WebKit-https.git
9 years ago[Windows, WinCairo] Unreviewed VS2010 Build Correction
bfulgham@webkit.org [Wed, 3 Apr 2013 18:27:53 +0000 (18:27 +0000)]
[Windows, WinCairo] Unreviewed VS2010 Build Correction

* win/tools/scripts/auto-version2010.sh: Added properties svn:eol-style and svn:executable.
* win/tools/scripts/feature-defines2010.sh: Added properties svn:eol-style and svn:executable.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147569 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago [WK2] XMLHttpRequest upload events don't work with NetworkProcess
ap@apple.com [Wed, 3 Apr 2013 18:23:34 +0000 (18:23 +0000)]
    [WK2] XMLHttpRequest upload events don't work with NetworkProcess
        https://bugs.webkit.org/show_bug.cgi?id=113887

        Reviewed by Brady Eidson.

        Just forward didSendData client call to WebProcess.

        * NetworkProcess/NetworkResourceLoader.cpp:
        (WebKit::NetworkResourceLoader::didSendData):
        (WebKit::NetworkResourceLoader::didCancelAuthenticationChallenge): Tweaked a comment.
        * WebProcess/Network/WebResourceLoader.cpp:
        (WebKit::WebResourceLoader::didSendData):
        * WebProcess/Network/WebResourceLoader.h:
        * WebProcess/Network/WebResourceLoader.messages.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147568 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Gstreamer] Use gst_buffer_extract() in copyGstreamerBuffersToAudioChannel()
ch.dumez@sisa.samsung.com [Wed, 3 Apr 2013 18:12:40 +0000 (18:12 +0000)]
[Gstreamer] Use gst_buffer_extract() in copyGstreamerBuffersToAudioChannel()
https://bugs.webkit.org/show_bug.cgi?id=113880

Reviewed by Philippe Normand.

copyGstreamerBuffersToAudioChannel() was mapping the GstBuffer content to memcpy
it to the AudioChannel buffer. This patch leverages gst_buffer_extract() to
simplify the code as it does exactly what we need: gst_buffer_map, memcpy,
gst_buffer_unmap and error handling.

Also replace GstBuffer NULL check by an assertion as we already make sure they
are not NULL before adding them to the list. Additionally, we now call
audioChannel->mutableData() only once instead of once per iteration. It is
a bit better as the getter does some work internally.

No new tests, no behavior change.

* platform/audio/gstreamer/AudioFileReaderGStreamer.cpp:
(WebCore::copyGstreamerBuffersToAudioChannel):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147567 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: DOM.highlightRect() and DOM.highlightQuad() should allow for page...
graouts@apple.com [Wed, 3 Apr 2013 17:30:32 +0000 (17:30 +0000)]
Web Inspector: DOM.highlightRect() and DOM.highlightQuad() should allow for page coordinates
https://bugs.webkit.org/show_bug.cgi?id=113869

Add a new "usePageCoordinates" parameter to DOM.highlightRect() and DOM.highlightQuad()
allowing to specify that the rect or quad to highlight should be converted to page
coordinates instead of using the default viewport coordinates.

Reviewed by Timothy Hatcher.

* inspector/Inspector.json:
* inspector/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::highlightRect):
(WebCore::InspectorDOMAgent::highlightQuad):
(WebCore::InspectorDOMAgent::innerHighlightQuad):
Store the "usePageCoordinates" parameter obtained from the protocol calls onto the
highlight configuration object so that it may be used within the InspectorOverlay.

* inspector/InspectorDOMAgent.h:
(InspectorDOMAgent):
* inspector/InspectorOverlay.cpp:
(WebCore::InspectorOverlay::highlightQuad):
Check the highlight configuration for the "usePageCoordinates" parameter and
apply the inverse of the page's scroll offset to the quad if it's true.

* inspector/InspectorOverlay.h:
(HighlightConfig):
(WebCore::Highlight::setDataFromConfig):
(Highlight):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147566 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd omitted reference operator of VisiblePosition in AccessibilityObject
kangil.han@samsung.com [Wed, 3 Apr 2013 16:43:32 +0000 (16:43 +0000)]
Add omitted reference operator of VisiblePosition in AccessibilityObject
https://bugs.webkit.org/show_bug.cgi?id=113879

Reviewed by Darin Adler.

Seems reference operator has been omitted in startOfStyleRange function, so add it.

* accessibility/AccessibilityObject.cpp:
(WebCore::startOfStyleRange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147565 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] Include Pango in jhbuild-optional.modules
commit-queue@webkit.org [Wed, 3 Apr 2013 16:35:47 +0000 (16:35 +0000)]
[GTK] Include Pango in jhbuild-optional.modules
https://bugs.webkit.org/show_bug.cgi?id=113878

Patch by Manuel Rego Casasnovas <rego@igalia.com> on 2013-04-03
Reviewed by Martin Robinson.

Since r147547 Pango 1.32.0 is required. Including the module in the
JHBuild optional moduleset.

* gtk/jhbuild-optional.modules: Add Pango module.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147564 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Add getter for JSC TestRunner to DumpRenderTreeQt.
zeno.albisser@digia.com [Wed, 3 Apr 2013 16:27:46 +0000 (16:27 +0000)]
[Qt] Add getter for JSC TestRunner to DumpRenderTreeQt.
https://bugs.webkit.org/show_bug.cgi?id=113864

Moving setDefersLoading from TestRunnerQt to TestRunner
requires access to the non-Qt TestRunner.
As long as we instantiate both TestRunner and TestRunnerQt,
we need a separate getter for the non-Qt TestRunner instance.

Reviewed by Simon Hausmann.

* DumpRenderTree/qt/DumpRenderTreeQt.cpp:
(WebPage::resetSettings):
* DumpRenderTree/qt/DumpRenderTreeQt.h:
(DumpRenderTree::jscTestRunner):
* DumpRenderTree/qt/TestRunnerQt.cpp:
(TestRunner::setDefersLoading):
    Moving setDefersLoading from TestRunnerQt
    to TestRunner.
* DumpRenderTree/qt/TestRunnerQt.h:
(TestRunnerQt):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147563 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdate the localizable strings file.
jberlin@webkit.org [Wed, 3 Apr 2013 16:12:21 +0000 (16:12 +0000)]
Update the localizable strings file.

Rubber-stamped by Michael Saboff.

* English.lproj/Localizable.strings:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147562 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBUILD FIX (r147493): Fix number of arguments to AudioDestinationIOS constructor
ddkilzer@apple.com [Wed, 3 Apr 2013 15:58:59 +0000 (15:58 +0000)]
BUILD FIX (r147493): Fix number of arguments to AudioDestinationIOS constructor

Fixes the following build failures:

    Source/WebCore/platform/audio/ios/AudioDestinationIOS.cpp:84:25: error: no matching constructor for initialization of 'WebCore::AudioDestinationIOS'
        return adoptPtr(new AudioDestinationIOS(callback, sampleRate));
                            ^                   ~~~~~~~~~~~~~~~~~~~~
    In file included from Source/WebCore/platform/audio/ios/AudioDestinationIOS.cpp:36:
    Source/WebCore/platform/audio/ios/AudioDestinationIOS.h:44:5: note: candidate constructor not viable: requires 3 arguments, but 2 were provided
        AudioDestinationIOS(AudioIOCallback&, unsigned numberOfOutputChannels, double sampleRate);
        ^
    Source/WebCore/platform/audio/ios/AudioDestinationIOS.h:42:7: note: candidate constructor (the implicit copy constructor) not viable: requires 1 argument, but 2 were provided
    class AudioDestinationIOS : public AudioDestination, public AudioSessionListener {
          ^
    Source/WebCore/platform/audio/ios/AudioDestinationIOS.cpp:101:22: error: out-of-line definition of 'AudioDestinationIOS' does not match any declaration in 'WebCore::AudioDestinationIOS'
    AudioDestinationIOS::AudioDestinationIOS(AudioIOCallback& callback, double sampleRate)
                         ^~~~~~~~~~~~~~~~~~~
    2 errors generated.

* platform/audio/ios/AudioDestinationIOS.h:
(AudioDestinationIOS): Remove numberOfOutputChannels argument.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147561 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBUILD FIX (r147494): -[AVAudioSession setCategory:error:] takes NSError**
ddkilzer@apple.com [Wed, 3 Apr 2013 15:51:20 +0000 (15:51 +0000)]
BUILD FIX (r147494): -[AVAudioSession setCategory:error:] takes NSError**

Fixes the following build failure:

    Source/WebCore/platform/audio/ios/AudioSessionIOS.mm:151:71: error: cannot initialize a parameter of type 'NSError **' with an lvalue of type 'NSError *'
        [[AVAudioSession sharedInstance] setCategory:categoryString error:error];
                                                                          ^~~~~

* platform/audio/ios/AudioSessionIOS.mm:
(WebCore::AudioSession::setCategory): Add '&' operator.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147560 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK][WK2] Unskip /webkit2/WebKitWebContext/uri-scheme
commit-queue@webkit.org [Wed, 3 Apr 2013 15:45:13 +0000 (15:45 +0000)]
[GTK][WK2] Unskip /webkit2/WebKitWebContext/uri-scheme
https://bugs.webkit.org/show_bug.cgi?id=104779

Patch by Manuel Rego Casasnovas <rego@igalia.com> on 2013-04-03
Reviewed by Carlos Garcia Campos.

* Scripts/run-gtk-tests:
(TestRunner): Unskip /webkit2/WebKitWebContext/uri-scheme as it is
passing in current trunk.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147559 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCSP blocks inline style when cloning a node
commit-queue@webkit.org [Wed, 3 Apr 2013 15:45:10 +0000 (15:45 +0000)]
CSP blocks inline style when cloning a node
https://bugs.webkit.org/show_bug.cgi?id=112270

Patch by Felipe Zimmerle <felipe@zimmerle.org> on 2013-04-03
Reviewed by Adam Barth.

Source/WebCore:

Checks if the Element is being cloned, if so, the application of the
style is allowed otherwise it relies on default permission mechanism.

Test: http/tests/security/contentSecurityPolicy/inline-style-allowed-while-cloning-objects.html

* dom/Element.cpp:
(WebCore::Element::attributeChanged): Added parameter
AttributeModificationReason to the method signature.
(WebCore::Element::cloneAttributesFromElement): It is now calling
attributeChanged with AttributeModificationReason parameter. In this scope
it is always set to ModifiedByCloning.
(WebCode::Element::attributeChangedFromParserOrByCloning):
Added parameter AttributeModificationReason.
* dom/Element.h: Added AttributeModificationReason enum. Used to specify
whenever an attribute was set by a cloned oject or directly.
AttributeModificationReason added to attributeChanged with the default
value set to ModifiedDirectly.
* dom/StyledElement.cpp:
(WebCore::StyledElement::attributeChanged): Added
AttributeModificationReason to the method signature.
(WebCore::StyledElement::styleAttributeChanged): Now it is checking the
reason of the update, if the reason is ModifiedDirectly check CSP
before set. If ModifiedByCloning set the attribute ignoring the CSP
policy.
* dom/StyledElement.h:
(StyledElement): Added AttributeModificationReason parameter to the methods:
attributeChanged and styleAttributeChanged, attributeChanged has
ModifiedDirectly as default value.
* svg/SVGElement.cpp:
(WebCore::SVGElement::attributeChanged): Added AttributeModificationReason
parameter to the method signature.
* svg/SVGElement.h:
(SVGElement): Added AttributeModificationReason parameter to the
attributeChanged signature with the default value set to:
ModifiedDirectly.

LayoutTests:

* http/tests/security/contentSecurityPolicy/inline-style-allowed-while-cloning-objects-expected.txt: Added.
* http/tests/security/contentSecurityPolicy/inline-style-allowed-while-cloning-objects.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147558 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] WebKitWebNavigationAction: fix mix between dispose and finalize
commit-queue@webkit.org [Wed, 3 Apr 2013 15:18:07 +0000 (15:18 +0000)]
[GTK] WebKitWebNavigationAction: fix mix between dispose and finalize
https://bugs.webkit.org/show_bug.cgi?id=113794

Patch by Alban Crequy <alban.crequy@collabora.co.uk> on 2013-04-03
Reviewed by Gustavo Noronha Silva.

WebKitWebNavigationAction's dispose function was named finalize and was calling
the parent class's finalise function, causing a memory corruption. Change it to
be a proper finalize function instead.

* webkit/webkitwebnavigationaction.cpp:
(webkit_web_navigation_action_class_init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147557 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed 32bit buildfix after r147542.
ossy@webkit.org [Wed, 3 Apr 2013 15:11:29 +0000 (15:11 +0000)]
Unreviewed 32bit buildfix after r147542.

* Modules/websockets/WebSocket.cpp:
(WebCore::WebSocket::didReceiveBinaryData):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147556 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Gstreamer] Avoid calls to g_slist_nth_data in webKitWebAudioSrcLoop()
ch.dumez@sisa.samsung.com [Wed, 3 Apr 2013 15:09:06 +0000 (15:09 +0000)]
[Gstreamer] Avoid calls to g_slist_nth_data in webKitWebAudioSrcLoop()
https://bugs.webkit.org/show_bug.cgi?id=113875

Reviewed by Philippe Normand.

webKitWebAudioSrcLoop() was iterating over 2 GSLists by using a counter and then
calling g_slist_nth_data() to get the element of each GSList. This is inefficient
because calling g_slist_nth_data() will iterate the GSList up until index 'n'.

This patch improves this by keeping pointers to the current element of each list
while iterating, so that we can simply use g_slist_next() to iterate instead
of g_slist_nth_data().

No new tests, no behavior change.

* platform/audio/gstreamer/WebKitWebAudioSourceGStreamer.cpp:
(webKitWebAudioSrcLoop):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147555 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoActions that require user gesture don't work in window.showModalDialog in Chromium
keishi@webkit.org [Wed, 3 Apr 2013 15:04:28 +0000 (15:04 +0000)]
Actions that require user gesture don't work in window.showModalDialog in Chromium
https://bugs.webkit.org/show_bug.cgi?id=113777

Reviewed by Jochen Eisinger.

In Chromium the window for showModalDialog is not modal which causes the
UserGestureIndicator to block actions in the sub window. We introduce a
RAII that saves the UserGestureIndicator and state, then resets it before
runModalDialog, and restore it when it's done.

No new tests. Can't cause user gesture inside modal dialog.

* dom/UserGestureIndicator.cpp:
(WebCore::UserGestureIndicatorDisabler::UserGestureIndicatorDisabler): Saves and resets the topmost indicator and state.
(WebCore):
(WebCore::UserGestureIndicatorDisabler::~UserGestureIndicatorDisabler): Restores topmost indicator and state.
* dom/UserGestureIndicator.h:
(WebCore):
(UserGestureIndicatorDisabler): RAII to temporarily disable UserGestureIndicator.
(UserGestureIndicator): Allow access to UserGestureIndicatorDisabler.
* page/DOMWindow.cpp:
(WebCore::DOMWindow::showModalDialog): Disable UserGestureIndicator while runModalDialog.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147554 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Fail gracefully if an OpenGL context could not be created.
zeno.albisser@digia.com [Wed, 3 Apr 2013 15:02:37 +0000 (15:02 +0000)]
[Qt] Fail gracefully if an OpenGL context could not be created.
https://bugs.webkit.org/show_bug.cgi?id=113784

In case we fail to create or adopt a valid platform OpenGL context
we do not want to provide a GraphicsContext3D instance.
Otherwise we would crash as soon as the instance is being used.

Reviewed by Jocelyn Turcotte.

* platform/graphics/qt/GraphicsContext3DQt.cpp:
(WebCore::GraphicsContext3DPrivate::GraphicsContext3DPrivate):
(WebCore::GraphicsContext3D::GraphicsContext3D):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147553 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[BlackBerry] Remove defaultViewportArguments check from call site of recomputeVirtual...
anilsson@rim.com [Wed, 3 Apr 2013 14:36:00 +0000 (14:36 +0000)]
[BlackBerry] Remove defaultViewportArguments check from call site of recomputeVirtualViewportFromViewportArguments
https://bugs.webkit.org/show_bug.cgi?id=113872

Reviewed by Carlos Garcia Campos.

Carlos pointed out that recomputeVirtualViewportFromViewportArguments
already contains the check, and returns an empty IntSize if it fails.
So pick up on that to make the code a bit more DRY.

* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::setViewportSize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147552 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Unreviewed gardening. Updated platform specific expected files after r147492.
kadam@inf.u-szeged.hu [Wed, 3 Apr 2013 14:23:34 +0000 (14:23 +0000)]
[Qt] Unreviewed gardening. Updated platform specific expected files after r147492.

* platform/qt/TestExpectations:
* platform/qt/fast/dom/HTMLImageElement/image-alt-text-expected.png:
* platform/qt/fast/dom/HTMLImageElement/image-alt-text-expected.txt:
* platform/qt/fast/dom/HTMLInputElement/input-image-alt-text-expected.png:
* platform/qt/fast/dom/HTMLInputElement/input-image-alt-text-expected.txt:
* platform/qt/fast/encoding/utf-16-big-endian-expected.png:
* platform/qt/fast/encoding/utf-16-big-endian-expected.txt:
* platform/qt/fast/encoding/utf-16-little-endian-expected.png:
* platform/qt/fast/encoding/utf-16-little-endian-expected.txt:
* platform/qt/fast/forms/input-value-expected.png: Added.
* platform/qt/fast/forms/input-value-expected.txt:
* platform/qt/fast/lists/inlineBoxWrapperNullCheck-expected.png:
* platform/qt/fast/lists/inlineBoxWrapperNullCheck-expected.txt:
* platform/qt/tables/mozilla/collapsing_borders/bug41262-3-expected.png:
* platform/qt/tables/mozilla/collapsing_borders/bug41262-3-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147551 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix HAVE(QTTESTSUPPORT) logic
pierre.rossi@gmail.com [Wed, 3 Apr 2013 14:15:49 +0000 (14:15 +0000)]
Fix HAVE(QTTESTSUPPORT) logic

build_qttestsupport has not been added to WEBKIT_CONFIG in detectFeatures,
resulting in HAVE(QTTESTSUPPORT) never being true.
Move that logic down in sanitizeFeatures.

Patch by Pierre Rossi <pierre.rossi@digia.com> on 2013-04-03
Rubber-stamped by Tor Arne Vestbø.

* qmake/mkspecs/features/features.prf:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147550 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK][AC] Implement matrix keyframe animations with clutter ac backend
commit-queue@webkit.org [Wed, 3 Apr 2013 13:23:26 +0000 (13:23 +0000)]
[GTK][AC] Implement matrix keyframe animations with clutter ac backend
https://bugs.webkit.org/show_bug.cgi?id=110314

Patch by ChangSeok Oh <changseok.oh@collabora.com> on 2013-04-03
Reviewed by Gustavo Noronha Silva.

Clutter 1.12 doesn't support additive transform animations yet, So clutter ac backend
uses matrix animations for the case. This patch follows changeset 143369, 143343
to support matrix keyframe animations. I believe this change is easy if you understand
above two changesets.

Covered by existing animation tests.

* platform/graphics/clutter/PlatformClutterAnimation.cpp:
(WebCore::PlatformClutterAnimation::setValues):
(WebCore::PlatformClutterAnimation::addClutterKeyframeTransitionForProperty):
(WebCore):
(WebCore::PlatformClutterAnimation::addTransformTransition):
* platform/graphics/clutter/PlatformClutterAnimation.h:
(PlatformClutterAnimation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147549 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoShould close select popup when the element loses focus
kov@webkit.org [Wed, 3 Apr 2013 13:20:29 +0000 (13:20 +0000)]
Should close select popup when the element loses focus
https://bugs.webkit.org/show_bug.cgi?id=113220

Patch by Gustavo Noronha Silva <gustavo.noronha@collabora.com> on 2013-04-03
Reviewed by Kent Tamura.

Source/WebCore:

Test: fast/forms/select-popup-closes-on-blur.html

* html/HTMLSelectElement.cpp:
(WebCore::HTMLSelectElement::menuListDefaultEventHandler): handle the blur event, closing
the popup menu if any.
* testing/Internals.cpp:
(WebCore::Internals::isSelectPopupVisible): utility to query the select node renderer to
find out whether the popup is open or closed.
(WebCore):
* testing/Internals.h: add the new method definition.
* testing/Internals.idl: ditto.

Source/WebKit/win:

* WebKit.vcproj/WebKitExports.def.in: export WebCore::HTMLNames::selectTag for Internals

LayoutTests:

* fast/forms/select/popup-closes-on-blur-expected.txt: Added.
* fast/forms/select/popup-closes-on-blur.html: Added.
* platform/mac/TestExpectations:
* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147548 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] Bump required versions of some dependencies
commit-queue@webkit.org [Wed, 3 Apr 2013 13:20:28 +0000 (13:20 +0000)]
[GTK] Bump required versions of some dependencies
https://bugs.webkit.org/show_bug.cgi?id=113282

Patch by Carlos Garcia Campos <cgarcia@igalia.com> on 2013-04-03
Reviewed by Martin Robinson.

.:

This patch updates the following dependencies:

 - libsoup 2.42.0: is the current stable release.

 - glib 2.36.0: libsoup 2.42 requires 2.35.0, but we don't want to
   depend on unstable versions and 2.36 is the next stable version
   after 2.35.

 - gtk+ 3.6.0: this is the previous stable release. We don't
   currently have any code using newer API, so we will be able to
   remove all the #ifdefed code.

 - at-spi2 2.5.3: is the version required by gtk+ 3.6.

 - GObject introspection 1.32.0: is the version required by gtk+ 3.6.

 - Pango 1.32.0: it's the current stable version and the first one
   depending on Harfbuzz.

 - Fontconfig 2.5: required by pango 1.32.0 when using Harfbuzz.

 - gtk+ 2.24.10: it allows us to remove a lot of deprecated code
   and #ifdefs.

* Source/autotools/Versions.m4:

Tools:

* gtk/jhbuild.modules: Update libsoup, glib, at-spi2 and gtk+ versions.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147547 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] gst-plugins-good fails to build with newer Linux(v4l2)
kov@webkit.org [Wed, 3 Apr 2013 13:12:22 +0000 (13:12 +0000)]
[GTK] gst-plugins-good fails to build with newer Linux(v4l2)
https://bugs.webkit.org/show_bug.cgi?id=113809

Patch by Gustavo Noronha Silva <gustavo.noronha@collabora.com> on 2013-04-03
Reviewed by Philippe Normand.

Newer Linux kernels have removed some deprecated macros from v4l2
headers, causing gst-plugins-good's build to break. We don't need it
right now, so just disable it.

* gtk/jhbuild.modules: disable v4l2 usage in gst-plugins-good.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147546 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Switch ContextMenu implementation to CROSS_PLATFORM_CONTEXT_MENU
jesus@webkit.org [Wed, 3 Apr 2013 12:44:36 +0000 (12:44 +0000)]
[Qt] Switch ContextMenu implementation to CROSS_PLATFORM_CONTEXT_MENU
https://bugs.webkit.org/show_bug.cgi?id=113535

Reviewed by Simon Hausmann.

Source/WebCore:

Now Qt uses the CROSS_PLATFORM_CONTEXT_MENUS
USE flag. This flag provides a full cross-platform
representation of a ContextMenu and a ContextMenuItem.
The embedder can then decide how to show this, so neither
WebCore nor WebCore/platform need to know any platform
specifics about menus.

No new tests, no behavior changes.

* Target.pri:
* platform/ContextMenu.h:
(ContextMenu):
* platform/ContextMenuItem.h:
* platform/qt/ContextMenuItemQt.cpp:
(WebCore::ContextMenuItem::platformContextMenuItem):
* platform/qt/ContextMenuQt.cpp:
(WebCore::ContextMenu::ContextMenu):
(WebCore::ContextMenu::getContextMenuItems):
(WebCore::ContextMenu::createPlatformContextMenuFromItems):
(WebCore::ContextMenu::platformContextMenu):

Source/WebKit/qt:

Adjust ContextMenuClientQt and QWebPageAdapter to use CROSS_PLATFORM_CONTEXT_MENUS
USE flag by implementing customizeMenu() instead of getCustomMenuFromDefaultItems()
and by calling menu->items() instead of menu->platformDescription().

* WebCoreSupport/ContextMenuClientQt.cpp:
(WebCore::ContextMenuClientQt::customizeMenu):
* WebCoreSupport/ContextMenuClientQt.h:
(ContextMenuClientQt):
* WebCoreSupport/QWebPageAdapter.cpp:
(descriptionForPlatformMenu):
(QWebPageAdapter::updatePositionDependentMenuActions):

Tools:

Adding USE_CROSS_PLATFORM_CONTEXT_MENUS to WEBKIT_CONFIG.

* qmake/mkspecs/features/features.prf:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147545 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Monospace font does not render in proper sizes
allan.jensen@digia.com [Wed, 3 Apr 2013 12:35:36 +0000 (12:35 +0000)]
[Qt] Monospace font does not render in proper sizes
https://bugs.webkit.org/show_bug.cgi?id=93263

Reviewed by Jocelyn Turcotte.

Remove this patch's artificial dependency on Qt 5.1.

* platform/graphics/qt/FontPlatformDataQt.cpp:
(WebCore::FontPlatformData::FontPlatformData):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147544 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Move logic of simple forwarding functions to TestRunner.
zeno.albisser@digia.com [Wed, 3 Apr 2013 12:33:41 +0000 (12:33 +0000)]
[Qt] Move logic of simple forwarding functions to TestRunner.
https://bugs.webkit.org/show_bug.cgi?id=113863

Reviewed by Simon Hausmann.

* DumpRenderTree/qt/TestRunnerQt.cpp:
(TestRunner::setTextDirection):
(TestRunner::goBack):
(TestRunner::evaluateScriptInIsolatedWorld):
(TestRunner::addUserStyleSheet):
* DumpRenderTree/qt/TestRunnerQt.h:
(TestRunnerQt):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147543 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRefine LOG messages in WebSocket related components
commit-queue@webkit.org [Wed, 3 Apr 2013 12:15:54 +0000 (12:15 +0000)]
Refine LOG messages in WebSocket related components
https://bugs.webkit.org/show_bug.cgi?id=113852

Patch by Takeshi Yoshino <tyoshino@chromium.org> on 2013-04-03
Reviewed by Kent Tamura.

Refine LOG messages by
- Make them contain method names to ease identifying where the event occurred.
- For consistency, make all messages print the pointer of the origin instance.
- Quote UTF8 data in LOG messages.
- For overloaded methods, add the type to clarify which one is run

* Modules/websockets/WebSocket.cpp:
(WebCore::WebSocket::connect):
(WebCore::WebSocket::send):
(WebCore::WebSocket::close):
(WebCore::WebSocket::contextDestroyed):
(WebCore::WebSocket::didConnect):
(WebCore::WebSocket::didReceiveMessage):
(WebCore::WebSocket::didReceiveBinaryData):
(WebCore::WebSocket::didReceiveMessageError):
(WebCore::WebSocket::didUpdateBufferedAmount):
(WebCore::WebSocket::didStartClosingHandshake):
(WebCore::WebSocket::didClose):
* Modules/websockets/WebSocketChannel.cpp:
(WebCore::WebSocketChannel::connect):
(WebCore::WebSocketChannel::subprotocol):
(WebCore::WebSocketChannel::extensions):
(WebCore::WebSocketChannel::send):
(WebCore::WebSocketChannel::bufferedAmount):
(WebCore::WebSocketChannel::close):
(WebCore::WebSocketChannel::fail):
(WebCore::WebSocketChannel::disconnect):
(WebCore::WebSocketChannel::willOpenSocketStream):
(WebCore::WebSocketChannel::didOpenSocketStream):
(WebCore::WebSocketChannel::didCloseSocketStream):
(WebCore::WebSocketChannel::didReceiveSocketStreamData):
(WebCore::WebSocketChannel::didFailSocketStream):
(WebCore::WebSocketChannel::didStartLoading):
(WebCore::WebSocketChannel::didReceiveData):
(WebCore::WebSocketChannel::didFinishLoading):
(WebCore::WebSocketChannel::didFail):
(WebCore::WebSocketChannel::appendToBuffer):
(WebCore::WebSocketChannel::processBuffer):
(WebCore::WebSocketChannel::startClosingHandshake):
(WebCore::WebSocketChannel::closingTimerFired):
* Modules/websockets/WebSocketDeflater.cpp:
(WebCore::WebSocketDeflater::~WebSocketDeflater):
(WebCore::WebSocketInflater::~WebSocketInflater):
* Modules/websockets/WebSocketHandshake.cpp:
(WebCore::WebSocketHandshake::readServerHandshake):
* platform/network/chromium/SocketStreamHandle.cpp:
(WebCore::SocketStreamHandleInternal::connect):
(WebCore::SocketStreamHandleInternal::send):
(WebCore::SocketStreamHandleInternal::close):
(WebCore::SocketStreamHandleInternal::didOpenStream):
(WebCore::SocketStreamHandleInternal::didSendData):
(WebCore::SocketStreamHandleInternal::didReceiveData):
(WebCore::SocketStreamHandleInternal::didClose):
(WebCore::SocketStreamHandleInternal::didFail):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147542 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Update QMAKE_MAC_SDK check for Qt 5.1
vestbo@webkit.org [Wed, 3 Apr 2013 11:56:44 +0000 (11:56 +0000)]
[Qt] Update QMAKE_MAC_SDK check for Qt 5.1

Relies on QMAKE_MAC_SDK_VERSION set in features.prf

Reviewed by Simon Hausmann.

* WebCore.pri:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147541 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Disable media support using QuickTime if deployment target != SDK
vestbo@webkit.org [Wed, 3 Apr 2013 11:56:24 +0000 (11:56 +0000)]
[Qt] Disable media support using QuickTime if deployment target != SDK

We link to a single libWebKitSystemInterface static lib, so unless the
deployment target matches the SDK let's not pretend like we'll be able
to run on anything lower than the SDK version.

Reviewed by Simon Hausmann.

* qmake/mkspecs/features/features.prf:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147540 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] assignToHTMLImageElement no longer exists in Qt5
commit-queue@webkit.org [Wed, 3 Apr 2013 11:15:23 +0000 (11:15 +0000)]
[Qt] assignToHTMLImageElement no longer exists in Qt5
https://bugs.webkit.org/show_bug.cgi?id=113639

Patch by Kristof Provost <kristof@codepro.be> on 2013-04-03
Reviewed by Simon Hausmann.

No new tests: covered by ./Source/WebKit/qt/tests/hybridPixmap/test.html

* bridge/qt/qt_pixmapruntime.cpp:
(JSC::Bindings::QtPixmapRuntime::getClassRef):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147539 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: crash in WebCore::InspectorLayerTreeAgent::buildObjectForLayer if...
graouts@apple.com [Wed, 3 Apr 2013 11:06:49 +0000 (11:06 +0000)]
Web Inspector: crash in WebCore::InspectorLayerTreeAgent::buildObjectForLayer if a layer is created for an anonymous RenderObject (:first-letter)
https://bugs.webkit.org/show_bug.cgi?id=113768

Source/WebCore:

The InspectorLayerTreeAgent now knows how to deal with anonymous RenderObjects
for the :first-letter and :first-line pseudo-elements.

Reviewed by Timothy Hatcher.

Test: inspector-protocol/layers/layers-anonymous.html

* inspector/Inspector.json:
Change the Layer type in the LayerTree domain to have more accurate terminology
(pseudo-element vs. pseudo-class) and a new optional isAnonymous flag for layers
associated to anonymous RenderObjects.

* inspector/InspectorLayerTreeAgent.cpp:
(WebCore::InspectorLayerTreeAgent::buildObjectForLayer):
Check for anonymous RenderObjects and set the parent renderer's node as the node
for this layer since anonymous renderers have the Document as their node and this
would not be satisfactory to show in a front-end. We also check for anonymous
RenderObjects for the :first-letter and :first-line pseudo-elements and set them
in the pseudoElement property of the Layer object created such that a front-end
could correctly identify what type of pseudo-element for the associated node
yielded this layer.

(WebCore::InspectorLayerTreeAgent::idForNode):
Here we fix the actual crash by first checking if the node provided is null and
returning 0 if there's no such node. This ensures that any scenario where there
is no node associated with the RenderObject simply informs of the front-end that
no such node exists and we do not crash.

LayoutTests:

Reviewed by Timothy Hatcher.

* inspector-protocol/layers/layers-anonymous-expected.txt: Added.
* inspector-protocol/layers/layers-anonymous.html: Added.
New test for layers created for CSS anonymous boxes or blocks.

* inspector-protocol/layers/layers-generated-content.html:
Update test to use the "pseudoElement" property instead of "pseudoClass"
which was the previous, less accurate name.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147538 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[BlackBerry] Decouple layout viewport from visual viewport
anilsson@rim.com [Wed, 3 Apr 2013 11:06:26 +0000 (11:06 +0000)]
[BlackBerry] Decouple layout viewport from visual viewport
https://bugs.webkit.org/show_bug.cgi?id=113829

Reviewed by Carlos Garcia Campos.

Setting the visual viewport would force the layout viewport equal to
the visual viewport. Decouple the viewports by allowing the user to
specify a layout viewport when setting the viewport(s). The layout
viewport could be the exact same value passed for visual viewport, if
the old behavior is desired.

Also reevaluate any viewport arguments when (only) the layout viewport
changes in WebPage::setDefaultLayoutSize() since the layout viewport is
used as input to the calculation.

* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::setScreenOrientation):
(BlackBerry::WebKit::WebPagePrivate::setViewportSize):
(BlackBerry::WebKit::WebPage::setViewportSize):
(WebKit):
(BlackBerry::WebKit::WebPagePrivate::setDefaultLayoutSize):
(BlackBerry::WebKit::WebPage::defaultLayoutSize):
* Api/WebPage.h:
* Api/WebPage_p.h:
(WebPagePrivate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147537 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Unreviewed gardening. Rebaselining after r147530.
zarvai@inf.u-szeged.hu [Wed, 3 Apr 2013 10:59:46 +0000 (10:59 +0000)]
[Qt] Unreviewed gardening. Rebaselining after r147530.

* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-deny-expected.txt:
* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-expected.txt:
* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body-expected.txt:
* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny-expected.txt:
* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-multiple-headers-sameorigin-deny-expected.txt:
* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-parent-same-origin-deny-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-deny-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-multiple-headers-sameorigin-deny-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-parent-same-origin-deny-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147536 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: [REGRESSION(r147117)][Elements] Copy/paste keyboard shortcuts broken...
apavlov@chromium.org [Wed, 3 Apr 2013 10:24:58 +0000 (10:24 +0000)]
Web Inspector: [REGRESSION(r147117)][Elements] Copy/paste keyboard shortcuts broken in "Edit as HTML"
https://bugs.webkit.org/show_bug.cgi?id=113798

Reviewed by Yury Semikhatsky.

Stop propagation of "copy" event from the multiline editor's
CodeMirror host element so that it is not handled by
WebInspector.documentCopy().

* inspector/front-end/UIUtils.js:
(.consumeCopy):
(.cleanUpAfterEditing):
* inspector/front-end/inspector.js:
(WebInspector.addMainEventListeners):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147535 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[EFL] accessibility/ignore-spacer-elements.html is failing
commit-queue@webkit.org [Wed, 3 Apr 2013 09:49:14 +0000 (09:49 +0000)]
[EFL] accessibility/ignore-spacer-elements.html is failing
https://bugs.webkit.org/show_bug.cgi?id=112001

Patch by Krzysztof Czech <k.czech@samsung.com> on 2013-04-03
Reviewed by Gyuyoung Kim.

Fixed failing test. The reason of failure is different expectation of childrenCount.

* accessibility/ignore-spacer-elements.html:
* platform/efl-wk1/TestExpectations:
* platform/efl-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147534 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWith borders of differing alpha, the corners become squared off
tasak@google.com [Wed, 3 Apr 2013 09:48:04 +0000 (09:48 +0000)]
With borders of differing alpha, the corners become squared off
https://bugs.webkit.org/show_bug.cgi?id=113744

Reviewed by Simon Fraser.

Source/WebCore:

paintTranslucentBorderSides should pick up borders with common color
in order: top, bottom, left and right (not top, right, bottom and
left), because willBeOverdrawn depends on the assumption, borders are
drawn in order: top, bottom, left and right.

Test: fast/borders/border-left-right-same-bottom-different-color.html

* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::paintTranslucentBorderSides):

LayoutTests:

* fast/borders/border-left-right-same-bottom-different-color-expected.txt: Added.
* fast/borders/border-left-right-same-bottom-different-color.html: Added.
* platform/chromium-linux/fast/borders/border-left-right-same-bottom-different-color-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147533 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Unreviewed gardening. Updated platform specific expected files.
kadam@inf.u-szeged.hu [Wed, 3 Apr 2013 09:45:49 +0000 (09:45 +0000)]
[Qt] Unreviewed gardening. Updated platform specific expected files.

* platform/qt/TestExpectations: Skipped failing reftest.
* platform/qt/compositing/filters/sw-layer-overlaps-hw-shadow-expected.txt: Added after r147502.
* platform/qt/compositing/filters/sw-nested-shadow-overlaps-hw-nested-shadow-expected.txt: Added after r147502.
* platform/qt/compositing/filters/sw-shadow-overlaps-hw-layer-expected.txt: Added after r147502.
* platform/qt/compositing/filters/sw-shadow-overlaps-hw-shadow-expected.txt: Added after r147502.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147532 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[New Multicolumn]Remove assert in RenderMultiColumnFlowThread::createAnonymous
mihnea@adobe.com [Wed, 3 Apr 2013 09:31:40 +0000 (09:31 +0000)]
[New Multicolumn]Remove assert in RenderMultiColumnFlowThread::createAnonymous
https://bugs.webkit.org/show_bug.cgi?id=113855

Reviewed by Antti Koivisto.

Remove CSSRegions specific assert from new multicolumn implementation.
No change in functionality, no new tests.

* rendering/RenderMultiColumnFlowThread.cpp:
(WebCore::RenderMultiColumnFlowThread::createAnonymous):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147531 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoX-Frame-Options: Blocked frames should not inherit their parent's SecurityOrigin.
mkwst@chromium.org [Wed, 3 Apr 2013 09:28:57 +0000 (09:28 +0000)]
X-Frame-Options: Blocked frames should not inherit their parent's SecurityOrigin.
https://bugs.webkit.org/show_bug.cgi?id=112903

Reviewed by Adam Barth.

Source/WebCore:

This change brings WebKit in line with IE and Gecko's behavior, both of
which treat the blocked frame as being cross-origin for the purposes of
access checks ('[frame].contentWindow.location.href' is inaccessible,
for example).

* dom/Document.cpp:
(WebCore::Document::processHttpEquiv):
    Rather than redirecting to 'about:blank', redirect to a URL which
    does not inherit the SecurityOrigin of the parent.
* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::responseReceived):
    Before calling cancelMainResourceLoad, ensure that the frame's
    document is sandboxed into a unique origin so that it doesn't inherit
    the parent's SecurityOrigin.

LayoutTests:

* http/tests/security/XFrameOptions/x-frame-options-deny-expected.txt:
* http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-expected.txt:
* http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body-expected.txt:
* http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body.html:
* http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny-expected.txt:
* http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny.html:
* http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag.html:
* http/tests/security/XFrameOptions/x-frame-options-deny.html:
* http/tests/security/XFrameOptions/x-frame-options-multiple-headers-sameorigin-deny-expected.txt:
* http/tests/security/XFrameOptions/x-frame-options-multiple-headers-sameorigin-deny.html:
* http/tests/security/XFrameOptions/x-frame-options-parent-same-origin-deny-expected.txt:
* http/tests/security/XFrameOptions/x-frame-options-parent-same-origin-deny.html:
* platform/chromium/http/tests/security/XFrameOptions/x-frame-options-deny-expected.txt:
* platform/chromium/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-expected.txt:
* platform/chromium/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body-expected.txt:
* platform/chromium/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny-expected.txt:
* platform/chromium/http/tests/security/XFrameOptions/x-frame-options-multiple-headers-sameorigin-deny-expected.txt:
* platform/chromium/http/tests/security/XFrameOptions/x-frame-options-parent-same-origin-deny-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147530 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Chromium] Implement target-densityDpi viewport property emulation
mnaganov@chromium.org [Wed, 3 Apr 2013 09:19:11 +0000 (09:19 +0000)]
[Chromium] Implement target-densityDpi viewport property emulation
https://bugs.webkit.org/show_bug.cgi?id=110835

The value is only used in the Chromium port, when the corresponding
setting is enabled.

This is needed for supporting existing WebView-based applications
that rely on this property.

Reviewed by Adam Barth.

Source/WebCore:

* dom/ViewportArguments.cpp:
(WebCore::findTargetDensityDPIValue):
(WebCore):
(WebCore::setViewportFeature):
* dom/ViewportArguments.h:
(WebCore::ViewportArguments::ViewportArguments):
(ViewportArguments):
(WebCore::ViewportArguments::operator==):

Source/WebKit/chromium:

* public/WebSettings.h:
(WebSettings):
* src/ChromeClientImpl.cpp:
(WebKit):
(WebKit::calculateTargetDensityDPIFactor):
(WebKit::ChromeClientImpl::dispatchViewportPropertiesDidChange):
* src/WebSettingsImpl.cpp:
(WebKit::WebSettingsImpl::WebSettingsImpl):
(WebKit::WebSettingsImpl::setSupportDeprecatedTargetDensityDPI):
(WebKit):
* src/WebSettingsImpl.h:
(WebSettingsImpl):
(WebKit::WebSettingsImpl::supportDeprecatedTargetDensityDPI):
* tests/WebFrameTest.cpp:
* tests/data/viewport-target-densitydpi-high.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147529 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rolling out r147409.
commit-queue@webkit.org [Wed, 3 Apr 2013 08:56:14 +0000 (08:56 +0000)]
Unreviewed, rolling out r147409.
http://trac.webkit.org/changeset/147409
https://bugs.webkit.org/show_bug.cgi?id=113856

The baselines should be reverted after r147450. (Requested by
kadam on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2013-04-03

* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-deny-expected.txt:
* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-expected.txt:
* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body-expected.txt:
* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny-expected.txt:
* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-multiple-headers-sameorigin-deny-expected.txt:
* platform/qt-5.0-wk2/http/tests/security/XFrameOptions/x-frame-options-parent-same-origin-deny-expected.txt:
* platform/qt/TestExpectations:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-deny-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-multiple-headers-sameorigin-deny-expected.txt:
* platform/qt/http/tests/security/XFrameOptions/x-frame-options-parent-same-origin-deny-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147528 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed. Rebaseline several tests for Chromium that have better output after
ch.dumez@sisa.samsung.com [Wed, 3 Apr 2013 08:38:25 +0000 (08:38 +0000)]
Unreviewed. Rebaseline several tests for Chromium that have better output after
r147492.

* platform/chromium-linux/editing/selection/select-missing-image-expected.png:
* platform/chromium-linux/fast/block/float/002-expected.png:
* platform/chromium-linux/fast/block/float/017-expected.png:
* platform/chromium-linux/fast/dom/HTMLImageElement/image-alt-text-expected.png:
* platform/chromium-linux/fast/dom/HTMLInputElement/input-image-alt-text-expected.png:
* platform/chromium-linux/fast/encoding/utf-16-big-endian-expected.png:
* platform/chromium-linux/fast/encoding/utf-16-little-endian-expected.png:
* platform/chromium-linux/fast/flexbox/023-expected.png:
* platform/chromium-linux/fast/flexbox/024-expected.png:
* platform/chromium-linux/fast/forms/input-value-expected.png:
* platform/chromium-linux/fast/forms/input-value-expected.txt:
* platform/chromium-linux/fast/invalid/012-expected.png:
* platform/chromium-linux/tables/mozilla/bugs/bug2997-expected.png:
* platform/chromium-linux/tables/mozilla/collapsing_borders/bug41262-3-expected.png:
* platform/chromium-mac-lion/editing/selection/select-missing-image-expected.png:
* platform/chromium-mac-lion/fast/block/float/002-expected.png:
* platform/chromium-mac-lion/fast/block/float/017-expected.png:
* platform/chromium-mac-lion/fast/dom/HTMLImageElement/image-alt-text-expected.png:
* platform/chromium-mac-lion/fast/dom/HTMLInputElement/input-image-alt-text-expected.png:
* platform/chromium-mac-lion/fast/encoding/utf-16-big-endian-expected.png:
* platform/chromium-mac-lion/fast/encoding/utf-16-little-endian-expected.png:
* platform/chromium-mac-lion/fast/flexbox/023-expected.png:
* platform/chromium-mac-lion/fast/flexbox/024-expected.png:
* platform/chromium-mac-lion/fast/forms/005-expected.png:
* platform/chromium-mac-lion/fast/forms/input-value-expected.png:
* platform/chromium-mac-lion/fast/invalid/012-expected.png:
* platform/chromium-mac-lion/fast/parser/comment-in-script-expected.png:
* platform/chromium-mac-lion/tables/mozilla/bugs/bug2997-expected.png:
* platform/chromium-mac-lion/tables/mozilla/collapsing_borders/bug41262-3-expected.png:
* platform/chromium-mac-snowleopard/fast/block/float/002-expected.png:
* platform/chromium-mac-snowleopard/fast/block/float/017-expected.png:
* platform/chromium-mac-snowleopard/fast/dom/HTMLImageElement/image-alt-text-expected.png:
* platform/chromium-mac-snowleopard/fast/dom/HTMLInputElement/input-image-alt-text-expected.png:
* platform/chromium-mac-snowleopard/fast/encoding/utf-16-big-endian-expected.png:
* platform/chromium-mac-snowleopard/fast/encoding/utf-16-little-endian-expected.png:
* platform/chromium-mac-snowleopard/fast/flexbox/023-expected.png:
* platform/chromium-mac-snowleopard/fast/flexbox/024-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/input-value-expected.png:
* platform/chromium-mac-snowleopard/fast/invalid/012-expected.png:
* platform/chromium-mac-snowleopard/fast/parser/comment-in-script-expected.png:
* platform/chromium-mac-snowleopard/tables/mozilla/bugs/bug2997-expected.png:
* platform/chromium-mac-snowleopard/tables/mozilla/collapsing_borders/bug41262-3-expected.png:
* platform/chromium-mac/editing/selection/select-missing-image-expected.png:
* platform/chromium-mac/fast/block/float/002-expected.png:
* platform/chromium-mac/fast/block/float/017-expected.png:
* platform/chromium-mac/fast/dom/HTMLImageElement/image-alt-text-expected.png:
* platform/chromium-mac/fast/dom/HTMLImageElement/image-alt-text-expected.txt: Renamed from LayoutTests/platform/chromium/fast/dom/HTMLImageElement/image-alt-text-expected.txt.
* platform/chromium-mac/fast/dom/HTMLInputElement/input-image-alt-text-expected.png:
* platform/chromium-mac/fast/dom/HTMLInputElement/input-image-alt-text-expected.txt: Renamed from LayoutTests/platform/chromium/fast/dom/HTMLInputElement/input-image-alt-text-expected.txt.
* platform/chromium-mac/fast/encoding/utf-16-big-endian-expected.png:
* platform/chromium-mac/fast/encoding/utf-16-big-endian-expected.txt:
* platform/chromium-mac/fast/encoding/utf-16-little-endian-expected.png:
* platform/chromium-mac/fast/encoding/utf-16-little-endian-expected.txt:
* platform/chromium-mac/fast/flexbox/023-expected.png:
* platform/chromium-mac/fast/flexbox/024-expected.png:
* platform/chromium-mac/fast/forms/005-expected.png:
* platform/chromium-mac/fast/forms/input-value-expected.png:
* platform/chromium-mac/fast/forms/input-value-expected.txt:
* platform/chromium-mac/fast/invalid/012-expected.png:
* platform/chromium-mac/fast/parser/comment-in-script-expected.png:
* platform/chromium-mac/tables/mozilla/bugs/bug2997-expected.png:
* platform/chromium-mac/tables/mozilla/collapsing_borders/bug41262-3-expected.png:
* platform/chromium-mac/tables/mozilla/collapsing_borders/bug41262-3-expected.txt: Renamed from LayoutTests/platform/chromium/tables/mozilla/collapsing_borders/bug41262-3-expected.txt.
* platform/chromium-win/editing/selection/select-missing-image-expected.png:
* platform/chromium-win/fast/block/float/002-expected.png:
* platform/chromium-win/fast/block/float/017-expected.png:
* platform/chromium-win/fast/dom/HTMLImageElement/image-alt-text-expected.png:
* platform/chromium-win/fast/dom/HTMLImageElement/image-alt-text-expected.txt:
* platform/chromium-win/fast/dom/HTMLInputElement/input-image-alt-text-expected.png:
* platform/chromium-win/fast/dom/HTMLInputElement/input-image-alt-text-expected.txt:
* platform/chromium-win/fast/encoding/utf-16-big-endian-expected.png:
* platform/chromium-win/fast/encoding/utf-16-big-endian-expected.txt:
* platform/chromium-win/fast/encoding/utf-16-little-endian-expected.png:
* platform/chromium-win/fast/encoding/utf-16-little-endian-expected.txt:
* platform/chromium-win/fast/flexbox/023-expected.png:
* platform/chromium-win/fast/flexbox/024-expected.png:
* platform/chromium-win/fast/forms/input-value-expected.png:
* platform/chromium-win/fast/forms/input-value-expected.txt:
* platform/chromium-win/fast/invalid/012-expected.png:
* platform/chromium-win/fast/parser/comment-in-script-expected.png:
* platform/chromium-win/tables/mozilla/bugs/bug2997-expected.png:
* platform/chromium-win/tables/mozilla/collapsing_borders/bug41262-3-expected.png:
* platform/chromium-win/tables/mozilla/collapsing_borders/bug41262-3-expected.txt:
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147527 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoExtract URL that doesn't inherit a parent's SecurityOrigin out into a constant.
mkwst@chromium.org [Wed, 3 Apr 2013 08:31:07 +0000 (08:31 +0000)]
Extract URL that doesn't inherit a parent's SecurityOrigin out into a constant.
https://bugs.webkit.org/show_bug.cgi?id=113780

Reviewed by Jochen Eisinger.

We're scheduling navigations to 'data:text/html,' in XSSAuditor in
order to end up on a page that doesn't inherit its parent's
SecurityOrigin. We'll be reusing this mechainsm to solve
http://wkbug.com/112903

This patch makes us more explicit about what we're doing and why.
It doesn't change any behavior: XSSAuditor tests should still pass.

* html/parser/XSSAuditorDelegate.cpp:
(WebCore::XSSAuditorDelegate::didBlockScript):
    Use SecurityOrigin::urlWithUniqueSecurityOrigin rather than a
    literal string to make our intentions clear.
* page/SecurityOrigin.cpp:
(WebCore::SecurityOrigin::urlWithUniqueSecurityOrigin):
* page/SecurityOrigin.h:
    Add the new static method.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147526 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRemove code for Mac 10.5 and earlier from WTF and WebKit2
rniwa@webkit.org [Wed, 3 Apr 2013 08:21:49 +0000 (08:21 +0000)]
Remove code for Mac 10.5 and earlier from WTF and WebKit2
https://bugs.webkit.org/show_bug.cgi?id=113844

Reviewed by Benjamin Poulain.

Removed the code for 10.5 and removed if-def for 10.6.

Source/WebKit2:

* Shared/mac/PasteboardTypes.mm:
(WebKit::PasteboardTypes::forEditing):

Source/WTF:

* wtf/FastMalloc.cpp:
(WTF::TCMalloc_ThreadCache::CreateCacheIfNecessary):
* wtf/Platform.h:
* wtf/ThreadingPthreads.cpp:
(WTF::initializeCurrentThreadInternal):
* wtf/unicode/icu/CollatorICU.cpp:
(WTF::Collator::userDefault):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147525 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rebaseline several tests cases for GTK that have better results
ch.dumez@sisa.samsung.com [Wed, 3 Apr 2013 07:54:04 +0000 (07:54 +0000)]
Unreviewed, rebaseline several tests cases for GTK that have better results
after r147492.

* platform/gtk/TestExpectations:
* platform/gtk/fast/dom/HTMLImageElement/image-alt-text-expected.png:
* platform/gtk/fast/dom/HTMLImageElement/image-alt-text-expected.txt:
* platform/gtk/fast/dom/HTMLInputElement/input-image-alt-text-expected.png:
* platform/gtk/fast/dom/HTMLInputElement/input-image-alt-text-expected.txt:
* platform/gtk/fast/encoding/utf-16-big-endian-expected.png:
* platform/gtk/fast/encoding/utf-16-big-endian-expected.txt:
* platform/gtk/fast/encoding/utf-16-little-endian-expected.png:
* platform/gtk/fast/encoding/utf-16-little-endian-expected.txt:
* platform/gtk/fast/forms/input-value-expected.png:
* platform/gtk/fast/forms/input-value-expected.txt:
* platform/gtk/tables/mozilla/collapsing_borders/bug41262-3-expected.png:
* platform/gtk/tables/mozilla/collapsing_borders/bug41262-3-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147524 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoclean up static casts in WebAccessibilityObjectWrapperMac.mm to methods like toAccess...
cfleizach@apple.com [Wed, 3 Apr 2013 06:56:15 +0000 (06:56 +0000)]
clean up static casts in WebAccessibilityObjectWrapperMac.mm to methods like toAccessibilityList()
https://bugs.webkit.org/show_bug.cgi?id=113828

Reviewed by Tim Horton.

static_cast changed to toAccessibilityList().

* accessibility/AccessibilityList.h:
(WebCore::toAccessibilityList):
* accessibility/mac/WebAccessibilityObjectWrapperMac.mm:
(-[WebAccessibilityObjectWrapper subrole]):
(-[WebAccessibilityObjectWrapper roleDescription]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147523 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdate expectations for fast/css3-text/css3-text-indent/text-indent-each-line.html
commit-queue@webkit.org [Wed, 3 Apr 2013 06:52:47 +0000 (06:52 +0000)]
Update expectations for fast/css3-text/css3-text-indent/text-indent-each-line.html
https://bugs.webkit.org/show_bug.cgi?id=113848

Mark as [ Failure ImageOnlyFailure ] for fast/css3-text/css3-text-indent/.

Unreviewed gardening.

Patch by Jaehun Lim <ljaehun.lim@samsung.com> on 2013-04-02

* platform/mac/TestExpectations:
* platform/qt-mac/TestExpectations:
* platform/qt/TestExpectations:
* platform/win/TestExpectations:
* platform/wincairo/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147522 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rolling out r147370.
commit-queue@webkit.org [Wed, 3 Apr 2013 06:40:30 +0000 (06:40 +0000)]
Unreviewed, rolling out r147370.
http://trac.webkit.org/changeset/147370
https://bugs.webkit.org/show_bug.cgi?id=113850

This patch is causing crashes in video tests on all Mac bots.
(Requested by jernoble_ on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2013-04-02

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::~HTMLMediaElement):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147521 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAXRoleDescripton for DL is "list"; should be "description list"
commit-queue@webkit.org [Wed, 3 Apr 2013 06:17:21 +0000 (06:17 +0000)]
AXRoleDescripton for DL is "list"; should be "description list"
https://bugs.webkit.org/show_bug.cgi?id=111665

Patch by James Craig <james@cookiecrook.com> on 2013-04-02
Reviewed by Chris Fleizach.

Source/WebCore:

Defining the role and description strings, and accessors for DL.
This previously relied on getting the role and description from the platform API.

Updated existing test coverage.

* English.lproj/Localizable.strings:
* accessibility/AccessibilityObject.h:
* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::determineAccessibilityRole):
* accessibility/mac/WebAccessibilityObjectWrapperMac.mm:
(createAccessibilityRoleMap):
(-[WebAccessibilityObjectWrapper subrole]):
(-[WebAccessibilityObjectWrapper roleDescription]):
* platform/LocalizedStrings.cpp:
(WebCore::AXDescriptionListText):
(WebCore):
* platform/LocalizedStrings.h:
(WebCore):
* platform/blackberry/LocalizedStringsBlackBerry.cpp:
(WebCore::AXDescriptionListText):
(WebCore):
* platform/chromium/LocalizedStringsChromium.cpp:
(WebCore::AXDescriptionListText):
(WebCore):
* platform/efl/LocalizedStringsEfl.cpp:
(WebCore::AXDescriptionListText):
(WebCore):
* platform/gtk/LocalizedStringsGtk.cpp:
(WebCore::AXDescriptionListText):
(WebCore):
* platform/qt/LocalizedStringsQt.cpp:
(WebCore::AXDescriptionListText):
(WebCore):

Source/WebKit/chromium:

Defining the role and description strings.

* public/WebAccessibilityRole.h:
* src/AssertMatchingEnums.cpp:

Tools:

Defining the role and description strings.

* DumpRenderTree/chromium/TestRunner/src/AccessibilityUIElementChromium.cpp:

LayoutTests:

Updating expectation strings for DL role description.

* platform/mac/accessibility/lists-expected.txt:
* platform/mac/accessibility/role-subrole-roledescription-expected.txt:
* platform/mac/accessibility/role-subrole-roledescription.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147520 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: [Network] Split business logic and presentation for initiator column.
eustas@chromium.org [Wed, 3 Apr 2013 06:00:21 +0000 (06:00 +0000)]
Web Inspector: [Network] Split business logic and presentation for initiator column.
https://bugs.webkit.org/show_bug.cgi?id=113486

Reviewed by Pavel Feldman.

Currently business logic and presentation are mixed.
This prevents making cell renderer static and applying
viewport rendering, because initiator sorting function
depends on data written when cell is rendered.

* inspector/front-end/NetworkPanel.js:
(WebInspector.NetworkDataGridNode.prototype._refreshInitiatorCell):
Removed business logic. Use "initiatorInfo".
(WebInspector.NetworkDataGridNode.InitiatorComparator):
Use "initiatorInfo".
* inspector/front-end/NetworkRequest.js:
(WebInspector.NetworkRequest.prototype.initiatorInfo):
Extracted business logic from "refreshInitiatorCell".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147519 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed gardening; rebaseline mac expected results after r147470.
jer.noble@apple.com [Wed, 3 Apr 2013 05:39:31 +0000 (05:39 +0000)]
Unreviewed gardening; rebaseline mac expected results after r147470.

* platform/mac/css3/filters/effect-reference-composite-expected.png: Added.
* platform/mac/css3/filters/effect-reference-composite-expected.txt: Added.
* platform/mac/css3/filters/effect-reference-composite-hw-expected.png: Added.
* platform/mac/css3/filters/effect-reference-composite-hw-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147518 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoneedsLeopardMailQuirks should be removed
rniwa@webkit.org [Wed, 3 Apr 2013 04:56:27 +0000 (04:56 +0000)]
needsLeopardMailQuirks should be removed
https://bugs.webkit.org/show_bug.cgi?id=113749

Reviewed by Benjamin Poulain.

Removed needsLeopardMailQuirks from settings.in after r147447.

* page/Settings.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147517 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[EFL] Unskip and update inspector tests
commit-queue@webkit.org [Wed, 3 Apr 2013 03:46:01 +0000 (03:46 +0000)]
[EFL] Unskip and update inspector tests
https://bugs.webkit.org/show_bug.cgi?id=113840

Unreviewed gardening.

Patch by Seokju Kwon <seokju.kwon@gmail.com> on 2013-04-02

* platform/efl/TestExpectations:
* platform/efl/http/tests/inspector/console-websocket-error-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147516 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnskip filter-fallback-to-software.html on Mac WK2 Release
benjamin@webkit.org [Wed, 3 Apr 2013 02:50:24 +0000 (02:50 +0000)]
Unskip filter-fallback-to-software.html on Mac WK2 Release

Patch by Benjamin Poulain <bpoulain@apple.com> on 2013-04-02

* platform/mac-wk2/TestExpectations:
The test seems to pass in Release. Let's see how the bots feel about that.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147515 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDFG should compile a little sooner
ggaren@apple.com [Wed, 3 Apr 2013 01:41:16 +0000 (01:41 +0000)]
DFG should compile a little sooner
https://bugs.webkit.org/show_bug.cgi?id=113835

Unreviewed.

Rolled out r147511 because it was based on incorrect performance
measurement.

* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::optimizationThresholdScalingFactor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSource/WebCore: [css3-text] Rendering -webkit-each-line value for text-indent from...
commit-queue@webkit.org [Wed, 3 Apr 2013 01:32:24 +0000 (01:32 +0000)]
Source/WebCore: [css3-text] Rendering -webkit-each-line value for text-indent from css3-text
https://bugs.webkit.org/show_bug.cgi?id=113259

Patch by Jaehun Lim <ljaehun.lim@samsung.com> on 2013-04-02
Reviewed by Levi Weintraub.

CSS3 TEXT introduces new value, "each-line" for text-indent.
Indentation affects the first line of the block container as well as each line
after a forced line break, but does not affect lines after a soft wrap break.

This patch is the rendering part to support each-line value for text-indent.
It's prefixed and guarded by CSS3_TEXT flag.

Test: fast/css3-text/css3-text-indent/text-indent-each-line.html

* rendering/RenderBlock.h:
(WebCore::RenderBlock::availableLogicalWidthForLine):
(WebCore::RenderBlock::logicalRightOffsetForLine):
(WebCore::RenderBlock::logicalLeftOffsetForLine):
(WebCore::RenderBlock::startOffsetForLine):
(WebCore::RenderBlock::endOffsetForLine):
(WebCore::RenderBlock::pixelSnappedLogicalLeftOffsetForLine):
(WebCore::RenderBlock::pixelSnappedLogicalRightOffsetForLine):
    Rename a parameter from firstLine to shouldIndentText.

* rendering/RenderBlockLineLayout.cpp:
(WebCore::LineWidth::LineWidth):
(WebCore::LineWidth::shouldIndentText): Return boolean according to m_shouldIndentText.
(WebCore::LineWidth::updateAvailableWidth):
(WebCore::LineWidth::shrinkAvailableWidthForNewFloatIfNeeded):
(WebCore::LineWidth::fitBelowFloats):
    Add enum IndentTextOrNot and m_shouldIndentText.
    Use m_shouldIndentText instead of m_isFirstLine.

(WebCore::requiresIndent): Decide whether text indentation is required.
(WebCore::updateLogicalInlinePositions):
(WebCore::RenderBlock::computeInlineDirectionPositionsForLine):
(WebCore::RenderBlock::LineBreaker::nextSegmentBreak):

LayoutTests: [css3-text] Rendering -webkit-each-line value for text-indent from css3-text
https://bugs.webkit.org/show_bug.cgi?id=113259

CSS3 TEXT introduces new value, "each-line" for text-indent.
Indentation affects the first line of the block container as well as each line
after a forced line break, but does not affect lines after a soft wrap break.

Patch by Jaehun Lim <ljaehun.lim@samsung.com> on 2013-04-02
Reviewed by Levi Weintraub.

* fast/css3-text/css3-text-indent/text-indent-each-line-expected.html: Added.
* fast/css3-text/css3-text-indent/text-indent-each-line.html: Added.
* platform/win/TestExpectations:
* platform/wincairo/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147513 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDFG should compile a little sooner
ggaren@apple.com [Wed, 3 Apr 2013 00:34:31 +0000 (00:34 +0000)]
DFG should compile a little sooner
https://bugs.webkit.org/show_bug.cgi?id=113835

Reviewed by Michael Saboff.

2% speedup on SunSpider.

2% speedup on JSRegress.

Neutral on Octane, v8, and Kraken.

The worst-hit single sub-test is kraken-stanford-crypto-ccm.js, which gets
18% slower. Since Kraken is neutral overall in its preferred mean, I
think that's OK for now.

(Our array indexing speculation fails pathologically on
kraken-stanford-crypto-ccm.js. Compiling sooner is a regression because
it triggers those failures sooner. I'm going to file some follow-up bugs
explaining how to fix our speculations on this sub-test, at which point
compiling earlier should become a slight speedup on Kraken overall.)

* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::optimizationThresholdScalingFactor): I experimented
with a few different options, including reducing the coefficient 'a'.
A simple linear reduction on instruction count worked best.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147511 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoCompositing tests failing after http://trac.webkit.org/changeset/147502
dino@apple.com [Wed, 3 Apr 2013 00:28:58 +0000 (00:28 +0000)]
Compositing tests failing after trac.webkit.org/changeset/147502
https://bugs.webkit.org/show_bug.cgi?id=113838

Temporarily marking as failing.

* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147510 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed. Disable running tests on Win EWS queue.
roger_fong@apple.com [Wed, 3 Apr 2013 00:08:49 +0000 (00:08 +0000)]
Unreviewed. Disable running tests on Win EWS queue.

* Scripts/webkitpy/tool/commands/earlywarningsystem.py:
(WinEWS):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBe more robust against empty message receiver names in incoming messages
andersca@apple.com [Tue, 2 Apr 2013 23:34:22 +0000 (23:34 +0000)]
Be more robust against empty message receiver names in incoming messages
https://bugs.webkit.org/show_bug.cgi?id=113833
<rdar://problem/13284433>

Reviewed by Beth Dakin.

Turns out that we either send or receive messages whose receiver names are empty. This leads to bad things when we try to look
up the message receiver name in a hash map since the empty name is used to represent an empty hash map value.

* Platform/CoreIPC/Connection.cpp:
(CoreIPC::Connection::addWorkQueueMessageReceiverOnConnectionWorkQueue):
Sprinkle assertions.

(CoreIPC::Connection::processIncomingMessage):
If the message receiver name is not valid, make sure to call didReceiveInvalidMessage on the client thread.

(CoreIPC::Connection::dispatchDidReceiveInvalidMessage):
Add new helper function.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147508 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoNeed japanese line layout performance test
glenn@skynav.com [Tue, 2 Apr 2013 23:31:18 +0000 (23:31 +0000)]
Need japanese line layout performance test
https://bugs.webkit.org/show_bug.cgi?id=113811

Reviewed by Ryosuke Niwa.

Will be used by https://bugs.webkit.org/show_bug.cgi?id=105692 to test potential performance regression
for Japanese (and other CJK) line break changes.

* Layout/LineLayoutJapanese.html: Added.
* Skipped: Skip this test by default.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoGeneralize prior line break context state and names.
glenn@skynav.com [Tue, 2 Apr 2013 23:28:32 +0000 (23:28 +0000)]
Generalize prior line break context state and names.
https://bugs.webkit.org/show_bug.cgi?id=113823

Reviewed by Darin Adler.

No new tests required due to no behavioral change.

* platform/text/TextBreakIterator.h:
(WebCore::LazyLineBreakIterator::LazyLineBreakIterator):
(WebCore::LazyLineBreakIterator::lastCharacter):
(WebCore::LazyLineBreakIterator::secondToLastCharacter):
(WebCore::LazyLineBreakIterator::setPriorContext):
(WebCore::LazyLineBreakIterator::updatePriorContext):
(WebCore::LazyLineBreakIterator::resetPriorContext):
(WebCore::LazyLineBreakIterator::resetStringAndReleaseIterator):
Store previous characters as prior context buffer used directly
by line break iterator implementation. Rename members to reflect
somewhat more general usage. Also fix bad indentations.
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::layoutRunsAndFloatsInRange):
(WebCore::RenderBlock::LineBreaker::nextSegmentBreak):
Use renamed line break iterator methods.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoPadding applied twice for empty generated RenderInlines
eae@chromium.org [Tue, 2 Apr 2013 23:17:18 +0000 (23:17 +0000)]
Padding applied twice for empty generated RenderInlines
https://bugs.webkit.org/show_bug.cgi?id=113800

Source/WebCore:

Reviewed by Levi Weintraub.

The padding for empty generated content render objects is
incorrectly applied twice. Regression from r140693.

Test: fast/css-generated-content/margin-empty-content.html

* rendering/RenderBlockLineLayout.cpp:
(WebCore::inlineLogicalWidth):

LayoutTests:

Reviewed by Levi Weintraub.

Add test for empty generated content with padding.

* fast/css-generated-content/margin-empty-content-expected.html: Added.
* fast/css-generated-content/margin-empty-content.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147505 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoSource/WebCore: InlineIterator needs to be updated when RenderCombineText shrinks...
leviw@chromium.org [Tue, 2 Apr 2013 23:10:49 +0000 (23:10 +0000)]
Source/WebCore: InlineIterator needs to be updated when RenderCombineText shrinks at a line break
https://bugs.webkit.org/show_bug.cgi?id=113733

Reviewed by Emil A Eklund.

Update stale InlineIterators that point beyond the end of a RenderCombineText that's
shrunk when RenderCombineText::combineText() is called.

Test: fast/text/international/combine-at-line-break-crash.html

* rendering/BidiRun.h:
(WebCore::BidiRun::BidiRun): Adding an assertion that runs are created within
the bounds of their associated RenderText.
* rendering/RenderBlockLineLayout.cpp:
(WebCore::iteratorIsBeyondEndOfRenderCombineText): Added.
(WebCore::RenderBlock::LineBreaker::nextSegmentBreak): Update stale iterators.

LayoutTests: InlineIterator needs to be updated when RenderCombineText shrinks at a line break
https://bugs.webkit.org/show_bug.cgi?id=113732

Reviewed by Emil A Eklund.

* fast/text/international/combine-at-line-break-crash-expected.txt: Added.
* fast/text/international/combine-at-line-break-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147504 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUse cross-port testRunner methods in fast/loader/url-selected-user-gesture.html
jochen@chromium.org [Tue, 2 Apr 2013 22:51:23 +0000 (22:51 +0000)]
Use cross-port testRunner methods in fast/loader/url-selected-user-gesture.html
https://bugs.webkit.org/show_bug.cgi?id=113827

Reviewed by Benjamin Poulain.

* fast/loader/url-selected-user-gesture-expected.txt:
* fast/loader/url-selected-user-gesture.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147503 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[CSS Filters] Filter outsets clipped on composited layers when filter is applied...
mvujovic@adobe.com [Tue, 2 Apr 2013 22:31:51 +0000 (22:31 +0000)]
[CSS Filters] Filter outsets clipped on composited layers when filter is applied after first layout
https://bugs.webkit.org/show_bug.cgi?id=109098

Reviewed by Dean Jackson.

Source/WebCore:

This patch addresses the following issues:
(1) Compositors who can compute their own filter outsets don't need WebKit to inflate the
    composited layer bounds. Eventually, all platforms' compositors should be able to
    compute their own filter outsets.
(2) When filter outsets change, we need to update composited layer bounds if the filters are
    painted in software.
(3) When filter outsets change, we need to recompute the overlap map to promote the correct
    layers to composited layers on all platforms.

Tests: css3/filters/composited-layer-bounds-after-sw-blur-animation.html
       css3/filters/composited-layer-bounds-with-composited-blur.html
       css3/filters/composited-layer-child-bounds-after-composited-to-sw-shadow-change.html
       css3/filters/composited-layer-promotion-after-outset-overlap-change-using-composited-shadow.html
       css3/filters/composited-layer-promotion-after-outset-overlap-change-using-sw-shadow.html

* platform/graphics/IntRectExtent.h:
(WebCore::IntRectExtent::isZero):
(IntRectExtent):

    Make isZero public on IntRectExtent.

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

    Helper function used to exit early from filters related code.

(WebCore::RenderLayer::updateFilters):

    Move filter updating code from RenderLayer::styleChanged into this new method.

    This method calls RenderLayerBacking::updateFilters instead of
    RenderLayerBacking::updateGraphicsLayerGeometry. RenderLayerBacking::updateFilters lets
    us know whether there was a filter paint mode change (i.e. the filters were painted in
    software and are now painted in the compositor, or vice-versa). We need to know this
    earlier, in RenderLayer::styleChanged, because we may need to rebuild the compositing
    layers due to a filter paint mode change.

(WebCore::RenderLayer::styleChanged):

    Add a condition for filters for setCompositingLayersNeedRebuild. If the filter outsets
    on a layer changed, we need to rebuild the compositing layers because there could be new
    overlap. Likewise, if there are or were filter outsets and the filter paint mode changed
    the layer bounds (and child positions) may have changed. In either case, we don't
    rebuild the compositing layers if there is an accelerated filter animation running.

    Remove old code to detect a filter paint mode change from the end of
    RenderLayer::styleChanged because it could never be executed. The old code counted on
    RenderLayerBacking::canCompositeFilters to change on its own, but it doesn't change
    unless updateFilters gets called. Thus, we could never enter the if statement that
    calls setBackingNeedsRepaint.

* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::updateFilters):

    Now return true if there was a filter paint mode change, so RenderLayer::styleChanged
    can decide if it should rebuild the compositing layers.

    Call updateOrRemoveFilterEffectRenderer and setContentsNeedDisplay if there was a
    render mode change. We need to either destroy or create the software renderer and we
    need to repaint the layer contents if there was a render mode change.

(WebCore::RenderLayerBacking::updateGraphicsLayerGeometry):

    Don't call updateFilters here. Call it earlier in RenderLayer::styleChanged, for reasons
    explained above.

* rendering/RenderLayerBacking.h:
(RenderLayerBacking):

    Make updateFilters public so RenderLayer::styleChanged can call it.

* rendering/RenderLayerCompositor.cpp:
(WebCore::RenderLayerCompositor::calculateCompositedBounds):

    If the platform's compositor computes filter outsets for its layers internally, do not
    inflate WebKit's composited bounds for filter outsets. This results in smaller layers
    on Safari. Note that we still need to include filter outsets in layer bounds for other
    situations like overlap testing in WebKit and when we're painting filters in software.

Source/WTF:

* wtf/Platform.h:
    Add flag for HAVE(COMPOSITOR_FILTER_OUTSETS) macro. This is used to check if a
    platform's compositor expands layers internally for filter outsets when it applies
    filters like drop-shadow and blur. Currently, only CoreAnimation does has this
    capability.

LayoutTests:

* compositing/filters/sw-layer-overlaps-hw-shadow-expected.txt:
* compositing/filters/sw-nested-shadow-overlaps-hw-nested-shadow-expected.txt:
* compositing/filters/sw-shadow-overlaps-hw-layer-expected.txt:
* compositing/filters/sw-shadow-overlaps-hw-shadow-expected.txt:

    The above tests needed rebaselining because we're now using smaller layers on Safari.
    We aren't inflating layers for filter outsets since CoreAnimation does it for us
    already.

* css3/filters/composited-layer-bounds-after-sw-blur-animation-expected.txt: Added.
* css3/filters/composited-layer-bounds-after-sw-blur-animation.html: Added.
* css3/filters/composited-layer-bounds-with-composited-blur-expected.txt: Added.
* css3/filters/composited-layer-bounds-with-composited-blur.html: Added.
* css3/filters/composited-layer-child-bounds-after-composited-to-sw-shadow-change-expected.txt: Added.
* css3/filters/composited-layer-child-bounds-after-composited-to-sw-shadow-change.html: Added.
* css3/filters/composited-layer-promotion-after-outset-overlap-change-using-composited-shadow-expected.txt: Added.
* css3/filters/composited-layer-promotion-after-outset-overlap-change-using-composited-shadow.html: Added.
* css3/filters/composited-layer-promotion-after-outset-overlap-change-using-sw-shadow-expected.txt: Added.
* css3/filters/composited-layer-promotion-after-outset-overlap-change-using-sw-shadow.html: Added.

    The above tests are all new. They mostly test edge cases with filter render mode
    changes, filter outset changes, and composited layer promotion due to overlap.

    Before this patch, composited-layer-bounds-after-sw-blur-animation.html breaks
    visibly on Safari WebKit nightly. The blur filter in the test appears clipped before
    this patch. The other tests render the same before and after this patch, but now their
    layer sizes are smaller.

* css3/filters/filter-change-repaint-composited-expected.png:
* css3/filters/filter-change-repaint-composited.html:

    This test has been expanded with two more cases, including the original reproduction of
    this bug. The test needs rebaselining on Chromium Linux and Win.

* platform/chromium-win/css3/filters/filter-change-repaint-composited-expected.png: Removed.
* platform/chromium/TestExpectations:
* platform/chromium/compositing/filters/sw-nested-shadow-overlaps-hw-nested-shadow-expected.txt: Added.
* platform/chromium/compositing/filters/sw-shadow-overlaps-hw-shadow-expected.txt: Added.
* platform/chromium/css3/filters/composited-layer-bounds-after-sw-blur-animation-expected.txt: Added.
* platform/chromium/css3/filters/composited-layer-bounds-with-composited-blur-expected.txt: Added.
* platform/chromium/css3/filters/composited-layer-child-bounds-after-composited-to-sw-shadow-change-expected.txt: Added.
* platform/chromium/css3/filters/composited-layer-promotion-after-outset-overlap-change-using-composited-shadow-expected.txt: Added.
* platform/chromium/css3/filters/composited-layer-promotion-after-outset-overlap-change-using-sw-shadow-expected.txt: Added.

    We need different expectations for Chromium for the new tests because layer bounds
    include filter outsets on Chromium but not on Safari.

* platform/chromium/css3/filters/filter-change-repaint-composited-expected.png: Added.
* platform/qt/TestExpectations:

    Note that we need baselines on Qt for the new and updated tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rolling out r147423.
benjamin@webkit.org [Tue, 2 Apr 2013 21:52:11 +0000 (21:52 +0000)]
Unreviewed, rolling out r147423.

The change r147402 was reverted by r147450.
The rebaseline introduced by r147423 needs to be reverted.

* platform/mac-wk2/http/tests/security/XFrameOptions: Removed.
* platform/mac-wk2/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny-expected.txt: Removed.
* platform/mac/http/tests/security/XFrameOptions: Removed.
* platform/mac/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-expected.txt: Removed.
* platform/mac/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body-expected.txt: Removed.
* platform/mac/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago [Mac][WK2] Donâ\80\99t let plug-ins use System V shared memory
ap@apple.com [Tue, 2 Apr 2013 21:49:34 +0000 (21:49 +0000)]
    [Mac][WK2] Donâ\80\99t let plug-ins use System V shared memory
        https://bugs.webkit.org/show_bug.cgi?id=113466
        <rdar://problem/13159030>

        Reviewed, tweaked and landed by Alexey Proskuryakov.

        Instead of allowing plug-ins to request System V shm, give them a
        temporary but usable alternative if they try to request it.

        * PluginProcess/mac/PluginProcessShim.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147500 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GTK] Make libsecret optional
mrobinson@webkit.org [Tue, 2 Apr 2013 21:24:14 +0000 (21:24 +0000)]
[GTK] Make libsecret optional
https://bugs.webkit.org/show_bug.cgi?id=113821

Reviewed by Gustavo Noronha Silva.

.:

Add a configuration option to disable credential storage and thus remove
the libsecret dependency. This should make it possible to build WebKit 2.x
on Windows again.

* Source/autotools/FindDependencies.m4: Only look for libsecret if credential storage is active.
* Source/autotools/PrintBuildConfiguration.m4: Print whether or not credential storage is active.
* Source/autotools/ReadCommandLineArguments.m4: Added an option to control credential storage.
* Source/autotools/SetupAutoconfHeader.m4: Expose credential storage setting to code.

Source/WebCore:

Don't try to use libsecret if credential storage is disabled.

* platform/gtk/GRefPtrGtk.cpp: Protect libsecret sections.
* platform/gtk/GRefPtrGtk.h: Ditto.
* platform/network/gtk/CredentialBackingStore.cpp: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147499 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, chromium gardening.
japhet@chromium.org [Tue, 2 Apr 2013 21:23:55 +0000 (21:23 +0000)]
Unreviewed, chromium gardening.

* platform/chromium-linux/css3/filters/effect-reference-composite-expected.png: Removed.
* platform/chromium-linux/css3/filters/effect-reference-composite-expected.txt: Removed.
* platform/chromium-linux/css3/filters/effect-reference-composite-hw-expected.png: Removed.
* platform/chromium-linux/css3/filters/effect-reference-composite-hw-expected.txt: Removed.
* platform/chromium-mac/css3/filters/effect-reference-composite-expected.png: Added.
* platform/chromium-mac/css3/filters/effect-reference-composite-expected.txt: Added.
* platform/chromium-mac/css3/filters/effect-reference-composite-hw-expected.png: Added.
* platform/chromium-mac/css3/filters/effect-reference-composite-hw-expected.txt: Added.
* platform/chromium-win/css3/filters/effect-reference-composite-expected.png: Added.
* platform/chromium-win/css3/filters/effect-reference-composite-expected.txt: Added.
* platform/chromium-win/css3/filters/effect-reference-composite-hw-expected.png: Added.
* platform/chromium-win/css3/filters/effect-reference-composite-hw-expected.txt: Added.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147498 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[BlackBerry] API update for IntRect
commit-queue@webkit.org [Tue, 2 Apr 2013 21:21:33 +0000 (21:21 +0000)]
[BlackBerry] API update for IntRect
https://bugs.webkit.org/show_bug.cgi?id=113824

Patch by Maxim Mogilnitsky <mmogilnitsky@rim.com> on 2013-04-02
Reviewed by Rob Buis.

Modifing IntRect and FloatRect API.

* WebKitSupport/SelectionHandler.cpp:
(BlackBerry::WebKit::SelectionHandler::regionForTextQuads):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147497 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rolling out r147423.
benjamin@webkit.org [Tue, 2 Apr 2013 21:19:16 +0000 (21:19 +0000)]
Unreviewed, rolling out r147423.
http://trac.webkit.org/changeset/147423
https://bugs.webkit.org/show_bug.cgi?id=113775

The change r147402 was reverted by r147450.
The rebaseline introduced by r147423 needs to be reverted.

* platform/mac-wk2/http/tests/security/XFrameOptions: Removed.
* platform/mac-wk2/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny-expected.txt: Removed.
* platform/mac/http/tests/security/XFrameOptions: Removed.
* platform/mac/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-expected.txt: Removed.
* platform/mac/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body-expected.txt: Removed.
* platform/mac/http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147496 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[CSS Exclusions] refactor shape-outside code to use isFloatingWithShapeOutside()...
commit-queue@webkit.org [Tue, 2 Apr 2013 21:18:09 +0000 (21:18 +0000)]
[CSS Exclusions] refactor shape-outside code to use isFloatingWithShapeOutside() helper method
https://bugs.webkit.org/show_bug.cgi?id=113799

Patch by Bem Jones-Bey <bjonesbe@adobe.com> on 2013-04-02
Reviewed by Dirk Schulze.

When reviewing my patch for bug 110349, Julien mentioned that the
common test for floating with shape outside should be factored out.
This patch does that.

No new functionality, so no new tests.

* rendering/ExclusionShapeOutsideInfo.cpp:
(WebCore::ExclusionShapeOutsideInfo::isEnabledFor): Use helper method.
* rendering/RenderBox.h:
(WebCore::RenderBox::exclusionShapeOutsideInfo): Ditto.
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::shouldBeNormalFlowOnly): Ditto.
* rendering/RenderObject.h:
(WebCore::RenderObject::hasPaintOffset):  Ditto.
(WebCore::RenderObject::isFloatingWithShapeOutside): Add method to
    encapsulate the common test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147495 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd platform AudioSession and iOS platform-specific code.
jer.noble@apple.com [Tue, 2 Apr 2013 20:49:19 +0000 (20:49 +0000)]
Add platform AudioSession and iOS platform-specific code.
https://bugs.webkit.org/show_bug.cgi?id=113813
<rdar://problem/11701792>

Reviewed by Eric Carlson.

Replace the deprecated AudioToolbox AudioSession APIs with the new AVAudioSession API.
To protect against exposing platform specific intrinsics to cross-platform code, add a
platform generic class AudioSession, with port-specific implementations. A default
implementation is provided.

Add a new listener interface to receive interruption notifications:
* platform/audio/AudioSessionListener.h: Added.
(WebCore::AudioSessionListener::AudioSessionListener):
(WebCore::AudioSessionListener::~AudioSessionListener):

Add a platform generic AudioSession class:
* platform/audio/AudioSession.cpp: Added.
(AudioSession::sharedSession): AudioSession is a singleton object, and this is its accessor.
(AudioSession::addListener): Add the parameter to the listeners set.
(AudioSession::removeListener): Remove the parameter from the listers set.
(AudioSession::beganAudioInterruption): Iterate over the parameters.
(AudioSession::endedAudioInterruption): Ditto.
(AudioSession::AudioSession): Empty stub.
(AudioSession::~AudioSession): Ditto.
(AudioSession::setCategory): Ditto.
(AudioSession::category): Ditto.
(AudioSession::sampleRate): Ditto.
(AudioSession::numberOfOutputChannels): Ditto.
(AudioSession::setActive): Ditto.
(AudioSession::preferredBufferDuration): Ditto.
(AudioSession::setPreferredBufferDuration): Ditto.
* platform/audio/AudioSession.h: Added.

Add IOS-specific AudioSession implementation:
* platform/audio/ios/AudioSessionIOS.mm: Added.
(-[WebAudioSessionHelper initWithCallback:WebCore::]): Register for interruption notifications.
(-[WebAudioSessionHelper dealloc]): Unregister for interruption notifications.
(-[WebAudioSessionHelper interruption:]): Pass to AudioSession.
(WebCore::AudioSessionPrivate::AudioSessionPrivate): Private storage for AudioSessionIOS.
(WebCore::AudioSession::AudioSession): Create AudioSessionPrivate and WebAudioSessionHelper.
(WebCore::AudioSession::~AudioSession): Simple destructor.
(WebCore::AudioSession::setCategory): Check the current categoryOverride, and refuse to overwrite
    "MediaPlayback" with "AmbientSound".
(WebCore::AudioSession::setCategoryOverride): Simple setter.
(WebCore::AudioSession::categoryOverride): Simple getter.
(WebCore::AudioSession::category): Pass to AVAudioSession.
(WebCore::AudioSession::sampleRate): Ditto.
(WebCore::AudioSession::numberOfOutputChannels): Ditto.
(WebCore::AudioSession::setActive): Ditto.
(WebCore::AudioSession::preferredBufferDuration): Ditto.
(WebCore::AudioSession::setPreferredBufferDuration): Ditto.

Use the new AudioSession class in place of explicit AudioToolbox or AVAudioSession calls:
* platform/audio/ios/AudioDestinationIOS.cpp:
(WebCore::AudioDestination::hardwareSampleRate): Pass to AudioSession.
(WebCore::AudioDestination::maxChannelCount): Ditto.
(WebCore::AudioDestinationIOS::AudioDestinationIOS): Register for listeners and activate.
(WebCore::AudioDestinationIOS::~AudioDestinationIOS): Unregister and deactivate.
(WebCore::AudioDestinationIOS::configure): Use AudioSession to set the buffer length.
* platform/audio/ios/AudioDestinationIOS.h:

Add the new files to the project:
* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147494 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpstream AudioDestinationIOS.
jer.noble@apple.com [Tue, 2 Apr 2013 20:47:12 +0000 (20:47 +0000)]
Upstream AudioDestinationIOS.
https://bugs.webkit.org/show_bug.cgi?id=113806

Reviewed by Eric Carlson.

Upstream AudioDestinationIOS from the iOS repository.

* WebCore.xcodeproj/project.pbxproj:
* platform/audio/ios/AudioDestinationIOS.cpp: Added.
(WebCore::audioDestinations):
(WebCore::AudioDestination::create):
(WebCore::AudioDestination::hardwareSampleRate):
(WebCore::AudioDestination::maxChannelCount):
(WebCore::AudioDestinationIOS::AudioDestinationIOS):
(WebCore::AudioDestinationIOS::~AudioDestinationIOS):
(WebCore::AudioDestinationIOS::configure):
(WebCore::AudioDestinationIOS::audioDestinationInterruptionListener):
(WebCore::AudioDestinationIOS::initializeAudioSession):
(WebCore::AudioDestinationIOS::start):
(WebCore::AudioDestinationIOS::stop):
(WebCore::AudioDestinationIOS::beganAudioInterruption):
(WebCore::AudioDestinationIOS::endedAudioInterruption):
(WebCore::AudioDestinationIOS::render):
(WebCore::AudioDestinationIOS::inputProc):
(WebCore::AudioDestinationIOS::frameSizeChangedProc):
* platform/audio/ios/AudioDestinationIOS.h: Added.
(AudioDestinationIOS):
(WebCore::AudioDestinationIOS::isPlaying):
(WebCore::AudioDestinationIOS::sampleRate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agofast/dom/HTMLImageElement/image-alt-text.html and fast/dom/HTMLInputElement/input...
ch.dumez@sisa.samsung.com [Tue, 2 Apr 2013 20:35:33 +0000 (20:35 +0000)]
fast/dom/HTMLImageElement/image-alt-text.html and fast/dom/HTMLInputElement/input-image-alt-text.html are failing
https://bugs.webkit.org/show_bug.cgi?id=94198

Reviewed by Darin Adler.

Source/WebCore:

Fix alt text not being displayed for img elements or input of type
"image" due to insufficient size. imageSizeForError() was adding
padding for the alt text but setImageSizeForAltText() was not. The
issue was occurring in paintReplaced() because the contentWidth() /
contentHeight() would be exactly the same size as the alt text.
However, paintReplaced() first draws an outline rect where the
image should be and then reduces the usable width / height for the
alt text by 2 so that the text does not get printed over the rect
borders. However, after decreasing the usable width / height, there
is no longer enough space to draw the text.

No new tests, already covered by existing tests.

* rendering/RenderImage.cpp:
(WebCore::RenderImage::setImageSizeForAltText): Add padding to the text size
so that the alt text can still be drawn, despite the outline rect
borders. Such padding was already used in imageSizeForError(). Also use
ceilf() for the value returned by Font::width() instead of truncating it.
This is needed because we need to fit the whole text in the image and this
may not be the case if we truncate.
(WebCore::RenderImage::paintReplaced):

LayoutTests:

Rebaseline several test cases that have better result now and add those
tests to TestExpectations for other ports since they will need a
rebaseline.

* platform/chromium/TestExpectations:
* platform/efl/fast/block/float/002-expected.png:
* platform/efl/fast/block/float/017-expected.png:
* platform/efl/fast/dom/HTMLImageElement/image-alt-text-expected.png:
* platform/efl/fast/dom/HTMLImageElement/image-alt-text-expected.txt:
* platform/efl/fast/dom/HTMLInputElement/input-image-alt-text-expected.png:
* platform/efl/fast/dom/HTMLInputElement/input-image-alt-text-expected.txt:
* platform/efl/fast/encoding/utf-16-big-endian-expected.png:
* platform/efl/fast/encoding/utf-16-big-endian-expected.txt:
* platform/efl/fast/encoding/utf-16-little-endian-expected.png:
* platform/efl/fast/encoding/utf-16-little-endian-expected.txt:
* platform/efl/fast/flexbox/023-expected.png:
* platform/efl/fast/flexbox/024-expected.png:
* platform/efl/fast/forms/input-value-expected.png:
* platform/efl/fast/forms/input-value-expected.txt:
* platform/efl/fast/invalid/012-expected.png:
* platform/efl/fast/lists/inlineBoxWrapperNullCheck-expected.png:
* platform/efl/fast/lists/inlineBoxWrapperNullCheck-expected.txt:
* platform/efl/fast/parser/comment-in-script-expected.png:
* platform/efl/fast/text/whitespace/normal-after-nowrap-breaking-expected.png:
* platform/efl/tables/mozilla/bugs/bug2997-expected.png:
* platform/efl/tables/mozilla/collapsing_borders/bug41262-3-expected.png:
* platform/efl/tables/mozilla/collapsing_borders/bug41262-3-expected.txt:
* platform/gtk/TestExpectations:
* platform/mac/TestExpectations:
* platform/qt/TestExpectations:
* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147492 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd support for using ARM FFT in WebAudio
rtoy@google.com [Tue, 2 Apr 2013 20:24:49 +0000 (20:24 +0000)]
Add support for using ARM FFT in WebAudio
https://bugs.webkit.org/show_bug.cgi?id=109755

Reviewed by Chris Rogers.

Source/WebCore:

No new tests.

* WebCore.gyp/WebCore.gyp: Add dependency on openmax_dl when use_openmax_dl_fft is enabled.
* WebCore.gypi: Add source FFTFrameOpenMAXDLAndroid.cpp
* platform/audio/AudioArray.h:
(WebCore::AudioArray::allocate): Need 32-byte aligntment with the
OpenMAX DL FFT.
* platform/audio/FFTFrame.h:
(FFTFrame): Support OpenMAX DL FFT
* platform/audio/FFTFrameStub.cpp: Support OpenMAX DL FFT
* platform/audio/chromium/FFTFrameOpenMAXDLAndroid.cpp: Added.  Implements the
necessary interface using the OpenMAX DL FFT routines.
(WebCore):
(WebCore::FFTFrame::FFTFrame):
(WebCore::FFTFrame::initialize):
(WebCore::FFTFrame::cleanup):
(WebCore::FFTFrame::~FFTFrame):
(WebCore::FFTFrame::multiply):
(WebCore::FFTFrame::doFFT):
(WebCore::FFTFrame::doInverseFFT):
(WebCore::FFTFrame::realData):
(WebCore::FFTFrame::imagData):
(WebCore::FFTFrame::contextForSize):

Source/WebKit/chromium:

* features.gypi: Support building with the OpenMAX DL FFT.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147491 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[GStreamer] Memory leaks in MediaPlayerPrivateGStreamer
commit-queue@webkit.org [Tue, 2 Apr 2013 20:04:00 +0000 (20:04 +0000)]
[GStreamer] Memory leaks in MediaPlayerPrivateGStreamer
https://bugs.webkit.org/show_bug.cgi?id=113762

Patch by Sudarsana Nagineni <sudarsana.nagineni@intel.com> on 2013-04-02
Reviewed by Philippe Normand.

Don't leak strings returned by gst_element_get_name() and
gst_missing_plugin_message_get_installer_detail().

No new tests. No change in behavior.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147490 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAdd WKContextSetInvalidMessageFunction
andersca@apple.com [Tue, 2 Apr 2013 20:00:25 +0000 (20:00 +0000)]
Add WKContextSetInvalidMessageFunction
https://bugs.webkit.org/show_bug.cgi?id=113820
<rdar://problem/12679817>

Reviewed by Tim Horton.

* UIProcess/API/C/WKContext.cpp:
(WKContextSetInvalidMessageFunction):
Call WebContext::didReceiveInvalidMessage.

* UIProcess/API/C/WKContextPrivate.h:
* UIProcess/WebContext.cpp:
(WebKit::WebContext::setInvalidMessageCallback):
Set the global variable.

(WebKit::WebContext::didReceiveInvalidMessage):
Construct a message name string and call the invalid message callback.

* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::didReceiveInvalidMessage):
Call WebContext::didReceiveInvalidMessage.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147489 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRebaseline overflowHidden.html's test results after r147245.
benjamin@webkit.org [Tue, 2 Apr 2013 19:57:58 +0000 (19:57 +0000)]
Rebaseline overflowHidden.html's test results after r147245.

Unreviewed.

The patch r147245 (https://bugs.webkit.org/show_bug.cgi?id=106143) updated results for

Patch by Benjamin Poulain <bpoulain@apple.com> on 2013-04-02
Reviewed by Tony Chang.
Chromium, but not for the other platforms.

The new results on Mac, EFL and Win match the common reference.

* platform/efl/fast/table/overflowHidden-expected.txt: Removed.
* platform/mac/fast/table/overflowHidden-expected.txt: Removed.
* platform/win-7sp0/fast/table: Removed.
* platform/win-7sp0/fast/table/overflowHidden-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147488 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix css style for CC button on android
commit-queue@webkit.org [Tue, 2 Apr 2013 19:53:35 +0000 (19:53 +0000)]
Fix css style for CC button on android
https://bugs.webkit.org/show_bug.cgi?id=113815

Patch by Min Qin <qinmin@chromium.org> on 2013-04-02
Reviewed by Adam Barth.

The default CC button size is 16px as defined in mediaControls.css. Android should use 35px.
This change fixed that problem.
Also fixed a problem that r142947 does not change the display of fullscreen button to -webkit-flex
Android does not run media layout tests currently. Will fix the test expectations when layout tests are run in content shell.

* css/mediaControlsChromiumAndroid.css:
(video::-webkit-media-controls-fullscreen-button):
(audio::-webkit-media-controls-toggle-closed-captions-button, video::-webkit-media-controls-toggle-closed-captions-button):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147486 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago [Mac] Remove WKCreateNSURLConnectionDelegateProxy
ap@apple.com [Tue, 2 Apr 2013 19:52:22 +0000 (19:52 +0000)]
    [Mac] Remove WKCreateNSURLConnectionDelegateProxy
        https://bugs.webkit.org/show_bug.cgi?id=113810

        Reviewed by Anders Carlsson.

        This WKSI function is not used any more.

        * WebCore.exp.in:
        * platform/mac/WebCoreSystemInterface.h:
        * platform/mac/WebCoreSystemInterface.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147485 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[CSS Regions] Remove font-smoothing from fast/regions/region-style-inline-background...
abucur@adobe.com [Tue, 2 Apr 2013 19:50:45 +0000 (19:50 +0000)]
[CSS Regions] Remove font-smoothing from fast/regions/region-style-inline-background-color.html
https://bugs.webkit.org/show_bug.cgi?id=113814

Reviewed by James Robinson.

The test uses font smoothing and makes the test fail on Chromium Mountain Lion. This happens
because of how regions handle overflow and clipping. For now, disable the font smoothing in
the test.

* fast/regions/region-style-inline-background-color-expected.html:
* fast/regions/region-style-inline-background-color.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147484 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoDisable running tests on the MAC WK2 EWS queue.
lforschler@apple.com [Tue, 2 Apr 2013 19:13:18 +0000 (19:13 +0000)]
Disable running tests on the MAC WK2 EWS queue.

Unreviewed.

* Scripts/webkitpy/tool/commands/earlywarningsystem.py:
(MacWK2EWS):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147483 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoProvide a user default that can be used to disable docking of the Web Inspector.
timothy@apple.com [Tue, 2 Apr 2013 19:06:39 +0000 (19:06 +0000)]
Provide a user default that can be used to disable docking of the Web Inspector.

The user default is "WebKit Web Inspector Setting - inspectorAttachDisabled".

https://webkit.org/b/113779
rdar://problem/13446021

Reviewed by Joseph Pecoraro.

Source/WebKit/cf:

* WebCoreSupport/WebInspectorClientCF.cpp:
(WebInspectorClient::inspectorAttachDisabled): Added.
(WebInspectorClient::setInspectorAttachDisabled): Added.

Source/WebKit/mac:

* WebCoreSupport/WebInspectorClient.h:
(WebInspectorClient::inspectorAttachDisabled): Added.
(WebInspectorClient::setInspectorAttachDisabled): Added.

* WebCoreSupport/WebInspectorClient.mm:
(WebInspectorClient::didResizeMainFrame):
(-[WebInspectorWindowController window]):
(-[WebInspectorWindowController showWindow:]):
Check inspectorAttachDisabled() in places where we attach or update the dock button.

Source/WebKit/win:

* WebCoreSupport/WebInspectorClient.h:
(WebInspectorClient::inspectorAttachDisabled): Added.
(WebInspectorClient::setInspectorAttachDisabled): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147482 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, rolling out r147464.
commit-queue@webkit.org [Tue, 2 Apr 2013 18:54:53 +0000 (18:54 +0000)]
Unreviewed, rolling out r147464.
http://trac.webkit.org/changeset/147464
https://bugs.webkit.org/show_bug.cgi?id=113812

JSSQLResultSet.cpp:131:24: error: implicit conversion loses
integer precision: 'int64_t' (aka 'long long') to 'int'
[-Werror,-Wshorten-64-to-32] (Requested by xenon on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2013-04-02

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateImplementation):
* bindings/scripts/test/JS/JSTestObj.cpp:
(WebCore::jsTestObjAttrWithGetterException):
(WebCore::jsTestObjStringAttrWithGetterException):
(WebCore::jsTestObjWithScriptStateAttributeRaises):
(WebCore::jsTestObjWithScriptExecutionContextAttributeRaises):
(WebCore::jsTestObjWithScriptExecutionContextAndScriptStateAttributeRaises):
(WebCore::jsTestObjNullableDoubleAttribute):
(WebCore::jsTestObjNullableLongAttribute):
(WebCore::jsTestObjNullableBooleanAttribute):
(WebCore::jsTestObjNullableStringAttribute):
(WebCore::jsTestObjNullableLongSettableAttribute):
(WebCore::jsTestObjNullableStringValue):
* bindings/scripts/test/JS/JSTestTypedefs.cpp:
(WebCore::jsTestTypedefsAttrWithGetterException):
(WebCore::jsTestTypedefsStringAttrWithGetterException):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147480 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoRebaseline svg/as-image/image-respects-pageScaleFactor expectations
pdr@google.com [Tue, 2 Apr 2013 18:42:10 +0000 (18:42 +0000)]
Rebaseline svg/as-image/image-respects-pageScaleFactor expectations

Unreviewed rebaseline of test expectations.

* platform/chromium-linux/svg/as-image/image-respects-pageScaleFactor-expected.png:
* platform/chromium-mac-lion/svg/as-image/image-respects-pageScaleFactor-expected.png:
* platform/chromium-mac-snowleopard/svg/as-image/image-respects-pageScaleFactor-expected.png:
* platform/chromium-mac/svg/as-image/image-respects-pageScaleFactor-expected.png:
* platform/chromium-win/svg/as-image/image-respects-pageScaleFactor-expected.png:
* platform/chromium-win/svg/as-image/image-respects-pageScaleFactor-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147479 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: Make Source Frame & status bar a layout boundary
commit-queue@webkit.org [Tue, 2 Apr 2013 18:37:44 +0000 (18:37 +0000)]
Web Inspector: Make Source Frame & status bar a layout boundary
https://bugs.webkit.org/show_bug.cgi?id=112353

Patch by Andrey Lushnikov <lushnikov@chromium.org> on 2013-04-02
Reviewed by Pavel Feldman.

No new tests: no change in behaviour.

- Add View.markAsLayoutBoundary method to mark a view as a layout
boundary. This will force this view to autoupdate its height to actual
pixel value.
- Mark both CodeMirrorTextEditor and DefaultTextEditor as a layout
boundary.
- Mark main-status-bar element with a layout-boundary class.

* inspector/front-end/CodeMirrorTextEditor.js:
(WebInspector.CodeMirrorTextEditor):
* inspector/front-end/DefaultTextEditor.js:
* inspector/front-end/View.js:
(WebInspector.View.prototype.markAsLayoutBoundary):
(WebInspector.View.prototype._processWillShow):
(WebInspector.View.prototype._processWasShown):
(WebInspector.View.prototype._processOnResize):
* inspector/front-end/inspector.css:
(.layout-boundary):
* inspector/front-end/inspector.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147478 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[chromium] Initialize all fields in WebFilterOperation
danakj@chromium.org [Tue, 2 Apr 2013 18:36:40 +0000 (18:36 +0000)]
[chromium] Initialize all fields in WebFilterOperation
https://bugs.webkit.org/show_bug.cgi?id=113807

Reviewed by James Robinson.

* chromium/public/WebFilterOperation.h:
(WebKit::WebFilterOperation::WebFilterOperation):
* chromium/src/WebFilterOperation.cpp:
(WebKit::WebFilterOperation::WebFilterOperation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147477 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago <rdar://problem/13551119> [WK2] Crashes in NetworkProcess when canceling...
ap@apple.com [Tue, 2 Apr 2013 18:36:35 +0000 (18:36 +0000)]
    <rdar://problem/13551119> [WK2] Crashes in NetworkProcess when canceling loads
        https://bugs.webkit.org/show_bug.cgi?id=113803

        Reviewed by Darin Adler.

        Stop using a delegate proxy. It hasn't been needed in years, and the extra layer
        just adds bugs.

        This specific issue gets fixed because connection retains its delegate is retained,
        but a proxy does not.

        * WebCore.exp.in:
        * platform/network/ResourceHandle.h:
        * platform/network/ResourceHandleInternal.h:
        * platform/network/mac/ResourceHandleMac.mm:
        (WebCore::ResourceHandle::start):
        (WebCore::ResourceHandle::releaseDelegate):
        (WebCore::ResourceHandle::platformLoadResourceSynchronously):
        Removed code that was dealing with delegate proxy.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147476 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: Encapsulate SetEmbedderData/GetEmbedderData
commit-queue@webkit.org [Tue, 2 Apr 2013 18:35:42 +0000 (18:35 +0000)]
Web Inspector: Encapsulate SetEmbedderData/GetEmbedderData
https://bugs.webkit.org/show_bug.cgi?id=113331

Patch by John J. Barton <johnjbarton@chromium.org> on 2013-04-02
Reviewed by Pavel Feldman.

Refactor to encapsulate uses of V8 Set/GetEmbedderData together
with uses of V8 Set/GetAlignedPointerInEmbedderData since these
functions operate on the same underyling array in V8.

* bindings/v8/ScriptController.cpp:
(WebCore::ScriptController::setContextDebugId):
(WebCore::ScriptController::contextDebugId):
* bindings/v8/V8DOMWindowShell.cpp:
(WebCore::setInjectedScriptContextDebugId):
* bindings/v8/V8PerContextData.cpp:
(WebCore::V8PerContextDebugData::setContextDebugData):
(WebCore):
(WebCore::V8PerContextDebugData::contextDebugId):
(WebCore::V8PerContextDebugData::createDebugData):
(WebCore::V8PerContextDebugData::getDebugData):
(WebCore::V8PerContextDebugData::setDebugData):
* bindings/v8/V8PerContextData.h:
(V8PerContextDebugData):
(WebCore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147475 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: Adjust toolbar labels position
commit-queue@webkit.org [Tue, 2 Apr 2013 18:32:57 +0000 (18:32 +0000)]
Web Inspector: Adjust toolbar labels position
https://bugs.webkit.org/show_bug.cgi?id=113805

Patch by Alexei Filippov <alph@chromium.org> on 2013-04-02
Reviewed by Pavel Feldman.

Move the labels 1px up to make them centered.
Also fix font centering in native memory distribution snapshot.

* inspector/front-end/inspector.css:
(.toolbar-label):
* inspector/front-end/nativeMemoryProfiler.css:
(.native-snapshot-view .data-grid td):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed, chromium expectations update.
japhet@chromium.org [Tue, 2 Apr 2013 18:30:53 +0000 (18:30 +0000)]
Unreviewed, chromium expectations update.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147473 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix Windows build.
timothy@apple.com [Tue, 2 Apr 2013 18:27:33 +0000 (18:27 +0000)]
Fix Windows build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147472 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoViewport constrained elements should work properly with headers and footers
bdakin@apple.com [Tue, 2 Apr 2013 18:25:41 +0000 (18:25 +0000)]
Viewport constrained elements should work properly with headers and footers
https://bugs.webkit.org/show_bug.cgi?id=113743

Reviewed by Simon Fraser.

Source/WebCore:

ScrollableArea::constrainScrollPositionForOverhang() needs to take headerHeight
and footerHeight into account. It also needs to use totalContentsSize() instead of
contentsSize().

Send totalContentsSize() instead of contentsSize() and also send headerHeight()
and footerHeight().
* page/FrameView.cpp:
(WebCore::FrameView::scrollOffsetForFixedPosition):
* page/FrameView.h:
(FrameView):

The scrolling tree will also need to know about header and footer heights since it
will need these values to call
ScrollableArea::constrainScrollPositionForOverhang().
* page/scrolling/ScrollingStateScrollingNode.cpp:
(WebCore::ScrollingStateScrollingNode::ScrollingStateScrollingNode):
(WebCore::ScrollingStateScrollingNode::setHeaderHeight):
(WebCore::ScrollingStateScrollingNode::setFooterHeight):
* page/scrolling/ScrollingStateScrollingNode.h:
(WebCore::ScrollingStateScrollingNode::headerHeight):
(ScrollingStateScrollingNode):
(WebCore::ScrollingStateScrollingNode::footerHeight):
* page/scrolling/ScrollingTreeScrollingNode.cpp:
(WebCore::ScrollingTreeScrollingNode::ScrollingTreeScrollingNode):
(WebCore::ScrollingTreeScrollingNode::updateBeforeChildren):
* page/scrolling/ScrollingTreeScrollingNode.h:
(WebCore::ScrollingTreeScrollingNode::headerHeight):
(WebCore::ScrollingTreeScrollingNode::footerHeight):
(ScrollingTreeScrollingNode):
* page/scrolling/mac/ScrollingCoordinatorMac.h:
(ScrollParameters):
* page/scrolling/mac/ScrollingCoordinatorMac.mm:
(WebCore::ScrollingCoordinatorMac::frameViewLayoutUpdated):
(WebCore::ScrollingCoordinatorMac::setScrollParametersForNode):
* page/scrolling/mac/ScrollingTreeScrollingNodeMac.mm:
(WebCore::ScrollingTreeScrollingNodeMac::setScrollLayerPosition):

Use footerHeight when calculating the maxValue and use headerHeight in determining
what the minimum value is.
* platform/ScrollableArea.cpp:
(WebCore::constrainedScrollPosition):
(WebCore::ScrollableArea::constrainScrollPositionForOverhang):
* platform/ScrollableArea.h:
(ScrollableArea):

LayoutTests:

* platform/mac-wk2/tiled-drawing/fixed/four-bars-with-header-and-footer-expected.txt: Added.
* platform/mac-wk2/tiled-drawing/fixed/four-bars-with-header-and-footer.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147471 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Skia] Enable GPU-accelerated Skia path for FEComposite filter.
senorblanco@chromium.org [Tue, 2 Apr 2013 18:23:30 +0000 (18:23 +0000)]
[Skia] Enable GPU-accelerated Skia path for FEComposite filter.
https://bugs.webkit.org/show_bug.cgi?id=113797

Reviewed by Adrienne Walker.

Source/WebCore:

Tests: css3/filters/effect-reference-composite-hw.html
       css3/filters/effect-reference-composite.html

* WebCore.gypi:
* platform/graphics/filters/FEComposite.h:
Enable Skia DAG hook for FEComposite.
* platform/graphics/filters/skia/FECompositeSkia.cpp: Added.
(WebCore::FEComposite::createImageFilter):
Implement skia DAG builder for FEComposite, and implement generic
skia version of composite filter.

LayoutTests:

* css3/filters/effect-reference-composite-hw.html: Added.
* css3/filters/effect-reference-composite.html: Added.
* platform/chromium-linux/css3/filters/effect-reference-composite-expected.png: Added.
* platform/chromium-linux/css3/filters/effect-reference-composite-expected.txt: Added.
* platform/chromium-linux/css3/filters/effect-reference-composite-hw-expected.png: Added.
* platform/chromium-linux/css3/filters/effect-reference-composite-hw-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147470 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: Avoid using style.display="inline-block"
commit-queue@webkit.org [Tue, 2 Apr 2013 18:20:14 +0000 (18:20 +0000)]
Web Inspector: Avoid using style.display="inline-block"
https://bugs.webkit.org/show_bug.cgi?id=113802

Patch by Andrey Lushnikov <lushnikov@chromium.org> on 2013-04-02
Reviewed by Pavel Feldman.

No new tests.

- Create class "inline-block" which will set "display: inline-block"
style.
- Use this class with DIV elements to get "inline-block" element.

* inspector/front-end/ObjectPopoverHelper.js:
(WebInspector.ObjectPopoverHelper.prototype.):
(WebInspector.ObjectPopoverHelper.prototype._showObjectPopover):
* inspector/front-end/ScriptsPanel.js:
(WebInspector.ScriptsPanel):
* inspector/front-end/inspector.css:
(.inline-block):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147469 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoWeb Inspector: [Timeline] warning markers are not propagated to coalesced records
caseq@chromium.org [Tue, 2 Apr 2013 18:18:48 +0000 (18:18 +0000)]
Web Inspector: [Timeline] warning markers are not propagated to coalesced records
https://bugs.webkit.org/show_bug.cgi?id=113789

Reviewed by Pavel Feldman.

Source/WebCore:

Test: inspector/timeline/timeline-coalescing.html

* inspector/front-end/TimelinePresentationModel.js: set childHasWarning on coalescing record if the record it replaces had warnings.
(WebInspector.TimelinePresentationModel.prototype._replaceWithCoalescedRecord):

LayoutTests:

* inspector/timeline/timeline-coalescing-expected.txt: Added.
* inspector/timeline/timeline-coalescing.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[BlackBerry] Create a selection from two points irregardless of the caret position.
commit-queue@webkit.org [Tue, 2 Apr 2013 18:17:10 +0000 (18:17 +0000)]
[BlackBerry] Create a selection from two points irregardless of the caret position.
https://bugs.webkit.org/show_bug.cgi?id=113795

Patch by Nima Ghanavatian <nghanavatian@blackberry.com> on 2013-04-02
Reviewed by Rob Buis.

Internally reviewed by Mike Fenton.

PR276577
If we are to set both the base and extent of a selection, we should maintain the base upon setting the extent.
This prevents us from overwriting the previous calculation when setting extent.
Further, instead of setting and manipulating the selection, simply set up start and end VisiblePositions and
create a new directional selection so as to set the base and extent simultaneously.

* WebKitSupport/SelectionHandler.cpp:
(BlackBerry::WebKit::SelectionHandler::setSelection):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@147467 268f45cc-cd09-0410-ab3c-d52691b4dbfc