Node propagation doesn't handle PutScopedVar
https://bugs.webkit.org/show_bug.cgi?id=68713

Reviewed by Sam Weinig.

This was causing assertion failures.

* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95846 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Clarify wording in cross origin XSL tests and enable on all platforms
https://bugs.webkit.org/show_bug.cgi?id=68683

The condition for passing the test should be more clear now. Also, since the redirect case works on all platforms, the regular case should also work on all platforms.

Reviewed by Alexey Proskuryakov.

* http/tests/security/cross-origin-xsl-BLOCKED-expected.txt:
* http/tests/security/cross-origin-xsl-BLOCKED.html:
* http/tests/security/cross-origin-xsl-redirect-BLOCKED-expected.txt:
* http/tests/security/cross-origin-xsl-redirect-BLOCKED.html:
* platform/chromium-cg-mac/http/tests/security/cross-origin-xsl-BLOCKED-expected.txt:
* platform/chromium-cg-mac/http/tests/security/cross-origin-xsl-redirect-BLOCKED-expected.txt:
* platform/chromium-mac/http/tests/security/cross-origin-xsl-BLOCKED-expected.txt:
* platform/chromium-mac/http/tests/security/cross-origin-xsl-redirect-BLOCKED-expected.txt:
* platform/chromium-win/http/tests/security/cross-origin-xsl-BLOCKED-expected.txt:
* platform/chromium-win/http/tests/security/cross-origin-xsl-redirect-BLOCKED-expected.txt:
* platform/gtk/Skipped:
* platform/gtk/http/tests/security/cross-origin-xsl-BLOCKED-expected.txt:
* platform/gtk/http/tests/security/cross-origin-xsl-redirect-BLOCKED-expected.txt:
* platform/mac-wk2/http/tests/security/cross-origin-xsl-BLOCKED-expected.txt:
* platform/mac-wk2/http/tests/security/cross-origin-xsl-redirect-BLOCKED-expected.txt:
* platform/qt-wk2/http/tests/security/cross-origin-xsl-BLOCKED-expected.txt:
* platform/qt-wk2/http/tests/security/cross-origin-xsl-redirect-BLOCKED-expected.txt:
* platform/win-wk2/http/tests/security/cross-origin-xsl-BLOCKED-expected.txt:
* platform/win-wk2/http/tests/security/cross-origin-xsl-redirect-BLOCKED-expected.txt:
* platform/wk2/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95845 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Make sure to define OVERRIDE and FINAL for older builds of clang.

* wtf/Compiler.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95844 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Mark two more tests as flaky.

* platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95843 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Implement op_resolve_global in the DFG JIT
https://bugs.webkit.org/show_bug.cgi?id=68704

Reviewed by Oliver Hunt.

This is performance neutral, but increases coverage.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::ByteCodeParser):
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGNode.h:
(JSC::DFG::Node::hasIdentifier):
(JSC::DFG::Node::resolveInfoIndex):
* dfg/DFGOperations.cpp:
* dfg/DFGOperations.h:
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95840 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[GTK] Remove extra space at the beginning of the last line.
https://bugs.webkit.org/show_bug.cgi?id=68710

Reviewed by Gustavo Noronha Silva.

* platform/network/soup/SocketStreamHandleSoup.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95839 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Mark a test as flaky.

* platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95838 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-09-23  Mario Sanchez Prada  <msanchez@igalia.com>

        [GTK] Fix coding style issues in ResourceRequestSoup.cpp
        https://bugs.webkit.org/show_bug.cgi?id=68707

        Reviewed by Xan Lopez.

        This patch fixes some issues as reported by check-webkit-style.

        * platform/network/soup/ResourceRequestSoup.cpp:
        (WebCore::ResourceRequest::updateFromSoupMessage):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95837 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add OVERRIDE to overrides of virtual member functions in client classes
https://bugs.webkit.org/show_bug.cgi?id=68702

Reviewed by Darin Adler.

Disable the warning in Xcode and VS2005 for using override and final when compiling as C++98,
and remove WebEditorClient::isEditable, WebChromeClient::formDidFocus,
WebChromeClient::formDidBlur and WebFrameLoaderClient::dispatchDidLoadResourceByXMLHttpRequest since
they have been removed from their respective base classes.

* Configurations/Base.xcconfig:
* WebProcess/WebCoreSupport/WebChromeClient.cpp:
* WebProcess/WebCoreSupport/WebChromeClient.h:
* WebProcess/WebCoreSupport/WebContextMenuClient.h:
* WebProcess/WebCoreSupport/WebDatabaseManager.h:
* WebProcess/WebCoreSupport/WebDragClient.h:
* WebProcess/WebCoreSupport/WebEditorClient.cpp:
* WebProcess/WebCoreSupport/WebEditorClient.h:
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:
* WebProcess/WebCoreSupport/WebGeolocationClient.h:
* WebProcess/WebCoreSupport/WebInspectorClient.h:
* WebProcess/WebCoreSupport/WebInspectorFrontendClient.h:
* WebProcess/WebCoreSupport/WebPlatformStrategies.h:
* WebProcess/WebCoreSupport/WebPopupMenu.h:
* WebProcess/WebCoreSupport/WebSearchPopupMenu.h:
* win/WebKit2Common.vsprops:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95834 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Define BUILDING_ON_LION / TARGETING_LION when appropriate in Platform.h.

* wtf/Platform.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95833 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix the build.

* TestWebKitAPI/Tests/WTF/StringBuilder.cpp: Define operator<<(std::ostream&, const WTF::String&) inside
the WTF namespace.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95832 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed: [Qt] Workaround MSVC2010 problems when linking QtWebKit

Include MSVC2010 in the list of compilers where incremental
build is disabled (INCREMENTAL:NO).

Change suggested by Simo Falt <simo.falt@nokia.com>, already present
in the Qt-4.8 repository where the problem was identified.

* WebCore.pri:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95831 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed. Normalize test names.

* inspector/performance/resources/first-open-elements.html:
* inspector/performance/resources/first-open-resources.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95828 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Try to fix warning seen in some 32-bit builds.

* rendering/RenderBoxModelObject.cpp:
(WebCore::backgroundRectAdjustedForBleedAvoidance):
Use an explicit cast for the conversion from a double to a LayoutUnit.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95826 268f45cc-cd09-0410-ab3c-d52691b4dbfc

We should add support for OVERRIDE and FINAL annotations
https://bugs.webkit.org/show_bug.cgi?id=68654

Reviewed by David Hyatt.

Add OVERRIDE and FINAL macros for compilers that support them.

* wtf/Compiler.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95823 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix typo in file name from trac.webkit.org/changeset/95813.

Unreviewed.

* platform/gtk/media/video-playing-and-pause-actual.txt: Removed.
* platform/gtk/media/video-playing-and-pause-expected.txt: Copied from LayoutTests/platform/gtk/media/video-playing-and-pause-actual.txt.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95822 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Rebaseline after trac.webkit.org/changeset/95799

Unreviewed.

* platform/mac-wk2/media/video-playing-and-pause-expected.txt: Added.
* platform/mac/media/video-playing-and-pause-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95818 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Rebaseline after trac.webkit.org/changeset/95799

Unreviewed.

* platform/gtk/media/video-playing-and-pause-actual.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95813 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-09-23  Xan Lopez  <xlopez@igalia.com>

        [GTK] Skip crashing test fast/workers/storage/interrupt-database.html
        https://bugs.webkit.org/show_bug.cgi?id=68700

        Unreviewed.

        * platform/gtk/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95809 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: Clicking on error doesn't point on the correct line if user searched something
https://bugs.webkit.org/show_bug.cgi?id=68627

Reviewed by Yury Semikhatsky.

* inspector/front-end/SearchController.js:
(WebInspector.SearchController.prototype.disableSearchUntilExplicitAction):
* inspector/front-end/inspector.js:
(WebInspector._showAnchorLocationInPanel):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95807 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: InspectorStyleSheet should use stylesheet's original url, not final one.
https://bugs.webkit.org/show_bug.cgi?id=68631

Reviewed by Yury Semikhatsky.

Source/WebCore:

Test: http/tests/inspector/styles/styles-redirected-css.html

* inspector/InspectorStyleSheet.cpp:
(WebCore::InspectorStyleSheet::resourceStyleSheetText):

LayoutTests:

* http/tests/inspector/elements-test.js:
(initialize_ElementTest.InspectorTest.getElementStylePropertyTreeItem):
(initialize_ElementTest.InspectorTest.getMatchedStylePropertyTreeItem):
(initialize_ElementTest.InspectorTest.getFirstPropertyTreeItemForSection):
* http/tests/inspector/styles/resources/styles-redirected-css.css: Added.
(div):
* http/tests/inspector/styles/resources/styles-redirected-css.php: Added.
* http/tests/inspector/styles/styles-redirected-css-expected.txt: Added.
* http/tests/inspector/styles/styles-redirected-css.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95806 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: XHRs requests with the same url as main resource should have correct category.
https://bugs.webkit.org/show_bug.cgi?id=68646

Reviewed by Yury Semikhatsky.

Source/WebCore:

Test: http/tests/inspector/network/network-xhr-same-url-as-main-resource.html

* inspector/InspectorResourceAgent.cpp:
(WebCore::InspectorResourceAgent::didReceiveResponse):

LayoutTests:

* http/tests/inspector/network/network-xhr-same-url-as-main-resource-expected.txt: Added.
* http/tests/inspector/network/network-xhr-same-url-as-main-resource.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95805 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt][WK2] QtFileDownloader ctor shouldn't call QtFileDownloader::onReadyRead()
https://bugs.webkit.org/show_bug.cgi?id=68696

Reviewed by Andreas Kling.

Instead of connecting signals and calling QtFileDownloader::onReadyRead()
on its constructor, QtFileDownloader will now handle this in start().

* WebProcess/Downloads/qt/DownloadQt.cpp:
(WebKit::Download::start):
* WebProcess/Downloads/qt/QtFileDownloader.cpp:
(WebKit::QtFileDownloader::QtFileDownloader):
(WebKit::QtFileDownloader::start):
* WebProcess/Downloads/qt/QtFileDownloader.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95804 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Update expectation for pservers-grad-14-b.svg on snowleopard
https://bugs.webkit.org/show_bug.cgi?id=68697

Unreviewed rebaseline.

* platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95802 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: Add context menu action to navigate from resources panel resource to network panel.
https://bugs.webkit.org/show_bug.cgi?id=68657

Reviewed by Yury Semikhatsky.

* English.lproj/localizedStrings.js:
* inspector/front-end/NetworkPanel.js:
(WebInspector.NetworkPanel.prototype.showAnchorLocation):
(WebInspector.NetworkPanel.prototype.revealAndHighlightResource):
* inspector/front-end/ResourcesPanel.js:
(WebInspector.FrameResourceTreeElement.prototype._handleContextMenuEvent):
(WebInspector.FrameResourceTreeElement.prototype._appendOpenInNetworkPanelAction):
* inspector/front-end/inspector.js:
(WebInspector.openInNetworkPanelLabel):
(WebInspector.openRequestInNetworkPanel):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95801 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Gtk] Title attribute is not respected on option elements.
https://bugs.webkit.org/show_bug.cgi?id=68615

Patch by Deepak Sherveghar <bpwv64@motorola.com> on 2011-09-23
Reviewed by Martin Robinson.

Set tooltip on PopupMenuGtk items created from GtkActions.

* platform/gtk/GtkPopupMenu.cpp:
(WebCore::GtkPopupMenu::appendItem): Added a call to gtk_widget_set_tooltip_text()
    to set the tooltip on menuitem. Tooltip text is retrieved from GtkAction.
* platform/gtk/PopupMenuGtk.cpp:
(WebCore::PopupMenuGtk::createGtkActionForMenuItem): Pass the tooltip text
    from client when creating a GtkAction.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95800 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt][Gtk] Wrong state when pausing a video in the "playing" event handler
https://bugs.webkit.org/show_bug.cgi?id=68589

Reviewed by Philippe Normand.

Source/WebCore:

Test: media/video-playing-and-pause.html

As suggested by Alexis Menard on IRC, query gstreamer's state instead of using a cached value
in MediaPlayerGstreamerPrivate::paused().

* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
(WebCore::MediaPlayerPrivateGStreamer::paused):

LayoutTests:

* media/video-playing-and-pause.html: Added.
* media/video-paint-test.js:
* platform/qt/media/video-playing-and-pause-expected.txt: Added.
* platform/qt/media/video-playing-and-pause-expected.png: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95799 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium, gtk, qt] Rebaseline following r95726
https://bugs.webkit.org/show_bug.cgi?id=68694

Unreviewed rebaseline.

* css2.1/20110323/abspos-non-replaced-width-margin-000-expected.txt: Renamed from LayoutTests/platform/gtk/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.txt.
* css2.1/20110323/abspos-replaced-width-margin-000-expected.txt: Renamed from LayoutTests/platform/gtk/css2.1/20110323/abspos-replaced-width-margin-000-expected.txt.
* platform/chromium/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.png: Added.
* platform/chromium/css2.1/20110323/abspos-replaced-width-margin-000-expected.png: Added.
* platform/qt/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.txt: Removed.
* platform/qt/css2.1/20110323/abspos-replaced-width-margin-000-expected.txt: Removed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95798 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: skip overlapping scripts when displaying concatenated content.
https://bugs.webkit.org/show_bug.cgi?id=68144

Source/WebCore:

Dynamically appended script tags have wrong offsets and cause troubles building concatenated content.

Reviewed by Pavel Feldman.

* inspector/front-end/SourceFile.js:
(WebInspector.ConcatenatedScriptsContentProvider.prototype._concatenateScriptsContent.appendChunk):
(WebInspector.ConcatenatedScriptsContentProvider.prototype._concatenateScriptsContent):

LayoutTests:

Reviewed by Pavel Feldman.

* inspector/debugger/content-providers-expected.txt:
* inspector/debugger/content-providers.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95797 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-09-23  Xan Lopez  <xlopez@igalia.com>

        [GTK] Rebaseline following r95745, r95725

        Unreviewed rebaseline.

        * platform/gtk/editing/deleting/merge-whitespace-pre-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95796 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[chromium] Rebaseline following r95745, r95725
https://bugs.webkit.org/show_bug.cgi?id=68685

Unreviewed rebaseline.

* platform/chromium-cg-mac-leopard/editing/deleting/merge-whitespace-pre-expected.png:
* platform/chromium-linux/editing/deleting/merge-whitespace-pre-expected.png:
* platform/chromium-mac/editing/deleting/merge-whitespace-pre-expected.png:
* platform/chromium-win/editing/deleting/merge-whitespace-pre-expected.png:
* platform/chromium-win/editing/deleting/merge-whitespace-pre-expected.txt:
* platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95795 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: file open dialog appears when user clicks on the timeline bar in timeline panel.
https://bugs.webkit.org/show_bug.cgi?id=68312

Reviewed by Yury Semikhatsky.

* inspector/front-end/TimelinePanel.js:
(WebInspector.TimelinePanel.prototype._createFileSelector):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95794 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, rolling out r95383.
http://trac.webkit.org/changeset/95383
https://bugs.webkit.org/show_bug.cgi?id=68690

It broke WebInspector.Timeline.LoadFromFile feature.
(Requested by loislo on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-09-23

* inspector/front-end/TimelinePanel.js:
(WebInspector.TimelinePanel.prototype._createFileSelector):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95793 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Web Inspector: UI performance test for Scripts panel first open experience.
https://bugs.webkit.org/show_bug.cgi?id=68541

The test's body has dummy scripts section copied from inspector-test.jss

Reviewed by Yury Semikhatsky.

* inspector/performance/resources/first-open-scripts.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95792 268f45cc-cd09-0410-ab3c-d52691b4dbfc

use after free in WebCore::SVGTRefElement::updateReferencedText
https://bugs.webkit.org/show_bug.cgi?id=67555

Patch by Rob Buis <rbuis@rim.com> on 2011-09-23
Reviewed by Nikolas Zimmermann.

Source/WebCore:

Event listeners can outlive the tref element that created them when
the tref is cloned and then garbage collected, causing a dangling pointer to the
tref. To fix this do not install event listener until the tref is inserted into the document.

Test: svg/custom/tref-clone-crash.html

* svg/SVGTRefElement.cpp:
(WebCore::SVGTRefElement::svgAttributeChanged):
(WebCore::SVGTRefElement::insertedIntoDocument):
* svg/SVGTRefElement.h:

LayoutTests:

Test that cloned tref does not cause a crash.

* svg/custom/tref-clone-crash-expected.txt: Added.
* svg/custom/tref-clone-crash.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95791 268f45cc-cd09-0410-ab3c-d52691b4dbfc

ASSERTION FAILED: documentLoader in WebKit/Source/WebCore/inspector/InspectorInstrumentation.cpp(597)
https://bugs.webkit.org/show_bug.cgi?id=68291

Reviewed by Tony Chang.

* dom/Document.cpp:
(WebCore::Document::finishedParsing):
* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::domContentLoadedEventFiredImpl):
(WebCore::InspectorInstrumentation::loadEventFiredImpl):
* inspector/InspectorInstrumentation.h:
(WebCore::InspectorInstrumentation::domContentLoadedEventFired):
(WebCore::InspectorInstrumentation::loadEventFired):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::dispatchLoadEvent):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95790 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-09-23  Xan Lopez  <xlopez@igalia.com>

        Crash on editing/pasteboard/drag-drop-input-in-svg.svg
        https://bugs.webkit.org/show_bug.cgi?id=68686

        Unreviewed, skip crashing test.

        * platform/gtk/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95789 268f45cc-cd09-0410-ab3c-d52691b4dbfc

plugins/mouse-click-iframe-to-plugin.html should not use flash plugin
https://bugs.webkit.org/show_bug.cgi?id=68601

Use the test plugin that we always have.

* platform/qt-wk2/Skipped: Unskip the test.
* plugins/mouse-click-iframe-to-plugin.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95788 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Source/JavaScriptCore: GetScopedVar should have value profiling
https://bugs.webkit.org/show_bug.cgi?id=68676

Patch by Filip Pizlo <fpizlo@apple.com> on 2011-09-22
Reviewed by Oliver Hunt.

Added GetScopedVar value profiling and predictin propagation.
Added GetScopeChain to CSE.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGGraph.h:
(JSC::DFG::Graph::predict):
* dfg/DFGNode.h:
(JSC::DFG::Node::hasPrediction):
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):
(JSC::DFG::Propagator::getScopeChainLoadElimination):
(JSC::DFG::Propagator::performNodeCSE):
* jit/JITPropertyAccess.cpp:
(JSC::JIT::emit_op_get_scoped_var):

LayoutTests: [Qt] Unreviewed gardening, update expected file after r95745.

Patch by Csaba Osztrogonác <ossy@webkit.org> on 2011-09-22

* platform/qt/editing/deleting/merge-whitespace-pre-expected.png:
* platform/qt/editing/deleting/merge-whitespace-pre-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95787 268f45cc-cd09-0410-ab3c-d52691b4dbfc

GetScopedVar should have value profiling
https://bugs.webkit.org/show_bug.cgi?id=68676

Reviewed by Oliver Hunt.

Added GetScopedVar value profiling and predictin propagation.
Added GetScopeChain to CSE.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGGraph.h:
(JSC::DFG::Graph::predict):
* dfg/DFGNode.h:
(JSC::DFG::Node::hasPrediction):
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):
(JSC::DFG::Propagator::getScopeChainLoadElimination):
(JSC::DFG::Propagator::performNodeCSE):
* jit/JITPropertyAccess.cpp:
(JSC::JIT::emit_op_get_scoped_var):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95786 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] Unreviewed gardening, update expected file after r95745.

* platform/qt/editing/deleting/merge-whitespace-pre-expected.png:
* platform/qt/editing/deleting/merge-whitespace-pre-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95785 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed fix after r95764, update expected fail to expected pass.

* ietestcenter/Javascript/15.3.4.5-0-2-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95784 268f45cc-cd09-0410-ab3c-d52691b4dbfc

implement -webkit-flex-order
https://bugs.webkit.org/show_bug.cgi?id=67432

Reviewed by Ojan Vafai.

Source/WebCore:

flex-order can be an int (including negative), but we disallow the two
smallest values so we can put the numbers into a hash set.

Also, create two iterators: one that goes in render tree order (we use
this for the first pass and to collect the possible flex order values)
and one that goes in flex order.

Test: css3/flexbox/flex-order.html

* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
* rendering/RenderFlexibleBox.cpp:
(WebCore::FlexOrderHashTraits::emptyValue):
(WebCore::FlexOrderHashTraits::constructDeletedValue):
(WebCore::FlexOrderHashTraits::isDeletedValue):
(WebCore::RenderFlexibleBox::TreeOrderIterator::TreeOrderIterator): A simple iterator
    that goes in render tree order.
(WebCore::RenderFlexibleBox::TreeOrderIterator::next):
(WebCore::RenderFlexibleBox::TreeOrderIterator::reset):
(WebCore::RenderFlexibleBox::TreeOrderIterator::flexOrderValues):
(WebCore::RenderFlexibleBox::FlexOrderIterator::FlexOrderIterator): An iterator that
    goes in flex-order order.  Creating this involves sorting, so only create it once
    and pass it around.
(WebCore::RenderFlexibleBox::FlexOrderIterator::first):
(WebCore::RenderFlexibleBox::FlexOrderIterator::next):
(WebCore::RenderFlexibleBox::FlexOrderIterator::reset):
(WebCore::RenderFlexibleBox::layoutHorizontalBlock):
(WebCore::RenderFlexibleBox::computePreferredLogicalWidth):
(WebCore::RenderFlexibleBox::runFreeSpaceAllocationAlgorithmInlineDirection):
(WebCore::RenderFlexibleBox::layoutAndPlaceChildrenInlineDirection):
* rendering/RenderFlexibleBox.h:

LayoutTests:

* css3/flexbox/flex-order-expected.png: Added.
* css3/flexbox/flex-order-expected.txt: Added.
* css3/flexbox/flex-order.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95783 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Use AffineTransform scale functions in ShadowBlur::adjustBlurRadius
https://bugs.webkit.org/show_bug.cgi?id=68667

Reviewed by Simon Fraser.

* platform/graphics/ShadowBlur.cpp:
(WebCore::ShadowBlur::adjustBlurRadius): Use AffineTransform::xScale and
AffineTransform::yScale instead of the custom code here that seems to do
the same thing.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95782 268f45cc-cd09-0410-ab3c-d52691b4dbfc

enable NRWT for run-webkit-tests on Lion
https://bugs.webkit.org/show_bug.cgi?id=68673

Reviewed by Eric Seidel.

* Scripts/run-webkit-tests:
(useNewRunWebKitTests):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95781 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Rebaseline for bug 65583 (path based border radius drawing on skia) part 6
https://bugs.webkit.org/show_bug.cgi?id=68671

Patch by Ben Wells <benwells@chromium.org> on 2011-09-22
Reviewed by James Robinson.

* platform/chromium-linux/tables/mozilla/marvin/: 9 changes
* platform/chromium-linux/tables/mozilla_expected_failures/marvin/: 14 changes
* platform/chromium-linux/transforms/svg-vs-css-expected.png:
* platform/chromium-win/tables/mozilla/marvin/: 9 changes
* platform/chromium-win/tables/mozilla_expected_failures/marvin/: 14 changes
* platform/chromium-win/transforms/svg-vs-css-expected.png:
* platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95780 268f45cc-cd09-0410-ab3c-d52691b4dbfc

PPC build fix, part 3.

* runtime/Executable.cpp:
(JSC::FunctionExecutable::compileForConstructInternal):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95779 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove unneeded type conversion from background bleed code
https://bugs.webkit.org/show_bug.cgi?id=68669

Reviewed by Geoffrey Garen.

* rendering/RenderBoxModelObject.cpp:
(WebCore::backgroundRectAdjustedForBleedAvoidance): Use ceil instead of ceilf
and eliminate superflous conversion to float and use of FloatSize. Also added
a "why" comment.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95778 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Refactor checks for antialiasing lines to share a single function
https://bugs.webkit.org/show_bug.cgi?id=68666

Reviewed by Dan Bernstein.

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::paintColumnRules): Call shouldAntialiasLines.
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::paintBorder): Ditto.
(WebCore::RenderBoxModelObject::shouldAntialiasLines): Added.
* rendering/RenderBoxModelObject.h: Added shouldAntialiasLines function.
* rendering/RenderInline.cpp:
(WebCore::RenderInline::paintOutlineForLine): Call shouldAntialiasLines.
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::paintCollapsedBorder): Ditto.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95777 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove unused members from LayerChromium.
These members are never set, and never used.
Other members that are set but not used have been left (we may use them
later, e.g. m_opaque).
https://bugs.webkit.org/show_bug.cgi?id=68297

Patch by Antoine Labour <piman@chromium.org> on 2011-09-22
Reviewed by James Robinson.

No functional change, no new test needed.

* platform/graphics/chromium/LayerChromium.cpp:
(WebCore::LayerChromium::LayerChromium):
* platform/graphics/chromium/LayerChromium.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95776 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Adding testTracks() function to video-test.js
https://bugs.webkit.org/show_bug.cgi?id=68643

Reviewed by Eric Carlson.

* media/video-test.js:
(testTracks):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95775 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Make XSSAuditor extract meaningful snippet from script blocks for comparison
against the URL when checking for reflection.  Avoids getting caugh up in
trailing comments.
https://bugs.webkit.org/show_bug.cgi?id=68094

Patch by Tom Sepez <tsepez@chromium.org> on 2011-09-22
Reviewed by Adam Barth.

Source/WebCore:

Tests: http/tests/security/xssAuditor/script-tag-with-trailing-comment.html
       http/tests/security/xssAuditor/script-tag-with-trailing-comment2.html
       http/tests/security/xssAuditor/script-tag-with-trailing-comment3.html

* html/parser/XSSAuditor.cpp:
(WebCore::XSSAuditor::filterTokenAfterScriptStartTag):
(WebCore::XSSAuditor::extractCodeFragment):
* html/parser/XSSAuditor.h:

LayoutTests:

* http/tests/security/xssAuditor/resources/echo-intertag.pl:
* http/tests/security/xssAuditor/script-tag-with-trailing-comment-expected.txt: Added.
* http/tests/security/xssAuditor/script-tag-with-trailing-comment.html: Added.
* http/tests/security/xssAuditor/script-tag-with-trailing-comment2-expected.txt: Added.
* http/tests/security/xssAuditor/script-tag-with-trailing-comment2.html: Added.
* http/tests/security/xssAuditor/script-tag-with-trailing-comment3-expected.txt: Added.
* http/tests/security/xssAuditor/script-tag-with-trailing-comment3.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95774 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] TouchWebView load test not running
https://bugs.webkit.org/show_bug.cgi?id=68620

Patch by Gopal Raghavan <gopal.1.raghavan@nokia.com> on 2011-09-22
Reviewed by Chang Shu.

Load test was not executed since it was missing in pro file. Added tst_load.qml to qmltests.pro

* UIProcess/API/qt/tests/qmltests/qmltests.pro:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95773 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Another PPC build fix.

* runtime/Executable.cpp:
* runtime/Executable.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95772 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Rebaseline for bug 65583 (path based border radius drawing on skia) part 5
https://bugs.webkit.org/show_bug.cgi?id=68608

Patch by Ben Wells <benwells@chromium.org> on 2011-09-22
Reviewed by James Robinson.

* platform/chromium-linux/css2.1/: 14 changes
* platform/chromium-linux/editing/execCommand/5138441-expected.png:
* platform/chromium-linux/editing/inserting/editing-empty-divs-expected.png:
* platform/chromium-linux/editing/selection/transformed-selection-rects-expected.png:
* platform/chromium-linux/fast/: 9 changes
* platform/chromium-win/css2.1/: 14 changes
* platform/chromium-win/editing/execCommand/5138441-expected.png:
* platform/chromium-win/editing/inserting/editing-empty-divs-expected.png:
* platform/chromium-win/editing/selection/transformed-selection-rects-expected.png:
* platform/chromium-win/fast/: 10 changes
* platform/chromium/test_expectations.txt:
* platform/win-xp/editing/selection/transformed-selection-rects-expected.txt: Renamed from LayoutTests/platform/win-wk2/editing/selection/transformed-selection-rects-expected.txt.
* platform/win/editing/selection/transformed-selection-rects-expected.txt: Renamed from LayoutTests/platform/chromium-win/editing/selection/transformed-selection-rects-expected.txt.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95771 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add unit test for existing StringBuilder
https://bugs.webkit.org/show_bug.cgi?id=67080

Patch by Xianzhu Wang <wangxianzhu@chromium.org> on 2011-09-22
Reviewed by Darin Adler.

* TestWebKitAPI/TestWebKitAPI.gypi:
* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WTF/StringBuilder.cpp: Added.
(operator<<):
(TestWebKitAPI::expectBuilderContent):
(TestWebKitAPI::expectEmpty):
(TestWebKitAPI::TEST):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95770 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Updated expectations and filed WK68674.

* platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95769 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove didReceiveAuthenticationChallenge() from SubresourceLoaderClient.
Instead, add a load-specific policy for showing the user authentication
challenge down to ResourceLoaderOptions and enforce it in ResourceLoader.
https://bugs.webkit.org/show_bug.cgi?id=65330

Reviewed by Alexey Proskuryakov.

No new tests, refactor only.

* loader/DocumentThreadableLoader.cpp:
* loader/DocumentThreadableLoader.h:
* loader/MainResourceLoader.cpp:
* loader/NetscapePlugInStreamLoader.cpp:
* loader/ResourceLoadScheduler.h:
* loader/ResourceLoader.cpp:
(WebCore::ResourceLoader::didReceiveAuthenticationChallenge):
   For resource types that always send a challenge to the embedder,
   this patch doesn't change anything. For those that don't, we will
   always try to continue without credentials when they are forbidden
   and the platform supports it.
   When continuing without credentials was initially implemented in
   DocumentThreadableLoader, we sent the ThreadableLoaderClient a didFail(),
   then canceled the SubresourceLoader. This was necessary because of the
   quirks of ThreadableLoader cancellation (we sever the client/loader connections
   before the load actually cancels), but a simple didFail() should suffice at
   the ResourceLoader layer.
* loader/ResourceLoaderOptions.h:
* loader/SubresourceLoader.cpp:
* loader/SubresourceLoader.h:
* loader/SubresourceLoaderClient.h:
* loader/cache/CachedResource.cpp:
* loader/cache/CachedResourceLoader.cpp:
* loader/cache/CachedResourceLoader.h:
* loader/icon/IconLoader.cpp: The ResourceLoader implementation of
    didReceiveAuthenticationChallege means that IconLoader will now
    try to continue with credentials on platforms that support it,
    rather than just canceling outright. We still will never prompt
    for authentication for icons.
* loader/icon/IconLoader.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95768 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[EFL] Use ewk_view to navigate in history instead of ewk_history.
https://bugs.webkit.org/show_bug.cgi?id=68455

Patch by Raphael Kubo da Costa <kubo@profusion.mobi> on 2011-09-22
Reviewed by Antonio Gomes.

BackForwardItem::invoke() was using ewk_history functions to navigate
in history; however, this will only call WebCore::BackForwardListImpl
and not move between pages at all. We now call ewk_view_navigate and
its siblings to make sure the navigation actually happens.

This should make tests like fast/dom/navigation-type-back-forward.html
stop timing out.

* DumpRenderTree/efl/WorkQueueItemEfl.cpp:
(BackForwardItem::invoke):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95767 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add ENABLE_CSS_FILTERS
https://bugs.webkit.org/show_bug.cgi?id=68652

Reviewed by Simon Fraser.

* Source/JavaScriptCore/Configurations/FeatureDefines.xcconfig:
* Source/WebCore/Configurations/FeatureDefines.xcconfig:
* Source/WebKit/mac/Configurations/FeatureDefines.xcconfig:
* Source/WebKit2/Configurations/FeatureDefines.xcconfig:
* WebKitLibraries/win/tools/vsprops/FeatureDefines.vsprops:
* WebKitLibraries/win/tools/vsprops/FeatureDefinesCairo.vsprops:
* Tools/Scripts/build-webkit:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95766 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Make sure to update scrollbars in ScrollView::setFrameRect
https://bugs.webkit.org/show_bug.cgi?id=68663

Reviewed by Sam Weinig.

Call updateScrollbars after setting the frame rect. This was previously done by the
call to setBoundsSize from RenderWidget::setWidgetGeometry but setBoundsSize was removed
completely in http://trac.webkit.org/changeset/95725.

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

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95765 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Function.prototype.bind.length shoudl be 1.

Rubber stamped by Olier Hunt.

Source/JavaScriptCore:

* runtime/FunctionPrototype.cpp:
(JSC::FunctionPrototype::addFunctionProperties):

LayoutTests:

* fast/js/function-bind-expected.txt:
* fast/js/script-tests/function-bind.js:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95764 268f45cc-cd09-0410-ab3c-d52691b4dbfc

PPC build fix.

* bytecode/CodeBlock.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95763 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Windows build fix pt. 2

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95762 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Windows build fix pt. 1

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95761 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt][Mac]Unreviewed gardening.

New Qt specific results were added by r95723, but they fail on Qt-Mac platform

* platform/qt-mac/Skipped: Skip css3/selectors3 tests.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95760 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add APIObjects for Size, Point, Rect
https://bugs.webkit.org/show_bug.cgi?id=68644

Reviewed by Sam Weinig.

Make APIObjects for size, point and rect so that they can be
used in WKArrayRefs.

Make some inline helper methods to create WKSize, WKPoint and WKRect
types.

* GNUmakefile.am:
* Shared/API/c/WKBase.h:
* Shared/API/c/WKGeometry.cpp: Copied from Source/WebKit2/Shared/API/c/WKGeometry.h.
(WKSizeGetTypeID):
(WKPointGetTypeID):
(WKRectGetTypeID):
(WKPointCreate):
(WKSizeCreate):
(WKRectCreate):
(WKSizeGetValue):
(WKPointGetValue):
(WKRectGetValue):
* Shared/API/c/WKGeometry.h:
(WKPoint::WKPointMake):
(WKPoint::WKSizeMake):
(WKPoint::WKRectMake):
* Shared/API/c/WKSharedAPICast.h:
* Shared/APIObject.h:
* Shared/UserMessageCoders.h:
(WebKit::UserMessageEncoder::baseEncode):
(WebKit::UserMessageDecoder::baseDecode):
* Shared/WebGeometry.h: Added.
(WebKit::WebSize::create):
(WebKit::WebSize::size):
(WebKit::WebSize::WebSize):
(WebKit::WebSize::type):
(WebKit::WebPoint::create):
(WebKit::WebPoint::point):
(WebKit::WebPoint::WebPoint):
(WebKit::WebPoint::type):
(WebKit::WebRect::create):
(WebKit::WebRect::rect):
(WebKit::WebRect::WebRect):
(WebKit::WebRect::type):
* WebKit2.pro:
* WebKit2.xcodeproj/project.pbxproj:
* WebKit2API.pri:
* win/WebKit2.vcproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95759 268f45cc-cd09-0410-ab3c-d52691b4dbfc

DFG JIT does not support to_primitive or strcat
https://bugs.webkit.org/show_bug.cgi?id=68582

Reviewed by Darin Adler.

This adds functional support for to_primitive and strcat. It focuses
on minimizing the amount of code emitted on to_primitive (if we know
that it is a primitive or can speculate cheaply, then we omit the
slow path) and on keeping the implementation of strcat simple while
leveraging whatever optimizations we have already. In particular,
unlike the Call and Construct nodes which require extending the size
of the DFG's callee registers, StrCat takes advantage of the fact
that no JS code can run while StrCat is in progress and uses a
scratch buffer, rather than the register file, to store the list of
values to concatenate. This was done mainly to keep the code simple,
but there are probably other benefits to keeping call frame sizes
down. Essentially, this patch ensures that the presence of an
op_strcat does not mess up any other optimizations we might do while
ensuring that if you do execute it, it'll work about as well as you'd
expect.

When combined with the previous patch for integer division, this is a
14% speed-up on Kraken. Without it, it would have been a 2% loss.

* assembler/AbstractMacroAssembler.h:
(JSC::AbstractMacroAssembler::TrustedImmPtr::TrustedImmPtr):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.h:
(JSC::DFG::canCompileOpcode):
* dfg/DFGJITCodeGenerator.h:
(JSC::DFG::JITCodeGenerator::callOperation):
* dfg/DFGJITCompiler.cpp:
(JSC::DFG::JITCompiler::exitSpeculativeWithOSR):
* dfg/DFGNode.h:
* dfg/DFGOperations.cpp:
* dfg/DFGOperations.h:
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):
(JSC::DFG::Propagator::performNodeCSE):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::JSGlobalData):
(JSC::JSGlobalData::~JSGlobalData):
* runtime/JSGlobalData.h:
(JSC::JSGlobalData::scratchBufferForSize):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95758 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Progress control gets cropped on the bottom
https://bugs.webkit.org/show_bug.cgi?id=68302
<rdar://problem/10069915>

Reviewed by Kent Tamura.

As it turns out the smaller control type does not get rendered either, so this patch
fixes both.

* manual-tests/dom/progressbar.html: Altered to show both sizes of controls
* rendering/RenderThemeMac.h:
* rendering/RenderThemeMac.mm: Added methods to inflate the drawing rect
(WebCore::RenderThemeMac::progressBarSizes):
(WebCore::RenderThemeMac::progressBarMargins):
(WebCore::RenderThemeMac::minimumProgressBarHeight):
(WebCore::RenderThemeMac::paintProgressBar): Inflate the rect based on minimum desired
control height and glow margin.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95757 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=68658

Make matchedEndLine smart enough to not match lines that have moved to new
regions with different available content logical widths. When this happens, we go ahead and treat
the line as failing to match.

Reviewed by Anders Carlsson.

Source/WebCore:

Added new tests in fast/regions.

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::lineWidthForPaginatedLineChanged):
* rendering/RenderBlock.h:
Enhanced to take an optional delta, so that you can check a new position without having to move
the line box.

* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::checkPaginationAndFloatsAtEndLine):
New function that refactors checking for line width changes from region movement as well as the
float checks. Since the float checks were duplicated twice, this is a nice refactoring.

(WebCore::RenderBlock::matchedEndLine):
Changed to call the new helper function that will check both floats and pagination.

LayoutTests:

* fast/regions/webkit-flow-inlines-dynamic.html: Added.
* platform/mac/fast/regions/webkit-flow-inlines-dynamic-expected.png: Added.
* platform/mac/fast/regions/webkit-flow-inlines-dynamic-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95756 268f45cc-cd09-0410-ab3c-d52691b4dbfc

EventSenderProxy::mouseUp should call -[WKView mouseUp:] instead of -[WKView mouseDown:]
https://bugs.webkit.org/show_bug.cgi?id=68660

Reviewed by Adam Barth.

* WebKitTestRunner/mac/EventSenderProxy.mm:
(WTR::EventSenderProxy::mouseUp):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95755 268f45cc-cd09-0410-ab3c-d52691b4dbfc

DFG JIT should support integer division
https://bugs.webkit.org/show_bug.cgi?id=68597

Reviewed by Darin Adler.

This adds support for ArithDiv speculating integer, and speculating
that the result is integer (i.e. remainder = 0).

This is a 4% win on Kraken and a 1% loss on V8.

* bytecode/CodeBlock.h:
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::makeDivSafe):
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGNode.h:
(JSC::DFG::Node::hasArithNodeFlags):
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateArithNodeFlags):
(JSC::DFG::Propagator::propagateNodePredictions):
(JSC::DFG::Propagator::fixupNode):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* jit/JITArithmetic.cpp:
(JSC::JIT::emit_op_div):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Implement put_scoped_var in the DFG jit
https://bugs.webkit.org/show_bug.cgi?id=68653

Reviewed by Gavin Barraclough.

Naive implementation of put_scoped_var.  Same story as the
get_scoped_var implementation, although I've hoisted scope
object acquisition into a separate dfg node.  Ideally in the
future we would reuse the resolved scope chain object, but
for now we don't.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.h:
(JSC::DFG::canCompileOpcode):
* dfg/DFGNode.h:
(JSC::DFG::Node::hasScopeChainDepth):
(JSC::DFG::Node::scopeChainDepth):
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95753 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Skip test that need new baselines.

* platform/mac/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95752 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Implement Function.prototype.bind
https://bugs.webkit.org/show_bug.cgi?id=26382

Reviewed by Sam Weinig.

Source/JavaScriptCore:

This patch provides a basic functional implementation
for Function.bind. It should (hopefully!) be fully
functionally correct, and the bound functions can be
called to quickly (since they are a subclass of
JSFunction, not InternalFunction), but we'll probably
want to follow up with some optimization work to keep
bound calls in JIT code.

* JavaScriptCore.JSVALUE32_64only.exp:
* JavaScriptCore.JSVALUE64only.exp:
* JavaScriptCore.exp:
* JavaScriptCore.xcodeproj/project.pbxproj:
* jit/JITStubs.cpp:
(JSC::JITThunks::hostFunctionStub):
* jit/JITStubs.h:
* jsc.cpp:
(GlobalObject::addFunction):
* runtime/CommonIdentifiers.h:
* runtime/ConstructData.h:
* runtime/Executable.h:
(JSC::NativeExecutable::NativeExecutable):
* runtime/FunctionPrototype.cpp:
(JSC::FunctionPrototype::addFunctionProperties):
(JSC::functionProtoFuncBind):
* runtime/FunctionPrototype.h:
* runtime/JSBoundFunction.cpp: Added.
(JSC::boundFunctionCall):
(JSC::boundFunctionConstruct):
(JSC::JSBoundFunction::create):
(JSC::JSBoundFunction::hasInstance):
(JSC::JSBoundFunction::getOwnPropertySlot):
(JSC::JSBoundFunction::getOwnPropertyDescriptor):
(JSC::JSBoundFunction::JSBoundFunction):
(JSC::JSBoundFunction::finishCreation):
* runtime/JSBoundFunction.h: Added.
(JSC::JSBoundFunction::targetFunction):
(JSC::JSBoundFunction::boundThis):
(JSC::JSBoundFunction::boundArgs):
(JSC::JSBoundFunction::createStructure):
* runtime/JSFunction.cpp:
(JSC::JSFunction::create):
(JSC::JSFunction::finishCreation):
(JSC::createDescriptorForThrowingProperty):
(JSC::JSFunction::getOwnPropertySlot):
* runtime/JSFunction.h:
* runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::getHostFunction):
* runtime/JSGlobalData.h:
* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::reset):
(JSC::JSGlobalObject::visitChildren):
* runtime/JSGlobalObject.h:
(JSC::JSGlobalObject::boundFunctionStructure):
* runtime/Lookup.cpp:
(JSC::setUpStaticFunctionSlot):

Source/WebCore:

Test: fast/js/function-bind.html

* bindings/js/JSDOMBinding.cpp:
(WebCore::objectToStringFunctionGetter):
* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::nonCachingStaticFunctionGetter):
* bindings/js/JSHistoryCustom.cpp:
(WebCore::nonCachingStaticBackFunctionGetter):
(WebCore::nonCachingStaticForwardFunctionGetter):
(WebCore::nonCachingStaticGoFunctionGetter):
* bindings/js/JSLocationCustom.cpp:
(WebCore::nonCachingStaticReplaceFunctionGetter):
(WebCore::nonCachingStaticReloadFunctionGetter):
(WebCore::nonCachingStaticAssignFunctionGetter):
    - Function::create no longer requires functionStructure() to be passed.

LayoutTests:

We now pass Function.bind tests.

* fast/js/Object-getOwnPropertyNames-expected.txt:
* fast/js/basic-strict-mode-expected.txt:
* fast/js/function-bind-expected.txt: Added.
* fast/js/function-bind.html: Added.
* fast/js/mozilla/strict/15.3.4.5-expected.txt:
* fast/js/script-tests/function-bind.js: Added.
* ietestcenter/Javascript/15.2.3.3-4-38-expected.txt:
* ietestcenter/Javascript/15.3.4.5-0-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-0-2-expected.txt:
* ietestcenter/Javascript/15.3.4.5-13.b-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-13.b-2-expected.txt:
* ietestcenter/Javascript/15.3.4.5-13.b-3-expected.txt:
* ietestcenter/Javascript/15.3.4.5-13.b-4-expected.txt:
* ietestcenter/Javascript/15.3.4.5-13.b-5-expected.txt:
* ietestcenter/Javascript/15.3.4.5-15-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-15-2-expected.txt:
* ietestcenter/Javascript/15.3.4.5-16-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-2-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-3-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-4-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-5-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-6-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-7-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-8-expected.txt:
* ietestcenter/Javascript/15.3.4.5-2-9-expected.txt:
* ietestcenter/Javascript/15.3.4.5-8-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-8-2-expected.txt:
* ietestcenter/Javascript/15.3.4.5-9-1-expected.txt:
* ietestcenter/Javascript/15.3.4.5-9-2-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=68650

Make determineStartPosition smart enough to not skip over clean lines that have moved to a new
region with a different available content logical width. When this happens, we go ahead and treat
the line as dirty.

Reviewed by Sam Weinig.

Source/WebCore:

Covered by existing fast/regions tests.

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::layoutBlock):
Disable region fitting if we detect that the regions all have uniform widths. In this case we know
the content width can't vary, so there's no reason to waste time worrying about it.

(WebCore::RenderBlock::computeLogicalLocationForFloat):
Pull the content width checking code into computeLogicalLocationForFloat, since we've incorporated
region fitting into the fixed offsets now rather than the line functions.

(WebCore::RenderBlock::positionNewFloats):
Changed because more code moved from here into computeLogicalLocationForFloat.

(WebCore::RenderBlock::adjustForRegionFittingIfNeeded):
No changes. Just moved the function above the left offset function.

(WebCore::RenderBlock::logicalLeftOffsetForContent):
(WebCore::RenderBlock::logicalRightOffsetForContent):
Added. These new functions give the fixed left and right offsets for lines in a particular region.
When no regions are present or when all the regions have a uniform width, they are identical
to the versions of the functions that take no arguments.

(WebCore::RenderBlock::logicalRightOffsetForLine):
Changed to no longer do region fitting, since that's built into logicalRightOffsetForContent now.

(WebCore::RenderBlock::lineWidthForPaginatedLineChanged):
New function that takes a root line box and compares its cached old content width with the width
available at the line's new location.

* rendering/RenderBlock.h:
(WebCore::RenderBlock::logicalRightOffsetForLine):
(WebCore::RenderBlock::logicalLeftOffsetForLine):
Modified to call the logicalXXXOffsetForContent functions that take a block direction position.

(WebCore::RenderBlock::logicalRightOffsetForContent):
(WebCore::RenderBlock::logicalLeftOffsetForContent):
Added the new functions that take a position so that they can fit to the correct region.

(WebCore::RenderBlock::availableLogicalWidthForContent):
New helper function that calls right - left (similar to the corresponding line function that includes
floats).

* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::determineStartPosition):
Patched to check if a line has moved to a place with a different available content width. If so, the line
is dirtied so that line layout can re-run.

* rendering/RootInlineBox.cpp:
(WebCore::RootInlineBox::RootInlineBox):
(WebCore::RootInlineBox::alignBoxesInBlockDirection):
Sets the paginated line width from the current block logical height.

* rendering/RootInlineBox.h:
(WebCore::RootInlineBox::paginationStrut):
(WebCore::RootInlineBox::setPaginationStrut):
Fixed paginationStrut to be LayoutUnit instead of int.

(WebCore::RootInlineBox::paginatedLineWidth):
(WebCore::RootInlineBox::setPaginatedLineWidth):
Added paginatedLineWidth getter/setter for caching the content width for a given root line.

LayoutTests:

* platform/mac/fast/regions/webkit-flow-floats-inside-regions-bounds-expected.png:
* platform/mac/fast/regions/webkit-flow-floats-inside-regions-bounds-expected.txt:
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.png:
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove Widget::beforeMouseDown and Widget::afterMouseDown
https://bugs.webkit.org/show_bug.cgi?id=68570

Reviewed by Darin Adler.

These two member functions were originally added to avoid crashes due to removing NSViews while they were
being tracked by AppKit. Since they were added, we've moved away from NSViews for form controls, and the bugs in
AppKit that lead to crashes have been fixed.

This patch was rolled out in r95743 because it broke a bunch of tests. The patch accidentally removed setting back
m_sendingEventToSubview to false.

* page/mac/EventHandlerMac.mm:
(WebCore::EventHandler::passMouseDownEventToWidget):
* platform/Widget.h:
* platform/mac/WidgetMac.mm:
(WebCore::Widget::Widget):
(WebCore::Widget::removeFromSuperview):

* rendering/RenderWidget.cpp:
Change widgetHierarchyUpdateSuspendCount to be an unsigned integer instead of a size_t.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    [WK2] UIProcess should check that WebProcess isn't sending unexpected file: URLs to it
        https://bugs.webkit.org/show_bug.cgi?id=68573

        Reviewed by Anders Carlsson.

        Re-landing with a slightly less aggressive check.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Platform-specific results for test introduced by r95726

* platform/gtk/fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks-expected.txt: Added.
* platform/mac/fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks-expected.txt: Added.
* platform/qt/fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, rolling out r95335 and r95645.
http://trac.webkit.org/changeset/95335
http://trac.webkit.org/changeset/95645
https://bugs.webkit.org/show_bug.cgi?id=68649

Wrong fix (Requested by rniwa on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2011-09-22

Source/WebCore:

* editing/DeleteSelectionCommand.cpp:
(WebCore::DeleteSelectionCommand::saveTypingStyleState):
* editing/EditingStyle.cpp:
(WebCore::EditingStyle::removeStyleAddedByNode):
* editing/EditingStyle.h:
* editing/ReplaceSelectionCommand.cpp:
(WebCore::ReplaceSelectionCommand::removeRedundantStylesAndKeepStyleSpanInline):
* editing/markup.cpp:
(WebCore::StyledMarkupAccumulator::serializeNodes):
(WebCore::ancestorToRetainStructureAndAppearance):
* editing/markup.h:

LayoutTests:

* editing/deleting/merge-paragraph-from-address-expected.txt: Removed.
* editing/deleting/merge-paragraph-from-address.html: Removed.
* editing/deleting/merge-paragraph-from-h6-expected.txt: Removed.
* editing/deleting/merge-paragraph-from-h6-with-style-expected.txt: Removed.
* editing/deleting/merge-paragraph-from-h6-with-style.html: Removed.
* editing/deleting/merge-paragraph-from-h6.html: Removed.
* editing/deleting/merge-paragraph-from-listing-expected.txt: Removed.
* editing/deleting/merge-paragraph-from-listing.html: Removed.
* editing/deleting/merge-paragraph-from-p-with-style-expected.txt: Removed.
* editing/deleting/merge-paragraph-from-p-with-style.html: Removed.
* editing/deleting/merge-paragraph-into-h1-expected.txt: Removed.
* editing/deleting/merge-paragraph-into-h1-style-expected.txt: Removed.
* editing/deleting/merge-paragraph-into-h1-with-style-expected.txt: Removed.
* editing/deleting/merge-paragraph-into-h1-with-style.html: Removed.
* editing/deleting/merge-paragraph-into-h1.html: Removed.
* editing/deleting/merge-paragraph-into-pre-expected.txt: Removed.
* editing/deleting/merge-paragraph-into-pre.html: Removed.
* platform/mac/editing/deleting/merge-whitespace-pre-expected.png:
* platform/mac/editing/deleting/merge-whitespace-pre-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, update expected results after r95726

* platform/chromium-win/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.png:
* platform/chromium-win/css2.1/20110323/abspos-replaced-width-margin-000-expected.png:
* platform/gtk/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.txt:
* platform/gtk/css2.1/20110323/abspos-replaced-width-margin-000-expected.txt:
* platform/qt/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.txt:
* platform/qt/css2.1/20110323/abspos-replaced-width-margin-000-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95744 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Revert r95673 since it caused hundreds of tests to fail on Leopard / Snow Leopard.

* page/mac/EventHandlerMac.mm:
(WebCore::EventHandler::passMouseDownEventToWidget):
* platform/Widget.h:
* platform/mac/WidgetMac.mm:
(WebCore::Widget::Widget):
(WebCore::Widget::removeFromSuperview):
(WebCore::Widget::beforeMouseDown):
(WebCore::Widget::afterMouseDown):
* rendering/RenderWidget.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Implement get_scoped_var in the DFG
https://bugs.webkit.org/show_bug.cgi?id=68640

Reviewed by Gavin Barraclough.

Naive implementation of get_scoped_var in the DFG.  Essentially this
is the bare minimum required to get correct behaviour, so there's no
load/store coalescing or type profiling involved, even though these
would be wins.  No impact on SunSpider or V8.

* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.h:
(JSC::DFG::canCompileOpcode):
* dfg/DFGNode.h:
(JSC::DFG::Node::hasVarNumber):
(JSC::DFG::Node::hasScopeChainDepth):
(JSC::DFG::Node::scopeChainDepth):
* dfg/DFGPropagator.cpp:
(JSC::DFG::Propagator::propagateNodePredictions):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compile):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=68645

Fix the layout test fast/regions directory. Remove platform-specific results that were
incorrectly checked in to cross-platform locations and add missing pixel test results.

Reviewed by Beth Dakin.

* fast/regions/content-flowed-into-regions-dynamically-inserted-expected.txt: Removed.
* fast/regions/flows-dependency-dynamic-remove-expected.txt: Removed.
* fast/regions/flows-dependency-same-flow-expected.txt: Removed.
* fast/regions/render-region-renderer-expected.txt: Removed.
* fast/regions/webkit-flow-floats-inside-regions-bounds-expected.txt: Removed.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.txt: Removed.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-expected.txt: Removed.
* fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-rl-expected.txt: Removed.
* fast/regions/webkit-flow-renderer-expected.txt: Removed.
* fast/regions/webkit-flow-renderer-layer-expected.txt: Removed.
* fast/regions/webkit-flow-renderer-nested-expected.txt: Removed.
* platform/mac/fast/regions/content-flowed-into-regions-dynamically-added-expected.png: Added.
* platform/mac/fast/regions/content-flowed-into-regions-dynamically-inserted-expected.txt: Added.
* platform/mac/fast/regions/content-flowed-into-regions-dynamically-removed-expected.png: Added.
* platform/mac/fast/regions/content-flowed-into-regions-expected.png: Added.
* platform/mac/fast/regions/content-flowed-into-regions-with-dyn-index-expected.png: Added.
* platform/mac/fast/regions/content-flowed-into-regions-with-index-dom-expected.png: Added.
* platform/mac/fast/regions/content-flowed-into-regions-with-index-expected.png: Added.
* platform/mac/fast/regions/flows-dependency-dynamic-remove-expected.png: Added.
* platform/mac/fast/regions/flows-dependency-dynamic-remove-expected.txt: Added.
* platform/mac/fast/regions/flows-dependency-same-flow-expected.png: Added.
* platform/mac/fast/regions/flows-dependency-same-flow-expected.txt: Added.
* platform/mac/fast/regions/no-split-line-box-expected.png: Added.
* platform/mac/fast/regions/render-region-renderer-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-floats-inside-regions-bounds-expected.png: Added.
* platform/mac/fast/regions/webkit-flow-floats-inside-regions-bounds-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.png: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-expected.png: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-rl-expected.png: Added.
* platform/mac/fast/regions/webkit-flow-inlines-inside-regions-bounds-vertical-rl-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-renderer-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-renderer-layer-expected.txt: Added.
* platform/mac/fast/regions/webkit-flow-renderer-nested-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95741 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=68638

Make RenderFlowThread cache whether or not it has regions of varying widths. This will
be relevant for performance as we begin adding code to do custom block painting and
layout based off regions not having the same width.

Reviewed by Dan Bernstein and Adam Roben.

* rendering/RenderFlowThread.cpp:
(WebCore::RenderFlowThread::RenderFlowThread):
(WebCore::RenderFlowThread::layout):
* rendering/RenderFlowThread.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

FrameView::invalidateRect and FrameView::setFrameRect shouldn't take LayoutRects
https://bugs.webkit.org/show_bug.cgi?id=68639

Reviewed by David Hyatt.

FrameView::invalidateRect and FrameView::setFrameRect override Widget and ScrollView
member functions that take IntRects, and Widget coordinates are currently defined in terms of
IntRects (and probably always should since we want them to be pixel-aligned).

* page/FrameView.cpp:
(WebCore::FrameView::invalidateRect):
(WebCore::FrameView::setFrameRect):
* page/FrameView.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed build fix (fixes warning about too many parens).

* rendering/RenderBox.cpp:
(WebCore::RenderBox::computePositionedLogicalWidthReplaced):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Widget::frameRect shouldn't be virtual
https://bugs.webkit.org/show_bug.cgi?id=68637

Reviewed by Sam Weinig.

Nobody overrides Widget::frameRect and nobody should.

* platform/Widget.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove FindSafari

It isn't used or built anymore.

Fixes <http://webkit.org/b/68628>

Reviewed by Darin Adler.

* FindSafari: Removed this directory and its contents.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove FindSafari from all our .sln files

It isn't used anymore, so there's no point in building it.

Part of <http://webkit.org/b/68628> Remove FindSafari

Reviewed by Steve Falkenburg.

Source/JavaScriptCore:

* JavaScriptCore.vcproj/JavaScriptCore.sln:

Source/WebKit/win:

* WebKit.vcproj/WebKit.sln: Also relinearized the build order while I was at it.

Tools:

* DumpRenderTree/DumpRenderTree.sln:
* WebKitTestRunner/WebKitTestRunner.sln:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove all uses of FindSafari

It isn't really needed anymore.

Fixes <http://webkit.org/b/68626>.

Reviewed by Darin Adler.

* Scripts/old-run-webkit-tests: Moved read/writeRegistryString from here to webkitdirs.pm.

* Scripts/run-webkit-nightly.cmd: Removed. This used to be used to launch nightly builds of
WebKit, but now we use WebKit.exe for that.

* Scripts/webkitdirs.pm:
(installedSafariPath): Changed to use readRegistryString instead of invoking FindSafari to
do the same thing.

(readRegistryString):
(writeRegistryString):
Moved here from old-run-webkit-tests.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Use the "direct object" form of system() to run WebKit.exe

This prevents the shell from parsing (and possibly splitting, if it contains spaces) the
path to WebKit.exe.

Fixes <http://webkit.org/b/68623> run-safari doesn't work if there are spaces in the path to
%WEBKITOUTPUTDIR%

Reviewed by Steve Falkenburg.

* Scripts/webkitdirs.pm:
(runSafari):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Quote the WEBKITLIBRARIESDIR path before passing it to the shell/cygpath

Fixes <http://webkit.org/b/68621> build-webkit goes totally haywire on Windows if there are
spaces in the path to the WebKit source tree

Reviewed by John Sullivan.

* Scripts/webkitdirs.pm:
(setupCygwinEnv):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Add testRegExp.vcproj to WebKit.sln

Fixes <http://webkit.org/b/68403> Windows bots don't build testRegExp.vcproj

Reviewed by Brian Weinstein.

* WebKit.vcproj/WebKit.sln: Added testRegExp. Made it build just after jsc, just like it
does when built as part of JavaScriptCore.sln.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

[Qt] Use same DPI for application font as rest of app in DRT and WTR

QApplication will initialize the default application font based
on the application DPI at construction time, but we then override
the application DPI using QX11Info (hard-coding it to 96 for
consistency). This hard-coding is not reflected in the application
font, so we explicitly have update the font ourselves.

The 6 test results that are updated were originally produced with
a DPI of 75, as this is the default fallback DPI when a QFont is
constructed before QApplication. This was wrong, and the results
are updated to match a DPI of 96.

Reviewed by Csaba Osztrogonác.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Ref protect shaders in V8WebGLRenderingContext::getAttachedShadersCallback
https://bugs.webkit.org/show_bug.cgi?id=68630

Patch by Sergey Glazunov <serg.glazunov@gmail.com> on 2011-09-22
Reviewed by Adam Barth.

Source/WebCore:

Test: fast/canvas/webgl/shader-deleted-by-accessor.html

* bindings/js/JSWebGLRenderingContextCustom.cpp:
(WebCore::JSWebGLRenderingContext::getAttachedShaders):
* bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
(WebCore::V8WebGLRenderingContext::getAttachedShadersCallback):
* html/canvas/WebGLRenderingContext.cpp:
(WebCore::WebGLRenderingContext::getAttachedShaders):
* html/canvas/WebGLRenderingContext.h:

LayoutTests:

* fast/canvas/webgl/shader-deleted-by-accessor-expected.txt: Added.
* fast/canvas/webgl/shader-deleted-by-accessor.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed, platform-specific results for r95721.

* platform/mac/fast/css/outline-narrowLine-expected.txt: Added.
* platform/gtk/fast/css/outline-narrowLine-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

CSS 2.1 failure: abspos-non-replaced-width-margin-000, abspos-replaced-width-margin-000
https://bugs.webkit.org/show_bug.cgi?id=47148

Reviewed by David Hyatt.

Source/WebCore:

* rendering/RenderBlock.h:
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::startOffsetForLine): Return offset from right, rather than offset from left + logicalwidth
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::startAlignedOffsetForLine):
      Return the correct alignment for RTL blocks too. Looks like neither FF or Opera do this correctly.
      Covered by new test left-right-center-inline-alignment-in-ltr-and-rtl-blocks.html
* rendering/RenderBox.cpp:
(WebCore::computeInlineStaticDistance):
    - The CSS 2.1 tests now require UAs to use the parent block's direction (rather than the
      containing block's direction) when computing static distance. The container block's direction
      is still used in the other width and offset related calculations.
    - In the context of this function, the logicalRight value is just an element that will be used later to calculate
      the correct logicalLeft position for the RTL block. So whereas an LTR block can substract containerBlock->borderLogicalLeft()
      in this function directly, an RTL block has to add it here so that will be later subtracted from availableSpace to get
      logicalLeft in computePositionedLogicalWidthUsing().

(WebCore::RenderBox::computePositionedLogicalWidth):
   - Update comments and use the container block's direction except when calculating the static distance.
   - Remove the notion of a quirks mode in this function and for calculating the position of replaced and non-replaced
     blocks generally. This is in line with both FF4 and Opera - neither of which have a quirks mode for them.
(WebCore::RenderBox::computePositionedLogicalWidthUsing):
   - Update comments and use the container block's direction except when calculating the static distance.
(WebCore::RenderBox::computePositionedLogicalHeight):
   - Update comments.
(WebCore::RenderBox::computePositionedLogicalWidthReplaced):
   - Update comments and use the container block's direction except when calculating the static distance.
   - If the containing block is RTL, always assume that values are over-constrained.
(WebCore::RenderBox::computePositionedLogicalHeightReplaced):
   - Update comments.

LayoutTests:

* platform/chromium-linux/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.png:
* platform/chromium-linux/css2.1/20110323/abspos-replaced-width-margin-000-expected.png:
  New results for CSS 2.1 reference tests. These match exactly the reference test results.

* fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks.html: Added.
* platform/chromium-linux/fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks-expected.png: Added.
* platform/chromium-linux/fast/inline/left-right-center-inline-alignment-in-ltr-and-rtl-blocks-expected.txt: Added.
  New test for left,right and center alignment of inline LTR and RTL blocks. FF and Opera don't pass this test.

* fast/block/positioning/absolute-position-direction-strict.html: Removed.
* fast/block/positioning/absolute-position-direction.html: Renamed from LayoutTests/fast/block/positioning/absolute-position-direction-quirk.html.
* platform/chromium-linux/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/chromium-linux/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/chromium-linux/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/chromium-win/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/chromium-win/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/chromium-win/fast/block/positioning/absolute-position-direction-expected.txt: Renamed from LayoutTests/platform/chromium-win/fast/block/positioning/absolute-position-direction-quirk-expected.txt.
* platform/chromium-win/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/chromium-win/fast/block/positioning/absolute-position-direction-strict-expected.txt: Removed.
* platform/gtk/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/gtk/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/gtk/fast/block/positioning/absolute-position-direction-expected.txt: Renamed from LayoutTests/platform/gtk/fast/block/positioning/absolute-position-direction-quirk-expected.txt.
* platform/gtk/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/gtk/fast/block/positioning/absolute-position-direction-strict-expected.txt: Removed.
* platform/mac-leopard/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/mac-leopard/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/mac-leopard/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/mac/css2.1/20110323/abspos-non-replaced-width-margin-000-expected.txt:
* platform/mac/css2.1/20110323/abspos-replaced-width-margin-000-expected.txt:
* platform/mac/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/mac/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/mac/fast/block/positioning/absolute-position-direction-expected.txt: Renamed from LayoutTests/platform/mac/fast/block/positioning/absolute-position-direction-quirk-expected.txt.
* platform/mac/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/mac/fast/block/positioning/absolute-position-direction-strict-expected.txt: Removed.
* platform/qt/fast/block/positioning/absolute-position-direction-expected.png: Renamed from LayoutTests/platform/qt/fast/block/positioning/absolute-position-direction-quirk-expected.png.
* platform/qt/fast/block/positioning/absolute-position-direction-expected.txt: Renamed from LayoutTests/platform/qt/fast/block/positioning/absolute-position-direction-quirk-expected.txt.
* platform/qt/fast/block/positioning/absolute-position-direction-strict-expected.png: Removed.
* platform/qt/fast/block/positioning/absolute-position-direction-strict-expected.txt: Removed.
  These tests are superseded by abspos-non-replaced-width-margin-000.htm. Using the parent() block's
  direction to compute static distance is now a strict requirement for CSS 2.1, rather than an IE quirk.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Get rid of Widget::setBoundsSize
https://bugs.webkit.org/show_bug.cgi?id=68491

Reviewed by Sam Weinig.

This basically reverts http://trac.webkit.org/changeset/79167 and http://trac.webkit.org/changeset/75897
which were added to make zoom gestures work with WebKit1, but since we don't support zoom gestures in WebKit1
anymore, it's better for code cleanliness to get rid of Widget::setBoundsSize and the associated code. While this
will unfortunately break transformed iframes in WebKit1, but Simon says that it's an acceptable tradeoff.

Source/WebCore:

* WebCore.exp.in:
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::open):
* page/FrameView.cpp:
(WebCore::FrameView::create):
* platform/ScrollView.cpp:
(WebCore::ScrollView::visibleContentRect):
(WebCore::ScrollView::updateScrollbars):
(WebCore::ScrollView::wheelEvent):
* platform/ScrollView.h:
* platform/Widget.cpp:
* platform/Widget.h:
(WebCore::Widget::resize):
* platform/chromium/PopupContainer.cpp:
(WebCore::PopupContainer::refresh):
* platform/gtk/ScrollViewGtk.cpp:
(WebCore::ScrollView::visibleContentRect):
* platform/mac/WidgetMac.mm:
(WebCore::Widget::paint):
* rendering/RenderWidget.cpp:
(WebCore::RenderWidget::setWidgetGeometry):
(WebCore::RenderWidget::setWidget):
(WebCore::RenderWidget::updateWidgetPosition):
* rendering/RenderWidget.h:

Source/WebKit/mac:

* Plugins/Hosted/WebHostedNetscapePluginView.mm:
(-[WebHostedNetscapePluginView updateAndSetWindow]):
* WebView/WebFrameView.mm:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

https://bugs.webkit.org/show_bug.cgi?id=68590

Floats pushed to next page, column or region don't reposition properly if the amount of
available logical width at the new position changes. Refactor the code so that we can
run the float placement algorithm again when this happens.

Source/WebCore:

Covered by an existing regions test that exposes the issue.

Reviewed by Adam Roben.

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::computeLogicalLocationForFloat):
(WebCore::RenderBlock::positionNewFloats):
* rendering/RenderBlock.h:

LayoutTests:

Reviewed by Adam Roben.

* fast/regions/webkit-flow-floats-inside-regions-bounds-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@95724 268f45cc-cd09-0410-ab3c-d52691b4dbfc