon::calculateDrawTransforms):
(WebCore::CCLayerTreeHostCommon::calculateVisibleAndScissorRects):
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.h:
(CCLayerTreeHostCommon):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::calculateRenderSurfaceLayerList):
(WebCore::CCLayerTreeHostImpl::calculateRenderPasses):
(WebCore::CCLayerTreeHostImpl::drawLayers):
(WebCore::CCLayerTreeHostImpl::swapBuffers):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCRenderPass.cpp:
(WebCore::CCRenderPass::appendQuadsForRenderSurfaceLayer):
* platform/graphics/chromium/cc/CCRenderPass.h:
(CCRenderPass):
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore::CCRenderSurface::createSharedQuadState):
(WebCore::CCRenderSurface::createReplicaSharedQuadState):
(WebCore):
(WebCore::CCRenderSurface::computeRootScissorRectInCurrentSurface):
(WebCore::CCRenderSurface::appendQuads):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(CCRenderSurface):
(WebCore::CCRenderSurface::setScissorRect):
(WebCore::CCRenderSurface::scissorRect):
* platform/graphics/chromium/cc/CCRenderSurfaceDrawQuad.cpp:
(WebCore::CCRenderSurfaceDrawQuad::create):
(WebCore::CCRenderSurfaceDrawQuad::CCRenderSurfaceDrawQuad):
* platform/graphics/chromium/cc/CCRenderSurfaceDrawQuad.h:
(CCRenderSurfaceDrawQuad):
* platform/graphics/chromium/cc/CCRenderer.h:
(CCRenderer):
* platform/graphics/chromium/cc/CCSharedQuadState.cpp:
(WebCore::CCSharedQuadState::create):
(WebCore::CCSharedQuadState::CCSharedQuadState):
* platform/graphics/chromium/cc/CCSharedQuadState.h:
(CCSharedQuadState):
(WebCore::CCSharedQuadState::scissorRect):
2012-06-04 Kentaro Hara <haraken@chromium.org>
Remove SelectorQueryCacheEntry from SelectorQuery.h
https://bugs.webkit.org/show_bug.cgi?id=88204
Reviewed by Hajime Morita.
As anttik@ pointed out in https://bugs.webkit.org/show_bug.cgi?id=87942#c15,
we can remove SelectorQueryCacheEntry by having SelectorQuery hold CSSSelectorList.
Tests: fast/dom/SelectorAPI/* No change in test results.
* dom/SelectorQuery.cpp:
(WebCore::SelectorQuery::SelectorQuery):
(WebCore::SelectorQueryCache::add):
* dom/SelectorQuery.h:
(SelectorQuery):
(SelectorQueryCache):
2012-06-04 Sudarsana Nagineni <sudarsana.nagineni@linux.intel.com>
[GTK] Memory leak in PasteboardHelper
https://bugs.webkit.org/show_bug.cgi?id=88216
Reviewed by Martin Robinson.
Fixed a memory leak in PasteboardHelper by using adoptRef instead
of just getting a new reference of TargetList.
No new tests because of no behavior changes.
* platform/gtk/PasteboardHelper.cpp:
(WebCore::PasteboardHelper::dropAtomsForContext):
2012-06-04 Rob Buis <rbuis@rim.com>
Hit ASSERT when editing attribute value in Element in SVG Document
https://bugs.webkit.org/show_bug.cgi?id=88179
Reviewed by Pavel Feldman.
Make sure the <span> is created as a HTML Element regardless of document in use.
* inspector/InspectorDOMAgent.cpp:
(WebCore):
(WebCore::InspectorDOMAgent::setAttributesAsText):
2012-06-04 Yoshifumi Inoue <yosin@chromium.org>
[Forms] Change function and variable names independent from "double" type
https://bugs.webkit.org/show_bug.cgi?id=88220
Reviewed by Kent Tamura.
This patch is a part of introducing decimal arithmetic to input type.
This patch renames function and variable names which contain term "double"
to "number" to separate concept and implementation.
No new tests. This patch doesn't change behavior.
* html/BaseDateAndTimeInputType.cpp:
(WebCore::BaseDateAndTimeInputType::valueAsDate): Replace parseToDouble to parseToNumber.
(WebCore::BaseDateAndTimeInputType::valueAsNumber): ditto.
(WebCore::BaseDateAndTimeInputType::parseToNumber): ditto.
* html/BaseDateAndTimeInputType.h:
(BaseDateAndTimeInputType):
* html/DateInputType.cpp:
(WebCore::DateInputType::createStepRange): Replace parseToDouble to parseToNumber.
* html/DateTimeInputType.cpp:
(WebCore::DateTimeInputType::createStepRange): Replace parseToDouble to parseToNumber.
* html/DateTimeLocalInputType.cpp:
(WebCore::DateTimeLocalInputType::createStepRange): Replace parseToDouble to parseToNumber.
* html/InputType.cpp:
(WebCore::InputType::rangeUnderflow): Replace parseToDouble to parseToNumber and doubleValue to numericValue.
(WebCore::InputType::rangeOverflow): ditto.
(WebCore::InputType::isInRange): ditto.
(WebCore::InputType::isOutOfRange): ditto.
(WebCore::InputType::stepMismatch): ditto.
(WebCore::InputType::validationMessage): ditto.
(WebCore::InputType::parseToNumber): ditto.
(WebCore::InputType::parseToDoubleWithDecimalPlaces): ditto.
(WebCore::InputType::stepUpFromRenderer): dttio.
* html/InputType.h:
(InputType):
* html/MonthInputType.cpp:
(WebCore::MonthInputType::createStepRange): Replace parseToDouble to parseToNumber.
(WebCore::MonthInputType::parseToNumber): ditto.
* html/MonthInputType.h:
(MonthInputType):
* html/NumberInputType.cpp:
(WebCore::NumberInputType::valueAsNumber): Replace parseToDouble to parseToNumber.
(WebCore::NumberInputType::createStepRange): ditto.
(WebCore::NumberInputType::parseToNumber): ditto.
* html/NumberInputType.h:
* html/RangeInputType.cpp:
(WebCore::ensureMaximum): Added helper function for createStepRange for avoiding cast in max() and using "const".
(WebCore::RangeInputType::valueAsNumber): Replace parseToDouble to parseToNumber.
(WebCore::RangeInputType::createStepRange): ditto.
(WebCore::RangeInputType::handleKeydownEvent): ditto.
(WebCore::RangeInputType::parseToNumber): ditto.
(WebCore::RangeInputType::sanitizeValue): ditto.
* html/RangeInputType.h:
(RangeInputType):
* html/StepRange.cpp:
(WebCore::StepRange::parseStep): Replace DoubleWithDecimalPlaces to NumberWithDecimalPlaces.
(WebCore::StepRange::stepMismatch): Replace parseToDouble to parseToNumber.
* html/StepRange.h:
(WebCore::StepRange::NumberWithDecimalPlaces): Renamed from DoubleWithDecimalPlaces.
(WebCore::StepRange::NumberWithDecimalPlacesOrMissing): Renamed from DoubleWithDecimalPlacesOrMissing.
* html/TimeInputType.cpp:
(WebCore::TimeInputType::createStepRange): Replace parseToDouble to parseToNumber.
* html/WeekInputType.cpp:
(WebCore::WeekInputType::createStepRange): Replace parseToDouble to parseToNumber.
2012-06-04 David Kilzer <ddkilzer@apple.com>
BUILD FIX: FeatureDefines.xcconfig should match across projects
* Configurations/FeatureDefines.xcconfig:
- Add missing ENABLE_CSS_IMAGE_RESOLUTION.
2012-06-04 Justin Novosad <junov@chromium.org>
Canvas 2D - Source rectangles that overlap the source image boundary, not supported by drawImage
https://bugs.webkit.org/show_bug.cgi?id=66574
Reviewed by Stephen White.
Test: fast/canvas/drawImage-clipped-source.html
When the source rectangle argument to drawImage is not entirely
contatained by the domain of the source image, the source rectangle
is clipped, and the destination rectangle is clipped proportionately
Old behavior was to silently fail, which did not comply with the spec.
* html/canvas/CanvasRenderingContext2D.cpp:
(WebCore::normalizeAndClipRects):
(WebCore):
(WebCore::CanvasRenderingContext2D::drawImage):
2012-06-04 Dominik Röttsches <dominik.rottsches@intel.com>
[Cairo] Implementation of GraphicsContext does not save/restore ImageInterpolation correctly
https://bugs.webkit.org/show_bug.cgi?id=87985
Reviewed by Gustavo Noronha Silva.
Moving image interpolation quality setting into state so that it's saved and restored.
Covered by existing test fast/canvas/canvas-imageSmoothingEnabled.html which is now passing.
* platform/graphics/cairo/GraphicsContextPlatformPrivateCairo.h:
(GraphicsContextPlatformPrivate): Removing stale member that was never used.
* platform/graphics/cairo/PlatformContextCairo.cpp:
(WebCore::PlatformContextCairo::State::State):
(PlatformContextCairo::State):
(WebCore::PlatformContextCairo::PlatformContextCairo):
(WebCore::PlatformContextCairo::drawSurfaceToContext):
(WebCore::PlatformContextCairo::setImageInterpolationQuality):
(WebCore):
(WebCore::PlatformContextCairo::imageInterpolationQuality):
* platform/graphics/cairo/PlatformContextCairo.h:
(PlatformContextCairo): Moving implementations for image interpolation quality setter into implementation file.
2012-06-04 Philip Rogers <pdr@google.com>
Fix assert in non-scaling strokes in patterns
https://bugs.webkit.org/show_bug.cgi?id=88198
When calculating the CTM for a rect or path in a pattern we need to use
an SVGPatternElement's local transform. Pattern did not expose the transform
because it does not extend from from SVGStyledLocatable or SVGStyledTransformable.
This patch exposes the pattern's transform in localCoordinateSpaceTransform.
SVGGradientElement does not need this special handling because the children
of a gradient element are restricted to not include shape and structural elements
that would need to calculate a CTM.
Before this patch the added test would assert.
This patch is a prerequisite to a major refactoring of RenderSVGShape &
friends, and the test will be unskipped in a followup patch.
Reviewed by Dirk Schulze.
Tests: svg/stroke/non-scaling-stroke-pattern-expected.svg
svg/stroke/non-scaling-stroke-pattern.svg
* svg/SVGPatternElement.cpp:
(WebCore::SVGPatternElement::localCoordinateSpaceTransform):
(WebCore):
* svg/SVGPatternElement.h:
(SVGPatternElement):
* svg/SVGStyledElement.cpp:
(WebCore::SVGStyledElement::localCoordinateSpaceTransform):
2012-06-04 Alexei Filippov <alexeif@chromium.org>
Web Inspector: speed up _calculateRetainedSizes function
https://bugs.webkit.org/show_bug.cgi?id=87863
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshot.prototype._calculateRetainedSizes):
2012-06-04 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: add unknown size to the memory pie-chart legend
https://bugs.webkit.org/show_bug.cgi?id=88081
Reviewed by Pavel Feldman.
Added "Unknown" part size to the memory pie-chart legend.
* inspector/front-end/NativeMemorySnapshotView.js:
(WebInspector.NativeMemoryProfileType.prototype.buttonClicked.didReceiveMemorySnapshot):
(WebInspector.NativeMemoryProfileType.prototype.buttonClicked):
(WebInspector.MemoryBlockViewProperties._initialize):
(WebInspector.NativeMemoryPieChart.prototype._paint):
2012-06-04 Mike West <mkwst@chromium.org>
Adding explicit console error for duplicated CSP directives.
https://bugs.webkit.org/show_bug.cgi?id=88193
This patch adds an explicit error message for duplicated CSP
directives instead of reusing the "unrecognized directive" error.
Reviewed by Adam Barth.
Test: http/tests/security/contentSecurityPolicy/duplicate-directive.html
* page/ContentSecurityPolicy.cpp:
(CSPDirectiveList):
(WebCore::CSPDirectiveList::logDuplicateDirective):
(WebCore):
(WebCore::CSPDirectiveList::parseReportURI):
(WebCore::CSPDirectiveList::setCSPDirective):
(WebCore::CSPDirectiveList::applySandboxPolicy):
(WebCore::CSPDirectiveList::addDirective):
2012-06-04 Yoshifumi Inoue <yosin@chromium.org>
REGRESSION(r117929) [Forms] input type=number thinks 0 is not a number
https://bugs.webkit.org/show_bug.cgi?id=88208
Reviewed by Kent Tamura.
This patch fixes invalid handling of return value of parseToDoubleForNumberType
in NumberInputType::isAcceptableValue to use isfinite() function for checking
whether specified string is valid number or not. Before r117929, return value
of parseToDoubleForNumberType was bool, but after r117929, it is double.
Test: fast/forms/number/input-number-from-renderer.html
* html/NumberInputType.cpp:
(WebCore::NumberInputType::isAcceptableValue): Use isfinite()
2012-06-04 Antti Koivisto <antti@apple.com>
Crash in WebCore::RenderView::getRetainedWidgets
https://bugs.webkit.org/show_bug.cgi?id=88217
Reviewed by Kenneth Rohde Christiansen.
There is no known repro, the crash has been seen on the field. It is likely that it happens with frame
flattening enabled only.
Null check root renderer in performPostLayoutTasks. The root can legitimely be null for several reasons and
is mostly null checked everywhere else.
* page/FrameView.cpp:
(WebCore::FrameView::performPostLayoutTasks):
2012-06-04 Kentaro Hara <haraken@chromium.org>
Use SelectorQuery in Element::webkitMatchesSelector()
https://bugs.webkit.org/show_bug.cgi?id=88209
Reviewed by Antti Koivisto.
By using SelectorQuery in Element::webkitMatchesSelector(),
we can not only simplify the code but also improve the performance
thanks to the cache mechanism of SelectorQuery which was implemented
in bug 87942.
Tests: fast/dom/SelectorAPI/*. No change in test results.
* dom/Element.cpp:
(WebCore::Element::webkitMatchesSelector):
* dom/SelectorQuery.cpp:
(WebCore::SelectorQuery::matches):
(WebCore):
* dom/SelectorQuery.h:
(SelectorQuery):
2012-05-31 Kinuko Yasuda <kinuko@chromium.org>
[chromium] DataTransferItem.webkitGetAsEntry() should synchronously return entry
https://bugs.webkit.org/show_bug.cgi?id=87969
Reviewed by Tony Chang.
Per discussion on whatwg we should make webkitGetAsEntry synchronous.
http://lists.w3.org/Archives/Public/public-whatwg-archive/2012Apr/0079.html
Tests: editing/pasteboard/data-transfer-items-drag-drop-entry.html
fast/filesystem/cross-filesystem-op.html
* Modules/filesystem/DataTransferItemFileSystem.h:
(DataTransferItemFileSystem):
* Modules/filesystem/DataTransferItemFileSystem.idl:
* Modules/filesystem/chromium/DataTransferItemFileSystemChromium.cpp:
(WebCore::DataTransferItemFileSystem::webkitGetAsEntry): Changed it to synchronously return entry.
2012-06-03 Dominic Cooney <dominicc@chromium.org>
EventTarget.idl should generate JS bindings
https://bugs.webkit.org/show_bug.cgi?id=88120
This is one step in a larger change to expose a constructor for
EventTarget and put EventTarget on the prototype chain of DOM
objects that are event targets, in line with DOM Core. See
<https://bugs.webkit.org/show_bug.cgi?id=67312>
Reviewed by Adam Barth.
No new tests. This change does not change any functionality.
* CMakeLists.txt: Generate JSEventTarget.
* DerivedSources.cpp:
* DerivedSources.make:
* DerivedSources.pri:
* GNUmakefile.list.am:
* Target.pri:
* UseJSC.cmake:
* WebCore.gyp/WebCore.gyp:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSBindingsAllInOne.cpp: JSEventTarget.cpp was renamed.
* bindings/js/JSEventTarget.h: Removed. Generated now.
* bindings/js/JSEventTargetCustom.cpp: Renamed from Source/WebCore/bindings/js/JSEventTarget.cpp.
* bindings/scripts/CodeGeneratorV8.pm:
(GetInternalFields): EventTarget is an EventTarget.
* dom/EventTarget.idl: Defer to toJS, toEventTarget in JSEventTargetCustom.cpp.
2012-06-03 Hayato Ito <hayato@chromium.org>
Supports traversing nodes in an orphaned shadow subtree in ComposedShadowTreeParentWalker.
https://bugs.webkit.org/show_bug.cgi?id=87924
Reviewed by Dimitri Glazkov.
Some events, such as a 'click' event, can be dispatched on nodes in an orphaned shadow subtree.
When resolving a parent node of a shadow root which is not assigned to any shadow insertion point,
we have to return its shadow host as a parent node.
The related shadow DOM spec section is:
https://dvcs.w3.org/hg/webcomponents/raw-file/tip/spec/shadow/index.html#events
Test: fast/dom/shadow/shadow-dom-event-dispatching.html
* dom/ComposedShadowTreeWalker.cpp:
(WebCore::ComposedShadowTreeParentWalker::traverseParentIncludingInsertionPointAndShadowRoot):
2012-06-03 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119359.
http://trac.webkit.org/changeset/119359
https://bugs.webkit.org/show_bug.cgi?id=88189
This broke platform/mac/fast/dom/wrapper-classes-objc.html.
(Requested by dominicc on #webkit).
* bindings/objc/PublicDOMInterfaces.h:
* bindings/scripts/CodeGeneratorObjC.pm:
(GetBaseClass):
* dom/MouseEvent.h:
(MouseEvent):
* dom/WheelEvent.cpp:
(WebCore::WheelEvent::WheelEvent):
* dom/WheelEvent.h:
(WheelEvent):
* dom/WheelEvent.idl:
2012-06-03 Dominic Cooney <dominicc@chromium.org>
WheelEvent should inherit from MouseEvent
https://bugs.webkit.org/show_bug.cgi?id=76104
Reviewed by Kentaro Hara.
The spec for WheelEvent is
<http://www.w3.org/TR/DOM-Level-3-Events/#webidl-events-WheelEvent>
Tests: fast/events/event-creation.html
http://samples.msdn.microsoft.com/ietestcenter/dominheritance/showdominheritancetest.htm?Prototype_WheelEvent
* bindings/objc/PublicDOMInterfaces.h: Remove redundant MouseEvent API from WheelEvent
* bindings/scripts/CodeGeneratorObjC.pm: MouseEvents are Events
* dom/MouseEvent.h: Expose no-arg constructor to WheelEvent
* dom/WheelEvent.cpp:
(WebCore::WheelEvent::WheelEvent): Call MouseEvent superconstructor
* dom/WheelEvent.h: Extend MouseEvent
* dom/WheelEvent.idl: "
2012-06-03 Varun Jain <varunjain@google.com>
[chromium] Add new gesture type (two finger tap) that triggers context menu.
https://bugs.webkit.org/show_bug.cgi?id=88173
Reviewed by Adam Barth.
Test: fast/events/touch/gesture/context-menu-on-two-finger-tap.html
* platform/PlatformEvent.h:
2012-06-02 Andy Estes <aestes@apple.com>
__ZNK7WebCore4Node27traverseNextAncestorSiblingEv should be exported even if ENABLE(FULLSCREEN_API) is false
https://bugs.webkit.org/show_bug.cgi?id=88184
Reviewed by Dan Bernstein.
Node::traverseNextAncestorSibling() is called in several inline
functions defined in ContainerNode.h (a private header), so its symbol
needs to be exported regardless of whether full-screen API is enabled.
* WebCore.exp.in:
2012-06-02 Xianzhu Wang <wangxianzhu@chromium.org>
SVGImageCache leaks image data
https://bugs.webkit.org/show_bug.cgi?id=87792
There are two functions to remove a client from a CachedImage:
- CachedResource::removeClient()
- CachedImage::removeClientForRenderer().
It's easy to make error to call the former which will leak the cached
image buffers in SVGImageCache.
This change combined the two by adding the virtual
CachedResource::didRemoveClient(). CachedImage will do SVGImageCache
cleanup in the function.
Reviewed by Nikolas Zimmermann.
Test: svg/as-image/svg-image-leak-cached-data.html
* loader/cache/CachedFont.h:
(WebCore::CachedFontClient::resourceClientType): Added 'const'.
* loader/cache/CachedImage.cpp:
(WebCore):
(WebCore::CachedImage::didRemoveClient): Removes the client from SVGImageCache.
(WebCore::CachedImage::lookupOrCreateImageForRenderer):
* loader/cache/CachedImage.h:
(CachedImage):
(WebCore::CachedImageClient::resourceClientType): Added 'const'.
* loader/cache/CachedRawResource.h:
(WebCore::CachedRawResourceClient::resourceClientType): Added 'const'.
* loader/cache/CachedResource.cpp:
(WebCore::CachedResource::removeClient): Added invocation of didRemoveClient().
* loader/cache/CachedResource.h:
(WebCore::CachedResource::didRemoveClient): Added for subclasses to do additional works.
* loader/cache/CachedResourceClient.h:
(WebCore::CachedResourceClient::resourceClientType): Added 'const'.
* loader/cache/CachedSVGDocument.h:
(WebCore::CachedSVGDocumentClient::resourceClientType): Added 'const'.
* loader/cache/CachedStyleSheetClient.h:
(WebCore::CachedStyleSheetClient::resourceClientType): Added 'const'.
* rendering/style/StyleCachedImage.cpp:
(WebCore::StyleCachedImage::removeClient):
* rendering/style/StyleCachedImageSet.cpp:
(WebCore::StyleCachedImageSet::removeClient):
* svg/graphics/SVGImageCache.cpp:
(WebCore::SVGImageCache::~SVGImageCache): Added checking for leaks.
(WebCore::SVGImageCache::removeClientFromCache):
(WebCore::SVGImageCache::setRequestedSizeAndScales):
(WebCore::SVGImageCache::requestedSizeAndScales):
(WebCore::SVGImageCache::lookupOrCreateBitmapImageForClient):
* svg/graphics/SVGImageCache.h:
(WebCore):
(SVGImageCache):
2012-06-02 Ryosuke Niwa <rniwa@webkit.org>
Windows build fix after r119330. Unnest the class for now.
* dom/SelectorQuery.cpp:
(WebCore::SelectorQueryCacheEntry::SelectorQueryCacheEntry):
(WebCore::SelectorQueryCache::add):
* dom/SelectorQuery.h:
(SelectorQueryCacheEntry):
(WebCore::SelectorQueryCacheEntry::selectorQuery):
(WebCore):
(SelectorQueryCache):
2012-06-02 Geoffrey Garen <ggaren@apple.com>
DOM string cache should hash pointers, not characters
https://bugs.webkit.org/show_bug.cgi?id=88175
Reviewed by Phil Pizlo and Sam Weinig.
Dromaeo DOM Core reports no change.
http://trac.webkit.org/changeset/84934 accidentally changed from hashing
pointers to hashing characters, due to template defaults. Let's change back.
Hashing characters is not so good because:
(1) It's not memory-safe with HashMap::set(). HashMap::set() replaces
the value but not the key. Since our values own our keys, we need to
ensure object identity between key and value, or the key can be freed
prematurely. (This is impossible to demonstrate with our current
eager sweep behavior, but it shows up as crashes in layout tests if you
change to lazy sweep.)
(2) It's slower.
* bindings/js/DOMWrapperWorld.h:
(WebCore): Override the default hash, which hashes based on characters.
2012-06-02 Eli Fidler <efidler@rim.com>
Don't crash if we ask for fonts that don't exist.
https://bugs.webkit.org/show_bug.cgi?id=88106
Reviewed by Dan Bernstein.
RIM PR 161219
If the site doesn't ask for a reasonable font and the system doesn't
provide a good fallback, FontFallBackList::primaryFontData can be 0,
which can cause a crash.
* platform/graphics/skia/FontCacheSkia.cpp:
(WebCore::FontCache::getLastResortFallbackFont):
2012-06-02 Kentaro Hara <haraken@chromium.org>
[Performance] Optimize querySelector() by caching SelectorQuery objects
https://bugs.webkit.org/show_bug.cgi?id=87942
Reviewed by Antti Koivisto.
This patch improves performance of Node::querySelector() by 7.1x in Safari/Mac
and by 8.5x in Chromium/Linux.
Performance test: Parser/query-selector-first.html, Parser/query-selector-last.html
[query-selector-first.html]
Safari/Mac 264.97 runs/s => 1872.78 runs/s (7.06x speed-up)
Chromium/Linux 244.84 runs/s => 2071.60 runs/s (8.46x speed-up)
[query-selector-last.html]
Safari/Mac 393.73 runs/s => 466.05 runs/s (1.18x speed-up)
Chromium/Linux 401.15 runs/s => 484.45 runs/s (1.20x speed-up)
Previously Node::querySelector() and Node::querySelectorAll() had been
parsing CSS queries every time. This patch optimizes the performance by caching
parsed results onto a Document.
The cache is invalidated when any of CSS related variables is updated.
As per the current implementation of CSSParserContext::operator==(), the CSS related
variables are as follows:
- baseURI
- charset
- mode
- isHTMLDocument
- isCSSCustomFilterEnabled
- isCSSRegionsEnabled
- needsSiteSpecificQuirks
- enforcesCSSMIMETypeInNoQuirksMode
Actually, we do not need to watch all of these variables:
- The current implementation does not watch the change of charset.
charset is always set to a null String by CSSParserContext::CSSParserContext().
- isHTMLDocument never changes.
- isCSSCustomFilterEnabled, isCSSRegionsEnabled, needsSiteSpecificQuirks and
enforcesCSSMIMETypeInNoQuirksMode are not flipped in a user scenario.
If someone changes them, it would be reasonable to expect them
to take the effect only on subsequent document loads.
Thus we do not need to invalidate the cache when these variables are updated.
Consequently, the condition under which we have to invalidate the cache is
that any of the following variables is updated:
- baseURI
- mode
Tests: fast/dom/SelectorAPI/*. No change in test results.
* dom/SelectorQuery.h: SelectorQueryCache is a cache from CSS selectors to parsed results.
SelectorQueryCache::Entry is an entry of the cache.
SelectorQueryCache::Entry holds a SelectorQuery object and a CSSSelectorList object.
The reason why SelectorQueryCache::Entry needs to hold the CSSSelectorList object
is that the CSSSelectorList object keeps the lifetime of CSSSelector objects
in the SelectorQuery object. Since the SelectorQuery object just holds pointers
to CSSSelector objects, the CSSSelectorList object must not be destructed
before the SelectorQuery object is destructed.
(WebCore):
(SelectorDataList):
(WebCore::SelectorQuery::SelectorQuery):
(SelectorQuery):
(SelectorQueryCache):
(WebCore::SelectorQueryCache::SelectorQueryCache):
(Entry):
(WebCore::SelectorQueryCache::Entry::selectorQuery):
* dom/SelectorQuery.cpp:
(WebCore::SelectorQuery::initialize):
(WebCore::SelectorQueryCache::Entry::Entry):
(WebCore::SelectorQueryCache::add): Returns a cached SelectorQuery object if any.
Otherwise, parses a given CSS selector, creates a SelectorQuery object,
adds the SelectorQuery object to a new entry in the cache, returns the SelectorQuery
object.
(WebCore::SelectorQueryCache::invalidate): Clears the cache.
* dom/Document.h:
(WebCore):
(Document):
* dom/Document.cpp:
(WebCore::Document::selectorQueryCache):
(WebCore):
(WebCore::Document::setCompatibilityMode): Invalidates the cache
when m_compatibilityMode is updated.
(WebCore::Document::updateBaseURL): Invalidates the cache
when m_baseURL is updated.
* dom/Node.h: Changed String to AtomicString, since the key of the cache
should be AtomicString.
(Node):
* dom/Node.cpp: Optimized the code by using the cache.
(WebCore::Node::querySelector):
(WebCore::Node::querySelectorAll):
2012-06-02 Dan Bernstein <mitz@apple.com>
Reverted the last change.
* WebCore.xcodeproj/project.pbxproj:
2012-06-02 Dan Bernstein <mitz@apple.com>
Stop treating Perl code as private headers.
* WebCore.xcodeproj/project.pbxproj: Removed CodeGenerator.pm, generate-bindings.pl,
IDLParser.pm, IDLStructure.pm, and preprocessor.pm from the Copy Headers build phase.
2012-06-02 Zeno Albisser <zeno@webkit.org>
Fix and enable WebGL for WebKit2 on Qt.
https://bugs.webkit.org/show_bug.cgi?id=86214
Make GraphicsContext3DPrivate use GraphicsSurfaces
for WK2. The GraphicsContext3D then uses the existing
RenderBuffer for multisample rendering.
When WebGraphicsLayer::syncCompositingState is being
executed, the canvas is being synced as well. This means
that the RenderBuffer contents are being blit onto
a GraphicsSurface, and the GraphicsSurface token is
being sent to the UIProcess.
The WebLayerTreeRenderer then creates a
TextureMapperSurfaceBackingStore for the canvas and
passes the GraphicsSurface token as an argument.
The token can then be used to identify the GraphicsSurface
from the UIProcess side.
Reviewed by Noam Rosenthal.
* platform/graphics/GraphicsContext3D.h:
Added createGraphicsSurfaces function. This is currently only
being used by the Qt port.
* platform/graphics/opengl/GraphicsContext3DOpenGLCommon.cpp:
(WebCore::GraphicsContext3D::reshape):
Calling the createGraphicsSurfaces function when the GraphicsContext3D
is reshaped. This is currently only relevant for the Qt port.
* platform/graphics/qt/GraphicsContext3DQt.cpp:
(GraphicsContext3DPrivate):
Added m_frontBufferGraphicsSurface, m_backBufferGraphicsSurface
and m_surfaceFlags members.
(WebCore::GraphicsContext3DPrivate::GraphicsContext3DPrivate):
In case of WK2, create a QOpenGLContext and two GraphicsSurface
for sharing the WebGL content with the UIProcess. One GraphicsSurface
is being used as the front, the other one as the backbuffer.
Creating a QOpenGLContext currently requires showing a QWindow.
For the moment we therefore create a minimal QWindow and place
it offscreen.
(WebCore::GraphicsContext3DPrivate::copyToGraphicsSurface):
This new function is called from the WebGraphicsLayer,
to blit the multisample framebuffer and copy its contents
onto the GraphicsSurface.
(WebCore::GraphicsContext3DPrivate::createGraphicsSurfaces):
Whenever the GraphicsContext3D is being reshaped,
new GraphicsSurfaces must be created with the updated dimensions.
(WebCore::GraphicsContext3D::createGraphicsSurfaces):
* platform/graphics/texmap/TextureMapperPlatformLayer.h:
Added a new virtual function copyToGraphicsSurface.
(TextureMapperPlatformLayer):
(WebCore::TextureMapperPlatformLayer::copyToGraphicsSurface):
2012-06-02 Tony Chang <tony@chromium.org>
Rename the flexbox CSS propery values from start to flex-start and end to flex-end
https://bugs.webkit.org/show_bug.cgi?id=88152
Reviewed by Ojan Vafai.
The spec changed. This is to differentiate from start/end which will
depend on writing mode direction, rather than flex direction.
No new tests, just updated the existing values.
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
(WebCore::CSSPrimitiveValue::operator EAlignItems):
(WebCore::CSSPrimitiveValue::operator EJustifyContent):
(WebCore::CSSPrimitiveValue::operator EAlignContent):
* css/CSSValueKeywords.in:
* rendering/RenderFlexibleBox.cpp:
(WebCore::initialJustifyContentOffset):
(WebCore::alignmentForChild):
(WebCore::initialAlignContentOffset):
(WebCore::RenderFlexibleBox::alignFlexLines):
(WebCore::RenderFlexibleBox::alignChildren):
* rendering/style/RenderStyle.h:
* rendering/style/RenderStyleConstants.h:
2012-06-01 Alexandre Elias <aelias@google.com>
[chromium] Software compositor initialization and base classes
https://bugs.webkit.org/show_bug.cgi?id=87920
Reviewed by James Robinson.
This introduces new wrapper class CCGraphicsContext and base class
CCRenderer which will be used for software compositing support.
If no GraphicsContext3D is available, early return for now.
No new tests. (No-op change in 3d mode.)
* platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.cpp:
(WebCore::BitmapCanvasLayerTextureUpdater::Texture::updateRect):
(WebCore::BitmapCanvasLayerTextureUpdater::updateTextureRect):
* platform/graphics/chromium/BitmapCanvasLayerTextureUpdater.h:
(Texture):
(BitmapCanvasLayerTextureUpdater):
* platform/graphics/chromium/BitmapSkPictureCanvasLayerTextureUpdater.cpp:
(WebCore::BitmapSkPictureCanvasLayerTextureUpdater::Texture::updateRect):
(WebCore::BitmapSkPictureCanvasLayerTextureUpdater::updateTextureRect):
* platform/graphics/chromium/BitmapSkPictureCanvasLayerTextureUpdater.h:
(Texture):
(BitmapSkPictureCanvasLayerTextureUpdater):
* platform/graphics/chromium/CanvasLayerTextureUpdater.h:
* platform/graphics/chromium/FrameBufferSkPictureCanvasLayerTextureUpdater.cpp:
(WebCore::FrameBufferSkPictureCanvasLayerTextureUpdater::Texture::updateRect):
(WebCore::FrameBufferSkPictureCanvasLayerTextureUpdater::updateTextureRect):
* platform/graphics/chromium/FrameBufferSkPictureCanvasLayerTextureUpdater.h:
(Texture):
(FrameBufferSkPictureCanvasLayerTextureUpdater):
* platform/graphics/chromium/ImageLayerChromium.cpp:
(WebCore::ImageLayerTextureUpdater::Texture::updateRect):
(WebCore::ImageLayerTextureUpdater::updateTextureRect):
* platform/graphics/chromium/LayerChromium.h:
(WebCore):
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::CCRenderer::toGLMatrix):
(WebCore):
(WebCore::LayerRendererChromium::create):
(WebCore::CCRenderer::CCRenderer):
(WebCore::LayerRendererChromium::LayerRendererChromium):
(WebCore::LayerRendererChromium::copyPlaneToTexture):
(WebCore::LayerRendererChromium::drawHeadsUpDisplay):
(WebCore::LayerRendererChromium::getFramebufferTexture):
(WebCore::LayerRendererChromium::bindFramebufferToTexture):
* platform/graphics/chromium/LayerRendererChromium.h:
(CCRenderer):
(WebCore::CCRenderer::~CCRenderer):
(WebCore::CCRenderer::ccContext):
(WebCore::CCRenderer::settings):
(WebCore::CCRenderer::viewportSize):
(WebCore::CCRenderer::viewportWidth):
(WebCore::CCRenderer::viewportHeight):
(WebCore::CCRenderer::projectionMatrix):
(WebCore::CCRenderer::windowMatrix):
(WebCore::CCRenderer::sharedGeometryQuad):
(WebCore):
(LayerRendererChromium):
* platform/graphics/chromium/LayerTextureSubImage.cpp:
(WebCore::LayerTextureSubImage::upload):
(WebCore::LayerTextureSubImage::uploadWithTexSubImage):
(WebCore::LayerTextureSubImage::uploadWithMapTexSubImage):
* platform/graphics/chromium/LayerTextureSubImage.h:
(WebCore):
(LayerTextureSubImage):
* platform/graphics/chromium/LayerTextureUpdater.h:
(WebCore):
(Texture):
* platform/graphics/chromium/ManagedTexture.cpp:
(WebCore::ManagedTexture::bindTexture):
(WebCore::ManagedTexture::framebufferTexture2D):
* platform/graphics/chromium/ManagedTexture.h:
(WebCore):
(ManagedTexture):
* platform/graphics/chromium/RenderSurfaceChromium.cpp:
* platform/graphics/chromium/TextureCopier.cpp:
(WebCore::AcceleratedTextureCopier::copyTexture):
* platform/graphics/chromium/TextureCopier.h:
(TextureCopier):
(AcceleratedTextureCopier):
* platform/graphics/chromium/TextureUploader.h:
(TextureUploader):
* platform/graphics/chromium/ThrottledTextureUploader.cpp:
(WebCore::ThrottledTextureUploader::uploadTexture):
* platform/graphics/chromium/ThrottledTextureUploader.h:
(ThrottledTextureUploader):
* platform/graphics/chromium/cc/CCGraphicsContext.h: Copied from Source/WebCore/platform/graphics/chromium/CanvasLayerTextureUpdater.h.
(WebCore):
(CCGraphicsContext):
(WebCore::CCGraphicsContext::create2D):
(WebCore::CCGraphicsContext::create3D):
(WebCore::CCGraphicsContext::context3D):
(WebCore::CCGraphicsContext::CCGraphicsContext):
* platform/graphics/chromium/cc/CCHeadsUpDisplay.cpp:
(WebCore::CCHeadsUpDisplay::draw):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::willDraw):
(WebCore::CCLayerImpl::bindContentsTexture):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(WebCore):
(CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::createContext):
(WebCore::CCLayerTreeHost::context):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(CCLayerTreeHostClient):
(WebCore::CCSettings::CCSettings):
(CCSettings):
(CCLayerTreeHost):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::context):
(WebCore::CCLayerTreeHostImpl::initializeLayerRenderer):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::layerRenderer):
* platform/graphics/chromium/cc/CCProxy.h:
(WebCore):
(CCProxy):
* platform/graphics/chromium/cc/CCScrollbarLayerImpl.cpp:
(WebCore::CCScrollbarLayerImpl::willDraw):
* platform/graphics/chromium/cc/CCScrollbarLayerImpl.h:
(CCScrollbarLayerImpl):
* platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
(WebCore::CCSingleThreadProxy::context):
(WebCore::CCSingleThreadProxy::initializeContext):
(WebCore::CCSingleThreadProxy::recreateContext):
* platform/graphics/chromium/cc/CCSingleThreadProxy.h:
(CCSingleThreadProxy):
* platform/graphics/chromium/cc/CCTextureUpdater.cpp:
(WebCore::CCTextureUpdater::update):
* platform/graphics/chromium/cc/CCTextureUpdater.h:
(CCTextureUpdater):
* platform/graphics/chromium/cc/CCThreadProxy.cpp:
(WebCore::CCThreadProxy::context):
(WebCore::CCThreadProxy::initializeContext):
(WebCore::CCThreadProxy::recreateContext):
(WebCore::CCThreadProxy::beginFrame):
(WebCore::CCThreadProxy::initializeContextOnImplThread):
(WebCore::CCThreadProxy::recreateContextOnImplThread):
* platform/graphics/chromium/cc/CCThreadProxy.h:
(CCThreadProxy):
* platform/graphics/chromium/cc/CCVideoLayerImpl.cpp:
(WebCore::CCVideoLayerImpl::willDraw):
(WebCore::CCVideoLayerImpl::willDrawInternal):
(WebCore::CCVideoLayerImpl::reserveTextures):
* platform/graphics/chromium/cc/CCVideoLayerImpl.h:
(CCVideoLayerImpl):
2012-06-01 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119283, r119287, and r119291.
http://trac.webkit.org/changeset/119283
http://trac.webkit.org/changeset/119287
http://trac.webkit.org/changeset/119291
https://bugs.webkit.org/show_bug.cgi?id=88159
Not only broke compilation in the initial commit but also
broke LayerChromiumTest.basicCreateAndDestroy (Requested by
rniwa on #webkit).
* WebCore.gypi:
* platform/graphics/chromium/AnimationTranslationUtil.cpp: Removed.
* platform/graphics/chromium/AnimationTranslationUtil.h: Removed.
* platform/graphics/chromium/GraphicsLayerChromium.cpp:
(WebCore::GraphicsLayerChromium::addAnimation):
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::addAnimation):
* platform/graphics/chromium/LayerChromium.h:
(LayerChromium):
* platform/graphics/chromium/LinkHighlight.cpp:
(WebCore::LinkHighlight::LinkHighlight):
* platform/graphics/chromium/cc/CCAnimationCurve.h:
(CCTransformAnimationCurve):
* platform/graphics/chromium/cc/CCKeyframedAnimationCurve.cpp:
(WebCore::CCTransformKeyframe::create):
(WebCore::CCTransformKeyframe::CCTransformKeyframe):
(WebCore::CCTransformKeyframe::value):
(WebCore::CCTransformKeyframe::clone):
(WebCore::CCKeyframedTransformAnimationCurve::getValue):
* platform/graphics/chromium/cc/CCKeyframedAnimationCurve.h:
(CCTransformKeyframe):
(CCKeyframedTransformAnimationCurve):
* platform/graphics/chromium/cc/CCLayerAnimationController.cpp:
(WebCore::CCLayerAnimationController::addAnimation):
(WebCore):
(WebCore::CCLayerAnimationController::add):
(WebCore::CCLayerAnimationController::pushNewAnimationsToImplThread):
(WebCore::CCLayerAnimationController::replaceImplThreadAnimations):
(WebCore::CCLayerAnimationController::tickAnimations):
* platform/graphics/chromium/cc/CCLayerAnimationController.h:
(CCLayerAnimationControllerClient):
(CCLayerAnimationController):
* platform/graphics/chromium/cc/CCLayerImpl.h:
2012-06-01 Levi Weintraub <leviw@chromium.org>
Large number constant in TransformationMatrix::projectPoint overflows FractionalLayoutUnits with sub-pixel layout enabled
https://bugs.webkit.org/show_bug.cgi?id=87896
Reviewed by James Robinson.
kLargeNumber is meant to avoid overflowing when projecting a point through a transform. Unfortunately,
due to FractionalLayoutUnit's diminished range compared to integers, we were overflowing anyways. This
change adjusts our large number by the same value, but adjusted for our denominator.
This was originally missed because transforms/3d is marked as pass/fail in Chromium's test_expectations.txt
https://bugs.webkit.org/show_bug.cgi?id=66989 tracks removing this problem.
No new tests. Covered by transforms/3d/hit-testing/perspective-clipped.html
* platform/graphics/transforms/TransformationMatrix.cpp:
(WebCore::TransformationMatrix::projectPoint):
2012-06-01 Tim Horton <timothy_horton@apple.com>
Cleanup GeneratorGeneratedImage/Gradient changes from r117858
https://bugs.webkit.org/show_bug.cgi?id=88063
Reviewed by Simon Fraser.
No new tests, code cleanup.
* platform/graphics/Generator.h:
(WebCore::Generator::hash): hash() should be const.
* platform/graphics/GeneratorGeneratedImage.cpp:
(WebCore::GeneratorGeneratedImage::drawPattern): Minor cleanup.
* platform/graphics/Gradient.cpp:
(WebCore::Gradient::Gradient): Rename m_hashCache to m_cachedHash.
(WebCore::Gradient::addColorStop): Rename clearHashCache to invalidateHash.
(WebCore::Gradient::sortStopsIfNecessary): Rename clearHashCache to invalidateHash.
(WebCore::Gradient::setSpreadMethod): Rename clearHashCache to invalidateHash.
(WebCore::Gradient::setGradientSpaceTransform): Rename clearHashCache to invalidateHash.
(WebCore::Gradient::hash): Use intHash instead of the pair hasher. Add compile time size checks
for structures being passed to StringHasher.
* platform/graphics/Gradient.h:
(WebCore::Gradient::setP0): Rename clearHashCache to invalidateHash.
(WebCore::Gradient::setP1): Rename clearHashCache to invalidateHash.
(WebCore::Gradient::setStartRadius): Rename clearHashCache to invalidateHash.
(WebCore::Gradient::setEndRadius): Rename clearHashCache to invalidateHash.
(WebCore::Gradient::invalidateHash): Rename clearHashCache to invalidateHash.
* platform/graphics/GraphicsContext.cpp:
(WebCore::scalesMatch): Added. Determine if the scale of two AffineTransforms match.
(WebCore::GraphicsContext::isCompatibleWithBuffer): Make use of scalesMatch to simplify the logic.
2012-06-01 Raymond Toy <rtoy@google.com>
Remove RefInfo class
https://bugs.webkit.org/show_bug.cgi?id=87904
Reviewed by Chris Rogers.
No new tests; covered by existing tests.
* Modules/webaudio/AudioBufferSourceNode.cpp:
(WebCore::AudioBufferSourceNode::~AudioBufferSourceNode): Clear
panner node.
(WebCore::AudioBufferSourceNode::totalPitchRate):
(WebCore::AudioBufferSourceNode::setPannerNode): Use
RefTypeConnection counting for panner node.
(WebCore):
(WebCore::AudioBufferSourceNode::clearPannerNode): New function to
clear panner node in AudioBufferSourceNode.
(WebCore::AudioBufferSourceNode::finish): Clear panner node when done.
* Modules/webaudio/AudioBufferSourceNode.h:
(AudioBufferSourceNode): Add virtual finish(), m_pannerNode is not
a RefPtr anymore.
* Modules/webaudio/AudioContext.cpp:
(WebCore::AudioContext::addDeferredFinishDeref): Remove unneeded RefTyp.
(WebCore::AudioContext::handleDeferredFinishDerefs): Remove
unneeded RefType.
* Modules/webaudio/AudioContext.h:
(AudioContext): Update addDeferredFinishDeref signature.
* Modules/webaudio/AudioNode.cpp:
(WebCore::AudioNode::deref): Update call to addDeferredFinishDeref.
* Modules/webaudio/AudioScheduledSourceNode.h:
(AudioScheduledSourceNode): Make finish() virtual.
2012-06-01 Tony Chang <tony@chromium.org>
rename -webkit-flex-pack and -webkit-flex-line-pack to -webkit-justify-content and -webkit-align-content
https://bugs.webkit.org/show_bug.cgi?id=88113
Reviewed by Ojan Vafai.
This recently changed in the spec:
http://dev.w3.org/csswg/css3-flexbox/#justify-content-property
http://dev.w3.org/csswg/css3-flexbox/#align-content-property
No new tests, updated existing tests.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore):
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
(WebCore::isKeywordPropertyID):
(WebCore::CSSParser::parseValue):
* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
(WebCore::CSSPrimitiveValue::operator EJustifyContent):
(WebCore::CSSPrimitiveValue::operator EAlignContent):
* css/CSSProperty.cpp:
(WebCore::CSSProperty::isInheritedProperty):
* css/CSSPropertyNames.in:
* css/StyleBuilder.cpp:
(WebCore::StyleBuilder::StyleBuilder):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::repositionLogicalHeightDependentFlexItems):
(WebCore::initialJustifyContentOffset):
(WebCore::justifyContentSpaceBetweenChildren):
(WebCore::RenderFlexibleBox::layoutAndPlaceChildren):
(WebCore::RenderFlexibleBox::layoutColumnReverse):
(WebCore::initialAlignContentOffset):
(WebCore::alignContentSpaceBetweenChildren):
(WebCore::RenderFlexibleBox::alignFlexLines):
* rendering/RenderFlexibleBox.h:
* rendering/style/RenderStyle.h:
* rendering/style/RenderStyleConstants.h:
* rendering/style/StyleFlexibleBoxData.cpp:
(WebCore::StyleFlexibleBoxData::StyleFlexibleBoxData):
(WebCore::StyleFlexibleBoxData::operator==):
* rendering/style/StyleFlexibleBoxData.h:
(StyleFlexibleBoxData):
* rendering/style/StyleRareNonInheritedData.cpp:
(WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
(WebCore::StyleRareNonInheritedData::operator==):
* rendering/style/StyleRareNonInheritedData.h:
(StyleRareNonInheritedData):
2012-06-01 Adam Barth <abarth@webkit.org>
Overflow scrolling doesn't need to create a stacking context is the overflow is hidden
https://bugs.webkit.org/show_bug.cgi?id=88057
Reviewed by James Robinson.
This patch causes us to match the iOS implementation of this feature.
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
2012-06-01 James Robinson <jamesr@chromium.org>
[chromium] Unreviewed compile fix for r119283
For the record, Dana Jensens <danakj@chromium.org> wrote this slightly faster than I did.
* platform/graphics/chromium/LayerChromium.h:
(WebCore::LayerChromium::bounds):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(WebCore::CCLayerImpl::bounds):
2012-06-01 Ian Vollick <vollick@chromium.org>
[chromium] Accelerated animations should use WebTransformOperations
https://bugs.webkit.org/show_bug.cgi?id=87686
Reviewed by James Robinson.
CCTransformKeyframe new owns a WebTransformOperations rather than a
TransformOperations. LayerChromium's API has been changed so that
LayerChromium::addAnimation should take only a CCActiveAnimation.
GraphicsLayerChromium is new responsible for translating to
WebTransformOperations and creating CCActiveAnimations. Tests that use
the public API (that is, they call addAnimation with KeyframeValueList
and Animation arguments) have been moved to GraphicsLayerChromiumTest.
Unit tests:
GraphicsLayerChromiumTest.createOpacityAnimation
GraphicsLayerChromiumTest.createTransformAnimation
GraphicsLayerChromiumTest.createTransformAnimationWithBigRotation
GraphicsLayerChromiumTest.createTransformAnimationWithSingularMatrix
GraphicsLayerChromiumTest.createReversedAnimation
GraphicsLayerChromiumTest.createAlternatingAnimation
GraphicsLayerChromiumTest.createReversedAlternatingAnimation
* WebCore.gypi:
* platform/graphics/chromium/AnimationTranslationUtil.cpp: Added.
(WebCore):
(WebCore::toWebTransformOperations):
(WebCore::appendKeyframe):
(WebCore::CCKeyframedTransformAnimationCurve):
(WebCore::createActiveAnimation):
* platform/graphics/chromium/AnimationTranslationUtil.h: Added.
(WebCore):
* platform/graphics/chromium/GraphicsLayerChromium.cpp:
(WebCore::GraphicsLayerChromium::addAnimation):
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::addAnimation):
* platform/graphics/chromium/LayerChromium.h:
(WebCore):
(LayerChromium):
* platform/graphics/chromium/LinkHighlight.cpp:
(WebCore::LinkHighlight::LinkHighlight):
* platform/graphics/chromium/cc/CCAnimationCurve.h:
(CCTransformAnimationCurve):
* platform/graphics/chromium/cc/CCKeyframedAnimationCurve.cpp:
(WebCore::CCTransformKeyframe::create):
(WebCore::CCTransformKeyframe::CCTransformKeyframe):
(WebCore::CCTransformKeyframe::value):
(WebCore::CCTransformKeyframe::clone):
(WebCore::CCKeyframedTransformAnimationCurve::getValue):
* platform/graphics/chromium/cc/CCKeyframedAnimationCurve.h:
(CCTransformKeyframe):
(CCKeyframedTransformAnimationCurve):
* platform/graphics/chromium/cc/CCLayerAnimationController.cpp:
(WebCore::CCLayerAnimationController::removeAnimation):
(WebCore):
(WebCore::CCLayerAnimationController::addAnimation):
(WebCore::CCLayerAnimationController::getActiveAnimation):
(WebCore::CCLayerAnimationController::pushNewAnimationsToImplThread):
(WebCore::CCLayerAnimationController::replaceImplThreadAnimations):
(WebCore::CCLayerAnimationController::tickAnimations):
* platform/graphics/chromium/cc/CCLayerAnimationController.h:
(CCLayerAnimationControllerClient):
(CCLayerAnimationController):
2012-06-01 Mihai Parparita <mihaip@chromium.org>
Improve synchronous XHR disabling
https://bugs.webkit.org/show_bug.cgi?id=88032
Reviewed by Alexey Proskuryakov.
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::setSyncXHRInDocumentsEnabled): Fix a
copy-and-paste error
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::open): Improves the disabled message, as
suggested in a comment after r118599 was put in the commit queue.
2012-06-01 Beth Dakin <bdakin@apple.com>
https://bugs.webkit.org/show_bug.cgi?id=87774
REGRESSION (r105515): reflection masks are truncated at zoom levels < 1
-and corresponding-
<rdar://problem/11387506>
Reviewed by Simon Fraser.
paintNinePieceImage() expects un-zoomed results from
calculateImageIntrinsicDimensions(). This was previously addressed by having
paintNinePieceImage() divide the effective zoom out of the result from
calculateImageIntrinsicDimensions(). However, that results in buggy behavior for
generated images and images with percentage sizes. In the end it seems best to
just send a parameter to calculateImageIntrinsicDimensions() indicating whether
the caller wants the result to be scaled by the effective zoom when appropriate.
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::calculateImageIntrinsicDimensions):
(WebCore::RenderBoxModelObject::calculateFillTileSize):
(WebCore::RenderBoxModelObject::paintNinePieceImage):
* rendering/RenderBoxModelObject.h:
(RenderBoxModelObject):
2012-06-01 Julien Chaffraix <jchaffraix@webkit.org>
Prepare table collapsed border computation to support mixed directionality on row group
https://bugs.webkit.org/show_bug.cgi?id=88110
Reviewed by Ojan Vafai.
No expected change in behavior.
One big issue with supporting mixed directionality inside a table is that the start / end
borders don't align between table parts anymore: the start border of a ltr table will have
to match the end border of a rtl row group for the purpose of collapsed border computation.
This change adds the concept of adjoining borders in the table direction so that we can safely
hide which exact borders we pick up for the collapsed border computation.
* rendering/RenderTable.cpp:
(WebCore::RenderTable::calcBorderStart):
(WebCore::RenderTable::calcBorderEnd):
Refactored those functions to use proper naming along with the new APIs. The name 'adjoining' is
used extensively as we cannot make any assumptions on which borders we will get.
(WebCore::RenderTable::recalcBordersInRowDirection):
Added a FIXME found during testing.
* rendering/RenderTableCell.h:
(WebCore::RenderTableCell::borderAdjoiningTableStart):
(WebCore::RenderTableCell::borderAdjoiningTableEnd):
* rendering/RenderTableRow.h:
(WebCore::RenderTableRow::borderAdjoiningTableStart):
(WebCore::RenderTableRow::borderAdjoiningTableEnd):
* rendering/RenderTableSection.h:
(WebCore::RenderTableSection::borderAdjoiningTableStart):
(WebCore::RenderTableSection::borderAdjoiningTableEnd):
Those functions are the same at the moment to match the existing code. They
will be changed to use the proper directionality in a follow up patch.
* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::firstRowCellAdjoiningTableStart):
(WebCore::RenderTableSection::firstRowCellAdjoiningTableEnd):
Those functions return the cells that is adjoining a table edge. Due to us flipping
the cells at layout to match the section's direction, those functions will need to
account for mixed direction in determining the right cell to consider.
2012-06-01 Shezan Baig <shezbaig.wk@gmail.com>
Indenting a paragraph that begins with a link 3 times breaks the paragraph into two paragraphs
https://bugs.webkit.org/show_bug.cgi?id=87428
Reviewed by Ryosuke Niwa.
Fix the way lastNode (our insertion point) is updated whenever
traverseNextSibling moves up to a new parent, so that the relative
depth between the next sibling and the original start node is
maintained in the clone. The divergence in depth broke the paragraph
into two paragraphs because the next sibling was inserted outside the
blockquote that was created for the indentation.
Note that the topNode is not required anymore because it is no longer
used anywhere.
Tests: editing/execCommand/indent-nested-inlines-1.html
editing/execCommand/indent-nested-inlines-2.html
* editing/CompositeEditCommand.cpp:
(WebCore::CompositeEditCommand::cloneParagraphUnderNewElement):
2012-06-01 Joe Thomas <joethomas@motorola.com>
getComputedStyle for background shorthand property does not return background-origin and background-clip.
https://bugs.webkit.org/show_bug.cgi?id=86155
Reviewed by Tony Chang.
Added background-origin and background-clip CSS Property values to the background shorthand CSSValueList.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::getBackgroundShorthandValue):
2012-06-01 Mark Pilgrim <pilgrim@chromium.org>
[Chromium] Call clipboard methods directly
https://bugs.webkit.org/show_bug.cgi?id=88038
Reviewed by Adam Barth.
Part of a refactoring series. See tracking bug 82948.
* WebCore.gypi:
* platform/Pasteboard.h:
(Pasteboard):
* platform/chromium/ChromiumDataObject.cpp:
(WebCore::ChromiumDataObject::createFromPasteboard):
* platform/chromium/ChromiumDataObjectItem.cpp:
(WebCore::ChromiumDataObjectItem::getAsFile):
(WebCore::ChromiumDataObjectItem::internalGetAsString):
* platform/chromium/ClipboardUtilitiesChromium.cpp:
(WebCore::currentPasteboardBuffer):
* platform/chromium/ClipboardUtilitiesChromium.h:
(WebCore):
* platform/chromium/PasteboardChromium.cpp:
(WebCore::Pasteboard::writeSelection):
(WebCore::Pasteboard::writePlainText):
(WebCore::Pasteboard::writeURL):
(WebCore::Pasteboard::writeImage):
(WebCore::Pasteboard::writeClipboard):
(WebCore::Pasteboard::canSmartReplace):
(WebCore::Pasteboard::plainText):
(WebCore::Pasteboard::documentFragment):
* platform/chromium/PasteboardPrivate.h: Removed.
* platform/chromium/PlatformSupport.h:
(WebCore):
(PlatformSupport):
2012-06-01 Hugo Parente Lima <hugo.lima@openbossa.org>
Improve handling of legacy viewport meta tags
https://bugs.webkit.org/show_bug.cgi?id=55874
Reviewed by Adam Barth.
We now support MobileOptimized and HandheldFriendly as well.
We set width equal to device-width for HandheldFriendly and
for MobileOptimized, the content value of MobileOptimized is
ignored and the initial-scale set to 1 to fit Android behavior.
The prioritizing is done the same way as on Windows Phone 7:
XHTML Mobile Profile found
HandheldFriendly
MobileOptimized (overrides HandheldFriendly)
Viewport (overrides all above)
Original patch by Kenneth Rohde Christiansen.
Tests: fast/viewport/viewport-legacy-handheldfriendly.html
fast/viewport/viewport-legacy-mobileoptimized-2.html
fast/viewport/viewport-legacy-mobileoptimized-3.html
fast/viewport/viewport-legacy-mobileoptimized.html
fast/viewport/viewport-legacy-ordering-1.html
fast/viewport/viewport-legacy-ordering-2.html
fast/viewport/viewport-legacy-ordering-3.html
fast/viewport/viewport-legacy-ordering-4.html
fast/viewport/viewport-legacy-ordering-5.html
fast/viewport/viewport-legacy-ordering-6.html
fast/viewport/viewport-legacy-ordering-7.html
fast/viewport/viewport-legacy-ordering-8.html
fast/viewport/viewport-legacy-ordering-9.html
* dom/Document.cpp:
(WebCore::Document::setDocType):
(WebCore::Document::processViewport): Add an origin parameter to
tell what is changing the viewport.
* dom/Document.h:
(Document):
* dom/ViewportArguments.h:
* html/HTMLMetaElement.cpp:
(WebCore::HTMLMetaElement::process):
2012-06-01 Tony Chang <tony@chromium.org>
rename -webkit-flex-order to -webkit-order
https://bugs.webkit.org/show_bug.cgi?id=88104
Reviewed by Ojan Vafai.
This recently changed in the spec:
http://dev.w3.org/csswg/css3-flexbox/#order-property
No new tests, just updated existing results.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore):
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
* css/CSSProperty.cpp:
(WebCore::CSSProperty::isInheritedProperty):
* css/CSSPropertyNames.in:
* css/StyleBuilder.cpp:
(WebCore::StyleBuilder::StyleBuilder):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::OrderIterator::OrderIterator):
(WebCore::RenderFlexibleBox::OrderIterator::next):
(WebCore::RenderFlexibleBox::layoutBlock):
(WebCore::RenderFlexibleBox::repositionLogicalHeightDependentFlexItems):
(WebCore::RenderFlexibleBox::layoutFlexItems):
(WebCore::RenderFlexibleBox::computeMainAxisPreferredSizes):
(WebCore::RenderFlexibleBox::computeNextFlexLine):
(WebCore::RenderFlexibleBox::packFlexLines):
(WebCore::RenderFlexibleBox::alignChildren):
(WebCore::RenderFlexibleBox::flipForRightToLeftColumn):
(WebCore::RenderFlexibleBox::flipForWrapReverse):
* rendering/RenderFlexibleBox.h:
* rendering/style/RenderStyle.h:
* rendering/style/StyleFlexibleBoxData.cpp:
(WebCore::StyleFlexibleBoxData::StyleFlexibleBoxData):
(WebCore::StyleFlexibleBoxData::operator==):
* rendering/style/StyleFlexibleBoxData.h:
(StyleFlexibleBoxData):
* rendering/style/StyleRareNonInheritedData.cpp:
(WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
(WebCore::StyleRareNonInheritedData::operator==):
* rendering/style/StyleRareNonInheritedData.h:
(StyleRareNonInheritedData):
2012-06-01 Zeno Albisser <zeno@webkit.org>
Unreviewed build fix after r119247.
Do not include TextureMapperGL.h when not building
with GRAPHICS_SURFACE.
* platform/graphics/texmap/TextureMapperBackingStore.cpp:
2012-06-01 Mario Sanchez Prada <msanchez@igalia.com>
[GTK] Add a new and reusable Geoclue-based geolocation provider in WebCore
https://bugs.webkit.org/show_bug.cgi?id=87800
Reviewed by Carlos Garcia Campos.
Added new and reusable Geoclue-based geolocation provider to WebCore.
* GNUmakefile.am:
* GNUmakefile.list.am:
* platform/geoclue/GeolocationProviderGeoclue.cpp: Added.
(getPositionCallback):
(positionChangedCallback):
(GeolocationProviderGeoclue::GeolocationProviderGeoclue):
(GeolocationProviderGeoclue::~GeolocationProviderGeoclue):
(GeolocationProviderGeoclue::startUpdating):
(GeolocationProviderGeoclue::stopUpdating):
(GeolocationProviderGeoclue::setEnableHighAccuracy):
(GeolocationProviderGeoclue::setGeoclueClient):
(GeolocationProviderGeoclue::setGeocluePosition):
(GeolocationProviderGeoclue::updateClientRequirements):
(GeolocationProviderGeoclue::positionChanged):
(GeolocationProviderGeoclue::errorOccured):
* platform/geoclue/GeolocationProviderGeoclue.h: Added.
(WebCore):
(GeolocationProviderGeoclue):
* platform/geoclue/GeolocationProviderGeoclueClient.h: Added.
(WebCore):
(GeolocationProviderGeoclueClient):
2012-06-01 Thiago Marcos P. Santos <thiago.santos@intel.com>
[Qt] Remove deprecated Q_GLOBAL_STATIC_WITH_INITIALIZER
https://bugs.webkit.org/show_bug.cgi?id=88100
Reviewed by Tor Arne Vestbø.
It will be deprecated on Qt5.
* platform/text/qt/TextBreakIteratorInternalICUQt.cpp:
(WebCore):
2012-06-01 Zeno Albisser <zeno@webkit.org>
Make TextureMapper work with GraphicsSurface.
https://bugs.webkit.org/show_bug.cgi?id=87738
Add TextureMapperSurfaceBackingStore, a new backing store
that allows to import textures from a GraphicsSurface.
On Mac the GraphicsSurface is backed by an IOSurface
which must be used with a GL_TEXTURE_RECTANGLE_ARB texture.
Therefore it is also necessary to add new shader programs
for directly painting these textures on screen.
Reviewed by Noam Rosenthal.
* platform/graphics/texmap/TextureMapperBackingStore.cpp:
Add a new TextureMapperBackingStore implementation that can directly
reuse textures as they are being passed from the GraphicsSurface.
(WebCore::TextureMapperSurfaceBackingStore::setGraphicsSurface):
(WebCore::TextureMapperSurfaceBackingStore::texture):
(WebCore::TextureMapperSurfaceBackingStore::paintToTextureMapper):
* platform/graphics/texmap/TextureMapperBackingStore.h:
(GraphicsSurfaceData):
(WebCore::GraphicsSurfaceData::setSurface):
(WebCore::GraphicsSurfaceData::GraphicsSurfaceData):
(TextureMapperSurfaceBackingStore):
(WebCore::TextureMapperSurfaceBackingStore::create):
(WebCore::TextureMapperSurfaceBackingStore::~TextureMapperSurfaceBackingStore):
(WebCore::TextureMapperSurfaceBackingStore::TextureMapperSurfaceBackingStore):
* platform/graphics/texmap/TextureMapperGL.cpp:
Add a drawing function for textures of type GL_TEXTURE_RECTANGLE_ARB.
(WebCore::TextureMapperGL::drawTextureRectangleARB):
* platform/graphics/texmap/TextureMapperShaderManager.cpp:
(WebCore::TextureMapperShaderManager::getShaderProgram):
(WebCore::TextureMapperShaderProgramSimple::create):
(WebCore::TextureMapperShaderProgramSimple::initialize):
(WebCore::TextureMapperShaderProgramSolidColor::create):
(WebCore::TextureMapperShaderProgramSolidColor::initialize):
(WebCore::TextureMapperShaderProgramRectSimple::create):
(WebCore::TextureMapperShaderProgramRectSimple::fragmentShaderSource):
(WebCore::TextureMapperShaderProgramOpacityAndMask::create):
(WebCore::TextureMapperShaderProgramOpacityAndMask::initialize):
(WebCore::TextureMapperShaderProgramRectOpacityAndMask::create):
(WebCore::TextureMapperShaderProgramRectOpacityAndMask::fragmentShaderSource):
* platform/graphics/texmap/TextureMapperShaderManager.h:
Add new shader programs that can be used with GL_TEXTURE_RECTANGLE_ARB textures.
This is mainly necessary, because this type of texture uses non-normalized coordinates.
Further move the calls to initializeProgram() from the constructor into a separate
initialize() function, as initializeProgram() calls vertexShaderSource() (and friends)
and we should not call virtual functions in the constructor.
(WebCore::TextureMapperShaderProgram::initialize):
(TextureMapperShaderProgramSimple):
(WebCore::TextureMapperShaderProgramSimple::TextureMapperShaderProgramSimple):
(TextureMapperShaderProgramRectSimple):
(WebCore::TextureMapperShaderProgramRectSimple::TextureMapperShaderProgramRectSimple):
(TextureMapperShaderProgramOpacityAndMask):
(WebCore::TextureMapperShaderProgramOpacityAndMask::TextureMapperShaderProgramOpacityAndMask):
(TextureMapperShaderProgramRectOpacityAndMask):
(WebCore::TextureMapperShaderProgramRectOpacityAndMask::TextureMapperShaderProgramRectOpacityAndMask):
(TextureMapperShaderProgramSolidColor):
(WebCore::TextureMapperShaderProgramSolidColor::TextureMapperShaderProgramSolidColor):
2012-06-01 Christophe Dumez <christophe.dumez@intel.com>
[EFL] EFL port does not enable WEB_INTENTS_TAG flag
https://bugs.webkit.org/show_bug.cgi?id=86866
Reviewed by Adam Barth.
Fix compilation error when WEB_INTENTS_TAG flag is enabled.
HTMLElement::insertedInto() takes a ContainerNode* in argument, not a
Node*.
Test: webintents/intent-tag.html
* CMakeLists.txt:
* html/HTMLIntentElement.cpp:
(WebCore::HTMLIntentElement::insertedInto):
* html/HTMLIntentElement.h:
(HTMLIntentElement):
2012-06-01 Dominik Röttsches <dominik.rottsches@intel.com>
[cairo] Pixel artifacts can be seen on reflections
https://bugs.webkit.org/show_bug.cgi?id=85483
Antialiased clipping in the simple rectangular clip method
leads to edge artifacts when transformations are applied to the layer.
Explicitly disabling antialiased clipping for this function solves this issue.
Comparing Cairo GraphicsContext to Qt and Skia GC, it seems that these
backends do not expect the platform context to clip antialiased in this
case either.
Reviewed by Martin Robinson.
No new tests, covered by fast/css/transformed-mask.html.
* platform/graphics/cairo/GraphicsContextCairo.cpp:
(WebCore::GraphicsContext::clip):
2012-06-01 Florin Malita <fmalita@chromium.org>
http://shinydemos.com/clock/ doesn't seem to work
https://bugs.webkit.org/show_bug.cgi?id=79682
Reviewed by Nikolas Zimmermann.
Test: svg/repaint/text-mask-update.svg
Currently, parent resources are not invalidated when SVGInlineText nodes
are added or removed. Adjusting SVGResourcesCache::clientWasAddedToTree()
and SVGResourcesCache::clientWillBeRemovedFromTree() to cover this case.
* rendering/svg/SVGResourcesCache.cpp:
(WebCore::SVGResourcesCache::clientWasAddedToTree):
(WebCore::SVGResourcesCache::clientWillBeRemovedFromTree):
2012-06-01 John Mellor <johnme@chromium.org>
Don't set scroll position twice in HistoryController::restoreScrollPositionAndViewState
https://bugs.webkit.org/show_bug.cgi?id=88068
Reviewed by Adam Barth.
In the case where pageScaleFactor changes, we were calling
setScrollPosition with an incorrect scroll position (i.e. a scroll
position scaled by the pageScaleFactor we hadn't yet applied), then
fixing it by setting the pageScaleFactor and scroll position together,
overwriting the old scroll position. It's cleaner to just set the
pageScaleFactor and scroll position together.
No new tests as this isn't expected to change the ultimate behavior, just clean up how it happens.
* loader/HistoryController.cpp:
(WebCore::HistoryController::restoreScrollPositionAndViewState):
2012-06-01 Thiago Marcos P. Santos <thiago.santos@intel.com>
[Qt] Remove deprecated to/fromAscii()
https://bugs.webkit.org/show_bug.cgi?id=88086
Reviewed by Simon Hausmann.
Replacing to/fromAscii with to/fromLatin1 since it
is deprecated on Qt5.
* bridge/qt/qt_class.cpp:
(JSC::Bindings::QtClass::fallbackObject):
* platform/network/qt/ResourceRequestQt.cpp:
(WebCore::ResourceRequest::toNetworkRequest):
2012-06-01 Dan Bernstein <mitz@apple.com>
Layout not updated after setting -webkit-line-clamp to none
https://bugs.webkit.org/show_bug.cgi?id=88049
Reviewed by Abhishek Arya.
Test: fast/flexbox/line-clamp-removed-dynamically.html
* rendering/RenderDeprecatedFlexibleBox.cpp:
(WebCore::RenderDeprecatedFlexibleBox::styleWillChange): Added. Calls clearLineClamp if
line-clamp will change to none.
(WebCore::RenderDeprecatedFlexibleBox::clearLineClamp): Added. Marks possibly-clamped
children for layout and clears truncation from blocks.
* rendering/RenderDeprecatedFlexibleBox.h:
2012-06-01 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119213.
http://trac.webkit.org/changeset/119213
https://bugs.webkit.org/show_bug.cgi?id=88084
This patch broke two tests on all platform except Chromium.
The authors are unavailable. (Requested by zherczeg on
#webkit).
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
2012-06-01 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119219.
http://trac.webkit.org/changeset/119219
https://bugs.webkit.org/show_bug.cgi?id=88088
This patch broke two tests on GTK/Qt. The authors are
unavailable. (Requested by kkristof on #webkit).
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::closeAndRemoveChild):
(WebCore::FrameLoader::detachFromParent):
* page/Frame.cpp:
(WebCore::Frame::willDetachPage):
* page/Frame.h:
(Frame):
(WebCore::Frame::detachFromPage):
(WebCore):
* page/Page.cpp:
(WebCore::Page::~Page):
2012-06-01 Alexei Filippov <alexeif@chromium.org>
Web Inspector: Add Closure compiler annotations to WorkerConsole
https://bugs.webkit.org/show_bug.cgi?id=88073
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshotWorker.js:
(WebInspector.WorkerConsole.prototype.log):
(WebInspector.WorkerConsole.prototype.error):
(WebInspector.WorkerConsole.prototype.info):
2012-06-01 MORITA Hajime <morrita@google.com>
Frame::willDetachPage() shouldn't be called more than once.
https://bugs.webkit.org/show_bug.cgi?id=88056
Reviewed by Ryosuke Niwa.
This change moved willDetachPage() to private and calls it from
detachFromParent(). Also, it checks m_page and call
willDetachPage() only once, if m_page is available.
No new tests. Covered by existing suites.
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::closeAndRemoveChild):
(WebCore::FrameLoader::detachFromParent):
* page/Frame.cpp:
(WebCore::Frame::detachFromPage):
(WebCore):
(WebCore::Frame::willDetachPage):
* page/Frame.h:
(Frame):
* page/Page.cpp:
(WebCore::Page::~Page):
2012-06-01 Pierre Rossi <pierre.rossi@gmail.com>
[Qt] Support drawing a pattern with a translation.
https://bugs.webkit.org/show_bug.cgi?id=87025
Drawing an SVG pattern that has its x or y attribute set
works by setting a transform on the Pattern.
We would ignore this by only taking into account the texture and
target rect.
Reviewed by Kenneth Rohde Christiansen.
test: svg/W3C-SVG-1.1-SE/pservers-pattern-04-f.svg
* platform/graphics/qt/GraphicsContextQt.cpp:
(WebCore::drawRepeatPattern): refactored to take into account the translation set on the brush.
(WebCore::GraphicsContext::fillRect):
2012-06-01 Yoshifumi Inoue <yosin@chromium.org>
REGRESSION(r109729) [Form] Rendering of select/optgroup/option combination is too slow.
https://bugs.webkit.org/show_bug.cgi?id=88059
Reviewed by Kent Tamura.
This patch changes to share RenderStyle object among the "option"
elements to improve rendering performance and reducing memory usage
of RenderStyle.
No new tests. This patch doesn't change behavior but rendering performance.
* css/StyleResolver.cpp:
(WebCore::StyleResolver::canShareStyleWithElement): Check attribute value
mismatching for "option" element.
2012-06-01 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: [REGRESSION] Bad layout of "Override device metrics" controls in the Settings dialog
https://bugs.webkit.org/show_bug.cgi?id=88074
Reviewed by Yury Semikhatsky.
The "Override device metrics" controls are placed in a table, which gets too narrow. Give it a
"white-space: nowrap" to avoid wrapping individual cells.
* inspector/front-end/SettingsScreen.js:
(WebInspector.SettingsScreen.prototype._createDeviceMetricsElement):
* inspector/front-end/elementsPanel.css:
* inspector/front-end/inspector.css:
(.nowrap):
2012-06-01 Sudarsana Nagineni <sudarsana.nagineni@linux.intel.com>
[EFL] Implement PlatformStrategies
https://bugs.webkit.org/show_bug.cgi?id=86946
Reviewed by Carlos Garcia Campos.
* PlatformEfl.cmake: Add PlatformStrategies.cpp file to the build system.
Also, remove PluginDataEfl.cpp because the same functionality is now
implemented in PlatformStrategiesEfl.cpp.
2012-06-01 Adam Barth <abarth@webkit.org>
ASSERT(m_fontSelector->document()) is bogus and should be removed
https://bugs.webkit.org/show_bug.cgi?id=88053
Reviewed by Abhishek Arya.
This ASSERT was introduced in http://trac.webkit.org/changeset/97402
together with a branch that handled the case of the ASSERT firing.
This ASSERT fires when running tests on Android (which runs with
ASSERTs enabled). The ASSERT appears to be bogus, so this patch removes
it.
* css/CSSSegmentedFontFace.cpp:
(WebCore::CSSSegmentedFontFace::getFontData):
2012-06-01 Yoshifumi Inoue <yosin@chromium.org>
[Platform][Decimal] UInt128::operator/= calls makeUInt128 with wrong argument order
https://bugs.webkit.org/show_bug.cgi?id=88044
Reviewed by Kent Tamura.
This patch fixed wrong argument of makeUInt128 in UInt128::operator/= to get right
result for decimal multiplication.
Test: WebKit/chromium/tests/DecimalTest.cpp: Add new a new test case.
* platform/Decimal.cpp:
(WebCore::DecimalPrivate::UInt128::operator/=):
2012-06-01 Adam Barth <abarth@webkit.org>
sandbox directive in X-WebKit-CSP header unable to create a unique origin
https://bugs.webkit.org/show_bug.cgi?id=88014
Reviewed by Ryosuke Niwa.
We process the HTTP headers for a response after we create a document
object for the response. Previously, the SecurityOrigin of a document
was determined when the document was created, which meant that the
sandbox directive in CSP couldn't create a unique origin.
In this patch, we transition to a unique origin when we start enforcing
the SandboxOrigin bit. This patch is more complicated than you might
expect because we redundantly store the SecurityOrigin object in
DOMWindow. Removing that redundant state is
https://bugs.webkit.org/show_bug.cgi?id=75793.
The CSP sandbox directive is defined in
http://dvcs.w3.org/hg/content-security-policy/raw-file/tip/csp-1.0-specification.html
by reference to
http://www.whatwg.org/specs/web-apps/current-work/#forced-sandboxing-flag-set.
The relation between sandbox and unique origin arises from
http://www.whatwg.org/specs/web-apps/current-work/#origin-0.
Test: http/tests/security/contentSecurityPolicy/sandbox-allow-scripts-in-http-header-control.html
http/tests/security/contentSecurityPolicy/sandbox-allow-scripts-in-http-header.html
http/tests/security/contentSecurityPolicy/sandbox-in-http-header-control.html
http/tests/security/contentSecurityPolicy/sandbox-in-http-header.html
* dom/Document.cpp:
(WebCore::Document::didUpdateSecurityOrigin):
(WebCore::Document::initContentSecurityPolicy):
* dom/Document.h:
(Document):
* dom/SecurityContext.cpp:
(WebCore::SecurityContext::enforceSandboxFlags):
(WebCore):
(WebCore::SecurityContext::didUpdateSecurityOrigin):
* dom/SecurityContext.h:
(SecurityContext):
2012-05-31 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: expose MemoryCache statistics in the inspector protocol
https://bugs.webkit.org/show_bug.cgi?id=87984
Reviewed by Pavel Feldman.
Added MemoryCache statistics to Memory.getProcessMemoryDistribution command
result. Corresponding sector is added to the memory pie-chart.
* inspector/InspectorMemoryAgent.cpp:
(MemoryBlockName):
(WebCore):
(WebCore::addMemoryBlockFor):
(WebCore::memoryCacheInfo):
(WebCore::InspectorMemoryAgent::getProcessMemoryDistribution):
* inspector/front-end/NativeMemorySnapshotView.js:
(WebInspector.MemoryBlockViewProperties._initialize):
2012-05-31 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: add RenderArena sizes to the memory pie-chart
https://bugs.webkit.org/show_bug.cgi?id=87971
Reviewed by Pavel Feldman.
Added allocated and used sizes of page render arenas to the memory
report returned by the InspectorMemoryAgent. Total allocated size of
the inspected page's render arenas will be displayed on the memory
pie-chart.
* inspector/InspectorMemoryAgent.cpp:
(MemoryBlockName):
(WebCore):
(WebCore::renderTreeInfo):
(WebCore::InspectorMemoryAgent::getProcessMemoryDistribution):
* inspector/front-end/NativeMemorySnapshotView.js:
(WebInspector.MemoryBlockViewProperties._initialize):
2012-05-30 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: change type of injected script id from long to int
https://bugs.webkit.org/show_bug.cgi?id=87837
Reviewed by Pavel Feldman.
Changed injected script id type from long to int so that we don't
lose precision when sending the id over the protocol.
* bindings/js/JSInjectedScriptManager.cpp:
(WebCore::InjectedScriptManager::createInjectedScript):
(WebCore::InjectedScriptManager::injectedScriptFor):
* bindings/v8/custom/V8InjectedScriptManager.cpp:
(WebCore::InjectedScriptManager::createInjectedScript):
(WebCore::InjectedScriptManager::injectedScriptFor):
* inspector/InjectedScriptManager.cpp:
(WebCore::InjectedScriptManager::injectedScriptForId):
(WebCore::InjectedScriptManager::injectedScriptIdFor):
(WebCore::InjectedScriptManager::injectScript):
* inspector/InjectedScriptManager.h:
(InjectedScriptManager):
* inspector/PageRuntimeAgent.cpp:
(WebCore::PageRuntimeAgent::notifyContextCreated):
2012-05-29 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: draw pie-chart based on memory data received from backend
https://bugs.webkit.org/show_bug.cgi?id=87737
Reviewed by Pavel Feldman.
Added pie chart for memory data received from inspector memory agent.
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* inspector/front-end/NativeMemorySnapshotView.js:
(WebInspector.NativeMemorySnapshotView):
(WebInspector.NativeMemoryProfileType):
(WebInspector.NativeMemoryProfileType.prototype.buttonClicked.didReceiveMemorySnapshot):
(WebInspector.NativeMemoryProfileType.prototype.buttonClicked):
(WebInspector.NativeMemoryProfileHeader):
(WebInspector.MemoryBlockViewProperties):
(WebInspector.MemoryBlockViewProperties._initialize):
(WebInspector.MemoryBlockViewProperties._forMemoryBlock):
(WebInspector.NativeMemoryPieChart):
(WebInspector.NativeMemoryPieChart.prototype.onResize):
(WebInspector.NativeMemoryPieChart.prototype._updateSize):
(WebInspector.NativeMemoryPieChart.prototype._addBlockLabels):
(WebInspector.NativeMemoryPieChart.prototype._paint.paintPercentAndLabel):
(WebInspector.NativeMemoryPieChart.prototype._paint):
(WebInspector.NativeMemoryPieChart.prototype._clear):
* inspector/front-end/WebKit.qrc:
* inspector/front-end/nativeMemoryProfiler.css: Added.
(.memory-pie-chart-container):
(.memory-pie-chart):
(.memory-blocks-list .swatch):
(.memory-blocks-list):
(.memory-blocks-list .item):
2012-05-31 Dominic Cooney <dominicc@chromium.org>
[V8] Expando properties on attribute nodes disappear
https://bugs.webkit.org/show_bug.cgi?id=87925
Reviewed by Adam Barth.
Test: fast/dom/gc-attribute-node.html
* bindings/v8/V8GCController.cpp:
(WebCore::calculateGroupId): Group attributes like other nodes.
2012-05-31 Alexander Shalamov <alexander.shalamov@intel.com>
[EFL] <input type="number"> is not a spinbutton
https://bugs.webkit.org/show_bug.cgi?id=86846
Reviewed by Kenneth Rohde Christiansen.
Implemented adjustInnerSpinButtonStyle and
paintInnerSpinButton in RenderThemeEfl.
* platform/efl/RenderThemeEfl.cpp:
(WebCore::RenderThemeEfl::applyEdjeStateFromForm):
(WebCore::RenderThemeEfl::edjeGroupFromFormType):
(WebCore::RenderThemeEfl::adjustInnerSpinButtonStyle):
(WebCore):
(WebCore::RenderThemeEfl::paintInnerSpinButton):
* platform/efl/RenderThemeEfl.h:
(RenderThemeEfl):
2012-05-31 Hajime Morrita <morrita@chromium.org>
REGRESSION(r117572): editing/spelling/spellcheck-async-remove-frame.html crashes on Mac
https://bugs.webkit.org/show_bug.cgi?id=86859
Reviewed by Ryosuke Niwa.
The test tries to reach an invalid SpellChecker object. Such an access should be guarded
beforehand.
Asynchronous spellchecking can return results after originated
frame is gone, which triggered an invalid access to the dead spellchecker
object. This chagne prevents it by marking request objects from
the spellchecker as invalid:
- Originally TextCheckerClient API was passed a SpellCheker object.
This change abstracted it behind TextCheckingRequest interface,
didSucceed() and didCancel() method specifically.
- TextCheckingRequest was turned from a plain old object into
a refcounted abstract class, which is now subclassed by SpellCheckRequest.
- SpellChecker now marks pending SpellCheckRequest objects as invalid
on its destructor.
Test: editing/spelling/spellcheck-async-remove-frame.html
* WebCore.exp.in:
* editing/SpellChecker.cpp:
(WebCore::SpellCheckRequest::SpellCheckRequest):
(WebCore::SpellCheckRequest::create):
(WebCore::SpellCheckRequest::didSucceed):
(WebCore):
(WebCore::SpellCheckRequest::didCancel):
(WebCore::SpellCheckRequest::wasRequestedBy):
(WebCore::SpellCheckRequest::requesterDestroyed):
(WebCore::SpellChecker::~SpellChecker):
(WebCore::SpellChecker::requestCheckingFor):
(WebCore::SpellChecker::invokeRequest):
(WebCore::SpellChecker::didCheckSucceed):
(WebCore::SpellChecker::didCheckCancel):
* editing/SpellChecker.h:
(WebCore):
(SpellCheckRequest):
(WebCore::SpellCheckRequest::isStarted):
(SpellChecker):
* loader/EmptyClients.h:
(WebCore::EmptyTextCheckerClient::requestCheckingOfString):
* platform/text/TextCheckerClient.h:
(TextCheckerClient):
* platform/text/TextChecking.h:
(GrammarDetail):
(TextCheckingResult):
(TextCheckingRequest):
(WebCore::TextCheckingRequest::~TextCheckingRequest):
2012-05-31 Tony Chang <tony@chromium.org>
initial flex value should be 1 1 auto
https://bugs.webkit.org/show_bug.cgi?id=86526
Reviewed by Ojan Vafai.
Also, if -webkit-flex is used to set the positive flex to 0 and
the negative flex is omitted, the negative flex is set to 0.
New test cases in: flex-property-parsing.html
* css/CSSParser.cpp:
(WebCore::CSSParser::parseFlex):
* css/StyleBuilder.cpp:
(WebCore::ApplyPropertyFlex::applyValue):
* rendering/style/RenderStyle.h:
2012-05-31 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119146.
http://trac.webkit.org/changeset/119146
https://bugs.webkit.org/show_bug.cgi?id=88035
android breakage fixed in http://crrev.com/139945 (Requested
by fischman on #webkit).
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::HTMLMediaElement):
(WebCore::HTMLMediaElement::startProgressEventTimer):
(WebCore::HTMLMediaElement::changeNetworkStateFromLoadingToIdle):
(WebCore::HTMLMediaElement::progressEventTimerFired):
(WebCore::HTMLMediaElement::startPlaybackProgressTimer):
* platform/graphics/MediaPlayer.cpp:
(WebCore::NullMediaPlayerPrivate::didLoadingProgress):
(WebCore::MediaPlayer::didLoadingProgress):
* platform/graphics/MediaPlayer.h:
* platform/graphics/MediaPlayerPrivate.h:
(MediaPlayerPrivateInterface):
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
(WebCore::MediaPlayerPrivateAVFoundation::MediaPlayerPrivateAVFoundation):
(WebCore::MediaPlayerPrivateAVFoundation::didLoadingProgress):
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:
(MediaPlayerPrivateAVFoundation):
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp:
(WebCore::MediaPlayerPrivate::didLoadingProgress):
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.h:
(MediaPlayerPrivate):
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
(WebCore::MediaPlayerPrivateGStreamer::MediaPlayerPrivateGStreamer):
(WebCore::MediaPlayerPrivateGStreamer::didLoadingProgress):
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
(MediaPlayerPrivateGStreamer):
* platform/graphics/mac/MediaPlayerPrivateQTKit.h:
(MediaPlayerPrivateQTKit):
* platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
(WebCore::MediaPlayerPrivateQTKit::MediaPlayerPrivateQTKit):
(WebCore::MediaPlayerPrivateQTKit::didLoadingProgress):
* platform/graphics/qt/MediaPlayerPrivateQt.cpp:
(WebCore::MediaPlayerPrivateQt::MediaPlayerPrivateQt):
(WebCore::MediaPlayerPrivateQt::didLoadingProgress):
* platform/graphics/qt/MediaPlayerPrivateQt.h:
(MediaPlayerPrivateQt):
* platform/graphics/win/MediaPlayerPrivateQuickTimeVisualContext.cpp:
(WebCore::MediaPlayerPrivateQuickTimeVisualContext::MediaPlayerPrivateQuickTimeVisualContext):
(WebCore::MediaPlayerPrivateQuickTimeVisualContext::didLoadingProgress):
* platform/graphics/win/MediaPlayerPrivateQuickTimeVisualContext.h:
(MediaPlayerPrivateQuickTimeVisualContext):
* platform/graphics/wince/MediaPlayerPrivateWinCE.h:
(MediaPlayerPrivate):
2012-05-31 Tom Sepez <tsepez@chromium.org>
XSSAuditor bypass with leading /*///*/ comment
https://bugs.webkit.org/show_bug.cgi?id=88002
Reviewed by Adam Barth.
Fixes issue in xssauditor's parsing of /*/.
Test: http/tests/security/xssAuditor/script-tag-with-trailing-comment4.html
* html/parser/XSSAuditor.cpp:
(WebCore::XSSAuditor::decodedSnippetForJavaScript):
2012-05-31 Hans Muller <hmuller@adobe.com>
Bug: Negative SVG rect rx,ry corner radii values aren't handled correctly
https://bugs.webkit.org/show_bug.cgi?id=87859
Reviewed by Dirk Schulze.
Added the rect element corner radius constraints specified in steps 2-4 of
http://www.w3.org/TR/SVG/shapes.html#RectElement to Path::addRoundedRect().
Support for steps 6 and 7 was already included. Steps 2-4:
2 - If neither rx nor ry are properly specified, then set both rx and ry to 0. (This will result in square corners.)
3 - Otherwise, if a properly specified value is provided for rx, but not for ry then set both rx and ry to the value of rx.
4 - Otherwise, if a properly specified value is provided for ry, but not for rx, then set both rx and ry to the value of ry.
Tests: svg/custom/rect-negative-corner-radii-expected.svg
svg/custom/rect-negative-corner-radii.svg
* platform/graphics/Path.cpp:
(WebCore::Path::addRoundedRect):
2012-05-31 Shawn Singh <shawnsingh@chromium.org>
[chromium] Migrate to WebTransformationMatrix
https://bugs.webkit.org/show_bug.cgi?id=87788
Reviewed by James Robinson.
Covered by all existing layout tests and unit tests.
This patch replaces all occurrences of WebCore::TransformationMatrix
with WebKit::WebTransformationMatrix in the chromium compositor code.
* platform/chromium/support/WebTransformationMatrix.cpp:
(WebKit::WebTransformationMatrix::WebTransformationMatrix):
(WebKit):
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::setSublayerTransform):
(WebCore::LayerChromium::setTransform):
(WebCore::LayerChromium::setTransformFromAnimation):
* platform/graphics/chromium/LayerChromium.h:
(LayerChromium):
(WebCore::LayerChromium::sublayerTransform):
(WebCore::LayerChromium::drawTransform):
(WebCore::LayerChromium::setDrawTransform):
(WebCore::LayerChromium::screenSpaceTransform):
(WebCore::LayerChromium::setScreenSpaceTransform):
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::LayerRendererChromium::drawCheckerboardQuad):
(WebCore::LayerRendererChromium::drawDebugBorderQuad):
(WebCore::LayerRendererChromium::drawBackgroundFilters):
(WebCore::LayerRendererChromium::drawRenderSurfaceQuad):
(WebCore::LayerRendererChromium::drawSolidColorQuad):
(WebCore::LayerRendererChromium::drawTileQuad):
(WebCore::LayerRendererChromium::drawHeadsUpDisplay):
(WebCore::LayerRendererChromium::toGLMatrix):
(WebCore::LayerRendererChromium::drawTexturedQuad):
(WebCore::LayerRendererChromium::copyTextureToFramebuffer):
* platform/graphics/chromium/LayerRendererChromium.h:
(WebCore::LayerRendererChromium::projectionMatrix):
(WebCore::LayerRendererChromium::windowMatrix):
(LayerRendererChromium):
* platform/graphics/chromium/LinkHighlight.cpp:
(WebCore::LinkHighlight::LinkHighlight):
* platform/graphics/chromium/RenderSurfaceChromium.h:
(WebCore::RenderSurfaceChromium::drawTransform):
(WebCore::RenderSurfaceChromium::setDrawTransform):
(WebCore::RenderSurfaceChromium::originTransform):
(WebCore::RenderSurfaceChromium::setOriginTransform):
(WebCore::RenderSurfaceChromium::screenSpaceTransform):
(WebCore::RenderSurfaceChromium::setScreenSpaceTransform):
(WebCore::RenderSurfaceChromium::replicaDrawTransform):
(WebCore::RenderSurfaceChromium::setReplicaDrawTransform):
(WebCore::RenderSurfaceChromium::replicaOriginTransform):
(WebCore::RenderSurfaceChromium::setReplicaOriginTransform):
(WebCore::RenderSurfaceChromium::replicaScreenSpaceTransform):
(WebCore::RenderSurfaceChromium::setReplicaScreenSpaceTransform):
(RenderSurfaceChromium):
* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::TiledLayerChromium::updateTiles):
* platform/graphics/chromium/cc/CCAnimationCurve.h:
(WebCore):
(CCTransformAnimationCurve):
* platform/graphics/chromium/cc/CCDamageTracker.cpp:
(WebCore::CCDamageTracker::extendDamageForLayer):
(WebCore::CCDamageTracker::extendDamageForRenderSurface):
* platform/graphics/chromium/cc/CCDrawQuad.h:
(WebCore::CCDrawQuad::quadTransform):
(WebCore::CCDrawQuad::layerTransform):
* platform/graphics/chromium/cc/CCKeyframedAnimationCurve.cpp:
(WebCore::CCTransformKeyframe::clone):
(WebCore::CCKeyframedTransformAnimationCurve::getValue):
* platform/graphics/chromium/cc/CCKeyframedAnimationCurve.h:
(CCKeyframedTransformAnimationCurve):
* platform/graphics/chromium/cc/CCLayerAnimationController.cpp:
(WebCore::CCLayerAnimationController::tickAnimations):
* platform/graphics/chromium/cc/CCLayerAnimationController.h:
(WebKit):
(WebCore):
(CCLayerAnimationControllerClient):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::quadTransform):
(WebCore::CCLayerImpl::setTransformFromAnimation):
(WebCore::CCLayerImpl::setSublayerTransform):
(WebCore::CCLayerImpl::setTransform):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(CCLayerImpl):
(WebCore::CCLayerImpl::sublayerTransform):
(WebCore::CCLayerImpl::drawTransform):
(WebCore::CCLayerImpl::setDrawTransform):
(WebCore::CCLayerImpl::screenSpaceTransform):
(WebCore::CCLayerImpl::setScreenSpaceTransform):
* platform/graphics/chromium/cc/CCLayerSorter.cpp:
(WebCore::CCLayerSorter::LayerShape::LayerShape):
(WebCore::CCLayerSorter::createGraphNodes):
* platform/graphics/chromium/cc/CCLayerSorter.h:
(WebKit):
(LayerShape):
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::updateLayers):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::CCLayerTreeHostCommon::calculateVisibleRect):
(WebCore::isSurfaceBackFaceVisible):
(WebCore::calculateVisibleLayerRect):
(WebCore::isScaleOrTranslation):
(WebCore::calculateDrawTransformsAndVisibilityInternal):
(WebCore::CCLayerTreeHostCommon::calculateDrawTransformsAndVisibility):
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.h:
(CCLayerTreeHostCommon):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::damageInSurfaceSpace):
(WebCore::CCLayerTreeHostImpl::calculateRenderSurfaceLayerList):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(WebCore):
* platform/graphics/chromium/cc/CCMathUtil.cpp:
(WebCore::projectPoint):
(WebCore::mapPoint):
(WebCore::CCMathUtil::mapClippedRect):
(WebCore::CCMathUtil::projectClippedRect):
(WebCore::CCMathUtil::mapClippedQuad):
(WebCore::CCMathUtil::mapQuad):
(WebCore::CCMathUtil::projectQuad):
* platform/graphics/chromium/cc/CCMathUtil.h:
(WebKit):
(WebCore):
(CCMathUtil):
* platform/graphics/chromium/cc/CCOcclusionTracker.cpp:
(WebCore::transformSurfaceOpaqueRegion):
(WebCore::reduceOcclusionBelowSurface):
(WebCore::contentToScreenSpaceTransform):
(WebCore::contentToTargetSurfaceTransform):
(WebCore::addOcclusionBehindLayer):
(WebCore::::markOccludedBehindLayer):
(WebCore::testContentRectOccluded):
(WebCore::computeUnoccludedContentRect):
(WebCore::::unoccludedContributingSurfaceContentRect):
* platform/graphics/chromium/cc/CCOcclusionTracker.h:
* platform/graphics/chromium/cc/CCOverdrawMetrics.cpp:
(WebCore::CCOverdrawMetrics::didUpload):
(WebCore::CCOverdrawMetrics::didCullForDrawing):
(WebCore::CCOverdrawMetrics::didDraw):
* platform/graphics/chromium/cc/CCOverdrawMetrics.h:
(WebKit):
(CCOverdrawMetrics):
* platform/graphics/chromium/cc/CCQuadCuller.cpp:
* platform/graphics/chromium/cc/CCRenderPass.cpp:
(WebCore::CCRenderPass::appendQuadsToFillScreen):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(WebCore::CCRenderSurface::setDrawTransform):
(WebCore::CCRenderSurface::drawTransform):
(WebCore::CCRenderSurface::setOriginTransform):
(WebCore::CCRenderSurface::originTransform):
(WebCore::CCRenderSurface::setScreenSpaceTransform):
(WebCore::CCRenderSurface::screenSpaceTransform):
(WebCore::CCRenderSurface::setReplicaDrawTransform):
(WebCore::CCRenderSurface::replicaDrawTransform):
(WebCore::CCRenderSurface::setReplicaOriginTransform):
(WebCore::CCRenderSurface::replicaOriginTransform):
(WebCore::CCRenderSurface::setReplicaScreenSpaceTransform):
(WebCore::CCRenderSurface::replicaScreenSpaceTransform):
(CCRenderSurface):
* platform/graphics/chromium/cc/CCSharedQuadState.cpp:
(WebCore::CCSharedQuadState::create):
(WebCore::CCSharedQuadState::CCSharedQuadState):
* platform/graphics/chromium/cc/CCSharedQuadState.h:
(CCSharedQuadState):
(WebCore::CCSharedQuadState::quadTransform):
(WebCore::CCSharedQuadState::layerTransform):
* platform/graphics/chromium/cc/CCSolidColorLayerImpl.cpp:
(WebCore::CCSolidColorLayerImpl::quadTransform):
* platform/graphics/chromium/cc/CCSolidColorLayerImpl.h:
(CCSolidColorLayerImpl):
* platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
(WebCore::CCTiledLayerImpl::quadTransform):
* platform/graphics/chromium/cc/CCTiledLayerImpl.h:
(CCTiledLayerImpl):
2012-05-31 Ian Vollick <vollick@chromium.org>
[chromium] Single thread proxy should not tick animations unless the layer renderer has been initialized
https://bugs.webkit.org/show_bug.cgi?id=87873
Reviewed by James Robinson.
When the layer renderer fails to initialize, be sure to stop the animation timer.
Unit test: CCLayerTreeHostTestInitializeLayerRendererFailsAfterAddAnimation.runSingleThread
* platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
(WebCore::CCSingleThreadProxy::initializeLayerRenderer):
(WebCore::CCSingleThreadProxy::didAddAnimation):
(WebCore::CCSingleThreadProxy::animationTimerDelay):
(WebCore):
* platform/graphics/chromium/cc/CCSingleThreadProxy.h:
2012-05-31 Simon Fraser <simon.fraser@apple.com>
RenderLayerCompositor cleanup: make RenderGeometryMap part of the OverlapMap
https://bugs.webkit.org/show_bug.cgi?id=88021
Reviewed by James Robinson.
We only ever use the RenderGeometryMap when we have an OverlapMap, so make
it a member of the OverlapMap.
No behavior change.
* rendering/RenderLayerCompositor.cpp:
(RenderLayerCompositor::OverlapMap):
(WebCore::RenderLayerCompositor::OverlapMap::geometryMap):
(WebCore::RenderLayerCompositor::updateCompositingLayers):
(WebCore::RenderLayerCompositor::addToOverlapMap):
(WebCore::RenderLayerCompositor::addToOverlapMapRecursive):
(WebCore::RenderLayerCompositor::computeCompositingRequirements):
* rendering/RenderLayerCompositor.h:
(WebCore):
(RenderLayerCompositor):
2012-05-31 Kentaro Hara <haraken@chromium.org>
[V8] Pass Isolate to v8Boolean()
https://bugs.webkit.org/show_bug.cgi?id=87948
Reviewed by Adam Barth.
The objective is to pass Isolate around in V8 bindings.
This patch passes Isolate to v8Boolean().
No tests. No change in behavior.
* bindings/v8/SerializedScriptValue.cpp:
* bindings/v8/custom/V8DOMStringMapCustom.cpp:
(WebCore::V8DOMStringMap::namedPropertyDeleter):
* bindings/v8/custom/V8InjectedScriptHostCustom.cpp:
(WebCore::V8InjectedScriptHost::isHTMLAllCollectionCallback):
* bindings/v8/custom/V8StorageCustom.cpp:
(WebCore::storageDeleter):
* bindings/v8/custom/V8WebSocketCustom.cpp:
(WebCore::V8WebSocket::sendCallback):
2012-05-31 Alec Flett <alecflett@chromium.org>
IndexedDB: Implement IDBTransaction.error and IDBRequest.error
https://bugs.webkit.org/show_bug.cgi?id=87865
Reviewed by Tony Chang.
Added "error" attribute to IDBRequest and IDBTransaction. Update
IDBDatabaseError to honor the IDBDatabaseException behavior
of dealing with IDB-specific throws of DOMException codes,
as per the spec.
Existing tests which previously tested 'errorCode' and
'webkitErrorMessage' have been updated to use the new attribute.
* Modules/indexeddb/IDBCursor.cpp:
(WebCore::IDBCursor::advance):
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::setVersion):
* Modules/indexeddb/IDBDatabaseError.h:
(WebCore::IDBDatabaseError::create):
(WebCore::IDBDatabaseError::code):
(WebCore::IDBDatabaseError::idbCode):
(WebCore::IDBDatabaseError::name):
(WebCore::IDBDatabaseError::IDBDatabaseError):
(IDBDatabaseError):
* Modules/indexeddb/IDBDatabaseException.cpp:
(WebCore):
(WebCore::getErrorEntry):
(WebCore::IDBDatabaseException::initializeDescription):
(WebCore::IDBDatabaseException::getErrorName):
(WebCore::IDBDatabaseException::getLegacyErrorCode):
* Modules/indexeddb/IDBDatabaseException.h:
(IDBDatabaseException):
* Modules/indexeddb/IDBObjectStore.cpp:
(WebCore::IDBObjectStore::createIndex):
* Modules/indexeddb/IDBRequest.cpp:
(WebCore::IDBRequest::error):
(WebCore):
(WebCore::IDBRequest::resetReadyState):
(WebCore::IDBRequest::abort):
(WebCore::IDBRequest::onError):
(WebCore::IDBRequest::onSuccess):
(WebCore::IDBRequest::onSuccessWithContinuation):
(WebCore::IDBRequest::dispatchEvent):
* Modules/indexeddb/IDBRequest.h:
(IDBRequest):
* Modules/indexeddb/IDBRequest.idl:
* Modules/indexeddb/IDBTransaction.cpp:
(WebCore::IDBTransaction::error):
(WebCore):
(WebCore::IDBTransaction::setError):
* Modules/indexeddb/IDBTransaction.h:
(IDBTransaction):
* Modules/indexeddb/IDBTransaction.idl:
2012-05-31 James Robinson <jamesr@chromium.org>
[chromium] Assertion failures during compositor startup in lost context situations
https://bugs.webkit.org/show_bug.cgi?id=87912
Reviewed by Adrienne Walker.
getShaderiv and getProgramiv may return 0 if the context is lost. We correctly recover in this case, so it's
bogus to fail an ASSERT in debug in this case.
Added new unit test in LayerRendererChromiumTest to cover this.
* platform/graphics/chromium/ProgramBinding.cpp:
(WebCore::contextLost):
(WebCore):
(WebCore::ProgramBindingBase::init):
(WebCore::ProgramBindingBase::createShaderProgram):
2012-05-31 Ben Murdoch <benm@google.com>
Fix crash in V8Document::createTouchListCallback.
https://bugs.webkit.org/show_bug.cgi?id=87085
Reviewed by Abhishek Arya.
Test: fast/events/touch/document-create-touch-list-crash.html
* bindings/v8/custom/V8DocumentCustom.cpp:
(WebCore::V8Document::createTouchListCallback): Verify the native type
of the arguments passed to createTouchList are Touches before appending
them to the TouchList. In the case of a non-Touch argument, insert
null into the TouchList. This consolidates V8 and JSC bindings
behaviour.
* dom/Document.cpp: Remove dead code.
* dom/Document.h: ditto.
2012-05-31 Simon Fraser <simon.fraser@apple.com>
Remove some duplicate code in RenderLayerCompositor
https://bugs.webkit.org/show_bug.cgi?id=88009
Reviewed by Anders Carlsson.
RenderLayerCompositor::hasNonAffineTransform() has the same code
as the existing RenderLayer::has3DTransform(), so just use the
RenderLayer method and remove hasNonAffineTransform().
No behavior change.
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::computeCompositingRequirements):
* rendering/RenderLayerCompositor.h:
(RenderLayerCompositor):
2012-05-31 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r119125.
http://trac.webkit.org/changeset/119125
https://bugs.webkit.org/show_bug.cgi?id=88007
Will break android build if rolled (Requested by rafaelw_ on
#webkit).
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::HTMLMediaElement):
(WebCore::HTMLMediaElement::startProgressEventTimer):
(WebCore::HTMLMediaElement::changeNetworkStateFromLoadingToIdle):
(WebCore::HTMLMediaElement::progressEventTimerFired):
(WebCore::HTMLMediaElement::startPlaybackProgressTimer):
* platform/graphics/MediaPlayer.cpp:
(WebCore::NullMediaPlayerPrivate::bytesLoaded):
(WebCore::MediaPlayer::bytesLoaded):
* platform/graphics/MediaPlayer.h:
* platform/graphics/MediaPlayerPrivate.h:
(MediaPlayerPrivateInterface):
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
(WebCore::MediaPlayerPrivateAVFoundation::MediaPlayerPrivateAVFoundation):
(WebCore::MediaPlayerPrivateAVFoundation::bytesLoaded):
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:
(MediaPlayerPrivateAVFoundation):
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp:
(WebCore::MediaPlayerPrivate::bytesLoaded):
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.h:
(MediaPlayerPrivate):
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
(WebCore::MediaPlayerPrivateGStreamer::MediaPlayerPrivateGStreamer):
(WebCore::MediaPlayerPrivateGStreamer::bytesLoaded):
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
(MediaPlayerPrivateGStreamer):
* platform/graphics/mac/MediaPlayerPrivateQTKit.h:
(MediaPlayerPrivateQTKit):
* platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
(WebCore::MediaPlayerPrivateQTKit::MediaPlayerPrivateQTKit):
(WebCore::MediaPlayerPrivateQTKit::bytesLoaded):
* platform/graphics/qt/MediaPlayerPrivateQt.cpp:
(WebCore::MediaPlayerPrivateQt::MediaPlayerPrivateQt):
(WebCore::MediaPlayerPrivateQt::bytesLoaded):
* platform/graphics/qt/MediaPlayerPrivateQt.h:
(MediaPlayerPrivateQt):
* platform/graphics/win/MediaPlayerPrivateQuickTimeVisualContext.cpp:
(WebCore::MediaPlayerPrivateQuickTimeVisualContext::MediaPlayerPrivateQuickTimeVisualContext):
(WebCore::MediaPlayerPrivateQuickTimeVisualContext::bytesLoaded):
* platform/graphics/win/MediaPlayerPrivateQuickTimeVisualContext.h:
(MediaPlayerPrivateQuickTimeVisualContext):
* platform/graphics/wince/MediaPlayerPrivateWinCE.h:
(MediaPlayerPrivate):
2012-05-31 Andy Estes <aestes@apple.com>
Disentangle code that relies on USE(AUTOMATIC_TEXT_REPLACEMENT) from that which merely relies on PLATFORM(MAC)
https://bugs.webkit.org/show_bug.cgi?id=87933
Reviewed by Dan Bernstein.
* editing/Editor.cpp:
* editing/Editor.h:
* loader/EmptyClients.h:
* page/ContextMenuController.cpp:
(WebCore::ContextMenuController::contextMenuItemSelected):
* page/EditorClient.h:
2012-05-31 Dana Jansens <danakj@chromium.org>
[chromium] Move drawing code for RenderSurfaces into LayerRendererChromium
https://bugs.webkit.org/show_bug.cgi?id=87877
Reviewed by James Robinson.
We add data to CCRenderSurfaceDrawQuad so that the only use of
CCRenderSurface directly from LayerRendererChromium is to
reserve, release, and use the surface's textures, and to set the
scissor rect. Each of these changes will be done independent
of this change.
Covered by existing tests.
* platform/graphics/chromium/LayerRendererChromium.cpp:
(WebCore::applyFilters):
(WebCore):
(WebCore::LayerRendererChromium::drawBackgroundFilters):
(WebCore::LayerRendererChromium::drawRenderSurfaceQuad):
(WebCore::LayerRendererChromium::copyTextureToFramebuffer):
* platform/graphics/chromium/LayerRendererChromium.h:
(LayerRendererChromium):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::contentsTextureId):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::calculateDrawTransformsAndVisibilityInternal):
* platform/graphics/chromium/cc/CCQuadCuller.cpp:
* platform/graphics/chromium/cc/CCQuadCuller.h:
(CCQuadCuller):
* platform/graphics/chromium/cc/CCRenderPass.cpp:
(WebCore::CCRenderPass::appendQuadsForRenderSurfaceLayer):
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore):
(WebCore::CCRenderSurface::CCRenderSurface):
(WebCore::CCRenderSurface::prepareContentsTexture):
(WebCore::CCRenderSurface::releaseContentsTexture):
(WebCore::CCRenderSurface::hasValidContentsTexture):
(WebCore::CCRenderSurface::releaseBackgroundTexture):
(WebCore::CCRenderSurface::hasValidBackgroundTexture):
(WebCore::CCRenderSurface::hasMask):
(WebCore::CCRenderSurface::replicaHasMask):
(WebCore::CCRenderSurface::appendQuads):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(WebCore):
(CCRenderSurface):
* platform/graphics/chromium/cc/CCRenderSurfaceDrawQuad.cpp:
(WebCore::CCRenderSurfaceDrawQuad::create):
(WebCore::CCRenderSurfaceDrawQuad::CCRenderSurfaceDrawQuad):
* platform/graphics/chromium/cc/CCRenderSurfaceDrawQuad.h:
(WebKit):
(CCRenderSurfaceDrawQuad):
(WebCore::CCRenderSurfaceDrawQuad::maskTextureId):
(WebCore::CCRenderSurfaceDrawQuad::filters):
(WebCore::CCRenderSurfaceDrawQuad::backgroundFilters):
* platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
(WebCore::CCTiledLayerImpl::contentsTextureId):
* platform/graphics/chromium/cc/CCTiledLayerImpl.h:
(CCTiledLayerImpl):
2012-05-31 Brady Eidson <beidson@apple.com>
<rdar://problem/11544454> and https://bugs.webkit.org/show_bug.cgi?id=87990
Crashes unregistering DOMWindowProperties while releasing CachedPages
Reviewed by Jessie Berlin.
This patch rewrites DOMWindowProperty to always keep direct track of the DOMWindow
it has registered with and to only ever unregister from that very same DOMWindow.
No new tests. (While the direct cause of the crash is understood, reproducing it is not)
* page/DOMWindowProperty.cpp:
(WebCore::DOMWindowProperty::DOMWindowProperty):
(WebCore::DOMWindowProperty::~DOMWindowProperty):
(WebCore::DOMWindowProperty::disconnectFrameForPageCache):
(WebCore::DOMWindowProperty::reconnectFrameFromPageCache):
(WebCore::DOMWindowProperty::willDestroyGlobalObjectInCachedFrame):
(WebCore::DOMWindowProperty::willDestroyGlobalObjectInFrame):
(WebCore::DOMWindowProperty::willDetachGlobalObjectFromFrame):
* page/DOMWindowProperty.h:
(DOMWindowProperty):
2012-05-31 Tony Chang <tony@chromium.org>
rename flexbox CSS properties from flex-align and flex-item-align to align-items and align-self
https://bugs.webkit.org/show_bug.cgi?id=87907
Reviewed by Ojan Vafai.
This was recently changed in the spec:
http://dev.w3.org/csswg/css3-flexbox/#align-items-property
No new tests, updated the tests to use the new naming.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore):
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
(WebCore::isKeywordPropertyID):
(WebCore::CSSParser::parseValue):
* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
(WebCore::CSSPrimitiveValue::operator EAlignItems):
* css/CSSProperty.cpp:
(WebCore::CSSProperty::isInheritedProperty):
* css/CSSPropertyNames.in:
* css/StyleBuilder.cpp:
(WebCore::StyleBuilder::StyleBuilder):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::sizesToIntrinsicLogicalWidth):
* rendering/RenderFlexibleBox.cpp:
(WebCore::alignmentForChild):
(WebCore::RenderFlexibleBox::layoutAndPlaceChildren):
(WebCore::RenderFlexibleBox::alignChildren):
* rendering/style/RenderStyle.h:
* rendering/style/RenderStyleConstants.h:
* rendering/style/StyleFlexibleBoxData.cpp:
(WebCore::StyleFlexibleBoxData::StyleFlexibleBoxData):
(WebCore::StyleFlexibleBoxData::operator==):
* rendering/style/StyleFlexibleBoxData.h:
(StyleFlexibleBoxData): Move member variables to StyleRareNonInheritedData
because these values will eventually apply to more than just flexbox.
* rendering/style/StyleRareNonInheritedData.cpp:
(WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
(WebCore::StyleRareNonInheritedData::operator==):
* rendering/style/StyleRareNonInheritedData.h:
(StyleRareNonInheritedData):
2012-05-31 Ami Fischman <fischman@chromium.org>
Replace WebMediaPlayer::bytesLoaded() with an explicit didLoadingProgress()
https://bugs.webkit.org/show_bug.cgi?id=86113
Reviewed by Eric Carlson.
No new functionality, so no new tests. Loading progress is already tested by existing layouttests.
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::HTMLMediaElement):
(WebCore::HTMLMediaElement::startProgressEventTimer):
(WebCore::HTMLMediaElement::changeNetworkStateFromLoadingToIdle):
(WebCore::HTMLMediaElement::progressEventTimerFired):
(WebCore::HTMLMediaElement::startPlaybackProgressTimer):
* platform/graphics/MediaPlayer.cpp:
(WebCore::NullMediaPlayerPrivate::didLoadingProgress):
(WebCore::MediaPlayer::didLoadingProgress):
* platform/graphics/MediaPlayer.h:
* platform/graphics/MediaPlayerPrivate.h:
(MediaPlayerPrivateInterface):
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
(WebCore::MediaPlayerPrivateAVFoundation::MediaPlayerPrivateAVFoundation):
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:
(MediaPlayerPrivateAVFoundation):
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp:
(WebCore::MediaPlayerPrivate::didLoadingProgress):
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.h:
(MediaPlayerPrivate):
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
(WebCore::MediaPlayerPrivateGStreamer::MediaPlayerPrivateGStreamer):
(WebCore::MediaPlayerPrivateGStreamer::didLoadingProgress):
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:
(MediaPlayerPrivateGStreamer):
* platform/graphics/mac/MediaPlayerPrivateQTKit.h:
(MediaPlayerPrivateQTKit):
* platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
(WebCore::MediaPlayerPrivateQTKit::MediaPlayerPrivateQTKit):
(WebCore::MediaPlayerPrivateQTKit::didLoadingProgress):
* platform/graphics/qt/MediaPlayerPrivateQt.cpp:
(WebCore::MediaPlayerPrivateQt::MediaPlayerPrivateQt):
(WebCore::MediaPlayerPrivateQt::didLoadingProgress):
* platform/graphics/qt/MediaPlayerPrivateQt.h:
(MediaPlayerPrivateQt):
* platform/graphics/win/MediaPlayerPrivateQuickTimeVisualContext.cpp:
(WebCore::MediaPlayerPrivateQuickTimeVisualContext::MediaPlayerPrivateQuickTimeVisualContext):
(WebCore::MediaPlayerPrivateQuickTimeVisualContext::didLoadingProgress):
* platform/graphics/win/MediaPlayerPrivateQuickTimeVisualContext.h:
(MediaPlayerPrivateQuickTimeVisualContext):
* platform/graphics/wince/MediaPlayerPrivateWinCE.h:
(MediaPlayerPrivate):
2012-05-31 Erik Arvidsson <arv@chromium.org>
Make DOM Exceptions Errors
https://bugs.webkit.org/show_bug.cgi?id=85078
Reviewed by Oliver Hunt.
WebIDL mandates that exceptions should have Error.prototype on its prototype chain.
For JSC we have access to the Error.prototype from the binding code.
For V8 we set a field in the WrapperTypeInfo and when the constructor function is created we
set the prototype as needed.
Updated test: fast/dom/DOMException/prototype-object.html
* ForwardingHeaders/runtime/ErrorPrototype.h: Added.
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateHeader):
(GenerateImplementation):
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateNamedConstructorCallback):
(GenerateImplementation):
* bindings/scripts/test/JS/JSTestException.cpp:
(WebCore::JSTestException::createPrototype):
* bindings/scripts/test/JS/JSTestException.h:
* bindings/scripts/test/V8/V8Float64Array.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestActiveDOMObject.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestCustomNamedGetter.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestEventConstructor.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestEventTarget.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestException.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestInterface.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestMediaQueryListListener.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestNamedConstructor.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestNode.cpp:
(WebCore):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore):
(WebCore::V8TestObj::installPerContextProperties):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.cpp:
(WebCore):
* bindings/v8/NPV8Object.cpp:
(WebCore::npObjectTypeInfo):
* bindings/v8/V8BindingPerContextData.cpp:
(WebCore):
(WebCore::V8BindingPerContextData::init):
(WebCore::V8BindingPerContextData::createWrapperFromCacheSlowCase):
(WebCore::V8BindingPerContextData::constructorForTypeSlowCase):
* bindings/v8/V8BindingPerContextData.h:
(V8BindingPerContextData):
* bindings/v8/V8HiddenPropertyName.h:
(WebCore):
* bindings/v8/WrapperTypeInfo.h:
(WebCore):
(WrapperTypeInfo):
* bindings/v8/custom/V8HTMLImageElementConstructor.cpp:
(WebCore):
2012-05-31 Ian Vollick <vollick@chromium.org>
[chromium] create WebTransformOperation interface for chromium platform
https://bugs.webkit.org/show_bug.cgi?id=87510
Reviewed by James Robinson.
Unit tests:
WebTransformOperationsTest.transformTypesAreUnique
WebTransformOperationsTest.matchesTypesSameLength
WebTransformOperationsTest.matchesTypesDifferentLength
WebTransformOperationsTest.applyTranslate
WebTransformOperationsTest.applyRotate
WebTransformOperationsTest.applyScale
WebTransformOperationsTest.applySkew
WebTransformOperationsTest.applyPerspective
WebTransformOperationsTest.applyMatrix
WebTransformOperationsTest.applyOrder
WebTransformOperationsTest.blendOrder
WebTransformOperationsTest.blendProgress
WebTransformOperationsTest.blendWhenTypesDoNotMatch
* WebCore.gypi:
* platform/chromium/support/WebTransformOperations.cpp: Added.
(WebKit):
(WebTransformOperationsPrivate):
(WebKit::WebTransformOperations::apply):
(WebKit::WebTransformOperations::blend):
(WebKit::WebTransformOperations::matchesTypes):
(WebKit::WebTransformOperations::appendTranslate):
(WebKit::WebTransformOperations::appendRotate):
(WebKit::WebTransformOperations::appendScale):
(WebKit::WebTransformOperations::appendSkew):
(WebKit::WebTransformOperations::appendPerspective):
(WebKit::WebTransformOperations::appendMatrix):
(WebKit::WebTransformOperations::reset):
(WebKit::WebTransformOperations::initialize):
2012-05-31 Yury Semikhatsky <yurys@chromium.org>
m_totalAllocated should be intialized in RenderArena constructor
https://bugs.webkit.org/show_bug.cgi?id=87967
Reviewed by Ojan Vafai.
* rendering/RenderArena.cpp:
(WebCore::RenderArena::RenderArena): added missing initializer.
2012-05-31 Julien Chaffraix <jchaffraix@webkit.org>
The difference between a column and a column group renderer is badly drawn
https://bugs.webkit.org/show_bug.cgi?id=87556
Reviewed by Abhishek Arya.
No expected change in behavior.
This patch tries to draw a better line between those 3 concepts:
- renderer is a RenderTableCol (RenderObject::isRenderTableCol).
- renderer is a column (RenderTableCol::isTableColumn()).
- renderer is a column group (RenderTableCol::isTableColumnGroup()).
(most of the code that needs to know the difference manipulate a RenderTableCol
so there was no need to move the 2 last helpers into RenderObject)
* rendering/AutoTableLayout.cpp:
(WebCore::AutoTableLayout::recalcColumn):
Factored the code to use the column / column group iteration we
set up in the rest of the code. Made the code more self-describing
by using more RenderTableCol functions.
* rendering/RenderObject.cpp:
(WebCore::RenderObject::addChild):
Factored the code to use the isTableColumn() and isTableColumnGroup() to
better underline what we check.
* rendering/RenderObject.h:
(WebCore::RenderObject::isRenderTableCol):
(WebCore::RenderObject::isTablePart):
Renamed isTableCol to isRenderTableCol to match the class as this doesn't
imply that the renderer is actually a column.
* html/HTMLTableColElement.cpp:
(WebCore::HTMLTableColElement::parseAttribute):
* rendering/AutoTableLayout.cpp:
(WebCore::AutoTableLayout::fullRecalc):
* rendering/RenderTable.cpp:
(WebCore::RenderTable::addChild):
(WebCore::RenderTable::layout):
(WebCore::RenderTable::firstColumn):
Updated after the isRenderTableCol rename.
* dom/Text.cpp:
(WebCore::Text::rendererIsNeeded):
* editing/htmlediting.cpp:
(WebCore::isTableStructureNode):
Ditto and renamed some variables.
* html/HTMLFormElement.cpp:
(WebCore::HTMLFormElement::rendererIsNeeded):
Ditto and added a FIXME as the logic doesn't seem very bullet proof.
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::computeCollapsedStartBorder):
(WebCore::RenderTableCell::computeCollapsedEndBorder):
(WebCore::RenderTableCell::computeCollapsedBeforeBorder):
(WebCore::RenderTableCell::computeCollapsedAfterBorder):
Changed to use the enclosingColumnGroup* helpers.
* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::paintCell):
* rendering/RenderTableCol.cpp:
(WebCore::RenderTableCol::isChildAllowed):
(WebCore::RenderTableCol::canHaveChildren):
(WebCore::RenderTableCol::enclosingColumnGroup):
(WebCore::RenderTableCol::nextColumn):
Updated to use the new functions.
* rendering/RenderTableCol.h:
(WebCore::RenderTableCol::isTableColumn):
(WebCore::RenderTableCol::isTableColumnGroup):
(WebCore::RenderTableCol::enclosingColumnGroupIfAdjacentBefore):
(WebCore::RenderTableCol::enclosingColumnGroupIfAdjacentAfter):
Added the following new helpers.
2012-05-31 Jessie Berlin <jberlin@apple.com>
Fix Lion build after r119073.
* platform/Decimal.cpp:
(WebCore::Decimal::toString):
Do not attempt to assign a unsigned long long to an int.
2012-05-31 Christophe Dumez <christophe.dumez@intel.com>
[EFL] Enable CSS_IMAGE_SET flag
https://bugs.webkit.org/show_bug.cgi?id=87727
Reviewed by Adam Roben.
Fix CMakeLists.txt so that CSSImageSetValue.cpp gets compiled if
CSS_IMAGE_SET flag is enabled.
* CMakeLists.txt:
2012-05-31 Keyar Hood <keyar@chromium.org>
the imageSmoothingEnabled flag needs to be in the state object
https://bugs.webkit.org/show_bug.cgi?id=87853
Reviewed by Darin Adler.
Updated fast/canvas/canvas-imageSmoothingEnabled.html instead of
adding a new test.
The imageSmoothingEnabled flag is saved in the draw state now.
* html/canvas/CanvasRenderingContext2D.cpp:
(WebCore::CanvasRenderingContext2D::CanvasRenderingContext2D):
(WebCore::CanvasRenderingContext2D::State::State):
(WebCore::CanvasRenderingContext2D::State::operator=):
(WebCore::CanvasRenderingContext2D::webkitImageSmoothingEnabled):
(WebCore::CanvasRenderingContext2D::setWebkitImageSmoothingEnabled):
* html/canvas/CanvasRenderingContext2D.h:
(State):
2012-05-31 Eugene Klyuchnikov <eustas.bug@gmail.com>
Web Inspector: Fix checkbox position on Settings screen.
https://bugs.webkit.org/show_bug.cgi?id=87007
Reviewed by Yury Semikhatsky.
Checkbox seems to be improperly aligned.
It is "shifted" up on 1px on Mac and 2px on Linux.
Also, keyboard-shortcuts screen has "wrapped" key descriptions on Mac.
UI changes, no new tests.
* inspector/front-end/helpScreen.css:
(.help-container): Fixed column width.
(.help-block): Ditto.
(.help-key-cell): Ditto.
(body.platform-mac .help-container): Ditto.
(body.platform-mac .help-block): Ditto.
(body.platform-mac .help-key-cell): Ditto.
(.help-content p): Ajdusted spacing between lines
(.help-content input[type=checkbox]): Fixed alignment.
(body.platform-mac .help-content input[type=checkbox]): Ditto.
(.help-content input[type=radio]): Fixed alignment.
(body.platform-mac .help-content input[type=radio]): Ditto.
2012-05-30 Andreas Kling <kling@webkit.org>
Have StylePropertySet constructor take array/length instead of vector.
<http://webkit.org/b/87876>
Reviewed by Antti Koivisto.
Remove the StylePropertyVector typedef and have StylePropertySet constructors
take CSSProperty*/length since we are copying the data into a tightly packed
array anyway. This frees up the call sites to use whatever storage they please
rather than being restricted to a vector with inlineCapacity=4.
Change said call sites to use an arbitrary high inline capacity (256) for their
stack-allocated temporary vectors.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::copyPropertiesInSet):
* css/CSSParser.cpp:
(WebCore::filterProperties):
(WebCore::CSSParser::createStylePropertySet):
* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::createImmutable):
(WebCore::StylePropertySet::StylePropertySet):
(WebCore::StylePropertySet::removePropertiesInSet):
(WebCore::StylePropertySet::copyPropertiesInSet):
* css/StylePropertySet.h:
(WebCore::StylePropertySet::create):
(StylePropertySet):
2012-05-31 Tor Arne Vestbø <tor.arne.vestbo@nokia.com>
[Qt] Simplify QT_VERSION_CHECKS for Qt5 by introducing HAVE(QT5)
https://bugs.webkit.org/show_bug.cgi?id=87955
Reviewed by Simon Hausmann.
* bridge/qt/qt_class.cpp:
* bridge/qt/qt_instance.cpp:
* bridge/qt/qt_runtime.h:
* page/qt/EventHandlerQt.cpp:
* platform/graphics/GraphicsContext3D.h:
* platform/graphics/OpenGLShims.cpp:
* platform/graphics/OpenGLShims.h:
* platform/graphics/gstreamer/PlatformVideoWindowPrivate.h:
* platform/graphics/gstreamer/PlatformVideoWindowQt.cpp:
* platform/graphics/qt/GraphicsContext3DQt.cpp:
* platform/graphics/qt/SimpleFontDataQt.cpp:
* platform/graphics/texmap/TextureMapper.h:
* platform/graphics/texmap/TextureMapperGL.cpp:
* platform/qt/DeviceMotionProviderQt.h:
* platform/qt/DeviceOrientationProviderQt.h:
* platform/qt/KURLQt.cpp:
* platform/qt/PlatformScreenQt.cpp:
* platform/qt/QWebPageClient.h:
* plugins/qt/PluginPackageQt.cpp:
* plugins/qt/PluginViewQt.cpp:
2012-05-30 Andreas Kling <kling@webkit.org>
Have StylePropertySet constructor take array/length instead of vector.
<http://webkit.org/b/87876>
Reviewed by Antti Koivisto.
Remove the StylePropertyVector typedef and have StylePropertySet constructors
take CSSProperty*/length since we are copying the data into a tightly packed
array anyway. This frees up the call sites to use whatever storage they please
rather than being restricted to a vector with inlineCapacity=4.
Change said call sites to use an arbitrary high inline capacity (256) for their
stack-allocated temporary vectors.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::copyPropertiesInSet):
* css/CSSParser.cpp:
(WebCore::filterProperties):
(WebCore::CSSParser::createStylePropertySet):
* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::createImmutable):
(WebCore::StylePropertySet::StylePropertySet):
(WebCore::StylePropertySet::removePropertiesInSet):
(WebCore::StylePropertySet::copyPropertiesInSet):
* css/StylePropertySet.h:
(WebCore::StylePropertySet::create):
(StylePropertySet):
2012-05-31 Mike West <mkwst@chromium.org>
Fixing compilation with SVG disabled.
https://bugs.webkit.org/show_bug.cgi?id=87944
This patch adds the missing enumeration value
'CSSPropertyWebkitBoxDecorationBreak' into switch statements to make
clang happy.
Reviewed by Alexis Menard.
* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
2012-05-31 Yoshifumi Inoue <yosin@chromium.org>
[Platform] Implementation of Decimal(int32_t) isn't portable.
https://bugs.webkit.org/show_bug.cgi?id=87941
Reviewed by Kent Tamura.
This patch changes negation of integer to make unsigned integer
in portable way.
No new tests. This patch doesn't change behavior.
* platform/Decimal.cpp:
(WebCore::Decimal::Decimal): Use uint64_t(-int64_t(i32)) instead of uint64_t(-i32)
2012-05-31 Gyuyoung Kim <gyuyoung.kim@samsung.com>
[CMAKE][EFL] Remove unneeded include path
https://bugs.webkit.org/show_bug.cgi?id=87927
Reviewed by Kenneth Rohde Christiansen.
Though *wtf* directory was moved to Source/WTF, PlatformEfl.cmake is still including the previous
path. In addition, files on EFL port are using "<wtf/gobject/XXX>" directly in #include line.
So, EFL port doesn't need to include ${JAVASCRIPTCORE_DIR}/wtf/gobject path anymore.
* PlatformEfl.cmake: Remove '${JAVASCRIPTCORE_DIR}/wtf/gobject' path from include path list.
2012-05-31 Yoshifumi Inoue <yosin@chromium.org>
[Platform] Introduce Decimal class for Number/Range input type.
https://bugs.webkit.org/show_bug.cgi?id=87360
Reviewed by Kent Tamura.
This patch added new class Decimal for decimal arithmatic in two
files: platform/Decimal.cpp and Decimal.h with unit test.
Test: WebKit/chromium/tests/DecimalTest.cpp
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* platform/Decimal.cpp: Added.
(WebCore):
(DecimalPrivate):
(SpecialValueHandler):
(WebCore::DecimalPrivate::SpecialValueHandler::SpecialValueHandler):
(WebCore::DecimalPrivate::SpecialValueHandler::handle):
(WebCore::DecimalPrivate::SpecialValueHandler::value):
(UInt128):
(WebCore::DecimalPrivate::UInt128::UInt128):
(WebCore::DecimalPrivate::UInt128::high):
(WebCore::DecimalPrivate::UInt128::low):
(WebCore::DecimalPrivate::UInt128::multiply):
(WebCore::DecimalPrivate::UInt128::highUInt32):
(WebCore::DecimalPrivate::UInt128::lowUInt32):
(WebCore::DecimalPrivate::UInt128::isZero):
(WebCore::DecimalPrivate::UInt128::makeUInt64):
(WebCore::DecimalPrivate::UInt128::operator/=):
(WebCore::DecimalPrivate::UInt128::multiplyHigh):
(WebCore::DecimalPrivate::countDigits):
(WebCore::DecimalPrivate::scaleDown):
(WebCore::DecimalPrivate::scaleUp):
(WebCore::Decimal::EncodedData::EncodedData):
(WebCore::Decimal::EncodedData::operator==):
(WebCore::Decimal::Decimal):
(WebCore::Decimal::operator=):
(WebCore::Decimal::operator+=):
(WebCore::Decimal::operator-=):
(WebCore::Decimal::operator*=):
(WebCore::Decimal::operator/=):
(WebCore::Decimal::operator-):
(WebCore::Decimal::operator+):
(WebCore::Decimal::operator*):
(WebCore::Decimal::operator/):
(WebCore::Decimal::operator==):
(WebCore::Decimal::operator!=):
(WebCore::Decimal::operator<):
(WebCore::Decimal::operator<=):
(WebCore::Decimal::operator>):
(WebCore::Decimal::operator>=):
(WebCore::Decimal::abs):
(WebCore::Decimal::alignOperands):
(WebCore::Decimal::ceiling):
(WebCore::Decimal::compareTo):
(WebCore::Decimal::floor):
(WebCore::Decimal::fromString):
(WebCore::Decimal::infinity):
(WebCore::Decimal::nan):
(WebCore::Decimal::remainder):
(WebCore::Decimal::round):
(WebCore::Decimal::toString):
(WebCore::Decimal::zero):
* platform/Decimal.h: Added.
(WebCore):
(DecimalPrivate):
(Decimal):
(EncodedData):
(WebCore::Decimal::EncodedData::operator!=):
(WebCore::Decimal::EncodedData::coefficient):
(WebCore::Decimal::EncodedData::exponent):
(WebCore::Decimal::EncodedData::isFinite):
(WebCore::Decimal::EncodedData::isNaN):
(WebCore::Decimal::EncodedData::isSpecial):
(WebCore::Decimal::EncodedData::isZero):
(WebCore::Decimal::EncodedData::sign):
(WebCore::Decimal::EncodedData::setSign):
(WebCore::Decimal::EncodedData::formatClass):
(WebCore::Decimal::isFinite):
(WebCore::Decimal::isNaN):
(WebCore::Decimal::isNegative):
(WebCore::Decimal::isPositive):
(WebCore::Decimal::isSpecial):
(WebCore::Decimal::isZero):
(WebCore::Decimal::value):
(AlignedOperands):
(WebCore::Decimal::invertSign):
(WebCore::Decimal::exponent):
(WebCore::Decimal::sign):
2012-05-31 Kentaro Hara <haraken@chromium.org>
[V8] Pass Isolate to v8String() in custom bindings
https://bugs.webkit.org/show_bug.cgi?id=87825
Reviewed by Adam Barth.
The objective is to pass Isolate around in V8 bindings.
This patch passes Isolate to v8String() in custom bindings.
No tests. No change in behavior.
* bindings/v8/custom/V8ArrayBufferViewCustom.cpp:
(WebCore::installFastSet):
* bindings/v8/custom/V8ArrayBufferViewCustom.h:
(WebCore):
(WebCore::setWebGLArrayHelper):
* bindings/v8/custom/V8CSSStyleDeclarationCustom.cpp:
(WebCore::V8CSSStyleDeclaration::namedPropertyEnumerator):
(WebCore::V8CSSStyleDeclaration::namedPropertyGetter):
* bindings/v8/custom/V8CanvasRenderingContext2DCustom.cpp:
(WebCore::toV8Object):
* bindings/v8/custom/V8ClipboardCustom.cpp:
(WebCore::V8Clipboard::typesAccessorGetter):
* bindings/v8/custom/V8DOMStringMapCustom.cpp:
(WebCore::V8DOMStringMap::namedPropertyGetter):
(WebCore::V8DOMStringMap::namedPropertyEnumerator):
* bindings/v8/custom/V8FileReaderCustom.cpp:
(WebCore::V8FileReader::resultAccessorGetter):
* bindings/v8/custom/V8HTMLCanvasElementCustom.cpp:
(WebCore::V8HTMLCanvasElement::toDataURLCallback):
* bindings/v8/custom/V8HTMLElementCustom.cpp:
(WebCore::toV8Object):
* bindings/v8/custom/V8HTMLInputElementCustom.cpp:
(WebCore::V8HTMLInputElement::selectionDirectionAccessorGetter):
* bindings/v8/custom/V8IDBAnyCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8IDBKeyCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8InjectedScriptHostCustom.cpp:
(WebCore::V8InjectedScriptHost::databaseIdCallback):
(WebCore::V8InjectedScriptHost::storageIdCallback):
* bindings/v8/custom/V8InspectorFrontendHostCustom.cpp:
(WebCore::V8InspectorFrontendHost::platformCallback):
* bindings/v8/custom/V8JavaScriptCallFrameCustom.cpp:
(WebCore::V8JavaScriptCallFrame::typeAccessorGetter):
* bindings/v8/custom/V8LocationCustom.cpp:
(WebCore::V8Location::toStringCallback):
* bindings/v8/custom/V8SQLResultSetRowListCustom.cpp:
(WebCore::V8SQLResultSetRowList::itemCallback):
* bindings/v8/custom/V8StorageCustom.cpp:
(WebCore::V8Storage::namedPropertyEnumerator):
(WebCore::storageGetter):
* bindings/v8/custom/V8XMLHttpRequestCustom.cpp:
(WebCore::V8XMLHttpRequest::responseTextAccessorGetter):
* bindings/v8/custom/V8XSLTProcessorCustom.cpp:
(WebCore::V8XSLTProcessor::getParameterCallback):
2012-05-31 Hayato Ito <hayato@chromium.org>
ComposedShadowTreeWalker should support traversing nodes in an orphaned shadow subtree.
https://bugs.webkit.org/show_bug.cgi?id=87493
Reviewed by Dimitri Glazkov.
ComposedShadowTreeWalker assumed that a visited shadow root is
always assigned to a shadow insertion point since it only
traverses nodes which are rendered. But there is an exceptional
use case such as an event dispatching. Some events, such as a
'click' event, may happen in an orphaned shadow subtree. In such
cases, traversal might start with a node in an orphaned shadow
subtree. So ComposedShadowTreeWalker can not assume that visited
shadow root is always assigned to a shadow insertion point.
This patch only fixes ComposedShadowTreeWalker.
ComposedShadowTreeParentWalker will be fixed in another patch with
an event dispatching test.
Test: fast/dom/shadow/composed-shadow-tree-walker.html
* dom/ComposedShadowTreeWalker.cpp:
(WebCore::ComposedShadowTreeWalker::traverseParentBackToYoungerShadowRootOrHost):
2012-05-30 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: add MemoryUsageSupport::processMemorySizesInBytes
https://bugs.webkit.org/show_bug.cgi?id=87830
Reviewed by James Robinson.
Added a method for getting process memory usage in bytes. It is
used in the inspector memory instrumentation to get process total
memory usage.
* inspector/InspectorMemoryAgent.cpp:
(WebCore::InspectorMemoryAgent::getProcessMemoryDistribution):
* platform/MemoryUsageSupport.cpp:
(WebCore::MemoryUsageSupport::processMemorySizesInBytes):
(WebCore):
* platform/MemoryUsageSupport.h:
(MemoryUsageSupport): provided embedders with a way to report WebKit process
memory usage.
* platform/chromium/MemoryUsageSupportChromium.cpp:
(WebCore::MemoryUsageSupport::processMemorySizesInBytes):
(WebCore):
2012-05-31 Kent Tamura <tkent@chromium.org>
Unreviewed, rolling out r119062 and r119064.
http://trac.webkit.org/changeset/119062
http://trac.webkit.org/changeset/119064
https://bugs.webkit.org/show_bug.cgi?id=87360
Broke build on Lion, SnowLoepard, Chromium Windows, and
Chromium Linux 32
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* platform/Decimal.cpp: Removed.
* platform/Decimal.h: Removed.
2012-05-31 Shane Stephens <shanestephens@google.com>
text-decoration should not be propagated through absolutely positioned elements to <a> tags
https://bugs.webkit.org/show_bug.cgi?id=86517
Reviewed by Darin Adler.
Test: fast/css/text-decoration-in-second-order-descendants.html
* rendering/RenderObject.cpp:
(WebCore::RenderObject::getTextDecorationColors):
Remove fix from 18611 as it doesn't work on second order descendants.
* css/StyleResolver.cpp:
(WebCore::StyleResolver::adjustRenderStyle):
Add floating and positioned checks when deciding whether to avoid propagating text decoration. Move all checks into new method to improve readability.
(WebCore::doesNotInheritTextDecoration): Added
2012-05-30 Yoshifumi Inoue <yosin@chromium.org>
Build fix for Linon/SnowLeopard after r119062
* platform/Decimal.cpp:
(WebCore::Decimal::Decimal):
2012-05-30 Yoshifumi Inoue <yosin@chromium.org>
[Platform] Introduce Decimal class for Number/Range input type.
https://bugs.webkit.org/show_bug.cgi?id=87360
Reviewed by Kent Tamura.
This patch added new class Decimal for decimal arithmatic in two
files: platform/Decimal.cpp and Decimal.h with unit test.
Test: WebKit/chromium/tests/DecimalTest.cpp
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* platform/Decimal.cpp: Added.
(WebCore):
(DecimalPrivate):
(SpecialValueHandler):
(WebCore::DecimalPrivate::SpecialValueHandler::SpecialValueHandler):
(WebCore::DecimalPrivate::SpecialValueHandler::handle):
(WebCore::DecimalPrivate::SpecialValueHandler::value):
(UInt128):
(WebCore::DecimalPrivate::UInt128::UInt128):
(WebCore::DecimalPrivate::UInt128::high):
(WebCore::DecimalPrivate::UInt128::low):
(WebCore::DecimalPrivate::UInt128::multiply):
(WebCore::DecimalPrivate::UInt128::highUInt32):
(WebCore::DecimalPrivate::UInt128::lowUInt32):
(WebCore::DecimalPrivate::UInt128::isZero):
(WebCore::DecimalPrivate::UInt128::makeUInt64):
(WebCore::DecimalPrivate::UInt128::operator/=):
(WebCore::DecimalPrivate::UInt128::multiplyHigh):
(WebCore::DecimalPrivate::countDigits):
(WebCore::DecimalPrivate::scaleDown):
(WebCore::DecimalPrivate::scaleUp):
(WebCore::Decimal::EncodedData::EncodedData):
(WebCore::Decimal::EncodedData::operator==):
(WebCore::Decimal::Decimal):
(WebCore::Decimal::operator=):
(WebCore::Decimal::operator+=):
(WebCore::Decimal::operator-=):
(WebCore::Decimal::operator*=):
(WebCore::Decimal::operator/=):
(WebCore::Decimal::operator-):
(WebCore::Decimal::operator+):
(WebCore::Decimal::operator*):
(WebCore::Decimal::operator/):
(WebCore::Decimal::operator==):
(WebCore::Decimal::operator!=):
(WebCore::Decimal::operator<):
(WebCore::Decimal::operator<=):
(WebCore::Decimal::operator>):
(WebCore::Decimal::operator>=):
(WebCore::Decimal::abs):
(WebCore::Decimal::alignOperands):
(WebCore::Decimal::ceiling):
(WebCore::Decimal::compareTo):
(WebCore::Decimal::floor):
(WebCore::Decimal::fromString):
(WebCore::Decimal::infinity):
(WebCore::Decimal::nan):
(WebCore::Decimal::remainder):
(WebCore::Decimal::round):
(WebCore::Decimal::toString):
(WebCore::Decimal::zero):
* platform/Decimal.h: Added.
(WebCore):
(DecimalPrivate):
(Decimal):
(EncodedData):
(WebCore::Decimal::EncodedData::operator!=):
(WebCore::Decimal::EncodedData::coefficient):
(WebCore::Decimal::EncodedData::exponent):
(WebCore::Decimal::EncodedData::isFinite):
(WebCore::Decimal::EncodedData::isNaN):
(WebCore::Decimal::EncodedData::isSpecial):
(WebCore::Decimal::EncodedData::isZero):
(WebCore::Decimal::EncodedData::sign):
(WebCore::Decimal::EncodedData::setSign):
(WebCore::Decimal::EncodedData::formatClass):
(WebCore::Decimal::isFinite):
(WebCore::Decimal::isNaN):
(WebCore::Decimal::isNegative):
(WebCore::Decimal::isPositive):
(WebCore::Decimal::isSpecial):
(WebCore::Decimal::isZero):
(WebCore::Decimal::value):
(AlignedOperands):
(WebCore::Decimal::invertSign):
(WebCore::Decimal::exponent):
(WebCore::Decimal::sign):
2012-05-30 Patrick Gansterer <paroga@webkit.org>
Build fix for WinCE after r116723.
* platform/graphics/wince/FontWinCE.cpp:
(WebCore::TextRunComponent::TextRunComponent):
2012-05-30 Kentaro Hara <haraken@chromium.org>
Implement CSSParser::determineNameInNamespace() as a helper function for CSSGrammar.y
https://bugs.webkit.org/show_bug.cgi?id=87799
Reviewed by Darin Adler.
As pointed out by darin@ in https://bugs.webkit.org/show_bug.cgi?id=87627#c12,
we should avoid repeating the following code in CSSGrammar.y:
if (p->m_styleSheet)
$$->setTag(QualifiedName(namespacePrefix, $2, p->m_styleSheet->determineNamespace(namespacePrefix)));
else
$$->setTag(QualifiedName(namespacePrefix, $2, p->m_defaultNamespace));
This patch implements CSSParser::determineNameInNamespace() as a helper function
and replaces the above code.
Tests: fast/dom/SelectorAPI/*. No change in test results.
* css/CSSGrammar.y:
* css/CSSParser.cpp:
(WebCore::CSSParser::determineNameInNamespace):
(WebCore):
* css/CSSParser.h:
2012-05-30 Patrick Gansterer <paroga@webkit.org>
Build fix for WinCE after r117697.
* page/wince/FrameWinCE.cpp:
(WebCore::computePageRectsForFrame):
2012-05-30 Shawn Singh <shawnsingh@chromium.org>
[chromium] Fix min/max bounds error in CCMathUtil.cpp
https://bugs.webkit.org/show_bug.cgi?id=87915
Reviewed by James Robinson.
Two unit tests added to CCMathUtilTest:
CCMathUtilTest.verifyEnclosingClippedRectUsesCorrectInitialBounds
CCMathUtilTest.verifyEnclosingRectOfVerticesUsesCorrectInitialBounds
While computing bounds, the initial values for xmax and ymax are
intended to be set to -float_max. It turns out that
std::numeric_limits<float>::min() actually returns the smallest
positive value close to zero, which is not what was intended. This
patch fixes the code to use -float_max instead, which is the
intended value.
* platform/graphics/chromium/cc/CCMathUtil.cpp:
(WebCore::CCMathUtil::mapClippedRect):
(WebCore::CCMathUtil::projectClippedRect):
(WebCore::CCMathUtil::computeEnclosingRectOfVertices):
(WebCore::CCMathUtil::computeEnclosingClippedRect):
(WebCore):
* platform/graphics/chromium/cc/CCMathUtil.h:
(WebCore::HomogeneousCoordinate::HomogeneousCoordinate):
(HomogeneousCoordinate):
(WebCore::HomogeneousCoordinate::shouldBeClipped):
(WebCore::HomogeneousCoordinate::cartesianPoint2d):
(WebCore):
(CCMathUtil):
2012-05-30 Patrick Gansterer <paroga@webkit.org>
Build fix for WinCE after r118568.
* platform/text/wince/TextBreakIteratorWinCE.cpp:
(WebCore::NonSharedCharacterBreakIterator::NonSharedCharacterBreakIterator):
2012-05-30 Abhishek Arya <inferno@chromium.org>
Crash in ContainerNode::parserAddChild.
https://bugs.webkit.org/show_bug.cgi?id=87903
Reviewed by Ryosuke Niwa.
Call the ChildNodeInsertionNotifier.notify call at the end since
it can destroy |this| and some of the local pointers like |last|.
This also matches the order of calls - childrenChanged precedes
ChildNodeInsertionNotifier.notify in updateTreeAfterInsertion and
ContainerNode::parserInsertBefore.
Also remove a FIXME since we use ChildNodeInsertionNotifier.notify
instead of ChildNodeInsertionNotifier.notifyInsertedIntoDocument
(as recommended in the FIXME).
Test: fast/dom/child-insertion-notify-crash.html
* dom/ContainerNode.cpp:
(WebCore::ContainerNode::parserAddChild):
2012-05-30 Rick Byers <rbyers@chromium.org>
[chromium] Implement pointer and hover CSS media features
https://bugs.webkit.org/show_bug.cgi?id=87403
Reviewed by Adam Barth.
Add initial support for the pointer and hover media features as defined
here:
http://dev.w3.org/csswg/css4-mediaqueries/#pointer
http://dev.w3.org/csswg/css4-mediaqueries/#hover
For now this only supports the case where we know a touch screen is
available (currently known only for the chromium port on certain
platforms). All other cases continue to behave exactly as if the media
feature isn't supported.
Tests: fast/media/mq-pointer.html
* css/CSSValueKeywords.in:
* css/MediaFeatureNames.h:
(MediaFeatureNames):
* css/MediaQueryEvaluator.cpp:
(WebCore::getLeastCapablePrimaryPointerDeviceType):
(WebCore):
(WebCore::hoverMediaFeatureEval):
(WebCore::pointerMediaFeatureEval):
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::InternalSettings):
(WebCore::InternalSettings::restoreTo):
(WebCore::InternalSettings::setDeviceSupportsTouch):
(WebCore):
* testing/InternalSettings.h:
(InternalSettings):
* testing/InternalSettings.idl:
2012-05-30 Julien Chaffraix <jchaffraix@webkit.org>
Add an helper function to get the style for a cell's flow computation
https://bugs.webkit.org/show_bug.cgi?id=87902
Reviewed by Ojan Vafai.
No expected change in behavior.
This change introduces styleForCellFlow that unify which function we use
to determine a cell's direction and writing mode.
This unification will make it easier to support direction on table row
group (section in WebKit land).
* rendering/RenderTableCell.h:
(WebCore::RenderTableCell::styleForCellFlow):
Added the new function along with a comment as to its purpose.
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::clippedOverflowRectForRepaint):
(WebCore::RenderTableCell::computeCollapsedStartBorder):
(WebCore::RenderTableCell::computeCollapsedEndBorder):
(WebCore::RenderTableCell::computeCollapsedBeforeBorder):
(WebCore::RenderTableCell::computeCollapsedAfterBorder):
(WebCore::RenderTableCell::borderHalfLeft):
(WebCore::RenderTableCell::borderHalfRight):
(WebCore::RenderTableCell::borderHalfTop):
(WebCore::RenderTableCell::borderHalfBottom):
(WebCore::RenderTableCell::borderHalfStart):
(WebCore::RenderTableCell::borderHalfEnd):
(WebCore::RenderTableCell::borderHalfBefore):
(WebCore::RenderTableCell::borderHalfAfter):
(WebCore::RenderTableCell::paintCollapsedBorders):
Mostly mechanical replacement.
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::cachedCollapsedLeftBorder):
(WebCore::RenderTableCell::cachedCollapsedRightBorder):
(WebCore::RenderTableCell::cachedCollapsedTopBorder):
(WebCore::RenderTableCell::cachedCollapsedBottomBorder):
Made the argument 'const' now.
* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::layoutRows):
Added a FIXME to unify with the rest.
2012-05-30 Christophe Dumez <christophe.dumez@intel.com>
[JSC] SerializedScriptValue.create() succeeds even if MessagePort object cannot be found in transferred ports
https://bugs.webkit.org/show_bug.cgi?id=87118
Reviewed by Adam Barth.
Make SerializedScriptValue.create() throw an exception if one of the
MessagePort objects cannot be found in the transferred ports. This
matches the behavior of the V8 implementation.
Test: webintents/web-intents-obj-constructor.html
* bindings/js/SerializedScriptValue.cpp:
(WebCore::CloneSerializer::dumpIfTerminal):
2012-05-30 Mark Pilgrim <pilgrim@chromium.org>
[Chromium] Call fileUtilities methods directly
https://bugs.webkit.org/show_bug.cgi?id=87852
Reviewed by Adam Barth.
Part of a refactoring series. See tracking bug 82948.
* platform/chromium/DragDataChromium.cpp:
(WebCore::DragData::asURL):
* platform/chromium/FileSystemChromium.cpp:
(WebCore::deleteFile):
(WebCore::deleteEmptyDirectory):
(WebCore::getFileSize):
(WebCore::getFileModificationTime):
(WebCore::getFileMetadata):
(WebCore::directoryName):
(WebCore::pathByAppendingComponent):
(WebCore::makeAllDirectories):
(WebCore::openFile):
(WebCore::closeFile):
(WebCore::seekFile):
(WebCore::truncateFile):
(WebCore::readFromFile):
(WebCore::writeToFile):
* platform/chromium/PlatformSupport.h:
(PlatformSupport):
2012-05-30 Kent Tamura <tkent@chromium.org>
Form controls in <fieldset disabled> should not be focusable.
https://bugs.webkit.org/show_bug.cgi?id=87380
Reviewed by Darin Adler.
Test: fast/forms/fieldset/focus-in-fieldset-disabled.html
* html/HTMLFormControlElement.cpp:
(WebCore::HTMLFormControlElement::supportsFocus):
Refer to disabled() instead of m_disabled to take account of wrapping
fieldset elements.
2012-05-30 Robin Dunn <robin@alldunn.com>
[wx] Implement HTML clipboard support.
https://bugs.webkit.org/show_bug.cgi?id=87883
Reviewed by Kevin Ollivier.
* platform/wx/ClipboardWx.cpp:
(WebCore::ClipboardWx::writeRange):
* platform/wx/PasteboardWx.cpp:
(WebCore::Pasteboard::writeSelection):
(WebCore::Pasteboard::documentFragment):
2012-05-30 Garrett Casto <gcasto@chromium.org>
TextFieldDecorationElement should respect style attribute
https://bugs.webkit.org/show_bug.cgi?id=87762
Reviewed by Kent Tamura.
* html/shadow/TextFieldDecorationElement.cpp: Use style set on the
Element when creating rendering style.
(WebCore::TextFieldDecorationElement::customStyleForRenderer):
2012-05-30 Ojan Vafai <ojan@chromium.org>
Avoid second layout for flex-direction:row, flex-basis:auto flex items
https://bugs.webkit.org/show_bug.cgi?id=87901
Reviewed by Tony Chang.
No new tests. This is strictly a performance optimization and has no other
web visible changes.
Also, add FIXMEs for other cases where we could avoid doing layouts.
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::computeMainAxisPreferredSizes):
(WebCore::RenderFlexibleBox::layoutAndPlaceChildren):
(WebCore::RenderFlexibleBox::applyStretchAlignmentToChild):
2012-05-30 Zhenyao Mo <zmo@google.com>
WebKit incorrectly clears the alpha channel on readPixels, even for Framebuffers
https://bugs.webkit.org/show_bug.cgi?id=87310
Reviewed by Kenneth Russell.
* html/canvas/WebGLRenderingContext.cpp:
(WebCore):
(WebCore::WebGLRenderingContext::getParameter): set DEPTH_BITS/STENCIL_BITS to 0 if related channels are not requested.
(WebCore::WebGLRenderingContext::readPixels): don't do the alpha value fix if the current bound is not the internal drawing buffer.
2012-05-30 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118986.
http://trac.webkit.org/changeset/118986
https://bugs.webkit.org/show_bug.cgi?id=87914
Caused several IndexedDB browser_test failures on Chromium
canary builders (Requested by rafaelw_ on #webkit).
* Modules/indexeddb/IDBCursor.cpp:
(WebCore::IDBCursor::advance):
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::setVersion):
* Modules/indexeddb/IDBDatabaseError.h:
(WebCore::IDBDatabaseError::create):
(IDBDatabaseError):
(WebCore::IDBDatabaseError::createWithoutOffset):
(WebCore::IDBDatabaseError::code):
(WebCore::IDBDatabaseError::setCode):
(WebCore::IDBDatabaseError::setMessage):
(WebCore::IDBDatabaseError::IDBDatabaseError):
* Modules/indexeddb/IDBDatabaseException.cpp:
(WebCore):
(WebCore::IDBDatabaseException::initializeDescription):
* Modules/indexeddb/IDBDatabaseException.h:
(IDBDatabaseException):
* Modules/indexeddb/IDBObjectStore.cpp:
(WebCore::IDBObjectStore::createIndex):
* Modules/indexeddb/IDBRequest.cpp:
(WebCore::IDBRequest::resetReadyState):
(WebCore::IDBRequest::abort):
(WebCore::IDBRequest::onError):
(WebCore::IDBRequest::onSuccess):
(WebCore::IDBRequest::onSuccessWithContinuation):
(WebCore::IDBRequest::dispatchEvent):
* Modules/indexeddb/IDBRequest.h:
(IDBRequest):
* Modules/indexeddb/IDBRequest.idl:
* Modules/indexeddb/IDBTransaction.cpp:
* Modules/indexeddb/IDBTransaction.h:
(IDBTransaction):
* Modules/indexeddb/IDBTransaction.idl:
2012-05-30 Dominic Mazzoni <dmazzoni@chromium.org>
[Chromium] AX: WebAccessibilityObject should check if an AccessibilityObject is detached
https://bugs.webkit.org/show_bug.cgi?id=87778
Reviewed by Adam Barth.
Chromium was never actually using AccessibilityObjectWrapper, so
this change deletes it. Instead, it's replaced with a simple bool
that keeps track of whether an AccessibilityObject was detached or
not. WebKit/chromium/public/WebAccessibilityObject can then use this
to determine if an object is still valid.
Test: accessibility/accessibility-object-detached.html
* WebCore.gypi:
* accessibility/AccessibilityObject.cpp:
(WebCore::AccessibilityObject::AccessibilityObject):
(WebCore::AccessibilityObject::detach):
(WebCore::AccessibilityObject::isDetached):
(WebCore):
* accessibility/AccessibilityObject.h:
(AccessibilityObject):
* accessibility/chromium/AXObjectCacheChromium.cpp:
(WebCore):
(WebCore::AXObjectCache::detachWrapper):
(WebCore::AXObjectCache::attachWrapper):
* accessibility/chromium/AccessibilityObjectWrapper.h: Removed.
2012-05-30 Shawn Singh <shawnsingh@chromium.org>
Simplify TransformationMatrix rotation code to improve precision
https://bugs.webkit.org/show_bug.cgi?id=86666
Reviewed by Adrienne Walker.
Change covered by existing layout tests;
Additional unit tests added.
The original rotation code for WebCore::TransformationMatrix was
using indirect trig identities to compute basic rotation
matrices. As far as I can imagine, this is unnecessary, and brief
informal experiments showed that we can gain a few extra bits of
precision by avoiding that indirection.
Indeed, precision improved such that it revealed a mistake on one
of the unit tests, where the test was actually degenerate so that
the layer was aligned with a viewport boundary. When it was
imprecise, it didnt appear to be aligned. Putting an additional
translation on that test fixes it so it is no longer a degenerate
test.
This patch also revealed a few chinks in the unit testing armor of
this code, an error that should have been caught in unit tests was
luckily caught by SVG layout tests. This chink is fixed by adding
2 extra unit tests.
* platform/graphics/transforms/TransformationMatrix.cpp:
(WebCore::TransformationMatrix::rotate3d):
2012-05-30 Alec Flett <alecflett@chromium.org>
IndexedDB: Remove old get/getKey implementations
https://bugs.webkit.org/show_bug.cgi?id=87879
Reviewed by Ojan Vafai.
Removal of dead code, was scaffolding for a complicated
Chromium landing.
* Modules/indexeddb/IDBIndexBackendImpl.cpp:
(WebCore::IDBIndexBackendImpl::getInternal):
(WebCore::IDBIndexBackendImpl::getKeyInternal):
(WebCore::IDBIndexBackendImpl::get):
(WebCore::IDBIndexBackendImpl::getKey):
* Modules/indexeddb/IDBIndexBackendImpl.h:
(IDBIndexBackendImpl):
* Modules/indexeddb/IDBObjectStoreBackendImpl.cpp:
(WebCore::IDBObjectStoreBackendImpl::get):
(WebCore::IDBObjectStoreBackendImpl::getInternal):
* Modules/indexeddb/IDBObjectStoreBackendImpl.h:
(IDBObjectStoreBackendImpl):
2012-05-30 Joe Mason <jmason@rim.com>
[BlackBerry] Fix assertion fail on redirect due to multiple jobs per handle
https://bugs.webkit.org/show_bug.cgi?id=87579
Reviewed by George Staikos.
RIM PR #158892:
When we start a redirect, we now call cancelJob instead of just deleting it immediately to make sure
that all cleanup is performed. However, we also reassign the ResourceHandle to the new job, and
since cancelJob is asynchronous it is now assigned to two jobs simultaneously.
Work around this by only returning handles that have not been cancelled from findJobForHandle.
Cancelled jobs still technically exist in the jobs list, but they're invisible to callers. This is
safe because there is literally nothing that can be done with a cancelled job - it is supposed to
merely consume any notifications that are already in progress and then kill itself off - so no
callers of findJobForHandle are expecting a cancelled job. (All existing callers call methods on the
returned job which are no-ops for cancelled jobs, so there is no behaviour change.)
No new tests because there is no behaviour change (fixes a regression).
* platform/network/blackberry/NetworkManager.cpp:
(WebCore::NetworkManager::findJobForHandle):
2012-05-30 Kentaro Hara <haraken@chromium.org>
[V8][Performance] Optimize DOM attribute getters that return an empty string
https://bugs.webkit.org/show_bug.cgi?id=87820
Reviewed by Adam Barth.
This patch optimizes the performance of DOM attribute getters that return an
empty string by 65.8% (e.g. div.id, div.className, div.title, div.lang ...etc)
Performance test: Bindings/undefined-id-getter.html
[Without the patch]
RESULT Bindings: undefined-id-getter= 441.984669644 runs/s
median= 445.544554455 runs/s, stdev= 11.8406070959 runs/s, min= 405.063291139 runs/s, max= 447.204968944 runs/s
[With the patch]
RESULT Bindings: undefined-id-getter= 738.840460474 runs/s
median= 738.786279683 runs/s, stdev= 2.00249797161 runs/s, min= 733.944954128 runs/s, max= 741.721854305 runs/s
- This patch passes Isolate to v8::String::Empty(), which removes Isolate
look-up inside v8::String::Empty().
- This patch inlines V8Binding::getElementStringAttr() and
V8Binding::setElementStringAttr().
No tests. No change in behavior.
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateNormalAttrGetter):
(GenerateNormalAttrSetter):
* bindings/v8/V8Binding.cpp:
(WebCore::StringCache::v8ExternalStringSlow):
* bindings/v8/V8Binding.h:
(WebCore::StringCache::v8ExternalString):
(StringCache):
(WebCore::v8ExternalString):
(WebCore):
2012-05-30 Kentaro Hara <haraken@chromium.org>
[V8] Implement V8Binding::v8BooleanWithCheck(isolate)
https://bugs.webkit.org/show_bug.cgi?id=87814
Reviewed by Adam Barth.
The objective is to pass Isolate to v8::True() and v8::False().
Similarly to v8NullWithCheck() (bug 87713), this patch implements
V8Binding::v8BooleanWithCheck(isolate), which calls v8::True()/v8::False()
or v8::True(isolate)/v8::False(isolate) depending on whether the
passed isolate is null or not.
This patch also passes Isolate to v8Boolean() in CodeGeneratorV8.pm.
No tests. No behavior change.
* bindings/scripts/CodeGeneratorV8.pm:
(NativeToJSValue):
* bindings/scripts/test/V8/V8TestEventTarget.cpp:
(WebCore::TestEventTargetV8Internal::dispatchEventCallback):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::TestObjV8Internal::createAttrGetter):
(WebCore::TestObjV8Internal::reflectedBooleanAttrAttrGetter):
(WebCore::TestObjV8Internal::reflectedCustomBooleanAttrAttrGetter):
* bindings/v8/V8Binding.h:
(WebCore::v8Boolean):
(WebCore):
(WebCore::v8BooleanWithCheck):
2012-05-30 Kentaro Hara <haraken@chromium.org>
[V8] Replace v8::True() and v8::False() with v8Boolean()
https://bugs.webkit.org/show_bug.cgi?id=87812
Reviewed by Adam Barth.
To avoid implementing all of v8TrueWithCheck(isolate), v8FalseWithCheck(isolate)
and v8BooleanWithCheck(isolate), we can first replace v8::True() and v8::False()
with v8Boolean(). Then we can just implement v8BooleanWithCheck(isolate).
No tests. No change in behavior.
* bindings/v8/SerializedScriptValue.cpp:
* bindings/v8/V8Binding.h:
(WebCore::v8StringOrFalse):
* bindings/v8/V8NPUtils.cpp:
(WebCore::convertNPVariantToV8Object):
* bindings/v8/custom/V8DOMStringMapCustom.cpp:
(WebCore::V8DOMStringMap::namedPropertyDeleter):
* bindings/v8/custom/V8InjectedScriptHostCustom.cpp:
(WebCore::V8InjectedScriptHost::isHTMLAllCollectionCallback):
* bindings/v8/custom/V8StorageCustom.cpp:
(WebCore::storageDeleter):
2012-05-30 Kentaro Hara <haraken@chromium.org>
[V8] Replace v8::Null() with v8NullWithCheck(isolate) in custom bindings where isolate can be 0
https://bugs.webkit.org/show_bug.cgi?id=87807
Reviewed by Adam Barth.
The objective is to pass Isolate around in V8 bindings.
This patch replaces v8::Null() with v8NullWithCheck(isolate)
in custom bindings where isolate can be 0.
'where isolate can be 0' means all the places where we cannot
retrieve Isolate directly from AccessorInfo or Arguments.
No tests. No behavior change.
* bindings/v8/custom/V8BlobCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8CSSRuleCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8CSSStyleSheetCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8CSSValueCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DOMStringMapCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DOMTokenListCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DOMWindowCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DataViewCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8EntryCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8EntrySyncCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8EventCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Float32ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Float64ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8HTMLDocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8HTMLElementCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8IDBAnyCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8IDBKeyCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ImageDataCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Int16ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Int32ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Int8ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8LocationCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8NamedNodeMapCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8NodeCustom.cpp:
(WebCore::toV8Slow):
* bindings/v8/custom/V8SVGDocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8SVGElementCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8SVGPathSegCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ScriptProfileCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ScriptProfileNodeCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8StyleSheetCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Uint16ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Uint32ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Uint8ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Uint8ClampedArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8WorkerContextCustom.cpp:
(WebCore::toV8):
2012-05-30 Kentaro Hara <haraken@chromium.org>
[V8] Replace v8::Null() with v8::Null(isolate) or v8NullWithCheck(isolate) in non-custom bindings
https://bugs.webkit.org/show_bug.cgi?id=87810
Reviewed by Adam Barth.
The objective is to pass Isolate around in V8 bindings.
For non-custom bindings, we replace v8::Null() with v8::Null(isolate)
where isolate cannot be 0, and replace v8::Null() with v8NullWithCheck(isolate)
where isolate can be 0.
This will be the final patch for passing Isolate to v8::Null().
No tests. No change in behavior.
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateNormalAttrGetter):
(GenerateFunctionCallback):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::TestObjV8Internal::contentDocumentAttrGetter):
(WebCore::TestObjV8Internal::getSVGDocumentCallback):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.cpp:
(WebCore::TestSerializedScriptValueInterfaceV8Internal::cachedValueAttrGetter):
(WebCore::TestSerializedScriptValueInterfaceV8Internal::cachedReadonlyValueAttrGetter):
* bindings/v8/SerializedScriptValue.cpp:
(WebCore::SerializedScriptValue::deserialize):
* bindings/v8/V8DOMWrapper.cpp:
(WebCore::V8DOMWrapper::convertEventTargetToV8Object):
2012-05-30 Kentaro Hara <haraken@chromium.org>
[V8] Replace v8::Null() with v8::Null(isolate) in custom bindings where isolate shouldn't be 0
https://bugs.webkit.org/show_bug.cgi?id=87809
Reviewed by Adam Barth.
The objective is to pass Isolate around in V8 bindings.
This patch replaces v8::Null() with v8::Null(isolate) in custom bindings
where isolate shouldn't be 0. 'where isolate shouldn't be 0' is the place
where isolate is retrieved directly from AccessorInfo and Arguments.
This patch includes a couple of refactorings:
- Makes V8DeviceMotionEvent::createAccelerationObject() and
V8DeviceMotionEvent::createRotationRateObject() static
- Moves V8HTMLElement::toV8Object() into the existing ENABLE(MICRODATA) flag.
No tests. No change in behavior.
* bindings/v8/custom/V8ClipboardCustom.cpp:
(WebCore::V8Clipboard::typesAccessorGetter):
* bindings/v8/custom/V8CoordinatesCustom.cpp:
(WebCore::V8Coordinates::altitudeAccessorGetter):
(WebCore::V8Coordinates::altitudeAccuracyAccessorGetter):
(WebCore::V8Coordinates::headingAccessorGetter):
(WebCore::V8Coordinates::speedAccessorGetter):
* bindings/v8/custom/V8DeviceMotionEventCustom.cpp:
(WebCore::V8DeviceMotionEvent::accelerationAccessorGetter):
(WebCore::V8DeviceMotionEvent::accelerationIncludingGravityAccessorGetter):
(WebCore::V8DeviceMotionEvent::rotationRateAccessorGetter):
(WebCore::V8DeviceMotionEvent::intervalAccessorGetter):
* bindings/v8/custom/V8DeviceOrientationEventCustom.cpp:
(WebCore::V8DeviceOrientationEvent::alphaAccessorGetter):
(WebCore::V8DeviceOrientationEvent::betaAccessorGetter):
(WebCore::V8DeviceOrientationEvent::gammaAccessorGetter):
(WebCore::V8DeviceOrientationEvent::absoluteAccessorGetter):
* bindings/v8/custom/V8DocumentLocationCustom.cpp:
(WebCore::V8Document::locationAccessorGetter):
* bindings/v8/custom/V8HTMLCanvasElementCustom.cpp:
(WebCore::V8HTMLCanvasElement::getContextCallback):
* bindings/v8/custom/V8HTMLElementCustom.cpp:
(WebCore::toV8):
(WebCore):
(WebCore::toV8Object):
* bindings/v8/custom/V8HistoryCustom.cpp:
(WebCore::V8History::stateAccessorGetter):
* bindings/v8/custom/V8MessageEventCustom.cpp:
(WebCore::V8MessageEvent::dataAccessorGetter):
* bindings/v8/custom/V8NodeCustom.cpp:
(WebCore::V8Node::insertBeforeCallback):
(WebCore::V8Node::replaceChildCallback):
(WebCore::V8Node::removeChildCallback):
(WebCore::V8Node::appendChildCallback):
* bindings/v8/custom/V8PopStateEventCustom.cpp:
(WebCore::V8PopStateEvent::stateAccessorGetter):
* bindings/v8/custom/V8SQLResultSetRowListCustom.cpp:
(WebCore::V8SQLResultSetRowList::itemCallback):
* bindings/v8/custom/V8TrackEventCustom.cpp:
(WebCore::V8TrackEvent::trackAccessorGetter):
* bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
(WebCore::toV8Object):
(WebCore::V8WebGLRenderingContext::getAttachedShadersCallback):
(WebCore::V8WebGLRenderingContext::getSupportedExtensionsCallback):
2012-05-30 Nico Weber <thakis@chromium.org>
Make the files attribute of HTMLInputElement writable
https://bugs.webkit.org/show_bug.cgi?id=87154
Reviewed by Adam Barth.
whatwg thread:
http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2012-May/036140.html
* bindings/objc/PublicDOMInterfaces.h:
Remove readonly on files property.
* html/FileInputType.cpp:
(FileInputType):
Add a NULL check to setFiles().
* html/FileInputType.h:
(FileInputType):
Let setFiles() overwrite the base class's method.
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::setFiles):
Add setFiles() implementation, delegate to input type.
* html/HTMLInputElement.h:
(HTMLInputElement):
* html/HTMLInputElement.idl:
* html/InputType.cpp:
(WebCore::InputType::setFiles):
setFiles() does nothing by default.
* html/InputType.h:
(InputType):
Add a virtual setFiles() method.
2012-05-30 Joe Mason <jmason@rim.com>
[BlackBerry] Remove useless string->char*->string conversion in NetworkJob.
https://bugs.webkit.org/show_bug.cgi?id=87861
Reviewed by Antonio Gomes.
RIM PR #152383: username is a WTF::String. ResourceHandleInternal::m_user is a string. Yet they're
assigned with "username = m_user.utf8().data()". Pointless: "username = m_user" will do.
No new tests since there is no behaviour change.
* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::sendRequestWithCredentials):
2012-05-30 Dan Bernstein <mitz@apple.com>
[Qt] REGRESSION(r118568): It broke fast/forms/textarea-maxlength.html on Qt 4.8
https://bugs.webkit.org/show_bug.cgi?id=87740
Reviewed by Darin Adler.
* platform/text/qt/TextBreakIteratorQt.cpp:
(WebCore::NonSharedCharacterBreakIterator::NonSharedCharacterBreakIterator): Changed to use
the return value of setUpIterator.
2012-05-30 Zeno Albisser <zeno@webkit.org>
GraphicsSurface: allow importing and exporting of textures directly on GPU side.
https://bugs.webkit.org/show_bug.cgi?id=87725
Add copyFromFramebuffer function to copy a texture from an FBO
onto the GraphicsSurface.
Add getTextureID function to export the texture ID.
This allows binding/blitting the texture directly on the GPU.
Reviewed by Noam Rosenthal.
* platform/graphics/surfaces/GraphicsSurface.cpp:
(WebCore::GraphicsSurface::getTextureID):
(WebCore):
(WebCore::GraphicsSurface::copyFromFramebuffer):
* platform/graphics/surfaces/GraphicsSurface.h:
(GraphicsSurface):
* platform/graphics/surfaces/mac/GraphicsSurfaceMac.cpp:
(WebCore::GraphicsSurface::platformGetTextureID):
(WebCore):
(WebCore::GraphicsSurface::platformCopyFromFramebuffer):
2012-05-30 Alec Flett <alecflett@chromium.org>
IndexedDB: Implement IDBTransaction.error and IDBRequest.error
https://bugs.webkit.org/show_bug.cgi?id=87865
Reviewed by Tony Chang.
Added "error" attribute to IDBRequest and IDBTransaction. Update
IDBDatabaseError to honor the IDBDatabaseException behavior
of dealing with IDB-specific throws of DOMException codes,
as per the spec.
Existing tests which previously tested 'errorCode' and
'webkitErrorMessage' have been updated to use the new attribute.
* Modules/indexeddb/IDBCursor.cpp:
(WebCore::IDBCursor::advance):
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::setVersion):
* Modules/indexeddb/IDBDatabaseError.h:
(WebCore::IDBDatabaseError::create):
(WebCore::IDBDatabaseError::code):
(WebCore::IDBDatabaseError::idbCode):
(WebCore::IDBDatabaseError::name):
(WebCore::IDBDatabaseError::IDBDatabaseError):
(IDBDatabaseError):
* Modules/indexeddb/IDBDatabaseException.cpp:
(WebCore):
(WebCore::getErrorEntry):
(WebCore::IDBDatabaseException::initializeDescription):
(WebCore::IDBDatabaseException::getErrorName):
(WebCore::IDBDatabaseException::getLegacyErrorCode):
* Modules/indexeddb/IDBDatabaseException.h:
(IDBDatabaseException):
* Modules/indexeddb/IDBObjectStore.cpp:
(WebCore::IDBObjectStore::createIndex):
* Modules/indexeddb/IDBRequest.cpp:
(WebCore::IDBRequest::error):
(WebCore):
(WebCore::IDBRequest::resetReadyState):
(WebCore::IDBRequest::abort):
(WebCore::IDBRequest::onError):
(WebCore::IDBRequest::onSuccess):
(WebCore::IDBRequest::onSuccessWithContinuation):
(WebCore::IDBRequest::dispatchEvent):
* Modules/indexeddb/IDBRequest.h:
(IDBRequest):
* Modules/indexeddb/IDBRequest.idl:
* Modules/indexeddb/IDBTransaction.cpp:
(WebCore::IDBTransaction::error):
(WebCore):
(WebCore::IDBTransaction::setError):
* Modules/indexeddb/IDBTransaction.h:
(IDBTransaction):
* Modules/indexeddb/IDBTransaction.idl:
2012-05-30 Allan Sandfeld Jensen <allan.jensen@nokia.com>
[Qt] ImageDecoderQt doesn't support some decoders.
https://bugs.webkit.org/show_bug.cgi?id=87851
Reviewed by Kenneth Rohde Christiansen.
If forced to decode GIF and JPEG, ImageDecoderQt show several
artifacts. This is caused by mismatched decoded image-formats.
Convert mismatched color spaces into ARGB color space, and perform
manual mem-copy when the image has been decoded into a new buffer.
This patch also add support for down-scaling in the decoder.
* platform/graphics/qt/ImageDecoderQt.cpp:
(WebCore::ImageDecoderQt::internalDecodeSize):
(WebCore::ImageDecoderQt::internalHandleCurrentImage):
2012-05-30 Emil A Eklund <eae@chromium.org>
Floats wraps incorrectly when top edge has subpixel precision
https://bugs.webkit.org/show_bug.cgi?id=87688
Reviewed by Levi Weintraub.
When computing the logical left and right offsets for a float we
incorrectly floored the logical top when passing it to the float search
adapter. This caused it to return incorrect values where the floored and
pixel snapped values are not the same.
Test: fast/sub-pixel/float-wrap-with-subpixel-top.html
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::logicalLeftOffsetForLine):
(WebCore::RenderBlock::logicalRightOffsetForLine):
2012-05-30 Emil A Eklund <eae@chromium.org>
Inline blocks with margin wrap incorrectly with subpixel layout turned on
https://bugs.webkit.org/show_bug.cgi?id=87897
Reviewed by Levi Weintraub.
Change WebCore::getBorderPaddingMargin and the helper function
getBPMWidth to return a LayoutUnit rather than an int as the padding and
margin values have subpixel precision. This fixes an issue where we would
incorrectly wrap inline blocks with a subpixel padding or margin.
Test: fast/sub-pixel/inline-block-with-margin.html
* rendering/RenderBlock.cpp:
(WebCore::getBPMWidth):
(WebCore::getBorderPaddingMargin):
(WebCore::RenderBlock::computeInlinePreferredLogicalWidths):
2012-05-30 Tony Chang <tony@chromium.org>
rename flexbox css values from justify and distribute to space-between and space-around
https://bugs.webkit.org/show_bug.cgi?id=87867
Reviewed by Ojan Vafai.
This was recently changed in the spec:
http://dev.w3.org/csswg/css3-flexbox/#justify-content
No new tests, updated existing tests and expectations.
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
(WebCore::CSSPrimitiveValue::operator EFlexPack):
(WebCore::CSSPrimitiveValue::operator EFlexLinePack):
* css/CSSValueKeywords.in:
* rendering/RenderFlexibleBox.cpp:
(WebCore::initialPackingOffset):
(WebCore::packingSpaceBetweenChildren):
(WebCore::initialLinePackingOffset):
(WebCore::linePackingSpaceBetweenChildren):
* rendering/style/RenderStyleConstants.h:
2012-05-30 Kevin Ollivier <kevino@theolliviers.com>
[wx] Unreviewed build fix. Make sure we get the file size with the proper native type.
* platform/wx/FileSystemWx.cpp:
(WebCore::getFileMetadata):
2012-05-30 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118913.
http://trac.webkit.org/changeset/118913
https://bugs.webkit.org/show_bug.cgi?id=87891
patch is incorrect (see comments in
https://bugs.webkit.org/show_bug.cgi?id=87889) (Requested by
thorton on #webkit).
* css/CSSFontFaceSrcValue.cpp:
(WebCore::CSSFontFaceSrcValue::addSubresourceStyleURLs):
* css/CSSFontFaceSrcValue.h:
(CSSFontFaceSrcValue):
* css/CSSPrimitiveValue.cpp:
(WebCore::CSSPrimitiveValue::addSubresourceStyleURLs):
* css/CSSPrimitiveValue.h:
(CSSPrimitiveValue):
* css/CSSReflectValue.cpp:
(WebCore::CSSReflectValue::addSubresourceStyleURLs):
* css/CSSReflectValue.h:
(CSSReflectValue):
* css/CSSValue.cpp:
(WebCore::CSSValue::addSubresourceStyleURLs):
* css/CSSValue.h:
(CSSValue):
* css/CSSValueList.cpp:
(WebCore::CSSValueList::addSubresourceStyleURLs):
* css/CSSValueList.h:
(CSSValueList):
* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::addSubresourceStyleURLs):
* css/StylePropertySet.h:
(StylePropertySet):
* css/StyleSheetContents.cpp:
(WebCore::StyleSheetContents::addSubresourceStyleURLs):
* css/StyleSheetContents.h:
(StyleSheetContents):
* dom/Element.h:
* dom/Node.cpp:
(WebCore::Node::getSubresourceURLs):
* dom/Node.h:
(WebCore::Node::addSubresourceAttributeURLs):
(WebCore):
* dom/ProcessingInstruction.cpp:
(WebCore::ProcessingInstruction::addSubresourceAttributeURLs):
* dom/ProcessingInstruction.h:
(ProcessingInstruction):
* dom/StyledElement.cpp:
(WebCore::StyledElement::addSubresourceAttributeURLs):
* dom/StyledElement.h:
(StyledElement):
* html/HTMLBodyElement.cpp:
(WebCore::HTMLBodyElement::addSubresourceAttributeURLs):
* html/HTMLBodyElement.h:
(HTMLBodyElement):
* html/HTMLEmbedElement.cpp:
(WebCore::HTMLEmbedElement::addSubresourceAttributeURLs):
* html/HTMLEmbedElement.h:
(HTMLEmbedElement):
* html/HTMLImageElement.cpp:
(WebCore::HTMLImageElement::addSubresourceAttributeURLs):
* html/HTMLImageElement.h:
(HTMLImageElement):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::addSubresourceAttributeURLs):
* html/HTMLInputElement.h:
(HTMLInputElement):
* html/HTMLLinkElement.cpp:
(WebCore::HTMLLinkElement::addSubresourceAttributeURLs):
* html/HTMLLinkElement.h:
(HTMLLinkElement):
* html/HTMLObjectElement.cpp:
(WebCore::HTMLObjectElement::addSubresourceAttributeURLs):
* html/HTMLObjectElement.h:
(HTMLObjectElement):
* html/HTMLParamElement.cpp:
(WebCore::HTMLParamElement::addSubresourceAttributeURLs):
* html/HTMLParamElement.h:
(HTMLParamElement):
* html/HTMLScriptElement.cpp:
(WebCore::HTMLScriptElement::addSubresourceAttributeURLs):
* html/HTMLScriptElement.h:
(HTMLScriptElement):
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::addSubresourceAttributeURLs):
* html/HTMLStyleElement.h:
(HTMLStyleElement):
* html/HTMLTableCellElement.cpp:
(WebCore::HTMLTableCellElement::addSubresourceAttributeURLs):
* html/HTMLTableCellElement.h:
(HTMLTableCellElement):
* html/HTMLTableElement.cpp:
(WebCore::HTMLTableElement::addSubresourceAttributeURLs):
* html/HTMLTableElement.h:
(HTMLTableElement):
* svg/SVGCursorElement.cpp:
(WebCore::SVGCursorElement::addSubresourceAttributeURLs):
* svg/SVGCursorElement.h:
(SVGCursorElement):
* svg/SVGFEImageElement.cpp:
(WebCore::SVGFEImageElement::addSubresourceAttributeURLs):
* svg/SVGFEImageElement.h:
(SVGFEImageElement):
* svg/SVGImageElement.cpp:
(WebCore::SVGImageElement::addSubresourceAttributeURLs):
* svg/SVGImageElement.h:
(SVGImageElement):
* svg/SVGScriptElement.cpp:
(WebCore::SVGScriptElement::addSubresourceAttributeURLs):
* svg/SVGScriptElement.h:
(SVGScriptElement):
2012-05-30 Shezan Baig <shezbaig.wk@gmail.com>
Rename offsetTopLeft in RenderBoxModelObject to something better
https://bugs.webkit.org/show_bug.cgi?id=85915
Reviewed by Darin Adler.
Renamed offsetTopLeft in RenderBoxModelObject to
adjustedPositionRelativeToOffsetParent, because it returns the given
startPoint after adjusting it to be relative to the top-left corner of
the offsetParent. The definition of offsetParent itself is non-trivial
and is documented within the body of RenderObject::offsetParent,
therefore I decided to reuse this term, as-is, in the name of this
function.
No new tests; no functional or visible changes.
* rendering/RenderBox.cpp:
(WebCore::RenderBox::offsetLeft):
(WebCore::RenderBox::offsetTop):
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::adjustedPositionRelativeToOffsetParent):
(WebCore::RenderBoxModelObject::offsetLeft):
(WebCore::RenderBoxModelObject::offsetTop):
* rendering/RenderBoxModelObject.h:
(RenderBoxModelObject):
* rendering/RenderInline.cpp:
(WebCore::RenderInline::offsetLeft):
(WebCore::RenderInline::offsetTop):
2012-05-29 Adrienne Walker <enne@google.com>
Transformed fixed position layers have an incorrect overlap map entry
https://bugs.webkit.org/show_bug.cgi?id=64201
Reviewed by Darin Adler.
Previously, layers that both had a transform and were fixed position
were not considered as being fixed position in RenderGeometryMap or in
RenderBox::mapLocalToContainer (although this case is not incorrect in
the case of painting, so an external caller likely adjusts for this).
Tests: compositing/layer-creation/fixed-position-and-transform.html
compositing/layer-creation/fixed-position-under-transform.html
* rendering/RenderBox.cpp:
(WebCore::RenderBox::mapLocalToContainer):
* rendering/RenderGeometryMap.cpp:
(WebCore::RenderGeometryMap::mapToAbsolute):
2012-05-30 Ulan Degenbaev <ulan@chromium.org>
[v8] Crash after redefining setter on typed array to a number
https://bugs.webkit.org/show_bug.cgi?id=87862
Reviewed by Kenneth Russell.
Remove calls to typed array set() method from C++.
Store the copying script as a hidden property of a typed array prototype.
Test: fast/canvas/webgl/array-override-set.html
* bindings/v8/custom/V8ArrayBufferViewCustom.cpp:
(WebCore):
(WebCore::getHiddenCopyMethod):
(WebCore::installHiddenCopyMethod):
(WebCore::copyElements):
* bindings/v8/custom/V8ArrayBufferViewCustom.h:
(WebCore):
(WebCore::constructWebGLArray):
(WebCore::setWebGLArrayHelper):
* bindings/v8/custom/V8ArrayBufferViewCustomScript.js:
2012-05-30 David Hyatt <hyatt@apple.com>
https://bugs.webkit.org/show_bug.cgi?id=87882
[New Multicolumn] Rename layoutColumns to relayoutForPagination and make it
virtual so that it can be implemented by RenderMultiColumnBlock.
Reviewed by Anders Carlsson.
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::layoutBlock):
(WebCore::RenderBlock::relayoutForPagination):
* rendering/RenderBlock.h:
(RenderBlock):
* rendering/RenderMultiColumnBlock.cpp:
(WebCore::RenderMultiColumnBlock::relayoutForPagination):
(WebCore):
* rendering/RenderMultiColumnBlock.h:
(RenderMultiColumnBlock):
2012-05-30 Daniel Drake <dsd@laptop.org>
Check for GTK2/GTK3 symbol mismatch earlier
https://bugs.webkit.org/show_bug.cgi?id=87687
Reviewed by Martin Robinson.
No new tests. Regressions in core behavior are covered by existing
plugin tests and the fix deals with particular aspects of the system
environment that are difficult to test.
Detect plugins that would mix GTK+ symbols earlier, so that the
WebKit can skip them and choose a more appropriate plugin module.
* plugins/gtk/PluginPackageGtk.cpp: Move this code from PluginViewGtk.
(WebCore::moduleMixesGtkSymbols):
(WebCore::PluginPackage::load):
* plugins/gtk/PluginViewGtk.cpp: Move this code to PluginPackageGtk.
(WebCore::PluginView::platformStart):
2012-05-30 Allan Sandfeld Jensen <allan.jensen@nokia.com>
[Qt] Support IMAGE_DECODER_DOWN_SAMPLING flag.
https://bugs.webkit.org/show_bug.cgi?id=87840
Based on patches by Qi Zhang and Jussi Lehto.
Reviewed by Kenneth Rohde Christiansen.
* platform/graphics/Image.cpp:
(WebCore::Image::adjustSourceRectForDownSampling):
* platform/graphics/Image.h:
(Image):
* platform/graphics/openvg/ImageOpenVG.cpp:
(WebCore::BitmapImage::draw):
* platform/graphics/qt/ImageQt.cpp:
(WebCore::Image::drawPattern):
(WebCore::BitmapImage::draw):
2012-05-30 Shinya Kawanaka <shinyak@chromium.org>
comparePositions in htmlediting should consider nested Shadow DOM.
https://bugs.webkit.org/show_bug.cgi?id=87623
Reviewed by Ryosuke Niwa.
comparePositions in htmlediting.cpp didn't consider nested Shadow DOM.
This patch makes it possible to compare the nodes in nested Shadow DOM.
The algorithm is: we calculate the common tree scope of node A and B, and
adjust the nodes to the treeScope, then compare them. If the adjusted nodes
are the same, we check the shadow descendent of each node.
Test: editing/shadow/compare-positions-in-nested-shadow.html
* dom/TreeScope.cpp:
(WebCore::listTreeScopes):
(WebCore):
(WebCore::commonTreeScope):
* dom/TreeScope.h:
(WebCore):
* editing/htmlediting.cpp:
(WebCore::comparePositions):
2012-05-30 Erik Arvidsson <arv@chromium.org>
[V8] Clean up V8LazyEventListener to use one less function call
https://bugs.webkit.org/show_bug.cgi?id=87785
Reviewed by Kentaro Hara.
Instead of relying on 'arguments' we rely on 'this'. 'this' cannot be intercepted by a
with-statement so it is safe to use that to bind the context objects. This allows us to
remove one layer of function indirection in the generated code.
Covered by existing tests.
* bindings/v8/V8LazyEventListener.cpp:
(WebCore::V8LazyEventListener::prepareListenerObject):
2012-05-30 Tim Horton <timothy_horton@apple.com>
Factor DeferrableOneShotTimer out of GraphicsContextCG/GeneratorGeneratedImage
https://bugs.webkit.org/show_bug.cgi?id=87824
Reviewed by Simon Fraser.
DeferrableOneShotTimer is a timer which takes a class and method to call (as Timer does),
and a delay. It exposes restart(), stop(), and isActive(), and none of the rest of
the TimerBase interface. On restart(), the timer is started with the given delay,
unless it's already running, in which case a flag is set to automatically restart
the timer when it fires instead of calling the given method. This provides a
low-overhead way to implement a deferrable timer.
No new tests, refactoring.
* platform/Timer.h:
(DeferrableOneShotTimer):
(WebCore::DeferrableOneShotTimer::DeferrableOneShotTimer): New class.
(WebCore::DeferrableOneShotTimer::restart):
(WebCore::DeferrableOneShotTimer::fired):
* platform/graphics/GeneratorGeneratedImage.cpp:
(WebCore::GeneratorGeneratedImage::invalidateCacheTimerFired):
* platform/graphics/GeneratorGeneratedImage.h:
(WebCore::GeneratorGeneratedImage::GeneratorGeneratedImage): Make use of DeferrableOneShotTimer.
* platform/graphics/cg/GraphicsContextCG.cpp:
(WebCore::SubimageCacheWithTimer::SubimageCacheWithTimer): Make use of DeferrableOneShotTimer.
(WebCore::SubimageCacheWithTimer::invalidateCacheTimerFired):
2012-05-30 Xueqing Huang <huangxueqing@baidu.com>
eventSender.beginDragWithFiles should be implemented in Windows, which blocked drag and drop related tests.
https://bugs.webkit.org/show_bug.cgi?id=86296
Reviewed by Tony Chang.
Test: This change was requirement for eventSender.beginDragWithFiles, update win/Skipped file to remove drag and drop related tests will valiate it.
* platform/win/DragDataWin.cpp:
(WebCore::DragData::numberOfFiles):
2012-05-30 Caio Marcelo de Oliveira Filho <caio.oliveira@openbossa.org>
HashTable.h has using directives for std::pair and std::make_pair
https://bugs.webkit.org/show_bug.cgi?id=29919
Reviewed by Darin Adler.
Change code to use std::pair and std::make_pair.
* accessibility/AXObjectCache.cpp:
(WebCore::AXObjectCache::postNotification):
* html/MediaFragmentURIParser.cpp:
(WebCore::MediaFragmentURIParser::parseFragments):
* html/parser/HTMLMetaCharsetParser.cpp:
(WebCore::HTMLMetaCharsetParser::processMeta):
* loader/cache/CachedImage.cpp:
(WebCore::CachedImage::brokenImage):
* page/PageSerializer.cpp:
(WebCore::isCharsetSpecifyingNode):
2012-05-30 Peter Rybin <peter.rybin@gmail.com>
Web Inspector: disable ExactlyInt feature in InspectorTypeBuilder since it breaks release Windows 7 WK2
https://bugs.webkit.org/show_bug.cgi?id=87857
Reviewed by Yury Semikhatsky.
Code generator is patched to support old style of int parameter.
* inspector/CodeGeneratorInspector.py:
(TypeModel.init_class):
2012-05-30 Alexei Filippov <alexeif@chromium.org>
Web Inspector: Add trace() method to the WorkerConsole
https://bugs.webkit.org/show_bug.cgi?id=87656
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshotProxy.js:
(WebInspector.HeapSnapshotRealWorker.prototype._messageReceived):
* inspector/front-end/HeapSnapshotWorker.js:
(WebInspector.WorkerConsole.prototype.trace):
2012-05-30 Kinuko Yasuda <kinuko@chromium.org>
File.lastModifiedDate must return null if the modified time info is not available
https://bugs.webkit.org/show_bug.cgi?id=87709
Reviewed by Kentaro Hara.
Per File API spec, File.lastModifiedDate must return null if the
modified time info is not available.
http://dev.w3.org/2006/webapi/FileAPI/#dfn-lastModifiedDate
Test: http/tests/local/fileapi/file-last-modified-after-delete.html
* fileapi/File.cpp:
(WebCore::File::lastModifiedDateForBinding): Added.
(WebCore):
* fileapi/File.h:
(File):
* fileapi/File.idl: Added ImplementedAs=lastModifiedDateForBinding.
2012-05-30 Jochen Eisinger <jochen@chromium.org>
Match Firefox restrictions to window.blur and window.focus
https://bugs.webkit.org/show_bug.cgi?id=86969
Reviewed by Adam Barth.
Disallow window.blur altogether, and only allow window.focus to be
invoked from the context that created this very window.
There's a new setting (windowFocusRestricted) that defaults to true. If
false, the new restrictions are lifted. This can be used by ports that
would prefer to stick with the old behavior.
For tests, this setting is accessible in window.internals.settings
To temporarily allow window.focus(), an object of type
WindowFocusAllowedIndicator can be created. It is currently used to
enable window.focus() during dispatch of the click event for
notifications.
Tests: fast/dom/Window/mozilla-focus-blur.html
fast/notifications/notifications-click-event-focus.html
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* notifications/Notification.cpp:
(WebCore::Notification::dispatchClickEvent):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::focus):
(WebCore::DOMWindow::blur):
* page/DOMWindow.h:
(DOMWindow):
* page/DOMWindow.idl:
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(Settings):
(WebCore::Settings::setWindowFocusRestricted):
(WebCore::Settings::windowFocusRestricted):
* page/WindowFocusAllowedIndicator.cpp: Added.
(WebCore):
(WebCore::WindowFocusAllowedIndicator::windowFocusAllowed):
(WebCore::WindowFocusAllowedIndicator::WindowFocusAllowedIndicator):
(WebCore::WindowFocusAllowedIndicator::~WindowFocusAllowedIndicator):
* page/WindowFocusAllowedIndicator.h: Added.
(WebCore):
(WindowFocusAllowedIndicator):
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::InternalSettings):
(WebCore::InternalSettings::restoreTo):
(WebCore::InternalSettings::setWindowFocusRestricted):
(WebCore):
* testing/InternalSettings.h:
(InternalSettings):
* testing/InternalSettings.idl:
2012-05-30 Mariusz Grzegorczyk <mariusz.g@samsung.com>, Ryuan Choi <ryuan.choi@samsung.com>
[EFL][WK2] Fix WebKit2-EFL build
https://bugs.webkit.org/show_bug.cgi?id=83693
Reviewed by Carlos Garcia Campos.
* CMakeLists.txt: Add GestureTapHighlighter.cpp.
2012-05-29 Antti Koivisto <antti@apple.com>
Rename Foo::addSubresourceStyleURLs and Node::addSubresourceAttributeURLs
https://bugs.webkit.org/show_bug.cgi?id=87739
Rubber-stamped by Andreas Kling.
Rename them to collectSubresourceURLs and collectSubresourceURLsFromAttributes to match
the naming conventions.
Move collectSubresourceURLsFromAttributes from Node to Element as only elements have attributes.
* css/CSSFontFaceSrcValue.cpp:
(WebCore::CSSFontFaceSrcValue::collectSubresourceURLs):
* css/CSSFontFaceSrcValue.h:
(CSSFontFaceSrcValue):
* css/CSSPrimitiveValue.cpp:
(WebCore::CSSPrimitiveValue::collectSubresourceURLs):
* css/CSSPrimitiveValue.h:
(CSSPrimitiveValue):
* css/CSSReflectValue.cpp:
(WebCore::CSSReflectValue::collectSubresourceURLs):
* css/CSSReflectValue.h:
(CSSReflectValue):
* css/CSSValue.cpp:
(WebCore::CSSValue::collectSubresourceURLs):
* css/CSSValue.h:
(CSSValue):
* css/CSSValueList.cpp:
(WebCore::CSSValueList::collectSubresourceURLs):
* css/CSSValueList.h:
(CSSValueList):
* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::collectSubresourceURLs):
* css/StylePropertySet.h:
(StylePropertySet):
* css/StyleSheetContents.cpp:
(WebCore::StyleSheetContents::collectSubresourceURLs):
* css/StyleSheetContents.h:
(StyleSheetContents):
* dom/Element.h:
(WebCore::Element::collectSubresourceURLsFromAttributes):
(Element):
* dom/Node.cpp:
(WebCore::Node::getSubresourceURLs):
* dom/Node.h:
(WebCore):
* dom/ProcessingInstruction.cpp:
(WebCore::ProcessingInstruction::collectSubresourceURLsFromAttributes):
* dom/ProcessingInstruction.h:
(ProcessingInstruction):
* dom/StyledElement.cpp:
(WebCore::StyledElement::collectSubresourceURLsFromAttributes):
* dom/StyledElement.h:
(StyledElement):
* html/HTMLBodyElement.cpp:
(WebCore::HTMLBodyElement::collectSubresourceURLsFromAttributes):
* html/HTMLBodyElement.h:
(HTMLBodyElement):
* html/HTMLEmbedElement.cpp:
(WebCore::HTMLEmbedElement::collectSubresourceURLsFromAttributes):
* html/HTMLEmbedElement.h:
(HTMLEmbedElement):
* html/HTMLImageElement.cpp:
(WebCore::HTMLImageElement::collectSubresourceURLsFromAttributes):
* html/HTMLImageElement.h:
(HTMLImageElement):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::collectSubresourceURLsFromAttributes):
* html/HTMLInputElement.h:
(HTMLInputElement):
* html/HTMLLinkElement.cpp:
(WebCore::HTMLLinkElement::collectSubresourceURLsFromAttributes):
* html/HTMLLinkElement.h:
(HTMLLinkElement):
* html/HTMLObjectElement.cpp:
(WebCore::HTMLObjectElement::collectSubresourceURLsFromAttributes):
* html/HTMLObjectElement.h:
(HTMLObjectElement):
* html/HTMLParamElement.cpp:
(WebCore::HTMLParamElement::collectSubresourceURLsFromAttributes):
* html/HTMLParamElement.h:
(HTMLParamElement):
* html/HTMLScriptElement.cpp:
(WebCore::HTMLScriptElement::collectSubresourceURLsFromAttributes):
* html/HTMLScriptElement.h:
(HTMLScriptElement):
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::collectSubresourceURLsFromAttributes):
* html/HTMLStyleElement.h:
(HTMLStyleElement):
* html/HTMLTableCellElement.cpp:
(WebCore::HTMLTableCellElement::collectSubresourceURLsFromAttributes):
* html/HTMLTableCellElement.h:
(HTMLTableCellElement):
* html/HTMLTableElement.cpp:
(WebCore::HTMLTableElement::collectSubresourceURLsFromAttributes):
* html/HTMLTableElement.h:
(HTMLTableElement):
* svg/SVGCursorElement.cpp:
(WebCore::SVGCursorElement::collectSubresourceURLsFromAttributes):
* svg/SVGCursorElement.h:
(SVGCursorElement):
* svg/SVGFEImageElement.cpp:
(WebCore::SVGFEImageElement::collectSubresourceURLsFromAttributes):
* svg/SVGFEImageElement.h:
(SVGFEImageElement):
* svg/SVGImageElement.cpp:
(WebCore::SVGImageElement::collectSubresourceURLsFromAttributes):
* svg/SVGImageElement.h:
(SVGImageElement):
* svg/SVGScriptElement.cpp:
(WebCore::SVGScriptElement::collectSubresourceURLsFromAttributes):
* svg/SVGScriptElement.h:
(SVGScriptElement):
2012-05-30 Eugene Klyuchnikov <eustas.bug@gmail.com>
Web Inspector: Fix JS compiler warning
https://bugs.webkit.org/show_bug.cgi?id=87836
Reviewed by Yury Semikhatsky.
* inspector/front-end/SettingsScreen.js: Fix method annotation.
2012-05-30 Zoltan Horvath <zoltan@webkit.org>
[Qt] Set WebCore imagedecoders as default and add fallback to QImageDecoder
https://bugs.webkit.org/show_bug.cgi?id=80400
This change modifies the default ImageDecoder for Qt-port from QImageDecoder to WebCore ImageDecoder.
The new behavior is to use QImageDecoder only if WebCoreImageDecoder - and the platform - doesn't support
the requested image type.
The WTF_USE_QT_IMAGE_DECODER macro has been removed, since it is no longer needed.
This change adds build depedency (except Qt-MAC and Qt-WIN platform) for libpng-dev and libjpeg-dev packages,
because PNG and JPEG imagedecoders need not only these libraries, but their headers also. Qmake-config
tests for these libraries were introduced in r110045.
Notice: with using libjpeg-turbo as a system library I measured 10% performance improvement on Methanol benchmark
(collection of popular sites, page-loading benchmark) with this patch.
Reviewed by Simon Hausmann.
No new tests needed.
* Target.pri: Move WebCore ImageDecoder files out of guards. Remove ImageFrameQt.cpp from sources.
* WebCore.pri: Move WebCore ImageDecoder include paths out of guards.
* platform/MIMETypeRegistry.cpp:
(WebCore::initializeSupportedImageMIMETypes): Add WebCore supported and Qt supported MIME types.
(WebCore::initializeSupportedImageMIMETypesForEncoding): Use Qt supported MIME types.
* platform/graphics/ImageSource.cpp: Remove unnecessary includes.
* platform/graphics/ImageSource.h: Remove unnecessary typedefs.
(WebCore):
* platform/graphics/qt/ImageDecoderQt.cpp: Remove ImageDecoder::create, since it is no longer needed.
(WebCore::ImageDecoderQt::filenameExtension): Remove unnecessary semicolon.
(WebCore::ImageDecoderQt::internalHandleCurrentImage): Use QImage and ImageFrame instead of QPixmap.
(WebCore):
(WebCore::ImageFrame::asNewNativeImage): Moved here from removed ImageFrameQt.cpp.
* platform/image-decoders/ImageDecoder.cpp: Reorganize the includes of the header.
(WebCore::ImageDecoder::create): Add platform macro guarded fallback case for QImageDecoder. Add guards
for JPEG, PNG, ICO decoder for Qt-MAC and Qt-WIN platform.
* platform/image-decoders/ImageDecoder.h: Remove Qt-specific codes.
(WebCore::ImageFrame::getAddr): Remove Qt-specific case, since it is no longer needed.
(ImageFrame):
* platform/image-decoders/qt/ImageFrameQt.cpp: Removed. Dead code, other code has been moved to
ImageDecoderQt.cpp.
2012-05-23 Kinuko Yasuda <kinuko@chromium.org>
REGRESSION: We should allow null modificationTime when snapshot metadata is given
https://bugs.webkit.org/show_bug.cgi?id=86811
Reviewed by Jian Li.
r117432 has introduced a new File constructor which allows the caller
to pass in a snapshot file metadata. In the change we had considered the
given metadata is valid if "metadata.length >= 0 AND metadata.lastModifiedDate != 0",
but we should drop the latter condition (lastModifiedDate != 0) because
1. the value 0 is used to indicate the time information is unavailable in File, and
2. it is valid per spec (http://dev.w3.org/2006/webapi/FileAPI/#dfn-lastModifiedDate says the UA must return null if the information is not available).
(Note: the current js/v8 binding returns Date(0) for the time value 0,
which is still valid as epoch time but would fail to indicate the
unavailability of the information. In this patch I added FIXME in
File.idl and filed a separate issue http://webkit.org/b/87709)
No new tests as this change does not affect regular files/filesystems behavior.
(Tests in Chrome OS port should be able to verify this)
* fileapi/File.cpp:
(WebCore::File::lastModifiedDate):
(WebCore::File::size):
(WebCore::File::captureSnapshot):
* fileapi/File.h:
(File):
2012-05-30 MORITA Hajime <morrita@google.com>
[Shadow DOM] HTMLContentElement and HTMLShadowElement constructors should be visible.
https://bugs.webkit.org/show_bug.cgi?id=87818
Reviewed by Kentaro Hara.
Added them as DOMWindow properties
Tests: fast/dom/shadow/content-element-api.html
fast/dom/shadow/shadow-element.html
* page/DOMWindow.idl:
2012-05-30 Mike Lawther <mikelawther@chromium.org>
CSS3 calc: Implement CSSOM support
https://bugs.webkit.org/show_bug.cgi?id=84551
Reviewed by Ryosuke Niwa.
Adds a pretty printer for the calc expression tree.
Test: css3/calc/cssom.html
* css/CSSCalculationValue.cpp:
(WebCore::CSSCalcValue::customCssText):
(WebCore::CSSCalcPrimitiveValue::customCssText):
(WebCore::CSSCalcBinaryOperation::customCssText):
(CSSCalcBinaryOperation):
* css/CSSCalculationValue.h:
(CSSCalcExpressionNode):
2012-05-29 Li Yin <li.yin@intel.com>
[FileAPI] The result attribute of FileReader shuold use null to replace empty string
https://bugs.webkit.org/show_bug.cgi?id=87578
Reviewed by Jian Li.
Spec: http://www.w3.org/TR/FileAPI/#dfn-result
If the readyState is EMPTY (no read method has been called) or an error
in reading the File or Blob has occurred (using any read method), then the
result attribute MUST return null.
Currently, Firefox11, Opera and IE 10 follows the spec, but Webkit based
browser doesn't.
WebKit should change the returned value empty string into null to keep
conformance with the spec.
Tests: fast/files/read-file-async.html
fast/files/workers/worker-read-file-async.html
* fileapi/FileReader.cpp:
(WebCore::FileReader::arrayBufferResult):
(WebCore::FileReader::stringResult):
2012-05-29 Mark Pilgrim <pilgrim@chromium.org>
[Chromium] Move fileExists to Platform.h
https://bugs.webkit.org/show_bug.cgi?id=87531
Reviewed by Adam Barth.
Call fileExists through fileUtilities(), not directly.
* platform/chromium/FileSystemChromium.cpp:
(WebCore::fileExists):
2012-05-29 Kentaro Hara <haraken@chromium.org>
Remove a dummy style sheet from CSSParser::parseFlowThread()
https://bugs.webkit.org/show_bug.cgi?id=87715
Reviewed by Darin Adler.
r118736 changed CSSGrammar.y so that it works fine with no style sheet.
r118736 removed a dummy style sheet from CSSParser::parserSelector().
Now we can safely remove a dummy style sheet from parserFlowThread().
Tests: fast/regions. No change in test results.
* css/CSSParser.cpp:
(WebCore::CSSParser::parseFlowThread):
2012-05-29 Yoshifumi Inoue <yosin@chromium.org>
REGRESSION(r111497): The "option" element doesn't match CSS pseudo class :enabled
https://bugs.webkit.org/show_bug.cgi?id=87719
Reviewed by Kent Tamura.
This patch added checking of "option" element for CSS pseudo class :enabled as same as
:disabled to selector checker. Before r111497, it was done by using isFormControlElement.
After that revision, HTMLOptionElement was no longer derived from HTMLFormControlElement.
Test: fast/form/select/optgroup-rendering.html
* css/SelectorChecker.cpp:
(WebCore::SelectorChecker::checkOneSelector): Checking element is option element as same
as PseudoDisabled in PseudoEnabled case.
2012-05-29 Hayato Ito <hayato@chromium.org>
Introduces ComposedShadowTreeParentWalker, extracted from ComposedShadowTreeWalker.
https://bugs.webkit.org/show_bug.cgi?id=87004
Reviewed by Dimitri Glazkov.
Introduces a ComposedShadowTreeParentWalker, which is only used
for traversing a parent node (including shadow roots and insertion
points) and get rid of an equivalent function from
ComposedShadowTreeWalker.
Before this patch, there is an inconsistency inside of
ComposedShadowTreeWalker. The Walker uses 'Policy' to decide
whether it should visit shadow roots or not, but
parentIncludingInsertionPointAndShadowRoot() member function
ignores the policy. We can not add an assertion in its
constructor due to this inconsistency. To resolve it, we could
add yet another special policy, but that makes the implementation
complex and may add some overhead in runtime. So separate the
functionality into another class as ComposedShadowTreeParentWalker.
No new tests, no new functionality except for assertion.
* dom/ComposedShadowTreeWalker.cpp:
(WebCore::ComposedShadowTreeWalker::ComposedShadowTreeWalker):
(WebCore::ComposedShadowTreeParentWalker::ComposedShadowTreeParentWalker):
(WebCore):
(WebCore::ComposedShadowTreeParentWalker::parentIncludingInsertionPointAndShadowRoot):
(WebCore::ComposedShadowTreeParentWalker::traverseParentIncludingInsertionPointAndShadowRoot):
* dom/ComposedShadowTreeWalker.h:
(ComposedShadowTreeWalker):
(WebCore::ComposedShadowTreeWalker::assertPrecondition):
(WebCore):
(ComposedShadowTreeParentWalker):
(WebCore::ComposedShadowTreeParentWalker::get):
* dom/EventDispatcher.cpp:
(WebCore::EventRelatedTargetAdjuster::adjust):
(WebCore::EventDispatcher::ensureEventAncestors):
* page/EventHandler.cpp:
(WebCore::EventHandler::updateMouseEventTargetNode):
2012-05-29 Hayato Ito <hayato@chromium.org>
Add assertions to make sure that event's target and relatedTarget are accessible.
https://bugs.webkit.org/show_bug.cgi?id=87641
Reviewed by Dimitri Glazkov.
No new tests. No new functionality except for assertions.
* dom/EventContext.cpp:
(WebCore::EventContext::EventContext):
* dom/EventContext.h:
(WebCore):
(EventContext):
(WebCore::EventContext::setRelatedTarget):
(WebCore::EventContext::accessible):
2012-05-29 Kent Tamura <tkent@chromium.org>
Fix inefficient usage of traverseNextNode()
https://bugs.webkit.org/show_bug.cgi?id=87804
Reviewed by Ryosuke Niwa.
foo->traverseNextNode(foo) is equivalent to foo->firstChild(), and
firstChild() is more efficient.
No new tests because of no behavior change.
* html/HTMLFieldSetElement.cpp:
(WebCore::HTMLFieldSetElement::invalidateDisabledStateUnder):
Replace traverseNextNode() with firstChild(), and shorten a variable name.
* svg/SVGSVGElement.cpp:
(WebCore::SVGSVGElement::getElementById):
Replace traverseNextNode() with firstChild().
2012-05-29 Jonathan Dong <jonathan.dong@torchmobile.com.cn>
[BlackBerry] Loading media data with http authentication
https://bugs.webkit.org/show_bug.cgi?id=84214
Reviewed by George Staikos.
RIM PR: 117618
Implemented http authentication feature for media by implementing
two interface functions in class MediaPlayerPrivate:
onAuthenticationNeeded(): this function is triggered when MMR
engine requires http authentication. We search the CredentialStorage
to see if we have already stored existing credential information,
or challenge user to provide it.
OnAuthenticationAccepted(): this function is triggered when MMR
engine accepts the credential information, and we need to save
it in CredentialStorage for later use.
Internally reviewed by Max Feil <mfeil@qnx.com>.
Manual test case: blackberry/video-load-with-authentication.html
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.cpp:
(WebCore::generateProtectionSpaceFromMMRAuthChallenge):
(WebCore):
(WebCore::MediaPlayerPrivate::onAuthenticationNeeded):
(WebCore::MediaPlayerPrivate::onAuthenticationAccepted):
* platform/graphics/blackberry/MediaPlayerPrivateBlackBerry.h:
(MediaPlayerPrivate):
2012-05-29 MORITA Hajime <morrita@google.com>
[Shadow DOM] Node distribution should be orthogonal from node attachment
https://bugs.webkit.org/show_bug.cgi?id=87223
Reviewed by Dimitri Glazkov.
This chagne reorganizes Shadow DOM subtree distribution implementation.
Originally, it was interleaved across attach() of several classes like
InsertionPoint and ShadowRoot. Its invalidation was also mixed as a part of
the style recalculation and detach()-es.
This change extracts these bits of code to a set of ContentDistributor methods, which are
facaded by two ElementShadow API. Following two API are the primary entry points:
- ElementShadow::ensureDistribution()
- ElementShadow::invalidateDistribution()
The actual implementations are ContentDistributor::distribute() and
ContentDistributor::invalidate() respectively.
When clients need to traverse composed tree, before attach() for
example, they should call ensureDistribution() to make sure that
the traversal data structure ("the distribution") is ready. When
there is any DOM mutation which can result a composed tree
mutation, then clients should call invalidateDistribution() to
mark the distribution being dated.
Here are such DOM mutations:
- The children of any ShadowRoots are changed,
- The children of any InsertionPoints are changed,
- The children of any host elements are changed,
- Any insertion point is inserted to or removed from the shadow tree,
- @select attribute of <content> is modified and
- New ShadowRoot is added to the shadow tree.
Note that the validity of the distribution is tracked and
unnecessary distribution requests are ignored.
After the invalidation, that shadow subtrees are detached once and
request their re-attachment through the style recalculation.
Then, on the responding style recalculation and attach(), new
distribution will be computed.
No new tests. Covered by existing tests.
* dom/Element.cpp:
(WebCore::Element::~Element):
(WebCore::Element::childrenChanged):
* dom/ElementShadow.cpp:
(WebCore::ElementShadow::~ElementShadow):
(WebCore::ElementShadow::addShadowRoot):
(WebCore::ElementShadow::removeAllShadowRoots):
(WebCore::ElementShadow::attach):
(WebCore::ElementShadow::recalcStyle):
(WebCore::ElementShadow::ensureDistribution):
(WebCore::ElementShadow::invalidateDistribution):
* dom/ElementShadow.h:
(ElementShadow):
* dom/NodeRenderingContext.cpp:
(WebCore::NodeRenderingContext::NodeRenderingContext):
(WebCore::NodeRendererFactory::createRendererIfNeeded):
* dom/NodeRenderingContext.h:
(NodeRenderingContext):
* dom/ShadowRoot.cpp:
(WebCore::ShadowRoot::setApplyAuthorStyles):
(WebCore::ShadowRoot::attach):
(WebCore::ShadowRoot::childrenChanged):
(WebCore):
* dom/ShadowRoot.h:
(ShadowRoot):
* html/HTMLFormControlElement.cpp:
* html/ValidationMessage.cpp:
(WebCore::ValidationMessage::buildBubbleTree):
* html/shadow/ContentDistributor.cpp:
(WebCore::ContentDistributor::ContentDistributor):
(WebCore::ContentDistributor::~ContentDistributor):
(WebCore::ContentDistributor::findInsertionPointFor):
(WebCore::ContentDistributor::distribute):
(WebCore::ContentDistributor::invalidate):
(WebCore::ContentDistributor::finishInivalidation):
(WebCore::ContentDistributor::distributeSelectionsTo):
(WebCore::ContentDistributor::distributeShadowChildrenTo):
(WebCore::ContentDistributor::invalidateDistributionIn):
* html/shadow/ContentDistributor.h:
(WebCore::ContentDistributor::needsInvalidation):
(ContentDistributor):
(WebCore::ContentDistributor::needsDistribution):
* html/shadow/HTMLContentElement.cpp:
(WebCore::HTMLContentElement::parseAttribute):
* html/shadow/InsertionPoint.cpp:
(WebCore::InsertionPoint::attach):
(WebCore::InsertionPoint::detach):
(WebCore::InsertionPoint::nextTo):
(WebCore::InsertionPoint::previousTo):
(WebCore::InsertionPoint::childrenChanged):
* html/shadow/InsertionPoint.h:
(WebCore::InsertionPoint::setDistribution):
(WebCore::InsertionPoint::clearDistribution):
(InsertionPoint):
2012-05-29 Luke Macpherson <macpherson@chromium.org>
Implement post-landing feedback for WebKitCSSTransformValue::customCSSText().
https://bugs.webkit.org/show_bug.cgi?id=87684
Reviewed by Darin Adler.
Darin provided some style suggesting on https://bugs.webkit.org/show_bug.cgi?id=87462 after it landed.
This patch incorporates those suggestions, namely using 0 instead of empty string for UnknownTransformOperation
and WTF_ARRAY_LENGTH instead of hard coding the array length.
Covered by existing transform tests.
* css/WebKitCSSTransformValue.cpp:
(WebCore):
(WebCore::WebKitCSSTransformValue::customCssText):
2012-05-29 Eric Seidel <eric@webkit.org>
Fix ENABLE_IFRAME_SEAMLESS to actually fully disable <iframe seamless>
https://bugs.webkit.org/show_bug.cgi?id=87646
Reviewed by Adam Barth.
In the process of moving the seamless feature out of github and into bugs.webkit.org
multiple versions of the shouldDisplaySeamlessly function got written
(and moved from HTMLIFrameElement to Document), but only one of them was wrapped
in ENABLE_IFRAME_SEAMLESS. HTMLIFrameElement was checking mayDisplaySeamlessly
directly (as was my original design), which got around the ENABLE_IFRAME_SEAMLESS check.
I've fixed this oversight, and the feature is now off when we tell it to be off.
This is covered by many existing tests. I've verified locally that
all tests fail when ENABLE_IFRAME_SEAMLESS is disabled instead of
only some of them.
* dom/SecurityContext.h:
(SecurityContext):
* html/HTMLIFrameElement.cpp:
(WebCore::HTMLIFrameElement::shouldDisplaySeamlessly):
2012-05-29 Scott Graham <scottmg@chromium.org>
Fix warning in preprocessing Source/WebCore/dom/DOMError.idl
https://bugs.webkit.org/show_bug.cgi?id=87763
Reviewed by Adam Barth.
Add newline at end of file to avoid warning:
../dom/DOMError.idl:35:2: warning: no newline at end of file
* dom/DOMError.idl:
2012-05-29 Mark Pilgrim <pilgrim@chromium.org>
[Chromium] Move fileExists to Platform.h
https://bugs.webkit.org/show_bug.cgi?id=87531
Reviewed by Adam Barth.
Part of a refactoring series. See tracking bug 82948.
* platform/chromium/FileSystemChromium.cpp:
(WebCore::fileExists):
* platform/chromium/PlatformSupport.h:
(PlatformSupport):
2012-05-29 Kentaro Hara <haraken@chromium.org>
[V8] Rename v8Null() to v8NullWithCheck()
https://bugs.webkit.org/show_bug.cgi?id=87713
Reviewed by Adam Barth.
In IRC, cmarcelo@ pointed out that v8::Null() and v8Null()
are confusing. This patch renames v8Null() to v8NullWithCheck().
No tests. No behavior change.
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateHeader):
* bindings/scripts/test/V8/V8TestActiveDOMObject.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestCustomNamedGetter.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestEventConstructor.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestEventTarget.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestException.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestInterface.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestMediaQueryListListener.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestNamedConstructor.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestNode.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestObj.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.h:
(WebCore::toV8):
* bindings/v8/V8Binding.h:
(WebCore::v8NullWithCheck):
(WebCore::v8DateOrNull):
2012-05-29 Christophe Dumez <christophe.dumez@intel.com>
Some Content Security Policy console messages are missing line numbers in JSC
https://bugs.webkit.org/show_bug.cgi?id=87558
Reviewed by Adam Barth.
Make ScriptController::eventHandlerPosition() behave the same way on
JSC than V8. This means that TextPosition::minimumPosition() is
returned if parser is null, instead of
TextPosition::belowRangePosition(). The line numbers are now correctly
printed in the output of Content Security Policy tests.
Tests: http/tests/security/contentSecurityPolicy/inline-script-blocked-javascript-url.html
http/tests/security/contentSecurityPolicy/javascript-url-blocked.html
* bindings/js/ScriptController.cpp:
(WebCore::ScriptController::eventHandlerPosition):
2012-05-29 Shawn Singh <shawnsingh@chromium.org>
[chromium] make WebTransformationMatrix object usable by non-webkit code
https://bugs.webkit.org/show_bug.cgi?id=87315
Reviewed by James Robinson.
This change is covered by existing tests and by whether it
compiles. There should be no changes in behavior.
* platform/chromium/support/WebTransformationMatrix.cpp:
(WebKit::WebTransformationMatrix::WebTransformationMatrix):
(WebKit::WebTransformationMatrix::operator=):
(WebKit::WebTransformationMatrix::operator==):
(WebKit::WebTransformationMatrix::inverse):
(WebKit::WebTransformationMatrix::to2dTransform):
(WebKit::WebTransformationMatrix::multiply):
(WebKit::WebTransformationMatrix::makeIdentity):
(WebKit::WebTransformationMatrix::translate):
(WebKit::WebTransformationMatrix::translate3d):
(WebKit::WebTransformationMatrix::translateRight3d):
(WebKit::WebTransformationMatrix::scale):
(WebKit::WebTransformationMatrix::scaleNonUniform):
(WebKit::WebTransformationMatrix::scale3d):
(WebKit::WebTransformationMatrix::rotate):
(WebKit::WebTransformationMatrix::rotate3d):
(WebKit::WebTransformationMatrix::skewX):
(WebKit::WebTransformationMatrix::skewY):
(WebKit::WebTransformationMatrix::applyPerspective):
(WebKit::WebTransformationMatrix::blend):
(WebKit::WebTransformationMatrix::hasPerspective):
(WebKit::WebTransformationMatrix::isInvertible):
(WebKit::WebTransformationMatrix::isBackFaceVisible):
(WebKit::WebTransformationMatrix::isIdentity):
(WebKit::WebTransformationMatrix::isIdentityOrTranslation):
(WebKit::WebTransformationMatrix::isIntegerTranslation):
(WebKit::WebTransformationMatrix::m11):
(WebKit::WebTransformationMatrix::setM11):
(WebKit::WebTransformationMatrix::m12):
(WebKit::WebTransformationMatrix::setM12):
(WebKit::WebTransformationMatrix::m13):
(WebKit::WebTransformationMatrix::setM13):
(WebKit::WebTransformationMatrix::m14):
(WebKit::WebTransformationMatrix::setM14):
(WebKit::WebTransformationMatrix::m21):
(WebKit::WebTransformationMatrix::setM21):
(WebKit::WebTransformationMatrix::m22):
(WebKit::WebTransformationMatrix::setM22):
(WebKit::WebTransformationMatrix::m23):
(WebKit::WebTransformationMatrix::setM23):
(WebKit::WebTransformationMatrix::m24):
(WebKit::WebTransformationMatrix::setM24):
(WebKit::WebTransformationMatrix::m31):
(WebKit::WebTransformationMatrix::setM31):
(WebKit::WebTransformationMatrix::m32):
(WebKit::WebTransformationMatrix::setM32):
(WebKit::WebTransformationMatrix::m33):
(WebKit::WebTransformationMatrix::setM33):
(WebKit::WebTransformationMatrix::m34):
(WebKit::WebTransformationMatrix::setM34):
(WebKit::WebTransformationMatrix::m41):
(WebKit::WebTransformationMatrix::setM41):
(WebKit::WebTransformationMatrix::m42):
(WebKit::WebTransformationMatrix::setM42):
(WebKit::WebTransformationMatrix::m43):
(WebKit::WebTransformationMatrix::setM43):
(WebKit::WebTransformationMatrix::m44):
(WebKit::WebTransformationMatrix::setM44):
(WebKit::WebTransformationMatrix::a):
(WebKit::WebTransformationMatrix::setA):
(WebKit::WebTransformationMatrix::b):
(WebKit::WebTransformationMatrix::setB):
(WebKit::WebTransformationMatrix::c):
(WebKit::WebTransformationMatrix::setC):
(WebKit::WebTransformationMatrix::d):
(WebKit::WebTransformationMatrix::setD):
(WebKit::WebTransformationMatrix::e):
(WebKit::WebTransformationMatrix::setE):
(WebKit::WebTransformationMatrix::f):
(WebKit::WebTransformationMatrix::setF):
(WebKit::WebTransformationMatrix::toWebCoreTransform):
(WebKit::WebTransformationMatrix::mapRect):
(WebKit::WebTransformationMatrix::mapPoint):
(WebKit::WebTransformationMatrix::mapQuad):
(WebKit::WebTransformationMatrix::projectPoint):
2012-05-29 Alexis Menard <alexis.menard@openbossa.org>
[CSS3 Backgrounds and Borders] Implement box-decoration-break parsing.
https://bugs.webkit.org/show_bug.cgi?id=87678
Reviewed by Eric Seidel.
Implement the parsing part of box-decoration-break. The specification
is located at http://www.w3.org/TR/css3-background/#box-decoration-break.
So far Opera implements it. This patch only implements the parsing part of
the property. A following patch will implement the rendering part. Despite
the rendering part missing it improves http://css3test.com by 1%.
Test: fast/box-decoration-break/box-decoration-break-parsing.html
* css/CSSComputedStyleDeclaration.cpp:
(WebCore):
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
(WebCore::isKeywordPropertyID):
* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
(WebCore):
(WebCore::CSSPrimitiveValue::operator EBoxDecorationBreak):
* css/CSSProperty.cpp:
(WebCore::CSSProperty::isInheritedProperty):
* css/CSSPropertyNames.in:
* css/CSSValueKeywords.in:
* css/StyleBuilder.cpp:
(WebCore::StyleBuilder::StyleBuilder):
* rendering/style/RenderStyle.h:
* rendering/style/RenderStyleConstants.h:
* rendering/style/StyleBoxData.cpp:
(WebCore::StyleBoxData::StyleBoxData):
(WebCore::StyleBoxData::operator==):
* rendering/style/StyleBoxData.h:
(WebCore::StyleBoxData::boxDecorationBreak):
(StyleBoxData):
2012-05-29 David Barr <davidbarr@chromium.org>
[text-overflow: ellipsis] WebKit should ignore the ellipsis for the purpose of dispatching mouse events
https://bugs.webkit.org/show_bug.cgi?id=63781
Reviewed by Eric Seidel.
"Ellipsing only affects rendering and must not affect layout nor dispatching of pointer events."
http://www.w3.org/TR/2012/WD-css3-ui-20120117/#text-overflow
Test: fast/css/text-overflow-ellipsis-hit-test.html
* rendering/EllipsisBox.cpp:
(WebCore::EllipsisBox::nodeAtPoint): Always return false.
2012-05-29 Gavin Peters <gavinp@chromium.org>
[Chromium] Remove assertions on state in Prerender.cpp
https://bugs.webkit.org/show_bug.cgi?id=87746
Chrome was crashing in the new Prerender API on Google Web Search; this was because the DOM would stop(), setting the
state of the Prerenders in the document to Inactive, but an asyncronous GC would later come come along and trigger the
removedFromDocument behaviour on each element, including the <link> element launching the prerender. This causes an
assertion failure.
I believe the late-deletion of the DOM, and the HTMLElement::removedFromDocument calls are legal, and it was the
overzealous assertions in Prerender that were inappropriate.
Reviewed by Adam Barth.
No new automated tests, working on a Chromium browser test. Manually test by entering many prerender and non-prerendered
queries into Google Web Search.
* platform/chromium/Prerender.cpp:
(WebCore::Prerender::Prerender):
(WebCore::Prerender::add):
(WebCore::Prerender::cancel):
(WebCore::Prerender::abandon):
* platform/chromium/Prerender.h:
(Prerender):
2012-05-29 Victor Carbune <victor@rosedu.org>
Display cues in the controls area
https://bugs.webkit.org/show_bug.cgi?id=87591
Reviewed by Eric Carlson.
Changed the behaviour to implicitly hide the volume slidebar
if the controls are hidden (due mouse moving out of the video
area or controls are toggled through JS).
This seems to be the simplest fix to this issue, otherwise
on toggling the display:none property on the controls the
volume slidebar might appear below the video.
Re-enabled failing tests for this functionality.
* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlPanelElement::makeTransparent): Re-enabled
timer to set the display:none property on the controls.
* html/shadow/MediaControlRootElement.cpp:
(WebCore::MediaControlRootElement::hide): Added a call to hide
the volume slider.
(WebCore::MediaControlRootElement::makeTransparent): Same.
* html/shadow/MediaControlRootElementChromium.cpp:
(WebCore::MediaControlRootElementChromium::hide): Same.
(WebCore::MediaControlRootElementChromium::makeTransparent): Same.
2012-05-29 Adrienne Walker <enne@google.com>
Fix transparent iframes with composited contents in WK1 on Mac
https://bugs.webkit.org/show_bug.cgi?id=76478
Reviewed by Simon Fraser.
Test: compositing/iframes/repaint-after-losing-scrollbars.html
* page/FrameView.cpp:
(WebCore::FrameView::useSlowRepaints):
2012-05-29 David Hyatt <hyatt@apple.com>
https://bugs.webkit.org/show_bug.cgi?id=87787
[New Multicolumn] Refactor the region and flow thread code so that it can pick up the correct column widths
and heights when determining the logical width and height of the flow thread itself.
This patch adds new virtual methods to RenderRegion: logicalWidthForFlowThreadContent and logicalHeightForFlowThreadContent.
For normal regions, these methods return the contentWidth and contentHeight of the regions themselves. For column sets,
however, these methods actually return the column widths and heights.
This allows the flow thread to properly lay out according to the column width.
Region invalidation is also done properly by using these new methods.
In some cases the code is actually simplified, since we can get rid of some isHorizontalWritingMode() switching.
With this patch the flow thread now actually lays out and paints at the right width (spilling out of the region set in a single unbroken
column).
Reviewed by Dan Bernstein.
* rendering/RenderFlowThread.cpp:
(WebCore::RenderFlowThread::layout):
(WebCore::RenderFlowThread::computeLogicalWidth):
(WebCore::RenderFlowThread::computeLogicalHeight):
* rendering/RenderMultiColumnBlock.cpp:
(WebCore::RenderMultiColumnBlock::ensureColumnSets):
* rendering/RenderMultiColumnBlock.h:
(WebCore::RenderMultiColumnBlock::isRenderMultiColumnBlock):
(RenderMultiColumnBlock):
(WebCore::toRenderMultiColumnBlock):
(WebCore):
* rendering/RenderMultiColumnSet.cpp:
(WebCore::RenderMultiColumnSet::computeLogicalWidth):
(WebCore):
(WebCore::RenderMultiColumnSet::computeLogicalHeight):
* rendering/RenderMultiColumnSet.h:
(WebCore::RenderMultiColumnSet::setColumnWidthAndCount):
(RenderMultiColumnSet):
* rendering/RenderObject.h:
(WebCore::RenderObject::isRenderMultiColumnBlock):
* rendering/RenderRegion.cpp:
(WebCore::RenderRegion::logicalWidthForFlowThreadContent):
(WebCore):
(WebCore::RenderRegion::logicalHeightForFlowThreadContent):
(WebCore::RenderRegion::layout):
* rendering/RenderRegion.h:
(RenderRegion):
2012-05-29 Alec Flett <alecflett@chromium.org>
IndexedDB: Align codes and names for IDB-specific and DOM-specific errors/exceptions
https://bugs.webkit.org/show_bug.cgi?id=87276
Reviewed by Tony Chang.
No new tests. Existing tests have been updated to use the new
exception constant values.
The IDB spec has evolved to DOM4-style DOMExceptions, which are
referred to with MixedCase, string-based style names for
exceptions rather than UNDERSCORE_CONSTANT_ERR codes. Change to
use MixedCase error/exception names, and tease out the
IDB-specific exception/error codes/names from the system-level
codes. Make sure that the legacy IDB-masked exception code values
match the DOM exception values. Make sure that
IDBDatabaseException[exception.name] still works.
* Modules/indexeddb/IDBDatabase.cpp:
(WebCore::IDBDatabase::createObjectStore):
(WebCore::IDBDatabase::transaction):
* Modules/indexeddb/IDBDatabaseBackendImpl.cpp:
(WebCore::IDBDatabaseBackendImpl::deleteObjectStore):
(WebCore::IDBDatabaseBackendImpl::setVersion):
(WebCore::IDBDatabaseBackendImpl::transaction):
* Modules/indexeddb/IDBDatabaseException.cpp:
(IDBDatabaseExceptionNameDescription):
(WebCore):
(WebCore::IDBDatabaseException::initializeDescription):
* Modules/indexeddb/IDBDatabaseException.h:
* Modules/indexeddb/IDBDatabaseException.idl:
* Modules/indexeddb/IDBObjectStore.cpp:
(WebCore::IDBObjectStore::createIndex):
* Modules/indexeddb/IDBObjectStoreBackendImpl.cpp:
(WebCore::IDBObjectStoreBackendImpl::index):
(WebCore::IDBObjectStoreBackendImpl::deleteIndex):
* Modules/indexeddb/IDBRequest.cpp:
(WebCore::IDBRequest::abort):
* Modules/indexeddb/IDBTransactionBackendImpl.cpp:
(WebCore::IDBTransactionBackendImpl::objectStore):
2012-05-29 David Hyatt <hyatt@apple.com>
https://bugs.webkit.org/show_bug.cgi?id=87775
[New Multicolumn] Make sure RenderMultiColumnSets have their own copy of the count, width and height of their columns.
Reviewed by Anders Carlsson.
* rendering/RenderMultiColumnBlock.cpp:
(WebCore::RenderMultiColumnBlock::RenderMultiColumnBlock):
(WebCore::RenderMultiColumnBlock::ensureColumnSets):
* rendering/RenderMultiColumnBlock.h:
(WebCore::RenderMultiColumnBlock::columnWidth):
(WebCore::RenderMultiColumnBlock::columnCount):
(RenderMultiColumnBlock):
* rendering/RenderMultiColumnSet.cpp:
(WebCore::RenderMultiColumnSet::RenderMultiColumnSet):
* rendering/RenderMultiColumnSet.h:
(WebCore::RenderMultiColumnSet::columnCount):
(WebCore::RenderMultiColumnSet::columnWidth):
(WebCore::RenderMultiColumnSet::columnHeight):
(RenderMultiColumnSet):
(WebCore::RenderMultiColumnSet::setColumnWidthAndCount):
(WebCore::RenderMultiColumnSet::setColumnHeight):
2012-05-29 Simon Fraser <simon.fraser@apple.com>
Incomplete repaint on twitter.com when replying to a tweet
https://bugs.webkit.org/show_bug.cgi?id=87553
Reviewed by Dean Jackson.
Style changes can cause a compositing layer to change between
requiring its own backing store or not, e.g. with the addition
or removal of a transform.
When that happens, we need to repaint the ancesetor layer that
this layer was, or will be drawing into.
Factored some code out of layerWillBeRemoved() to be able to
also call it from setRequiresOwnBackingStore().
New manual test, ManualTests/compositing/requires-backing-change.html.
I was not able to get an automated pixel test to work.
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::setRequiresOwnBackingStore):
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::updateCompositingLayers): Remove trailing whitespace.
(WebCore::RenderLayerCompositor::repaintInCompositedAncestor):
(WebCore::RenderLayerCompositor::layerWillBeRemoved):
* rendering/RenderLayerCompositor.h:
2012-05-29 David Hyatt <hyatt@apple.com>
https://bugs.webkit.org/show_bug.cgi?id=87771
[New Multicolumn] Make sure we have one column set by default and add it to the flow thread's region list.
Reviewed by Anders Carlsson.
* rendering/RenderMultiColumnBlock.cpp:
(WebCore::RenderMultiColumnBlock::ensureColumnSets):
* rendering/RenderRegion.cpp:
(WebCore::RenderRegion::RenderRegion):
2012-05-29 John Sullivan <sullivan@apple.com>
Update label for blacklisted plug-in
https://bugs.webkit.org/show_bug.cgi?id=87767
rdar://problem/11550048
Reviewed by Kevin Decker.
* English.lproj/Localizable.strings:
Regenerated.
* platform/LocalizedStrings.cpp:
(WebCore::insecurePluginVersionText):
Changed this string.
2012-05-29 Julien Chaffraix <jchaffraix@webkit.org>
REGRESSION(r111742): box-sizing: border-box doesn't work on fixed table layout
https://bugs.webkit.org/show_bug.cgi?id=87536
Reviewed by Tony Chang.
Tests: fast/table/fixed-table-layout/cell-box-sizing-fixed-table-layout.html
fast/table/fixed-table-layout/column-box-sizing-fixed-table-layout.html
fast/table/fixed-table-layout/column-group-box-sizing-fixed-table-layout.html
fast/table/fixed-table-layout/column-in-column-group-box-sizing-fixed-table-layout.html
The change in r111742 completely ignored border-sizing (following the table's code lead unfortunately).
The issue is that we would count the borders and paddings twice for the border-box case which would lead
to the content-box including them too. From a web-author, this behavior is equivalent to ignoring box-sizing.
* rendering/FixedTableLayout.cpp:
(WebCore::FixedTableLayout::calcWidthArray):
Reworked the function to properly use computeBorderBoxLogicalWidth() for correctness. This matches
what RenderBox and the rendering code does generally. Also refactored the code to avoid the need
for local variables and make it more readable.
2012-05-25 Ojan Vafai <ojan@chromium.org>
Absolute positioned elements in a relative positioned CSS3 Flexbox fail to display properly
https://bugs.webkit.org/show_bug.cgi?id=58453
Reviewed by Tony Chang.
Test: css3/flexbox/align-absolute-child.html
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::prepareChildForPositionedLayout):
(WebCore::RenderFlexibleBox::layoutAndPlaceChildren):
(WebCore::RenderFlexibleBox::adjustAlignmentForChild):
When adjusting positioned items, adjust their static position, not their location.
(WebCore::RenderFlexibleBox::flipForRightToLeftColumn):
(WebCore::RenderFlexibleBox::flipForWrapReverse):
* rendering/RenderFlexibleBox.h:
2012-05-29 Abhishek Arya <inferno@chromium.org>
Crash due to text fragment destruction when updating first-letter block.
https://bugs.webkit.org/show_bug.cgi?id=87751
Reviewed by Eric Seidel.
Test: fast/text/text-fragment-first-letter-update-crash.html
* rendering/RenderObject.cpp:
(WebCore::RenderObject::setStyle):
2012-05-29 David Hyatt <hyatt@apple.com>
https://bugs.webkit.org/show_bug.cgi?id=87764
Add implementation of checkForPaginationLogicalHeightChange to RenderMultiColumnBlock.
Also stub out ensureColumnSets() and write a comment describing how it's going to work.
Reviewed by Anders Carlsson.
* rendering/RenderMultiColumnBlock.cpp:
(WebCore::RenderMultiColumnBlock::RenderMultiColumnBlock):
(WebCore::RenderMultiColumnBlock::checkForPaginationLogicalHeightChange):
(WebCore):
(WebCore::RenderMultiColumnBlock::ensureColumnSets):
* rendering/RenderMultiColumnBlock.h:
(WebCore::RenderMultiColumnBlock::columnHeight):
(WebCore::RenderMultiColumnBlock::setColumnHeight):
(RenderMultiColumnBlock):
2012-05-29 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118620.
http://trac.webkit.org/changeset/118620
https://bugs.webkit.org/show_bug.cgi?id=87760
This patch causes regression to read an empty file/blob via
FileReader (Requested by jianli on #webkit).
* fileapi/FileReader.cpp:
(WebCore::FileReader::stringResult):
2012-05-29 Andreas Kling <kling@webkit.org>
Make StylePropertySet a variable-sized object to reduce memory use.
<http://webkit.org/b/87506>
Reviewed by Antti Koivisto.
Make immutable StylePropertySet objects variable-sized and allocate exactly
the space needed to hold the CSSProperty objects passed at construction.
This lets us avoid the memory overhead of Vector<CSSProperty, 4>.
Renamed StylePropertySet::adopt* to create and createImmutable since they no
longer adopt the incoming vector. More cleanup will follow in this area.
This reduces memory usage across the board - the more CSS declarations,
the bigger the benefit. As an example, we save around ~0.5MB on wsj.com.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::copyPropertiesInSet):
* css/CSSParser.cpp:
(WebCore::CSSParser::createStylePropertySet):
* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::createImmutable):
(WebCore::StylePropertySet::StylePropertySet):
(WebCore::StylePropertySet::~StylePropertySet):
(WebCore::StylePropertySet::copyPropertiesFrom):
(WebCore::StylePropertySet::removeProperty):
(WebCore::StylePropertySet::setProperty):
(WebCore::StylePropertySet::parseDeclaration):
(WebCore::StylePropertySet::addParsedProperties):
(WebCore::StylePropertySet::asText):
(WebCore::StylePropertySet::merge):
(WebCore::StylePropertySet::addSubresourceStyleURLs):
(WebCore::StylePropertySet::removePropertiesInSet):
(WebCore::StylePropertySet::findPropertyWithId):
(WebCore::StylePropertySet::removeEquivalentProperties):
(WebCore::StylePropertySet::copyPropertiesInSet):
(WebCore::StylePropertySet::averageSizeInBytes):
(SameSizeAsStylePropertySet):
(WebCore::StylePropertySet::append):
* css/StylePropertySet.h:
(WebCore::StylePropertySet::create):
(StylePropertySet):
(WebCore::StylePropertySet::propertyAt):
(WebCore::StylePropertySet::propertyCount):
(WebCore::StylePropertySet::isEmpty):
(WebCore::StylePropertySet::array):
* dom/StyledElement.cpp:
(WebCore::StyledElement::updateAttributeStyle):
2012-05-29 David Hyatt <hyatt@apple.com>
https://bugs.webkit.org/show_bug.cgi?id=87759
Pull the initial page logical height checking for columns into its own separate function. This
will allow the derived RenderMultiColumnBlock class to do its own setup there. It also has the
effect of cleaning up layoutBlock even for the existing code.
Reviewed by Anders Carlsson.
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::checkForPaginationLogicalHeightChange):
(WebCore):
(WebCore::RenderBlock::layoutBlock):
* rendering/RenderBlock.h:
(RenderBlock):
2012-05-29 Victor Carbune <victor@rosedu.org>
Update cue event dispatching for zero and negative duration cues
https://bugs.webkit.org/show_bug.cgi?id=85143
Reviewed by Eric Carlson.
Updated existing test.
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::updateActiveTextTrackCues): Explicitly
treated event dispatching for zero and negative duration cues.
(WebCore::HTMLMediaElement::textTrackAddCue): Interval tree doesn't
handle invalid cues with intervals - negative cues need to be treated
as zero-length cues as they are always missed cues anyway.
(WebCore::HTMLMediaElement::textTrackRemoveCue): Same.
2012-05-29 MORITA Hajime <morrita@google.com>
ShadowRoot.getElementById() returns removed node.
https://bugs.webkit.org/show_bug.cgi?id=62184
Deregitration of m_elementsById map is done at Element::removedFrom()
But the tree scope of the removed node is already reset from a shadow root
to a document.
This change lookup the old shadow root through its parent node and
deregitrater the node against it.
Reviewed by Dimitri Glazkov.
Test: fast/dom/shadow/get-element-by-id-in-shadow-mutation.html
* dom/Element.cpp:
(WebCore::treeScopeOfParent):
(WebCore):
(WebCore::Element::removedFrom):
* dom/Element.h:
(Element):
(WebCore::Element::updateId):
(WebCore):
2012-05-29 Dana Jansens <danakj@chromium.org>
[chromium] Unoccluded area in surfaces should always be clipped to the rootScissorRect
https://bugs.webkit.org/show_bug.cgi?id=87677
Reviewed by Adrienne Walker.
When a contributing surface is visited, and we have not visited another
surface, then we know the surface is completely unoccluded. However, the
root scissor rect should still be applied, and anything outside of it
should be considered occluded.
Unit test: WebKitTests::CCOcclusionTrackerTestTopmostSurfaceIsClippedToScissor
* platform/graphics/chromium/RenderSurfaceChromium.cpp:
(WebCore::RenderSurfaceChromium::targetRenderSurface):
(WebCore):
* platform/graphics/chromium/RenderSurfaceChromium.h:
(RenderSurfaceChromium):
* platform/graphics/chromium/cc/CCOcclusionTracker.cpp:
(WebCore::::unoccludedContributingSurfaceContentRect):
* platform/graphics/chromium/cc/CCRenderSurface.cpp:
(WebCore::CCRenderSurface::targetRenderSurface):
(WebCore):
* platform/graphics/chromium/cc/CCRenderSurface.h:
(CCRenderSurface):
2012-05-29 Jonathan Dong <jonathan.dong@torchmobile.com.cn>
[BlackBerry] Refactoring http/proxy authentication to use authcallback mechanism introduced with libcurl
https://bugs.webkit.org/show_bug.cgi?id=87706
Reviewed by Antonio Gomes.
Because we use authentication callback mechanism of libcurl,
which will take care of all the authentication communication
between UA and server, so the only thing we need to do is to
provide credential information when authentication callback
is triggered. By using this we don't need to process
intermediate headers for each http authentication scheme.
Implemented interface function notifyAuthReceived() in
NetworkJob to handle the authentication notification from
networking thread.
When the authentication callback is triggered, we should
cancel the current loading request and restart a new one with
credential information attached which comes from
credentialStorage or user input.
Internally reviewed by Joe Mason <jmason@rim.com>.
No new tests since there is no behavior change.
* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::notifyAuthReceived):
(WebCore):
(WebCore::NetworkJob::handleNotifyHeaderReceived):
(WebCore::NetworkJob::sendRequestWithCredentials):
* platform/network/blackberry/NetworkJob.h:
(NetworkJob):
2012-05-29 Jonathan Dong <jonathan.dong@torchmobile.com.cn>
[BlackBerry] http authentication challenge issue when loading subresource
https://bugs.webkit.org/show_bug.cgi?id=87660
Reviewed by Rob Buis.
We should try to store the credential information as soon as we
receive http status notification which indicates that the
authentication succeeds, instead of storing credentials when the
main resource receives notify done which is too late for loading
subresource.
Internally reviewed by Joe Mason <jmason@rim.com>
No new tests since there is no change in behavior.
* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::handleNotifyStatusReceived):
(WebCore::NetworkJob::handleNotifyClose):
2012-05-29 Alexey Proskuryakov <ap@apple.com>
[Mac] Move DefaultFonts.plist to C++ code
https://bugs.webkit.org/show_bug.cgi?id=87560
<rdar://problem/11466119>
Reviewed by Dan Bernstein.
* DerivedSources.make:
* Resources/DefaultFonts.plist.in: Removed.
* WebCore.xcodeproj/project.pbxproj:
* page/mac/SettingsMac.mm: (WebCore::Settings::initializeDefaultFontFamilies):
Instead of having a separate configuration file, just hardcode the settings in C++.
2012-05-29 Alexei Filippov <alexeif@chromium.org>
Web Inspector: Retainers are missing for all objects
https://bugs.webkit.org/show_bug.cgi?id=87655
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshotDataGrids.js:
(WebInspector.HeapSnapshotContainmentDataGrid.prototype.setDataSource):
2012-05-29 Philippe Normand <pnormand@igalia.com>
[GStreamer] Video sink doesn't respect WebKit code style
https://bugs.webkit.org/show_bug.cgi?id=87693
Reviewed by Alexis Menard.
No change in behavior of the sink, this patch only adapts the code
style to WebKit's.
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
* platform/graphics/gstreamer/VideoSinkGStreamer.cpp:
(_WebKitVideoSinkPrivate):
(webkitVideoSinkTimeoutCallback):
(webkitVideoSinkRender):
(webkitVideoSinkDispose):
(unlockBufferMutex):
(webkitVideoSinkUnlock):
(webkitVideoSinkUnlockStop):
(webkitVideoSinkStop):
(webkitVideoSinkStart):
(webkitVideoSinkMarshalVoidAndMiniObject):
(webkit_video_sink_class_init):
(webkitVideoSinkNew):
* platform/graphics/gstreamer/VideoSinkGStreamer.h:
(_WebKitVideoSink):
(_WebKitVideoSinkClass):
2012-05-29 Roland Takacs <takacs.roland@stud.u-szeged.hu>
[Qt] Uninitialized shader compiler when GLES is used
https://bugs.webkit.org/show_bug.cgi?id=87712
The '#if !defined(QT_OPENGL_ES_2)' has been removed before the initialization
of shader compiler to compile shader codes for ES-platforms as well.
Reviewed by Noam Rosenthal.
* platform/graphics/qt/GraphicsContext3DQt.cpp:
(WebCore::GraphicsContext3D::GraphicsContext3D):
2012-05-29 Taiju Tsuiki <tzik@chromium.org>
Add InspectorFileSystemAgent::FrontendProvider
https://bugs.webkit.org/show_bug.cgi?id=87635
InspectorFileSystemAgent needs weak reference to its frontend to perform
asynchronous operation.
Reviewed by Yury Semikhatsky.
* inspector/InspectorFileSystemAgent.cpp:
(InspectorFileSystemAgent::FrontendProvider):
(WebCore):
(WebCore::InspectorFileSystemAgent::~InspectorFileSystemAgent):
(WebCore::InspectorFileSystemAgent::enable):
(WebCore::InspectorFileSystemAgent::disable):
(WebCore::InspectorFileSystemAgent::setFrontend):
(WebCore::InspectorFileSystemAgent::clearFrontend):
(WebCore::InspectorFileSystemAgent::restore):
(WebCore::InspectorFileSystemAgent::InspectorFileSystemAgent):
* inspector/InspectorFileSystemAgent.h:
(InspectorFileSystemAgent):
2012-05-29 Yoshifumi Inoue <yosin@chromium.org>
REGRESSION(r109729): The optgroup element's "disabled" attribute has no effect to rendering and selection
https://bugs.webkit.org/show_bug.cgi?id=87614
Reviewed by Kent Tamura.
This patch adds HTMLOptGroupElement::disabled and isEnabledFormControl
methods which was implemented in HTMLFormControlElement at r109729 and
updates CSS pseudo class checker to check "optgroup" element.
Tests: fast/forms/select/optgroup-clicking.html
fast/forms/select/optgroup-disabled.html
fast/forms/select/optgroup-rendering.html
* css/SelectorChecker.cpp:
(WebCore::SelectorChecker::checkOneSelector): On disabled and enabled pseudo class checker.
We checked tag name "optgroup" in addition isFormControlElement.
* css/StyleResolver.cpp:
(WebCore::StyleResolver::canShareStyleWithElement): Checks optgroup.disabled.
* html/HTMLOptGroupElement.cpp:
(WebCore::HTMLOptGroupElement::disabled): Added.
* html/HTMLOptGroupElement.h:
(HTMLOptGroupElement): Added isEnabledFormControl.
2012-05-29 Alexander Pavlov <apavlov@chromium.org>
[Chromium] [Regression] SELECT popup misplaced when there is not enough vertical space
https://bugs.webkit.org/show_bug.cgi?id=87735
Reviewed by Kent Tamura.
This change fixes the obsolete WebWidget Y-repositioning code for the insufficient vertical space case, not removed in r113418.
No new tests, as popups are not testable in Chromium.
* platform/chromium/PopupContainer.cpp:
(WebCore::PopupContainer::layoutAndCalculateWidgetRect):
2012-05-29 Andrei Bucur <abucur@adobe.com>
[CSS Regions]Rename NamedFlow::overflow to NamedFlow::overset
https://bugs.webkit.org/show_bug.cgi?id=87491
This patch updates the CSSOM property name overflow to overset. There are no changes in functionality.
Link to spec: http://www.w3.org/TR/2012/WD-css3-regions-20120503/
Reviewed by Andreas Kling.
Test: fast/regions/webkit-named-flow-overset.html
* dom/WebKitNamedFlow.cpp:
(WebCore::WebKitNamedFlow::overset):
* dom/WebKitNamedFlow.h:
(WebKitNamedFlow):
* dom/WebKitNamedFlow.idl:
* rendering/RenderFlowThread.cpp:
(WebCore::RenderFlowThread::RenderFlowThread):
(WebCore::RenderFlowThread::computeOverflowStateForRegions):
* rendering/RenderFlowThread.h:
2012-05-29 Andreas Kling <kling@webkit.org>
Give StylePropertySet a proper mutability flag.
<http://webkit.org/b/87675>
Reviewed by Antti Koivisto.
Add a flag to StylePropertySet that determines whether it's mutable or not. Only objects
constructed from the CSS parser (by way of StylePropertySet::adopt()) are made immutable.
This is clean-up and prep work for implementing a more efficient storage scheme for immutable
StylePropertySets (fixed-size array instead of vector) as the vast majority of them are
unlikely to ever mutate during their lifetime.
Sprinkled ASSERT(isMutable()) all over StylePropertySet to catch any attempts to modify
an immutable object.
Also made properties() const on StyleRule and friends, which brought some additional code
churn along with it.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::copyPropertiesInSet):
* css/CSSFontFaceRule.cpp:
(WebCore::CSSFontFaceRule::style):
(WebCore::CSSFontFaceRule::reattach):
* css/CSSFontFaceSrcValue.cpp:
(WebCore::CSSFontFaceSrcValue::addSubresourceStyleURLs):
* css/CSSFontFaceSrcValue.h:
(CSSFontFaceSrcValue):
* css/CSSPageRule.cpp:
(WebCore::CSSPageRule::style):
(WebCore::CSSPageRule::reattach):
* css/CSSPrimitiveValue.cpp:
(WebCore::CSSPrimitiveValue::addSubresourceStyleURLs):
* css/CSSPrimitiveValue.h:
(CSSPrimitiveValue):
* css/CSSReflectValue.cpp:
(WebCore::CSSReflectValue::addSubresourceStyleURLs):
* css/CSSReflectValue.h:
(CSSReflectValue):
* css/CSSStyleRule.cpp:
(WebCore::CSSStyleRule::style):
(WebCore::CSSStyleRule::reattach):
* css/CSSValue.cpp:
(WebCore::CSSValue::addSubresourceStyleURLs):
* css/CSSValue.h:
(CSSValue):
* css/CSSValueList.cpp:
(WebCore::CSSValueList::addSubresourceStyleURLs):
* css/CSSValueList.h:
(CSSValueList):
* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::StylePropertySet):
(WebCore::StylePropertySet::setCSSParserMode):
(WebCore):
(WebCore::StylePropertySet::copyPropertiesFrom):
(WebCore::StylePropertySet::removeShorthandProperty):
(WebCore::StylePropertySet::removeProperty):
(WebCore::StylePropertySet::setProperty):
(WebCore::StylePropertySet::parseDeclaration):
(WebCore::StylePropertySet::addParsedProperties):
(WebCore::StylePropertySet::addParsedProperty):
(WebCore::StylePropertySet::merge):
(WebCore::StylePropertySet::addSubresourceStyleURLs):
(WebCore::StylePropertySet::removePropertiesInSet):
(WebCore::StylePropertySet::findPropertyWithId):
(WebCore::StylePropertySet::removeEquivalentProperties):
(WebCore::StylePropertySet::copyPropertiesInSet):
* css/StylePropertySet.h:
(WebCore::StylePropertySet::adopt):
(WebCore::StylePropertySet::adoptMutable):
(StylePropertySet):
(WebCore::StylePropertySet::isMutable):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::addMatchedProperties):
(WebCore::StyleResolver::collectMatchingRulesForList):
* css/StyleResolver.h:
(StyleResolver):
* css/StyleRule.cpp:
(WebCore::StyleRule::mutableProperties):
(WebCore):
(WebCore::StyleRulePage::mutableProperties):
(WebCore::StyleRuleFontFace::mutableProperties):
* css/StyleRule.h:
(WebCore::StyleRule::properties):
(StyleRule):
(WebCore::StyleRuleFontFace::properties):
(StyleRuleFontFace):
(WebCore::StyleRulePage::properties):
(StyleRulePage):
* dom/ElementAttributeData.cpp:
(WebCore::ElementAttributeData::ensureInlineStyle):
(WebCore::ElementAttributeData::updateInlineStyleAvoidingMutation):
* editing/EditingStyle.cpp:
(WebCore::styleFromMatchedRulesForElement):
* svg/SVGFontFaceElement.cpp:
(WebCore::SVGFontFaceElement::parseAttribute):
(WebCore::SVGFontFaceElement::rebuildFontFace):
(WebCore::SVGFontFaceElement::removedFrom):
2012-05-29 Mikhail Pozdnyakov <mikhail.pozdnyakov@intel.com>
[EFL] Refactoring. Get rid of unnecessary singleCharacterString() function
https://bugs.webkit.org/show_bug.cgi?id=87654
Reviewed by Csaba Osztrogonác.
Evas key events already provide key string, so no practical need for singleCharacterString()
evaluating the event string from keyname.
Modified code related functionality is already covered with fast/events and fast/forms tests.
* platform/efl/EflKeyboardUtilities.cpp:
* platform/efl/PlatformKeyboardEventEfl.cpp:
(WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent):
2012-05-29 Ilya Tikhonovsky <loislo@chromium.org>
Unreviewed: Web Inspector: single line fix after r118742.
* inspector/front-end/HeapSnapshotView.js:
(WebInspector.HeapProfileHeader.prototype.saveToFile):
2012-05-29 Kentaro Hara <haraken@chromium.org>
Remove m_rootNode and m_selectorChecker from SelectorQuery.h
https://bugs.webkit.org/show_bug.cgi?id=87650
Reviewed by Antti Koivisto.
To avoid parsing CSS selectors every time, I am planning to cache
SelectorQuery objects into a HashMap, which will be defined in
Node::querySelector(). To make the size of SelectorQuery small,
this patch removes m_rootNode and m_selectorChecker from SelectorQuery.h.
Tests: fast/dom/SelectorAPI/*. No change in test results.
* dom/Node.cpp:
(WebCore::Node::querySelector):
(WebCore::Node::querySelectorAll):
* dom/SelectorQuery.cpp:
(WebCore::SelectorQuery::SelectorQuery):
(WebCore):
(WebCore::SelectorQuery::queryAll):
(WebCore::SelectorQuery::queryFirst):
* dom/SelectorQuery.h:
(SelectorQuery):
2012-05-29 Eugene Klyuchnikov <eustas.bug@gmail.com>
Web Inspector: Tabbed Settings Screen
https://bugs.webkit.org/show_bug.cgi?id=87497
Reviewed by Yury Semikhatsky.
Combine settings screen and shortcuts screen
to a single tabbed settings screen.
This will make shortcuts screen more discoverable.
* inspector/front-end/HelpScreen.js:
(WebInspector.HelpScreen): Make titled tamplate optional.
(WebInspector.HelpScreen.prototype._createCloseButton): Extract method.
* inspector/front-end/SettingsScreen.js:
(WebInspector.SettingsScreen): Split initialization and UI construction.
(WebInspector.SettingsScreen.prototype._createSettingsTabView.appendSection):
(WebInspector.SettingsScreen.prototype._createSettingsTabView):
Extract GUI construction method.
(WebInspector.SettingsScreen.prototype._getOrCreateTabbedPane):
Make GUI construction lazy.
(WebInspector.SettingsScreen.prototype.selectTab):
Add method for external tab swithching.
(WebInspector.SettingsScreen.prototype.wasShown):
Make GUI construction lazy.
(WebInspector.SettingsScreenTabbedPane):
Extend class to access protected member.
(WebInspector.SettingsController.prototype._buttonClicked):
Change "showSettingsScreen" visibility and parameters.
(WebInspector.SettingsController.prototype._onHideSettingsScreen):
Preserve settings screen state.
(WebInspector.SettingsController.prototype.showSettingsScreen):
Change method visibility and parameters.
* inspector/front-end/ShortcutsScreen.js:
(WebInspector.ShortcutsScreen): Remove inheritance from HelpScreen.
(WebInspector.ShortcutsScreen.prototype._createShortcutsTabView):
Refactoring of GUI construction.
(WebInspector.ShortcutsSection.prototype.renderSection):
Ditto.
* inspector/front-end/helpScreen.css:
(.help-window-main ::-webkit-scrollbar): Rebind scrollbar settings.
(.help-window-main ::-webkit-resizer): Ditto.
(.help-window-main ::-webkit-scrollbar-thumb:vertical): Ditto.
(.help-window-main ::-webkit-scrollbar-thumb:vertical:active): Ditto.
(.help-window-main ::-webkit-scrollbar-track:vertical): Ditto.
(body:not(.compact) #settings-screen .tabbed-pane): Fix container height.
(#settings-screen .tabbed-pane-header): Ditto.
(#settings-screen .tabbed-pane-header-tabs): Adjust spacing.
(#settings-screen .tabbed-pane-header-tab): Adjust style.
(#settings-screen .help-close-button): Adjust spacing.
(#settings-screen .tabbed-pane-header-tab.selected):
Erase line below active tab.
(#settings-screen .tabbed-pane-content): Adjust spacing.
(#settings-screen .help-content): Ditto.
* inspector/front-end/inspector.js:
(WebInspector.documentKeyDown): Use Settings screen to show shortcuts.
2012-05-29 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: REGRESSION: load heap snapshot doesn't work.
https://bugs.webkit.org/show_bug.cgi?id=87642
HeapSnapshotReceiver interface was introduced.
It declares the API for HeapSnapshotLoader, HeapSnapshotLoaderProxy and HeapSnapshotSaveToFileReceiver.
The HeapProfileHeader was refactored and tests were added.
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshotLoader.js:
(WebInspector.HeapSnapshotLoader):
(WebInspector.HeapSnapshotLoader.prototype.startLoading):
(WebInspector.HeapSnapshotLoader.prototype.dispose):
(WebInspector.HeapSnapshotLoader.prototype._reset):
(WebInspector.HeapSnapshotLoader.prototype.finishLoading):
* inspector/front-end/HeapSnapshotProxy.js:
(WebInspector.HeapSnapshotWorker.prototype.startCheckingForLongRunningCalls):
(WebInspector.HeapSnapshotLoaderProxy.prototype.startLoading):
(WebInspector.HeapSnapshotLoaderProxy.prototype.pushJSONChunk):
(WebInspector.HeapSnapshotLoaderProxy.prototype.finishLoading):
* inspector/front-end/HeapSnapshotView.js:
(WebInspector.HeapSnapshotReceiver):
(WebInspector.HeapSnapshotReceiver.prototype.startLoading):
(WebInspector.HeapSnapshotReceiver.prototype.pushJSONChunk):
(WebInspector.HeapSnapshotReceiver.prototype.finishLoading):
(WebInspector.HeapSnapshotReceiver.prototype.dispose):
(WebInspector.HeapProfileHeader):
(WebInspector.HeapProfileHeader.prototype.load):
(WebInspector.HeapProfileHeader.prototype._setupWorker):
(WebInspector.HeapProfileHeader.prototype.dispose):
(WebInspector.HeapProfileHeader.prototype._saveStatusUpdate):
(WebInspector.HeapProfileHeader.prototype.pushJSONChunk):
(WebInspector.HeapProfileHeader.prototype._parsed):
(WebInspector.HeapProfileHeader.prototype.finishHeapSnapshot):
(WebInspector.HeapProfileHeader.prototype.saveToFile):
(WebInspector.HeapProfileHeader.prototype.loadFromFile.onLoad):
(WebInspector.HeapProfileHeader.prototype.loadFromFile):
(WebInspector.HeapProfileHeader.prototype._loadNextChunk):
(WebInspector.HeapProfileHeader.prototype._nextChunkLoaded):
(WebInspector.HeapProfileHeader.prototype._createFileReader):
(WebInspector.HeapSnapshotSaveToFileReceiver):
(WebInspector.HeapSnapshotSaveToFileReceiver.prototype.startLoading):
(WebInspector.HeapSnapshotSaveToFileReceiver.prototype.pushJSONChunk):
(WebInspector.HeapSnapshotSaveToFileReceiver.prototype.finishLoading):
(WebInspector.HeapSnapshotSaveToFileReceiver.prototype.dispose):
(WebInspector.HeapSnapshotSaveToFileReceiver.prototype._startSavingSnapshot):
(WebInspector.HeapSnapshotSaveToFileReceiver.prototype._saveStatusUpdate):
* inspector/front-end/ProfilesPanel.js:
(WebInspector.ProfilesPanel.prototype._createFileSelectorElement.onChange):
(WebInspector.ProfilesPanel.prototype._createFileSelectorElement):
(WebInspector.ProfilesPanel.prototype._loadFromFile):
2012-05-29 Eric Seidel <eric@webkit.org>
Add HTMLIFrameElement.seamless property accessor now that seamless is enabled and works
https://bugs.webkit.org/show_bug.cgi?id=87708
Reviewed by Adam Barth.
Lots of tests in fast/frames/seamless cover this property.
* html/HTMLIFrameElement.idl:
2012-05-29 Kentaro Hara <haraken@chromium.org>
[Performance] Optimize querySelector() and querySelectorAll() by removing
redundant dummy style sheet creations
https://bugs.webkit.org/show_bug.cgi?id=87627
Reviewed by Antti Koivisto.
This patch improves performance of querySelector("#id") by 4.2% ~ 17.7%
in Safari and Chromium.
Performance tests: https://bugs.webkit.org/attachment.cgi?id=144288
The performance test results in my desktop:
[Safari/Mac]
document.querySelector("#headId") : 13.60 ms => 12.40 ms +9.68%
document.querySelector("#duplicatedId") : 14.80 ms => 14.20 ms +4.23%
document.querySelector("#tailId") : 13.80 ms => 12.20 ms +13.11%
document.querySelectorAll("#headId") : 17.80 ms => 14.60 ms +21.92%
document.querySelectorAll("#duplicatedId") : 513.80 ms => 471.60 ms +8.95%
document.querySelectorAll("#tailId") : 19.20 ms => 16.40 ms +17.07%
document.querySelector("h1") : 14.40 ms => 13.60 ms +5.88%
document.querySelector("h2") : 14.80 ms => 13.80 ms +7.25%
document.querySelector("h3") : 312.00 ms => 306.60 ms +1.76%
document.querySelectorAll("h1") : 322.00 ms => 314.60 ms +2.35%
document.querySelectorAll("h2") : 481.00 ms => 420.60 ms +14.36%
document.querySelectorAll("h3") : 326.80 ms => 315.80 ms +3.48%
document.querySelector(".headClass") : 16.20 ms => 15.40 ms +5.19%
document.querySelector(".duplicatedClass") : 17.00 ms => 15.40 ms +10.39%
document.querySelector(".tailClass") : 369.60 ms => 362.60 ms +1.93%
document.querySelectorAll(".headClass") : 388.20 ms => 377.60 ms +2.81%
document.querySelectorAll(".duplicatedClass") : 605.00 ms => 581.20 ms +4.09%
document.querySelectorAll(".tailClass") : 401.60 ms => 391.40 ms +2.61%
[Chromium/Linux]
ocument.querySelector("#headId") : 16.00 ms => 13.60 ms +17.65%
document.querySelector("#duplicatedId") : 16.80 ms => 15.40 ms +9.09%
document.querySelector("#tailId") : 16.00 ms => 13.60 ms +17.65%
document.querySelectorAll("#headId") : 29.00 ms => 25.00 ms +16.00%
document.querySelectorAll("#duplicatedId") : 561.40 ms => 505.20 ms +11.12%
document.querySelectorAll("#tailId") : 34.00 ms => 26.80 ms +26.87%
document.querySelector("h1") : 18.20 ms => 15.80 ms +15.19%
document.querySelector("h2") : 19.00 ms => 16.60 ms +14.46%
document.querySelector("h3") : 280.40 ms => 274.00 ms +2.34%
document.querySelectorAll("h1") : 300.20 ms => 365.20 ms -17.80%
document.querySelectorAll("h2") : 487.60 ms => 504.20 ms -3.29%
document.querySelectorAll("h3") : 349.40 ms => 324.00 ms +7.84%
document.querySelector(".headClass") : 18.40 ms => 16.00 ms +15.00%
document.querySelector(".duplicatedClass") : 19.40 ms => 16.20 ms +19.75%
document.querySelector(".tailClass") : 349.60 ms => 345.40 ms +1.22%
document.querySelectorAll(".headClass") : 382.80 ms => 378.20 ms +1.22%
document.querySelectorAll(".duplicatedClass") : 550.80 ms => 574.20 ms -4.08%
document.querySelectorAll(".tailClass") : 381.00 ms => 387.00 ms -1.55%
I've found that StyleSheetInternal::create() in CSSParser::parseSelector()
is a bottleneck of querySelector() and querySelectorAll().
void CSSParser::parseSelector(const String& string, CSSSelectorList& selectorList)
{
RefPtr<StyleSheetInternal> dummyStyleSheet = StyleSheetInternal::create();
setStyleSheet(dummyStyleSheet.get());
...;
}
This patch removes a dummy style sheet. Previously the dummy style sheet
was needed to avoid null checks of m_styleSheet in CSSGrammer.y.
This patch fixes CSSGrammer.y so that it works fine with a null m_styleSheet.
Tests: fast/dom/SelectorAPI/*. No change in the test results.
* css/CSSGrammar.y:
* css/CSSParser.cpp:
(WebCore::CSSParser::parseSelector):
2012-05-29 Nikolas Zimmermann <nzimmermann@rim.com>
SVGViewSpec DOM bindings aka SVGSVGElement.currentView is unimplemented
https://bugs.webkit.org/show_bug.cgi?id=15495
Reviewed by Rob Buis.
Finish SVGViewSpec implementation and expose it to the Web. Numerous smaller bugs are fixed:
- SVGViewSpec and all of its contents should be read-only. Enforce that and test it.
Add a new enum to SVGPropertyInfo so that each SVGAnimatedProperty knows if its content is supposed to be read-write or read-only.
- Make the 'transform' attribute of SVGViewSpec exposable by switching to a SVGTransformListProperty - it can now be accessed from JS.
- Assure that following XXXString methods are in sync with the corresponding SVG DOM interfaces
readonly attribute DOMString viewBoxString;
readonly attribute DOMString preserveAspectRatioString;
readonly attribute DOMString transformString;
readonly attribute DOMString viewTargetString;
- Cleanup code in svg/properties that was intended to handle SVGViewSpec. Turns out lot of the complexity was unncessary, and can be removed.
- Move SVGFitToViewBox to the same pattern SVGZoomAndPan uses since two days, that avoids any virtual function calls for attribute parsing.
- Create/expose ObjectiveC/JS bindings for SVGViewSpec.
- Enable SVGViewSpec constructor.
Tests: svg/custom/linking-a-03-b-all-expected.svg
svg/dom/SVGViewSpec-defaults.html
svg/dom/SVGViewSpec.html
svg/dom/complex-svgView-specification.html
svg/dom/viewspec-parser.html
* CMakeLists.txt:
* DerivedSources.cpp:
* DerivedSources.make:
* DerivedSources.pri:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gyp/WebCore.gyp:
* WebCore.gypi:
* WebCore.order:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSSVGLengthCustom.cpp:
(WebCore::JSSVGLength::setValue):
(WebCore::JSSVGLength::convertToSpecifiedUnits):
* bindings/objc/DOMSVG.h:
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateImplementation):
(NativeToJSValue):
* bindings/scripts/CodeGeneratorObjC.pm:
(GenerateImplementation):
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateNormalAttrGetter):
(GenerateNormalAttrSetter):
(GenerateFunctionCallback):
* bindings/v8/custom/V8SVGLengthCustom.cpp:
(WebCore::V8SVGLength::valueAccessorSetter):
(WebCore::V8SVGLength::convertToSpecifiedUnitsCallback):
* page/DOMWindow.idl:
* rendering/svg/RenderSVGResourcePattern.cpp:
(WebCore::RenderSVGResourcePattern::buildTileImageTransform):
* svg/SVGFitToViewBox.cpp:
* svg/SVGFitToViewBox.h:
(SVGFitToViewBox):
(WebCore::SVGFitToViewBox::parseAttribute):
* svg/SVGMarkerElement.cpp:
(WebCore::SVGMarkerElement::orientTypePropertyInfo):
(WebCore::SVGMarkerElement::parseAttribute):
(WebCore::SVGMarkerElement::synchronizeOrientType):
(WebCore::SVGMarkerElement::lookupOrCreateOrientTypeWrapper):
* svg/SVGPathElement.cpp:
(WebCore::SVGPathElement::dPropertyInfo):
(WebCore::SVGPathElement::svgAttributeChanged):
(WebCore::SVGPathElement::pathByteStream):
(WebCore::SVGPathElement::lookupOrCreateDWrapper):
(WebCore::SVGPathElement::synchronizeD):
* svg/SVGPathSegWithContext.h:
(WebCore::SVGPathSegWithContext::animatedProperty):
* svg/SVGPatternElement.cpp:
(WebCore::SVGPatternElement::parseAttribute):
* svg/SVGPolyElement.cpp:
(WebCore::SVGPolyElement::pointsPropertyInfo):
(WebCore::SVGPolyElement::parseAttribute):
(WebCore::SVGPolyElement::synchronizePoints):
(WebCore::SVGPolyElement::lookupOrCreatePointsWrapper):
* svg/SVGRect.h:
* svg/SVGSVGElement.cpp:
(WebCore::SVGSVGElement::~SVGSVGElement):
(WebCore::SVGSVGElement::currentView):
(WebCore::SVGSVGElement::parseAttribute):
(WebCore::SVGSVGElement::currentViewBoxRect):
(WebCore::SVGSVGElement::viewBoxToViewTransform):
(WebCore::SVGSVGElement::setupInitialView):
(WebCore::SVGSVGElement::inheritViewAttributes):
* svg/SVGSVGElement.h:
(SVGSVGElement):
* svg/SVGSVGElement.idl:
* svg/SVGSymbolElement.cpp:
(WebCore::SVGSymbolElement::parseAttribute):
* svg/SVGTests.cpp:
(WebCore::SVGTests::requiredFeaturesPropertyInfo):
(WebCore::SVGTests::requiredExtensionsPropertyInfo):
(WebCore::SVGTests::systemLanguagePropertyInfo):
(WebCore::SVGTests::synchronizeRequiredFeatures):
(WebCore::SVGTests::synchronizeRequiredExtensions):
(WebCore::SVGTests::synchronizeSystemLanguage):
* svg/SVGTextContentElement.cpp:
(WebCore::SVGTextContentElement::textLengthPropertyInfo):
(WebCore::SVGTextContentElement::synchronizeTextLength):
(WebCore::SVGTextContentElement::lookupOrCreateTextLengthWrapper):
* svg/SVGViewElement.cpp:
(WebCore::SVGViewElement::parseAttribute):
* svg/SVGViewElement.h:
* svg/SVGViewSpec.cpp:
(WebCore::SVGViewSpec::viewBoxPropertyInfo):
(WebCore::SVGViewSpec::preserveAspectRatioPropertyInfo):
(WebCore::SVGViewSpec::transformPropertyInfo):
(WebCore::SVGViewSpec::SVGViewSpec):
(WebCore::SVGViewSpec::viewBoxIdentifier):
(WebCore::SVGViewSpec::preserveAspectRatioIdentifier):
(WebCore::SVGViewSpec::transformIdentifier):
(WebCore::SVGViewSpec::setZoomAndPan):
(WebCore::SVGViewSpec::setTransformString):
(WebCore::SVGViewSpec::transformString):
(WebCore::SVGViewSpec::viewBoxString):
(WebCore::SVGViewSpec::preserveAspectRatioString):
(WebCore::SVGViewSpec::viewTarget):
(WebCore::SVGViewSpec::transform):
(WebCore::SVGViewSpec::lookupOrCreateViewBoxWrapper):
(WebCore::SVGViewSpec::lookupOrCreatePreserveAspectRatioWrapper):
(WebCore::SVGViewSpec::lookupOrCreateTransformWrapper):
(WebCore::SVGViewSpec::reset):
(WebCore::SVGViewSpec::parseViewSpec):
* svg/SVGViewSpec.h:
(WebCore::SVGViewSpec::~SVGViewSpec):
(SVGViewSpec):
(WebCore::SVGViewSpec::create):
(WebCore::SVGViewSpec::setViewTargetString):
(WebCore::SVGViewSpec::setZoomAndPan):
(WebCore::SVGViewSpec::contextElement):
(WebCore::SVGViewSpec::resetContextElement):
(WebCore::SVGViewSpec::transformBaseValue):
(WebCore::SVGViewSpec::viewBoxAnimated):
(WebCore::SVGViewSpec::viewBox):
(WebCore::SVGViewSpec::viewBoxBaseValue):
(WebCore::SVGViewSpec::setViewBoxBaseValue):
(WebCore::SVGViewSpec::preserveAspectRatioAnimated):
(WebCore::SVGViewSpec::preserveAspectRatio):
(WebCore::SVGViewSpec::preserveAspectRatioBaseValue):
(WebCore::SVGViewSpec::setPreserveAspectRatioBaseValue):
* svg/SVGViewSpec.idl:
* svg/properties/SVGAnimatedProperty.h:
(WebCore::SVGAnimatedProperty::isReadOnly):
(WebCore::SVGAnimatedProperty::setIsReadOnly):
(SVGAnimatedProperty):
(WebCore::SVGAnimatedProperty::lookupOrCreateWrapper):
(WebCore::SVGAnimatedProperty::lookupWrapper):
(WebCore::SVGAnimatedProperty::SVGAnimatedProperty):
* svg/properties/SVGAnimatedPropertyMacros.h:
(WebCore::SVGSynchronizableAnimatedProperty::synchronize):
(SVGSynchronizableAnimatedProperty):
* svg/properties/SVGAnimatedPropertySynchronizer.h: Removed.
* svg/properties/SVGAttributeToPropertyMap.h:
* svg/properties/SVGListProperty.h:
* svg/properties/SVGListPropertyTearOff.h:
(SVGListPropertyTearOff):
(WebCore::SVGListPropertyTearOff::isReadOnly):
* svg/properties/SVGPathSegListPropertyTearOff.h:
(SVGPathSegListPropertyTearOff):
(WebCore::SVGPathSegListPropertyTearOff::isReadOnly):
* svg/properties/SVGProperty.h:
(SVGProperty):
* svg/properties/SVGPropertyInfo.h:
(WebCore::SVGPropertyInfo::SVGPropertyInfo):
(SVGPropertyInfo):
* svg/properties/SVGPropertyTearOff.h:
(WebCore::SVGPropertyTearOff::isReadOnly):
* svg/properties/SVGStaticListPropertyTearOff.h:
(SVGStaticListPropertyTearOff):
(WebCore::SVGStaticListPropertyTearOff::isReadOnly):
2012-05-29 Kent Tamura <tkent@chromium.org>
REGRESSION (r90971): the cursor is painted “behind” the placeholder text
https://bugs.webkit.org/show_bug.cgi?id=87155
Reviewed by Hajime Morita.
This regression happened only on platforms on which
RenderTheme::shouldShowPlaceholderWhenFocused() returns true.
Because the order of renderers for the editable node and the placeholder
node was:
- A renderer for the editable node
- A renderer for the placeholder node,
The text caret was painted, then the palceholder was painted.
We should not use z-index in the built-in shadow nodes. So the patch
fixes this bug by re-ordering these renderers.
Tests: fast/forms/input-placeholder-paint-order-2.html
fast/forms/input-placeholder-paint-order.html
fast/forms/textarea/textarea-placeholder-paint-order-2.html
fast/forms/textarea/textarea-placeholder-paint-order.html
* html/HTMLTextFormControlElement.cpp:
(WebCore::HTMLTextFormControlElement::fixPlaceholderRenderer):
Added. Reorder the order of renderers so that the placeholder renderer
precedes the inner text renderer.
* html/HTMLTextFormControlElement.h: Add fixPlaceholderRenderer() declaration.
* html/HTMLTextAreaElement.cpp:
(WebCore::HTMLTextAreaElement::attach): Calls fixPlaceholderRenderer().
(WebCore::HTMLTextAreaElement::updatePlaceholderText):
ditto. Also, use innerTextElement() to improvde code readability.
* html/HTMLTextAreaElement.h:
(HTMLTextAreaElement): Overrides attach().
* html/TextFieldInputType.cpp:
(WebCore::TextFieldInputType::updatePlaceholderText):
Calls fixPlaceholderRenderer().
(WebCore::TextFieldInputType::attach): ditto.
* html/TextFieldInputType.h:
(TextFieldInputType): Overrides attach().
2012-05-28 Kentaro Hara <haraken@chromium.org>
[V8] Implement V8Binding::v8Null(isolate) and use it in CodeGeneratorV8.pm
https://bugs.webkit.org/show_bug.cgi?id=87692
Reviewed by Adam Barth.
Since v8::Null(isolate) crashes if we pass a NULL isolate, we are planning
to pass Isolate to v8::Null() in the following steps:
[1] Implement V8Bindings::v8Null(isolate). v8Null(isolate) does the NULL check.
If isolate is NULL, v8Null(isolate) calls v8::Null(). Otherwise,
v8Null(isolate) calls v8::Null(isolate).
[2] In V8 bindings, we replace v8::Null() with v8::Null(isolate) for a non-optional
'isolate' parameter. (e.g. void foo(..., Isolate* isolate) { v8::Null(); } )
[3] In V8 bindings, we replace v8::Null() with v8Null(isolate) for an optional
'isolate' parameter. (e.g. void foo(..., Isolate* isolate = 0) { v8::Null(); } )
This bug fixes [1] by implementing V8Binding::v8Null(isolate). Also this patch uses
V8Binding::v8Null(isolate) in bindings/v8/*.{h,cpp}.
No tests. No behavior change.
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateHeader):
(NativeToJSValue):
* bindings/scripts/test/V8/V8Float64Array.h:
* bindings/scripts/test/V8/V8TestActiveDOMObject.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestCustomNamedGetter.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestEventConstructor.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestEventTarget.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestException.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestInterface.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestMediaQueryListListener.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestNamedConstructor.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestNode.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestObj.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.cpp:
(WebCore::TestSerializedScriptValueInterfaceV8Internal::valueAttrGetter):
(WebCore::TestSerializedScriptValueInterfaceV8Internal::readonlyValueAttrGetter):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.h:
(WebCore::toV8):
* bindings/v8/V8Binding.h:
(WebCore::v8Null):
(WebCore):
(WebCore::v8DateOrNull):
* bindings/v8/V8DOMWrapper.cpp:
* bindings/v8/V8DOMWrapper.h:
(WebCore):
2012-05-28 Kent Tamura <tkent@chromium.org>
Fix a crash in HTMLFormControlElement::disabled().
https://bugs.webkit.org/show_bug.cgi?id=86534
Reviewed by Ryosuke Niwa.
Stop to hold pointers of fildset and legend elements. We can avoid it by
holding ancestor's disabled state.
The ancesotr's disabled state should be invalidated when
- fieldset's disabled value is changed.
- fieldset's children is updated because a legend position might be changed.
- A form control is attached to or detached from a tree.
No new tests. It's almost impossible to make a reliable test.
* html/HTMLFieldSetElement.cpp:
(WebCore::HTMLFieldSetElement::invalidateDisabledStateUnder):
Added. Invalidate disabled state of form controls under the specified node.
(WebCore::HTMLFieldSetElement::disabledAttributeChanged):
Uses invalidateDisabledStateUnder().
(WebCore::HTMLFieldSetElement::childrenChanged):
Added new override function. We need invalidate disabled state of form
controls under legend elements.
* html/HTMLFieldSetElement.h:
(HTMLFieldSetElement): Add invalidateDisabledStateUnder() and childrenChanged().
* html/HTMLFormControlElement.cpp:
(WebCore::HTMLFormControlElement::HTMLFormControlElement):
Remove initialization of the removed data members.
Initialize m_ancestorDisabledState.
(WebCore::HTMLFormControlElement::updateAncestorDisabledState):
Update m_ancestorDisabledState. It should be
AncestorDisabledStateDisabled if the control is under a disabled
fieldset and not under the first legend child of the disabled filedset.
(WebCore::HTMLFormControlElement::ancestorDisabledStateWasChanged):
Invalidate m_ancestorDisabledState.
(WebCore::HTMLFormControlElement::insertedInto): ditto.
(WebCore::HTMLFormControlElement::removedFrom): ditto.
(WebCore::HTMLFormControlElement::disabled):
Calls updateAncestorDisabledState() if needed.
(WebCore::HTMLFormControlElement::recalcWillValidate):
Remove unnecessary check for m_legendAncestor.
* html/HTMLFormControlElement.h:
(HTMLFormControlElement):
- Rename updateFieldSetAndLegendAncestor() to updateAncestorDisabledState(), and make it private.
- Remove m_fieldSetAncestor, m_legendAncestor, and m_fieldSetAncestorValid.
- Add m_ancestorDisabledState.
2012-05-28 Takashi Toyoshima <toyoshim@chromium.org>
[WebSocket] Receiving reserved close codes, 1005, 1006, and 1015 must appear as code=1006 and wasClean=false
https://bugs.webkit.org/show_bug.cgi?id=87084
Reviewed by Kent Tamura.
Status codes 1005, 1006, and 1015 are forbidden to be sent in actual close frames.
If a client received these frames, the client should handle them as broken.
Close frames containing invalid body size are the same as these forbidden cases.
Update close-code-and-reason tests to verify this patch.
* Modules/websockets/WebSocket.cpp: Handle AbnormalClosure as wasClean == false
(WebCore::WebSocket::didClose):
* Modules/websockets/WebSocketChannel.cpp: Handle close frames' status code carefully
(WebCore::WebSocketChannel::processFrame):
* Modules/websockets/WebSocketChannel.h: Update on newly defined close event codes
2012-05-28 Kentaro Hara <haraken@chromium.org>
[V8] Avoid passing NULL to an 'isolate' parameter
https://bugs.webkit.org/show_bug.cgi?id=87689
Reviewed by Adam Barth.
v8::Null(isolate) crashes if we pass a NULL isolate.
Thus we are planning to replace v8::Null()s in a following way:
- Implement V8Bindings::v8Null(isolate). v8Null(isolate) does the NULL check.
If isolate is NULL, v8Null(isolate) calls v8::Null(). Otherwise,
v8Null(isolate) calls v8::Null(isolate).
- In V8 bindings, we replace v8::Null() with v8::Null(isolate) for a
non-optional 'isolate' parameter.
(e.g. void foo(..., Isolate* isolate) { v8::Null(); } )
- In V8 bindings, we replace v8::Null() with v8Null(isolate) for an
optional 'isolate' parameter.
(e.g. void foo(..., Isolate* isolate = 0) { v8::Null(); } )
However, currently we cannot do the replacement mechanically, since some code
pass NULL to a non-optional 'isolate' parameter. In other words, currently
"non-optional" does not guarantee that 'isolate' is not NULL.
This patch removes all the code that passes NULL to a non-optional 'isolate'
parameter. This will enable us to achieve the replacement mechanically.
No tests. No behavior change.
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateCallbackImplementation):
(NativeToJSValue):
* bindings/scripts/test/V8/V8TestCallback.cpp:
(WebCore::V8TestCallback::callbackWithClass1Param):
(WebCore::V8TestCallback::callbackWithClass2Param):
(WebCore::V8TestCallback::callbackWithStringList):
(WebCore::V8TestCallback::callbackRequiresThisToPass):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::V8TestObj::installPerContextProperties):
* bindings/v8/custom/V8CustomSQLStatementErrorCallback.cpp:
(WebCore::V8SQLStatementErrorCallback::handleEvent):
* bindings/v8/custom/V8MutationCallbackCustom.cpp:
(WebCore::V8MutationCallback::handleEvent):
2012-05-28 Kent Tamura <tkent@chromium.org>
Form controls in <fieldset disabled> should not be validated.
https://bugs.webkit.org/show_bug.cgi?id=87381
Reviewed by Hajime Morita.
We need to use disabeld() instead of m_disabled to calculate
willValidate property. Also, we need to update willValidate if
necessary.
Test: fast/forms/fieldset/validation-in-fieldset.html
* html/HTMLFieldSetElement.cpp:
(WebCore::HTMLFieldSetElement::disabledAttributeChanged):
- Do not traverse this.
- Calls ancestorDisabledStateWasChanged() instead of
setNeedsStyleRecalc() because we'd like to do additional tasks.
* html/HTMLFormControlElement.cpp:
(WebCore::HTMLFormControlElement::ancestorDisabledStateWasChanged):
Added. Just calls disabledAttributeChanged().
(WebCore::HTMLFormControlElement::parseAttribute):
Do not call setNeedsWillValidateCheck() whenever an attribute is updated.
It should be called only if disabled or readonly attribute is updated.
(WebCore::HTMLFormControlElement::disabledAttributeChanged):
Add setNeedsWillValidateCheck(). It was moved from parseAttribute().
(WebCore::HTMLFormControlElement::insertedInto):
Invalidate ancestor information.
(WebCore::HTMLFormControlElement::recalcWillValidate):
Use disabled() instead of m_disabled. disabled() takes care of
ancestor's disabled state.
* html/HTMLFormControlElement.h:
(HTMLFormControlElement):
2012-05-28 Rakesh KN <rakesh.kn@motorola.com>
[Forms] HTMLFieldSetElement.idl doesn't have elements attribute.
https://bugs.webkit.org/show_bug.cgi?id=80110
Reviewed by Kent Tamura.
Implemented elements attribute for HTMLFieldSetElement. This is spec'ed at
http://www.whatwg.org/specs/web-apps/current-work/multipage/forms.html#dom-fieldset-elements
Tests: fast/forms/fieldset/fieldset-elements.html
fast/forms/fieldset/fieldset-form-collection-radionode-list.html
* dom/Node.cpp:
(WebCore::Node::radioNodeList):
Extended ASSERT to assert if not HTMLFieldSetElement or HTMLFormElement.
* html/HTMLFieldSetElement.cpp:
(WebCore::HTMLFieldSetElement::elements):
Elements attribute implementation.
(WebCore::HTMLFieldSetElement::refreshElementsIfNeeded):
Update the formcontrol elements collections if dom tree got modified.
(WebCore::HTMLFieldSetElement::associatedElements):
FormControl elements collection accessor.
(WebCore::HTMLFieldSetElement::length):
Number of elements in the fieldset group.
* html/HTMLFieldSetElement.h:
Added elements collection member and form control collection members.
* html/HTMLFieldSetElement.idl:
Added elements attribute.
* html/HTMLFormCollection.cpp:
(WebCore::HTMLFormCollection::formControlElements):
Added support for HTMLFieldSetElement, based on base element type gets its associated elements.
(WebCore::HTMLFormCollection::numberOfFormControlElements): Ditto
(WebCore::HTMLFormCollection::getNamedFormItem): Process image elements only for form element.
(WebCore::HTMLFormCollection::updateNameCache): Ditto
* html/RadioNodeList.cpp:
(WebCore::RadioNodeList::RadioNodeList):
DynamicSubTree root element is decided based on the type whether base element is form or fieldset element.
Renamed m_formElement to m_baseElement.
(WebCore::RadioNodeList::~RadioNodeList):
Renamed m_formElement to m_baseElement.
(WebCore::RadioNodeList::checkElementMatchesRadioNodeListFilter):
Form element specific changes moved under form element check.
* html/RadioNodeList.h:
(WebCore::RadioNodeList::create):
(RadioNodeList):
Renamed m_formElement to m_baseElement.
2012-05-28 Jonathan Dong <jonathan.dong@torchmobile.com.cn>
[BlackBerry] http authenticate dialog popup only once no matter authentication pass or fail
https://bugs.webkit.org/show_bug.cgi?id=80135
Reviewed by Rob Buis.
RIM PR: 145660
Fixed a regression introduced by r111810, we should cancel the new
request when user press cancel button in http authentication challenge
dialog, and we should also allow sending empty username and password
with the request.
Also removed redundant codes which checked the existence of the
FrameLoaderClient pointer, as we've already moved authenticationChallenge()
out of class FrameLoaderClient, it is not needed.
Manual test added. Testing http authentication dialog relies on user interaction.
Resubmit the patch reverted by r115104 after the digest infinite loop
issue for BlackBerry porting get identified and fixed.
Internally reviewed by Joe Mason <jmason@rim.com>
* platform/blackberry/PageClientBlackBerry.h:
* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::sendRequestWithCredentials):
2012-05-28 Jonathan Dong <jonathan.dong@torchmobile.com.cn>
[BlackBerry] http authentication challenge issue when loading favicon
https://bugs.webkit.org/show_bug.cgi?id=87665
Reviewed by Rob Buis.
Provide the TargetType when generating a favicon loading
request. Loading favicons is triggered after the main resource
has been loaded and parsed, so if we cancel the authentication
challenge when loading main resource, we should also cancel
loading the favicon when it starts to load. If not we will
receive another challenge after we canceled the main resource
loading, which may confuse the user.
Internally reviewed by Joe Mason <jmason@rim.com>
No new tests because of no behavior changes.
* loader/icon/IconLoader.cpp:
(WebCore::IconLoader::startLoading):
* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::sendRequestWithCredentials):
2012-05-28 MORITA Hajime <morrita@google.com>
Rename FrameLoaderClient::shadowDOMAllowed() to allowShadowDOM()
https://bugs.webkit.org/show_bug.cgi?id=87101
Reviewed by Kentaro Hara.
No new tests. No behavior change.
* bindings/generic/ContextEnabledFeatures.cpp:
(WebCore::ContextEnabledFeatures::shadowDOMEnabled):
* loader/FrameLoaderClient.h:
(WebCore::FrameLoaderClient::allowShadowDOM):
2012-05-28 David Barton <dbarton@mathscribe.com>
mathml/presentation/mo-stretch.html and mroot-pref-width.html tests fail on Mac
https://bugs.webkit.org/show_bug.cgi?id=86786
Reviewed by Darin Adler.
The fix to https://bugs.webkit.org/show_bug.cgi?id=84167 changed inline-table baselines,
including for <mtable>. We therefore have to correct <mtable>'s vertical-align CSS
property.
Tested by existing tests mo-stretch.html, row-alignment.xhtml, and tables.xhtml.
* css/mathml.css:
(mtable):
2012-05-28 Luke Macpherson <macpherson@chromium.org>
Make CSSParser::filteredProperties() O(n) instead of O(n^2) and improve readability.
https://bugs.webkit.org/show_bug.cgi?id=87078
Reviewed by Darin Adler.
This patch implements a number of improvements to filteredProperties:
1) Make the code more linearly readable by separating out handling of important and non-important properties.
2) Eliminate one BitArray instance (reduces hot memory so more cache friendly).
3) Remove O(n^2) behavior caused by scanning for and removing previously encountered definitions of each property.
The key algorithmic change is to add properties in decreasing precedence:
a) Iterating once per (important, !important) so that important properties are visited first.
b) Reverse iteration of m_parsedProperties visits the properties in decreasing precedence.
Covered by loads of existing tests - getting CSS property precedence wrong results in too many errors to list.
In particular fast/css contains test cases for important corner cases like duplicated important properties.
* css/CSSParser.cpp:
(WebCore::CSSParser::createStylePropertySet):
* css/CSSProperty.h:
Add vector traits so that CSSProperty can just be memset by vector without calling constructor.
2012-05-28 MORITA Hajime <morrita@google.com>
Can't edit <input> elements with :first-letter
https://bugs.webkit.org/show_bug.cgi?id=87615
Reviewed by Kent Tamura.
A check using canHaveGeneratedChildren() should cover not only
button and menulist, but also RenderTextControl.
This change pulles canHaveGeneratedChildren() up
from RenderDeprecatedFlexibleBox to RenderObject,
and lets RenderTextControl override it.
Test: fast/forms/input-first-letter-edit.html
* rendering/RenderBlock.cpp:
(WebCore):
(WebCore::RenderBlock::styleDidChange):
(WebCore::RenderBlock::updateFirstLetter):
* rendering/RenderBlock.h: Removed a static function canHaveGeneratedChildren()
(RenderBlock):
* rendering/RenderDeprecatedFlexibleBox.h:
* rendering/RenderObject.cpp:
(WebCore):
(WebCore::RenderObject::canHaveGeneratedChildren): Added.
* rendering/RenderObject.h:
(RenderObject):
* rendering/RenderTextControl.h:
2012-05-28 Arvid Nilsson <anilsson@rim.com>
[BlackBerry] Add support for layers with scale invariant size
https://bugs.webkit.org/show_bug.cgi?id=87601
Reviewed by Rob Buis.
To support layers that have a "floating" appearance, i.e. don't change size
when the web page is drawn at a different scale, we add a new layer property
named "sizeIsScaleInvariant".
The anchor position will still be given in document coordinates for these
"floating" layers, so this is well suited for interface elements like selection
handles whose size is always the same but move with the web page contents.
PR #156812
* platform/graphics/blackberry/LayerCompositingThread.cpp:
(WebCore::LayerCompositingThread::setDrawTransform):
(WebCore::LayerCompositingThread::drawTextures):
(WebCore::LayerCompositingThread::drawMissingTextures):
* platform/graphics/blackberry/LayerCompositingThread.h:
(LayerCompositingThread):
* platform/graphics/blackberry/LayerCompositingThreadClient.h:
(LayerCompositingThreadClient):
(WebCore::LayerCompositingThreadClient::drawMissingTextures):
* platform/graphics/blackberry/LayerData.h:
(WebCore::LayerData::LayerData):
(WebCore::LayerData::sizeIsScaleInvariant):
(LayerData):
* platform/graphics/blackberry/LayerRenderer.cpp:
(WebCore::LayerRenderer::LayerRenderer):
(WebCore::LayerRenderer::compositeLayers):
(WebCore::LayerRenderer::updateLayersRecursive):
(WebCore::LayerRenderer::compositeLayersRecursive):
* platform/graphics/blackberry/LayerRenderer.h:
(LayerRenderer):
* platform/graphics/blackberry/LayerTiler.cpp:
(WebCore::LayerTiler::updateTextureContentsIfNeeded):
(WebCore::LayerTiler::drawTextures):
(WebCore::LayerTiler::drawMissingTextures):
(WebCore::LayerTiler::drawTexturesInternal):
* platform/graphics/blackberry/LayerTiler.h:
(LayerTiler):
* platform/graphics/blackberry/LayerWebKitThread.h:
(WebCore::LayerWebKitThread::setSizeIsScaleInvariant):
(LayerWebKitThread):
2012-05-28 Arvid Nilsson <anilsson@rim.com>
[BlackBerry] Make it possible to manipulate layers on the compositing thread
https://bugs.webkit.org/show_bug.cgi?id=87602
Reviewed by Rob Buis.
Normally, layers are manipulated on the WebKit thread, and the changes are
synced to the compositing thread during the next accelerated compositing commit
operation.
However, for overlay layers the ability to manipulate layers on the compositing
thread is necessary for adequate user interface responsiveness.
Two mechanisms are added:
1. For a layer with a WebKit-thread counterpart, you can temporarily override
attributes or add animations whose output override attributes.
2. For a layer with no WebKit-thread counterpart, you can now directly set the
compositing thread values for attributes. If you attempt to do this for layers
that do have a WebKit-thread counterpart, the compositing thread values will
be overwritten at the time of the next commit, which makes the override
mechanism more useful there.
PR #156812
* platform/graphics/blackberry/LayerCompositingThread.cpp:
(WebCore::LayerCompositingThread::addSublayer):
(WebCore):
(WebCore::LayerCompositingThread::updateAnimations):
(WebCore::LayerCompositingThread::removeAnimation):
(WebCore::LayerCompositingThread::override):
(WebCore::LayerCompositingThread::clearOverride):
* platform/graphics/blackberry/LayerCompositingThread.h:
(LayerOverride):
(WebCore::LayerOverride::create):
(WebCore::LayerOverride::setPosition):
(WebCore::LayerOverride::setAnchorPoint):
(WebCore::LayerOverride::setBounds):
(WebCore::LayerOverride::setTransform):
(WebCore::LayerOverride::setOpacity):
(WebCore::LayerOverride::addAnimation):
(WebCore::LayerOverride::LayerOverride):
(WebCore):
(LayerCompositingThread):
(WebCore::LayerCompositingThread::setPosition):
(WebCore::LayerCompositingThread::setAnchorPoint):
(WebCore::LayerCompositingThread::setBounds):
(WebCore::LayerCompositingThread::setSizeIsScaleInvariant):
(WebCore::LayerCompositingThread::setTransform):
(WebCore::LayerCompositingThread::setOpacity):
(WebCore::LayerCompositingThread::setNeedsTexture):
* platform/graphics/blackberry/LayerWebKitThread.cpp:
(WebCore::LayerWebKitThread::LayerWebKitThread):
(WebCore::LayerWebKitThread::commitOnCompositingThread):
* platform/graphics/blackberry/LayerWebKitThread.h:
(LayerWebKitThread):
(WebCore::LayerWebKitThread::clearOverride):
2012-05-28 Rob Flack <flackr@chromium.org>
[chromium] Only increase size of Combo Box Options when displayed on touch screen
https://bugs.webkit.org/show_bug.cgi?id=85921
Reviewed by Adam Barth.
Adds a flag to set whether the current device is a touch screen, independent of whether touch events are supported and use this for the combo box sizing.
No new tests as this is a flag change and covered by existing tests: WebKit/chromium/tests/PopupMenuTest.cpp
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(WebCore::Settings::setDeviceSupportsTouch):
(WebCore::Settings::deviceSupportsTouch):
(Settings):
* platform/chromium/PopupListBox.cpp:
(WebCore::PopupListBox::getRowHeight):
* platform/chromium/PopupListBox.h:
(PopupContainerSettings):
* platform/chromium/PopupMenuChromium.cpp:
(WebCore::PopupMenuChromium::show):
2012-05-28 Arvid Nilsson <anilsson@rim.com>
[BlackBerry] Make custom compositing thread layers more flexible
https://bugs.webkit.org/show_bug.cgi?id=87600
Reviewed by Rob Buis.
Introduce a LayerCompositingThreadClient that's used to fine tune the
behaviour of custom layers. Let the LayerTiler be a
LayerCompositingThreadClient and thus decouple it from
LayerCompositingThread. Adjust method signatures to allow a one-to-many
relationship between Client and Layer.
Remove the old LayerCompositingThread::drawCustom() in favour of this new
Client interface.
PR #156812
* platform/graphics/blackberry/LayerCompositingThread.cpp:
(WebCore::LayerCompositingThread::create):
(WebCore::LayerCompositingThread::LayerCompositingThread):
(WebCore::LayerCompositingThread::~LayerCompositingThread):
(WebCore::LayerCompositingThread::deleteTextures):
(WebCore::LayerCompositingThread::drawTextures):
(WebCore::LayerCompositingThread::hasMissingTextures):
(WebCore::LayerCompositingThread::drawMissingTextures):
(WebCore::LayerCompositingThread::updateTextureContentsIfNeeded):
(WebCore::LayerCompositingThread::bindContentsTexture):
(WebCore::LayerCompositingThread::setVisible):
(WebCore::LayerCompositingThread::scheduleCommit):
* platform/graphics/blackberry/LayerCompositingThread.h:
(WebCore):
(LayerCompositingThread):
* platform/graphics/blackberry/LayerCompositingThreadClient.h: Added.
(WebCore):
(LayerCompositingThreadClient):
(WebCore::LayerCompositingThreadClient::~LayerCompositingThreadClient):
(WebCore::LayerCompositingThreadClient::bindContentsTexture):
(WebCore::LayerCompositingThreadClient::hasMissingTextures):
(WebCore::LayerCompositingThreadClient::drawMissingTextures):
(WebCore::LayerCompositingThreadClient::scheduleCommit):
* platform/graphics/blackberry/LayerRenderer.cpp:
(WebCore::LayerRenderer::drawDebugBorder):
* platform/graphics/blackberry/LayerTiler.cpp:
(WebCore::LayerTiler::LayerTiler):
(WebCore::LayerTiler::layerCompositingThreadDestroyed):
(WebCore::LayerTiler::layerVisibilityChanged):
(WebCore::LayerTiler::uploadTexturesIfNeeded):
(WebCore::LayerTiler::deleteTextures):
(WebCore::LayerTiler::scheduleCommit):
(WebCore):
(WebCore::LayerTiler::bindContentsTexture):
* platform/graphics/blackberry/LayerTiler.h:
(LayerTiler):
(WebCore::LayerTiler::hasMissingTextures):
* platform/graphics/blackberry/LayerWebKitThread.cpp:
(WebCore::LayerWebKitThread::LayerWebKitThread):
* platform/graphics/blackberry/LayerWebKitThread.h:
(LayerWebKitThread):
2012-05-25 Jesus Sanchez-Palencia <jesus.palencia@openbossa.org>
WebKitTestRunner needs to support layoutTestController.setJavaScriptProfilingEnabled
https://bugs.webkit.org/show_bug.cgi?id=42328
Reviewed by Eric Seidel.
Add setJavaScriptProfilingEnabled() to window.internals.settings. No new tests, but this
change will allow more tests to run in WebKitTestRunner and DRT for ports that weren't
implementing this function before.
This patch also refactors InspectorController::enableProfiler() and
InspectorController::disableProfiler() to InspectorController::setProfilerEnabled(bool).
* WebCore.exp.in:
* inspector/InspectorController.cpp:
(WebCore::InspectorController::profilerEnabled):
(WebCore::InspectorController::setProfilerEnabled):
* inspector/InspectorController.h:
(InspectorController):
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::InternalSettings):
(WebCore::InternalSettings::restoreTo):
(WebCore::InternalSettings::setJavaScriptProfilingEnabled):
(WebCore):
* testing/InternalSettings.h:
(InternalSettings):
* testing/InternalSettings.idl:
2012-05-28 Yong Li <yoli@rim.com>
Crash on incomplete :not().
https://bugs.webkit.org/show_bug.cgi?id=86673
Reviewed by Antti Koivisto.
Add back null-checks for incomplete :not() class
which were dropped by r81845.
* css/CSSSelector.cpp:
(WebCore::CSSSelector::specificityForOneSelector):
(WebCore::CSSSelector::selectorText):
* css/SelectorChecker.cpp:
(WebCore::SelectorChecker::checkOneSelector):
(WebCore::SelectorChecker::determineLinkMatchType):
2012-05-28 Leo Yang <leo.yang@torchmobile.com.cn>
FileWriterSync binding should have no static table
https://bugs.webkit.org/show_bug.cgi?id=87645
Reviewed by George Staikos.
FileWriterSync could be used in filesystem FileWriter in worker thread, so we should add ProgressEvent in idl file.
Covered by existing test: fast/filesystem/workers/sync-operations.html.
* Modules/filesystem/FileWriterSync.idl:
2012-05-28 Arvid Nilsson <anilsson@rim.com>
[BlackBerry] Add an overlay layer
https://bugs.webkit.org/show_bug.cgi?id=87567
Reviewed by Antonio Gomes.
The overlay layer allows us to have compositing layers even though the
web page is not currently using accelerated compositing.
These layers can be used to implement tap highlight, inspector overlay
and more.
The WebCore changes support the overlay layer functionality in WebKit
by making it possible to add animations from the compositing thread
without them being instantly overwritten during the next commit.
Also a new custom layer type is added, which allow an overlay layer to
draw itself using raw OpenGL calls. An instance of a custom subclass of
LayerCompositingThread must be provided when creating the
LayerWebKitThread. Then, the custom layer can be used as the content
layer of a GraphicsLayer, or by itself.
Reviewed internally by Filip Spacek.
PR #154335
* platform/graphics/blackberry/LayerCompositingThread.cpp:
(WebCore::LayerCompositingThread::~LayerCompositingThread):
(WebCore::LayerCompositingThread::deleteTextures):
(WebCore::LayerCompositingThread::drawTextures):
(WebCore::LayerCompositingThread::drawMissingTextures):
(WebCore::LayerCompositingThread::updateTextureContentsIfNeeded):
(WebCore::LayerCompositingThread::setVisible):
(WebCore::LayerCompositingThread::scheduleCommit):
* platform/graphics/blackberry/LayerCompositingThread.h:
(WebCore::LayerCompositingThread::addAnimation):
(WebCore::LayerCompositingThread::setRunningAnimations):
(WebCore::LayerCompositingThread::setSuspendedAnimations):
(LayerCompositingThread):
(WebCore::LayerCompositingThread::drawCustom):
* platform/graphics/blackberry/LayerData.h:
(LayerData):
* platform/graphics/blackberry/LayerWebKitThread.cpp:
(WebCore::LayerWebKitThread::LayerWebKitThread):
(WebCore):
(WebCore::LayerWebKitThread::~LayerWebKitThread):
(WebCore::LayerWebKitThread::updateTextureContentsIfNeeded):
(WebCore::LayerWebKitThread::startAnimations):
(WebCore::LayerWebKitThread::commitOnCompositingThread):
(WebCore::LayerWebKitThread::setNeedsDisplayInRect):
(WebCore::LayerWebKitThread::setNeedsDisplay):
(WebCore::LayerWebKitThread::setIsMask):
(WebCore::LayerWebKitThread::setRunningAnimations):
(WebCore::LayerWebKitThread::setSuspendedAnimations):
* platform/graphics/blackberry/LayerWebKitThread.h:
(LayerWebKitThread):
2012-05-28 Leo Yang <leo.yang@torchmobile.com.cn>
ProgressEvent JSC binding should have no static table
https://bugs.webkit.org/show_bug.cgi?id=87365
Reviewed by George Staikos.
ProgressEvent could be used in filesystem FileWriter in worker thread, so we should add ProgressEvent in idl file.
Covered by existing test: fast/filesystem/workers/file-writer-events.html.
* dom/ProgressEvent.idl:
2012-05-28 Arvid Nilsson <anilsson@rim.com>
[BlackBerry] Update WebPageCompositor::render() API
https://bugs.webkit.org/show_bug.cgi?id=87565
Reviewed by Rob Buis.
The new API allows the embedder to specify the root transform and many
OpenGL related parameters.
Also refactor the code to allow several sets of layers to be rendered,
and to allow interleaving the rendering of layers with rendering of
buffers and checkerboard.
Reviewed internally by Filip Spacek.
PR #154334
* platform/graphics/blackberry/LayerCompositingThread.cpp:
(WebCore::LayerCompositingThread::drawTextures):
(WebCore::LayerCompositingThread::drawSurface):
* platform/graphics/blackberry/LayerRenderer.cpp:
(WebCore::LayerRenderer::orthoMatrix):
(WebCore::LayerRenderer::LayerRenderer):
(WebCore::LayerRenderer::prepareFrame):
(WebCore):
(WebCore::LayerRenderer::setViewport):
(WebCore::LayerRenderer::compositeLayers):
(WebCore::LayerRenderer::compositeBuffer):
(WebCore::LayerRenderer::drawCheckerboardPattern):
(WebCore::LayerRenderer::drawLayersOnSurfaces):
(WebCore::LayerRenderer::prepareFrameRecursive):
(WebCore::LayerRenderer::updateLayersRecursive):
(WebCore::LayerRenderer::compositeLayersRecursive):
(WebCore::LayerRenderer::updateScissorIfNeeded):
(WebCore::LayerRenderingResults::addHolePunchRect):
* platform/graphics/blackberry/LayerRenderer.h:
(LayerRenderer):
2012-05-28 Antti Koivisto <antti@apple.com>
REGRESSION(r96517): Attribute selector fails to match dynamically modified style attribute
https://bugs.webkit.org/show_bug.cgi?id=87349
Reviewed by Andreas Kling.
Selector fast path does not trigger lazy style attribute generation. Since attribute selectors matching
style attribute are rare, disallow them from the fast path rather than making it more branchy.
Test: fast/css/dynamic-style-attribute-query.html
* css/SelectorChecker.cpp:
(WebCore::isFastCheckableMatch):
2012-05-28 Peter Rybin <peter.rybin@gmail.com>
Web Inspector: Expose function (closure) scopes in remote protocol
https://bugs.webkit.org/show_bug.cgi?id=86861
Reviewed by Yury Semikhatsky.
A data transfer from V8's FunctionMirror via DebuggerScript.js via InjectedScriptHost is built.
Scope field is added to protocol declaration similar to scopes of stack call frame.
Test for function details is extended.
JSC code binging got fixme for implemting the corresponding feature.
* bindings/js/JSInjectedScriptHostCustom.cpp:
(WebCore::JSInjectedScriptHost::functionDetails):
* bindings/v8/DebuggerScript.js:
* bindings/v8/ScriptDebugServer.cpp:
(WebCore::ScriptDebugServer::functionScopes):
(WebCore):
* bindings/v8/ScriptDebugServer.h:
(ScriptDebugServer):
* bindings/v8/custom/V8InjectedScriptHostCustom.cpp:
(WebCore::V8InjectedScriptHost::functionDetailsCallback):
* inspector/InjectedScriptHost.cpp:
(WebCore):
(WebCore::InjectedScriptHost::scriptDebugServer):
* inspector/InjectedScriptHost.h:
(WebCore):
(WebCore::InjectedScriptHost::init):
(InjectedScriptHost):
* inspector/InjectedScriptSource.js:
(.):
* inspector/Inspector.json:
* inspector/InspectorController.cpp:
(WebCore::InspectorController::InspectorController):
* inspector/WorkerInspectorController.cpp:
(WebCore::WorkerInspectorController::WorkerInspectorController):
2012-05-28 Arvid Nilsson <anilsson@rim.com>
[BlackBerry] Add a constructor to create a Path from an SkPath
https://bugs.webkit.org/show_bug.cgi?id=87566
Reviewed by Antonio Gomes.
Skia allows the creation of more complex paths than WebCore would
allow. This will be used in the BlackBerry port to implement a default
tap highlight appearance.
Reviewed internally by Mike Lattanzio.
PR #154329
* platform/graphics/Path.h:
(Path):
* platform/graphics/skia/PathSkia.cpp:
(WebCore):
(WebCore::Path::Path):
2012-05-28 Alexis Menard <alexis.menard@openbossa.org>
Unreviewed build fix for Mac on Lion.
* inspector/CodeGeneratorInspector.py:
2012-05-28 Keishi Hattori <keishi@webkit.org>
Expose value localization function of HTMLInputElement
https://bugs.webkit.org/show_bug.cgi?id=84356
Reviewed by Kent Tamura.
No new tests.
We want to localize the values that are defined in the datalist element.
This adds HTMLInputElement::localizeValue() which will localize a given
value.
* html/BaseDateAndTimeInputType.cpp:
(WebCore::BaseDateAndTimeInputType::localizeValue):
(WebCore):
(WebCore::BaseDateAndTimeInputType::visibleValue):
* html/BaseDateAndTimeInputType.h:
(BaseDateAndTimeInputType):
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::localizeValue):
(WebCore):
* html/HTMLInputElement.h:
(HTMLInputElement):
* html/InputType.cpp:
(WebCore::InputType::localizeValue):
(WebCore):
* html/InputType.h:
(InputType):
* html/NumberInputType.cpp:
(WebCore::NumberInputType::localizeValue):
(WebCore):
(WebCore::NumberInputType::visibleValue):
* html/NumberInputType.h:
(NumberInputType):
2012-05-28 Yury Semikhatsky <yurys@chromium.org>
Unreviewed. Test fix after r118670: saved timeline data may
not have DOM counters data and MemoryStatistics object should
take this into account.
* inspector/front-end/MemoryStatistics.js:
(WebInspector.MemoryStatistics.prototype._onRecordAdded):
2012-05-28 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118580.
http://trac.webkit.org/changeset/118580
https://bugs.webkit.org/show_bug.cgi?id=87647
Caused webkit_unit_tests to crash on chromium. (Requested by
bulach on #webkit).
* platform/graphics/chromium/ContentLayerChromium.cpp:
(WebCore::ContentLayerPainter::create):
(WebCore::ContentLayerPainter::paint):
(WebCore::ContentLayerPainter::ContentLayerPainter):
(ContentLayerPainter):
(WebCore::ContentLayerChromium::createTextureUpdaterIfNeeded):
* platform/graphics/chromium/TiledLayerChromium.cpp:
(UpdatableTile):
(WebCore::UpdatableTile::UpdatableTile):
(WebCore::TiledLayerChromium::updateTiles):
* platform/graphics/chromium/TiledLayerChromium.h:
(TiledLayerChromium):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCSettings::CCSettings):
(CCSettings):
2012-05-28 Peter Beverloo <peter@chromium.org>
&AElig doesn't get rendered as U+00C6
https://bugs.webkit.org/show_bug.cgi?id=87465
Reviewed by Adam Barth.
The if-statement to check whether the first entry of a certain letter
already exists in the index used .get(). The very first alphabetical
entry is in position "0", which evaluates to false, causing the first
entry to be ignored. Instead, use a "x not in y" check here.
Also update WebCore.gyp to list the create-html-entity-table script as
an input for the action, to make sure the table will be recreated.
Test: html5lib/resources/entities02.dat
* WebCore.gyp/WebCore.gyp:
* html/parser/create-html-entity-table:
2012-05-28 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: dom counters graphs vanish on switching between timeline views
https://bugs.webkit.org/show_bug.cgi?id=87628
Reviewed by Vsevolod Vlasov.
MemoryStatistics object now listens to the TimelineModel events instead of being
populated by timeline panel. This behavior mathes that of the overview pane.
* inspector/front-end/MemoryStatistics.js:
(WebInspector.MemoryStatistics.prototype.addTimlineEvent):
* inspector/front-end/TimelinePanel.js:
(WebInspector.TimelinePanel.prototype._onTimelineEventRecorded):
(WebInspector.TimelinePanel.prototype._innerAddRecordToTimeline):
2012-05-28 Marcus Bulach <bulach@chromium.org>
[chromium] Buildfix: remove obsolete file from gypi.
https://bugs.webkit.org/show_bug.cgi?id=87643
Reviewed by no-reviewer.
Following http://trac.webkit.org/changeset/118610.
Removes reference to platform/graphics/gstreamer/ImageGStreamerCG.mm
* WebCore.gypi:
2012-05-28 MORITA Hajime <morrita@google.com>
HTMLFormControlElement::m_validationMessage shouldn't be cleared on detach()
https://bugs.webkit.org/show_bug.cgi?id=87608
Reviewed by Kent Tamura.
Moved m_validationMessage clearance from detach() to removedFrom() and
remove detach().
No new tests. This change has no visible difference. Upcoming changes rely on this though.
* html/HTMLFormControlElement.cpp:
(WebCore::HTMLFormControlElement::removedFrom):
2012-05-28 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: "Record CPU profile" button doesn't change its state after finishing profiling
https://bugs.webkit.org/show_bug.cgi?id=87624
Reviewed by Vsevolod Vlasov.
Keep record button state in sync with the current profiling state.
* inspector/front-end/ProfilesPanel.js:
(WebInspector.ProfilesPanel.prototype.toggleRecordButton):
2012-05-28 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118650.
http://trac.webkit.org/changeset/118650
https://bugs.webkit.org/show_bug.cgi?id=87639
Patch caused massive failures throughout the builders
(Requested by zdobersek on #webkit).
* css/MediaQuery.cpp:
(WebCore):
(WebCore::MediaQuery::serialize):
(WebCore::MediaQuery::MediaQuery):
(WebCore::MediaQuery::cssText):
* css/MediaQuery.h:
(WebCore::MediaQuery::expressions):
(WebCore::MediaQuery::mediaType):
(MediaQuery):
(WebCore::MediaQuery::copy):
* css/MediaQueryEvaluator.cpp:
(WebCore::MediaQueryEvaluator::MediaQueryEvaluator):
(WebCore::MediaQueryEvaluator):
(WebCore::MediaQueryEvaluator::eval):
(WebCore::aspect_ratioMediaFeatureEval):
(WebCore::device_aspect_ratioMediaFeatureEval):
(WebCore::transform_3dMediaFeatureEval):
(WebCore::view_modeMediaFeatureEval):
(WebCore::createFunctionMap):
* css/MediaQueryEvaluator.h:
(WebCore):
(MediaQueryEvaluator):
* css/StyleResolver.cpp:
(WebCore):
(WebCore::StyleResolver::collectMatchingRulesForList):
* css/StyleResolver.h:
(MediaQueryResult):
(WebCore::MediaQueryResult::MediaQueryResult):
(WebCore):
(StyleResolver):
2012-05-28 Gyuyoung Kim <gyuyoung.kim@samsung.com>
Move allowRoundingHacks to Internals interface
https://bugs.webkit.org/show_bug.cgi?id=87328
Reviewed by Hajime Morita.
Add allowRoundingHacks function, because it is able to work in the
cross-port way by means of the Internals interface.
In addition, a function is added in order to restore default setting values.
No new tests, since we are improving here the infra-structure for testing
a specific method.
* testing/Internals.cpp:
(WebCore::Internals::reset):
(WebCore):
(WebCore::Internals::resetDefaultsToConsistentValues):
(WebCore::Internals::allowRoundingHacks):
* testing/Internals.h:
(Internals):
* testing/Internals.idl:
2012-05-28 Jonathan Dong <jonathan.dong@torchmobile.com.cn>
[BlackBerry] http authentication crash the browser when user commit or cancel the http authentication dialog
https://bugs.webkit.org/show_bug.cgi?id=87579
Reviewed by George Staikos.
In function NetworkJob::startNewJobWithRequest, We should cancel the
NetworkJob first before we start a new NetworkJob which reuses the
resource handle of the old one. If we only set the m_handle = 0 of the
old NetworkJob without cancelling itself, it will still receives data
notification from network thread and handles the received data with
the released resource handler, which will cause the crash.
No new tests because no behavior has changed.
* platform/network/blackberry/NetworkJob.cpp:
(WebCore::NetworkJob::startNewJobWithRequest):
2012-05-27 MORITA Hajime <morrita@google.com>
[Refactoring][ShadowDOM] Some ElementShadow methods can be inlined.
https://bugs.webkit.org/show_bug.cgi?id=87617
Reviewed by Kentaro Hara.
Inlined ElementShadow::attachHost(), ElementShadow::detachHost() and
ElementShadow::reattach()
No new tests. No behavior change.
* dom/Element.cpp:
(WebCore::Element::attach):
(WebCore::Element::detach):
* dom/ElementShadow.cpp:
(WebCore::ElememtnShadow::reattachHostChildrenAndShadow):
* dom/ElementShadow.h:
(ElementShadow):
2012-05-28 Peter Rybin <peter.rybin@gmail.com>
Web Inspector: CodeGeneratorInspector.py: protect typed API from C++ implicit float to int cast
https://bugs.webkit.org/show_bug.cgi?id=87183
Reviewed by Yury Semikhatsky.
An intermediate C++ class is introduced that uses C++ template technique to control actual type
of its constructor argument.
All input parameters of type "int" now have type ExactlyInt.
All usage sites are fixed accordingly.
* inspector/CodeGeneratorInspector.py:
(TypeModel.RefPtrBased):
(TypeModel.Enum):
(TypeModel.ValueType):
(TypeModel.ValueType.get_opt_output_type_):
(TypeModel.ValueType.ValueOptional.get_command_return_pass_model):
(TypeModel.ExactlyInt):
(TypeModel.ExactlyInt.__init__):
(TypeModel.ExactlyInt.get_input_param_type_text):
(TypeModel.ExactlyInt.get_opt_output_type_):
(TypeModel.init_class):
(ExactlyInt):
* inspector/InspectorAgent.cpp:
(WebCore::InspectorAgent::enable):
(WebCore::InspectorAgent::didCreateWorker):
(WebCore::InspectorAgent::didDestroyWorker):
(WebCore::InspectorAgent::evaluateForTestInFrontend):
* inspector/InspectorApplicationCacheAgent.cpp:
(WebCore::InspectorApplicationCacheAgent::updateApplicationCacheStatus):
(WebCore::InspectorApplicationCacheAgent::getFramesWithManifests):
* inspector/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::buildObjectForNode):
* inspector/InspectorMemoryAgent.cpp:
(WebCore::jsHeapInfo):
(WebCore::InspectorMemoryAgent::getProcessMemoryDistribution):
* inspector/PageRuntimeAgent.cpp:
(WebCore::PageRuntimeAgent::notifyContextCreated):
2012-05-28 Kentaro Hara <haraken@chromium.org>
Unreviewed. Rebaselined run-binding-tests results.
* bindings/scripts/test/JS/JSTestActiveDOMObject.cpp:
(WebCore::JSTestActiveDOMObjectOwner::finalize):
* bindings/scripts/test/JS/JSTestCustomNamedGetter.cpp:
(WebCore::JSTestCustomNamedGetterOwner::finalize):
* bindings/scripts/test/JS/JSTestEventConstructor.cpp:
(WebCore::JSTestEventConstructorOwner::finalize):
* bindings/scripts/test/JS/JSTestEventTarget.cpp:
(WebCore::JSTestEventTargetOwner::finalize):
* bindings/scripts/test/JS/JSTestException.cpp:
(WebCore::JSTestExceptionOwner::finalize):
* bindings/scripts/test/JS/JSTestInterface.cpp:
(WebCore::JSTestInterfaceOwner::finalize):
* bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:
(WebCore::JSTestMediaQueryListListenerOwner::finalize):
* bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
(WebCore::JSTestNamedConstructorOwner::finalize):
* bindings/scripts/test/JS/JSTestObj.cpp:
(WebCore::JSTestObjOwner::finalize):
* bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:
(WebCore::JSTestSerializedScriptValueInterfaceOwner::finalize):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::V8TestObj::installPerContextProperties):
2012-05-28 Darin Adler <darin@apple.com>
StyleResolver need not allocate each MediaQueryResult on the heap
https://bugs.webkit.org/show_bug.cgi?id=75223
Reviewed by Daniel Bates.
* css/MediaQuery.cpp: Removed some comments that pointed to CSS documents.
There is no guarantee these links will be valid over time.
(WebCore::MediaQuery::MediaQuery): Rewrote for clarity, conventional WebKit coding
style, and simplicity.
(WebCore::MediaQuery::copy): Moved out of line; not performance critical, and this
allows us to cut down header dependencies.
(WebCore::MediaQuery::cssText): Updated for change to data member name.
* css/MediaQuery.h: Removed unneeded includes. Removed non-helpful argument name
"exprs". Changed expressions function to return a reference instead of
a pointer. Changed mediaType and cssText functions to return a reference. Renamed
m_serializationCache to m_serializedQuery. Moved copy function out of header.
* css/MediaQueryEvaluator.cpp: Renamed EvalFunc to MediaFeatureEvaluationFunction.
Broke a FIXME into three and reworded for clarity.
(WebCore::MediaQueryEvaluator): Updated for name changes.
(WebCore::MediaQueryEvaluator::eval): Rewrote this for clarity and to regularize
the logic a bit.
(WebCore::aspect_ratioMediaFeatureEval): Got rid of a != 0 that is contrary to the
normal WebKit style.
(WebCore::device_aspect_ratioMediaFeatureEval): Ditto.
(WebCore::transform_3dMediaFeatureEval): Fixed mangled #if that was here.
(WebCore::view_modeMediaFeatureEval): Replaced UNUSED_PARAM usage with ASSERT_UNUSED.
(WebCore::createFunctionMap): Changed this so it returns the map so we can use a
cleaner style in the caller.
(WebCore::MediaQueryEvaluator::eval): Updated to take a reference and improved the
comments and coding style a bit.
* css/MediaQueryEvaluator.h: Updated comment style. Removed unused constructor.
Removed unneeded destructor declaration. Renamed m_expResult to m_mediaFeatureResult.
* css/StyleResolver.cpp: Moved the MediaQueryResult class into this file
and made it a structure rather than a class.
(WebCore::StyleResolver::addViewportDependentMediaQueryResult): Updated to take
a reference argument instead of a pointer and for the new vector type.
(WebCore::StyleResolver::affectedByViewportChange): Updated for above changes.
* css/StyleResolver.h: Removed many unneeded includes and forward declarations of
classes, including now-unneeded include of MediaQueryExp.h. Replaced MediaQueryResult
definition with a forward declaration. Changed addViewportDependentMediaQueryResult
to take a reference instead of a pointer. Changed m_viewportDependentMediaQueryResults
to be a vector of values rather than of pointers.
2012-05-27 Shinya Kawanaka <shinyak@chromium.org>
cut, copy or paste event won't be fired in Shadow DOM.
https://bugs.webkit.org/show_bug.cgi?id=87352
Reviewed by Dimitri Glazkov.
cut, copy or paste event was not fired in Shadow DOM.
Since event re-targeting has been implemented, it should be safe to fire them in Shadow DOM now.
Tests: fast/dom/shadow/cppevent-in-shadow.html
fast/dom/shadow/cppevent-input-in-shadow.html
* editing/Editor.cpp:
(WebCore::Editor::findEventTargetFrom):
2012-05-27 Hayato Ito <hayato@chromium.org>
Support multiple shadow roots in event dispatching.
https://bugs.webkit.org/show_bug.cgi?id=87470
Reviewed by Dimitri Glazkov.
The current implementation does not set event's target correctly
if an event happens on DOM tree with multiple shadow roots. So
this patch updates an event re-targeting algorithm so that it
works even if it is applied to multiple shadow roots.
Tests: fast/dom/shadow/shadow-dom-event-dispatching.html
* dom/ComposedShadowTreeWalker.cpp:
(WebCore::ComposedShadowTreeWalker::traverseParentIncludingInsertionPointAndShadowRoot):
* dom/EventDispatcher.cpp:
(WebCore::EventDispatcher::ensureEventAncestors):
2012-05-27 Benjamin Poulain <benjamin@webkit.org>
Minor code cleaning for the interface of MainResourceLoader
https://bugs.webkit.org/show_bug.cgi?id=87607
Reviewed by Darin Adler.
* loader/MainResourceLoader.cpp:
(WebCore::MainResourceLoader::MainResourceLoader):
(WebCore::MainResourceLoader::handleSubstituteDataLoadNow):
(WebCore::MainResourceLoader::handleSubstituteDataLoadSoon):
* loader/MainResourceLoader.h:
(MainResourceLoader):
-Add the OVERRIDE keyword for the methods inherited from ResourceLoader.
-Remove the useless declaration of handleDataLoad(ResourceRequest&), this method does not exist.
-Rename handleDataLoadNow() to handleSubstituteDataLoadNow() and make the method private. This should
only be used by handleSubstituteDataLoadSoon() and indirectly through the timer.
2012-05-27 Yoshifumi Inoue <yosin@chromium.org>
[WTF] Introduce UINT64_C to MathExtras.h
https://bugs.webkit.org/show_bug.cgi?id=87485
Reviewed by Kent Tamura.
This patch introduces UINT64_C for all platforms to avoid using
conditional compilation.
No new tests. This patch doesn't change behavior.
* Modules/websockets/WebSocketFrame.cpp:
(WebCore::WebSocketFrame::parseFrame): Replace conditional compilation with UINT64_C.
2012-05-27 Luke Macpherson <macpherson@chromium.org>
Use StringBuilder in WebKitCSSTransformValue::customCssText() to allow code reuse with CSS Variables.
https://bugs.webkit.org/show_bug.cgi?id=87462
Reviewed by Dimitri Glazkov.
Factor out strings into a const char* array, and use a StringBuilder instead of String concatenation.
This will allow future code to re-use the array of transform names, and StringBuilder is generally faster.
Covered by existing CSS transform tests.
* css/WebKitCSSTransformValue.cpp:
(WebCore):
(WebCore::WebKitCSSTransformValue::customCssText):
2012-05-27 Arvid Nilsson <anilsson@rim.com>
[BlackBerry] Update color for tap highlight and selection
https://bugs.webkit.org/show_bug.cgi?id=87606
Reviewed by Antonio Gomes.
PR #154813
* platform/blackberry/RenderThemeBlackBerry.cpp:
(WebCore::RenderThemeBlackBerry::platformTapHighlightColor):
(WebCore::RenderThemeBlackBerry::platformActiveSelectionBackgroundColor):
2012-05-27 Benjamin Poulain <bpoulain@apple.com>
When pages are loaded from AppCache with DeferredLoading, willSendRequest() is never called
https://bugs.webkit.org/show_bug.cgi?id=87582
Reviewed by Darin Adler.
Previously, there was a shortcut when a deferred MainResourceLoader is resumed: If the data
was coming from AppCache we could jump directly to startDataLoadTimer().
The problem with the shortcut is willSendRequest() is never called in that particular case
(substituteData + deferred-resume). The imbalance between willSendRequest() and didReceiveResponse()
causes problems.
This patch removes the shortcut so that MainResourceLoader::loadNow() is used regardless of
the deferred loading. The method MainResourceLoader::loadNow() handle the substituteData as if the loading
was not deferred.
Test: http/tests/appcache/load-from-appcache-defer-resume-crash.html
* loader/MainResourceLoader.cpp:
(WebCore::MainResourceLoader::continueAfterNavigationPolicy):
(WebCore::MainResourceLoader::handleSubstituteDataLoadSoon):
Rename the method to be consistent with the attribute it uses, making the naming more explicit.
(WebCore::MainResourceLoader::loadNow):
(WebCore::MainResourceLoader::setDefersLoading):
* loader/MainResourceLoader.h:
(MainResourceLoader):
2012-05-27 David Kilzer <ddkilzer@apple.com>
Use xcrun to find gperf path on platforms that use Xcode
<http://webkit.org/b/87587>
Reviewed by Dan Bernstein.
* WebCore.xcodeproj/project.pbxproj:
(Generate Derived Sources): Set GPERF environment variable using
xcrun.
* css/makeprop.pl: Use GPERF environment variable if set, else
"gperf".
* css/makevalues.pl: Ditto.
* make-hash-tools.pl: Ditto.
2012-05-27 Li Yin <li.yin@intel.com>
[FileAPI] FileReader should fire progress event when blob has been completely read into memory
https://bugs.webkit.org/show_bug.cgi?id=87585
Reviewed by Kentaro Hara.
From Spec: http://www.w3.org/TR/FileAPI/#dfn-progress-event
One progress event will fire when blob has been completely read into memory.
Firefox, Opera and IE follows the spec.
Webkit based browser doesn't do that, it only fires progress event at interval of 50ms.
WebKit should add the behavior to make the conformance with the spec.
Tests: fast/files/file-reader-event-listener.html
* fileapi/FileReader.cpp:
(WebCore::FileReader::didFinishLoading):
2012-05-26 Li Yin <li.yin@intel.com>
[FileAPI] The result attribute of FileReader shuold use null to replace empty string
https://bugs.webkit.org/show_bug.cgi?id=87578
Reviewed by Kentaro Hara.
From Spec: http://www.w3.org/TR/FileAPI/#filedata-attr
Before read method has been called or an error in reading has occurred,
the result attribute should be null, not empty string.
Currently, Firefox, Opera and IE 10 follows the spec, but Webkit based
browser don't.
WebKit should change the returned value empty string into null to keep
conformance with the spec.
Tests: fast/files/read-file-async.html
fast/files/blob-slice-test.html
fast/files/read-blob-async.html
fast/files/workers/worker-read-blob-async.html
fast/files/workers/worker-read-file-async.html
* fileapi/FileReader.cpp:
(WebCore::FileReader::stringResult):
2012-05-26 Andy Estes <aestes@apple.com>
Fix the build when NETSCAPE_PLUGIN_API is disabled by marking a
parameter as unused.
* plugins/PluginData.cpp:
(WebCore::PluginData::initPlugins):
2012-05-26 Nate Chapin <japhet@chromium.org>
Cancel CachedResource loads when the last client is removed.
https://bugs.webkit.org/show_bug.cgi?id=35377
Reviewed by Darin Adler.
Test: http/tests/cache/cancel-in-progress-load.html
* loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::errorLoadingResource):
* loader/cache/CachedCSSStyleSheet.cpp:
(WebCore::CachedCSSStyleSheet::allClientsRemoved):
* loader/cache/CachedFont.cpp:
(WebCore::CachedFont::allClientsRemoved):
* loader/cache/CachedImage.cpp:
(WebCore::CachedImage::allClientsRemoved):
* loader/cache/CachedRawResource.cpp:
(WebCore):
* loader/cache/CachedRawResource.h:
(WebCore::CachedRawResource::shouldIgnoreHTTPStatusCodeErrors):
* loader/cache/CachedResource.cpp:
(WebCore::CachedResource::allClientsRemoved):
(WebCore):
* loader/cache/CachedResource.h:
(CachedResource):
* loader/cache/CachedScript.cpp:
(WebCore::CachedScript::allClientsRemoved):
2012-05-26 Simon Fraser <simon.fraser@apple.com>
fast/block/inline-children-root-linebox-crash.html asserts after r118567
https://bugs.webkit.org/show_bug.cgi?id=87544
Reviewed by Darin Adler.
RenderInline::offsetFromContainer() set offsetDependsOnPoint to true based
on the container's flipped writing mode. However, offsetFromContainer() would
then overwrite that, since it only checked for columns.
Fix by having RenderInline::offsetFromContainer() check for flipping on
the container. This fixes the assertion.
The new testcase exercises fixes another issue; unlike mapLocalToAbsolute(),
RenderGeometryMap::absoluteRect() didn't pass the rect center point through
the mapping, which resulted in a different result in some flipping cases.
Test: compositing/geometry/flipped-blocks-inline-mapping.html
* rendering/RenderGeometryMap.cpp:
(WebCore::RenderGeometryMap::absoluteRect):
* rendering/RenderInline.cpp:
(WebCore::RenderInline::offsetFromContainer):
(WebCore::RenderInline::pushMappingToContainer):
2012-05-26 Geoffrey Garen <ggaren@apple.com>
WebKit should be lazy-finalization-safe (esp. the DOM) v2
https://bugs.webkit.org/show_bug.cgi?id=87581
Reviewed by Oliver Hunt.
* bindings/js/JSDOMGlobalObject.cpp:
(WebCore::JSDOMGlobalObject::destroy):
* bindings/js/JSDOMWindowBase.cpp:
(WebCore::JSDOMWindowBase::destroy):
* bindings/js/JSDOMWindowShell.cpp:
(WebCore::JSDOMWindowShell::destroy):
* bindings/js/JSNodeCustom.cpp:
(WebCore::JSNodeOwner::finalize):
* bindings/js/JSWorkerContextBase.cpp:
(WebCore::JSWorkerContextBase::destroy):
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateImplementation):
* bindings/scripts/test/JS/JSTestActiveDOMObject.cpp:
(WebCore::JSTestActiveDOMObject::destroy):
(WebCore::JSTestActiveDOMObjectOwner::finalize):
* bindings/scripts/test/JS/JSTestCustomNamedGetter.cpp:
(WebCore::JSTestCustomNamedGetter::destroy):
(WebCore::JSTestCustomNamedGetterOwner::finalize):
* bindings/scripts/test/JS/JSTestEventConstructor.cpp:
(WebCore::JSTestEventConstructor::destroy):
(WebCore::JSTestEventConstructorOwner::finalize):
* bindings/scripts/test/JS/JSTestEventTarget.cpp:
(WebCore::JSTestEventTarget::destroy):
(WebCore::JSTestEventTargetOwner::finalize):
* bindings/scripts/test/JS/JSTestException.cpp:
(WebCore::JSTestException::destroy):
(WebCore::JSTestExceptionOwner::finalize):
* bindings/scripts/test/JS/JSTestInterface.cpp:
(WebCore::JSTestInterface::destroy):
(WebCore::JSTestInterfaceOwner::finalize):
* bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:
(WebCore::JSTestMediaQueryListListener::destroy):
(WebCore::JSTestMediaQueryListListenerOwner::finalize):
* bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
(WebCore::JSTestNamedConstructor::destroy):
(WebCore::JSTestNamedConstructorOwner::finalize):
* bindings/scripts/test/JS/JSTestObj.cpp:
(WebCore::JSTestObj::destroy):
(WebCore::JSTestObjOwner::finalize):
* bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:
(WebCore::JSTestSerializedScriptValueInterface::destroy):
(WebCore::JSTestSerializedScriptValueInterfaceOwner::finalize):
* bridge/objc/objc_runtime.mm:
(JSC::Bindings::ObjcFallbackObjectImp::destroy):
* bridge/qt/qt_runtime.cpp:
(JSC::Bindings::QtRuntimeMethod::destroy):
* bridge/qt/qt_runtime_qt4.cpp:
(JSC::Bindings::QtRuntimeMethod::destroy):
* bridge/runtime_array.cpp:
(JSC::RuntimeArray::destroy):
* bridge/runtime_method.cpp:
(JSC::RuntimeMethod::destroy):
* bridge/runtime_object.cpp:
(JSC::Bindings::RuntimeObject::destroy):
* bridge/runtime_root.cpp:
(JSC::Bindings::RootObject::finalize): Use static_cast instead of jsCast because
jsCast does Structure-based validation, and our Structure is not guaranteed
to be alive when we get finalized.
2012-05-26 Simon Fraser <simon.fraser@apple.com>
Clip rects assertion when hovering div with transform
https://bugs.webkit.org/show_bug.cgi?id=87580
Reviewed by Eric Seidel.
Hit testing used to use temporary clip rects in composited documents,
until r118562. Now that we cache clip rects for hit testing, we need
to clear the cache on descendant layers when a layer gains or loses
a transform.
Test: fast/layers/clip-rects-assertion.html
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::updateTransform):
2012-05-25 Dan Bernstein <mitz@apple.com>
<rdar://problem/11439771> WebProcess sends many synchronous messages to the UI process while scrolling beneath ScrollView::contentsToScreen()
https://bugs.webkit.org/show_bug.cgi?id=87571
Reviewed by Anders Carlsson.
fakeMouseEventTimerFired() uses the last known mouse position for the fake mouse event, but
calls contentsToScreen() to compute a corresponding position in screen coordinates. Avoid
this by also recording the last known mouse position in screen coordinates, and using that
value.
* page/EventHandler.cpp:
(WebCore::EventHandler::clear): Added resetting m_currentMouseGlobalPosition.
(WebCore::EventHandler::handleMousePressEvent): Added updating m_currentMouseGlobalPosition
when updating m_currentMousePosition.
(WebCore::EventHandler::handleMouseDoubleClickEvent): Ditto.
(WebCore::EventHandler::handleMouseMoveEvent): Ditto.
(WebCore::EventHandler::handleMouseReleaseEvent): Ditto.
(WebCore::EventHandler::fakeMouseMoveEventTimerFired): Changed to use m_currentMouseGlobalPosition
in the fake event instead of calling contentsToScreen().
* page/EventHandler.h: Added m_currentMouseGlobalPosition data member.
2012-05-25 Philippe Normand <pnormand@igalia.com>
[GStreamer] Remove ImageGStreamerCG implementation
https://bugs.webkit.org/show_bug.cgi?id=87559
The ImageGStreamerCG abstraction is being removed until I manage to
port my gst-mac WebKit branch over to the WebKit2 mac port. No
need to update the XCode project because this file is not
referenced there anyway.
Reviewed by Martin Robinson.
* platform/graphics/gstreamer/ImageGStreamer.h:
(ImageGStreamer):
* platform/graphics/gstreamer/ImageGStreamerCG.mm: Removed.
2012-05-26 Rob Buis <rwlbuis@webkit.org>
Bug 15799: textPath element does not re-render when referenced path changes
https://bugs.webkit.org/show_bug.cgi?id=15799
Reviewed by Nikolas Zimmermann.
Support textPath updating to changes on the referenced path. To make this possible
use the target reference functionality also used by SVGFEImageElement.
Tests: svg/custom/textPath-change-id-expected.svg
svg/custom/textPath-change-id-pattern-expected.svg
svg/custom/textPath-change-id-pattern.svg
svg/custom/textPath-change-id.svg
svg/custom/textPath-change-id2-expected.svg
svg/custom/textPath-change-id2-pattern-expected.svg
svg/custom/textPath-change-id2-pattern.svg
svg/custom/textPath-change-id2.svg
svg/custom/textPath-change-reference-expected.svg
svg/custom/textPath-change-reference-pattern-expected.svg
svg/custom/textPath-change-reference-pattern.svg
svg/custom/textPath-change-reference-using-baseval-expected.svg
svg/custom/textPath-change-reference-using-baseval-pattern-expected.svg
svg/custom/textPath-change-reference-using-baseval-pattern.svg
svg/custom/textPath-change-reference-using-baseval.svg
svg/custom/textPath-change-reference.svg
svg/custom/textPath-change-reference2-expected.svg
svg/custom/textPath-change-reference2-pattern-expected.svg
svg/custom/textPath-change-reference2-pattern.svg
svg/custom/textPath-change-reference2-using-baseval-expected.svg
svg/custom/textPath-change-reference2-using-baseval-pattern-expected.svg
svg/custom/textPath-change-reference2-using-baseval-pattern.svg
svg/custom/textPath-change-reference2-using-baseval.svg
svg/custom/textPath-change-reference2.svg
svg/custom/textPath-insert-path-expected.svg
svg/custom/textPath-insert-path-pattern-expected.svg
svg/custom/textPath-insert-path-pattern.svg
svg/custom/textPath-insert-path.svg
svg/custom/textPath-modify-child-expected.svg
svg/custom/textPath-modify-child-pattern-expected.svg
svg/custom/textPath-modify-child-pattern.svg
svg/custom/textPath-modify-child.svg
svg/custom/textPath-path-change-expected.svg
svg/custom/textPath-path-change-pattern-expected.svg
svg/custom/textPath-path-change-pattern.svg
svg/custom/textPath-path-change-using-svg-dom-expected.svg
svg/custom/textPath-path-change-using-svg-dom-pattern-expected.svg
svg/custom/textPath-path-change-using-svg-dom-pattern.svg
svg/custom/textPath-path-change-using-svg-dom.svg
svg/custom/textPath-path-change.svg
svg/custom/textPath-path-change2-expected.svg
svg/custom/textPath-path-change2-pattern-expected.svg
svg/custom/textPath-path-change2-pattern.svg
svg/custom/textPath-path-change2.svg
svg/custom/textPath-remove-path-expected.svg
svg/custom/textPath-remove-path-pattern-expected.svg
svg/custom/textPath-remove-path-pattern.svg
svg/custom/textPath-remove-path.svg
svg/custom/textPath-set-id-expected.svg
svg/custom/textPath-set-id.svg
svg/custom/textPath-startoffset-expected.svg
svg/custom/textPath-startoffset-pattern-expected.svg
svg/custom/textPath-startoffset-pattern.svg
svg/custom/textPath-startoffset.svg
* svg/SVGTextPathElement.cpp:
(WebCore::SVGTextPathElement::~SVGTextPathElement):
(WebCore):
(WebCore::SVGTextPathElement::clearResourceReferences):
(WebCore::SVGTextPathElement::svgAttributeChanged):
(WebCore::SVGTextPathElement::buildPendingResource):
(WebCore::SVGTextPathElement::insertedInto):
(WebCore::SVGTextPathElement::removedFrom):
* svg/SVGTextPathElement.h:
2012-05-26 Nikolas Zimmermann <nzimmermann@rim.com>
Avoid updateFromElement() usage in SVG
https://bugs.webkit.org/show_bug.cgi?id=87573
Stop relying on updateFromElement() - instead rely on addChild/removeChild, which
allows us to optimize the resources re-fetching. When a child is added to the tree
we don't need to remove existing resources from the SVGResourcesCache - the renderer
can't be in the cache yet. Similary, remove the entry from the cache earlier: as soon
as the renderer is removed from the tree, instead of waiting for willBeDestroyed().
No new tests, refactoring only.
* rendering/svg/RenderSVGBlock.cpp:
* rendering/svg/RenderSVGBlock.h:
(RenderSVGBlock):
* rendering/svg/RenderSVGContainer.cpp:
(WebCore::RenderSVGContainer::addChild):
(WebCore):
(WebCore::RenderSVGContainer::removeChild):
* rendering/svg/RenderSVGContainer.h:
(RenderSVGContainer):
* rendering/svg/RenderSVGInline.cpp:
(WebCore::RenderSVGInline::addChild):
(WebCore::RenderSVGInline::removeChild):
* rendering/svg/RenderSVGInline.h:
(RenderSVGInline):
* rendering/svg/RenderSVGModelObject.cpp:
* rendering/svg/RenderSVGModelObject.h:
(RenderSVGModelObject):
* rendering/svg/RenderSVGResourceContainer.cpp:
(WebCore::RenderSVGResourceContainer::registerResource):
* rendering/svg/RenderSVGRoot.cpp:
(WebCore::RenderSVGRoot::addChild):
(WebCore):
(WebCore::RenderSVGRoot::removeChild):
* rendering/svg/RenderSVGRoot.h:
(RenderSVGRoot):
* rendering/svg/RenderSVGText.cpp:
(WebCore::RenderSVGText::addChild):
(WebCore::RenderSVGText::removeChild):
* rendering/svg/SVGResourcesCache.cpp:
(WebCore::SVGResourcesCache::clientStyleChanged):
(WebCore::rendererCanHaveResources):
(WebCore):
(WebCore::SVGResourcesCache::clientWasAddedToTree):
(WebCore::SVGResourcesCache::clientWillBeRemovedFromTree):
* rendering/svg/SVGResourcesCache.h:
(SVGResourcesCache):
* svg/SVGStyledElement.cpp:
* svg/SVGStyledElement.h:
(SVGStyledElement):
2012-05-25 Nat Duca <nduca@chromium.org>
[chromium] Instrument V8 GC with TraceEvent
https://bugs.webkit.org/show_bug.cgi?id=87530
Reviewed by Kentaro Hara.
We sometimes get performance issues where performance stalls can
be attributed to badly timed GC operations, especially ones that
happen just before a frame running. This adds tracing calls around
GC so that we can better understand these kinds of hangs.
* bindings/v8/V8GCController.cpp:
(WebCore::V8GCController::gcPrologue):
(WebCore::V8GCController::gcEpilogue):
2012-05-25 Garrett Casto <gcasto@chromium.org>
Allow WebTextFieldDecoratorClient to see applied decorations.
https://bugs.webkit.org/show_bug.cgi?id=86557
Reviewed by Kent Tamura.
* html/shadow/TextFieldDecorationElement.cpp:
(WebCore::TextFieldDecorationElement::fromShadowRoot): A function
that will extract a TextFielDecorationElement from a ShadowRoot, if
there is one.
* html/shadow/TextFieldDecorationElement.h:
(WebCore):
(TextFieldDecorator):
2012-05-25 Tony Chang <tony@chromium.org>
implement new negative flexing algorithm
https://bugs.webkit.org/show_bug.cgi?id=86528
Reviewed by Ojan Vafai.
Rather than just scale by the negative flexibility, we also take the
flex-basis (preferred size) into consideration. That means items with
a larger preferred size will shrink faster.
Test: css3/flexbox/flex-algorithm.html (new test cases added)
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::preferredMainAxisContentExtentForChild): Handle overflow.
(WebCore::RenderFlexibleBox::layoutFlexItems):
(WebCore::RenderFlexibleBox::computeNextFlexLine): Sum weighted negative flex.
(WebCore::RenderFlexibleBox::freezeViolations):
(WebCore::RenderFlexibleBox::resolveFlexibleLengths): Shrink by weighted amount.
Also handle large values by making sure the flex values are finite.
* rendering/RenderFlexibleBox.h:
2012-05-25 Mihai Parparita <mihaip@chromium.org>
Allow synchronous XHRs to be disabled in documents
https://bugs.webkit.org/show_bug.cgi?id=87540
Reviewed by Eric Seidel.
Test: fast/xmlhttprequest/xmlhttprequest-sync-disabled.html
Synchronous XMLHttpRequests are a problematic API, since they result
in blocked UI threads. Some clients may wish to always disable them;
give them a setting to do so (see also r103629 for other cases where
synchronous XHRs are disabled).
* page/Settings.cpp:
(WebCore):
(WebCore::Settings::Settings):
* page/Settings.h:
(Settings):
(WebCore::Settings::setSyncXHRInDocumentsEnabled):
(WebCore::Settings::syncXHRInDocumentsEnabled):
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::InternalSettings):
(WebCore::InternalSettings::restoreTo):
(WebCore::InternalSettings::setSyncXHRInDocumentsEnabled):
(WebCore):
* testing/InternalSettings.h:
(InternalSettings):
* testing/InternalSettings.idl:
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::open):
2012-05-25 Kinuko Yasuda <kinuko@chromium.org>
[chromium] Deprecate FileUtilities::getFileSize and getFileModifiedTime in favor of getFileMetadata
https://bugs.webkit.org/show_bug.cgi?id=87492
Reviewed by Adam Barth.
No new tests: existing tests (http/tests/local/fileapi/* and fast/files/*) should pass.
* platform/chromium/FileSystemChromium.cpp:
(WebCore::getFileSize):
(WebCore::getFileModificationTime):
(WebCore::getFileMetadata):
* platform/chromium/PlatformSupport.h:
(PlatformSupport):
2012-05-25 Abhishek Arya <inferno@chromium.org>
Crash in RenderTableSection::paintCell.
https://bugs.webkit.org/show_bug.cgi?id=87445
Reviewed by Eric Seidel and Julien Chaffraix.
Fix the crash by preventing table parts from being set
as layout root. This prevents us from accessing removed
table cells which can happen if RenderTableSection::layout
is called directly without calling RenderTable::layout first
(in case of cell recalc).
Add ASSERTs to RenderTableSection::layout to prevent
layout to happen when we are already pending cell recalc
or our table is pending section recalc. In those cases,
RenderTable::layout should be called first to relayout
the entire table.
Test: tables/table-section-overflow-clip-crash.html
* rendering/RenderObject.cpp:
(WebCore::objectIsRelayoutBoundary):
* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::layout):
2012-05-25 Philip Rogers <pdr@google.com>
Fix for self-closing <use> tags
https://bugs.webkit.org/show_bug.cgi?id=87504
Reviewed by Adam Barth.
This change causes self-closing non-html tags to behave the same
as tags immediately followed by the closing tag.
Test: svg/custom/svg-self-closing-use.html
* html/parser/HTMLConstructionSite.cpp:
(WebCore::HTMLConstructionSite::attachLater):
(WebCore::HTMLConstructionSite::insertSelfClosingHTMLElement):
(WebCore::HTMLConstructionSite::insertForeignElement):
* html/parser/HTMLConstructionSite.h:
(HTMLConstructionSite):
2012-05-25 Dan Bernstein <mitz@apple.com>
Make the ICU-based implementation of NonSharedCharacterBreakIterator work in configurations
that do not have COMPARE_AND_SWAP enabled.
Reviewed by Jessie Berlin.
* platform/text/TextBreakIteratorICU.cpp:
(WebCore::compareAndSwapNonSharedCharacterBreakIterator): Added this helper. It uses
weakCompareAndSwap when COMPARE_AND_SWAP is enabled, and uses a mutex to do the atomic
compare and swap otherwise.
(WebCore::NonSharedCharacterBreakIterator::NonSharedCharacterBreakIterator): Changed to use
compareAndSwapNonSharedCharacterBreakIterator().
(WebCore::NonSharedCharacterBreakIterator::~NonSharedCharacterBreakIterator): Ditto.
2012-05-25 Tommy Widenflycht <tommyw@google.com>
MediaStream API: Make sure IceCallback is valid for PeerConnection00
https://bugs.webkit.org/show_bug.cgi?id=87480
Reviewed by Adam Barth.
Existing tests have been extended to cover this change.
* Modules/mediastream/PeerConnection00.cpp:
(WebCore::PeerConnection00::create):
* Modules/mediastream/PeerConnection00.h:
* Modules/mediastream/PeerConnection00.idl:
2012-05-25 Mike West <mkwst@chromium.org>
Inline script and style blocked by Content Security Policy should provide more detailed console errors.
https://bugs.webkit.org/show_bug.cgi?id=86848
Reviewed by Adam Barth.
This change adds a URL and line number for context to each call to
`ContentSecurityPolicy::allowInline*`, and pipes it through to the
console message generation in `CSPDirectiveList::reportViolation`.
Line numbers are not added for injected scripts (`document.write(...)`,
`document.body.appendChild`, and etc.).
Tests: http/tests/security/contentSecurityPolicy/injected-inline-script-allowed.html
http/tests/security/contentSecurityPolicy/injected-inline-script-blocked.html
http/tests/security/contentSecurityPolicy/injected-inline-style-allowed.html
http/tests/security/contentSecurityPolicy/injected-inline-style-blocked.html
* bindings/ScriptControllerBase.cpp:
(WebCore::ScriptController::executeIfJavaScriptURL):
* bindings/js/JSLazyEventListener.cpp:
(WebCore::JSLazyEventListener::initializeJSFunction):
* bindings/v8/V8LazyEventListener.cpp:
(WebCore::V8LazyEventListener::prepareListenerObject):
* dom/ScriptElement.cpp:
(WebCore::ScriptElement::ScriptElement):
(WebCore::ScriptElement::executeScript):
* dom/ScriptElement.h:
(ScriptElement):
* dom/StyleElement.cpp:
(WebCore::StyleElement::StyleElement):
(WebCore::StyleElement::createSheet):
* dom/StyleElement.h:
(StyleElement):
* dom/StyledElement.cpp:
(WebCore::StyledElement::StyledElement):
(WebCore):
(WebCore::StyledElement::style):
(WebCore::StyledElement::styleAttributeChanged):
* dom/StyledElement.h:
(StyledElement):
* page/ContentSecurityPolicy.cpp:
(CSPDirectiveList):
(WebCore::CSPDirectiveList::reportViolation):
(WebCore::CSPDirectiveList::checkInlineAndReportViolation):
(WebCore::CSPDirectiveList::checkEvalAndReportViolation):
(WebCore::CSPDirectiveList::allowJavaScriptURLs):
(WebCore::CSPDirectiveList::allowInlineEventHandlers):
(WebCore::CSPDirectiveList::allowInlineScript):
(WebCore::CSPDirectiveList::allowInlineStyle):
(WebCore::CSPDirectiveList::allowEval):
(WebCore):
(WebCore::isAllowedByAllWithCallStack):
(WebCore::isAllowedByAllWithContext):
(WebCore::ContentSecurityPolicy::allowJavaScriptURLs):
(WebCore::ContentSecurityPolicy::allowInlineEventHandlers):
(WebCore::ContentSecurityPolicy::allowInlineScript):
(WebCore::ContentSecurityPolicy::allowInlineStyle):
* page/ContentSecurityPolicy.h:
(WTF):
2012-05-25 Tim Horton <timothy_horton@apple.com>
ENABLE_CSS3_FLEXBOX is insufficient to disable all web-facing bits of the feature
https://bugs.webkit.org/show_bug.cgi?id=87537
<rdar://problem/11524921>
Reviewed by Simon Fraser.
Allow the feature flag to disable more web-facing parts of the CSS3 flexbox
implementation (primarily fallout from hiding it from computed style).
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
(WebCore::isKeywordPropertyID):
(WebCore::CSSParser::parseValue):
* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
* css/CSSProperty.cpp:
(WebCore::CSSProperty::isInheritedProperty):
* css/CSSPropertyNames.in:
* css/CSSValueKeywords.in:
* css/StyleBuilder.cpp:
(WebCore::StyleBuilder::StyleBuilder):
* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::getPropertyValue):
(WebCore::StylePropertySet::asText):
* css/StylePropertyShorthand.cpp:
(WebCore::shorthandForProperty):
* css/StylePropertyShorthand.h:
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* page/animation/CSSPropertyAnimation.cpp:
(WebCore::CSSPropertyAnimation::ensurePropertyMap):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::createObject):
* rendering/style/RenderStyleConstants.h:
2012-05-25 Adrienne Walker <enne@google.com>
[chromium] Add setting for painting debug info onto tiles
https://bugs.webkit.org/show_bug.cgi?id=75763
Reviewed by James Robinson.
Add a compile-time CCSetting to paint debug information onto tiles. This
can help to understand paint counts and layer indices. This setting is
off by default.
* platform/graphics/chromium/ContentLayerChromium.cpp:
(WebCore::ContentLayerPainter::create):
(WebCore::ContentLayerPainter::paint):
(WebCore::ContentLayerPainter::ContentLayerPainter):
(WebCore::ContentLayerChromium::createTextureUpdater):
* platform/graphics/chromium/TiledLayerChromium.cpp:
(WebCore::UpdatableTile::UpdatableTile):
(WebCore::UpdatableTile::setUpdateFrame):
(WebCore::UpdatableTile::incrementPaintCount):
(WebCore::UpdatableTile::updateFrame):
(WebCore::UpdatableTile::paintCount):
(WebCore::TiledLayerChromium::TiledLayerChromium):
(WebCore::TiledLayerChromium::prepareToUpdateTiles):
(WebCore::TiledLayerChromium::paintDebugTileInfo):
* platform/graphics/chromium/TiledLayerChromium.h:
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(WebCore::CCSettings::CCSettings):
2012-05-25 Ami Fischman <fischman@chromium.org>
[chromium] Default media controls should render only the currentTime-containing buffered range
https://bugs.webkit.org/show_bug.cgi?id=85925
Reviewed by Eric Carlson.
Test: http/tests/media/video-buffered-range-contains-currentTime.html
* rendering/RenderMediaControlsChromium.cpp:
(WebCore::paintMediaSlider):
2012-05-25 Simon Fraser <simon.fraser@apple.com>
Build fix: add TransformationMatrix ctor from an AffineTransform.
* platform/graphics/transforms/TransformationMatrix.cpp:
(WebCore::TransformationMatrix::TransformationMatrix):
(WebCore):
* platform/graphics/transforms/TransformationMatrix.h:
(TransformationMatrix):
2012-05-24 Ryosuke Niwa <rniwa@webkit.org>
createContextualFragment and insertAdjacentHTML should throw syntax error
https://bugs.webkit.org/show_bug.cgi?id=87454
Reviewed by Darin Adler.
Before this patch, createContextualFragment threw NOT_SUPPORTED_ERR and insertAdjacentHTML didn't throw any errors.
Make them throw SYNTAX_ERR to be consistent with the spec and Firefox:
http://html5.org/specs/dom-parsing.html#parsing
http://www.whatwg.org/specs/web-apps/current-work/multipage/the-xhtml-syntax.html#xml-fragment-parsing-algorithm
Also reduced the code duplication.
Test: fast/dom/xhtml-fragment-parsing-exceptions.xhtml
* dom/Range.cpp:
(WebCore::Range::createContextualFragment):
* dom/ShadowRoot.cpp:
(WebCore::ShadowRoot::setInnerHTML): Explicitly pass AllowScriptingContent.
* editing/markup.cpp:
(WebCore::createFragmentFromMarkup):
(WebCore::createFragmentForInnerOuterHTML): Takes ExceptionCode now.
(WebCore::createContextualFragment): Share code with createFragmentForInnerOuterHTML
and propagate the exception code.
* editing/markup.h:
* html/HTMLElement.cpp:
(WebCore::HTMLElement::setInnerHTML): Explicitly pass AllowScriptingContent.
(WebCore::HTMLElement::setOuterHTML): Ditto.
(WebCore::HTMLElement::insertAdjacentHTML): Ditto; also rename ignoredEc to ignoredEC
per Darin's comment on the bug 87339.
2012-05-25 John Knottenbelt <jknotten@chromium.org>
Body scrollWidth() and scrollHeight() should be page scale-invariant
https://bugs.webkit.org/show_bug.cgi?id=87494
RenderView::documentRect() is calculating the "scaled" document rect by applying
the current transformation matrix to the unscaledDocumentRect() and then
returning the rounded-out IntRect result.
This rounding out is incorrect because it allows the scaled rectangle to
represent an area that is not actually covered by the document.
We fix this by applying the current transform to the document rect
as a FloatRect and then explicitly converting to IntRect, which
takes the floor of the resulting rectangle coordinates instead of
rounding them out.
This is evidenced by the document.body.scrollWidth() and
document.body.scrollHeight() changing under page scale factor when
they are expected to remain invariant.
Reviewed by James Robinson.
Test: fast/dom/window-scroll-scaling.html
* rendering/RenderView.cpp:
(WebCore::RenderView::documentRect):
2012-05-25 Dan Bernstein <mitz@apple.com>
characterBreakIterator() is not safe to use reentrantly or from multiple threads
https://bugs.webkit.org/show_bug.cgi?id=87521
Reviewed by Darin Adler.
Replaced characterBreakIterator() with a NonSharedCharacterBreakIterator class, which
obtains a unique TextBreakIterator. Replaced the global shared instance with a single-entry
cache.
* dom/CharacterData.cpp:
(WebCore::CharacterData::parserAppendData): Changed to use NonSharedCharacterBreakIterator.
* platform/graphics/StringTruncator.cpp:
(WebCore::centerTruncateToBuffer): Ditto.
(WebCore::rightTruncateToBuffer): Ditto.
* platform/text/String.cpp:
(WebCore::numGraphemeClusters): Ditto.
(WebCore::numCharactersInGraphemeClusters): Ditto.
* platform/text/TextBreakIterator.h: Removed the declaration of characterBreakIterator().
(NonSharedCharacterBreakIterator): Added. An instance of this class has a character break
iterator instance that is unique to it for the lifetime of the instance.
(WebCore::NonSharedCharacterBreakIterator::operator TextBreakIterator*): Added.
* platform/text/TextBreakIteratorICU.cpp:
(WebCore::NonSharedCharacterBreakIterator::NonSharedCharacterBreakIterator): Added. Tries
to swap the m_iterator member variable with the cached instance. If that fails, initializes
m_iterator to a new character break iterator.
(WebCore::NonSharedCharacterBreakIterator::~NonSharedCharacterBreakIterator): Added. Tries
to put the m_iterator member variable back in the cache. If that fails, meaning there is
already something in the cache, destroys m_iterator.
* platform/text/gtk/TextBreakIteratorGtk.cpp:
(WebCore::NonSharedCharacterBreakIterator::NonSharedCharacterBreakIterator): Same as in
TextBreakIteratorICU.cpp.
(WebCore::NonSharedCharacterBreakIterator::~NonSharedCharacterBreakIterator): Ditto.
(WebCore::cursorMovementIterator): Moved the old implementation of characterBreakIterator()
here.
* platform/text/qt/TextBreakIteratorQt.cpp:
(WebCore::NonSharedCharacterBreakIterator::NonSharedCharacterBreakIterator): Same as in
TextBreakIteratorICU.cpp.
(WebCore::NonSharedCharacterBreakIterator::~NonSharedCharacterBreakIterator): Ditto.
(WebCore::cursorMovementIterator): Moved the old implementation of characterBreakIterator()
here.
* platform/text/wince/TextBreakIteratorWinCE.cpp:
(WebCore::NonSharedCharacterBreakIterator::NonSharedCharacterBreakIterator): Same as in
TextBreakIteratorICU.cpp.
(WebCore::NonSharedCharacterBreakIterator::~NonSharedCharacterBreakIterator): Ditto.
(WebCore::cursorMovementIterator): Moved the old implementation of characterBreakIterator()
here.
2012-05-25 Simon Fraser <simon.fraser@apple.com>
Terrible performance on http://alliances.commandandconquer.com/ and http://www.lordofultima.com/
https://bugs.webkit.org/show_bug.cgi?id=84410
Reviewed by Dave Hyatt.
First part of fixing O(N^2) issues when walking the RenderLayer tree
for computeCompositingRequirements().
For each layer that goes into the OverlapMap, we were computing an absolute
layer bounds, which requires walking back to the root of the tree.
Optimize this when possible by storing a stack of offsets as we walk
the tree, and using this stack to do the mapping.
The stack of offsets and transforms is managed by RenderGeometryMap.
When visiting a RenderLayer, RenderLayerCompositor pushes onto
the geometry map stack data about offsets and transforms between
the current layer and its stacking-parent. RenderGeometryMap handles
the case where the previous renderer pushed is between the current
renderer and its container. RenderGeometryMap can also handle callers
pushing renderers with multiple containers between them.
RenderGeometryMap stores some flags about whether the set of mapping
steps in the stack involve transforms, fixed position, or special non-uniform
mappings like CSS columns. In some cases, it falls back to mapping via
renderers.
Once constructed, the RenderGeometryMap stack can be used to map multiple
rects or points efficiently. Stacks consisting of simple offsets are
collapsed to a single offset.
Mappings between renderers and their containers are pushed by pushMappingToContainer()
methods, which are similar to mapLocalToContainer() methods. Having this code
in RenderObjects was deemed preferable to handling columns, transforms etc. all in
RenderLayer code.
Tested by assertions in RenderGeometryMap code that its mapping matches
mapping via localToAbsolute() calls.
RenderLayerCompositor::updateCompositingLayers() creates a RenderGeometryMap,
and pushes and pops layer renderers as it visits them. The geometry map is used
by RenderLayerCompositor::addToOverlapMap() when computing absolute layer bounds.
Futher optimizations in RenderGeometryMap are possible, especially with stacks that
have many offsets and a few transforms.
Tests: compositing/geometry/composited-in-columns.html
compositing/geometry/flipped-writing-mode.html
* CMakeLists.txt: Add RenderGeometryMap
* GNUmakefile.list.am: Ditt
* Target.pri: Ditto
* WebCore.gypi: Ditto
* WebCore.vcproj/WebCore.vcproj: Ditto
* WebCore.xcodeproj/project.pbxproj: Ditto
* rendering/RenderBox.cpp:
(WebCore::RenderBox::absoluteContentBox):
(WebCore::RenderBox::pushMappingToContainer):
(WebCore::RenderBox::offsetFromContainer):
* rendering/RenderBox.h:
* rendering/RenderGeometryMap.cpp: Added.
(RenderGeometryMapStep):
(WebCore::RenderGeometryMapStep::RenderGeometryMapStep):
(WebCore::RenderGeometryMapStep::mapPoint):
(WebCore::RenderGeometryMapStep::mapQuad):
(WebCore::RenderGeometryMap::RenderGeometryMap):
(WebCore::RenderGeometryMap::~RenderGeometryMap):
(WebCore::RenderGeometryMap::absolutePoint):
(WebCore::RenderGeometryMap::absoluteRect):
(WebCore::RenderGeometryMap::mapToAbsolute):
(WebCore::RenderGeometryMap::pushMappingsToAncestor):
(WebCore::RenderGeometryMap::push):
(WebCore::RenderGeometryMap::pushView):
(WebCore::RenderGeometryMap::popMappingsToAncestor):
(WebCore::RenderGeometryMap::stepInserted):
(WebCore::RenderGeometryMap::stepRemoved):
* rendering/RenderGeometryMap.h: Added.
(RenderGeometryMap):
(WebCore::RenderGeometryMap::hasNonUniformStep):
(WebCore::RenderGeometryMap::hasTransformStep):
(WebCore::RenderGeometryMap::hasFixedPositionStep):
* rendering/RenderInline.cpp:
(WebCore::RenderInline::offsetFromContainer):
(WebCore::RenderInline::pushMappingToContainer):
* rendering/RenderInline.h:
(RenderInline):
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::updateCompositingLayers):
(WebCore::RenderLayerCompositor::addToOverlapMap):
(WebCore::RenderLayerCompositor::addToOverlapMapRecursive):
(WebCore::RenderLayerCompositor::computeCompositingRequirements):
* rendering/RenderLayerCompositor.h:
(RenderLayerCompositor):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::mapLocalToContainer):
(WebCore::RenderObject::pushMappingToContainer):
(WebCore::RenderObject::offsetFromContainer):
(WebCore::RenderObject::container):
* rendering/RenderObject.h:
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::offsetFromContainer):
* rendering/RenderTableCell.h:
(RenderTableCell):
* rendering/RenderView.cpp:
(WebCore::RenderView::pushMappingToContainer):
* rendering/RenderView.h:
* rendering/svg/RenderSVGForeignObject.cpp:
(WebCore::RenderSVGForeignObject::pushMappingToContainer):
* rendering/svg/RenderSVGForeignObject.h:
(RenderSVGForeignObject):
* rendering/svg/RenderSVGInline.cpp:
(WebCore::RenderSVGInline::pushMappingToContainer):
* rendering/svg/RenderSVGInline.h:
(RenderSVGInline):
* rendering/svg/RenderSVGModelObject.cpp:
(WebCore::RenderSVGModelObject::pushMappingToContainer):
* rendering/svg/RenderSVGModelObject.h:
(RenderSVGModelObject):
* rendering/svg/RenderSVGRoot.cpp:
(WebCore::RenderSVGRoot::pushMappingToContainer):
* rendering/svg/RenderSVGRoot.h:
(RenderSVGRoot):
* rendering/svg/RenderSVGText.cpp:
(WebCore::RenderSVGText::pushMappingToContainer):
* rendering/svg/RenderSVGText.h:
(RenderSVGText):
* rendering/svg/SVGRenderSupport.cpp:
(WebCore::SVGRenderSupport::pushMappingToContainer):
* rendering/svg/SVGRenderSupport.h:
(SVGRenderSupport):
2012-05-25 Simon Fraser <simon.fraser@apple.com>
Cache absolute clip rects on RenderLayer for compositing overlap testing
https://bugs.webkit.org/show_bug.cgi?id=87212
Reviewed by Dave Hyatt.
Enhance the cache of ClipRects on RenderLayers to store three
different types of ClipRects, rather than just one.
We need to compute clip rects relative to different layers
for different purposes. For painting, we compute relative to
the compositing layer which is acting as a painting root.
For hit testing, we compute relative to the root, except
for transformed layers. For composting overlap testing, we
compute relative to the root ("absolute"). At other times, we do one-off
computation which we never want to cache ("temporary clip rects").
This change allows us to cache rects for hit testing, and for
compositing overlap testing. This has huge performance benefits
on some pages (bug 84410).
This change also makes ClipRects not arena-allocated, so we
can use RefPtr<ClipRect>.
No testable behavior change.
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::willBeDestroyed): No need for the
explicit clipRects teardown, since clipRects don't need a live
RenderObject for arena-based destruction.
* rendering/RenderLayer.cpp: Remove arena-related new and delete.
(WebCore::RenderLayer::RenderLayer): No need to explicitly initialize m_clipRects,
since it's an OwnPtr now.
(WebCore::RenderLayer::~RenderLayer): No explicit clipRect teardown required.
(WebCore::RenderLayer::clippingRootForPainting): Renamed to make its purpose
more obvious.
(WebCore::RenderLayer::paintLayer): Use the TemporaryClipRects type when necessary.
(WebCore::RenderLayer::paintLayerContents): Ditto
(WebCore::RenderLayer::hitTestLayer): No longer need to use temporary clipRects when
hit testing since we cache clip rects for hit testing.
(WebCore::RenderLayer::updateClipRects): Take a ClipRectsType and pass it through.
(WebCore::RenderLayer::calculateClipRects): Ditto
(WebCore::RenderLayer::parentClipRects): Ditto
(WebCore::RenderLayer::backgroundClipRect): Ditto
(WebCore::RenderLayer::calculateRects): Take ClipRectsType, which obviates temporaryClipRects.
(WebCore::RenderLayer::childrenClipRect): Use clippingRootForPainting().
(WebCore::RenderLayer::selfClipRect): Ditto
(WebCore::RenderLayer::localClipRect): Ditto
(WebCore::RenderLayer::clearClipRectsIncludingDescendants): Take a type of clip rect to clear
(include all). Allows us to just clear painting clip rects.
(WebCore::RenderLayer::clearClipRects):
* rendering/RenderLayer.h:
(WebCore::ClipRects::create): We don't use RefCounted<> in order to use a bit in
the refCount for a flag. Add create() method.
(WebCore::ClipRects::deref): No longer arena-allocated.
(WebCore::ClipRectsCache::ClipRectsCache): Struct that holds a small
array of the 3 types of clipRects (and, in debug, the layer relative
to which they were computed).
(WebCore::RenderLayer::clipRects):
* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateCompositedBounds): Use AbsoluteClipRects; rootLayer
is always the RenderView's layer here.
(WebCore::RenderLayerBacking::updateGraphicsLayerGeometry): Use TemporaryClipRects.
(WebCore::RenderLayerBacking::setRequiresOwnBackingStore): When this variable changes,
we need to invalidate painting clipRects, since it affects the ancestor relative to which
those rects are computed.
* rendering/RenderLayerBacking.h:
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::updateBacking): When the composited state
of a layer changes, we have to clear all descendant clip rects, since this
can affect the layers relative to which clip rects are computed.
(WebCore::RenderLayerCompositor::addToOverlapMap): Use AbsoluteClipRects.
(WebCore::RenderLayerCompositor::computeCompositingRequirements): No need
to call updateLayerPosition(), since that should have always happened after
layout. That call cleared clip rects, so removing it is very beneficial.
(WebCore::RenderLayerCompositor::clippedByAncestor): Use TemporaryClipRects.
* rendering/RenderTreeAsText.cpp:
(WebCore::writeLayers): Use TemporaryClipRects.
2012-05-25 Dean Jackson <dino@apple.com>
Unreviewed, rolling out r112155.
http://trac.webkit.org/changeset/112155
https://bugs.webkit.org/show_bug.cgi?id=79389
Hitch (due to style recalc?) when starting CSS3 animation
This caused a number of issues, including:
https://bugs.webkit.org/show_bug.cgi?id=87146
https://bugs.webkit.org/show_bug.cgi?id=84194
<rdar://problem/11506629>
<rdar://problem/11267408>
<rdar://problem/11531859>
* dom/Element.cpp:
(WebCore::Element::recalcStyle):
2012-05-25 David Hyatt <hyatt@apple.com>
https://bugs.webkit.org/show_bug.cgi?id=87525
For the new multi-column layout, create a flow thread and make sure the children get put inside it.
Reviewed by Eric Seidel.
* rendering/RenderMultiColumnBlock.cpp:
(WebCore::RenderMultiColumnBlock::RenderMultiColumnBlock):
(WebCore::RenderMultiColumnBlock::addChild):
(WebCore):
* rendering/RenderMultiColumnBlock.h:
(WebCore):
(RenderMultiColumnBlock):
(WebCore::RenderMultiColumnBlock::flowThread):
* rendering/RenderMultiColumnFlowThread.cpp:
(WebCore::RenderMultiColumnFlowThread::~RenderMultiColumnFlowThread):
(WebCore):
* rendering/RenderMultiColumnFlowThread.h:
(RenderMultiColumnFlowThread):
* rendering/RenderMultiColumnSet.h:
* rendering/RenderObject.h:
(RenderObject):
(WebCore::RenderObject::isRenderMultiColumnSet):
2012-05-25 Emil A Eklund <eae@chromium.org>
Change RenderBoxModelObject to compute relativePositionOffset as size
https://bugs.webkit.org/show_bug.cgi?id=87447
Reviewed by Eric Seidel.
Compute relativePositionOffset as size instead of doing one axis at a
time as all call sites uses the size version of the method. This avoids
having to walk the DOM twice to accumulate the offsets.
Also remove the relativePositionOffsetX and Y methods as they are no
longer used.
No new tests, covered by existing tests.
* rendering/RenderBox.cpp:
(WebCore::RenderBox::layoutOverflowRectForPropagation):
* rendering/RenderBoxModelObject.cpp:
(WebCore::accumulateRelativePositionOffsets):
(WebCore::RenderBoxModelObject::relativePositionOffset):
* rendering/RenderBoxModelObject.h:
(RenderBoxModelObject):
2012-05-25 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118395.
http://trac.webkit.org/changeset/118395
https://bugs.webkit.org/show_bug.cgi?id=87526
Breaking sites including GMail and Yahoo mail (Requested by
jsbell on #webkit).
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::collapseMargins):
2012-05-25 Ken Buchanan <kenrb@chromium.org>
Layout root not getting cleared for anonymous renderers geting destroyed
https://bugs.webkit.org/show_bug.cgi?id=84002
Reviewed by Abhishek Arya.
This is a follow-up to r109406, which added a check to clear layout
roots when they point to a renderer that is being destroyed. The
thinking was that layout roots would never be anonymous renderers,
but there are some cases where this is not true (in particular,
generated content containers with overflow clips can be layout roots).
As in r109406, this patch has no layout test. This is because any test
that exercises this behavior is caused by an existing layout bug where
a child is not properly getting layout (or a renderer is getting dirtied
out of order during layout) and will fail multiple ASSERTs:
in particular, ASSERT(!m_layoutRoot->container() || !m_layoutRoot->
container()->needsLayout()) in FrameView::scheduleRelayoutOfSubtree(),
and ASSERT_NOT_REACHED() in RenderObject::clearLayoutRootIfNeeded().
We are preventing those bugs from manifesting as security issues with
this patch.
This also removes an ASSERT from the RenderObject destructor. This is
redundant with the condition in RenderObject::clearLayoutRootIfNeeded()
which is always called in RenderObject::willBeDestroyed(), so the check
is not needed. It had to be removed because it fails when I try to
adjust the ASSERT condition by removing the !node()
check, due to RenderWidget clearing its node() during destruction.
* rendering/RenderObject.cpp:
(WebCore::RenderObject::~RenderObject):
(WebCore::RenderObject::willBeDestroyed):
2012-05-25 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: Hangup when continuously changing a css width value in Inspector
https://bugs.webkit.org/show_bug.cgi?id=85802
Reviewed by Vsevolod Vlasov.
An error in the property whitespace prefix detection algorithm would append the previous line trailing whitespace,
thereby enormously increasing the actual prefix during multiple incremental property changes.
* inspector/InspectorStyleSheet.cpp:
(WebCore::InspectorStyle::newLineAndWhitespaceDelimiters):
2012-05-25 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: Clean up Inspector.json after r118367
https://bugs.webkit.org/show_bug.cgi?id=87499
Reviewed by Yury Semikhatsky.
This cleans up the semantic inconsistencies introduced into type/field names r118367.
No new tests, as this is a refactoring.
* inspector/Inspector.json:
* inspector/InspectorCSSAgent.cpp:
(WebCore::InspectorCSSAgent::asInspectorStyleSheet):
(WebCore::InspectorCSSAgent::viaInspectorStyleSheet):
(WebCore::InspectorCSSAgent::detectOrigin):
* inspector/InspectorCSSAgent.h:
(InspectorCSSAgent):
* inspector/InspectorStyleSheet.cpp:
(WebCore::InspectorStyleSheet::create):
(WebCore::InspectorStyleSheet::InspectorStyleSheet):
(WebCore::InspectorStyleSheet::buildObjectForRule):
(WebCore::InspectorStyleSheet::resourceStyleSheetText):
(WebCore::InspectorStyleSheetForInlineStyle::create):
(WebCore::InspectorStyleSheetForInlineStyle::InspectorStyleSheetForInlineStyle):
* inspector/InspectorStyleSheet.h:
(InspectorStyleSheet):
(WebCore::InspectorStyleSheet::canBind):
(InspectorStyleSheetForInlineStyle):
2012-05-25 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: speed-up HeapSnapshot._bfs method.
https://bugs.webkit.org/show_bug.cgi?id=87502
It had containmentEdges.length call in the loop that forced deoptimization.
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshot.prototype._calculateObjectToWindowDistance):
(WebInspector.HeapSnapshot.prototype._bfs):
2012-05-25 Alexei Filippov <alexeif@chromium.org>
Web Inspector: Speed up edges iteration in heap profiler
https://bugs.webkit.org/show_bug.cgi?id=87286
Add an extra node to nodes array that points to the end of edges array.
It allows to eliminate a check for the last node in iteration code.
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshotNode.prototype._edgeIndexesStart):
(WebInspector.HeapSnapshotNode.prototype._edgeIndexesEnd):
(WebInspector.HeapSnapshotNodeIterator):
(WebInspector.HeapSnapshot.prototype._buildRetainers):
(WebInspector.HeapSnapshot.prototype._bfs):
(WebInspector.HeapSnapshot.prototype._buildAggregates):
(WebInspector.HeapSnapshot.prototype._buildPostOrderIndex):
(WebInspector.HeapSnapshot.prototype._buildDominatorTree):
(WebInspector.HeapSnapshot.prototype._markQueriableHeapObjects):
2012-05-25 Andrey Kosyakov <caseq@chromium.org>
Web Inspector: put paint and compositing timeline events in a new category of their own
https://bugs.webkit.org/show_bug.cgi?id=86852
Reviewed by Pavel Feldman.
- add forth timeline category, "Painting"; make it light-purple;
- assign paint and compositing events to Painting category;
* WebCore.gypi: added timelineBarLightPurple.png;
* inspector/front-end/Images/timelineBarLightPurple.png: Added.
* inspector/front-end/Images/timelineCheckmarks.png: added light-purple icon;
* inspector/front-end/Images/timelineDots.png: ditto.
* inspector/front-end/TimelineOverviewPane.js:
(WebInspector.TimelineCategoryStrips.prototype.update.appendRecord): do not merge bars in same raw if these are from different categories;
(WebInspector.TimelineCategoryStrips.prototype.update):
* inspector/front-end/TimelinePresentationModel.js:
(WebInspector.TimelinePresentationModel.categories):
(WebInspector.TimelinePresentationModel.recordStyle):
* inspector/front-end/WebKit.qrc: added timelineBarLightPurple.png;
* inspector/front-end/timelinePanel.css: added styles for painting category;
(.timeline-category-statusbar-item.timeline-category-painting .timeline-category-checkbox):
(.timeline-category-painting .timeline-graph-bar):
(.popover .timeline-painting):
(.timeline-category-scripting .timeline-tree-icon):
(.timeline-category-rendering .timeline-tree-icon):
(.timeline-category-painting .timeline-tree-icon):
2012-05-25 W. James MacLean <wjmaclean@chromium.org>
[chromium] LayerChromium should recognise existing layer active animations when the layer is added.
https://bugs.webkit.org/show_bug.cgi?id=87166
Reviewed by Adrienne Walker.
Unit test added.
LayerChromium needs to correctly recognize if a newly added layer has an existing
active animation.
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::setLayerTreeHost):
(WebCore::LayerChromium::notifyAnimationFinished):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(CCLayerTreeHost):
2012-05-25 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: HeapSnapshot: introduce performance counter for HeapSnapshotConstructorsDataGrid._aggregatesReceived method.
https://bugs.webkit.org/show_bug.cgi?id=87393
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshotDataGrids.js:
(WebInspector.HeapSnapshotConstructorsDataGrid.prototype._aggregatesReceived):
(WebInspector.HeapSnapshotConstructorsDataGrid.prototype._populateChildren):
2012-05-25 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: drop obsolete WebInspector.Uint32Array and adjust snapshot chunk size for better transfer-snapshot metric.
https://bugs.webkit.org/show_bug.cgi?id=87490
Originally WebInspector.Uint32Array was used for dynamic array
reallocation because we had no information about expected arrays sizes.
Now we have these sizes and allocates array precisely.
Reviewed by Yury Semikhatsky.
* bindings/v8/ScriptHeapSnapshot.cpp:
(WebCore):
* inspector/front-end/HeapSnapshot.js:
* inspector/front-end/HeapSnapshotLoader.js:
(WebInspector.HeapSnapshotLoader.prototype._parseUintArray):
(WebInspector.HeapSnapshotLoader.prototype.pushJSONChunk):
2012-05-25 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: drop obsolete WebInspector.Uint32Array and adjust snapshot chunk size for better transfer-snapshot metric.
https://bugs.webkit.org/show_bug.cgi?id=87490
Originally WebInspector.Uint32Array was used for dynamic array
reallocation because we had no information about expected arrays sizes.
Now we have these sizes and allocates array precisely.
Reviewed by Yury Semikhatsky.
* bindings/v8/ScriptHeapSnapshot.cpp:
(WebCore):
* inspector/front-end/HeapSnapshot.js:
* inspector/front-end/HeapSnapshotLoader.js:
(WebInspector.HeapSnapshotLoader.prototype._parseUintArray):
(WebInspector.HeapSnapshotLoader.prototype.pushJSONChunk):
2012-05-25 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: cmd-[ shortcut navigates page and is fr-keyboard incompatible
https://bugs.webkit.org/show_bug.cgi?id=85312
Reviewed by Vsevolod Vlasov.
Suppress the handling of panel history navigation events if the corresponding keyboard activities produce
the "keypress" event (which is the case on French keyboards, where AltGr+[ is translated into Ctrl+Alt+[ on Windows).
The event is also told to preventDefault() to avoid browser history navigation on Mac while traversing the Inspector panel history.
* inspector/front-end/InspectorView.js:
(WebInspector.InspectorView):
(WebInspector.InspectorView.prototype._keyPress):
(WebInspector.InspectorView.prototype._keyDown):
(WebInspector.InspectorView.prototype._keyDownInternal):
* inspector/front-end/UIUtils.js:
(WebInspector.isWin):
2012-05-25 Yury Semikhatsky <yurys@google.com>
Unreviewed. Fixed closure compiler warnings.
* inspector/front-end/NativeMemorySnapshotView.js:
(WebInspector.NativeMemoryProfileType.prototype.buttonClicked):
(WebInspector.NativeMemoryProfileHeader.prototype.createView):
2012-05-25 Lu Guanqun <guanqun.lu@intel.com>
[GTK] fix compilation warning in GtkInputMethodFilter.cpp
https://bugs.webkit.org/show_bug.cgi?id=87475
Reviewed by Martin Robinson.
* platform/gtk/GtkInputMethodFilter.cpp:
(WebCore::GtkInputMethodFilter::setWidget):
2012-05-25 Zalan Bujtas <zbujtas@gmail.com>
[Qt] Broken controls rendering when transform is applied.
https://bugs.webkit.org/show_bug.cgi?id=87483
Reviewed by Simon Hausmann.
Use only the scaling transform value to determine the size of
the control to be drawn. When other transforms present such as
rotate or skew, ignore them, unless scaling also involved. In
that case, calculate the scaling value out of the transformation.
* ManualTests/qt/control_paiting_with_transforms.html: Added.
* platform/qt/RenderThemeQtMobile.cpp:
(WebCore::painterScale):
(WebCore):
(WebCore::StylePainterMobile::sizeForPainterScale):
2012-05-17 Andrey Kosyakov <caseq@chromium.org>
[chromium] add instrumentation for compositing
https://bugs.webkit.org/show_bug.cgi?id=83928
Reviewed by James Robinson.
- plumb willCommit() and didBeginFrame() from CCSingleThreadProxy and CCThreadProxy to inspector instrumentation;
- note for threaded case, didBeginFrame() is invoked upon unblocking of main thread;
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(CCLayerTreeHostClient):
(WebCore::CCLayerTreeHost::didBeginFrame):
(WebCore::CCLayerTreeHost::willCommit):
* platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
(WebCore::CCSingleThreadProxy::commitAndComposite):
* platform/graphics/chromium/cc/CCThreadProxy.cpp:
(WebCore::CCThreadProxy::beginFrame):
2012-05-25 Taiju Tsuiki <tzik@chromium.org>
Web Inspector: Drop InspectorFileSystemInstrumentation
https://bugs.webkit.org/show_bug.cgi?id=87460
Inspector does not need to track DOMFileSystem object now. So we can
drop InspectorFileSystemInstrumentation.
Reviewed by Vsevolod Vlasov.
* GNUmakefile.list.am:
* Modules/filesystem/DOMFileSystem.cpp:
(WebCore::DOMFileSystem::create):
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* inspector/InspectorFileSystemAgent.cpp:
(WebCore::InspectorFileSystemAgent::enable):
* inspector/InspectorFileSystemAgent.h:
(InspectorFileSystemAgent):
* inspector/InspectorFileSystemInstrumentation.h: Removed.
* inspector/InspectorInstrumentation.cpp:
(WebCore):
* inspector/InspectorInstrumentation.h:
(WebCore):
(InspectorInstrumentation):
2012-05-25 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: HeapProfiler: speed-up _calculateClassesRetainedSize and _buildAggregates.
https://bugs.webkit.org/show_bug.cgi?id=87482
Engine didn't inline node's classIndex method because the switch statement in it wasn't inlineable.
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshotNode.prototype.classIndex):
2012-05-24 Andreas Kling <kling@webkit.org>
Dodge style recalc when id attribute is overwritten with same value.
<http://webkit.org/b/87211>
Reviewed by Eric Seidel.
Don't force style recalc when the id attribute is set to the same value it already had.
~3.5% improvement on Dromaeo's "dom-attr" locally.
* dom/Element.cpp:
(WebCore::Element::attributeChanged):
2012-05-24 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: add profile type for native memory snapshots
https://bugs.webkit.org/show_bug.cgi?id=87400
Reviewed by Vsevolod Vlasov.
Introduced new profile type and view classes for native memory snapshots.
All the stuff is hidden behind an experimental setting.
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* inspector/compile-front-end.py:
* inspector/front-end/NativeMemorySnapshotView.js: Added.
(WebInspector.NativeMemorySnapshotView):
(WebInspector.NativeMemorySnapshotView.prototype.dispose):
(WebInspector.NativeMemorySnapshotView.prototype.get statusBarItems):
(WebInspector.NativeMemorySnapshotView.prototype.get profile):
(WebInspector.NativeMemoryProfileType):
(WebInspector.NativeMemoryProfileType.prototype.get buttonTooltip):
(WebInspector.NativeMemoryProfileType.prototype.buttonClicked):
(WebInspector.NativeMemoryProfileType.prototype.get treeItemTitle):
(WebInspector.NativeMemoryProfileType.prototype.get description):
(WebInspector.NativeMemoryProfileType.prototype.createTemporaryProfile):
(WebInspector.NativeMemoryProfileType.prototype.createProfile):
(WebInspector.NativeMemoryProfileHeader):
(WebInspector.NativeMemoryProfileHeader.prototype.createSidebarTreeElement):
(WebInspector.NativeMemoryProfileHeader.prototype.createView):
* inspector/front-end/ProfilesPanel.js:
* inspector/front-end/Settings.js:
(WebInspector.ExperimentsSettings):
* inspector/front-end/WebKit.qrc:
* inspector/front-end/inspector.html:
2012-05-25 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: make some profiler methods private
https://bugs.webkit.org/show_bug.cgi?id=87479
Reviewed by Vsevolod Vlasov.
- Removed unused methods.
- Renamed private methods so that their names start with underscore.
- ProfileLauncherView methods are called directly from ProfilesPanel instead
of sending events.
* inspector/front-end/CPUProfileView.js:
* inspector/front-end/CSSSelectorProfileView.js:
(WebInspector.CSSSelectorProfileType.prototype.buttonClicked):
(WebInspector.CSSSelectorProfileType.prototype._startRecordingProfile):
(WebInspector.CSSSelectorProfileType.prototype.createTemporaryProfile):
* inspector/front-end/ProfileLauncherView.js:
(WebInspector.ProfileLauncherView.prototype.profileStarted):
(WebInspector.ProfileLauncherView.prototype.profileFinished):
* inspector/front-end/ProfilesPanel.js:
(WebInspector.ProfileType.prototype.createProfile):
(WebInspector.ProfilesPanel.prototype.toggleRecordButton):
(WebInspector.ProfilesPanel.prototype._reset):
(WebInspector.ProfilesPanel.prototype._populateProfiles.populateCallback.var):
(WebInspector.ProfilesPanel.prototype._populateProfiles.populateCallback):
(WebInspector.ProfilesPanel.prototype._populateProfiles):
(WebInspector.ProfilesPanel.prototype.setRecordingProfile):
(WebInspector.ProfilesPanel.prototype.takeHeapSnapshot.done):
(WebInspector.ProfilesPanel.prototype.takeHeapSnapshot):
2012-05-25 Keishi Hattori <keishi@webkit.org>
Bad checkValidity result on recently "enabled" form fields
https://bugs.webkit.org/show_bug.cgi?id=85704
Reviewed by Kent Tamura.
Test: fast/forms/disabled-attr-checkvalidity.html
We were tripping on the assertion m_isValid == validity()->valid()
inside HTMLFormControlElement::isValidFormControlElement.
m_isValid was becoming stale because setNeedsValidityCheck wasn't called
after m_willValidate changed.
* html/HTMLFormControlElement.cpp:
(WebCore::HTMLFormControlElement::willValidate): We used const_cast
because it was too difficult to remove const. We are checking if the
value from willValidate has changed to avoid an infinite loop.
(WebCore::HTMLFormControlElement::setNeedsWillValidateCheck):
2012-05-24 Kinuko Yasuda <kinuko@chromium.org>
[chromium] DataTransferItem.webkitGetAsEntry() shouldn't be exposed without flag yet
https://bugs.webkit.org/show_bug.cgi?id=87457
Reviewed by Kent Tamura.
No new tests, marking some tests SKIP as we stop exposing this by default.
* Modules/filesystem/DataTransferItemFileSystem.idl: Removed webkitGetAsEntry
2012-05-25 Rakesh KN <rakesh.kn@motorola.com>
RadioNodeList does not include a object element
https://bugs.webkit.org/show_bug.cgi?id=87371
Reviewed by Kent Tamura.
As per spec http://www.whatwg.org/specs/web-apps/current-work/multipage/forms.html#category-listed
Object element should also be listed in RadioNodeList object.
Updated existing test.
* html/RadioNodeList.cpp:
(WebCore::RadioNodeList::checkElementMatchesRadioNodeListFilter):
Compares if test element's name/id and form matches that of the RadioNodeList filter.
(WebCore::RadioNodeList::nodeMatches):
Added support for Object element, check if object elements name/id matches the RadioNodeList filter.
* html/RadioNodeList.h:
(RadioNodeList): Ditto
2012-05-25 Jan Keromnes <janx@linux.com>
Web Inspector: Resource object has no methods getContent and setContent
https://bugs.webkit.org/show_bug.cgi?id=87424
This fixes the extension API so that it uses the `new Resource()`
constructor, which adds the missing `getContent` and `setContent`
methods to a `Resource` object.
Reviewed by Vsevolod Vlasov.
The tests were fixed accordingly in:
LayoutTests/inspector/extensions/extensions-resources-expected.txt
* inspector/front-end/ExtensionAPI.js:
(injectedExtensionAPI.Panels.prototype.setOpenResourceHandler.else.callbackWrapper):
(injectedExtensionAPI.Panels.prototype.setOpenResourceHandler):
2012-05-24 Tim Horton <timothy_horton@apple.com>
Add feature defines for web-facing parts of CSS Regions and Exclusions
https://bugs.webkit.org/show_bug.cgi?id=87442
<rdar://problem/10887709>
Reviewed by Dan Bernstein.
* Configurations/FeatureDefines.xcconfig:
* GNUmakefile.am:
* bindings/generic/RuntimeEnabledFeatures.cpp:
* bindings/generic/RuntimeEnabledFeatures.h:
(RuntimeEnabledFeatures):
(WebCore::RuntimeEnabledFeatures::setCSSExclusionsEnabled):
(WebCore::RuntimeEnabledFeatures::cssExclusionsEnabled):
* bindings/js/JSCSSRuleCustom.cpp:
(WebCore::toJS):
* bindings/objc/DOMCSS.mm:
(kitClass):
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSParser.cpp:
(WebCore::isSimpleLengthPropertyID):
(WebCore::isValidKeywordPropertyAndValue):
(WebCore::isKeywordPropertyID):
(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::detectAtToken):
* css/CSSProperty.cpp:
(WebCore::CSSProperty::isInheritedProperty):
* css/CSSPropertyNames.in:
* css/CSSRule.cpp:
(WebCore::CSSRule::cssText):
(WebCore::CSSRule::destroy):
(WebCore::CSSRule::reattach):
* css/CSSRule.h:
(WebCore::CSSRule::isRegionRule):
* css/CSSRule.idl:
* css/StyleBuilder.cpp:
(WebCore::StyleBuilder::StyleBuilder):
* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::getPropertyValue):
(WebCore::StylePropertySet::asText):
* css/StylePropertyShorthand.cpp:
(WebCore::shorthandForProperty):
* css/StylePropertyShorthand.h:
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* css/StyleRule.cpp:
(WebCore::StyleRuleBase::destroy):
(WebCore::StyleRuleBase::copy):
(WebCore::StyleRuleBase::createCSSOMWrapper):
* css/WebKitCSSRegionRule.cpp:
* css/WebKitCSSRegionRule.h:
* css/WebKitCSSRegionRule.idl:
* dom/Document.cpp:
* dom/Document.h:
* dom/Document.idl:
* page/DOMWindow.idl:
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(WebCore::Settings::setCSSRegionsEnabled):
(WebCore::Settings::cssRegionsEnabled):
2012-05-24 Geoffrey Garen <ggaren@apple.com>
WebKit should be lazy-finalization-safe (esp. the DOM)
https://bugs.webkit.org/show_bug.cgi?id=87456
Reviewed by Filip Pizlo.
* bindings/js/DOMWrapperWorld.cpp:
(WebCore::JSStringOwner::finalize):
* bindings/js/JSDOMBinding.cpp:
(WebCore::jsStringSlowCase):
* bindings/js/JSDOMBinding.h:
(WebCore::cacheWrapper):
(WebCore::uncacheWrapper): Use the new idioms.
(WebCore::jsString): Use get instead of find because get is simpler in
the case of entries that are logically null.
(WebCore::domObjectWrapperMapFor): Removed, since it was unused.
* bindings/js/ScriptWrappable.h:
(WebCore::ScriptWrappable::clearWrapper): Use the new idioms.
* bridge/runtime_root.cpp:
(JSC::Bindings::RootObject::invalidate): Check for null while iterating,
since that's possible now.
(JSC::Bindings::RootObject::addRuntimeObject):
(JSC::Bindings::RootObject::removeRuntimeObject):
(JSC::Bindings::RootObject::finalize): Use the new idioms.
* bridge/runtime_root.h:
(RootObject): Clarified the word "need".
2012-05-24 Kent Tamura <tkent@chromium.org>
PAGE_POPUP: window.setValueAndClosePopup should be moved to a
per-context property of DOMWindow.
https://bugs.webkit.org/show_bug.cgi?id=87086
Reviewed by Adam Barth.
- Introduce window.pagePagePopupController property as V8EnabledPerContext.
- Move window.setValueAndClosePopup() to window.pagePopupController.
So, we can remove ad-hoc ScriptController::installFunctionsForPagePopup().
No behavior change.
Test: fast/forms/date/no-page-popup-controller.html
* Resources/calendarPicker.js:
(submitValue): Use window.pagePopupController.
(handleCancel): ditto.
* WebCore.gypi: Add new files.
* bindings/generic/ContextEnabledFeatures.cpp:
(WebCore::ContextEnabledFeatures::pagePopupEnabled): Added.
* bindings/generic/ContextEnabledFeatures.h:
(ContextEnabledFeatures): Added.
* bindings/v8/ScriptController.cpp: Remove installFunctionsForPagePopup().
* bindings/v8/ScriptController.h: ditto.
* loader/FrameLoaderClient.h:
(WebCore::FrameLoaderClient::allowPagePopup): Added.
* page/DOMWindowPagePopup.cpp:
(WebCore::DOMWindowPagePopup::DOMWindowPagePopup):
Creates a PagePopupController object for the specified PagePopupClient.
(WebCore::DOMWindowPagePopup::~DOMWindowPagePopup):
Empty destructor to generate the RefPtr<PagePopupController> destructor.
(WebCore::DOMWindowPagePopup::pagePopupController):
* page/DOMWindowPagePopup.h:
(DOMWindowPagePopup):
- Remove setValueAndClosePopup()
- Add pagePopupController()
- Change the data member from PagePopupClient to PagePopupController.
* page/DOMWindowPagePopup.idl: Supply per-context window.pagePopupController.
* page/PagePopupController.cpp: Added. This object is attached to window.
* page/PagePopupController.h: ditto.
* page/PagePopupController.idl: ditto.
2012-05-21 Kinuko Yasuda <kinuko@chromium.org>
Cleanup: add a file system call which captures the file metadata at once.
https://bugs.webkit.org/show_bug.cgi?id=86995
Reviewed by David Levin.
Current File.slice() (webkitSlice()) implementation calls two separate platform calls,
getFileSize() and getFileModificationTime() [both are defined in platform/FileSystem.h],
to capture the file metadata, but we should have a single file system call to get them at once
for two reasons: 1. save additional system call costs, and 2. atomically obtain the file metadata.
No new tests: existing tests (http/tests/local/fileapi/* and fast/files/*) should pass.
* fileapi/File.cpp:
(WebCore::File::captureSnapshot):
* platform/FileMetadata.h:
* platform/FileSystem.h:
* platform/chromium/FileSystemChromium.cpp:
(WebCore::getFileMetadata): Added.
* platform/chromium/PlatformSupport.h:
(PlatformSupport):
* platform/gtk/FileSystemGtk.cpp:
(WebCore::getFileMetadata): Added.
* platform/posix/FileSystemPOSIX.cpp:
(WebCore::getFileMetadata): Added.
* platform/qt/FileSystemQt.cpp:
(WebCore::getFileMetadata): Added.
* platform/win/FileSystemWin.cpp:
(WebCore::getFileSizeFromFindData):
(WebCore::getFileModificationTimeFromFindData):
(WebCore::getFileSize):
(WebCore::getFileModificationTime):
(WebCore::getFileMetadata): Added.
* platform/wince/FileSystemWinCE.cpp:
(WebCore::getFileSizeFromFileInfo):
(WebCore::getFileModificationTimeFromFileInfo):
(WebCore::getFileSize):
(WebCore::getFileModificationTime):
(WebCore::getFileMetadata): Added.
* platform/wx/FileSystemWx.cpp:
(WebCore::getFileMetadata): Added.
(WebCore):
2012-05-24 Hironori Bono <hbono@chromium.org>
Enable grammar checking on Chromium when we paste text (Take 2)
https://bugs.webkit.org/show_bug.cgi?id=74393
Reviewed by Ryosuke Niwa.
This change enables grammar checking on Chromium and implements a mock grammar
checker to fix a failing test.
Test: editing/spelling/grammar-markers.html
* platform/graphics/skia/GraphicsContextSkia.cpp:
(WebCore::GraphicsContext::drawLineForDocumentMarker): render grammar markers in gray on Windows and Linux or in green on Mac.
2012-05-24 Dominic Mazzoni <dmazzoni@google.com>
Crash in WebCore::AccessibilityTable::isDataTable
https://bugs.webkit.org/show_bug.cgi?id=87409
Reviewed by Abhishek Arya.
Use Node::rendererIsEditable everywhere rather than
Node::isContentEditable because the latter can trigger a layout
and destroy the renderer. New test covers the change to
AccessibilityTable.cpp, changes to AccessibilityRenderObject.cpp
are covered by existing tests.
Test: accessibility/contenteditable-table-check-causes-crash.html
* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::isReadOnly):
(WebCore::AccessibilityRenderObject::contentChanged):
* accessibility/AccessibilityTable.cpp:
(WebCore::AccessibilityTable::isDataTable):
2012-05-24 Yoshifumi Inoue <yosin@chromium.org>
[Forms][TextArea] Too long validation message doesn't count LF as CRLF
https://bugs.webkit.org/show_bug.cgi?id=87458
Reviewed by Kent Tamura.
This patch changes current number of characters in "too long" validation message
parameter to counting newline as 2 characters (CR and LF) as submission data.
No new tests. To have test for this change, we need to change localization
text handling during DRT. We'll try.
* html/HTMLTextAreaElement.cpp:
(WebCore::HTMLTextAreaElement::validationMessage): Use computeLengthForSubmission instead of numGraphmeClusters.
2012-05-24 Hayato Ito <hayato@chromium.org>
Fix crashes caused by a DOMCharacterDataModified event on a text node.
https://bugs.webkit.org/show_bug.cgi?id=86953
Reviewed by Dimitri Glazkov.
TextNode can be released while CharacterData::setData() will dispatch a mutation event.
So protect it.
Mutation event itself should not be dispatched on the test case.
This is being tracked by webkit bug https://bugs.webkit.org/show_bug.cgi?id=87372.
Test: fast/events/dom-character-data-modified-textarea-crash.html
* dom/CharacterData.cpp:
(WebCore::CharacterData::setData):
2012-05-24 Philippe Normand <pnormand@igalia.com>
[GTK] Add --enable-css3-flexbox configure option after r118304.
https://bugs.webkit.org/show_bug.cgi?id=87455
Reviewed by Xan Lopez.
* GNUmakefile.am:
2012-05-24 MORITA Hajime <morrita@google.com>
Scoped stylesheet should be per-document-configurable.
https://bugs.webkit.org/show_bug.cgi?id=86985
Reviewed by Kent Tamura.
This change replaced RuntimeEnabledFeatures::styleScopedEnabled() callsites
with newly introduced ContextEnabledFeatures::styleScopedEnabled().
Clients can override the decision by implementing FrameLoaderClient::allowStyleScoped().
No new tests. This isn't testable on DRT.
* bindings/generic/ContextEnabledFeatures.cpp:
(WebCore):
(WebCore::ContextEnabledFeatures::styleScopedEnabled):
* bindings/generic/ContextEnabledFeatures.h:
(WebCore):
(ContextEnabledFeatures):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::determineScope):
* css/StyleResolver.h:
(StyleResolver):
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::registerWithScopingNode):
(WebCore::HTMLStyleElement::unregisterWithScopingNode):
* loader/FrameLoaderClient.h:
(WebCore::FrameLoaderClient::allowStyleScoped):
2012-05-24 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118452.
http://trac.webkit.org/changeset/118452
https://bugs.webkit.org/show_bug.cgi?id=87446
Causes many tests to assert on Mac, NRWT bails out (Requested
by sundiamonde on #webkit).
* rendering/RenderObject.cpp:
(WebCore::RenderObject::~RenderObject):
(WebCore::RenderObject::clearLayoutRootIfNeeded):
2012-05-24 Kevin Ollivier <kevino@theolliviers.com>
[wx] Unreviewed build fix. Add contextMenuItemVector stub.
* platform/wx/ContextMenuWx.cpp:
(WebCore):
(WebCore::contextMenuItemVector):
2012-05-24 Raymond Toy <rtoy@google.com>
Use 32-byte alignment in AudioArray if using WEBAUDIO_FFMPEG
https://bugs.webkit.org/show_bug.cgi?id=87430
Reviewed by Chris Rogers.
Covered by existing tests.
* platform/audio/AudioArray.h:
(WebCore::AudioArray::allocate):
2012-05-24 Antoine Labour <piman@chromium.org>
[chromium] Add a setForceRenderSurface to WebLayer for test/bench purpose
https://bugs.webkit.org/show_bug.cgi?id=87436
Reviewed by James Robinson.
Tested by CCLayerTreeHostCommonTest.verifyForceRenderSurface
* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::LayerChromium):
(WebCore::LayerChromium::setForceRenderSurface):
(WebCore):
(WebCore::LayerChromium::pushPropertiesTo):
* platform/graphics/chromium/LayerChromium.h:
(WebCore::LayerChromium::forceRenderSurface):
(LayerChromium):
* platform/graphics/chromium/cc/CCLayerImpl.cpp:
(WebCore::CCLayerImpl::CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerImpl.h:
(WebCore::CCLayerImpl::forceRenderSurface):
(WebCore::CCLayerImpl::setForceRenderSurface):
(CCLayerImpl):
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::subtreeShouldRenderToSeparateSurface):
2012-05-24 Ken Buchanan <kenrb@chromium.org>
Layout root not getting cleared for anonymous renderers geting destroyed
https://bugs.webkit.org/show_bug.cgi?id=84002
Reviewed by Abhishek Arya.
This is a follow-up to r109406, which added a check to clear layout
roots when they point to a renderer that is being destroyed. The
thinking was that layout roots would never be anonymous renderers,
but there are some cases where this is not true (in particular,
generated content containers with overflow clips can be layout roots).
As in r109406, this patch has no layout test. This is because any test
that exercises this behavior is caused by an existing layout bug where
a child is not properly getting layout (or a renderer is getting dirtied
out of order during layout) and will fail multiple ASSERTs:
in particular, ASSERT(!m_layoutRoot->container() || !m_layoutRoot->
container()->needsLayout()) in FrameView::scheduleRelayoutOfSubtree(),
and ASSERT_NOT_REACHED() in RenderObject::clearLayoutRootIfNeeded().
We are preventing those bugs from manifesting as security issues with
this patch.
* rendering/RenderObject.cpp:
(WebCore::RenderObject::~RenderObject):
(WebCore::RenderObject::willBeDestroyed):
2012-05-24 Anders Carlsson <andersca@apple.com>
Corrupted pages rendering when images are zoomed on Google+
https://bugs.webkit.org/show_bug.cgi?id=87439
<rdar://problem/11503078>
Reviewed by Beth Dakin.
The rect that's given to scrollContentsSlowPath is in frame view coordinates, but if we end up
passing them to RenderLayer::setBackingNeedsRepaintInRect we need to account for the frame scale factor.
* page/FrameView.cpp:
(WebCore::FrameView::scrollContentsSlowPath):
2012-05-24 Ryosuke Niwa <rniwa@webkit.org>
REGRESSION (r112399): insertHTML doesn't respect current selection range and inserts HTML to incorrect position
https://bugs.webkit.org/show_bug.cgi?id=87195
Reviewed by Darin Adler.
The bug was caused by our passing insertionPos.anchorNode() to splitTreeToNode's start node even when
the position's type was an offset in a container. Fixed the bug by passing the node after the insert position
or the container node if the position is at the end of the container.
Test: editing/pasteboard/paste-at-end-of-node-followed-by-inline-element.html
* editing/ReplaceSelectionCommand.cpp:
(WebCore::ReplaceSelectionCommand::doApply):
2012-05-24 Emil A Eklund <eae@chromium.org>
REGRESSION (115573): Incorrect rounding of margins for floats
https://bugs.webkit.org/show_bug.cgi?id=87319
Reviewed by Eric Seidel.
In RenderBlock::computeInlinePreferredLogicalWidths we used a float to
accumulate margins for floating children while the children themselves
represent their margins as LayoutUnits. Due to lack of rounding this can
cause the block to be too small at certain certain zoom levels, causing
unwanted wrapping.
This patch changes computeInlinePreferredLogicalWidths to use a
LayoutUnit to accumulate the margins and thus ensures that the margin
values are rounded the same way.
Test: fast/block/float/floats-with-margin-should-not-wrap.html
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::computeInlinePreferredLogicalWidths):
2012-05-24 Christophe Dumez <christophe.dumez@intel.com>
postMessage and webkitPostMessage should behave the same way
https://bugs.webkit.org/show_bug.cgi?id=87384
Reviewed by Adam Barth.
Make postMessage behave the same way as webkitPostMessage, meaning
that it supports transfer of MessagePorts and ArrayBuffers as per
the spec. Both V8 and JSC implementations have been updated.
Test: fast/dom/Window/window-postmessage-args.html
* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::handlePostMessage):
(WebCore::JSDOMWindow::postMessage):
(WebCore::JSDOMWindow::webkitPostMessage):
* bindings/v8/custom/V8DOMWindowCustom.cpp:
(WebCore::handlePostMessageCallback):
(WebCore::V8DOMWindow::postMessageCallback):
(WebCore::V8DOMWindow::webkitPostMessageCallback):
* bindings/v8/custom/V8DedicatedWorkerContextCustom.cpp:
(WebCore::handlePostMessageCallback):
(WebCore::V8DedicatedWorkerContext::postMessageCallback):
(WebCore::V8DedicatedWorkerContext::webkitPostMessageCallback):
* bindings/v8/custom/V8MessagePortCustom.cpp:
(WebCore::handlePostMessageCallback):
(WebCore::V8MessagePort::postMessageCallback):
(WebCore::V8MessagePort::webkitPostMessageCallback):
* bindings/v8/custom/V8WorkerCustom.cpp:
(WebCore::handlePostMessageCallback):
(WebCore::V8Worker::postMessageCallback):
(WebCore::V8Worker::webkitPostMessageCallback):
2012-05-24 Pablo Flouret <pablof@motorola.com>
Submit button doesn't submit the form if the form is wrapped by an anchor tag
https://bugs.webkit.org/show_bug.cgi?id=86719
Reviewed by Ryosuke Niwa.
When a form's button is clicked or activated with the keyboard a
DOMActivate event is dispatched internally and the default handler for
it takes care of processing the form submission, but the underlying
event that prompted it is not set as handled and so it ends up
navigating the anchor, thereby cancelling the form submission.
This patch sets the original click event as handled if the DOMActivate
event was handled. This matches the rest of the browsers for form
controls that submit a form (input type=submit, button type=submit,
input type=image, etc), and matches IE for the rest of the controls
(basically, IE never activates the anchor when clicking on form
controls, Presto and Gecko mostly don't either, except in a few cases.
Test: fast/forms/form-in-anchor-controls-activation.html
* dom/Node.cpp:
(WebCore::Node::dispatchDOMActivateEvent):
(WebCore::Node::defaultEventHandler):
* dom/Node.h:
(Node):
* html/HTMLButtonElement.cpp:
(WebCore::HTMLButtonElement::defaultEventHandler):
2012-05-24 Crystal Zhang <haizhang@rim.com>
[BlackBerry] Implement select popup and remove old hook to air popup
https://bugs.webkit.org/show_bug.cgi?id=87419
Reviewed by Rob Buis.
Add new files to make file, add css file for select popup.
* PlatformBlackBerry.cmake:
* Resources/blackberry/popupControlBlackBerry.css: Added.
(html):
(body):
(.bottombuttonOK):
(.bottombuttonCancel):
(.tablebutton):
2012-05-24 Levi Weintraub <leviw@chromium.org>
Avoid creating InlineBoxes for floating and positioned objects in isolates.
https://bugs.webkit.org/show_bug.cgi?id=87277
Reviewed by Eric Seidel.
We currently will create a placeholder run for the first object we encounter inside an isolate. Then
in RenderBlockLineLayout's constructBidiRuns, we replace that run with the contents of the Isolate.
We run into problems when there are no valid contents in the Isolate. We can't simply remove the
placeholder if there's nothing to replace it with since it may be the logically last run, which we
track but can't rebuild by the time we're handling isolates (we've already shuffled the BidiRuns around).
With this change, we avoid creating a placeholder altogether until we hit contents in the isolate
that would warrant a BidiRun in the first place.
Test: fast/text/international/float-as-only-child-of-isolate-crash.html
* rendering/InlineIterator.h:
(WebCore::IsolateTracker::addFakeRunIfNecessary):
* rendering/RenderBlock.h:
(RenderBlock):
(WebCore::RenderBlock::shouldSkipCreatingRunsForObject):
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::appendRunsForObject):
2012-05-24 Ryosuke Niwa <rniwa@webkit.org>
There are too many poorly named functions to create a fragment from markup
https://bugs.webkit.org/show_bug.cgi?id=87339
Reviewed by Eric Seidel.
Moved all functions that create a fragment from markup to markup.h/cpp.
There should be no behavioral change.
* dom/Range.cpp:
(WebCore::Range::createContextualFragment):
* dom/Range.h: Removed createDocumentFragmentForElement.
* dom/ShadowRoot.cpp:
(WebCore::ShadowRoot::setInnerHTML):
* editing/markup.cpp:
(WebCore::createFragmentFromMarkup):
(WebCore::createFragmentForInnerOuterHTML): Renamed from createFragmentFromSource.
(WebCore::createFragmentForTransformToFragment): Moved from XSLTProcessor.
(WebCore::removeElementPreservingChildren): Moved from Range.
(WebCore::createContextualFragment): Ditto.
* editing/markup.h:
* html/HTMLElement.cpp:
(WebCore::HTMLElement::setInnerHTML):
(WebCore::HTMLElement::setOuterHTML):
(WebCore::HTMLElement::insertAdjacentHTML):
* inspector/DOMPatchSupport.cpp:
(WebCore::DOMPatchSupport::patchNode): Added a FIXME since this code should be using
one of the functions listed in markup.h
* xml/XSLTProcessor.cpp:
(WebCore::XSLTProcessor::transformToFragment):
2012-05-24 Jer Noble <jer.noble@apple.com>
MediaControlTimelineElement is adjusting time 3 times per click
https://bugs.webkit.org/show_bug.cgi?id=58160
Reviewed by Eric Carlson.
No new tests; we intentionally throttle timeupdate events for the same
movie time, so there is no way to write a layout test for this case.
Only call setCurrentTime() on mousedown or mousemove events.
* html/shadow/MediaControlElements.cpp:
(WebCore::MediaControlTimelineElement::defaultEventHandler):
2012-05-24 John Mellor <johnme@chromium.org>
Font Boosting: Add compile flag and runtime setting
https://bugs.webkit.org/show_bug.cgi?id=87394
Reviewed by Adam Barth.
Add ENABLE_FONT_BOOSTING compile flag and fontBoostingEnabled runtime setting.
No functionality yet, so no new tests.
* Configurations/FeatureDefines.xcconfig:
* GNUmakefile.am:
* Target.pri:
* page/Settings.cpp:
(WebCore::Settings::Settings):
(WebCore::Settings::setFontBoostingEnabled):
(WebCore):
* page/Settings.h:
(Settings):
(WebCore::Settings::fontBoostingEnabled):
2012-05-24 Greg Spencer <gspencer@chromium.org>
MHTML files should be loadable from all schemes considered local,
not just "file:"
https://bugs.webkit.org/show_bug.cgi?id=86540
Reviewed by Adam Barth.
Existing tests should verify correct function.
* loader/MainResourceLoader.cpp:
(WebCore::MainResourceLoader::continueAfterContentPolicy):
* loader/archive/mhtml/MHTMLArchive.cpp:
(WebCore::MHTMLArchive::create):
2012-05-24 Dana Jansens <danakj@chromium.org>
[chromium] Remove some leftover references to LayerTilerChromium
https://bugs.webkit.org/show_bug.cgi?id=87405
Reviewed by James Robinson.
* platform/graphics/chromium/LayerChromium.h:
(LayerChromium):
* platform/graphics/chromium/SolidColorLayerChromium.h:
2012-05-24 Jessie Berlin <jberlin@apple.com>
REGRESSION(r109663) All the the dom/html/level2/html/HTMLFrameElement* tests crash on Windows
https://bugs.webkit.org/show_bug.cgi?id=87410
Reviewed by Anders Carlsson.
Do not pass a reference type to va_start (see r75435).
* platform/LocalizedStrings.cpp:
(WebCore::formatLocalizedString):
2012-05-24 Yael Aharon <yael.aharon@nokia.com>
[Qt] Stop using the flag FIXED_POSITION_CREATES_STACKING_CONTEXT
https://bugs.webkit.org/show_bug.cgi?id=87392
Reviewed by Antonio Gomes.
Remove our dependency on a build flag and use the new setting.
No new tests.
* css/StyleResolver.cpp:
2012-05-24 Alexey Proskuryakov <ap@apple.com>
[WK2] Let the client give local files universal access on a case by case basis
https://bugs.webkit.org/show_bug.cgi?id=87174
<rdar://problem/11024330>
Reviewed by Maciej Stachowiak.
* dom/Document.cpp: (WebCore::Document::initSecurityContext): When settings->allowUniversalAccessFromFileURLs()
is false, also try asking the client for an indulgence.
* loader/FrameLoaderClient.h: (WebCore::FrameLoaderClient::shouldForceUniversalAccessFromLocalURL):
Default implementation doesn't change anything.
2012-05-24 Tony Chang <tony@chromium.org>
improve StyleRareNonInheritedData bit packing on Windows
https://bugs.webkit.org/show_bug.cgi?id=87322
Reviewed by Eric Seidel.
Accessors for m_runningAcceleratedAnimation and m_hasAspectRatio are on RenderStyle already.
Also reorder the variables in operator== to be consistent with the constructor and header file.
This makes it easier to add or remove values.
No new tests, just refactoring.
* rendering/style/StyleRareNonInheritedData.cpp:
(WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
(WebCore::StyleRareNonInheritedData::operator==):
* rendering/style/StyleRareNonInheritedData.h:
(StyleRareNonInheritedData):
2012-05-24 Robert Hogan <robert@webkit.org>
Negative margin block doesn't properly clear a float enclosed by a previous sibling
https://bugs.webkit.org/show_bug.cgi?id=10900
Reviewed by David Hyatt.
Tests: fast/css/clear-float-sibling.html
Parent blocks keep a list of child floats that extend out of the parent block and
by implication overhang into the parent's siblings. But this doesn't work if the
sibling has collapsing margins - it will not find the float in the previous block's
list so will ignore the float and fail to clear it.
RenderBlock:collapseMargins() needs to check if a child's collapsing margin has
reduced the height of the parent up past the bottom of its previous sibling's lowest float
and add the now overhanging float to the parent's float list if appropriate.
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::collapseMargins):
2012-05-24 Kinuko Yasuda <kinuko@chromium.org>
Cleanup: introduce toFile() to reduce static cast from Blob to File
https://bugs.webkit.org/show_bug.cgi?id=87234
Reviewed by Eric Seidel.
No new tests as this must have no side effect.
* bindings/v8/custom/V8BlobCustom.cpp:
(WebCore::toV8):
* fileapi/Blob.cpp:
(WebCore::Blob::webkitSlice):
* fileapi/Blob.h:
(Blob):
* fileapi/File.h:
(WebCore::toFile): Added.
(WebCore):
* fileapi/FileReader.cpp:
(WebCore::FileReader::readAsArrayBuffer):
(WebCore::FileReader::readAsBinaryString):
(WebCore::FileReader::readAsText):
(WebCore::FileReader::readAsDataURL):
* fileapi/WebKitBlobBuilder.cpp:
(WebCore::WebKitBlobBuilder::append):
* platform/chromium/ClipboardChromium.cpp:
(WebCore::ClipboardChromium::files):
* platform/network/FormData.cpp:
(WebCore::FormData::appendKeyValuePairItems):
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::send):
2012-05-24 Darin Adler <darin@apple.com>
SVGElement::addEventListener has peculiar RefPtr usage
https://bugs.webkit.org/show_bug.cgi?id=86497
Reviewed by Andy Estes.
* svg/SVGElement.cpp:
(WebCore::SVGElement::addEventListener): Switch to a more-standard style of RefPtr usage,
getting rid of some unneeded reference count churn; also removed an unneeded special case
for zero listeners.
2012-05-24 Raphael Kubo da Costa <rakuco@webkit.org>
[EFL] Modify keycode conversion functions to return keycodes with location information after r118001.
https://bugs.webkit.org/show_bug.cgi?id=87203
Reviewed by Andreas Kling.
Add the required changes to make
fast/events/keydown-leftright-keys.html pass after r118001.
* platform/efl/EflKeyboardUtilities.cpp:
(WebCore::createWindowsKeyMap): Translate the keycodes for
"{left,right}{Shift,Alt,Control}" into the right windows keyboard
definitions.
2012-05-24 Claudio Saavedra <csaavedra@igalia.com>
[Gtk] Wrong cursor used for ne-resize
https://bugs.webkit.org/show_bug.cgi?id=87366
Reviewed by Eric Seidel.
* platform/gtk/CursorGtk.cpp:
(WebCore::Cursor::ensurePlatformCursor): Use GDK_TOP_RIGHT_CORNER
for Cursor::NorthEastPanning.
2012-05-24 Gabor Ballabas <gaborb@inf.u-szeged.hu>
[Qt] Fix Webkit1 + V8 build.
https://bugs.webkit.org/show_bug.cgi?id=87368
Reviewed by Eric Seidel.
No new tests, because this is a buildfix.
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateImplementation):
* bindings/v8/npruntime_impl.h:
* bindings/v8/npruntime_priv.h:
2012-05-24 Dana Jansens <danakj@chromium.org>
[chromium] Only display frames created with memory allocations meant to be displayed
https://bugs.webkit.org/show_bug.cgi?id=85108
Reviewed by Adrienne Walker.
In this patch we remove the setVisible() code paths from CCProxy, and
instead commit visiblity state along with a frame. We also commit a
flag with a frame that indicates if the frame is one that can be drawn.
The impl host is set to visible during commit instead of using a
special channel through the proxy, and the scheduler allows
commits when we are waiting to draw our first frame but blocked on
being non-visible.
canDraw is gated on a new flag that indicates if the frame is one meant
for display. A frame is meant for display if the frame was generated
with a memory allocation meant for display. At this time, any non-zero
memory allocation is considered meant for display.
We prevent races by not changing the memory allocation at any time
except during a commit. So we force a commit when the memory
allocation needs to be adjusted and we are not visible. Similarly,
we force a commit when visibility changes so that we are able to
commit the visibility change to the impl tree.
In order to prevent drawing frames that are not meant for display
with a single thread, we prevent compositing when the impl tree
is not visible, with an early out in CCSingleThreadProxy.
Unit tests: CCLayerTreeHostTestVisibilityAndAllocationControlDrawing
* platform/graphics/chromium/LayerChromium.h:
* platform/graphics/chromium/TiledLayerChromium.cpp:
* platform/graphics/chromium/TiledLayerChromium.h:
* platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
(WebCore::CCLayerTreeHost::CCLayerTreeHost):
(WebCore::CCLayerTreeHost::initializeLayerRenderer):
(WebCore::CCLayerTreeHost::finishCommitOnImplThread):
(WebCore::CCLayerTreeHost::setNeedsCommit):
(WebCore):
(WebCore::CCLayerTreeHost::setNeedsForcedCommit):
(WebCore::CCLayerTreeHost::setVisible):
(WebCore::CCLayerTreeHost::setContentsMemoryAllocationLimitBytes):
(WebCore::CCLayerTreeHost::scheduleComposite):
(WebCore::CCLayerTreeHost::updateLayers):
* platform/graphics/chromium/cc/CCLayerTreeHost.h:
(CCLayerTreeHost):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
(WebCore::CCLayerTreeHostImpl::CCLayerTreeHostImpl):
(WebCore::CCLayerTreeHostImpl::canDraw):
* platform/graphics/chromium/cc/CCLayerTreeHostImpl.h:
(WebCore::CCLayerTreeHostImpl::sourceFrameCanBeDrawn):
(WebCore::CCLayerTreeHostImpl::setSourceFrameCanBeDrawn):
(CCLayerTreeHostImpl):
* platform/graphics/chromium/cc/CCProxy.h:
(CCProxy):
* platform/graphics/chromium/cc/CCSchedulerStateMachine.cpp:
(WebCore::CCSchedulerStateMachine::drawSuspendedUntilCommit):
(WebCore::CCSchedulerStateMachine::scheduledToDraw):
(WebCore::CCSchedulerStateMachine::updateState):
* platform/graphics/chromium/cc/CCSingleThreadProxy.cpp:
(WebCore::CCSingleThreadProxy::doCommit):
(WebCore::CCSingleThreadProxy::setNeedsCommit):
(WebCore):
(WebCore::CCSingleThreadProxy::setNeedsForcedCommit):
(WebCore::CCSingleThreadProxy::doComposite):
* platform/graphics/chromium/cc/CCSingleThreadProxy.h:
* platform/graphics/chromium/cc/CCThreadProxy.cpp:
(WebCore::CCThreadProxy::CCThreadProxy):
(WebCore::CCThreadProxy::setNeedsForcedCommit):
(WebCore):
(WebCore::CCThreadProxy::setNeedsForcedCommitOnImplThread):
(WebCore::CCThreadProxy::forceBeginFrameOnImplThread):
(WebCore::CCThreadProxy::beginFrame):
(WebCore::CCThreadProxy::scheduledActionCommit):
* platform/graphics/chromium/cc/CCThreadProxy.h:
(CCThreadProxy):
2012-05-24 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: [regression] search in formatted scripts is broken.
https://bugs.webkit.org/show_bug.cgi?id=87377
Reviewed by Vsevolod Vlasov.
Using formatted content in search.
Test: inspector/debugger/script-formatter-search.html
* inspector/front-end/JavaScriptSource.js:
(WebInspector.JavaScriptSource.prototype.workingCopyCommitted):
(WebInspector.JavaScriptSource.prototype.searchInContent.callbackWrapper):
(WebInspector.JavaScriptSource.prototype.searchInContent):
2012-05-24 Philip Rogers <pdr@google.com>
Refactor SVGAnimateTransformElement to avoid expensive determineAnimatedPropertyType call
https://bugs.webkit.org/show_bug.cgi?id=87309
Reviewed by Nikolas Zimmermann.
This is a simple refactor that matches the work done in
https://bugs.webkit.org/show_bug.cgi?id=87309 to avoid an expensive call.
No new tests, no actual effects beyond performance improvement.
* svg/SVGAnimateElement.h:
(SVGAnimateElement):
* svg/SVGAnimateTransformElement.cpp:
(WebCore::SVGAnimateTransformElement::hasValidAttributeType):
2012-05-24 Csaba Osztrogonác <ossy@webkit.org>
[Qt] Unreviewed trvial fixes.
* Target.pri: Typo fix after r118226.
* WebCore.pri: Warning fix after r117291.
2012-05-24 Antti Koivisto <antti@apple.com>
Move StyleRuleImport to a file of its own
https://bugs.webkit.org/show_bug.cgi?id=87386
Rubber-stamped by Andreas Kling.
Move StyleRuleImport out from CSSImportRule.cpp/.h
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* css/CSSImportRule.cpp:
(WebCore):
(WebCore::CSSImportRule::href):
* css/CSSImportRule.h:
(WebCore):
(CSSImportRule):
* css/CSSParser.cpp:
* css/StyleRuleImport.cpp: Copied from Source/WebCore/css/CSSImportRule.cpp.
(WebCore):
* css/StyleRuleImport.h: Copied from Source/WebCore/css/CSSImportRule.h.
(WebCore):
* css/StyleSheetContents.cpp:
2012-05-24 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118352.
http://trac.webkit.org/changeset/118352
https://bugs.webkit.org/show_bug.cgi?id=87390
Caused 6 editing/spelling tests crash/fail on chromium in
debug. (Requested by vsevik on #webkit).
* platform/graphics/skia/GraphicsContextSkia.cpp:
(WebCore::GraphicsContext::drawLineForDocumentMarker):
2012-05-24 Alexander Pavlov <apavlov@chromium.org>
Web Inspector: Support hierarchical context menus
https://bugs.webkit.org/show_bug.cgi?id=86847
Reviewed by Pavel Feldman.
This patch makes use of the WebMenuItemInfo SubMenu type to expose the capability of building submenu items
in the Web Inspector's context menu. ContextMenuItems are also passed/stored by reference/value rather than pointer
in order to be consistent with the PlatformMenuDescription typedef.
* bindings/js/JSInspectorFrontendHostCustom.cpp:
(WebCore::populateContextMenuItems): Enable submenu item population.
(WebCore):
(WebCore::JSInspectorFrontendHost::showContextMenu): Extract the menu population part into populateContextMenuItems().
* bindings/v8/custom/V8InspectorFrontendHostCustom.cpp:
(WebCore::populateContextMenuItems): Enable submenu item population.
(WebCore):
(WebCore::V8InspectorFrontendHost::showContextMenuCallback): Extract the menu population part into populateContextMenuItems().
* inspector/InspectorFrontendHost.cpp:
(WebCore::FrontendMenuProvider::create): Use reference instead of pointer for ContextMenuItems.
(WebCore::FrontendMenuProvider::FrontendMenuProvider): Use reference instead of pointer for ContextMenuItems.
(WebCore::FrontendMenuProvider::populateContextMenu): Use reference instead of pointer for ContextMenuItems.
(WebCore::FrontendMenuProvider::contextMenuCleared):
(FrontendMenuProvider):
(WebCore::InspectorFrontendHost::showContextMenu): Use reference instead of pointer for ContextMenuItems.
* inspector/InspectorFrontendHost.h:
(InspectorFrontendHost):
* inspector/front-end/ContextMenu.js: Support the tree-like structure of context menus.
(WebInspector.ContextMenuItem):
(WebInspector.ContextMenuItem.prototype.id):
(WebInspector.ContextMenuItem.prototype.type):
(WebInspector.ContextMenuItem.prototype._buildDescriptor):
(WebInspector.ContextSubMenuItem):
(WebInspector.ContextSubMenuItem.prototype.appendItem):
(WebInspector.ContextSubMenuItem.prototype.appendSubMenuItem):
(WebInspector.ContextSubMenuItem.prototype.appendCheckboxItem):
(WebInspector.ContextSubMenuItem.prototype.appendSeparator):
(WebInspector.ContextSubMenuItem.prototype._buildDescriptor):
(WebInspector.ContextMenu):
(WebInspector.ContextMenu.prototype.nextId):
(WebInspector.ContextMenu.prototype.show):
(WebInspector.ContextMenu.prototype._setHandler):
(WebInspector.ContextMenu.prototype._buildDescriptor):
* inspector/front-end/SoftContextMenu.js:
(.WebInspector.SoftContextMenu): Support sub-menus.
(.WebInspector.SoftContextMenu.prototype.show):
(.WebInspector.SoftContextMenu.prototype._parentGlassPaneElement):
(.WebInspector.SoftContextMenu.prototype._createMenuItem):
(.WebInspector.SoftContextMenu.prototype._createSubMenu):
(.WebInspector.SoftContextMenu.prototype._createSeparator):
(.WebInspector.SoftContextMenu.prototype._menuItemMouseUp):
(.WebInspector.SoftContextMenu.prototype._focus):
(.WebInspector.SoftContextMenu.prototype._triggerAction):
(.WebInspector.SoftContextMenu.prototype._showSubMenu):
(.WebInspector.SoftContextMenu.prototype._buildMouseEventForSubMenu):
(.WebInspector.SoftContextMenu.prototype._hideSubMenu):
(.WebInspector.SoftContextMenu.prototype._menuItemMouseOut):
(.WebInspector.SoftContextMenu.prototype._highlightMenuItem):
(.WebInspector.SoftContextMenu.prototype._menuKeyDown):
(.WebInspector.SoftContextMenu.prototype._glassPaneMouseUp):
(.WebInspector.SoftContextMenu.prototype._discardMenu):
(.WebInspector.SoftContextMenu.prototype._discardSubMenus):
* inspector/front-end/inspector.css: Support for sub-menus, separator improvement.
(.soft-context-menu-separator):
(.soft-context-menu-separator > .separator-line):
(.soft-context-menu-item-submenu-arrow):
* platform/chromium/ContextMenuChromium.cpp:
(WebCore::contextMenuItemVector): Implemented.
(WebCore):
2012-05-24 Vivek Galatage <vivekgalatage@gmail.com>
Web Inspector: Breakpoints Pane should not show context menu with no breakpoints
https://bugs.webkit.org/show_bug.cgi?id=87340
Reviewed by Pavel Feldman.
Removed the method _contextMenu as this would be redundant call
because _breakpointContextMenu would take care of showing the remove
all breakpoints option.
* inspector/front-end/BreakpointsSidebarPane.js:
(WebInspector.JavaScriptBreakpointsSidebarPane):
(WebInspector.JavaScriptBreakpointsSidebarPane.prototype._breakpointContextMenu):
2012-05-24 Yury Semikhatsky <yurys@chromium.org>
Unreviewed. Chromium Mac build fix after r118357.
Use full name specifier instead of "using" directive.
* inspector/InspectorMemoryAgent.cpp:
(WebCore::jsHeapInfo):
(WebCore::InspectorMemoryAgent::getProcessMemoryDistribution):
2012-05-24 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: introduce virtual resource for inspector stylesheet.
https://bugs.webkit.org/show_bug.cgi?id=87357
Reviewed by Vsevolod Vlasov.
This change introduces virtual resource that represents inspector stylesheet (that is added for styles added via inspector).
New resource has url of form "inspector://<page url folder>/inspector-stylesheet. It enables live editing and revision
history of the inspector stylesheet.
Test: inspector/styles/edit-inspector-stylesheet.html
* inspector/Inspector.json:
* inspector/InspectorCSSAgent.cpp:
(WebCore::InspectorCSSAgent::asInspectorStyleSheet):
(WebCore::InspectorCSSAgent::bindStyleSheet):
(WebCore::InspectorCSSAgent::viaInspectorStyleSheet):
(WebCore::InspectorCSSAgent::detectOrigin):
* inspector/InspectorCSSAgent.h:
(InspectorCSSAgent):
* inspector/InspectorDOMAgent.h:
(InspectorDOMAgent):
(WebCore::InspectorDOMAgent::pageAgent):
* inspector/InspectorStyleSheet.cpp:
(WebCore::InspectorStyleSheet::create):
(WebCore::InspectorStyleSheet::InspectorStyleSheet):
(WebCore::InspectorStyleSheet::buildObjectForStyleSheetInfo):
(WebCore::InspectorStyleSheet::buildObjectForRule):
(WebCore::InspectorStyleSheet::resourceStyleSheetText):
(WebCore::InspectorStyleSheetForInlineStyle::create):
(WebCore::InspectorStyleSheetForInlineStyle::InspectorStyleSheetForInlineStyle):
* inspector/InspectorStyleSheet.h:
(WebCore):
(InspectorStyleSheet):
(WebCore::InspectorStyleSheet::canBind):
(InspectorStyleSheetForInlineStyle):
* inspector/front-end/CSSStyleModel.js:
(WebInspector.CSSStyleModel):
(WebInspector.CSSStyleModel.prototype._undoRedoCompleted):
(WebInspector.CSSStyleModel.prototype.getViaInspectorResourceForRule):
(WebInspector.CSSStyleModelResourceBinding):
(WebInspector.CSSStyleModelResourceBinding.prototype.setContent):
(WebInspector.CSSStyleModelResourceBinding.prototype._inspectedURLChanged):
(WebInspector.CSSStyleModelResourceBinding.prototype._loadStyleSheetHeaders):
(WebInspector.CSSStyleModelResourceBinding.prototype._innerStyleSheetChanged):
(WebInspector.CSSStyleModelResourceBinding.prototype._getViaInspectorResource.hadersLoaded):
(WebInspector.CSSStyleModelResourceBinding.prototype._getViaInspectorResource):
(WebInspector.CSSStyleModelResourceBinding.prototype._viaInspectorResource.overrideRequestContent.callbackWrapper):
(WebInspector.CSSStyleModelResourceBinding.prototype._viaInspectorResource.overrideRequestContent):
(WebInspector.CSSStyleModelResourceBinding.prototype._viaInspectorResource):
(WebInspector.CSSStyleModelResourceBinding.prototype._viaInspectorResourceURL):
* inspector/front-end/Resource.js:
(WebInspector.Resource):
(WebInspector.Resource.prototype.isHidden):
* inspector/front-end/ResourceTreeModel.js:
(WebInspector.ResourceTreeModel.prototype._onRequestUpdateDropped):
(WebInspector.ResourceTreeModel.prototype._addFramesRecursively):
(WebInspector.ResourceTreeFrame.prototype._navigate):
(WebInspector.ResourceTreeFrame.prototype.addResource):
* inspector/front-end/ResourcesPanel.js:
(WebInspector.FrameTreeElement.prototype.appendResource):
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel.prototype._createSourceFrame):
* inspector/front-end/StylesPanel.js:
(WebInspector.InspectorStyleSource):
* inspector/front-end/StylesSidebarPane.js:
(WebInspector.StylePropertiesSection.prototype._createRuleOriginNode.callback):
(WebInspector.StylePropertiesSection.prototype._createRuleOriginNode):
(WebInspector.BlankStylePropertiesSection.prototype.editingSelectorCommitted.successCallback):
(WebInspector.BlankStylePropertiesSection.prototype.editingSelectorCommitted):
2012-05-24 Robin Cao <robin.cao@torchmobile.com.cn>
[BlackBerry] Possible deadlock in the WebGL code path
https://bugs.webkit.org/show_bug.cgi?id=87375
Reviewed by Rob Buis.
Fix a deadlock happening in the WebGL code path. The mutex is locked
but is not unlocked afterword. Also fix a possible double delete,
the mutex m_frontBufferLock will be destroyed in the base class.
Covered by tests in fast/canvas/webgl.
* platform/graphics/blackberry/LayerCompositingThread.cpp:
(WebCore::LayerCompositingThread::releaseTextureResources):
* platform/graphics/blackberry/WebGLLayerWebKitThread.cpp:
(WebCore::WebGLLayerWebKitThread::~WebGLLayerWebKitThread):
2012-05-24 Ian Vollick <vollick@chromium.org>
[chromium] Forcibly sync running animations in the waiting state when synchronized start times are needed.
https://bugs.webkit.org/show_bug.cgi?id=87153
Reviewed by James Robinson.
Unit test: CCLayerAnimationControllerTest.ForceSyncWhenSynchronizedStartTimeNeeded
* platform/graphics/chromium/cc/CCLayerAnimationController.cpp:
(WebCore::CCLayerAnimationController::replaceImplThreadAnimations):
2012-05-24 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: convert HeapSnapshotGridNode._provider into getter.
https://bugs.webkit.org/show_bug.cgi?id=87382
I found that we create a provider for child nodes for the each DataGrid node in advance.
It cost us one async call to the HeapSnapshot's worker per each such the node.
I converted the property into the getter which is initializing the provider lazily.
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshot.js:
* inspector/front-end/HeapSnapshotGridNodes.js:
(WebInspector.HeapSnapshotGridNode):
(WebInspector.HeapSnapshotGridNode.prototype.createProvider):
(WebInspector.HeapSnapshotGridNode.prototype._provider):
(WebInspector.HeapSnapshotGridNode.prototype.dispose):
(WebInspector.HeapSnapshotGridNode.prototype._populate):
(WebInspector.HeapSnapshotGridNode.prototype.expandWithoutPopulate):
(WebInspector.HeapSnapshotGridNode.prototype._populateChildren.serializeNextChunk):
(WebInspector.HeapSnapshotGridNode.prototype.sort):
(WebInspector.HeapSnapshotGenericObjectNode):
(WebInspector.HeapSnapshotGenericObjectNode.prototype.updateHasChildren):
(WebInspector.HeapSnapshotObjectNode):
(WebInspector.HeapSnapshotInstanceNode):
(WebInspector.HeapSnapshotConstructorNode):
(WebInspector.HeapSnapshotConstructorNode.prototype.createProvider):
(WebInspector.HeapSnapshotConstructorNode.prototype.revealNodeBySnapshotObjectId):
(WebInspector.HeapSnapshotDiffNode):
(WebInspector.HeapSnapshotDiffNode.prototype.createProvider):
(WebInspector.HeapSnapshotDominatorObjectNode):
(WebInspector.HeapSnapshotDominatorObjectNode.prototype.createProvider):
(WebInspector.HeapSnapshotDominatorObjectNode.prototype.retrieveChildBySnapshotObjectId):
2012-05-24 Rakesh KN <rakesh.kn@motorola.com>
Filter for RadioNodeList should be case sensitive.
https://bugs.webkit.org/show_bug.cgi?id=87369
Reviewed by Kent Tamura.
Element's id/name attribute matching criteria is case sensitive now.
Modified existing test.
* html/RadioNodeList.cpp:
(WebCore::RadioNodeList::nodeMatches):
Element matching criteria is case sensitive.
2012-05-23 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: add a command to InspectorMemoryAgent for getting process memory break down
https://bugs.webkit.org/show_bug.cgi?id=87263
Reviewed by Pavel Feldman.
Introduced new protocol command Memory.getProcessMemoryDistribution which returns
memory distribution for the inspected process. Currently only JS allocated and used
heap size is included.
* inspector/Inspector.json:
* inspector/InspectorMemoryAgent.cpp:
(WebCore::jsHeapInfo):
(WebCore):
(WebCore::InspectorMemoryAgent::getProcessMemoryDistribution):
* inspector/InspectorMemoryAgent.h:
(InspectorMemoryAgent):
* platform/chromium/PlatformSupport.h:
(PlatformSupport):
2012-05-24 Nikolas Zimmermann <nzimmermann@rim.com>
SVGZoomAndPan constants are missing from window object
https://bugs.webkit.org/show_bug.cgi?id=15494
Reviewed by Rob Buis.
Generate the SVGZoomAndPanConstructor, so that the constants defined in the IDL
can be reached from the bindings. Provide a stub-implementation of ref/deref
that's never used, as we don't actually use JSSVGZoomAndPan which needs this, but
only the JSSVGZoomAndPanConstructor.
Add a new IDL flag "SuppressToJSObject" which disables generation of toJS/toV8
methods for classes that are only used in SVGs interfaces via multiple inheritance.
This affects: SVGFitToViewBox, SVGTests, SVGLangSapce, SVGExternalResourcesRequired, etc.
Unlike those classes SVGZoomAndPan defines constants, and thus needs a generated Constructor.
That requires us to build JSSVGZoomAndPan.* (JSSVGFitToViewBox/etc. is generated, but not built!).
Unfortunately this causes compilation problems on Windows, as it builds all sources in a single-file.
MSVC can't decide whether it should call toJS(Node*) or toJS(SVGZoomAndPan*) for a SVGSVGElement.
To avoid these problems stop generating toJS/toV8 completely for all SVG MI types. They were
never used before, so there's no point in actually generating them, as it's now causing problems.
This is a preparation towards bug 15495, which covers implementing
SVGSVGElement.currentView and the SVGViewSpec interface.
No new tests. The SVGZoomAndPan constructor is covered by existing tests, now that its enabled.
* CMakeLists.txt:
* DerivedSources.cpp:
* DerivedSources.pri:
* GNUmakefile.list.am:
* WebCore.gyp/WebCore.gyp:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/scripts/CodeGeneratorJS.pm:
(ShouldGenerateToJSDeclaration):
(ShouldGenerateToJSImplementation):
(GenerateHeader):
(GenerateImplementation):
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateHeader):
* bindings/scripts/IDLAttributes.txt:
* gyp/WebCore.gyp:
* page/DOMWindow.idl:
* svg/SVGDocument.cpp:
(WebCore::SVGDocument::zoomAndPanEnabled):
* svg/SVGExternalResourcesRequired.idl:
* svg/SVGFitToViewBox.idl:
* svg/SVGLangSpace.idl:
* svg/SVGLocatable.idl:
* svg/SVGRenderingIntent.idl:
* svg/SVGSVGElement.cpp:
(WebCore::SVGSVGElement::SVGSVGElement):
(WebCore::SVGSVGElement::parseAttribute):
(WebCore::SVGSVGElement::viewBoxToViewTransform):
(WebCore::SVGSVGElement::setupInitialView):
(WebCore::SVGSVGElement::inheritViewAttributes):
* svg/SVGSVGElement.h:
(SVGSVGElement):
(WebCore::SVGSVGElement::useCurrentView):
(WebCore::SVGSVGElement::setUseCurrentView):
(WebCore::SVGSVGElement::zoomAndPan):
(WebCore::SVGSVGElement::setZoomAndPan):
* svg/SVGStylable.idl:
* svg/SVGTests.idl:
* svg/SVGURIReference.idl:
* svg/SVGUnitTypes.idl:
* svg/SVGViewElement.cpp:
(WebCore::SVGViewElement::SVGViewElement):
(WebCore::SVGViewElement::parseAttribute):
* svg/SVGViewElement.h:
(SVGViewElement):
(WebCore::SVGViewElement::zoomAndPan):
(WebCore::SVGViewElement::setZoomAndPan):
* svg/SVGViewSpec.cpp:
(WebCore::SVGViewSpec::SVGViewSpec):
(WebCore::SVGViewSpec::setTransformString):
(WebCore::SVGViewSpec::parseViewSpec):
* svg/SVGViewSpec.h:
(SVGViewSpec):
(WebCore::SVGViewSpec::transformBaseValue):
(WebCore::SVGViewSpec::zoomAndPan):
(WebCore::SVGViewSpec::setZoomAndPanBaseValue):
* svg/SVGZoomAndPan.cpp:
(WebCore::SVGZoomAndPan::isKnownAttribute):
(WebCore::SVGZoomAndPan::addSupportedAttributes):
(WebCore):
(WebCore::SVGZoomAndPan::parseZoomAndPan):
(WebCore::SVGZoomAndPan::ref):
(WebCore::SVGZoomAndPan::deref):
(WebCore::SVGZoomAndPan::setZoomAndPan):
* svg/SVGZoomAndPan.h:
(SVGZoomAndPan):
(WebCore::SVGZoomAndPan::parseFromNumber):
(WebCore::SVGZoomAndPan::parseAttribute):
(WebCore::SVGZoomAndPan::zoomAndPan):
* svg/SVGZoomAndPan.idl:
2012-05-24 Hironori Bono <hbono@chromium.org>
Enable grammar checking on Chromium when we paste text.
https://bugs.webkit.org/show_bug.cgi?id=74393
Reviewed by Hajime Morita.
This change enables grammar checking on Chromium and implements a mock grammar
checker to fix a failing test.
Test: editing/spelling/grammar-markers.html
* platform/graphics/skia/GraphicsContextSkia.cpp:
(WebCore::GraphicsContext::drawLineForDocumentMarker): render grammar markers in gray on Windows and Linux or in green on Mac.
2012-05-17 Andrey Kosyakov <caseq@chromium.org>
Web Inspector: add inspector instrumentation interface for compositing
https://bugs.webkit.org/show_bug.cgi?id=83842
Reviewed by Pavel Feldman.
- added instrumentation methods to mark start/end of layer compositing;
- added associated timeline record type (CompositeLayers);
- plumbed the calls from instrumentation up to the timeline panel;
- actual call-sites are platform-specific and will come as separate patches.
* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::willCompositeImpl):
(WebCore):
(WebCore::InspectorInstrumentation::didCompositeImpl):
* inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
(WebCore::InspectorInstrumentation::willComposite):
(WebCore):
(WebCore::InspectorInstrumentation::didComposite):
* inspector/InspectorTimelineAgent.cpp:
(TimelineRecordType):
(WebCore::InspectorTimelineAgent::willComposite):
(WebCore):
(WebCore::InspectorTimelineAgent::didComposite):
* inspector/InspectorTimelineAgent.h:
(InspectorTimelineAgent):
* inspector/front-end/TimelineModel.js:
* inspector/front-end/TimelinePresentationModel.js:
(WebInspector.TimelinePresentationModel.recordStyle):
2012-05-24 Ilya Tikhonovsky <loislo@chromium.org>
Unreviewed Web Inspector: remove obsolete isDetailedSnapshot method from HeapSnapshotView.
* inspector/front-end/HeapSnapshotView.js:
2012-05-24 Vivek Galatage <vivekgalatage@gmail.com>
Web Inspector: localStorage items are not updated when the storage changes
https://bugs.webkit.org/show_bug.cgi?id=83012
Reviewed by Pavel Feldman.
Renamed the inspector protocol UpdateDOMStorage to domStorageUpdated.
Instrumented StorageEventDispatcher to send the update event to the
front-end. Also removed the way DOM Storage Agent used to listen for
the storage events. Added new test to verify the update notifications.
Test: inspector/storage-panel-dom-storage-update.html
* dom/EventListener.h:
* inspector/Inspector.json:
* inspector/InspectorDOMStorageAgent.cpp:
(WebCore):
(WebCore::InspectorDOMStorageAgent::getDOMStorageEntries):
(WebCore::InspectorDOMStorageAgent::storageId):
(WebCore::InspectorDOMStorageAgent::didUseDOMStorage):
(WebCore::InspectorDOMStorageAgent::didDispatchDOMStorageEvent):
* inspector/InspectorDOMStorageAgent.h:
(InspectorDOMStorageAgent):
* inspector/InspectorDOMStorageResource.cpp:
(WebCore::InspectorDOMStorageResource::InspectorDOMStorageResource):
(WebCore::InspectorDOMStorageResource::isSameHostAndType):
(WebCore::InspectorDOMStorageResource::unbind):
* inspector/InspectorDOMStorageResource.h:
(WebCore):
(InspectorDOMStorageResource):
(WebCore::InspectorDOMStorageResource::create):
(WebCore::InspectorDOMStorageResource::id):
(WebCore::InspectorDOMStorageResource::storageArea):
(WebCore::InspectorDOMStorageResource::frame):
* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::didDispatchDOMStorageEventImpl):
(WebCore):
* inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
(WebCore::InspectorInstrumentation::didUseDOMStorage):
(WebCore):
(WebCore::InspectorInstrumentation::didDispatchDOMStorageEvent):
* inspector/front-end/DOMStorage.js:
(WebInspector.DOMStorageDispatcher.prototype.domStorageUpdated):
* inspector/front-end/ResourcesPanel.js:
(WebInspector.ResourcesPanel.prototype.domStorageUpdated):
* storage/StorageEventDispatcher.cpp:
(WebCore::StorageEventDispatcher::dispatch):
2012-05-23 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: introduce console proxy object for HeapSnapshot worker.
https://bugs.webkit.org/show_bug.cgi?id=87346
There is no console object in html5 workers at the moment.
This makes me sad when I'm troubleshooting the HeapSnapshot stuff.
I wrote a simplest proxy object which sends the log, info and error messages
to the WebInspector window.
Reviewed by Pavel Feldman.
* inspector/front-end/HeapSnapshotProxy.js:
(WebInspector.HeapSnapshotRealWorker.prototype._messageReceived):
* inspector/front-end/HeapSnapshotWorker.js:
(WebInspector.WorkerConsole):
(WebInspector.WorkerConsole.prototype.log):
(WebInspector.WorkerConsole.prototype.error):
(WebInspector.WorkerConsole.prototype.info):
(WebInspector.WorkerConsole.prototype._postMessage):
2012-05-24 Shinya Kawanaka <shinyak@chromium.org>
Elements in Shadow DOM are not resizable.
https://bugs.webkit.org/show_bug.cgi?id=87342
Reviewed by Ryosuke Niwa.
In RenderLayer::resize(), shadowAncestorNode() was used but if the ancestor node is used,
it is not possible to resize elements in Shadow DOM.
The comments said it is necessary for textarea, however actually it is not necessary now.
Existing test (fast/css/resize-corner-tracking) covers it.
Test: fast/dom/shadow/resize-in-shadow-dom.html
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::resize):
2012-05-24 Antti Koivisto <antti@apple.com>
Move StyleSheetContents to a separate file
https://bugs.webkit.org/show_bug.cgi?id=87354
Reviewed by Eric Seidel.
Move StyleSheetContents class to StyleSheetContents.h/cpp.
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* css/CSSFontFaceSrcValue.cpp:
* css/CSSGrammar.y:
* css/CSSImportRule.cpp:
* css/CSSParser.cpp:
* css/CSSPrimitiveValue.cpp:
* css/CSSRule.cpp:
* css/CSSStyleSheet.cpp:
(WebCore::CSSStyleSheet::create):
(WebCore::CSSStyleSheet::href):
(WebCore):
(WebCore::CSSStyleSheet::baseURL):
(WebCore::CSSStyleSheet::isLoading):
* css/CSSStyleSheet.h:
(WebCore):
(CSSStyleSheet):
* css/StylePropertySet.cpp:
* css/StyleResolver.cpp:
* css/StyleSheetContents.cpp: Copied from css/CSSStyleSheet.cpp.
(WebCore):
* css/StyleSheetContents.h: Copied from css/CSSStyleSheet.h.
(WebCore):
* dom/DOMImplementation.cpp:
(WebCore::XMLMIMETypeRegExp::XMLMIMETypeRegExp):
* dom/Document.cpp:
* dom/ProcessingInstruction.cpp:
* dom/StyleElement.cpp:
* html/HTMLLinkElement.cpp:
* html/HTMLStyleElement.cpp:
* inspector/InspectorStyleSheet.cpp:
* loader/cache/CachedCSSStyleSheet.cpp:
* page/PageSerializer.cpp:
2012-05-24 Kristóf Kosztyó <kkristof@inf.u-szeged.hu>
[Qt] Buildfix for the minimal build because the r118197 broke it.
https://bugs.webkit.org/show_bug.cgi?id=87096
Reviewed by Csaba Osztrogonác.
* plugins/PluginData.cpp:
(WebCore):
(WebCore::PluginData::refresh):
(WebCore::PluginData::initPlugins):
2012-05-23 Ilya Tikhonovsky <loislo@chromium.org>
Unreviewed: Web Inspector single line fix for r118162.
* inspector/front-end/HeapSnapshotDataGrids.js:
(WebInspector.HeapSnapshotConstructorsDataGrid.prototype._populateChildren):
2012-05-23 Lu Guanqun <guanqun.lu@intel.com>
sort the array 'non_wrapper_types'
https://bugs.webkit.org/show_bug.cgi?id=87335
Reviewed by Kentaro Hara.
* bindings/scripts/CodeGeneratorV8.pm:
2012-05-23 Shinya Kawanaka <shinyak@chromium.org>
Document.elementFromPoint exposes inner element of Shadow DOM.
https://bugs.webkit.org/show_bug.cgi?id=87235
Reviewed by Dimitri Glazkov.
Document::elementFromPoint didn't consider nested Shadow DOM.
The container node should be adjusted up to document TreeScope beyond ShadowRoot TreeScope.
Test: fast/dom/shadow/element-from-point-in-nested-shadow.html
* dom/Document.cpp:
(WebCore::Document::elementFromPoint):
2012-05-23 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118297.
http://trac.webkit.org/changeset/118297
https://bugs.webkit.org/show_bug.cgi?id=87338
It caused several layout failures on Mac (Requested by
sundiamonde on #webkit).
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::computeInlinePreferredLogicalWidths):
2012-05-23 Shinya Kawanaka <shinyak@chromium.org>
Document.caretRangeFromPoint exposes inner elemnet of Shadow DOM.
https://bugs.webkit.org/show_bug.cgi?id=87231
Reviewed by Dimitri Glazkov.
Document::caretRangeFromPoint didn't consider nested Shadow DOM.
The container node should be adjusted up to document TreeScope beyond ShadowRoot TreeScope.
Test: fast/dom/shadow/caret-range-from-point-in-nested-shadow.html
* dom/Document.cpp:
(WebCore::Document::caretRangeFromPoint):
2012-05-23 Julien Chaffraix <jchaffraix@webkit.org>
Crash in RenderTableCol::nextColumn
https://bugs.webkit.org/show_bug.cgi?id=87314
Reviewed by Abhishek Arya.
Tests: fast/table/canvas-column-in-column-group.html
fast/table/columngroup-inside-columngroup.html
The issue comes from elements not abiding by the display property (e.g. canvas). This means
that any renderer with display: table-column would pass the current isChildAllowed check and
would confuse our algorithm to iterate.
We were getting away with allowing those children as table columns or column groups don't
paint themselves but it's better to just not allow such children in the first place.
* rendering/RenderTableCol.cpp:
(WebCore::RenderTableCol::isChildAllowed):
Fixed the logic to only accept proper column renderer (RenderTableCol with display: column
to ignore column-groups). Also removed an unneeded NULL-check.
2012-05-23 Jer Noble <jer.noble@apple.com>
REGRESSION: compositing/video/video-poster.html fails on Mac
https://bugs.webkit.org/show_bug.cgi?id=87199
Reviewed by Maciej Stachowiak.
No new tests; fixes failing compositing/video/video-poster.html test.
Instead of creating the video layer directly, simply allow the layer
to be created in updateStates() by changing the definition of
isReadyForVideoSetup() to bypass the m_isAllowedToRender check if
the player reports a video track is present. This causes the video layer
to be created and for future calls to prepareForRendering() to result
in calls to mediaPlayerRenderingModeChanged().
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
(WebCore::MediaPlayerPrivateAVFoundation::isReadyForVideoSetup):
(WebCore::MediaPlayerPrivateAVFoundation::metadataLoaded):
2012-05-23 Nico Weber <thakis@chromium.org>
[chromium/mac] Unbreak popup menus.
https://bugs.webkit.org/show_bug.cgi?id=87325
Reviewed by David Levin.
http://svn.webkit.org/repository/webkit/trunk@117607 added this function, but the
Chromium/Mac build doesn't use the file added in that change and hence didn't have
the category method.. Adding it doesn't help, as the category calls a
10.7-only function. Just don't use the category for chromium/mac. Fixes
http://crbug.com/129418
* platform/mac/WebCoreNSCellExtras.h:
2012-05-23 Ojan Vafai <ojan@chromium.org>
add back the ability to disable flexbox
https://bugs.webkit.org/show_bug.cgi?id=87147
Reviewed by Tony Chang.
* Configurations/FeatureDefines.xcconfig:
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
2012-05-23 Shinya Kawanaka <shinyak@chromium.org>
[Shadow] mousewheel event isn't fired on nodes in Shadow DOM.
https://bugs.webkit.org/show_bug.cgi?id=87218
Reviewed by Dimitri Glazkov.
EventHander::handleWheelEvent re-targets an event to a shadow ancestor node,
however it should be done by adjusting event target. So we don't need to have it.
Tests: fast/dom/shadow/wheel-event-in-shadow-dom.html
fast/dom/shadow/wheel-event-on-input-in-shadow-dom.html
* page/EventHandler.cpp:
(WebCore::EventHandler::handleWheelEvent):
2012-05-23 Huang Dongsung <luxtella@company100.net>
Remove an unused setter in CachedResource.h.
https://bugs.webkit.org/show_bug.cgi?id=87324
CachedResource::setInLiveDecodedResourcesList was defined in r24925, but it has
been unused.
Reviewed by Darin Adler.
* loader/cache/CachedResource.h:
2012-05-23 Hayato Ito <hayato@chromium.org>
Adjust a target node of a mouse event correctly when a distributed text node is clicked.
https://bugs.webkit.org/show_bug.cgi?id=86999
Reviewed by Dimitri Glazkov.
Make EventHandler::updateMouseEventTargetNode() be aware of shadow dom subtree
so that an insertion point can receive an event when a distributed text node is clicked.
Tests: fast/dom/shadow/shadow-dom-event-dispatching.html
* page/EventHandler.cpp:
(WebCore::EventHandler::updateMouseEventTargetNode):
2012-05-23 Shinya Kawanaka <shinyak@chromium.org>
[Shadow] drop event is not fired on nodes in Shadow DOM
https://bugs.webkit.org/show_bug.cgi?id=85774
Reviewed by Dimitri Glazkov.
EventHander::updateDragAndDrop re-targets an event to a shadow ancestor node,
however it should be done by adjusting event target. So we don't need to have it.
Tests: fast/dom/shadow/drop-event-for-input-in-shadow.html
fast/dom/shadow/drop-event-in-shadow.html
* page/EventHandler.cpp:
(WebCore::EventHandler::updateDragAndDrop):
2012-05-23 Emil A Eklund <eae@chromium.org>
REGRESSION (115573): Incorrect rounding of margins for floats
https://bugs.webkit.org/show_bug.cgi?id=87319
Reviewed by Eric Seidel.
In RenderBlock::computeInlinePreferredLogicalWidths we used a float to
accumulate margins for floating children while the children themselves
represent their margins as LayoutUnits. Due to lack of rounding this can
cause the block to be too small at certain certain zoom levels, causing
unwanted wrapping.
This patch changes computeInlinePreferredLogicalWidths to use a
LayoutUnit to accumulate the margins and thus ensures that the margin
values are rounded the same way.
Test: fast/block/float/floats-with-margin-should-not-wrap.html
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::computeInlinePreferredLogicalWidths):
2012-05-23 Huang Dongsung <luxtella@company100.net>
Remove unused declarations in MemoryCache.h
https://bugs.webkit.org/show_bug.cgi?id=87323
It is originated from r75912.
Reviewed by Darin Adler.
* loader/cache/MemoryCache.h:
2012-05-23 Eric Seidel <eric@webkit.org>
Add seamless layout code (and pass most of the remaining seamless tests)
https://bugs.webkit.org/show_bug.cgi?id=86608
Reviewed by Ojan Vafai.
This patch contains almost all the layout changes needed for seamless iframes.
I removed the scroll-bar avoiding code at the last moment, as it didn't
work for platforms other than mac-lion. I'll add that, as well as the
HTMLIFrameElement.seamless idl attribute in a follow-up patch.
Seamless iframes piggy-back a bit on the existing frame-flattening
logic, however seamless is different from frame-flattening in a few ways:
- Frame flattening can only ever make an iframe larger (seamless just behaves like a normal div).
- Frame flattening disables scrollbars (seamless frames behave like normal overflow: auto divs).
- Seamless only has to work with iframes (flattening works with frame/frameset as well).
- Seamless support shrink-wrap size negotiation when the iframe is inline.
Test: fast/frames/seamless/seamless-percent-height.html
* css/StyleResolver.cpp:
(WebCore::StyleResolver::adjustRenderStyle): map inline -> inline-block for seamless iframes.
* dom/Document.cpp:
(WebCore::Document::scheduleStyleRecalc):
- Seamless iframes don't manage their own style recalc.
(WebCore::Document::recalcStyle):
- We should make sure our parent is resolved before we are, but currently that causes some tests to crash
I don't have a test to demonstrate this need yet, as presumably it's fulfilled through other codepaths atm.
* page/FrameView.cpp:
(WebCore::FrameView::scheduleRelayout): Do the cheaper check first.
(WebCore::FrameView::isInChildFrameWithFrameFlattening): Make frameview layout abort child layouts like how frame flattening does.
* rendering/RenderBox.h:
(WebCore::RenderBox::stretchesToViewport): Disable the stretch-to-viewport quirk for seamless iframes (it makes no sense and breaks the layout code).
* rendering/RenderIFrame.cpp:
(WebCore::RenderIFrame::computeLogicalHeight):
- This function is needed for the child document to participate in the normal block shrink-wrap algorithm.
Thankfully all the shrink-wrap logic is in RenderBox instead of RenderBlock. In the future we may make
RenderIframe a RenderBlock for the seamless case. We may just split RenderIframe into two renderers.
(WebCore::RenderIFrame::computeLogicalWidth):
(WebCore::RenderIFrame::shouldComputeSizeAsReplaced):
- seamless iframes behave like blocks, not inline replaced elements.
(WebCore):
(WebCore::RenderIFrame::isInlineBlockOrInlineTable):
- Behave like an inline-block when marked inline.
(WebCore::RenderIFrame::minPreferredLogicalWidth):
- When asked for our pref widths, return those of our child document.
(WebCore::RenderIFrame::maxPreferredLogicalWidth):
(WebCore::RenderIFrame::isSeamless): helper function
(WebCore::RenderIFrame::contentRootRenderer): helper function
(WebCore::RenderIFrame::flattenFrame): seamless iframes never use the frame-flattening feature.
(WebCore::RenderIFrame::layoutSeamlessly): The guts of seamless layout.
(WebCore::RenderIFrame::layout):
* rendering/RenderIFrame.h:
(WebCore):
(RenderIFrame):
2012-05-23 Rafael Brandao <rafael.lobo@openbossa.org>
[Qt] GraphicsContext should avoid converting rotation angle to degrees
https://bugs.webkit.org/show_bug.cgi?id=87317
Reviewed by Noam Rosenthal.
The angle for rotate is already in radians, so we should take advantage of it
by using QTransform::rotateRadians, rather than converting it back internally.
* platform/graphics/qt/GraphicsContextQt.cpp:
(WebCore::GraphicsContext::rotate):
2012-05-23 David Tseng <dtseng@google.com>
AX: Prevents assertion from potentially dereferencing null pointer.
https://bugs.webkit.org/show_bug.cgi?id=87290
Reviewed by Chris Fleizach.
Changes to assertions. Coverage by existing tests.
* accessibility/AccessibilityTable.cpp:
(WebCore::AccessibilityTable::cellForColumnAndRow):
2012-05-23 Tony Chang <tony@chromium.org>
auto margins on flexbox should allocate space in the cross direction
https://bugs.webkit.org/show_bug.cgi?id=86913
Reviewed by Ojan Vafai.
Test: css3/flexbox/auto-margins.html
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::hasAutoMarginsInCrossAxis):
(WebCore):
(WebCore::RenderFlexibleBox::updateAutoMarginsInCrossAxis):
(WebCore::RenderFlexibleBox::layoutAndPlaceChildren): If we have flex-align: baseline and auto margins, we ignore flex-align: baseline.
(WebCore::RenderFlexibleBox::alignChildren): auto margins apply before flex-align.
* rendering/RenderFlexibleBox.h:
2012-05-23 Kentaro Hara <haraken@chromium.org>
Unreviewed, rolling out r118120.
http://trac.webkit.org/changeset/118120
https://bugs.webkit.org/show_bug.cgi?id=87193
some LayoutTests crash since the patch passes a NULL to
Null(isolate)
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateHeader):
* bindings/scripts/test/V8/V8TestActiveDOMObject.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestCustomNamedGetter.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestEventConstructor.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestEventTarget.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestException.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestInterface.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestMediaQueryListListener.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestNamedConstructor.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestNode.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestObj.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.h:
(WebCore::toV8):
2012-05-23 Kentaro Hara <haraken@chromium.org>
Unreviewed, rolling out r118129.
http://trac.webkit.org/changeset/118129
https://bugs.webkit.org/show_bug.cgi?id=87202
some LayoutTests crash since the patch passes a NULL to
Null(isolate)
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateNormalAttrGetter):
(GenerateFunctionCallback):
(NativeToJSValue):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::TestObjV8Internal::contentDocumentAttrGetter):
(WebCore::TestObjV8Internal::getSVGDocumentCallback):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.cpp:
(WebCore::TestSerializedScriptValueInterfaceV8Internal::valueAttrGetter):
(WebCore::TestSerializedScriptValueInterfaceV8Internal::readonlyValueAttrGetter):
(WebCore::TestSerializedScriptValueInterfaceV8Internal::cachedValueAttrGetter):
(WebCore::TestSerializedScriptValueInterfaceV8Internal::cachedReadonlyValueAttrGetter):
2012-05-23 Kentaro Hara <haraken@chromium.org>
Unreviewed, rolling out r118133.
http://trac.webkit.org/changeset/118133
https://bugs.webkit.org/show_bug.cgi?id=87207
some LayoutTests crash since the patch passes a NULL to
Null(isolate)
* bindings/v8/custom/V8BlobCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8CSSRuleCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8CSSStyleSheetCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8CSSValueCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ClipboardCustom.cpp:
(WebCore::V8Clipboard::typesAccessorGetter):
* bindings/v8/custom/V8CoordinatesCustom.cpp:
(WebCore::V8Coordinates::altitudeAccessorGetter):
(WebCore::V8Coordinates::altitudeAccuracyAccessorGetter):
(WebCore::V8Coordinates::headingAccessorGetter):
(WebCore::V8Coordinates::speedAccessorGetter):
* bindings/v8/custom/V8DOMStringMapCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DOMTokenListCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DOMWindowCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DataViewCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DeviceMotionEventCustom.cpp:
(WebCore::V8DeviceMotionEvent::accelerationAccessorGetter):
(WebCore::V8DeviceMotionEvent::accelerationIncludingGravityAccessorGetter):
(WebCore::V8DeviceMotionEvent::rotationRateAccessorGetter):
(WebCore::V8DeviceMotionEvent::intervalAccessorGetter):
* bindings/v8/custom/V8DeviceOrientationEventCustom.cpp:
(WebCore::V8DeviceOrientationEvent::alphaAccessorGetter):
(WebCore::V8DeviceOrientationEvent::betaAccessorGetter):
(WebCore::V8DeviceOrientationEvent::gammaAccessorGetter):
(WebCore::V8DeviceOrientationEvent::absoluteAccessorGetter):
* bindings/v8/custom/V8DocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DocumentLocationCustom.cpp:
(WebCore::V8Document::locationAccessorGetter):
* bindings/v8/custom/V8EntryCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8EntrySyncCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8EventCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Float32ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Float64ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8HTMLCanvasElementCustom.cpp:
(WebCore::V8HTMLCanvasElement::getContextCallback):
* bindings/v8/custom/V8HTMLDocumentCustom.cpp:
(WebCore::toV8):
2012-05-23 Kentaro Hara <haraken@chromium.org>
Unreviewed, rolling out r118134.
http://trac.webkit.org/changeset/118134
https://bugs.webkit.org/show_bug.cgi?id=87209
some LayoutTests crash since the patch passes a NULL to
Null(isolate)
* bindings/v8/custom/V8HTMLElementCustom.cpp:
(WebCore::toV8Object):
(WebCore::toV8):
* bindings/v8/custom/V8HistoryCustom.cpp:
(WebCore::V8History::stateAccessorGetter):
* bindings/v8/custom/V8IDBAnyCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8IDBKeyCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ImageDataCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Int16ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Int32ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Int8ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8LocationCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8MessageEventCustom.cpp:
(WebCore::V8MessageEvent::dataAccessorGetter):
* bindings/v8/custom/V8NamedNodeMapCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8NodeCustom.cpp:
(WebCore::V8Node::insertBeforeCallback):
(WebCore::V8Node::replaceChildCallback):
(WebCore::V8Node::removeChildCallback):
(WebCore::V8Node::appendChildCallback):
(WebCore::toV8Slow):
* bindings/v8/custom/V8PopStateEventCustom.cpp:
(WebCore::V8PopStateEvent::stateAccessorGetter):
* bindings/v8/custom/V8SQLResultSetRowListCustom.cpp:
(WebCore::V8SQLResultSetRowList::itemCallback):
* bindings/v8/custom/V8SVGDocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8SVGElementCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8SVGPathSegCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ScriptProfileCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ScriptProfileNodeCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8StyleSheetCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8TrackEventCustom.cpp:
(WebCore::V8TrackEvent::trackAccessorGetter):
* bindings/v8/custom/V8Uint16ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Uint32ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Uint8ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Uint8ClampedArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
(WebCore::toV8Object):
(WebCore::V8WebGLRenderingContext::getAttachedShadersCallback):
(WebCore::V8WebGLRenderingContext::getSupportedExtensionsCallback):
* bindings/v8/custom/V8WorkerContextCustom.cpp:
(WebCore::toV8):
2012-05-23 James Robinson <jamesr@chromium.org>
Add a Setting to make position:fixed form a new stacking context
https://bugs.webkit.org/show_bug.cgi?id=87186
Reviewed by Adam Barth.
position:fixed elements form a new stacking context in several mobile WebKit ports and in the Qt port always. To
be consistent across devices, we want to try turning it on everywhere. This will likely cause compatibility
issues, so we need a runtime setting in order to stage the change carefully.
Tests: fast/block/positioning/fixed-position-stacking-context.html
fast/block/positioning/fixed-position-stacking-context2.html
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(WebCore::Settings::setFixedPositionCreatesStackingContext):
(WebCore::Settings::fixedPositionCreatesStackingContext):
(Settings):
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::InternalSettings):
(WebCore::InternalSettings::restoreTo):
(WebCore::InternalSettings::setFixedPositionCreatesStackingContext):
(WebCore):
* testing/InternalSettings.h:
(InternalSettings):
* testing/InternalSettings.idl:
2012-05-23 Michael Nordman <michaeln@google.com>
[chromium] DomStorage events handling needs TLC (3)
https://bugs.webkit.org/show_bug.cgi?id=87031
https://code.google.com/p/chromium/issues/detail?id=128482
Create WebCore::Storage instances as a side effect of attaching storage event handlers.
This allows storage events in chromium to be propagated with less IPC chatter.
Worth noting that in non-chromium ports, creation of the localStorage instance can have
an additional side effect of scheduling a background task to read the area's values
from disk, which given the interest in events is probably a beneficial side effect.
Reviewed by Adam Barth.
No new tests. Existing tests cover this.
* page/DOMWindow.cpp:
(WebCore::didAddStorageEventListener):
(WebCore::DOMWindow::addEventListener):
2012-05-23 Michael Saboff <msaboff@apple.com>
Crash in fast/files/read tests during Garbage Collection
https://bugs.webkit.org/show_bug.cgi?id=87165
Reviewed by Alexey Proskuryakov.
Changed hasPendingActivity() processing to use ActiveDOMObject::hasPendingActivity()
and associated setPendingActivity() / unsetPendingActivity().
Fixed two existing tests with change.
* Modules/filesystem/FileWriter.cpp:
(WebCore::FileWriter::stop):
(WebCore::FileWriter::write):
(WebCore::FileWriter::truncate):
(WebCore::FileWriter::signalCompletion):
* Modules/filesystem/FileWriter.h:
* fileapi/FileReader.cpp:
(WebCore::FileReader::readInternal):
(WebCore::FileReader::doAbort):
(WebCore::FileReader::didFinishLoading):
(WebCore::FileReader::didFail):
* fileapi/FileReader.h:
2012-05-23 Abhishek Arya <inferno@chromium.org>
Crash in run-ins with continuations while moving back to original position.
https://bugs.webkit.org/show_bug.cgi?id=87264
Reviewed by Julien Chaffraix.
Run-in that are now placed in sibling block can break up into continuation
chains when new children are added to it. We cannot easily send them back to their
original place since that requires writing integration logic with RenderInline::addChild
and all other places that might cause continuations to be created (without blowing away
|this|). Disabling this feature for now to prevent crashes.
Test: fast/runin/runin-continuations-crash.html
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::moveRunInToOriginalPosition):
2012-05-23 Abhishek Arya <inferno@chromium.org>
Crash in RenderInline::linesVisualOverflowBoundingBox.
https://bugs.webkit.org/show_bug.cgi?id=85804
Reviewed by Dave Hyatt.
Defer layout of replaced elements to the next line break function.
We shouldn't do it while we are clearing our inline chilren
lineboxes in full layout mode.
Test: fast/block/inline-children-root-linebox-crash.html
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::layoutInlineChildren):
(WebCore::RenderBlock::LineBreaker::nextLineBreak):
2012-05-23 Chris Rogers <crogers@google.com>
AudioParam must support fan-in (multiple audio connections) (take 2)
https://bugs.webkit.org/show_bug.cgi?id=83610
Reviewed by Kenneth Russell.
This re-lands bug: https://bugs.webkit.org/show_bug.cgi?id=83610
Now that a fix has been landed: http://trac.webkit.org/changeset/118099
Test: webaudio/audioparam-summingjunction.html
* Modules/webaudio/AudioParam.cpp:
* Modules/webaudio/AudioParam.h:
(WebCore::AudioParam::calculateSampleAccurateValues):
(WebCore::AudioParam::calculateAudioRateSignalValues):
Sums intrinsic parameter value with all audio-rate connections.
(WebCore::AudioParam::connect):
(WebCore::AudioParam::disconnect):
Support multiple connections.
(WebCore::AudioParam::hasSampleAccurateValues):
If we have one or more audio-rate connections.
(WebCore::AudioParam::AudioParam):
AudioParam now sub-classes AudioSummingJunction.
2012-05-23 Christophe Dumez <christophe.dumez@intel.com>
[EFL] webintents/web-intents-delivery-reuse.html is failing on EFL port
https://bugs.webkit.org/show_bug.cgi?id=87092
Reviewed by Adam Barth.
Remove readonly keyword from the webkitIntent attribute. This is
needed otherwise the [Replaceable] extended attribute has not effect
with JSC. This is tested by webintents/web-intents-delivery-reuse.html
test case.
* Modules/intents/DOMWindowIntents.idl:
2012-05-23 Abhishek Arya <inferno@chromium.org>
ASSERT failure toRenderProgress in HTMLProgressElement::didElementStateChange
https://bugs.webkit.org/show_bug.cgi?id=87274
Reviewed by Darin Adler.
Progress bar can't run-in. Prevent it from becoming a run-in, leading to an
unworkable RenderInline.
Test: fast/runin/progress-run-in-crash.html
* html/HTMLProgressElement.cpp:
(WebCore::HTMLProgressElement::didElementStateChange):
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::moveRunInUnderSiblingBlockIfNeeded):
2012-05-23 Andrew Lo <anlo@rim.com>
[BlackBerry] UI thread unnecessarily blocks on WebKit thread when servicing requestAnimationFrames
https://bugs.webkit.org/show_bug.cgi?id=87289
Reviewed by Antonio Gomes.
requestAnimationFrame already covered by tests in LayoutTests/fast/animation.
If UI thread cannot acquire DisplayRefreshMonitor mutex immediately,
avoid blocking on the mutex, it can trigger the frame change on the next
animation tick instead.
* platform/graphics/blackberry/DisplayRefreshMonitorBlackBerry.cpp:
(WebCore::DisplayRefreshMonitor::displayLinkFired):
2012-05-23 Alec Flett <alecflett@chromium.org>
Implement DOM4 DOMError
https://bugs.webkit.org/show_bug.cgi?id=87055
Reviewed by Adam Barth.
Implement the DOMError class and it's IDL interface.
No new tests. Not testable yet, as it's not used.
* WebCore.gypi:
* dom/DOMError.cpp: Added.
(WebCore):
(WebCore::DOMError::DOMError):
* dom/DOMError.h: Added.
(WebCore):
(DOMError):
(WebCore::DOMError::create):
(WebCore::DOMError::name):
* dom/DOMError.idl: Added.
2012-05-23 Kenneth Russell <kbr@google.com>
REGRESSION(117918) - 4 webgl/tex-iomage-and-sub-image-2d-with-image tests failing on Mac bots.
https://bugs.webkit.org/show_bug.cgi?id=87136
Reviewed by Stephen White.
Fixed longstanding bug in GraphicsContext3D constructor in Mac
port, and logic error in tracking of texture bound to unit 0. Also
fixed up Cairo port's constructor by code inspection. This code is
ripe for cleanup; Bug 87198 filed for this.
Covered by existing tests.
* platform/graphics/cairo/GraphicsContext3DCairo.cpp:
(WebCore::GraphicsContext3D::GraphicsContext3D):
Initialize missing members.
* platform/graphics/mac/GraphicsContext3DMac.mm:
(WebCore::GraphicsContext3D::GraphicsContext3D):
Correctly initialize m_activeTexture.
* platform/graphics/opengl/GraphicsContext3DOpenGLCommon.cpp:
(WebCore::GraphicsContext3D::bindTexture):
Fix logic error in tracking of texture bound to unit 0.
2012-05-23 Dana Jansens <danakj@chromium.org>
[chromium] Layers on main thread should get a RenderSurface for animating transform only if masksToBounds is true also
https://bugs.webkit.org/show_bug.cgi?id=87184
Reviewed by James Robinson.
Layers on main thread get a RenderSurface if their impl counterpart may
have a RenderSurface due to its animating transform. We currently create
a RenderSurface for all layers with an animating transform and
a descendant that drawsContents. But a RenderSurface on impl also
implies that the layer masksToBounds. So only make a RenderSurface on
main thread when the layer also masksToBounds.
Modified unit test CCLayerTreeHostCommonTest.verifyAnimationsForRenderSurfaceHierarchy.
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::subtreeShouldRenderToSeparateSurface):
2012-05-23 Chris Fleizach <cfleizach@apple.com>
Regression(r112694): Crash in WebCore::AXObjectCache::postNotification
https://bugs.webkit.org/show_bug.cgi?id=86029
Reviewed by Abhishek Arya.
Test: accessibility/content-changed-notification-causes-crash.html
* accessibility/AccessibilityObject.h:
(WebCore::AccessibilityObject::isDetached):
(AccessibilityObject):
* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::contentChanged):
2012-05-23 Robin Dunn <robin@alldunn.com>
[wx] Fix memory leak in FontHolder.
https://bugs.webkit.org/show_bug.cgi?id=87281
Reviewed by Kevin Ollivier.
* platform/graphics/wx/FontPlatformData.h:
2012-05-23 Robin Dunn <robin@alldunn.com>
[wx] Make sure we do not create a substitute font if the facename is not found.
https://bugs.webkit.org/show_bug.cgi?id=87280
Reviewed by Kevin Ollivier.
* platform/graphics/wx/FontCacheWx.cpp:
(WebCore::FontCache::createFontPlatformData):
2012-05-23 Robin Dunn <robin@alldunn.com>
[wx] Implement transparency layer support.
https://bugs.webkit.org/show_bug.cgi?id=87279
Reviewed by Kevin Ollivier.
* platform/graphics/wx/GraphicsContextWx.cpp:
(WebCore::GraphicsContext::beginPlatformTransparencyLayer):
(WebCore::GraphicsContext::endPlatformTransparencyLayer):
(WebCore::GraphicsContext::supportsTransparencyLayers):
2012-05-23 Jer Noble <jer.noble@apple.com>
GenericClock loses precision due to float->double conversion.
https://bugs.webkit.org/show_bug.cgi?id=87227
Reviewed by Darin Adler.
No new tests; fixes media/media-controller-time.html on Chromium.
Only narrow to float after the entire calculation is complete. This requires now() and m_offset
to be double-precision.
* platform/ClockGeneric.cpp:
(ClockGeneric::currentTime):
(ClockGeneric::now): Make double-precision.
* platform/ClockGeneric.h:
(ClockGeneric): Make m_offset double-precision.
2012-05-22 Jer Noble <jer.noble@apple.com>
PlatformClockCM has uninitialized m_rate member.
https://bugs.webkit.org/show_bug.cgi?id=87217
Reviewed by Eric Carlson.
Test: media/media-controller-time.html
Initialize the m_rate member to a default of 1 (second-per-second), matching the implementation
of ClockGeneric.
* platform/mac/PlatformClockCM.mm:
(PlatformClockCM::PlatformClockCM):
2012-05-23 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118096.
http://trac.webkit.org/changeset/118096
https://bugs.webkit.org/show_bug.cgi?id=87285
caused multiple crashes on the mac bots (Requested by
jessieberlin on #webkit).
* dom/ContainerNodeAlgorithms.h:
2012-05-23 Darin Adler <darin@apple.com>
Speed up traverseNextNode when called on a ContainerNode or Element
https://bugs.webkit.org/show_bug.cgi?id=87224
Reviewed by Geoffrey Garen.
* dom/ContainerNode.h: Added overrides of traverseNextNode for ContainerNode.
That way, when we know at compile time something is ContainerNode, we can
skip a branch at the start of this hot function.
2012-05-23 Sudarsana Nagineni <sudarsana.nagineni@linux.intel.com>
Use PluginDataNone.cpp always when NETSCAPE_PLUGIN_API is OFF
https://bugs.webkit.org/show_bug.cgi?id=87096
Reviewed by Andy Estes.
Fix a build error by adding an extra check in PluginData.cpp to use
PluginStrategy only when NETSCAPE_PLUGIN_API is ON.
No new tests. This is only a build fix.
* plugins/PluginData.cpp:
(WebCore):
2012-05-23 Malcolm MacLeod <malcolm.macleod@tshwanedje.com>
[wx] Fix scrollbar drawing under wxGTK.
https://bugs.webkit.org/show_bug.cgi?id=87150
Reviewed by Kevin Ollivier.
* platform/wx/wxcode/gtk/scrollbar_render.cpp:
(wxRenderer_DrawScrollbar):
2012-05-23 Antti Koivisto <antti@apple.com>
Rename StyleSheetInternal to StyleSheetContents
https://bugs.webkit.org/show_bug.cgi?id=87272
Reviewed by Andreas Kling.
The name is more descriptive and the code reads better.
Plain "StyleSheet" would be good too but that is a CSSOM type and the name can't be used without
changing idl code generation and resolving bunch of other issues.
* css/CSSFontFaceSrcValue.cpp:
(WebCore::CSSFontFaceSrcValue::addSubresourceStyleURLs):
* css/CSSFontFaceSrcValue.h:
(CSSFontFaceSrcValue):
* css/CSSImportRule.cpp:
(WebCore::StyleRuleImport::setCSSStyleSheet):
(WebCore::StyleRuleImport::requestStyleSheet):
* css/CSSImportRule.h:
(WebCore):
(WebCore::StyleRuleImport::parentStyleSheet):
(WebCore::StyleRuleImport::setParentStyleSheet):
(WebCore::StyleRuleImport::styleSheet):
(StyleRuleImport):
* css/CSSMediaRule.cpp:
(WebCore::CSSMediaRule::insertRule):
* css/CSSParser.cpp:
(WebCore::CSSParser::parseSheet):
(WebCore::CSSParser::parseRule):
(WebCore::CSSParser::parseKeyframeRule):
(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::parseSelector):
(WebCore::CSSParser::parseDeclaration):
(WebCore::CSSParser::setStyleSheet):
(WebCore::CSSParser::parseFlowThread):
* css/CSSParser.h:
(WebCore):
(CSSParser):
* css/CSSPrimitiveValue.cpp:
(WebCore::CSSPrimitiveValue::addSubresourceStyleURLs):
* css/CSSPrimitiveValue.h:
(CSSPrimitiveValue):
* css/CSSReflectValue.cpp:
(WebCore::CSSReflectValue::addSubresourceStyleURLs):
* css/CSSReflectValue.h:
(CSSReflectValue):
* css/CSSRule.cpp:
(WebCore::CSSRule::parserContext):
* css/CSSStyleSheet.cpp:
(WebCore::StyleSheetContents::estimatedSizeInBytes):
(WebCore::StyleSheetContents::StyleSheetContents):
(WebCore::StyleSheetContents::~StyleSheetContents):
(WebCore::StyleSheetContents::isCacheable):
(WebCore::StyleSheetContents::parserAppendRule):
(WebCore::StyleSheetContents::ruleAt):
(WebCore::StyleSheetContents::ruleCount):
(WebCore::StyleSheetContents::clearCharsetRule):
(WebCore::StyleSheetContents::clearRules):
(WebCore::StyleSheetContents::parserSetEncodingFromCharsetRule):
(WebCore::StyleSheetContents::wrapperInsertRule):
(WebCore::StyleSheetContents::wrapperDeleteRule):
(WebCore::StyleSheetContents::parserAddNamespace):
(WebCore::StyleSheetContents::determineNamespace):
(WebCore::StyleSheetContents::parseAuthorStyleSheet):
(WebCore::StyleSheetContents::parseString):
(WebCore::StyleSheetContents::parseStringAtLine):
(WebCore::StyleSheetContents::isLoading):
(WebCore::StyleSheetContents::checkLoaded):
(WebCore::StyleSheetContents::notifyLoadedSheet):
(WebCore::StyleSheetContents::startLoadingDynamicSheet):
(WebCore::StyleSheetContents::rootStyleSheet):
(WebCore::StyleSheetContents::singleOwnerNode):
(WebCore::StyleSheetContents::singleOwnerDocument):
(WebCore::StyleSheetContents::completeURL):
(WebCore::StyleSheetContents::addSubresourceStyleURLs):
(WebCore::StyleSheetContents::parentStyleSheet):
(WebCore::StyleSheetContents::registerClient):
(WebCore::StyleSheetContents::unregisterClient):
(WebCore::StyleSheetContents::addedToMemoryCache):
(WebCore::StyleSheetContents::removedFromMemoryCache):
(WebCore::CSSStyleSheet::createInline):
(WebCore::CSSStyleSheet::CSSStyleSheet):
(WebCore::CSSStyleSheet::~CSSStyleSheet):
(WebCore::CSSStyleSheet::willMutateRules):
(WebCore::CSSStyleSheet::didMutateRules):
(WebCore::CSSStyleSheet::reattachChildRuleCSSOMWrappers):
(WebCore::CSSStyleSheet::length):
(WebCore::CSSStyleSheet::item):
(WebCore::CSSStyleSheet::rules):
(WebCore::CSSStyleSheet::insertRule):
(WebCore::CSSStyleSheet::deleteRule):
(WebCore::CSSStyleSheet::cssRules):
* css/CSSStyleSheet.h:
(WebCore::StyleSheetContents::create):
(StyleSheetContents):
(WebCore::StyleSheetContents::copy):
(WebCore::CSSStyleSheet::create):
(WebCore::CSSStyleSheet::contents):
(CSSStyleSheet):
* css/CSSValue.cpp:
(WebCore::CSSValue::addSubresourceStyleURLs):
* css/CSSValue.h:
(WebCore):
(CSSValue):
* css/CSSValueList.cpp:
(WebCore::CSSValueList::addSubresourceStyleURLs):
* css/CSSValueList.h:
(CSSValueList):
* css/PropertySetCSSStyleDeclaration.cpp:
(WebCore::PropertySetCSSStyleDeclaration::contextStyleSheet):
* css/PropertySetCSSStyleDeclaration.h:
(WebCore):
(PropertySetCSSStyleDeclaration):
* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::setProperty):
(WebCore::StylePropertySet::parseDeclaration):
(WebCore::StylePropertySet::addSubresourceStyleURLs):
* css/StylePropertySet.h:
(WebCore):
(StylePropertySet):
* css/StyleResolver.cpp:
(RuleSet):
(WebCore):
(WebCore::StyleResolver::StyleResolver):
(WebCore::StyleResolver::addAuthorRulesAndCollectUserRulesFromSheets):
(WebCore::StyleResolver::appendAuthorStylesheets):
(WebCore::parseUASheet):
(WebCore::StyleResolver::collectMatchingRulesForList):
* css/StyleResolver.h:
(StyleResolver):
* css/WebKitCSSKeyframesRule.cpp:
(WebCore::WebKitCSSKeyframesRule::insertRule):
* dom/DOMImplementation.cpp:
(WebCore::DOMImplementation::createCSSStyleSheet):
* dom/Document.cpp:
(WebCore::Document::recalcStyle):
(WebCore::Document::updateBaseURL):
(WebCore::Document::pageUserSheet):
(WebCore::Document::pageGroupUserSheets):
(WebCore::Document::addUserSheet):
(WebCore::Document::testAddedStylesheetRequiresStyleRecalc):
(WebCore::Document::analyzeStylesheetChange):
(WebCore::styleSheetsUseRemUnits):
* dom/Document.h:
(WebCore):
(Document):
* dom/ElementAttributeData.cpp:
(WebCore::ElementAttributeData::updateInlineStyleAvoidingMutation):
* dom/ProcessingInstruction.cpp:
(WebCore::ProcessingInstruction::setCSSStyleSheet):
(WebCore::ProcessingInstruction::parseStyleSheet):
* dom/StyleElement.cpp:
(WebCore::StyleElement::createSheet):
* dom/StyledElement.cpp:
(WebCore::StyledElement::setInlineStyleProperty):
(WebCore::StyledElement::addSubresourceAttributeURLs):
(WebCore::StyledElement::addPropertyToAttributeStyle):
* html/HTMLLinkElement.cpp:
(WebCore::HTMLLinkElement::setCSSStyleSheet):
(WebCore::HTMLLinkElement::styleSheetIsLoading):
(WebCore::HTMLLinkElement::addSubresourceAttributeURLs):
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::addSubresourceAttributeURLs):
* inspector/InspectorStyleSheet.cpp:
(WebCore::fillMediaListChain):
(WebCore::InspectorStyle::setPropertyText):
(WebCore::InspectorStyleSheet::styleSheetURL):
(WebCore::InspectorStyleSheet::reparseStyleSheet):
(WebCore::InspectorStyleSheet::ensureSourceData):
(WebCore::InspectorStyleSheetForInlineStyle::getStyleAttributeRanges):
* loader/cache/CachedCSSStyleSheet.cpp:
(WebCore::CachedCSSStyleSheet::restoreParsedStyleSheet):
(WebCore::CachedCSSStyleSheet::saveParsedStyleSheet):
* loader/cache/CachedCSSStyleSheet.h:
(WebCore):
(CachedCSSStyleSheet):
* page/PageSerializer.cpp:
(WebCore::PageSerializer::serializeCSSStyleSheet):
2012-05-23 Darin Adler <darin@apple.com>
Use ContainerNode instead of Node where possible
https://bugs.webkit.org/show_bug.cgi?id=87220
Reviewed by Geoffrey Garen.
It's better to use a more specific type; in some cases we even generate
more efficient code if we have a more specific type. Also, we want any
type casts to be as close as possible to the corresponding type checks,
so eliminating these uses of toContainerNode is a plus, also.
* dom/ContainerNodeAlgorithms.h: Changed insertionPoint to be a ContainerNode
instead of a Node. Fixed spelling error "inseretions". Changed
(WebCore::ChildFrameDisconnector::Target::Target): Changed type of frame owner
element to HTMLFrameOwnerElement from Node.
* dom/DocumentType.cpp:
(WebCore::DocumentType::insertedInto):
(WebCore::DocumentType::removedFrom):
* dom/DocumentType.h:
* dom/Element.cpp:
(WebCore::Element::insertedInto):
(WebCore::Element::removedFrom):
* dom/Element.h:
* dom/Node.cpp:
(WebCore::Node::insertedInto):
(WebCore::Node::removedFrom):
* dom/Node.h:
* dom/ProcessingInstruction.cpp:
(WebCore::ProcessingInstruction::insertedInto):
(WebCore::ProcessingInstruction::removedFrom):
* dom/ProcessingInstruction.h:
* dom/ScriptElement.cpp:
(WebCore::ScriptElement::insertedInto):
* dom/ScriptElement.h:
* html/FormAssociatedElement.cpp:
(WebCore::FormAssociatedElement::insertedInto):
(WebCore::FormAssociatedElement::removedFrom):
* html/FormAssociatedElement.h:
* html/HTMLBaseElement.cpp:
(WebCore::HTMLBaseElement::insertedInto):
(WebCore::HTMLBaseElement::removedFrom):
* html/HTMLBaseElement.h:
* html/HTMLBodyElement.cpp:
(WebCore::HTMLBodyElement::insertedInto):
(WebCore::HTMLBodyElement::didNotifyDescendantInsertions):
* html/HTMLBodyElement.h:
* html/HTMLFormControlElement.cpp:
(WebCore::HTMLFormControlElement::insertedInto):
(WebCore::HTMLFormControlElement::removedFrom):
* html/HTMLFormControlElement.h:
* html/HTMLFormElement.cpp:
(WebCore::HTMLFormElement::insertedInto):
(WebCore::HTMLFormElement::didNotifyDescendantInsertions):
(WebCore::HTMLFormElement::removedFrom):
* html/HTMLFormElement.h:
* html/HTMLFrameElementBase.cpp:
(WebCore::HTMLFrameElementBase::insertedInto):
(WebCore::HTMLFrameElementBase::didNotifyDescendantInsertions):
* html/HTMLFrameElementBase.h:
* html/HTMLFrameSetElement.cpp:
(WebCore::HTMLFrameSetElement::insertedInto):
(WebCore::HTMLFrameSetElement::removedFrom):
* html/HTMLFrameSetElement.h:
* html/HTMLIFrameElement.cpp:
(WebCore::HTMLIFrameElement::insertedInto):
(WebCore::HTMLIFrameElement::removedFrom):
* html/HTMLIFrameElement.h:
* html/HTMLImageElement.cpp:
(WebCore::HTMLImageElement::insertedInto):
(WebCore::HTMLImageElement::removedFrom):
* html/HTMLImageElement.h:
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::insertedInto):
(WebCore::HTMLInputElement::removedFrom):
* html/HTMLInputElement.h:
* html/HTMLLinkElement.cpp:
(WebCore::HTMLLinkElement::insertedInto):
(WebCore::HTMLLinkElement::removedFrom):
* html/HTMLLinkElement.h:
* html/HTMLMapElement.cpp:
(WebCore::HTMLMapElement::insertedInto):
(WebCore::HTMLMapElement::removedFrom):
* html/HTMLMapElement.h:
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::insertedInto):
(WebCore::HTMLMediaElement::removedFrom):
* html/HTMLMediaElement.h:
* html/HTMLMetaElement.cpp:
(WebCore::HTMLMetaElement::insertedInto):
* html/HTMLMetaElement.h:
* html/HTMLObjectElement.cpp:
(WebCore::HTMLObjectElement::insertedInto):
(WebCore::HTMLObjectElement::removedFrom):
* html/HTMLObjectElement.h:
* html/HTMLOptionElement.cpp:
(WebCore::HTMLOptionElement::insertedInto):
* html/HTMLOptionElement.h:
* html/HTMLQuoteElement.cpp:
(WebCore::HTMLQuoteElement::insertedInto):
* html/HTMLQuoteElement.h:
* html/HTMLScriptElement.cpp:
(WebCore::HTMLScriptElement::insertedInto):
* html/HTMLScriptElement.h:
* html/HTMLSelectElement.cpp:
(WebCore::HTMLSelectElement::insertedInto):
* html/HTMLSelectElement.h:
* html/HTMLSourceElement.cpp:
(WebCore::HTMLSourceElement::insertedInto):
(WebCore::HTMLSourceElement::removedFrom):
* html/HTMLSourceElement.h:
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::insertedInto):
(WebCore::HTMLStyleElement::removedFrom):
* html/HTMLStyleElement.h:
* html/HTMLTextFormControlElement.cpp:
(WebCore::HTMLTextFormControlElement::insertedInto):
* html/HTMLTextFormControlElement.h:
* html/HTMLTitleElement.cpp:
(WebCore::HTMLTitleElement::insertedInto):
(WebCore::HTMLTitleElement::removedFrom):
* html/HTMLTitleElement.h:
* html/HTMLTrackElement.cpp:
(WebCore::HTMLTrackElement::insertedInto):
(WebCore::HTMLTrackElement::removedFrom):
* html/HTMLTrackElement.h:
* mathml/MathMLMathElement.cpp:
(WebCore::MathMLMathElement::insertedInto):
* mathml/MathMLMathElement.h:
* svg/SVGElement.cpp:
(WebCore::SVGElement::removedFrom):
* svg/SVGElement.h:
* svg/SVGFEImageElement.cpp:
(WebCore::SVGFEImageElement::insertedInto):
(WebCore::SVGFEImageElement::removedFrom):
* svg/SVGFEImageElement.h:
* svg/SVGFontFaceElement.cpp:
(WebCore::SVGFontFaceElement::insertedInto):
(WebCore::SVGFontFaceElement::removedFrom):
* svg/SVGFontFaceElement.h:
* svg/SVGFontFaceUriElement.cpp:
(WebCore::SVGFontFaceUriElement::insertedInto):
* svg/SVGFontFaceUriElement.h:
* svg/SVGGlyphElement.cpp:
(WebCore::SVGGlyphElement::insertedInto):
(WebCore::SVGGlyphElement::removedFrom):
* svg/SVGGlyphElement.h:
* svg/SVGHKernElement.cpp:
(WebCore::SVGHKernElement::insertedInto):
(WebCore::SVGHKernElement::removedFrom):
* svg/SVGHKernElement.h:
* svg/SVGImageElement.cpp:
(WebCore::SVGImageElement::insertedInto):
* svg/SVGImageElement.h:
* svg/SVGSVGElement.cpp:
(WebCore::SVGSVGElement::insertedInto):
(WebCore::SVGSVGElement::removedFrom):
* svg/SVGSVGElement.h:
* svg/SVGScriptElement.cpp:
(WebCore::SVGScriptElement::insertedInto):
* svg/SVGScriptElement.h:
* svg/SVGStyleElement.cpp:
(WebCore::SVGStyleElement::insertedInto):
(WebCore::SVGStyleElement::removedFrom):
* svg/SVGStyleElement.h:
* svg/SVGStyledElement.cpp:
(WebCore::SVGStyledElement::insertedInto):
(WebCore::SVGStyledElement::removedFrom):
* svg/SVGStyledElement.h:
* svg/SVGTRefElement.cpp:
(WebCore::SVGTRefElement::insertedInto):
(WebCore::SVGTRefElement::removedFrom):
* svg/SVGTRefElement.h:
* svg/SVGTextPathElement.cpp:
(WebCore::SVGTextPathElement::insertedInto):
* svg/SVGTextPathElement.h:
* svg/SVGTitleElement.cpp:
(WebCore::SVGTitleElement::insertedInto):
(WebCore::SVGTitleElement::removedFrom):
* svg/SVGTitleElement.h:
* svg/SVGUseElement.cpp:
(WebCore::SVGUseElement::insertedInto):
(WebCore::SVGUseElement::removedFrom):
* svg/SVGUseElement.h:
* svg/SVGVKernElement.cpp:
(WebCore::SVGVKernElement::insertedInto):
(WebCore::SVGVKernElement::removedFrom):
* svg/SVGVKernElement.h:
* svg/animation/SVGSMILElement.cpp:
(WebCore::SVGSMILElement::insertedInto):
(WebCore::SVGSMILElement::removedFrom):
* svg/animation/SVGSMILElement.h:
Changed arguments of insertedInto and removedFrom to ContainerNode instead of Node.
Did the same with didNotifyDescendantInsertions, while fixing the typo in its name.
* editing/ReplaceSelectionCommand.cpp:
(WebCore::ReplaceSelectionCommand::doApply): Put a typecast toHTMLElement here.
The check for isListElement and isLegacyAppleStyleSpan takes care of the type checking.
(WebCore::ReplaceSelectionCommand::insertAsListItems): Changed this function to take
an HTMLElement instead of a Node, then we can drop use of the toContainerNode function.
* editing/ReplaceSelectionCommand.h:
* editing/TextIterator.cpp: Fixed a typo in a comment.
2012-05-23 Alexei Filippov <alexeif@chromium.org>
Web Inspector: filter out debugging stuff from the heap class view
https://bugs.webkit.org/show_bug.cgi?id=86723
We mark objects that are accessible from user roots (e.g. Window)
with pageObject attribute. This fix filters out all non-pageObjects
from the classes view.
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshotNode.prototype.isPageObject):
(WebInspector.HeapSnapshot.prototype._buildAggregates):
(WebInspector.HeapSnapshot.prototype._calculateClassesRetainedSize):
(WebInspector.HeapSnapshot.prototype._buildPostOrderIndex):
(WebInspector.HeapSnapshot.prototype._markPageOwnedNodes):
(WebInspector.HeapSnapshot.prototype.createNodesProviderForClass):
2012-05-23 Andrei Bucur <abucur@adobe.com>
[CSSRegions]Implement NamedFlow::name attribute
https://bugs.webkit.org/show_bug.cgi?id=79645
Implementation for the NamedFlow::name attribute per the CSS Regions WD 3-May-2012 recommendation:
http://www.w3.org/TR/2012/WD-css3-regions-20120503/
Reviewed by Andreas Kling.
Test: fast/regions/webkit-named-flow-name.html
* dom/WebKitNamedFlow.cpp:
(WebCore::WebKitNamedFlow::name):
(WebCore):
* dom/WebKitNamedFlow.h:
(WebKitNamedFlow):
* dom/WebKitNamedFlow.idl:
2012-05-23 Andreas Kling <kling@webkit.org>
Let the compiler generate FontFamily's copy ctor and assignment operator.
Blindly rubberstamped by the brave Antti Koivisto.
* platform/graphics/FontFamily.cpp:
* platform/graphics/FontFamily.h:
2012-05-23 Andreas Kling <kling@webkit.org>
CSSParser: Simplify StylePropertySet creation.
<http://webkit.org/b/87259>
Reviewed by Alexis Menard.
Turn CSSParser's filteredProperties() into createStylePropertySet() which filters the parsed
properties and creates a new StylePropertySet, adopting the vector to avoid the Vector copy we
were previously incurring.
Added a StylePropertyVector typedef for Vector<CSSProperty, 4> since we need to pass this type
to the constructor for Vector::swap() to work.
And thus StylePropertySet::create(const Vector<CSSProperty>&, CSSParserMode) became
StylePropertySet::adopt(StylePropertyVector&, CSSParserMode).
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::copyPropertiesInSet):
* css/CSSParser.cpp:
(WebCore::CSSParser::createStylePropertySet):
(WebCore::CSSParser::createStyleRule):
(WebCore::CSSParser::createFontFaceRule):
(WebCore::CSSParser::createPageRule):
(WebCore::CSSParser::createKeyframe):
* css/CSSParser.h:
* css/StylePropertySet.cpp:
(WebCore::StylePropertySet::StylePropertySet):
(WebCore::StylePropertySet::removePropertiesInSet):
(WebCore::StylePropertySet::copyPropertiesInSet):
(SameSizeAsStylePropertySet):
* css/StylePropertySet.h:
(WebCore::StylePropertySet::adopt):
(StylePropertySet):
2012-05-23 Andreas Kling <kling@webkit.org>
Kill some frivolous null-checking of Element::m_attributeData.
<http://webkit.org/b/87248>
Reviewed by Antti Koivisto.
Replace unnecessary Element::m_attributeData with assertions where it makes sense.
Others are flat out removed because it should be obvious from context that it's non-null.
E.g if an attribute has changed, we're obviously storing it somewhere.
Also added a short-circuit to Element::normalize() if there are no Attr objects
associated with the element.
* dom/Element.cpp:
(WebCore::Element::~Element):
(WebCore::Element::attributeChanged):
(WebCore::Element::insertedInto):
(WebCore::Element::removedFrom):
(WebCore::Element::removeAttributeNode):
(WebCore::Element::normalizeAttributes):
* dom/StyledElement.cpp:
(WebCore::StyledElement::classAttributeChanged):
2012-05-23 Andreas Kling <kling@webkit.org>
Let Element::detachAttribute() handle removal of the attribute.
<http://webkit.org/b/87242>
Reviewed by Antti Koivisto.
All the call sites were removing the attribute after detaching it, so move
that into detachAttribute(). The function name already made it sound like it
removes the attribute anyway.
Also sprinkle some assertions wrt being called with guaranteed presence of
attributeData() and the provided index in the attribute vector.
* dom/Element.cpp:
(WebCore::Element::detachAttribute):
(WebCore::Element::removeAttributeNode):
* dom/NamedNodeMap.cpp:
(WebCore::NamedNodeMap::removeNamedItem):
(WebCore::NamedNodeMap::removeNamedItemNS):
2012-05-23 Andreas Kling <kling@webkit.org>
Streamline Element teardown for the no-Attr-objects case.
<http://webkit.org/b/87228>
Reviewed by Antti Koivisto.
Dodge one function call in ~Element() if the element has no Attr objects
associated (the typical case.)
Renamed detachAttributes to detachAttrObjectsFromElement to reduce ambiguity.
* dom/Element.cpp:
(WebCore::Element::~Element):
* dom/ElementAttributeData.cpp:
(WebCore::ElementAttributeData::detachAttrObjectsFromElement):
(WebCore::ElementAttributeData::clearAttributes):
* dom/ElementAttributeData.h:
(ElementAttributeData):
2012-05-23 Jochen Eisinger <jochen@chromium.org>
Fix incomplete switch statement in CSSParser after r118082
https://bugs.webkit.org/show_bug.cgi?id=87250
Reviewed by Alexis Menard.
* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
2012-05-23 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118105.
http://trac.webkit.org/changeset/118105
https://bugs.webkit.org/show_bug.cgi?id=87255
New LayoutTest is causing several cookie tests to fail on
numerous bots (Requested by steveblock on #webkit).
* platform/blackberry/CookieJarBlackBerry.cpp:
(WebCore::setCookies):
* platform/blackberry/CookieManager.cpp:
(WebCore::CookieManager::setCookies):
(WebCore::CookieManager::checkAndTreatCookie):
(WebCore::CookieManager::addCookieToMap):
(WebCore::CookieManager::setPrivateMode):
(WebCore):
* platform/blackberry/CookieManager.h:
* platform/blackberry/CookieMap.cpp:
(WebCore::CookieMap::addOrReplaceCookie):
(WebCore::CookieMap::removeCookie):
* platform/blackberry/CookieMap.h:
(CookieMap):
2012-05-23 Zan Dobersek <zandobersek@gmail.com>
[Gtk] EventSender should properly handle (left|right)(control|shift|alt) after r118001
https://bugs.webkit.org/show_bug.cgi?id=87221
Reviewed by Martin Robinson.
No new tests - covered by fast/events/keydown-leftright-keys.html.
Windows key code for shift, control and alt keys should respect the
left or right position of that key. This way the DOM location code of
a key event is properly determined for these keys.
* platform/gtk/PlatformKeyboardEventGtk.cpp:
(WebCore::PlatformKeyboardEvent::windowsKeyCodeForGdkKeyCode):
2012-05-23 Ilya Tikhonovsky <loislo@chromium.org>
Web Inspector: eliminate getters from HeapSnapshot.js
https://bugs.webkit.org/show_bug.cgi?id=87241
getters can't be inlined by engine at the moment.
I found that conversion getters into functions gives us about 600 ms
RESULT heap-snapshot: switch-to-containment-view= 1920 ms
vs
RESULT heap-snapshot: switch-to-containment-view= 1367 ms
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshot.js:
(WebInspector.HeapSnapshotEdge.prototype.hasStringName):
(WebInspector.HeapSnapshotEdge.prototype.isElement):
(WebInspector.HeapSnapshotEdge.prototype.isHidden):
(WebInspector.HeapSnapshotEdge.prototype.isWeak):
(WebInspector.HeapSnapshotEdge.prototype.isInternal):
(WebInspector.HeapSnapshotEdge.prototype.isInvisible):
(WebInspector.HeapSnapshotEdge.prototype.isShortcut):
(WebInspector.HeapSnapshotEdge.prototype.name):
(WebInspector.HeapSnapshotEdge.prototype.node):
(WebInspector.HeapSnapshotEdge.prototype.nodeIndex):
(WebInspector.HeapSnapshotEdge.prototype.rawEdges):
(WebInspector.HeapSnapshotEdge.prototype.toString):
(WebInspector.HeapSnapshotEdge.prototype.type):
(WebInspector.HeapSnapshotEdge.prototype._hasStringName):
(WebInspector.HeapSnapshotEdge.prototype._name):
(WebInspector.HeapSnapshotEdge.prototype._nameOrIndex):
(WebInspector.HeapSnapshotEdgeIterator.prototype.index):
(WebInspector.HeapSnapshotEdgeIterator.prototype.setIndex):
(WebInspector.HeapSnapshotEdgeIterator.prototype.item):
(WebInspector.HeapSnapshotRetainerEdge):
(WebInspector.HeapSnapshotRetainerEdge.prototype.clone):
(WebInspector.HeapSnapshotRetainerEdge.prototype.hasStringName):
(WebInspector.HeapSnapshotRetainerEdge.prototype.isElement):
(WebInspector.HeapSnapshotRetainerEdge.prototype.isHidden):
(WebInspector.HeapSnapshotRetainerEdge.prototype.isInternal):
(WebInspector.HeapSnapshotRetainerEdge.prototype.isInvisible):
(WebInspector.HeapSnapshotRetainerEdge.prototype.isShortcut):
(WebInspector.HeapSnapshotRetainerEdge.prototype.isWeak):
(WebInspector.HeapSnapshotRetainerEdge.prototype.name):
(WebInspector.HeapSnapshotRetainerEdge.prototype.node):
(WebInspector.HeapSnapshotRetainerEdge.prototype.nodeIndex):
(WebInspector.HeapSnapshotRetainerEdge.prototype.retainerIndex):
(WebInspector.HeapSnapshotRetainerEdge.prototype.setRetainerIndex):
(WebInspector.HeapSnapshotRetainerEdge.prototype._node):
(WebInspector.HeapSnapshotRetainerEdge.prototype._edge):
(WebInspector.HeapSnapshotRetainerEdge.prototype.toString):
(WebInspector.HeapSnapshotRetainerEdge.prototype.type):
(WebInspector.HeapSnapshotRetainerEdgeIterator.prototype.first):
(WebInspector.HeapSnapshotRetainerEdgeIterator.prototype.hasNext):
(WebInspector.HeapSnapshotRetainerEdgeIterator.prototype.index):
(WebInspector.HeapSnapshotRetainerEdgeIterator.prototype.setIndex):
(WebInspector.HeapSnapshotRetainerEdgeIterator.prototype.item):
(WebInspector.HeapSnapshotRetainerEdgeIterator.prototype.next):
(WebInspector.HeapSnapshotNode.prototype.canBeQueried):
(WebInspector.HeapSnapshotNode.prototype.distanceToWindow):
(WebInspector.HeapSnapshotNode.prototype.className):
(WebInspector.HeapSnapshotNode.prototype.dominatorIndex):
(WebInspector.HeapSnapshotNode.prototype.edges):
(WebInspector.HeapSnapshotNode.prototype.edgesCount):
(WebInspector.HeapSnapshotNode.prototype.flags):
(WebInspector.HeapSnapshotNode.prototype.id):
(WebInspector.HeapSnapshotNode.prototype.isHidden):
(WebInspector.HeapSnapshotNode.prototype.isNative):
(WebInspector.HeapSnapshotNode.prototype.isSynthetic):
(WebInspector.HeapSnapshotNode.prototype.isWindow):
(WebInspector.HeapSnapshotNode.prototype.isDetachedDOMTreesRoot):
(WebInspector.HeapSnapshotNode.prototype.isDetachedDOMTree):
(WebInspector.HeapSnapshotNode.prototype.isRoot):
(WebInspector.HeapSnapshotNode.prototype.name):
(WebInspector.HeapSnapshotNode.prototype.rawEdges):
(WebInspector.HeapSnapshotNode.prototype.retainedSize):
(WebInspector.HeapSnapshotNode.prototype.retainers):
(WebInspector.HeapSnapshotNode.prototype.selfSize):
(WebInspector.HeapSnapshotNode.prototype.type):
(WebInspector.HeapSnapshotNode.prototype._name):
(WebInspector.HeapSnapshotNode.prototype._edgeIndexesEnd):
(WebInspector.HeapSnapshotNode.prototype._nextNodeIndex):
(WebInspector.HeapSnapshotNode.prototype._type):
(WebInspector.HeapSnapshotNodeIterator):
(WebInspector.HeapSnapshotNodeIterator.prototype.hasNext):
(WebInspector.HeapSnapshotNodeIterator.prototype.index):
(WebInspector.HeapSnapshotNodeIterator.prototype.setIndex):
(WebInspector.HeapSnapshotNodeIterator.prototype.item):
(WebInspector.HeapSnapshotNodeIterator.prototype.next):
(WebInspector.HeapSnapshot.prototype._allNodes):
(WebInspector.HeapSnapshot.prototype.rootNode):
(WebInspector.HeapSnapshot.prototype.get totalSize):
(WebInspector.HeapSnapshot.prototype._dominatedNodesOfNode):
(WebInspector.HeapSnapshot.prototype.aggregatesForDiff):
(WebInspector.HeapSnapshot.prototype._calculateObjectToWindowDistance):
(WebInspector.HeapSnapshot.prototype._buildAggregates):
(WebInspector.HeapSnapshot.prototype._calculateClassesRetainedSize):
(WebInspector.HeapSnapshot.prototype._sortAggregateIndexes):
(WebInspector.HeapSnapshot.prototype._markInvisibleEdges):
(WebInspector.HeapSnapshot.prototype._markDetachedDOMTreeNodes):
(WebInspector.HeapSnapshot.prototype._markQueriableHeapObjects):
(WebInspector.HeapSnapshot.prototype._calculateDiffForClass):
(WebInspector.HeapSnapshot.prototype._nodeForSnapshotObjectId):
(WebInspector.HeapSnapshot.prototype.nodeClassName):
(WebInspector.HeapSnapshot.prototype.dominatorIdsForNode):
(WebInspector.HeapSnapshot.prototype.createEdgesProvider):
(WebInspector.HeapSnapshot.prototype.createRetainingEdgesProvider):
(WebInspector.HeapSnapshotFilteredOrderedIterator.prototype._createIterationOrder):
(WebInspector.HeapSnapshotFilteredOrderedIterator.prototype.isEmpty):
(WebInspector.HeapSnapshotFilteredOrderedIterator.prototype.item):
(WebInspector.HeapSnapshotFilteredOrderedIterator.prototype.serializeItemsRange):
(WebInspector.HeapSnapshotEdgesProvider.prototype.serializeItem):
(WebInspector.HeapSnapshotEdgesProvider.prototype.sort.compareEdgeFieldName):
(WebInspector.HeapSnapshotEdgesProvider.prototype.sort.compareNodeField):
(WebInspector.HeapSnapshotNodesProvider):
(WebInspector.HeapSnapshotNodesProvider.prototype.nodePosition):
(WebInspector.HeapSnapshotNodesProvider.prototype.serializeItem):
(WebInspector.HeapSnapshotNodesProvider.prototype.sort.sortByNodeField):
* inspector/front-end/HeapSnapshotGridNodes.js:
* inspector/front-end/HeapSnapshotProxy.js:
(WebInspector.HeapSnapshotProviderProxy.prototype.isEmpty):
2012-05-22 Yury Semikhatsky <yurys@chromium.org>
Web Inspector: Debugger should correctly show catch scope
https://bugs.webkit.org/show_bug.cgi?id=87197
Reviewed by Vsevolod Vlasov.
* bindings/v8/DebuggerScript.js: In case of Catch scope the scope object
is a synthetic one which means that it contains all standard object properties.
When passing this object further we need to leave only properties available in
the catch block(only variable thar stores the exception in fact).
2012-05-23 Allan Sandfeld Jensen <allan.jensen@nokia.com>
[Qt] FloatSize Qt Support
https://bugs.webkit.org/show_bug.cgi?id=87237
Reviewed by Andreas Kling.
No new functionality. No new tests.
* Target.pri:
* platform/graphics/FloatSize.h:
(FloatSize):
* platform/graphics/qt/FloatSizeQt.cpp: Added.
(WebCore::FloatSize::FloatSize):
(WebCore::FloatSize::operator QSizeF):
2012-05-23 Alexei Filippov <alexeif@chromium.org>
Web Inspector: Enable HeapSnapshotRealWorker in production mode.
https://bugs.webkit.org/show_bug.cgi?id=87240
Reviewed by Yury Semikhatsky.
* inspector/front-end/HeapSnapshotProxy.js:
(WebInspector.HeapSnapshotWorker):
2012-05-23 'Pavel Feldman' <pfeldman@chromium.org>
Not reviewed: inspector's Find is not consuming event as of r117974.
* inspector/front-end/AdvancedSearchController.js:
(WebInspector.AdvancedSearchController.prototype.handleShortcut):
* inspector/front-end/SearchController.js:
(WebInspector.SearchController.prototype.handleShortcut):
2012-05-23 Pavel Feldman <pfeldman@chromium.org>
Web Inspector: enable drag-an-drop from the navigator panel.
https://bugs.webkit.org/show_bug.cgi?id=87238
Reviewed by Vsevolod Vlasov.
Making tree item draggable.
* inspector/front-end/NavigatorView.js:
(WebInspector.NavigatorSourceTreeElement.prototype.onattach):
(WebInspector.NavigatorSourceTreeElement.prototype._onmousedown.callback):
(WebInspector.NavigatorSourceTreeElement.prototype._onmousedown):
(WebInspector.NavigatorSourceTreeElement.prototype._ondragstart):
2012-05-23 Ryosuke Niwa <rniwa@webkit.org>
WebKit spends ~20% of time in HTMLTextAreaElement::defaultValue() when opening a review page
https://bugs.webkit.org/show_bug.cgi?id=86813
Reviewed by Eric Seidel.
Replace String and calls to String::operator+= by StringBuilder and calls to append respectively.
While there are other optimizations we can implement such as not replacing all nodes in the shadow DOM,
or not serializing the value when detecting whether the value had changed, they require considerable
amount of refactoring and this change alone seems to buy us a lot so let us defer those 'til later time.
Also avoid the redundant call to setValue in childrenChanged when the value is dirty.
This should halve the time we spend in setting the value.
* html/HTMLTextAreaElement.cpp:
(WebCore::HTMLTextAreaElement::childrenChanged):
(WebCore::HTMLTextAreaElement::defaultValue):
2012-05-23 Alice Boxhall <aboxhall@chromium.org>
AX: Contenteditable tables should always be exposed as data tables
https://bugs.webkit.org/show_bug.cgi?id=87052
Reviewed by Chris Fleizach.
Tested in accessibility/table-detection.html.
* accessibility/AccessibilityTable.cpp:
(WebCore::AccessibilityTable::isDataTable): Return true if the node is contenteditable.
2012-05-23 Ryosuke Niwa <rniwa@webkit.org>
Fix a typo per Darin's comment.
* rendering/RenderBlockLineLayout.cpp:
(WebCore::constructBidiRuns):
2012-05-23 Kentaro Hara <haraken@chromium.org>
[V8] Pass Isolate to v8::Null() in custom bindings (Part 2)
https://bugs.webkit.org/show_bug.cgi?id=87209
Reviewed by Adam Barth.
The objective is to pass Isolate around in V8 bindings.
This patch passes Isolate to v8::Null() in custom bindings.
No tests. No change in behavior.
* bindings/v8/custom/V8HTMLElementCustom.cpp:
(WebCore::toV8Object):
(WebCore::toV8):
* bindings/v8/custom/V8HistoryCustom.cpp:
(WebCore::V8History::stateAccessorGetter):
* bindings/v8/custom/V8IDBAnyCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8IDBKeyCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ImageDataCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Int16ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Int32ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Int8ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8LocationCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8MessageEventCustom.cpp:
(WebCore::V8MessageEvent::dataAccessorGetter):
* bindings/v8/custom/V8NamedNodeMapCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8NodeCustom.cpp:
(WebCore::V8Node::insertBeforeCallback):
(WebCore::V8Node::replaceChildCallback):
(WebCore::V8Node::removeChildCallback):
(WebCore::V8Node::appendChildCallback):
(WebCore::toV8Slow):
* bindings/v8/custom/V8PopStateEventCustom.cpp:
(WebCore::V8PopStateEvent::stateAccessorGetter):
* bindings/v8/custom/V8SQLResultSetRowListCustom.cpp:
(WebCore::V8SQLResultSetRowList::itemCallback):
* bindings/v8/custom/V8SVGDocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8SVGElementCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8SVGPathSegCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ScriptProfileCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ScriptProfileNodeCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8StyleSheetCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8TrackEventCustom.cpp:
(WebCore::V8TrackEvent::trackAccessorGetter):
* bindings/v8/custom/V8Uint16ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Uint32ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Uint8ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Uint8ClampedArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
(WebCore::toV8Object):
(WebCore::V8WebGLRenderingContext::getAttachedShadersCallback):
(WebCore::V8WebGLRenderingContext::getSupportedExtensionsCallback):
* bindings/v8/custom/V8WorkerContextCustom.cpp:
(WebCore::toV8):
2012-05-23 Kentaro Hara <haraken@chromium.org>
[V8] Pass Isolate to v8::Null() in custom bindings (Part 1)
https://bugs.webkit.org/show_bug.cgi?id=87207
Reviewed by Adam Barth.
The objective is to pass Isolate around in V8 bindings.
This patch passes Isolate to v8::Null() in custom bindings.
Also this patch makes V8DeviceMotionEventCustom::createAccelerationObject()
and V8DeviceMotionEventCustom::createRotationRateObject() static.
No tests. No change in behavior.
* bindings/v8/custom/V8BlobCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8CSSRuleCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8CSSStyleSheetCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8CSSValueCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8ClipboardCustom.cpp:
(WebCore::V8Clipboard::typesAccessorGetter):
* bindings/v8/custom/V8CoordinatesCustom.cpp:
(WebCore::V8Coordinates::altitudeAccessorGetter):
(WebCore::V8Coordinates::altitudeAccuracyAccessorGetter):
(WebCore::V8Coordinates::headingAccessorGetter):
(WebCore::V8Coordinates::speedAccessorGetter):
* bindings/v8/custom/V8DOMStringMapCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DOMTokenListCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DOMWindowCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DataViewCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DeviceMotionEventCustom.cpp:
(WebCore::V8DeviceMotionEvent::accelerationAccessorGetter):
(WebCore::V8DeviceMotionEvent::accelerationIncludingGravityAccessorGetter):
(WebCore::V8DeviceMotionEvent::rotationRateAccessorGetter):
(WebCore::V8DeviceMotionEvent::intervalAccessorGetter):
* bindings/v8/custom/V8DeviceOrientationEventCustom.cpp:
(WebCore::V8DeviceOrientationEvent::alphaAccessorGetter):
(WebCore::V8DeviceOrientationEvent::betaAccessorGetter):
(WebCore::V8DeviceOrientationEvent::gammaAccessorGetter):
(WebCore::V8DeviceOrientationEvent::absoluteAccessorGetter):
* bindings/v8/custom/V8DocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8DocumentLocationCustom.cpp:
(WebCore::V8Document::locationAccessorGetter):
* bindings/v8/custom/V8EntryCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8EntrySyncCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8EventCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Float32ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8Float64ArrayCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8HTMLCanvasElementCustom.cpp:
(WebCore::V8HTMLCanvasElement::getContextCallback):
* bindings/v8/custom/V8HTMLDocumentCustom.cpp:
(WebCore::toV8):
2012-05-22 MORITA Hajime <morrita@google.com>
[Refactoring] ElementShadow::m_needsRedistributing should be on ContentDistributor
https://bugs.webkit.org/show_bug.cgi?id=87216
Reviewed by Kentaro Hara.
Moved ElementShadow::m_needsRedistributing to ContentDistributor::m_needsRedistributing
No new tests. No behavior change.
* dom/ElementShadow.cpp:
(WebCore::ElementShadow::recalcStyle):
(WebCore::ElementShadow::needsRedistributing):
(WebCore::ElementShadow::setNeedsRedistributing):
* dom/ElementShadow.h:
(ElementShadow):
* dom/ShadowRoot.h: Removed outdated declarations.
* html/shadow/ContentDistributor.cpp:
(WebCore::ContentDistributor::ContentDistributor):
* html/shadow/ContentDistributor.h:
(WebCore::ContentDistributor::needsRedistributing):
(WebCore::ContentDistributor::setNeedsRedistributing):
(WebCore::ContentDistributor::clearNeedsRedistributing):
2012-05-22 Kentaro Hara <haraken@chromium.org>
[V8] Pass Isolate to v8::Null() in CodeGeneratorV8.pm
https://bugs.webkit.org/show_bug.cgi?id=87202
Reviewed by Adam Barth.
The objective is to pass Isolate around in V8 bindings.
This patch passes Isolate to v8::Null() in CodeGeneratorV8.pm.
No tests. No change in behavior.
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateNormalAttrGetter):
(GenerateFunctionCallback):
(NativeToJSValue):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::TestObjV8Internal::contentDocumentAttrGetter):
(WebCore::TestObjV8Internal::getSVGDocumentCallback):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.cpp:
(WebCore::TestSerializedScriptValueInterfaceV8Internal::valueAttrGetter):
(WebCore::TestSerializedScriptValueInterfaceV8Internal::readonlyValueAttrGetter):
(WebCore::TestSerializedScriptValueInterfaceV8Internal::cachedValueAttrGetter):
(WebCore::TestSerializedScriptValueInterfaceV8Internal::cachedReadonlyValueAttrGetter):
2012-05-22 Andreas Kling <kling@webkit.org>
Short-circuit Element::attrIfExists() when the Element has no Attr list.
<http://webkit.org/b/87214>
Reviewed by Ryosuke Niwa.
This function was lukewarm on Dromaeo/dom-attr (0.4% of samples.)
Added an early return if !hasAttrList() to avoid the function call overhead of
ElementAttributeData::attrIfExists().
* dom/Element.cpp:
(WebCore::Element::attrIfExists):
2012-05-22 Hayato Ito <hayato@chromium.org>
Remove an assertion since an event's target might be a shadow root if a text node is clicked.
https://bugs.webkit.org/show_bug.cgi?id=87072
Reviewed by Dimitri Glazkov.
Tests: fast/dom/shadow/shadow-dom-event-dispatching.html
* dom/ComposedShadowTreeWalker.cpp:
(WebCore::ComposedShadowTreeWalker::ComposedShadowTreeWalker):
2012-05-22 Dongwoo Im <dw.im@samsung.com>
[EFL] Implements the registerProtocolHandler method and option.
https://bugs.webkit.org/show_bug.cgi?id=73638
Reviewed by Andreas Kling.
The registerProtocolHandler() method allows Web sites to register themselves
as possible handlers for particular schemes.
http://dev.w3.org/html5/spec/Overview.html#dom-navigator-registerprotocolhandler
* CMakeLists.txt: Add page/NavigatorRegisterProtocolHandler.idl into WebCore_IDL_FILES.
2012-05-22 Kentaro Hara <haraken@chromium.org>
[V8][Performance] Pass Isolate to v8::Null() in V8Node::toV8()
https://bugs.webkit.org/show_bug.cgi?id=87193
Reviewed by Adam Barth.
This patch improves the performance of Bindings/undefined-first-child.html by 58.7%,
Bindings/undefined-get-element-by-id.html by 26.8%,
Dromaeo/dom-query.html(getElementById (not in document)) by 9.8%,
and Dromaeo/dom-query.html(getElementsByName (not in document)) by 2.1%.
These benchmarks test the performance of accessing an undefined Node object.
Performance test results in my Linux desktop:
[Bindings/undefined-first-child.html]
- without my patch
RESULT Bindings: undefined-first-child= 481.843775891 runs/s
median= 474.934036939 runs/s, stdev= 17.7753278263 runs/s, min= 471.204188482 runs/s, max= 534.629404617 runs/s
- with my patch
RESULT Bindings: undefined-first-child= 764.720531908 runs/s
median= 803.515696424 runs/s, stdev= 54.7402312963 runs/s, min= 688.806888069 runs/s, max= 807.061790668 runs/s
[Bindings/undefined-get-element-by-id.html]
- without my patch
RESULT Bindings: undefined-get-element-by-id= 236.642705579 runs/s
median= 234.527786832 runs/s, stdev= 5.77090297761 runs/s, min= 231.362467866 runs/s, max= 246.305418719 runs/s
- with my patch
RESULT Bindings: undefined-get-element-by-id= 300.031032067 runs/s
median= 303.797468354 runs/s, stdev= 6.99060409266 runs/s, min= 287.581699346 runs/s, max= 304.955527319 runs/s
[Dromaeo/dom-query.html(getElementById (not in document))]
- without my patch
2855.80runs/s ±0.30%
- with my patch
3136.20runs/s ±0.22%
[Dromaeo/dom-query.html(getElementsByName (not in document))]
- without my patch
2198.00runs/s ±0.50%
- with my patch
2243.40runs/s ±3.39%
Currently v8::Null() looks up Isolate, which is a bottleneck of
V8Node::toV8() for undefined Nodes. This patch removes the Isolate
look-up by passing Isolate to v8::Null(), like v8::Null(isolate).
* bindings/scripts/CodeGeneratorV8.pm: Modified as described above.
(GenerateHeader):
* bindings/scripts/test/V8/V8TestActiveDOMObject.h: Updated run-bindings-tests results.
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestCustomNamedGetter.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestEventConstructor.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestEventTarget.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestException.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestInterface.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestMediaQueryListListener.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestNamedConstructor.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestNode.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestObj.h:
(WebCore::toV8):
* bindings/scripts/test/V8/V8TestSerializedScriptValueInterface.h:
(WebCore::toV8):
2012-05-22 MORITA Hajime <morrita@google.com>
REGRESSION(r116629) tests for <style scoped> is crashing
https://bugs.webkit.org/show_bug.cgi?id=87191
Reviewed by Kent Tamura.
unregisterWithScopingNode() and registerWithScopingNode() doesn't follow
refactoring around removedFrom(), insertedInto() and willRemove().
This change aligns them to such recent changes.
No new tests. Covered by existing tests.
* html/HTMLStyleElement.cpp:
(WebCore::HTMLStyleElement::unregisterWithScopingNode)
(WebCore::HTMLStyleElement::insertedInto):
- Allowed registerWithScopingNode() only for in-document traversal.
That is how it originally did.
(WebCore::HTMLStyleElement::removedFrom):
- Gave ex-parent to unregisterWithScopingNode().
* html/HTMLStyleElement.h:
(WebCore::HTMLStyleElement::unregisterWithScopingNode):
(HTMLStyleElement):
2012-05-21 Shawn Singh <shawnsingh@chromium.org>
Improve W3C conformance of backface-visibility
https://bugs.webkit.org/show_bug.cgi?id=84195
Reviewed by Adrienne Walker.
Tests: compositing/backface-visibility/backface-visibility-3d.html
compositing/backface-visibility/backface-visibility-hierarchical-transform.html
compositing/backface-visibility/backface-visibility-non3d.html
compositing/backface-visibility/backface-visibility-simple.html
Unit Tests: CCLayerTreeHostCommonTest::verifyBackFaceCullingWithoutPreserves3d
CCLayerTreeHostCommonTest::verifyBackFaceCullingWithPreserves3d
Also updated a few other layout and unit tests.
This patch updates chromium to conform to the W3C spec on CSS
transforms, Section 12 about backface visibility. This patch also
updates layout tests to reflect the latest W3C spec. In
particular, rule #1 on section 12, http://dev.w3.org/csswg/css3-transforms/#backface-visibility-property
states which transforms should be used to determine backface-visibility.
Because chromium has an additional quirk of the concept of render
surfaces, unit tests needed to be updated as well.
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::layerIsRootOfNewRenderingContext):
(WebCore):
(WebCore::isLayerBackFaceVisible):
(WebCore::isSurfaceBackFaceVisible):
(WebCore::calculateVisibleLayerRect):
(WebCore::layerShouldBeSkipped):
(WebCore::calculateDrawTransformsAndVisibilityInternal):
2012-05-22 Sheriff Bot <webkit.review.bot@gmail.com>
Unreviewed, rolling out r118095.
http://trac.webkit.org/changeset/118095
https://bugs.webkit.org/show_bug.cgi?id=87201
"Causing crashes on Mac in
WebCore::StyleResolver::adjustRenderStyle (Requested by
sundiamonde on #webkit).
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(Settings):
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::InternalSettings):
(WebCore::InternalSettings::restoreTo):
* testing/InternalSettings.h:
(InternalSettings):
* testing/InternalSettings.idl:
2012-05-22 Levi Weintraub <leviw@chromium.org>
NULL ptr in WebCore::RenderBlock::layoutRunsAndFloatsInRange
https://bugs.webkit.org/show_bug.cgi?id=77786
Reviewed by Ryosuke Niwa.
InlineBidiResolver adds one fake TextRun for isolated inlines in the process of creating the
list of TextRuns to send to the UBA. After the UBA has been run and the TextRuns reordered,
we re-run InlineBidiResolver rooted in the isolate and replace the fake run with those
generated by the subsequent pass by calling the method BidiRunList::replaceRunWithRuns.
This method assumes there are runs to replace the fake run with.
Positioned inline children are ignored when creating TextRuns, so when an isolated inline
has only positioned children we end up with an empty set of runs to pass to replaceRunWithRuns.
Ideally, we'd remove the fake run and not replace it with anything, but BidiRunList keeps
a pointer to the logically last run, which we're unable to easily re-determine after the UBA
has been run. Instead, we leave the fake run in the list and simply avoid calling
replaceRunWithRuns when we don't have any replacement runs.
Test: fast/block/line-layout/crash-in-isolate-with-positioned-child.html
* rendering/RenderBlockLineLayout.cpp:
(WebCore::constructBidiRuns):
2012-05-22 Kent Tamura <tkent@chromium.org>
[V8] Refactor generation code for non-standard functions
https://bugs.webkit.org/show_bug.cgi?id=87185
Reviewed by Kentaro Hara.
No behavior change. Just a refactoring.
* bindings/scripts/CodeGeneratorV8.pm:
(IsStandardFunction): Introduce a new function to check if a
function can be in a batch registration.
(GenerateNonStandardFunction): Factor out a code to generate
non-standard functions.
(GenerateImplementation):
Use IsStandardFunction and GenerateNonStandardFunction.
2012-05-22 Hayato Ito <hayato@chromium.org>
Make ComposedShadowTreeWalker traverse inactive insertion points correctly.
https://bugs.webkit.org/show_bug.cgi?id=86830
Reviewed by Dimitri Glazkov.
Fixed InsertionPoint::isActive() issue, which may return true
wrongly even if the insertion point is not in Shadow DOM subtree.
Now ComposedShadowTreeWalker can traverse inactive insertion
points correctly using InsertionPoint::isActive().
Test: fast/dom/shadow/composed-shadow-tree-walker.html
* dom/ComposedShadowTreeWalker.cpp:
(WebCore::ComposedShadowTreeWalker::traverseNode):
(WebCore::ComposedShadowTreeWalker::escapeFallbackContentElement):
(WebCore::ComposedShadowTreeWalker::traverseNodeEscapingFallbackContents):
* html/shadow/InsertionPoint.cpp:
(WebCore::InsertionPoint::isActive):
* html/shadow/InsertionPoint.h:
(WebCore::isActiveInsertionPoint):
(WebCore):
2012-05-22 Mark Pilgrim <pilgrim@chromium.org>
[Chromium] Call canAccelerate2dCanvas directly
https://bugs.webkit.org/show_bug.cgi?id=86893
Reviewed by James Robinson.
Part of a refactoring series. See tracking bug 82948.
* html/HTMLCanvasElement.cpp:
(WebCore::HTMLCanvasElement::shouldAccelerate):
* platform/chromium/PlatformSupport.h:
2012-05-22 Jason Liu <jason.liu@torchmobile.com.cn>
[BlackBerry] Possible to clobber httponly cookie.
https://bugs.webkit.org/show_bug.cgi?id=86067
Reviewed by Rob Buis.
If a cookie is set by javaScript and there is already a same httpOnly cookie in cookieManager,
we should reject it. If it has a httpOnly property, we reject it, too.
Test: http/tests/cookies/js-get-and-set-http-only-cookie.php
* platform/blackberry/CookieJarBlackBerry.cpp:
(WebCore::setCookies):
* platform/blackberry/CookieManager.cpp:
(WebCore::CookieManager::setCookies):
(WebCore::CookieManager::shouldRejectNotHttpCookie):
(WebCore):
* platform/blackberry/CookieManager.h:
2012-05-22 Dana Jansens <danakj@chromium.org>
[chromium] Don't drop children of a RenderSurface from the renderSurfaceLayerList when then position of the surface in its clipRect is not known
https://bugs.webkit.org/show_bug.cgi?id=87181
Reviewed by Adrienne Walker.
We want to avoid dropping things from the renderSurfaceLayerList on main
thread when they may be in the renderSurfaceLayerList on impl thread due
to animation. A render surface would drop all its children if its
contentRect was clipped to become empty by its parent. But when the
surface is being animated, then we can't be sure how its parent will
clip the surface's content, so we don't clip it at all.
Chromium bug: http://crbug.com/128739
Unit test: CCLayerTreeHostCommonTest.verifyClipRectCullsSurfaceWithoutVisibleContent
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::calculateDrawTransformsAndVisibilityInternal):
2012-05-22 Kangil Han <kangil.han@samsung.com>
[EFL][DRT] Implement touch event
https://bugs.webkit.org/show_bug.cgi?id=86720
Reviewed by Hajime Morita.
Currently EFL DRT doesn't support touch event.
Therefore, this patch enabled it and implemented some eventSender function callbacks.
* platform/PlatformTouchEvent.h:
(PlatformTouchEvent): Implement meta state.
* platform/efl/PlatformTouchEventEfl.cpp:
(WebCore::PlatformTouchEvent::PlatformTouchEvent): Implement meta state.
2012-05-22 Chris Rogers <crogers@google.com>
AudioNodeOutput::disconnectAllParams() must take care when iterating through its parameter list
https://bugs.webkit.org/show_bug.cgi?id=87190
Reviewed by Kenneth Russell.
No new tests. Covered by existing tests.
* Modules/webaudio/AudioNodeOutput.cpp:
(WebCore::AudioNodeOutput::disconnectAllParams):
2012-05-22 MORITA Hajime <morrita@google.com>
<shadow> and <content> should be hidden behind ContextEnabledFeatures
https://bugs.webkit.org/show_bug.cgi?id=87063
This change repalces "runtimeConditional" tag definition property
with new "contextConditional" property, which refers
ContextEnabledFeatures instead of RuntimeEnabledFeatures.
Reviewed by Dimitri Glazkov.
No new tests. DRT cannot test this. No behaviour change on DRT.
* dom/make_names.pl:
(defaultTagPropertyHash):
(printConstructorInterior):
(printFactoryCppFile):
(printWrapperFunctions):
(printWrapperFactoryCppFile):
* html/HTMLTagNames.in:
2012-05-22 Kent Tamura <tkent@chromium.org>
Fix a crash in HTMLFormControlElement::disabled().
https://bugs.webkit.org/show_bug.cgi?id=86534
Reviewed by Hajime Morita.
Node::removedFrom() was not called in some cases. This patch fixes it.
No new tests. It's almost impossible to make a reliable test.
* dom/ContainerNodeAlgorithms.h:
Remove inDocument() check so that Node::removedFrom() is called even if
a parent node in an orphan tree is deleted but a child node remains,
2012-05-22 James Robinson <jamesr@chromium.org>
Add a Setting to make position:fixed form a new stacking context
https://bugs.webkit.org/show_bug.cgi?id=87186
Reviewed by Adam Barth.
position:fixed elements form a new stacking context in several mobile WebKit ports and in the Qt port always. To
be consistent across devices, we want to try turning it on everywhere. This will likely cause compatibility
issues, so we need a runtime setting in order to stage the change carefully.
Tests: fast/block/positioning/fixed-position-stacking-context.html
fast/block/positioning/fixed-position-stacking-context2.html
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(WebCore::Settings::setFixedPositionCreatesStackingContext):
(WebCore::Settings::fixedPositionCreatesStackingContext):
(Settings):
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::InternalSettings):
(WebCore::InternalSettings::restoreTo):
(WebCore::InternalSettings::setFixedPositionCreatesStackingContext):
(WebCore):
* testing/InternalSettings.h:
(InternalSettings):
* testing/InternalSettings.idl:
2012-05-22 Adam Barth <abarth@webkit.org>
RenderInline::absoluteRects does some incorrect layout math
https://bugs.webkit.org/show_bug.cgi?id=87036
Reviewed by Eric Seidel.
http://trac.webkit.org/changeset/88297/trunk/Source/WebCore/rendering/RenderInline.cpp
contained a typo. We should accumulate offsets, not add sizes to
offsets.
Test: fast/inline/boundingBox-with-continuation.html
* rendering/RenderInline.cpp:
(WebCore::RenderInline::absoluteRects):
(WebCore):
2012-05-22 Dana Jansens <danakj@chromium.org>
[chromium] Don't force the visibleLayerRect to be empty for animating layers whose front face is not visible
https://bugs.webkit.org/show_bug.cgi?id=86886
Reviewed by Adrienne Walker.
Previously we would force the visibleLayerRect to be empty for these
animating layers, and then special case the pre-painting code to
paint them if they were within some size limit, to avoid having the
layer steal too much memory.
Instead, just use the visibleLayerRect given to the layer. Since the
transform flips the layer such that its back face is visible, it will
also compute a visibleLayerRect in the front face of the layer, and
its size will be bounded by the viewport. This will cause the
layer to be painted with visible priority instead of prepainted, but
ensures the visible parts of the layer is painted and ready to go when
the animation flips the layer around.
* platform/graphics/chromium/cc/CCLayerTreeHostCommon.cpp:
(WebCore::calculateVisibleLayerRect):
2012-05-22 Jer Noble <jer.noble@apple.com>
REGRESSION (r98359): Video does not render on http://panic.com/dietcoda/
https://bugs.webkit.org/show_bug.cgi?id=87171
Reviewed by Maciej Stachowiak.
No new tests; behavior is very timing specific and only occurs on a subset of all platforms.
Instead of calling prepareForRendering() from metadataLoaded(), which may fail and cause subsequent
calls to prepareForRendering() to short circuit, call createVideoLayer() directly, which achieves
the original goals of the fix for http://webkit.org/b/70448, but without breaking subsequent calls
to prepareForRendering() if called at the wrong time.
* platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
(WebCore::MediaPlayerPrivateAVFoundation::metadataLoaded):
2012-05-22 Tim Horton <timothy_horton@apple.com>
Add a quirk for applications that depend on the relative ordering of progressCompleted/didFinishLoad
https://bugs.webkit.org/show_bug.cgi?id=87178
<rdar://problem/11468434>
Reviewed by Maciej Stachowiak.
Some applications depend on the relative ordering of progressCompleted/didFinishLoad, which was changed
to be more correct in http://trac.webkit.org/changeset/94105. For applications built before 94105, we can
provide the old behavior. For the time being, this will only apply to Mail.app.
No new tests, will not affect behavior for any application except Mail.
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::checkLoadCompleteForThisFrame):
* page/Settings.cpp:
(WebCore::Settings::Settings):
* page/Settings.h:
(WebCore::Settings::setNeedsDidFinishLoadOrderQuirk):
(WebCore::Settings::needsDidFinishLoadOrderQuirk):
2012-05-22 Alexis Menard <alexis.menard@openbossa.org>
Move some CSS regions properties to CSSParser::isValidKeywordPropertyAndValue.
https://bugs.webkit.org/show_bug.cgi?id=87169
Reviewed by Tony Chang.
Move some CSS regions properties to the fast path implemented by
CSSParser::isValidKeywordPropertyAndValue.
No new tests : Existing tests should cover.
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
(WebCore::isKeywordPropertyID):
(WebCore::CSSParser::parseValue):
2012-05-22 Emil A Eklund <eae@chromium.org>
Represents margins as box and simplify writing mode logic
https://bugs.webkit.org/show_bug.cgi?id=86952
Reviewed by Eric Seidel.
Add a new class, FractionalLayoutBox, that represents the four sides of a
box and change RenderBox to use it to represent margins.
The new box class is intended to be used to represent things like
margins, paddings and borders where the current rect class isn't a good
fit. It also encapsulated a lot of the horizontal/vertical writing mode
and bidi logic allowing it to be reused.
No new tests, no change in functionality.
* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* platform/graphics/FractionalLayoutBoxExtent.cpp: Added.
* platform/graphics/FractionalLayoutBoxExtent.h: Added.
* rendering/LayoutTypes.h:
* rendering/RenderBlock.cpp:
* rendering/RenderBox.cpp:
* rendering/RenderBox.h:
* rendering/RenderScrollbarPart.cpp:
* rendering/RenderScrollbarPart.h:
2012-05-22 Kentaro Hara <haraken@chromium.org>
[V8] Replace 'throwTypeError(); return Undefined();' with 'return throwTypeError();'
https://bugs.webkit.org/show_bug.cgi?id=87106
Reviewed by Adam Barth.
This patch replaces
throwTypeError();
return v8::Undefined();
with
return throwTypeError();
No tests. No change in behavior.
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateOverloadedFunctionCallback):
(GenerateParametersCheck):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::TestObjV8Internal::overloadedMethodCallback):
(WebCore::TestObjV8Internal::overloadedMethod1Callback):
2012-05-22 Dan Bernstein <mitz@apple.com>
REGRESSION (r105513): CSS max-width is not applied to SVG element
https://bugs.webkit.org/show_bug.cgi?id=87050
Reviewed by Darin Adler.
Tests: svg/css/max-height.html
svg/css/max-width.html
* rendering/svg/RenderSVGRoot.cpp:
(WebCore::RenderSVGRoot::computeReplacedLogicalWidth): Changed to also call into the base
class implementation of this function if max-width was specified.
(WebCore::RenderSVGRoot::computeReplacedLogicalHeight): Changed to call into the bas class
implementation of this function if either height or max-height were specified.
2012-05-22 Kentaro Hara <haraken@chromium.org>
[V8] Pass Isolate to throwTypeError()
https://bugs.webkit.org/show_bug.cgi?id=87111
Reviewed by Adam Barth.
The objective is to pass Isolate around in V8 bindings.
This patch passes Isolate to throwTypeError() in custom bindings.
No tests. No change in behavior.
* bindings/v8/custom/V8ArrayBufferCustom.cpp:
(WebCore::V8ArrayBuffer::constructorCallback):
* bindings/v8/custom/V8ArrayBufferViewCustom.h:
(WebCore::constructWebGLArrayWithArrayBufferArgument):
(WebCore::constructWebGLArray):
* bindings/v8/custom/V8AudioBufferSourceNodeCustom.cpp:
(WebCore::V8AudioBufferSourceNode::bufferAccessorSetter):
* bindings/v8/custom/V8AudioContextCustom.cpp:
(WebCore::V8AudioContext::constructorCallback):
* bindings/v8/custom/V8BlobCustom.cpp:
(WebCore::V8Blob::constructorCallback):
* bindings/v8/custom/V8ClipboardCustom.cpp:
(WebCore::V8Clipboard::setDragImageCallback):
* bindings/v8/custom/V8DOMFormDataCustom.cpp:
(WebCore::V8DOMFormData::constructorCallback):
* bindings/v8/custom/V8DataViewCustom.cpp:
(WebCore::V8DataView::constructorCallback):
* bindings/v8/custom/V8HTMLDocumentCustom.cpp:
(WebCore::V8HTMLDocument::openCallback):
* bindings/v8/custom/V8HTMLImageElementConstructor.cpp:
(WebCore::v8HTMLImageElementConstructorCallback):
* bindings/v8/custom/V8HTMLInputElementCustom.cpp:
(WebCore::V8HTMLInputElement::selectionStartAccessorGetter):
(WebCore::V8HTMLInputElement::selectionStartAccessorSetter):
(WebCore::V8HTMLInputElement::selectionEndAccessorGetter):
(WebCore::V8HTMLInputElement::selectionEndAccessorSetter):
(WebCore::V8HTMLInputElement::selectionDirectionAccessorGetter):
(WebCore::V8HTMLInputElement::selectionDirectionAccessorSetter):
(WebCore::V8HTMLInputElement::setSelectionRangeCallback):
* bindings/v8/custom/V8HTMLMediaElementCustom.cpp:
(WebCore::V8HTMLMediaElement::controllerAccessorSetter):
* bindings/v8/custom/V8InjectedScriptHostCustom.cpp:
(WebCore::V8InjectedScriptHost::inspectedObjectCallback):
* bindings/v8/custom/V8IntentConstructor.cpp:
(WebCore::V8Intent::constructorCallback):
* bindings/v8/custom/V8MessageChannelConstructor.cpp:
(WebCore::V8MessageChannel::constructorCallback):
* bindings/v8/custom/V8NotificationCenterCustom.cpp:
(WebCore::V8NotificationCenter::requestPermissionCallback):
* bindings/v8/custom/V8SQLResultSetRowListCustom.cpp:
(WebCore::V8SQLResultSetRowList::itemCallback):
* bindings/v8/custom/V8WebKitMutationObserverCustom.cpp:
(WebCore::V8WebKitMutationObserver::constructorCallback):
* bindings/v8/custom/V8WebKitPointConstructor.cpp:
(WebCore::V8WebKitPoint::constructorCallback):
* bindings/v8/custom/V8WebSocketCustom.cpp:
(WebCore::V8WebSocket::constructorCallback):
* bindings/v8/custom/V8XMLHttpRequestConstructor.cpp:
(WebCore::V8XMLHttpRequest::constructorCallback):
2012-05-22 Kentaro Hara <haraken@chromium.org>
[V8] Replace 'setDOMException(); return Undefined();' with 'return setDOMException();'
https://bugs.webkit.org/show_bug.cgi?id=87102
Reviewed by Adam Barth.
This patch replaces
setDOMException();
return v8::Undefined();
with
return setDOMException();
In addition, just in case, this patch replaces
if (...)
setDOMException();
return v8::Undefined();
with
if (...)
return setDOMException();
return v8::Undefined();
because people might insert some code just above
'return v8::Undefined()' in the future.
No tests. No change in behavior.
* bindings/v8/custom/V8ArrayBufferViewCustom.h:
(WebCore::constructWebGLArrayWithArrayBufferArgument):
(WebCore::setWebGLArrayHelper):
* bindings/v8/custom/V8DOMWindowCustom.cpp:
(WebCore::WindowSetTimeoutImpl):
* bindings/v8/custom/V8DataViewCustom.cpp:
(WebCore::V8DataView::getInt8Callback):
(WebCore::V8DataView::getUint8Callback):
(WebCore::V8DataView::setInt8Callback):
(WebCore::V8DataView::setUint8Callback):
* bindings/v8/custom/V8DirectoryEntrySyncCustom.cpp:
(WebCore::V8DirectoryEntrySync::getDirectoryCallback):
(WebCore::V8DirectoryEntrySync::getFileCallback):
* bindings/v8/custom/V8HTMLOptionsCollectionCustom.cpp:
(WebCore::V8HTMLOptionsCollection::addCallback):
* bindings/v8/custom/V8NodeCustom.cpp:
(WebCore::V8Node::insertBeforeCallback):
(WebCore::V8Node::replaceChildCallback):
(WebCore::V8Node::removeChildCallback):
(WebCore::V8Node::appendChildCallback):
* bindings/v8/custom/V8SVGLengthCustom.cpp:
(WebCore::V8SVGLength::valueAccessorGetter):
(WebCore::V8SVGLength::convertToSpecifiedUnitsCallback):
* bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
(WebCore::getObjectParameter):
(WebCore::V8WebGLRenderingContext::getFramebufferAttachmentParameterCallback):
(WebCore::V8WebGLRenderingContext::getParameterCallback):
(WebCore::V8WebGLRenderingContext::getProgramParameterCallback):
(WebCore::V8WebGLRenderingContext::getShaderParameterCallback):
(WebCore::V8WebGLRenderingContext::getUniformCallback):
(WebCore::vertexAttribAndUniformHelperf):
(WebCore::uniformHelperi):
(WebCore::uniformMatrixHelper):
* bindings/v8/custom/V8WebKitMutationObserverCustom.cpp:
(WebCore::V8WebKitMutationObserver::observeCallback):
* bindings/v8/custom/V8XMLHttpRequestCustom.cpp:
(WebCore::V8XMLHttpRequest::responseAccessorGetter):
2012-05-22 Martin Robinson <mrobinson@igalia.com>
[TextureMapper] [Cairo] Implement repaint counters
https://bugs.webkit.org/show_bug.cgi?id=87062
Reviewed by Gustavo Noronha Silva.
No new tests. This just adds a debugging feature that shouldn't
affect the normal drawing of a page.
* platform/graphics/texmap/TextureMapperLayer.cpp:
(WebCore::TextureMapperLayer::updateBackingStore): Now, if repaint counters are
enabled, increment the repaint count and call drawRepaintCounter.
(WebCore::TextureMapperLayer::drawRepaintCounter): Added.
* platform/graphics/texmap/TextureMapperLayer.h:
(TextureMapperLayer): Added declaration for drawRepaintCounter.
2012-05-22 Simon Fraser <simon.fraser@apple.com>
Use the same animation begin time while updating compositing layers
https://bugs.webkit.org/show_bug.cgi?id=87080
Reviewed by Dean Jackson.
Tidied up some beginAnimationUpdate/endAnimationUpdate logic,
but further cleaning is required. We currently bracket some calls
to recaclcStyle() with begin/endAnimationUpdate, but not all.
We don't bracket layout, or compositing updates. This patch
adds bracketing around compositing updates, in order to be
able to add some assertions in a later patch that rely on
RenderLayer::currentTransform() giving consistent results,
since currentTransform() depends on the animation begin time.
Made AnimationController::beginAnimationUpdate() and
AnimationController::endAnimationUpdate() nestable by adding
a counter.
Added a stack-based class, AnimationUpdateBlock, which
calls begin/endAnimationUpdate.
Replaced calls to begin/endAnimationUpdate in Document::updateStyleIfNeeded()
and Document::styleResolverChanged() with AnimationUpdateBlock. Filed a bug
to cover other calls to recalcStyle which might need this (87159).
Removed a call to beginAnimationUpdate() in Document::implicitOpen(),
which was added in r42415. I believe that r57390 makes this unnecessary,
and was not able to reproduce the original bug after removing it. It
was the only unpaired beginAnimationUpdate call.
Made AnimationControllerPrivate::getAnimatedStyleForRenderer() not
use a new beginTime each time it was called, so its beginTime now
depends on some earlier beginAnimationUpdate.
Fixed CSSComputedStyleDeclaration::getPropertyCSSValue() to use
an AnimationUpdateBlock so that calls to computed style do
use the current time to compute an animating value (as before).
Not testable.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* dom/Document.cpp:
(WebCore::Document::updateStyleIfNeeded):
(WebCore::Document::implicitOpen):
(WebCore::Document::styleResolverChanged):
* page/animation/AnimationController.cpp:
(WebCore::AnimationControllerPrivate::getAnimatedStyleForRenderer):
(WebCore::AnimationController::AnimationController):
(WebCore::AnimationController::beginAnimationUpdate):
(WebCore::AnimationController::endAnimationUpdate):
* page/animation/AnimationController.h:
(AnimationController):
(WebCore):
(AnimationUpdateBlock):
(WebCore::AnimationUpdateBlock::AnimationUpdateBlock):
(WebCore::AnimationUpdateBlock::~AnimationUpdateBlock):
* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::updateCompositingLayers):
2012-05-22 Tony Chang <tony@chromium.org>
REGRESSION(r117613): Dromaeo/jslib-style-prototype regressed by 20%
https://bugs.webkit.org/show_bug.cgi?id=87142
Reviewed by Alexis Menard.
Move CSSPropertyDisplay back into isValidKeywordPropertyAndValue.
Pass in the parser context so we can check whether we're grid or not.
No new tests. Covered by Dromaeo/jslib-style-prototype perf test.
* css/CSSParser.cpp:
(WebCore::isValidKeywordPropertyAndValue):
(WebCore::isKeywordPropertyID):
(WebCore::parseKeywordValue):
(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::parseFont):
== Rolled over to ChangeLog-2012-05-22 ==