WebKit.git
12 years ago2010-09-22 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Wed, 22 Sep 2010 21:56:20 +0000 (21:56 +0000)]
2010-09-22  Dirk Pranke  <dpranke@chromium.org>

        Unreviewed, build fix.

        Fix bustage of rebaseline-chromium-webkit-tests resulting from
        r67974. Really need better unit tests for this tool :(

        * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68085 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoUnbreak test-webkitpy
aroben@apple.com [Wed, 22 Sep 2010 21:43:43 +0000 (21:43 +0000)]
Unbreak test-webkitpy

* Scripts/webkitpy/test/main.py:
(Tester.run_tests): Add a line that mysteriously got deleted.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68084 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agohttps://bugs.webkit.org/show_bug.cgi?id=46304
hyatt@apple.com [Wed, 22 Sep 2010 21:37:26 +0000 (21:37 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=46304

Reviewed by Dan Bernstein.

display:inline should become display:inline-block when an object's block flow does not match its parent's block flow.

Added fast/blockflow/display-mutation.html

WebCore:

* css/CSSStyleSelector.cpp:
(WebCore::CSSStyleSelector::styleForElement):
(WebCore::CSSStyleSelector::pseudoStyleForElement):
(WebCore::CSSStyleSelector::adjustRenderStyle):
* css/CSSStyleSelector.h:

LayoutTests:

* fast/blockflow/display-mutation-expected.txt: Added.
* fast/blockflow/display-mutation.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68083 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoAdd a ViewportConfiguration class for the Qt WebKit2 API, which
kenneth@webkit.org [Wed, 22 Sep 2010 21:37:12 +0000 (21:37 +0000)]
Add a ViewportConfiguration class for the Qt WebKit2 API, which
currently calculated a fallback viewport configuration.

Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2010-09-22
Reviewed by Antonio Gomes.

* UIProcess/API/qt/qwkpage.cpp:
(QWKPage::ViewportConfiguration::ViewportConfiguration):
(QWKPage::ViewportConfiguration::~ViewportConfiguration):
(QWKPage::ViewportConfiguration::operator=):
(QWKPage::viewportConfigurationForSize):
* UIProcess/API/qt/qwkpage.h:
* UIProcess/API/qt/qwkpage_p.h:
(QtViewportConfigurationPrivate::QtViewportConfigurationPrivate):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68082 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoWindows build fix
aroben@apple.com [Wed, 22 Sep 2010 21:14:41 +0000 (21:14 +0000)]
Windows build fix

* WebProcess/WebKitMain.cpp:
(WebKitMain): Fix typo

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68081 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoMake test-webkitpy test WebKit2's scripts
aroben@apple.com [Wed, 22 Sep 2010 20:44:10 +0000 (20:44 +0000)]
Make test-webkitpy test WebKit2's scripts

These scripts can't be in WebKitTools due to limitations of Apple's
build process. But that doesn't mean we can't test them!

Fixes <http://webkit.org/b/46297> test-webkitpy should test code in
WebKit2/Scripts

Reviewed by Adam Barth.

* Scripts/test-webkitpy:
(_clean_packages_with_test): Renamed from _clean_webkitpy_with_test.
Now takes an external_package_paths parameter and cleans both webkitpy
and any external packages.
(init): Added an external_package_paths parameter which we pass along
to _clean_packages_with_test.
(top level): Add WebKit2/Scripts/webkit2 as our only external package
and pass it along to init and Tester.run_tests.

* Scripts/webkitpy/test/main.py:
(Tester.run_tests): Added an optional external_package_paths
parameter. We modify sys.path so that the external packages can be
imported, and search for unittest files inside all external packages
in addition to inside webkitpy.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68080 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoAutogenerate WebPage's message-receiving code and message types
aroben@apple.com [Wed, 22 Sep 2010 20:43:49 +0000 (20:43 +0000)]
Autogenerate WebPage's message-receiving code and message types

This patch encompasses several changes that allow message types and
receiving code to be generated by a script, and add some type-safety
as a bonus. Messages are now represented by structs instead of an ID +
ArgumentCoder. The struct contains the arguments and the ID together,
and has a constructor that enforces the use of correct types.
Correspondingly, a new overload of WebProcessProxy::send that takes a
message struct (instead of a message ID and separate arguments) has
been added. Eventually all callers should use this overload and the
old one can be removed.

This patch only touches WebPage's messages. We should transition other
message receivers over to this new system eventually.

Fixes <http://webkit.org/b/43636> <rdar://problem/8282462> Add a
type-safe IPC mechanism to WebKit2

Reviewed by Anders Carlsson.

* DerivedSources.make: Added. Calls generate-message-receiver.py and
generate-messages-header.py for each message receiver it knows about
(just WebPage for now).

* Platform/CoreIPC/Arguments.h: Added First/Second/ThirdArgumentType
typedefs for use in handleMessage.

* Platform/CoreIPC/HandleMessage.h: Added.
(CoreIPC::handleMessage): This overloaded function template decodes
arguments and passes them along to the specified function.

* Scripts/generate-message-receiver.py: Added.
* Scripts/generate-messages-header.py: Added.
These scripts just wrap functionality in messages.py.

* Scripts/webkit2/__init__.py: Added. This just exists so that Python
will treat this directory as a package.

* Scripts/webkit2/messages.py: Added. Contains the code to parse
messages files and generate .cpp/.h files from them.
(MessageReceiver.__init__): This class represents a single receiver of
messages.
(MessageReceiver.iterparameters): Returns a generator that can be used
to iterate over all the parameters of all the messages of this
receiver.
(MessageReceiver.parse): Reads a messages file from a file-like object
and parses it into a MessageReceiver object.
(Message.__init__): This class represents a single message.
(Message.id): Returns the ID name for this message.
(Parameter.__init__): This class represents a single parameter for a
message.
(messages_header_filename): Returns the name of the header that
defines the messages for a given receiver.
(surround_in_condition): Surrounds the given string in #if/#endif if
there is an associated condition.
(messages_to_kind_enum): Returns a string that defines the Kind enum
for these messages.
(function_parameter_type): Returns the type that should be used when
passing a value of the given type as a parameter to a function.
(base_class): Returns the base class for a message struct.
(message_to_struct_declaration): Returns a string that declares the
struct for this message.
(forward_declarations_for_namespace): Returns a string that contains
forward-declarations for a set of types in a given namespace.
(forward_declarations): Returns a string that contains all the
forward-declarations needed in order to declare all the message
structs for this receiver.
(generate_header_file): Returns a string containing the messages
header file for this receiver.
(handler_function): Returns the name of the function that handles a
given message for a given receiver.
(case_statement): Returns a string containing a case statement for
handling the given message.
(header_for_type): Returns the header needed to define a given type,
enclosed in quotes or angle brackets as needed.
(generate_message_handler): Returns a string containing the contents
of a .cpp file that defines a didReceive*Message function.

* Scripts/webkit2/messages_unittest.py: Added. Contains tests for
messages.py.

* Shared/CoreIPCSupport/WebPageMessageKinds.h: Removed. This has been
replaced by a generated WebPageMessages.h header.

* UIProcess/WebEditCommandProxy.cpp:
* UIProcess/WebPageProxy.cpp:
Updated to use the new message structs and WebProcessProxy::send
overload.

* UIProcess/WebProcessProxy.h:
(WebKit::WebProcessProxy::send): Added this new overload that takes a
message struct.

* WebKit2.xcodeproj/project.pbxproj: Added a Derived Sources shell
script target that invokes DerivedSources.make. Added "Derived
Sources" and "Scripts" groups that contain the various new files.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::didReceivePolicyDecision):
(WebKit::WebPage::getSourceForFrame):
Changed these functions to take the raw IPC types and do the necessary
translation themselves. This keeps the generated code from need to
know how to perform the translation.

(WebKit::WebPage::didReceiveMessage): Replaced handling of WebPage
messages with a call to didReceiveWebPageMessage, whose implementation
is generated by the scripts.

* WebProcess/WebPage/WebPage.h: Added didReceiveWebPageMessage.

* WebProcess/WebPage/WebPage.messages.in: Added. This file declares
all of the messages that WebPage receives, roughly grouped by
functionality.

* win/WebKit2.vcproj: Added "Derived Sources" and "Scripts" filters
that contain the various new files. Let VS resort some other files.

* win/WebKit2Common.vsprops: Added
$(WebKitOutputDir)\obj\$(ProjectName)\DerivedSources to the include
path so that the generated messages header can be found.

* win/WebKit2.make: Copy the generated source files to $(DSTROOT).

* win/WebKit2Generated.make: Added a call to build-generated-files.sh.

* win/WebKit2Generated.vcproj: Added build-generated-files.sh.

* win/build-generated-files.sh: Added. Invokes DerivedSources.make.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68079 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Mario Sanchez Prada <msanchez@igalia.com>
commit-queue@webkit.org [Wed, 22 Sep 2010 20:43:26 +0000 (20:43 +0000)]
2010-09-22 Mario Sanchez Prada <msanchez@igalia.com>

        Reviewed by Martin Robinson.

        [Gtk] object:text-changed events should be emitted for entries and password text
        https://bugs.webkit.org/show_bug.cgi?id=25898

        Implement proper 'text-changed' signal emission for the GTK port

        Call deleteTextFromNode() when needed while removing text nodes.
        Do it even when removeNode() is going to be called afterwards, in
        order to allow accessibility to get properly notified about the
        text being removed alongside with that node.

        * editing/DeleteSelectionCommand.cpp:
        (WebCore::DeleteSelectionCommand::handleGeneralDelete):

        New function in AXObjectCache to call when text changes in a node.
        Added one new function to allow notifying something changed in a
        text node through the associated RenderObject, making such a
        function dependant on the platform-specific implementation,
        provided through a protected function (provided a proper
        implementation for the GTK port and a dummy one for the others).

        * accessibility/AXObjectCache.cpp:
        (WebCore::AXObjectCache::nodeTextChangeNotification): New
        * accessibility/AXObjectCache.h:
        (WebCore::AXObjectCache::AXTextChange): New enumeration
        (WebCore::AXObjectCache::nodeTextChangeNotification): New
        (WebCore::AXObjectCache::nodeTextChangePlatformNotification): New
        * accessibility/AccessibilityRenderObject.h:
        (WebCore::toAccessibilityRenderObject):
        * accessibility/chromium/AXObjectCacheChromium.cpp:
        (WebCore::AXObjectCache::nodeTextChangePlatformNotification): New
        * accessibility/gtk/AXObjectCacheAtk.cpp:
        (WebCore::emitTextChanged):
        (WebCore::AXObjectCache::nodeTextChangePlatformNotification): New
        * accessibility/mac/AXObjectCacheMac.mm:
        (WebCore::AXObjectCache::nodeTextChangePlatformNotification): New
        * accessibility/win/AXObjectCacheWin.cpp:
        (WebCore::AXObjectCache::nodeTextChangePlatformNotification): New

        Notify accessibility when something changes in a text node.
        Call to AXObjectCache::nodeTextChangeNotification() to notify when
        text was inserted/deleted when applying/unapplying a text edition
        command, along with the offset in the original text where the
        change took place and the number of characters that got affected.

        * editing/AppendNodeCommand.cpp:
        (WebCore::sendAXTextChangedIgnoringLineBreaks):
        (WebCore::AppendNodeCommand::doApply):
        (WebCore::AppendNodeCommand::doUnapply):
        * editing/DeleteFromTextNodeCommand.cpp:
        (WebCore::DeleteFromTextNodeCommand::doApply):
        (WebCore::DeleteFromTextNodeCommand::doUnapply):
        * editing/InsertIntoTextNodeCommand.cpp:
        (WebCore::InsertIntoTextNodeCommand::doApply):
        (WebCore::InsertIntoTextNodeCommand::doUnapply):
        * editing/InsertNodeBeforeCommand.cpp:
        (WebCore::InsertNodeBeforeCommand::doApply):
        (WebCore::InsertNodeBeforeCommand::doUnapply):
        (WebCore::DeleteSelectionCommand::handleGeneralDelete):
2010-09-22  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Martin Robinson.

        [Gtk] object:text-changed events should be emitted for entries and password text
        https://bugs.webkit.org/show_bug.cgi?id=25898

        New unit test to make sure text-changed signals are emitted

        * tests/testatk.c:
        (textChangedCb): New. Signal handler for the
        text-changed::insert and text-changed::delete signals.
        (checkTextChangesAndBailOut): New. Source function to check
        the global result of the test and quit from the main loop.
        (testWebkitAtkTextChangedNotifications): New test.
        (main):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68078 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agohttps://bugs.webkit.org/show_bug.cgi?id=46298, add logicalWidth/Height getters/setter...
hyatt@apple.com [Wed, 22 Sep 2010 20:27:21 +0000 (20:27 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=46298, add logicalWidth/Height getters/setters to RenderBox.

Reviewed by Dan Bernstein.

* rendering/RenderBox.cpp:
(WebCore::RenderBox::logicalWidth):
(WebCore::RenderBox::logicalHeight):
(WebCore::RenderBox::setLogicalWidth):
(WebCore::RenderBox::setLogicalHeight):
* rendering/RenderBox.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68077 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Oliver Hunt <oliver@apple.com>
oliver@apple.com [Wed, 22 Sep 2010 20:24:23 +0000 (20:24 +0000)]
2010-09-22  Oliver Hunt  <oliver@apple.com>

        Reviewed by Gavin Barraclough.

        [INTERPRETER] Two tests fail with SputnikError: #1.1: if argArray is neither an array nor an arguments object (see 10.1.8), a TypeError exception is thrown
        https://bugs.webkit.org/show_bug.cgi?id=44245

        Remove incorrect code from op_load_varargs in the interpreter.

        * interpreter/Interpreter.cpp:
        (JSC::Interpreter::privateExecute):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68076 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoEnhance ProcessLauncher to be able to launch plug-in processes
andersca@apple.com [Wed, 22 Sep 2010 20:14:43 +0000 (20:14 +0000)]
Enhance ProcessLauncher to be able to launch plug-in processes
https://bugs.webkit.org/show_bug.cgi?id=46295

Reviewed by Darin Adler.

* UIProcess/Launcher/ProcessLauncher.cpp:
(WebKit::ProcessLauncher::ProcessLauncher):
Make the ProcessLauncher constructor take a LaunchOptions struct.

(WebKit::ProcessLauncher::processTypeAsString):
Given a process type, return its string representation.

(WebKit::ProcessLauncher::getProcessTypeFromString):
And vice versa.

* UIProcess/Launcher/mac/ProcessLauncherMac.mm:
(WebKit::ProcessLauncher::launchProcess):
Rename "mode" to "type" and pass the type as a string based on the LaunchOptions struct.
Also, set the launch architecture from the LaunchOptions struct.

* UIProcess/Launcher/win/ProcessLauncherWin.cpp:
(WebKit::ProcessLauncher::launchProcess):
Rename "mode" to "type."

* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::connect):
Initialize a LaunchOptions struct.

* WebProcess/WebKitMain.cpp:
(WebKitMain):
Factor code that can be shared between the mac and windows WebKitMain code out into a
WebKitMain overload that takes a CommandLine.

* WebProcess/WebProcessMain.h:
* WebProcess/mac/WebProcessMainMac.mm:
(WebKit::WebProcessMain):
* WebProcess/win/WebProcessMainWin.cpp:
(WebKit::WebProcessMain):
Change WebProcessMain to take a const CommandLine reference instead of a pointer.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68075 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Luiz Agostini <luiz.agostini@openbossa.org>
commit-queue@webkit.org [Wed, 22 Sep 2010 19:46:33 +0000 (19:46 +0000)]
2010-09-22  Luiz Agostini  <luiz.agostini@openbossa.org>

        Reviewed by Oliver Hunt.

        Script engine agnostic ScriptCallback class
        https://bugs.webkit.org/show_bug.cgi?id=43216

        Refactoring class ScriptFunctionCall to create a class (ScriptCallback) that
        receives a script function as a ScriptValue and calls that function with provided parameters.

        * bindings/js/ScriptFunctionCall.cpp:
        (WebCore::ScriptCallArgumentHandler::appendArgument):
        (WebCore::ScriptFunctionCall::ScriptFunctionCall):
        (WebCore::ScriptCallback::ScriptCallback):
        (WebCore::ScriptCallback::call):
        * bindings/js/ScriptFunctionCall.h:
        (WebCore::ScriptCallArgumentHandler::ScriptCallArgumentHandler):
        * bindings/v8/ScriptFunctionCall.cpp:
        (WebCore::ScriptCallArgumentHandler::appendArgument):
        (WebCore::ScriptFunctionCall::ScriptFunctionCall):
        (WebCore::ScriptCallback::ScriptCallback):
        (WebCore::ScriptCallback::call):
        * bindings/v8/ScriptFunctionCall.h:
        (WebCore::ScriptCallArgumentHandler::ScriptCallArgumentHandler):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68074 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Wed, 22 Sep 2010 19:38:45 +0000 (19:38 +0000)]
2010-09-22  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Tony Chang.

        new-run-webkit-tests: r68063 broke linux python tests

        Add a missing "from __future__ import with_statement" to this
        new file; the perils of testing only on the Mac and by review :(

        https://bugs.webkit.org/show_bug.cgi?id=46293

        * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68073 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoUnreviewed.
bfulgham@webkit.org [Wed, 22 Sep 2010 19:38:27 +0000 (19:38 +0000)]
Unreviewed.

Build fix for invalid assertion after r68059.

* editing/ApplyStyleCommand.cpp:
(WebCore::getPropertiesNotIn): Correct assertion target after
name change.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68072 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Abhishek Arya <inferno@chromium.org>
inferno@chromium.org [Wed, 22 Sep 2010 19:37:52 +0000 (19:37 +0000)]
2010-09-22  Abhishek Arya  <inferno@chromium.org>

        Unreviewed.

        Rebaseline multicol/span tests for GTK, QT platforms as part of r68062.

        * platform/gtk/fast/multicol/span/span-as-immediate-child-property-removal-expected.txt:
        * platform/gtk/fast/multicol/span/span-as-immediate-columns-child-removal-expected.txt:
        * platform/qt/fast/multicol/span/span-as-immediate-child-property-removal-expected.txt:
        * platform/qt/fast/multicol/span/span-as-immediate-columns-child-removal-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68071 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoRename GetPluginHostConnection to GetPluginPath
andersca@apple.com [Wed, 22 Sep 2010 19:37:25 +0000 (19:37 +0000)]
Rename GetPluginHostConnection to GetPluginPath
https://bugs.webkit.org/show_bug.cgi?id=46292

Reviewed by Adam Roben.

Rename GetPluginHostConnection to GetPluginPath since this message will always get the
plug-in path. We'll add another message to actually get the connection.

* Shared/CoreIPCSupport/WebProcessProxyMessageKinds.h:
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::getPluginPath):
(WebKit::WebProcessProxy::didReceiveMessage):
(WebKit::WebProcessProxy::didReceiveSyncMessage):
* UIProcess/WebProcessProxy.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::createPlugin):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68070 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agohttps://bugs.webkit.org/show_bug.cgi?id=45957, fix regression in column rules stackin...
hyatt@apple.com [Wed, 22 Sep 2010 19:35:22 +0000 (19:35 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=45957, fix regression in column rules stacking test.

Reviewed by Simon Fraser.

WebCore:

Rewrite how the pagination bit gets set in RenderLayers, since it was totally broken and not
accounting for whether or not the columns block was even in the layer's containing block
hierarchy.

* rendering/RenderLayer.cpp:
(WebCore::checkContainingBlockChainForPagination):
(WebCore::RenderLayer::updatePagination):

LayoutTests:

* platform/mac/fast/multicol/column-rules-stacking-expected.checksum:
* platform/mac/fast/multicol/column-rules-stacking-expected.png:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68069 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Kenneth Russell <kbr@google.com>
kbr@google.com [Wed, 22 Sep 2010 19:29:21 +0000 (19:29 +0000)]
2010-09-22  Kenneth Russell  <kbr@google.com>

        Unreviewed, Chromium build fix. Fix build when accelerated
        compositing is not enabled at compile time.

        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::resize):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68068 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 No'am Rosenthal <noam.rosenthal@nokia.com>
noam.rosenthal@nokia.com [Wed, 22 Sep 2010 19:18:30 +0000 (19:18 +0000)]
2010-09-22  No'am Rosenthal  <noam.rosenthal@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Move the accelerated compositing build flag to the right place
        https://bugs.webkit.org/show_bug.cgi?id=43882

        * WebKit.pri: Removed the redundant version check.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68067 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoUnreviewed.
bfulgham@webkit.org [Wed, 22 Sep 2010 19:17:54 +0000 (19:17 +0000)]
Unreviewed.

[WinCairo] Build fix for platform/cairo stuff after r67981.

* WebCore.vcproj/WebCore.vcproj: Add new files needed by all cairo
ports.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68066 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoMove plug-in creation to WebPage::createPlugin
andersca@apple.com [Wed, 22 Sep 2010 19:12:50 +0000 (19:12 +0000)]
Move plug-in creation to WebPage::createPlugin
https://bugs.webkit.org/show_bug.cgi?id=46289

Reviewed by Adam Roben.

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::createPlugin):
Call WebPage::createPlugin.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::createPlugin):
Create the plug-in.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68065 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Erik Arvidsson <arv@chromium.org>
arv@chromium.org [Wed, 22 Sep 2010 19:11:06 +0000 (19:11 +0000)]
2010-09-22  Erik Arvidsson  <arv@chromium.org>

        Reviewed by Nate Chapin.

        [Chromium] Fix memory model for dataset
        https://bugs.webkit.org/show_bug.cgi?id=46280

        This is covered by existing tests.

        * bindings/v8/custom/V8DOMStringMapCustom.cpp:
        (WebCore::toV8):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68064 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Wed, 22 Sep 2010 19:06:51 +0000 (19:06 +0000)]
2010-09-22  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Tony Chang.

        Adds platform/chromium-gpu-* directories for holding the
        GPU-accelerated baselines and expectations for Chromium.

        https://bugs.webkit.org/show_bug.cgi?id=46225

        * platform/chromium-gpu/test_expectations.txt: Added.
        * platform/chromium-gpu-mac/README.txt: Added.
        * platform/chromium-gpu-win/README.txt: Added.
        * platform/chromium-gpu-linux/README.txt: Added.
2010-09-22  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Tony Chang.

        Add a 'chromium-gpu' set of ports that will test the accelerated
        GPU paths. This patch adds:

        - support for the '--accelerated-compositing' and
          'accelerated-2d-canvas' flags to new-run-webkit-tests (and the
          'no-' flags)
        - adds a new set of Ports that will look under
          platform/chromium-gpu-$OS/ for baselines before looking in the
          regular chromium search path
        - adds a new test_expectations.txt file in platform/chromium-gpu
          that skips all but the tests we actually want to run with
          acceleration.

        This patch will allow us to run both with and without acceleration
        and to change the defaults for both the regular and -gpu options
        as the code evolves.

        We plan to add both --chromium-$OS and --chromium-gpu-$OS runs to
        each test bot.

        https://bugs.webkit.org/show_bug.cgi?id=46225

        * Scripts/webkitpy/layout_tests/port/chromium.py:
        * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
        * Scripts/webkitpy/layout_tests/port/factory.py:
        * Scripts/webkitpy/layout_tests/port/webkit.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68063 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Abhishek Arya <inferno@chromium.org>
inferno@chromium.org [Wed, 22 Sep 2010 19:03:50 +0000 (19:03 +0000)]
2010-09-22  Abhishek Arya  <inferno@chromium.org>

        Reviewed by Dave Hyatt.

        Fix wrong casts for table cell since its parent is not a RenderBlock. Remove
        the redundant destroy call for empty anonymous columns or column span block.
        https://bugs.webkit.org/show_bug.cgi?id=46245

        Tests: fast/table/table-anonymous-block-destroy-crash.html
               fast/table/table-columns-blocks-calc-crash.html

        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::addChildToAnonymousColumnBlocks):
        (WebCore::RenderBlock::columnsBlockForSpanningElement):
        (WebCore::RenderBlock::removeChild):
2010-09-22  Abhishek Arya  <inferno@chromium.org>

        Reviewed by Dave Hyatt.

        Tests that removing an anonymous block from a table cell does not result in crash.
        Also, tests that calculating columns block for a table cell does not result in crash.
        Rebaseline existing multicol span tests since we don't explicitly remove an empty
        anonymous columns or column span blocks.
        https://bugs.webkit.org/show_bug.cgi?id=46245

        * fast/table/table-anonymous-block-destroy-crash-expected.txt: Added.
        * fast/table/table-anonymous-block-destroy-crash.html: Added.
        * fast/table/table-columns-blocks-calc-crash-expected.txt: Added.
        * fast/table/table-columns-blocks-calc-crash.html: Added.
        * platform/mac/fast/multicol/span/span-as-immediate-child-property-removal-expected.checksum:
        * platform/mac/fast/multicol/span/span-as-immediate-child-property-removal-expected.png:
        * platform/mac/fast/multicol/span/span-as-immediate-child-property-removal-expected.txt:
        * platform/mac/fast/multicol/span/span-as-immediate-columns-child-removal-expected.checksum:
        * platform/mac/fast/multicol/span/span-as-immediate-columns-child-removal-expected.png:
        * platform/mac/fast/multicol/span/span-as-immediate-columns-child-removal-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68062 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Matt Perry <mpcomplete@chromium.org>
mpcomplete@chromium.org [Wed, 22 Sep 2010 18:59:06 +0000 (18:59 +0000)]
2010-09-22  Matt Perry  <mpcomplete@chromium.org>

        Reviewed by Darin Fisher.

        Trying to reland a version of r67749:
        Have V8DOMWindowShell ask the embedder whether to run a V8 extension
        in a particular script context.
        https://bugs.webkit.org/show_bug.cgi?id=45721

        * bindings/v8/V8DOMWindowShell.cpp:
        (WebCore::V8DOMWindowShell::createNewContext):
        * bindings/v8/V8Proxy.cpp:
        (WebCore::V8Proxy::registerExtension):
        * bindings/v8/V8Proxy.h:
        * loader/EmptyClients.h:
        (WebCore::EmptyFrameLoaderClient::allowScriptExtension):
        * loader/FrameLoaderClient.h:
2010-09-22  Matt Perry  <mpcomplete@chromium.org>

        Reviewed by Darin Fisher.

        Trying to reland a version of r67749:
        Have V8DOMWindowShell ask the embedder whether to run a V8 extension
        in a particular script context.
        https://bugs.webkit.org/show_bug.cgi?id=45721

        * public/WebFrameClient.h:
        (WebKit::WebFrameClient::allowScriptExtension):
        * public/WebScriptController.h:
        * src/FrameLoaderClientImpl.cpp:
        (WebKit::FrameLoaderClientImpl::allowScriptExtension):
        * src/FrameLoaderClientImpl.h:
        * src/WebScriptController.cpp:
        (WebKit::WebScriptController::registerExtension):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68061 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Oliver Hunt <oliver@apple.com>
oliver@apple.com [Wed, 22 Sep 2010 18:59:03 +0000 (18:59 +0000)]
2010-09-22  Oliver Hunt  <oliver@apple.com>

        Reviewed by Gavin Barraclough.

        [JIT] fast/js/sputnik/Conformance/15_Native_Objects/15.3_Function/15.3.5/S15.3.5.3_A2_T6.html fails
        https://bugs.webkit.org/show_bug.cgi?id=44246

        JIT code generated for instanceof was not checking to ensure that the prototype property was
        an object, this patch ensures that it does.

        * jit/JITOpcodes.cpp:
        (JSC::JIT::emit_op_instanceof):
        (JSC::JIT::emitSlow_op_instanceof):
        * jit/JITOpcodes32_64.cpp:
        (JSC::JIT::emit_op_instanceof):
        (JSC::JIT::emitSlow_op_instanceof):
2010-09-22  Oliver Hunt  <oliver@apple.com>

        Reviewed by Gavin Barraclough.

        [JIT] fast/js/sputnik/Conformance/15_Native_Objects/15.3_Function/15.3.5/S15.3.5.3_A2_T6.html fails
        https://bugs.webkit.org/show_bug.cgi?id=44246

        Correct expected output from this test.

        * fast/js/sputnik/Conformance/15_Native_Objects/15.3_Function/15.3.5/S15.3.5.3_A2_T6-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68060 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Wed, 22 Sep 2010 18:30:34 +0000 (18:30 +0000)]
2010-09-22  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Darin Adler.

        queryCommandState returns false for Underline command when no selection is made
        https://bugs.webkit.org/show_bug.cgi?id=17594

        The bug was caused by selectionComputedStyle's adding -webkit-text-decorations-in-effect property
        to a dummy span used to obtain the computed style when there is a typing style.
        Since this property is for internal use only, the CSS parser always stripped the property.
        As a result, we were always ignoring the text decorations set by the typing style.

        Fixed the bug by making selectionComputedStyle directly merge the computed style of the current
        caret position and the typing style. This also eliminates the need for a dummy span element.
        Due to the merge, the return value of selectionComputedStyle had to be changed to CSSMutableStyleDeclaration
        from CSSComputedStyleDeclaration.

        Test: editing/execCommand/query-text-decoration-with-typing-style.html

        * editing/ApplyStyleCommand.cpp:
        (WebCore::StyleChange::init): Calls getPropertiesNotIn.
        (WebCore::getPropertiesNotIn): Renamed from getPropertiesNotInComputedStyle since it takes
        CSSStyleDeclaration* instead of CSSComputedStyleDeclaration* for the second argument.
        (WebCore::ApplyStyleCommand::removeNonEditingProperties): Extracted from editingStyleAtPosition.
        (WebCore::ApplyStyleCommand::editingStyleAtPosition): Calls removeNonEditingProperties.
        * editing/ApplyStyleCommand.h:
        * editing/Editor.cpp:
        (WebCore::triStateOfStyle): Calls getPropertiesNotIn. Renamed from triStateOfStyleInComputedStyle
        since it no longer takes CSSComputedStyleDeclaration.
        (WebCore::Editor::selectionStartHasStyle): Calls selectionComputedStyle and triStateOfStyle.
        (WebCore::Editor::selectionHasStyle): Ditto.
        (WebCore::Editor::selectionStartCSSPropertyValue): Calls selectionComputedStyle.
        (WebCore::Editor::selectionComputedStyle): See above.
        * editing/Editor.h:
        * editing/EditorCommand.cpp:
        (WebCore::executeToggleStyleInList): Calls selectionComputedStyle.
2010-09-22  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Darin Adler.

        queryCommandState returns false for Underline command when no selection is made
        https://bugs.webkit.org/show_bug.cgi?id=17594

        Added a test to ensure queryCommandValue('underline') and queryCommandValue('strikeThrough')
        correctly incorporates the typing style.

        * editing/execCommand/query-text-decoration-with-typing-style-expected.txt: Added.
        * editing/execCommand/query-text-decoration-with-typing-style.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68059 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoRandom plug-in cleanups
andersca@apple.com [Wed, 22 Sep 2010 18:14:59 +0000 (18:14 +0000)]
Random plug-in cleanups
https://bugs.webkit.org/show_bug.cgi?id=46279

Reviewed by Darin Adler.

* WebProcess/Plugins/Plugin.cpp:
(WebKit::Plugin::Parameters::encode):
(WebKit::Plugin::Parameters::decode):
Add CoreIPC coding support for the Plugin::Parameters struct.

* WebProcess/Plugins/PluginController.h:
Add pluginProcessCrashed pure virtual member function.

* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::handleEvent):
handleEvent can be called when m_plugin is null.

(WebKit::PluginView::pluginProcessCrashed):
Ask the renderer to show the crashed plug-in indicator.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68058 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoDon't listen for some window notifications on all windows
andersca@apple.com [Wed, 22 Sep 2010 17:48:40 +0000 (17:48 +0000)]
Don't listen for some window notifications on all windows
https://bugs.webkit.org/show_bug.cgi?id=46277

Reviewed by John Sullivan.

Only listen for NSWindowDidBecomeKeyNotification and NSWindowDidResignKeyNotification notifications
on all windows; this matches old WebKit and fixes a crash where we would send notifications to views whose
underlying pages were invalid.

* UIProcess/API/mac/WKView.mm:
(-[WKView addWindowObserversForWindow:]):
(-[WKView removeWindowObservers]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68057 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Jamey Hicks <jamey.hicks@nokia.com>
commit-queue@webkit.org [Wed, 22 Sep 2010 17:33:56 +0000 (17:33 +0000)]
2010-09-22  Jamey Hicks  <jamey.hicks@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Web Inspector: Remote Web Inspector support for QtWebKit
        https://bugs.webkit.org/show_bug.cgi?id=43988

        Runs a web debug server on port specified by QtTestBrowser
        argument -remote-inspector-port. Property
        _q_webInspectorServerPort of the QWebPage instance will be set
        according to the argument. All pages with that property set will
        be remotely inspectable.

        URL for remote inspection of first QWebPage is
            http://localhost:9222/webkit/inspector/inspector.html?page=1
        where 1 is the number of the QWebPage instance.

        The base URL yields an index page with links to the individual inspectors:
            http://localhost:9222/

        * WebCore.pro
2010-09-22  Jamey Hicks  <jamey.hicks@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Web Inspector: Remote Web Inspector support for QtWebKit
        https://bugs.webkit.org/show_bug.cgi?id=43988

        Runs a web debug server on port specified by QtTestBrowser
        argument -remote-inspector-port. Property
        _q_webInspectorServerPort of the QWebPage instance will be set
        according to the argument. All pages with that property set will
        be remotely inspectable.

        URL for remote inspection of first QWebPage is
            http://localhost:9222/webkit/inspector/inspector.html?page=1
        where 1 is the number of the QWebPage instance.

        The base URL yields an index page with links to the individual inspectors:
            http://localhost:9222/

        * Api/qwebinspector.cpp:
        (QWebInspectorPrivate::attachAndReplaceRemoteFrontend):
        (QWebInspectorPrivate::detachRemoteFrontend):
        * Api/qwebinspector_p.h:
        (QWebInspectorPrivate::QWebInspectorPrivate):
        * Api/qwebpage.cpp:
        (QWebPagePrivate::dynamicPropertyChangeEvent):
        (QWebPagePrivate::inspectorServerPort):
        * Api/qwebpage.h:
        * Api/qwebpage_p.h:
        * WebCoreSupport/InspectorClientQt.cpp:
        (WebCore::InspectorClientQt::InspectorClientQt):
        (WebCore::InspectorClientQt::inspectorDestroyed):
        (WebCore::InspectorClientQt::openInspectorFrontend):
        (WebCore::InspectorClientQt::attachAndReplaceRemoteFrontend):
        (WebCore::InspectorClientQt::detachRemoteFrontend):
        (WebCore::InspectorClientQt::sendMessageToFrontend):
        * WebCoreSupport/InspectorClientQt.h:
        * WebCoreSupport/InspectorServerQt.cpp: Added.
        (WebCore::generateWebSocketChallengeResponse):
        (WebCore::parseWebSocketChallengeNumber):
        (WebCore::InspectorServerQt::server):
        (WebCore::InspectorServerQt::InspectorServerQt):
        (WebCore::InspectorServerQt::~InspectorServerQt):
        (WebCore::InspectorServerQt::listen):
        (WebCore::InspectorServerQt::close):
        (WebCore::InspectorServerQt::inspectorClientForPage):
        (WebCore::InspectorServerQt::registerClient):
        (WebCore::InspectorServerQt::unregisterClient):
        (WebCore::InspectorServerQt::newConnection):
        (WebCore::InspectorServerRequestHandlerQt::InspectorServerRequestHandlerQt):
        (WebCore::InspectorServerRequestHandlerQt::~InspectorServerRequestHandlerQt):
        (WebCore::InspectorServerRequestHandlerQt::tcpReadyRead):
        (WebCore::InspectorServerRequestHandlerQt::tcpConnectionDisconnected):
        (WebCore::InspectorServerRequestHandlerQt::webSocketSend):
        (WebCore::InspectorServerRequestHandlerQt::webSocketReadyRead):
        (WebCore::RemoteFrontendChannel::RemoteFrontendChannel):
        (WebCore::RemoteFrontendChannel::sendMessageToFrontend):
        * WebCoreSupport/InspectorServerQt.h: Added.
2010-09-22  Jamey Hicks  <jamey.hicks@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Web Inspector: Remote Web Inspector support for QtWebKit
        https://bugs.webkit.org/show_bug.cgi?id=43988

        Runs a web debug server on port specified by QtTestBrowser
        argument -remote-inspector-port. Property
        _q_webInspectorServerPort of the QWebPage instance will be set
        according to the argument. All pages with that property set will
        be remotely inspectable.

        URL for remote inspection of first QWebPage is
            http://localhost:9222/webkit/inspector/inspector.html?page=1
        where 1 is the number of the QWebPage instance.

        The base URL yields an index page with links to the individual inspectors:
            http://localhost:9222/

        * QtTestBrowser/launcherwindow.cpp:
        (LauncherWindow::init):
        * QtTestBrowser/launcherwindow.h:
        * QtTestBrowser/main.cpp:
        (LauncherApplication::handleUserOptions):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68056 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Wed, 22 Sep 2010 17:28:52 +0000 (17:28 +0000)]
2010-09-22  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Darin Adler.

        Inline UTF8SequenceLength
        https://bugs.webkit.org/show_bug.cgi?id=45589

        * wtf/unicode/UTF8.cpp:
        (WTF::Unicode::convertUTF8ToUTF16): Use inline version of UTF8SequenceLength to improve performance.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68055 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Darin Adler <darin@apple.com>
darin@apple.com [Wed, 22 Sep 2010 17:26:02 +0000 (17:26 +0000)]
2010-09-21  Darin Adler  <darin@apple.com>

        Reviewed by Anders Carlsson.

        Fix some Objective-C GC problems and use RetainPtr instead of HardRetain/Release
        https://bugs.webkit.org/show_bug.cgi?id=46220

        * Misc/WebNSFileManagerExtras.m:
        (setMetaData): Use CFRelease instead of HardRelease.
        (-[NSFileManager _webkit_setMetadataURL:referrer:atPath:]): Use CFStringCreateCopy
        instead of -[NSObject copy] combined with HardRetainWithNSRelease.
2010-09-21  Darin Adler  <darin@apple.com>

        Reviewed by Anders Carlsson.

        Fix some Objective-C GC problems and use RetainPtr instead of HardRetain/Release
        https://bugs.webkit.org/show_bug.cgi?id=46220

        * WebCore.exp.in: Updated.

        * bindings/objc/ObjCEventListener.h: Use a RetainPtr for m_listener, since that
        will do a CFRetain, which is right for GC rather than an -[NSObject retain],
        which is not.
        * bindings/objc/ObjCEventListener.mm:
        (WebCore::ObjCEventListener::ObjCEventListener): Removed explicit retain.
        (WebCore::ObjCEventListener::~ObjCEventListener): Removed explicit release and
        added get().
        (WebCore::ObjCEventListener::handleEvent): Added get();

        * bridge/objc/objc_instance.mm:
        (ObjcInstance::setGlobalException): Use copy/release instead of
        HardRetain/HardRelease.
        (ObjcInstance::moveGlobalExceptionToExecState): Ditto.

        * platform/Cursor.h: Use RetainPtr for PlatformCursor on Mac.

        * platform/Cursor.cpp:
        (WebCore::Cursor::platformCursor): Moved this to CursorMac for the Mac.

        * platform/Widget.cpp: Removed retain/releasePlatformWidget and made them
        both inlines in the header file.

        * platform/Widget.h: Fixed includes and added an include of RetainPtr.
        Made platformWidget and setPlatformWidget non-inline on the Mac platform.
        Made releasePlatformWidget and retainPlatformWidget empty inline functions
        for non-GTK platforms.

        * platform/graphics/mac/ImageMac.mm: Removed unneeded FoundationExtras.h
        include.

        * platform/mac/ClipboardMac.mm:
        (WebCore::cocoaTypeFromHTMLClipboardType): use RetainPtr for the return type.
        (WebCore::ClipboardMac::clearData): Use RetainPtr.
        (WebCore::ClipboardMac::getData): Ditto.
        (WebCore::ClipboardMac::setData): Ditto.

        * platform/mac/CursorMac.mm:
        (WebCore::createCustomCursor): Use RetainPtr for the return type.
        (WebCore::Cursor::ensurePlatformCursor): Removed all the calls to HardRetain
        since m_platformCursor is now a RetainPtr.
        (WebCore::Cursor::Cursor): Ditto.
        (WebCore::Cursor::operator=): Ditto. Also HardRelease.
        (WebCore::Cursor::~Cursor): Ditto.
        (WebCore::Cursor::platformCursor): Use get here.

        * platform/mac/FoundationExtras.h: Removed all the functions except for
        HardAutorelease. Later we can remove that one too.

        * platform/mac/ThemeMac.mm:
        (WebCore::setUpButtonCell): Changed this function to have a return value.
        Otherwise we end up taking a pointer (actually a reference) to a global
        variable, which does not do the right thing for GC.
        (WebCore::nonDefaultButton): Added.
        (WebCore::defaultButton): Added.
        (WebCore::button): Changed to call two separate functions with two separate
        global variables to avoid the problem with pointers to globals.

        * platform/mac/WidgetMac.mm:
        (WebCore::Widget::~Widget): Removed unneeded call to releasePlatformWidget.
        (WebCore::Widget::platformWidget): Non-inline so we can comipile the header
        plain non-Objective-C C++.
        (WebCore::Widget::setPlatformWidget): Ditto.

        * platform/network/mac/WebCoreURLResponse.mm:
        (mimeTypeFromUTITree): Use RetainPtr for the return type.
        (-[NSURLResponse adjustMIMETypeIfNecessary]): Use RetainPtr.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68054 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoFix webkit-patch failure-reason now that build.webkit.org has been updated
aroben@apple.com [Wed, 22 Sep 2010 17:18:56 +0000 (17:18 +0000)]
Fix webkit-patch failure-reason now that build.webkit.org has been updated

Fixes <http://webkit.org/b/46273> webkit-patch failure-reason crashes
every time

Reviewed by Eric Seidel.

* Scripts/webkitpy/common/net/buildbot.py:
(BuildBot._file_cell_text): Added. Travels down the firstChild chain
looking for an element that contains text, then returns it.
(BuildBot._parse_twisted_file_row): Use _file_cell_text to get the
text out of the cells. This way it doesn't matter whether the cells
have <b> children (as for cells in directory rows) or not (as for
cells in file rows)
(BuildBot._parse_twisted_directory_listing): Look for rows that have
the "directory" or "file" class, rather than rows with any class,
since header rows now have a class attribute.

* Scripts/webkitpy/common/net/buildbot_unittest.py:
(BuildBotTest._example_directory_listing): Updated to more closely
match the markup that build.webkit.org is producing now.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68053 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoCrash when running plug-in layout tests
andersca@apple.com [Wed, 22 Sep 2010 17:00:53 +0000 (17:00 +0000)]
Crash when running plug-in layout tests
https://bugs.webkit.org/show_bug.cgi?id=46269

Reviewed by Adam Roben.

Store the web page directly in the PluginView class, instead of getting it from the frame
of the plug-in element; this fixes two crashes, one where the plug-in element had been deallocated
before the plug-in view, and another where the plug-in element's document did not have a frame.

This also makes things more robust since we're asserting in the WebPage destructor that no plug-ins have
an outstanding reference to it.

* WebProcess/Plugins/PluginView.cpp:
(WebKit::webPage):
(WebKit::PluginView::PluginView):
(WebKit::PluginView::~PluginView):
(WebKit::PluginView::initializePlugin):
* WebProcess/Plugins/PluginView.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68052 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Jeremy Orlow <jorlow@chromium.org>
jorlow@chromium.org [Wed, 22 Sep 2010 16:57:24 +0000 (16:57 +0000)]
2010-09-22  Jeremy Orlow  <jorlow@chromium.org>

        Reviewed by Darin Fisher.

        [Chromium] Change WebKit API for IDBTransaciton to new style
        https://bugs.webkit.org/show_bug.cgi?id=46263

        * public/WebIDBIndex.h:
        (WebKit::WebIDBIndex::openObjectCursor):
        (WebKit::WebIDBIndex::openCursor):
        (WebKit::WebIDBIndex::getObject):
        (WebKit::WebIDBIndex::get):
        * public/WebIDBObjectStore.h:
        (WebKit::WebIDBObjectStore::get):
        (WebKit::WebIDBObjectStore::put):
        (WebKit::WebIDBObjectStore::remove):
        (WebKit::WebIDBObjectStore::openCursor):
        * src/WebIDBIndexImpl.cpp:
        (WebKit::WebIDBIndexImpl::openCursor):
        (WebKit::WebIDBIndexImpl::openObjectCursor):
        (WebKit::WebIDBIndexImpl::getObject):
        (WebKit::WebIDBIndexImpl::get):
        * src/WebIDBIndexImpl.h:
        * src/WebIDBObjectStoreImpl.cpp:
        (WebKit::WebIDBObjectStoreImpl::get):
        (WebKit::WebIDBObjectStoreImpl::put):
        (WebKit::WebIDBObjectStoreImpl::remove):
        (WebKit::WebIDBObjectStoreImpl::openCursor):
        * src/WebIDBObjectStoreImpl.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68051 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Wed, 22 Sep 2010 16:25:25 +0000 (16:25 +0000)]
2010-09-22  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] r68027 caused many crashes on the bots
        https://bugs.webkit.org/show_bug.cgi?id=46268

        Add the appropriate OwnPtrCairo.h include which has the OwnPtr
        specialization for Cairo paths. Without it, we will have memory
        corruption.

        * platform/graphics/cairo/PathCairo.cpp: Add OwnPtrCairo.h include.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68050 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Pavel Podivilov <podivilov@chromium.org>
podivilov@chromium.org [Wed, 22 Sep 2010 16:06:24 +0000 (16:06 +0000)]
2010-09-22  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: implement pausing on XHR
        https://bugs.webkit.org/show_bug.cgi?id=46086

        * bindings/v8/custom/V8XMLHttpRequestCustom.cpp:
        (WebCore::V8XMLHttpRequest::sendCallback):
        * inspector/Inspector.idl:
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::InspectorController):
        (WebCore::InspectorController::setNativeBreakpoint):
        (WebCore::InspectorController::removeNativeBreakpoint):
        (WebCore::InspectorController::willInsertDOMNodeImpl):
        (WebCore::InspectorController::willRemoveDOMNodeImpl):
        (WebCore::InspectorController::willModifyDOMAttrImpl):
        (WebCore::InspectorController::instrumentWillSendXMLHttpRequestImpl):
        * inspector/InspectorController.h:
        (WebCore::InspectorController::instrumentWillSendXMLHttpRequest):
        (WebCore::InspectorController::inspectorControllerForScriptExecutionContext):
        (WebCore::InspectorController::inspectorControllerForNode):
        (WebCore::InspectorController::inspectorControllerForDocument):
        * inspector/InspectorDebuggerAgent.cpp:
        (WebCore::InspectorDebuggerAgent::InspectorDebuggerAgent):
        (WebCore::InspectorDebuggerAgent::didPause):
        (WebCore::InspectorDebuggerAgent::breakProgram):
        * inspector/InspectorDebuggerAgent.h:
        * inspector/InspectorTimelineAgent.h:
        * inspector/front-end/CallStackSidebarPane.js:
        (WebInspector.CallStackSidebarPane.prototype.updateStatus):
        * inspector/front-end/ElementsPanel.js:
        (WebInspector.ElementsPanel):
        (WebInspector.ElementsPanel.prototype.reset):
        * inspector/front-end/ElementsTreeOutline.js:
        (WebInspector.ElementsTreeElement.prototype._populateTagContextMenu):
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel):
        (WebInspector.ScriptsPanel.prototype.debuggerPaused):
        (WebInspector.ScriptsPanel.prototype.reset):
        * inspector/front-end/Settings.js:
        * xml/XMLHttpRequest.h:
        (WebCore::XMLHttpRequest::url):
2010-09-22  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: implement pausing on XHR
        https://bugs.webkit.org/show_bug.cgi?id=46086

        * src/js/DevTools.js:
        ():

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68049 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoChange a bunch of Vector<char> to Vector<uint8_t> to better indicate that we're
andersca@apple.com [Wed, 22 Sep 2010 15:57:35 +0000 (15:57 +0000)]
Change a bunch of Vector<char> to Vector<uint8_t> to better indicate that we're
dealing with raw bytes.

Reviewed by Adam Roben.

* WebProcess/Plugins/Netscape/NetscapeBrowserFuncs.cpp:
(WebKit::parsePostBuffer):
(WebKit::NPN_GetURL):
(WebKit::NPN_PostURL):
(WebKit::NPN_GetURLNotify):
(WebKit::NPN_PostURLNotify):
* WebProcess/Plugins/Netscape/NetscapePlugin.cpp:
(WebKit::NetscapePlugin::loadURL):
(WebKit::NetscapePlugin::initialize):
* WebProcess/Plugins/Netscape/NetscapePlugin.h:
* WebProcess/Plugins/Netscape/NetscapePluginStream.cpp:
(WebKit::NetscapePluginStream::deliverData):
(WebKit::NetscapePluginStream::deliverDataToPlugin):
* WebProcess/Plugins/Netscape/NetscapePluginStream.h:
* WebProcess/Plugins/PluginController.h:
* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::loadURL):
* WebProcess/Plugins/PluginView.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68048 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Brett Wilson <brettw@chromium.org>
brettw@chromium.org [Wed, 22 Sep 2010 15:46:22 +0000 (15:46 +0000)]
2010-09-21  Brett Wilson  <brettw@chromium.org>

        Reviewed by Darin Fisher.

        Add caps log and num lock toggle state to WebInputEvent.
        https://bugs.webkit.org/show_bug.cgi?id=46229

        * public/WebInputEvent.h:
        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::handleInputEvent):
        * src/WebViewImpl.h:
        * src/gtk/WebInputEventFactory.cpp:
        (WebKit::gdkStateToWebEventModifiers):
        * src/win/WebInputEventFactory.cpp:
        (WebKit::SetToggleKeyState):
        (WebKit::WebInputEventFactory::keyboardEvent):
        (WebKit::WebInputEventFactory::mouseEvent):
        (WebKit::WebInputEventFactory::mouseWheelEvent):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68047 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoAssertion fails in [WKView _updateWindowFrame] when closing a WebKit2 window
andersca@apple.com [Wed, 22 Sep 2010 15:37:27 +0000 (15:37 +0000)]
Assertion fails in [WKView _updateWindowFrame] when closing a WebKit2 window
https://bugs.webkit.org/show_bug.cgi?id=46264
<rdar://problem/8463534>

Reviewed by Adam Roben.

* UIProcess/API/mac/WKView.mm:
(-[WKView viewDidMoveToWindow]):
Only update window visibility and frame when we're moving to a window.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68046 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Kenneth Russell <kbr@google.com>
kbr@google.com [Wed, 22 Sep 2010 15:31:10 +0000 (15:31 +0000)]
2010-09-22  Kenneth Russell  <kbr@google.com>

        Reviewed by James Robinson.

        Add local triangulation of cubic curve control points
        https://bugs.webkit.org/show_bug.cgi?id=45252

        Adding a localized triangulation algorithm which takes as input
        the four control points of a cubic curve segment and provides both
        triangles as well as the ability to walk the interior edges. The
        latter will be used later to fill the interior of shapes bounded
        by these cubic curves, quadratic curves and line segments.

        * platform/graphics/gpu/LoopBlinnLocalTriangulator.cpp: Added.
        (WebCore::LoopBlinnLocalTriangulator::Triangle::contains):
        (WebCore::LoopBlinnLocalTriangulator::Triangle::nextVertex):
        (WebCore::LoopBlinnLocalTriangulator::Triangle::indexForVertex):
        (WebCore::LoopBlinnLocalTriangulator::Triangle::makeCounterClockwise):
        (WebCore::LoopBlinnLocalTriangulator::LoopBlinnLocalTriangulator):
        (WebCore::LoopBlinnLocalTriangulator::reset):
        (WebCore::LoopBlinnLocalTriangulator::triangulate):
        (WebCore::LoopBlinnLocalTriangulator::triangulateHelper):
        (WebCore::LoopBlinnLocalTriangulator::addTriangle):
        (WebCore::LoopBlinnLocalTriangulator::addInteriorVertex):
        (WebCore::LoopBlinnLocalTriangulator::isSharedEdge):
        * platform/graphics/gpu/LoopBlinnLocalTriangulator.h: Added.
        (WebCore::LoopBlinnLocalTriangulator::Vertex::Vertex):
        (WebCore::LoopBlinnLocalTriangulator::Vertex::xyCoordinates):
        (WebCore::LoopBlinnLocalTriangulator::Vertex::klmCoordinates):
        (WebCore::LoopBlinnLocalTriangulator::Vertex::set):
        (WebCore::LoopBlinnLocalTriangulator::Vertex::end):
        (WebCore::LoopBlinnLocalTriangulator::Vertex::setEnd):
        (WebCore::LoopBlinnLocalTriangulator::Vertex::marked):
        (WebCore::LoopBlinnLocalTriangulator::Vertex::setMarked):
        (WebCore::LoopBlinnLocalTriangulator::Vertex::interior):
        (WebCore::LoopBlinnLocalTriangulator::Vertex::setInterior):
        (WebCore::LoopBlinnLocalTriangulator::Vertex::resetFlags):
        (WebCore::LoopBlinnLocalTriangulator::Triangle::Triangle):
        (WebCore::LoopBlinnLocalTriangulator::Triangle::getVertex):
        (WebCore::LoopBlinnLocalTriangulator::Triangle::setVertices):
        (WebCore::LoopBlinnLocalTriangulator::getVertex):
        (WebCore::LoopBlinnLocalTriangulator::numberOfTriangles):
        (WebCore::LoopBlinnLocalTriangulator::getTriangle):
        (WebCore::LoopBlinnLocalTriangulator::numberOfInteriorVertices):
        (WebCore::LoopBlinnLocalTriangulator::getInteriorVertex):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68045 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Lucas De Marchi <lucas.demarchi@profusion.mobi>
demarchi@webkit.org [Wed, 22 Sep 2010 15:27:03 +0000 (15:27 +0000)]
2010-09-22  Lucas De Marchi  <lucas.demarchi@profusion.mobi>

        Unreviewed, EFL build fix.

        * CMakeListsEfl.txt:
        * platform/efl/ClipboardEfl.cpp:
        (WebCore::ClipboardEfl::ClipboardEfl):
2010-09-22  Lucas De Marchi  <lucas.demarchi@profusion.mobi>

        Unreviewed, EFL build fix.

        * CMakeLists.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68044 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Wed, 22 Sep 2010 15:18:36 +0000 (15:18 +0000)]
2010-09-22  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        Added information about drop shadow improvements to the NEWS file.

        * NEWS: Added information about drop shadow improvements.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68043 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 İsmail Dönmez <ismail@namtrac.org>
andreas.kling@nokia.com [Wed, 22 Sep 2010 15:03:24 +0000 (15:03 +0000)]
2010-09-22  İsmail Dönmez  <ismail@namtrac.org>

        Reviewed by Andreas Kling.

        [Qt] Build fix for WinCE

        Add two missing stubs:
        - privateBrowsingStateChanged(bool)
        - bindingInstance()

        * platform/qt/TemporaryLinkStubsQt.cpp:
        (PluginView::privateBrowsingStateChanged):
        (PluginView::bindingInstance):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68042 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Wed, 22 Sep 2010 15:01:22 +0000 (15:01 +0000)]
2010-09-22  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] FontCustomPlatformData.cpp leaks FT_Faces
        https://bugs.webkit.org/show_bug.cgi?id=16941

        Attach lifetime of the FontCustomPlatformData FT_Face to the resulting cairo_font_face_t.
        This will free the memory associated with the FT_Face once the cairo_font_face_t is
        destroyed.

        No new tests as functionality should not change.

        * platform/graphics/cairo/FontCustomPlatformData.cpp:
        (WebCore::releaseCustomFontData): Moved this helper method to the top of the file
        and gave it a more descriptive name.
        (WebCore::FontCustomPlatformData::FontCustomPlatformData): Changed the constructor to
        take the FT_Face and the buffer, so to better encapsulate the creation of the cairo
        face.
        (WebCore::FontCustomPlatformData::~FontCustomPlatformData): Added a comment about the
        fate of the FT_Face to the destructor.
        (WebCore::createFontCustomPlatformData): Cleaned up the code here and moved some of the
        logic into the constructor.
        * platform/graphics/cairo/FontCustomPlatformData.h: Updated method definitions.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68041 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Wed, 22 Sep 2010 14:40:34 +0000 (14:40 +0000)]
2010-09-21  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Fix transfer annotations for WebKitGTK+ static API
        https://bugs.webkit.org/show_bug.cgi?id=46244

        Deal with all the warnings related to transfer issues. Mostly the
        problem was the transfer being missing, but there were also typos
        in the function name or missing ':' at the end of the function
        name in the gtk-doc.

        * webkit/webkitdownload.cpp:
        * webkit/webkitnetworkrequest.cpp:
        * webkit/webkitnetworkresponse.cpp:
        * webkit/webkitsecurityorigin.cpp:
        * webkit/webkitwebdatabase.cpp:
        * webkit/webkitwebdatasource.cpp:
        * webkit/webkitwebframe.cpp:
        * webkit/webkitwebhistoryitem.cpp:
        * webkit/webkitwebinspector.cpp:
        * webkit/webkitwebresource.cpp:
        * webkit/webkitwebsettings.cpp:
        * webkit/webkitwebview.cpp:
        (webkit_web_view_class_init):
        (webkit_web_view_get_main_frame):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68040 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Balazs Kelemen <kb@inf.u-szeged.hu>
andreas.kling@nokia.com [Wed, 22 Sep 2010 14:25:51 +0000 (14:25 +0000)]
2010-09-22  Balazs Kelemen  <kb@inf.u-szeged.hu>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] PluginStrategy implementation is broken
        https://bugs.webkit.org/show_bug.cgi?id=46078

        Remove the QWebPage* member from WebPlatformStrategies.
        Get the originator QWebPage of the getPluginInfo call via
        the Page* argument.
        * Api/qwebpage.cpp:
        (QWebPagePrivate::QWebPagePrivate):
        * WebCoreSupport/WebPlatformStrategies.cpp:
        (WebPlatformStrategies::initialize):
        (WebPlatformStrategies::WebPlatformStrategies):
        (WebPlatformStrategies::getPluginInfo):
        Use the ChromeClient for accessing the originator QWebPage.
        * WebCoreSupport/WebPlatformStrategies.h:
2010-09-22  Balazs Kelemen  <kb@inf.u-szeged.hu>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] PluginStrategy implementation is broken
        https://bugs.webkit.org/show_bug.cgi?id=46078

        * UIProcess/API/qt/qwkpage.cpp:
        (QWKPagePrivate::QWKPagePrivate):
        Initialize the WebPlatformStrategies at the UI side as well because
        we are using the LocalizationStrategy in the UI process.
        Added a FIXME since this should be fixed in the future.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68039 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Balazs Kelemen <kb@inf.u-szeged.hu>
andreas.kling@nokia.com [Wed, 22 Sep 2010 14:17:55 +0000 (14:17 +0000)]
2010-09-22  Balazs Kelemen  <kb@inf.u-szeged.hu>

        Reviewed by Kenneth Rohde Christiansen.

        PluginStrategy should satisfy the needs of Qt
        https://bugs.webkit.org/show_bug.cgi?id=45857
        No new functionality so no new tests.

        * WebCoreSupport/WebPlatformStrategies.cpp:
        (WebPlatformStrategies::getPluginInfo):
        * WebCoreSupport/WebPlatformStrategies.h:
2010-09-22  Balazs Kelemen  <kb@inf.u-szeged.hu>

        Reviewed by Kenneth Rohde Christiansen.

        PluginStrategy should satisfy the needs of Qt
        https://bugs.webkit.org/show_bug.cgi?id=45857
        No new functionality so no new tests.

        * WebCoreSupport/WebPlatformStrategies.h:
        * WebCoreSupport/WebPlatformStrategies.mm:
        (WebPlatformStrategies::getPluginInfo):
2010-09-22  Balazs Kelemen  <kb@inf.u-szeged.hu>

        Reviewed by Kenneth Rohde Christiansen.

        PluginStrategy should satisfy the needs of Qt
        https://bugs.webkit.org/show_bug.cgi?id=45857
        No new functionality so no new tests.

        * plugins/PluginData.cpp:
        (WebCore::PluginData::initPlugins):
        * plugins/PluginStrategy.h: Added a |const Page*| argument to getPluginInfo.
2010-09-22  Balazs Kelemen  <kb@inf.u-szeged.hu>

        Reviewed by Kenneth Rohde Christiansen.

        PluginStrategy should satisfy the needs of Qt
        https://bugs.webkit.org/show_bug.cgi?id=45857
        No new functionality so no new tests.

        * WebProcess/WebCoreSupport/WebPlatformStrategies.cpp:
        (WebKit::WebPlatformStrategies::getPluginInfo):
        * WebProcess/WebCoreSupport/WebPlatformStrategies.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68038 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 İsmail Dönmez <ismail@namtrac.org>
andreas.kling@nokia.com [Wed, 22 Sep 2010 13:53:46 +0000 (13:53 +0000)]
2010-09-22  İsmail Dönmez  <ismail@namtrac.org>

        Reviewed by Andreas Kling.

        [Qt] Buildfix for WinCE

        * Api/DerivedSources.pro:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68037 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Wed, 22 Sep 2010 13:02:58 +0000 (13:02 +0000)]
2010-09-22  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: Elements panel does not reflect changes to the text node values.
        https://bugs.webkit.org/show_bug.cgi?id=46166

        * dom/CharacterData.cpp:
        (WebCore::CharacterData::dispatchModifiedEvent):
        * inspector/Inspector.idl:
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::characterDataModifiedImpl):
        * inspector/InspectorController.h:
        (WebCore::InspectorController::characterDataModified):
        * inspector/InspectorDOMAgent.cpp:
        (WebCore::InspectorDOMAgent::characterDataModified):
        * inspector/InspectorDOMAgent.h:
        (WebCore::EventListenerInfo::EventListenerInfo):
        (WebCore::InspectorDOMAgent::create):
        (WebCore::InspectorDOMAgent::cast):
        (WebCore::InspectorDOMAgent::cssStore):
        * inspector/front-end/DOMAgent.js:
        (WebInspector.DOMAgent.prototype._characterDataModified):
        (WebInspector.characterDataModified):
        * inspector/front-end/ElementsPanel.js:
        (WebInspector.ElementsPanel.prototype._characterDataModified):
        * inspector/front-end/ElementsTreeOutline.js:
        (WebInspector.ElementsTreeElement.prototype._textNodeEditingCommitted):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68036 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago[Qt] On Maemo platform, web style is missing for some inputs
diegohcg@webkit.org [Wed, 22 Sep 2010 12:10:25 +0000 (12:10 +0000)]
[Qt] On Maemo platform, web style is missing for some inputs
https://bugs.webkit.org/show_bug.cgi?id=46182

Patch by Ragner Magalhaes <ragner.magalhaes@openbossa.org> on 2010-09-21
Reviewed by Kenneth Rohde Christiansen.

* css/themeQtMaemo5.css:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68035 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Wed, 22 Sep 2010 10:26:48 +0000 (10:26 +0000)]
2010-09-22  Adam Barth  <abarth@webkit.org>

        Reviewed by Alexey Proskuryakov.

        REGRESSION: HTMLElement.localName returns lowercase, breaks Jira
        https://bugs.webkit.org/show_bug.cgi?id=14114

        Add tests from bug.

        * fast/parser/localname-case-expected.txt: Added.
        * fast/parser/localname-case-strict-expected.txt: Added.
        * fast/parser/localname-case-strict.html: Added.
        * fast/parser/localname-case.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68034 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Wed, 22 Sep 2010 10:19:50 +0000 (10:19 +0000)]
2010-09-22  Patrick Gansterer  <paroga@webkit.org>

        Unreviewed.

        [WINCE] Build fix for ResourceHandleWin after r67291.

        * platform/network/win/ResourceHandleWin.cpp:
        (WebCore::ResourceHandle::loadResourceSynchronously): Replace Frame* parameter with NetworkingContext*.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68033 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Balazs Kelemen <kbalazs@webkit.org>
kbalazs@webkit.org [Wed, 22 Sep 2010 09:34:50 +0000 (09:34 +0000)]
2010-09-22  Balazs Kelemen  <kbalazs@webkit.org>

        Unreviewed HTML syntax fix for my recent change in the coding-style.

        * coding/coding-style.html: Change "&gt" and "&lt" to "&gt;" and "&lt;".

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68032 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Chris Guillory <chris.guillory@google.com>
commit-queue@webkit.org [Wed, 22 Sep 2010 08:58:55 +0000 (08:58 +0000)]
2010-09-22  Chris Guillory   <chris.guillory@google.com>

        Reviewed by Chris Fleizach.

        Crash due to bad cast in AXObjectCacheChromium.cpp.
        https://bugs.webkit.org/show_bug.cgi?id=46237

        * accessibility/chromium/AXObjectCacheChromium.cpp:
        (WebCore::AXObjectCache::postPlatformNotification):
        * page/ChromeClient.h:
        (WebCore::ChromeClient::postAccessibilityNotification):
        (WebCore::ChromeClient::didChangeAccessibilityObjectState):
        (WebCore::ChromeClient::didChangeAccessibilityObjectChildren):
        * page/chromium/ChromeClientChromium.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68031 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Balazs Kelemen <kb@inf.u-szeged.hu>
kbalazs@webkit.org [Wed, 22 Sep 2010 08:55:17 +0000 (08:55 +0000)]
2010-09-22  Balazs Kelemen  <kb@inf.u-szeged.hu>

        Reviewed by Darin Adler.

        check-webkit-style and the coding style guidelines page are inconsistent
        https://bugs.webkit.org/show_bug.cgi?id=46099

        * coding/coding-style.html:
        Making clear that includes of system headers must come after includes of other headers.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68030 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Chris Rogers <crogers@google.com>
crogers@google.com [Wed, 22 Sep 2010 08:45:48 +0000 (08:45 +0000)]
2010-09-22  Chris Rogers  <crogers@google.com>

        Reviewed by James Robinson.

        Add HRTFElevation files
        https://bugs.webkit.org/show_bug.cgi?id=45864

        No new tests since audio API is not yet implemented.

        * platform/audio/HRTFElevation.cpp: Added.
        (WebCore::HRTFElevation::calculateSymmetricKernelsForAzimuthElevation):
        (WebCore::HRTFElevation::calculateKernelsForAzimuthElevation):
        (WebCore::HRTFElevation::createForSubject):
        (WebCore::HRTFElevation::createByInterpolatingSlices):
        (WebCore::HRTFElevation::getKernelsFromAzimuth):
        * platform/audio/HRTFElevation.h: Added.
        (WebCore::HRTFElevation::kernelListL):
        (WebCore::HRTFElevation::kernelListR):
        (WebCore::HRTFElevation::elevationAngle):
        (WebCore::HRTFElevation::numberOfAzimuths):
        (WebCore::HRTFElevation::sampleRate):
        (WebCore::HRTFElevation::HRTFElevation):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68029 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Paul Knight <pknight@apple.com>
commit-queue@webkit.org [Wed, 22 Sep 2010 08:37:01 +0000 (08:37 +0000)]
2010-09-22  Paul Knight  <pknight@apple.com>

        Reviewed by Simon Fraser.

        -[WebView _scheduleCompositingLayerSync] should wake the run loop
        https://bugs.webkit.org/show_bug.cgi?id=46226

        Call CFRunLoopWakeUp to make sure the run loop is not sleeping, which could delay layer painting.

        * WebView/WebView.mm:
        (-[WebView _scheduleCompositingLayerSync]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68028 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Alejandro G. Castro <alex@igalia.com>
alex@webkit.org [Wed, 22 Sep 2010 08:23:20 +0000 (08:23 +0000)]
2010-09-22  Alejandro G. Castro  <alex@igalia.com>

        Reviewed by Martin Robinson.

        [gtk] Use the smart pointers to handle cairo_path_t
        https://bugs.webkit.org/show_bug.cgi?id=46212

        Modified the code to use smart pointers with cairo_path_t
        variables.

        * platform/graphics/cairo/GraphicsContextCairo.cpp:
        (WebCore::appendPathToCairoContext):
        (WebCore::GraphicsContext::clip):
        (WebCore::GraphicsContext::drawTiledShadow):
        * platform/graphics/cairo/PathCairo.cpp:
        (WebCore::Path::Path):
        (WebCore::Path::operator=):
        (WebCore::Path::apply):
        (WebCore::Path::debugString):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68027 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Wed, 22 Sep 2010 08:11:51 +0000 (08:11 +0000)]
2010-09-22  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Kent Tamura.

        [BREWMP] Implement Image::loadPlatformResource
        https://bugs.webkit.org/show_bug.cgi?id=45873

        Load images from the resource directory.

        * platform/graphics/brew/ImageBrew.cpp: Added.
        (WebCore::Image::loadPlatformResource):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68026 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Chris Rogers <crogers@google.com>
crogers@google.com [Wed, 22 Sep 2010 08:10:47 +0000 (08:10 +0000)]
2010-09-22  Chris Rogers  <crogers@google.com>

        Reviewed by James Robinson.

        Add HRTFKernel files
        https://bugs.webkit.org/show_bug.cgi?id=45863

        No new tests since audio API is not yet implemented.

        * platform/audio/HRTFKernel.cpp: Added.
        (WebCore::extractAverageGroupDelay):
        (WebCore::HRTFKernel::HRTFKernel):
        (WebCore::HRTFKernel::createImpulseResponse):
        (WebCore::HRTFKernel::createInterpolatedKernel):
        * platform/audio/HRTFKernel.h: Added.
        (WebCore::HRTFKernel::create):
        (WebCore::HRTFKernel::fftFrame):
        (WebCore::HRTFKernel::fftSize):
        (WebCore::HRTFKernel::frameDelay):
        (WebCore::HRTFKernel::sampleRate):
        (WebCore::HRTFKernel::nyquist):
        (WebCore::HRTFKernel::HRTFKernel):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68025 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Wed, 22 Sep 2010 08:09:43 +0000 (08:09 +0000)]
2010-09-22  Kent Tamura  <tkent@chromium.org>

        Reviewed by Chris Fleizach.

        Support keyboard operations for <input type=range>.
        https://bugs.webkit.org/show_bug.cgi?id=45803

        * fast/forms/range-keyoperation-expected.txt: Added.
        * fast/forms/range-keyoperation.html: Added.
2010-09-22  Kent Tamura  <tkent@chromium.org>

        Reviewed by Chris Fleizach.

        Support keyboard operations for <input type=range>.
        https://bugs.webkit.org/show_bug.cgi?id=45803

        Increasing the value with Up/Right arrow keys by its step value,
        and decreasing with Down/Left arrow keys. If an input element has
        step=any attribute, increasing/decreasing by 1/100 of max-min.

        Note: This change is not useful on Mac because users can't set
        focus on range controls.

        Test: fast/forms/range-keyoperation.html

        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::defaultEventHandler):
          Calls handleKeyEventForRange() for RANGE and key events.
        (WebCore::HTMLInputElement::handleKeyEventForRange):
        * html/HTMLInputElement.h: Add handleKeyEventForRange() declaration.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68024 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Mario Sanchez Prada <msanchez@igalia.com>
commit-queue@webkit.org [Wed, 22 Sep 2010 07:49:34 +0000 (07:49 +0000)]
2010-09-22  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Chris Fleizach.

        [Gtk] Incorrect exposure of list items whose children are elements
        https://bugs.webkit.org/show_bug.cgi?id=45383

        Check that list items will be always exposed like that for GTK

        * platform/gtk/accessibility/list-items-always-exposed-expected.txt: Added.
        * platform/gtk/accessibility/list-items-always-exposed.html: Added.
2010-09-22  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Chris Fleizach.

        [Gtk] Incorrect exposure of list items whose children are elements
        https://bugs.webkit.org/show_bug.cgi?id=45383

        Set ATK_ROLE_LIST_ITEM looking for the role of the parent object

        Test: platform/gtk/accessibility/list-items-always-exposed.html

        * accessibility/gtk/AccessibilityObjectAtk.cpp:
        (WebCore::AccessibilityObject::accessibilityPlatformIncludesObject):
        Always include objects with ListItemRole in accessibility for GTK.
        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (getInterfaceMaskFromObject): Add some specific code to handle the
        special case of list markers by implementing AtkText when needed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68023 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Dirk Schulze <krit@webkit.org>
krit@webkit.org [Wed, 22 Sep 2010 07:42:12 +0000 (07:42 +0000)]
2010-09-22  Dirk Schulze  <krit@webkit.org>

        Reviewed by Nikolas Zimmermann.

        Make FilterEffects independent of SVG
        https://bugs.webkit.org/show_bug.cgi?id=46177

        Moving all FilterEffects from svg/graphics/filters to platform/graphics/filters, now that they
        are independent of SVG. Removed the SVG prefix of the files and updated the headers.

        * Android.mk:
        * CMakeLists.txt:
        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * platform/graphics/filters/DistantLightSource.h: Copied from WebCore/svg/graphics/filters/SVGDistantLightSource.h.
        * platform/graphics/filters/FEConvolveMatrix.cpp: Copied from WebCore/svg/graphics/filters/SVGFEConvolveMatrix.cpp.
        * platform/graphics/filters/FEConvolveMatrix.h: Copied from WebCore/svg/graphics/filters/SVGFEConvolveMatrix.h.
        * platform/graphics/filters/FEDiffuseLighting.cpp: Copied from WebCore/svg/graphics/filters/SVGFEDiffuseLighting.cpp.
        * platform/graphics/filters/FEDiffuseLighting.h: Copied from WebCore/svg/graphics/filters/SVGFEDiffuseLighting.h.
        * platform/graphics/filters/FEDisplacementMap.cpp: Copied from WebCore/svg/graphics/filters/SVGFEDisplacementMap.cpp.
        * platform/graphics/filters/FEDisplacementMap.h: Copied from WebCore/svg/graphics/filters/SVGFEDisplacementMap.h.
        * platform/graphics/filters/FEFlood.cpp: Copied from WebCore/svg/graphics/filters/SVGFEFlood.cpp.
        * platform/graphics/filters/FEFlood.h: Copied from WebCore/svg/graphics/filters/SVGFEFlood.h.
        * platform/graphics/filters/FELighting.cpp: Copied from WebCore/svg/graphics/filters/SVGFELighting.cpp.
        * platform/graphics/filters/FELighting.h: Copied from WebCore/svg/graphics/filters/SVGFELighting.h.
        * platform/graphics/filters/FEMerge.cpp: Copied from WebCore/svg/graphics/filters/SVGFEMerge.cpp.
        * platform/graphics/filters/FEMerge.h: Copied from WebCore/svg/graphics/filters/SVGFEMerge.h.
        * platform/graphics/filters/FEMorphology.cpp: Copied from WebCore/svg/graphics/filters/SVGFEMorphology.cpp.
        * platform/graphics/filters/FEMorphology.h: Copied from WebCore/svg/graphics/filters/SVGFEMorphology.h.
        * platform/graphics/filters/FEOffset.cpp: Copied from WebCore/svg/graphics/filters/SVGFEOffset.cpp.
        * platform/graphics/filters/FEOffset.h: Copied from WebCore/svg/graphics/filters/SVGFEOffset.h.
        * platform/graphics/filters/FESpecularLighting.cpp: Copied from WebCore/svg/graphics/filters/SVGFESpecularLighting.cpp.
        * platform/graphics/filters/FESpecularLighting.h: Copied from WebCore/svg/graphics/filters/SVGFESpecularLighting.h.
        * platform/graphics/filters/FETile.cpp: Copied from WebCore/svg/graphics/filters/SVGFETile.cpp.
        * platform/graphics/filters/FETile.h: Copied from WebCore/svg/graphics/filters/SVGFETile.h.
        * platform/graphics/filters/FETurbulence.cpp: Copied from WebCore/svg/graphics/filters/SVGFETurbulence.cpp.
        * platform/graphics/filters/FETurbulence.h: Copied from WebCore/svg/graphics/filters/SVGFETurbulence.h.
        * platform/graphics/filters/LightSource.cpp: Copied from WebCore/svg/graphics/filters/SVGLightSource.cpp.
        * platform/graphics/filters/LightSource.h: Copied from WebCore/svg/graphics/filters/SVGLightSource.h.
        * platform/graphics/filters/PointLightSource.h: Copied from WebCore/svg/graphics/filters/SVGPointLightSource.h.
        * platform/graphics/filters/SpotLightSource.h: Copied from WebCore/svg/graphics/filters/SVGSpotLightSource.h.
        * svg/SVGFEConvolveMatrixElement.h:
        * svg/SVGFEDiffuseLightingElement.cpp:
        * svg/SVGFEDisplacementMapElement.h:
        * svg/SVGFEDistantLightElement.cpp:
        * svg/SVGFEFloodElement.h:
        * svg/SVGFELightElement.h:
        * svg/SVGFEMergeElement.h:
        * svg/SVGFEMorphologyElement.h:
        * svg/SVGFEOffsetElement.h:
        * svg/SVGFEPointLightElement.cpp:
        * svg/SVGFESpecularLightingElement.h:
        * svg/SVGFESpotLightElement.cpp:
        * svg/SVGFETileElement.h:
        * svg/SVGFETurbulenceElement.h:
        * svg/graphics/filters/SVGDistantLightSource.h: Removed.
        * svg/graphics/filters/SVGFEConvolveMatrix.cpp: Removed.
        * svg/graphics/filters/SVGFEConvolveMatrix.h: Removed.
        * svg/graphics/filters/SVGFEDiffuseLighting.cpp: Removed.
        * svg/graphics/filters/SVGFEDiffuseLighting.h: Removed.
        * svg/graphics/filters/SVGFEDisplacementMap.cpp: Removed.
        * svg/graphics/filters/SVGFEDisplacementMap.h: Removed.
        * svg/graphics/filters/SVGFEFlood.cpp: Removed.
        * svg/graphics/filters/SVGFEFlood.h: Removed.
        * svg/graphics/filters/SVGFELighting.cpp: Removed.
        * svg/graphics/filters/SVGFELighting.h: Removed.
        * svg/graphics/filters/SVGFEMerge.cpp: Removed.
        * svg/graphics/filters/SVGFEMerge.h: Removed.
        * svg/graphics/filters/SVGFEMorphology.cpp: Removed.
        * svg/graphics/filters/SVGFEMorphology.h: Removed.
        * svg/graphics/filters/SVGFEOffset.cpp: Removed.
        * svg/graphics/filters/SVGFEOffset.h: Removed.
        * svg/graphics/filters/SVGFESpecularLighting.cpp: Removed.
        * svg/graphics/filters/SVGFESpecularLighting.h: Removed.
        * svg/graphics/filters/SVGFETile.cpp: Removed.
        * svg/graphics/filters/SVGFETile.h: Removed.
        * svg/graphics/filters/SVGFETurbulence.cpp: Removed.
        * svg/graphics/filters/SVGFETurbulence.h: Removed.
        * svg/graphics/filters/SVGLightSource.cpp: Removed.
        * svg/graphics/filters/SVGLightSource.h: Removed.
        * svg/graphics/filters/SVGPointLightSource.h: Removed.
        * svg/graphics/filters/SVGSpotLightSource.h: Removed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68022 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Michael Saboff <msaboff@apple.com>
msaboff@apple.com [Wed, 22 Sep 2010 07:40:40 +0000 (07:40 +0000)]
2010-09-22  Michael Saboff  <msaboff@apple.com>

        Reviewed by Darin Adler.

        New test for ^ (BOL) anchored regular expressions.  Some of the
        tests inspired by https://bugs.webkit.org/show_bug.cgi?id=46049.

        * fast/js/regexp-bol-expected.txt: Added.
        * fast/js/regexp-bol.html: Added.
        * fast/js/script-tests/regexp-bol.js: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68021 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Wed, 22 Sep 2010 07:37:18 +0000 (07:37 +0000)]
2010-09-22  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Adam Roben.

        [WIN] Add internetHandle to WebCoreSynchronousLoader.
        https://bugs.webkit.org/show_bug.cgi?id=46185

        For loading content via network the WebCoreSynchronousLoader needs its own (synchronous) internetHandle.
        This also adds asynchronousInternetHandle() to create a static asynchronous internetHandle.

        * platform/network/win/ResourceHandleWin.cpp:
        (WebCore::createInternetHandle):
        (WebCore::asynchronousInternetHandle):
        (WebCore::WebCoreSynchronousLoader::internetHandle):
        (WebCore::WebCoreSynchronousLoader::WebCoreSynchronousLoader): Create synchronous internetHandle.
        (WebCore::WebCoreSynchronousLoader::~WebCoreSynchronousLoader): Delete synchronous internetHandle.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68020 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Wed, 22 Sep 2010 07:26:19 +0000 (07:26 +0000)]
2010-09-22  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Chris Fleizach.

        [chromium] Remove unused sys/types.h in VDMXParser.cpp
        https://bugs.webkit.org/show_bug.cgi?id=46136

        sys/types.h is not needed here.

        * platform/graphics/chromium/VDMXParser.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68019 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Wed, 22 Sep 2010 07:25:03 +0000 (07:25 +0000)]
2010-09-22  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Kent Tamura.

        [BREWMP] Change the home directory path to the WebKit module directory
        https://bugs.webkit.org/show_bug.cgi?id=45895

        AEEFS_HOME_DIR denotes the currently running application's directory.
        Change it to fs:/~<clsid> to use the WebKit module directory as home.

        * platform/brew/FileSystemBrew.cpp:
        (WebCore::homeDirectoryPath):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68018 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-22 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Wed, 22 Sep 2010 07:12:11 +0000 (07:12 +0000)]
2010-09-22  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Kent Tamura.

        [BREWMP] Add dummy Pasteboard
        https://bugs.webkit.org/show_bug.cgi?id=46107

        Pasteboard is not supported yet. Add dummy implementation to avoid link errors.

        * platform/brew/PasteboardBrew.cpp: Added.
        (WebCore::Pasteboard::generalPasteboard):
        (WebCore::Pasteboard::Pasteboard):
        (WebCore::Pasteboard::clear):
        (WebCore::Pasteboard::writeSelection):
        (WebCore::Pasteboard::writePlainText):
        (WebCore::Pasteboard::writeURL):
        (WebCore::Pasteboard::writeImage):
        (WebCore::Pasteboard::canSmartReplace):
        (WebCore::Pasteboard::plainText):
        (WebCore::Pasteboard::documentFragment):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68017 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Wed, 22 Sep 2010 05:56:52 +0000 (05:56 +0000)]
2010-09-21  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Kent Tamura.

        Each EntityMaskIn* needs explanation
        https://bugs.webkit.org/show_bug.cgi?id=44833

        Added a description as to why we escape nbsp when serializing HTML documents but not when serializing XML documents.

        * editing/markup.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68016 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 James Robinson <jamesr@chromium.org>
jamesr@google.com [Wed, 22 Sep 2010 05:53:17 +0000 (05:53 +0000)]
2010-09-21  James Robinson  <jamesr@chromium.org>

        Reviewed by Kenneth Russell.

        [chromium] Check if the acceleratedCompositingEnabled setting is active in WebViewImpl::graphicsContext3D()
        https://bugs.webkit.org/show_bug.cgi?id=46239

        This adds a check for whether acceleratedCompositingEnabled is true on the Page's Settings
        object before creating and vending a GraphicsContext3D object.  This is needed because when
        an offscreen WebGraphicsContext3DCommandBufferImpl is initialized it first asks the WebViewImpl
        for the compositor context.  WebGraphicsContext3DCommandBufferImpl::initialize checks if the
        disable accelerated compositing command line switch is set, but this is insufficient as sometimes
        the Setting object does not agree exactly with the command line switches.  For example,
        we explicitly toggle accelerated compositing for some chrome UI URLs regardless of what the
        flag says.

        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::graphicsContext3D):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68015 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Wed, 22 Sep 2010 05:36:45 +0000 (05:36 +0000)]
2010-09-21  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Kent Tamura.

        cleanup: removeInlineStyleFromElement and extractInlineStyleToPushDown should be merged
        https://bugs.webkit.org/show_bug.cgi?id=46205

        Cleanup required to fix the bug 27818. Added the style extraction mechanism to removeInlineStyleFromElement
        and removeCSSStyle and replaced the call to extractInlineStyleToPushDown by a call to removeInlineStyleFromElement.

        * editing/ApplyStyleCommand.cpp:
        (WebCore::ApplyStyleCommand::removeInlineStyleFromElement): Added extractedStyle argument.
        (WebCore::ApplyStyleCommand::removeCSSStyle): Added extractedStyle argument.
        (WebCore::ApplyStyleCommand::pushDownInlineStyleAroundNode): Calls removeInlineStyleFromElement instead of
        extractInlineStyleToPushDown which has been deleted.
        * editing/ApplyStyleCommand.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68014 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Wed, 22 Sep 2010 05:22:27 +0000 (05:22 +0000)]
2010-09-21  Kent Tamura  <tkent@chromium.org>

        Unreviewed, text expectation update.

        * platform/chromium/drt_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68013 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Wed, 22 Sep 2010 03:45:35 +0000 (03:45 +0000)]
2010-09-21  Kent Tamura  <tkent@chromium.org>

        Reviewed by David Levin.

        [Chromium] Remove WebInputElement::InputType and inputType()
        https://bugs.webkit.org/show_bug.cgi?id=46238

        * DEPS: Roll Chromium revision to 60132, which removed all usage of inputType().
        * public/WebInputElement.h: Remove InputType and inputType().
        * src/AssertMatchingEnums.cpp: Remove WebInputElement::InputType -
          HTMLInputElement::InputType matching tests.
        * src/WebInputElement.cpp: Remove inputType().

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68012 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Wed, 22 Sep 2010 03:35:51 +0000 (03:35 +0000)]
2010-09-21  Kent Tamura  <tkent@chromium.org>

        Unreviewed, test expectation update.

        * platform/chromium/drt_expectations.txt:
          Remove passing tests, and fix typo.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68011 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Wed, 22 Sep 2010 03:31:56 +0000 (03:31 +0000)]
2010-09-21  Kent Tamura  <tkent@chromium.org>

        Unreviewed, test expectation update.

        * platform/chromium/drt_expectations.txt: Update DRT results.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68010 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Wed, 22 Sep 2010 03:13:38 +0000 (03:13 +0000)]
2010-09-21  Dirk Pranke  <dpranke@chromium.org>

        Unreviewed, build fix.

        new-run-webkit-tests: fix bug introduced in r68008 where if you
        specify --chromium and no --platform, and you're running on
        windows, you use the 'chromium-win' port by default instead of the
        version-specific port. This breaks the buildbots.

        It will be good for this refactoring to settle down so I can
        rewrite the logic for default ports and the unit tests to be
        clearer (and the testing more comprehensive).

        * Scripts/webkitpy/layout_tests/port/chromium_win.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68009 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Wed, 22 Sep 2010 02:24:51 +0000 (02:24 +0000)]
2010-09-21  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Adam Barth.

        Modify the Port interface to take only a series of keyword arguments
        in the constructor, and modify Port/factory.get() to accomodate that,
        and to accept user=XXX as an argument so we can pass
        webkitpy.common.system.user.User objects in.

        Then, modify new-run-webkit-tests and rebaseline-chromium-webkit-tests
        to use the common routine in webkitpy.common.system.user.open_url()
        to display HTML files.

        There was a routine in the Port interface to do the same thing,
        but I see no need for a port-specific hook for this, since it is
        something that will always be executed by the host environment
        and displaying web pages has nothing to do with running layout tests.

        Note that new-run-webkit-tests used to use test_shell to display
        the page; this is potentially useful so that you can actually click
        from a result to the broken page; however, since DumpRenderTree
        doesn't support this functionality, it will be going away eventually.

        https://bugs.webkit.org/show_bug.cgi?id=46128

        * Scripts/webkitpy/layout_tests/port/base.py:
        * Scripts/webkitpy/layout_tests/port/base_unittest.py:
        * Scripts/webkitpy/layout_tests/port/chromium.py:
        * Scripts/webkitpy/layout_tests/port/chromium_linux.py:
        * Scripts/webkitpy/layout_tests/port/chromium_mac.py:
        * Scripts/webkitpy/layout_tests/port/chromium_unittest.py:
        * Scripts/webkitpy/layout_tests/port/chromium_win.py:
        * Scripts/webkitpy/layout_tests/port/dryrun.py:
        * Scripts/webkitpy/layout_tests/port/factory.py:
        * Scripts/webkitpy/layout_tests/port/google_chrome.py:
        * Scripts/webkitpy/layout_tests/port/google_chrome_unittest.py:
        * Scripts/webkitpy/layout_tests/port/gtk.py:
        * Scripts/webkitpy/layout_tests/port/mac.py:
        * Scripts/webkitpy/layout_tests/port/qt.py:
        * Scripts/webkitpy/layout_tests/port/test.py:
        * Scripts/webkitpy/layout_tests/port/webkit.py:
        * Scripts/webkitpy/layout_tests/port/win.py:
        * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests.py:
        * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
        * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68008 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Oliver Hunt <oliver@apple.com>
oliver@apple.com [Wed, 22 Sep 2010 01:22:25 +0000 (01:22 +0000)]
2010-09-21  Oliver Hunt  <oliver@apple.com>

        RS=Gavin Barraclough.

        Fix codeblock dumping

        * bytecode/CodeBlock.cpp:
        (JSC::CodeBlock::dump):
        * runtime/Executable.h:
        (JSC::ScriptExecutable::ScriptExecutable):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68006 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoRevert "2010-09-21 Chris Guillory <chris.guillory@google.com>"
atwilson@chromium.org [Wed, 22 Sep 2010 01:10:54 +0000 (01:10 +0000)]
Revert "2010-09-21  Chris Guillory   <chris.guillory@google.com>"

This reverts commit 347aab9430be62b827cd1130adbfdf99d18bfaf0.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68005 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Cosmin Truta <ctruta@chromium.org>
jamesr@google.com [Wed, 22 Sep 2010 01:08:21 +0000 (01:08 +0000)]
2010-09-21  Cosmin Truta  <ctruta@chromium.org>

        Reviewed by James Robinson.

        RenderImage::intrinsicSizeChanged crashes when m_imageResource is missing
        https://bugs.webkit.org/show_bug.cgi?id=46120

        Test a style change applied to a zoomed image.

        * css3/style-zoomed-image.html: Added.
        * css3/style-zoomed-image-expected.txt: Added.
2010-09-21  Cosmin Truta  <ctruta@chromium.org>

        Reviewed by James Robinson.

        RenderImage::intrinsicSizeChanged crashes when m_imageResource is missing
        https://bugs.webkit.org/show_bug.cgi?id=46120

        Avoid applying operator -> to m_imageResource when its underlying pointer is NULL.

        * rendering/RenderImage.h:
        (WebCore::RenderImage::intrinsicSizeChanged): Checked m_imageResource.
        * html/HTMLImageElement.cpp:
        (HTMLImageElement::createRenderer): Fixed indentation.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68004 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoUpdated test after r67994.
mitz@apple.com [Wed, 22 Sep 2010 01:07:13 +0000 (01:07 +0000)]
Updated test after r67994.

Reviewed by Geoffrey Garen.

* fast/dom/Range/range-expand.html: Do not expect sentence selections to include
trailing newline characters.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68003 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Lucas De Marchi <lucas.demarchi@profusion.mobi>
demarchi@webkit.org [Wed, 22 Sep 2010 00:38:41 +0000 (00:38 +0000)]
2010-09-21  Lucas De Marchi  <lucas.demarchi@profusion.mobi>

        Reviewed by Kenneth Rohde Christiansen.

        [EFL] Get fresh theme when running EWebLauncher
        https://bugs.webkit.org/show_bug.cgi?id=46210

        If user tried to run EWebLauncher outside of the build tree, it would
        not find the theme and fallback to the installed one. However, since
        EWebLauncher is not installed, we always want to take the theme from
        the just compiled source code. If user had never installed WebKit's
        EFL port it could even receive a segv since no theme would be found.

        Now EWebLauncher does not fallback to the installed theme and CMake
        gives as DATA_DIR the directory of the theme it has just built.

        * CMakeListsEfl.txt: export DATA_DIR to parent CMakeLists.txt in order
        to be used by EWebLauncher.
2010-09-21  Lucas De Marchi  <lucas.demarchi@profusion.mobi>

        Reviewed by Kenneth Rohde Christiansen.

        [EFL] Get fresh theme when running EWebLauncher
        https://bugs.webkit.org/show_bug.cgi?id=46210

        If user tried to run EWebLauncher outside of the build tree, it would
        not find the theme and fallback to the installed one. However, since
        EWebLauncher is not installed, we always want to take the theme from
        the just compiled source code. If user had never installed WebKit's
        EFL port it could even receive a segv since no theme would be found.

        Now EWebLauncher does not fallback to the installed theme and CMake
        gives as DATA_DIR the directory of the theme it has just built.

        * CMakeListsEfl.txt: Pass the build directory as DATA_DIR to
        EWebLauncher.
        * EWebLauncher/main.c: Use only the theme from build director. Do not
        fallback to others as this could hide real bugs.
        (quit):
        (browserCreate):
        (findThemePath):
        (main):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68002 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 John Gregg <johnnyg@google.com>
johnnyg@google.com [Wed, 22 Sep 2010 00:35:23 +0000 (00:35 +0000)]
2010-09-21  John Gregg  <johnnyg@google.com>

        Unreviewed, chromium test expecations.

        [chromium] update test expectations
        https://bugs.webkit.org/show_bug.cgi?id=46235

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68001 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Kenneth Russell <kbr@google.com>
kbr@google.com [Wed, 22 Sep 2010 00:34:32 +0000 (00:34 +0000)]
2010-09-21  Kenneth Russell  <kbr@google.com>

        Reviewed by James Robinson.

        [chromium] Remove GLES2Context and WebGLES2Context
        https://bugs.webkit.org/show_bug.cgi?id=46131

        Removed now-obsolete GLES2Context and WebGLES2Context classes, and
        dependency on command buffer client code from WebCore. Built and
        tested 3D CSS and WebGL content on Mac OS X to test.

        * WebKit.gyp:
        * public/WebGLES2Context.h: Removed.
        * public/WebGraphicsContext3D.h:
        * public/WebKitClient.h:
        * public/WebView.h:
        * src/ChromeClientImpl.cpp:
        * src/GLES2Context.cpp: Removed.
        * src/GLES2ContextInternal.cpp: Removed.
        * src/GLES2ContextInternal.h: Removed.
        * src/WebGraphicsContext3DDefaultImpl.cpp:
        * src/WebGraphicsContext3DDefaultImpl.h:
        * src/WebViewImpl.cpp:
        * src/WebViewImpl.h:
2010-09-21  Kenneth Russell  <kbr@google.com>

        Reviewed by James Robinson.

        [chromium] Remove GLES2Context and WebGLES2Context
        https://bugs.webkit.org/show_bug.cgi?id=46131

        Removed now-obsolete GLES2Context and WebGLES2Context classes, and
        dependency on command buffer client code from WebCore. Built and
        tested 3D CSS and WebGL content on Mac OS X to test.

        * WebCore.gypi:
        * platform/chromium/GLES2Context.h: Removed.
        * platform/graphics/skia/ImageSkia.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68000 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Wed, 22 Sep 2010 00:33:05 +0000 (00:33 +0000)]
2010-09-21  Dirk Pranke  <dpranke@chromium.org>

        Unreviewed, build fix.

        Don't pass the --test-shell arg to the Chromium Mac port of DRT;
        it just confuses it.

        https://bugs.webkit.org/show_bug.cgi?id=46230

        * Scripts/webkitpy/layout_tests/port/webkit.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67999 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoTry to fix the Windows build.
ggaren@apple.com [Wed, 22 Sep 2010 00:31:49 +0000 (00:31 +0000)]
Try to fix the Windows build.

* loader/Cache.cpp:
(WebCore::Cache::makeResourcePurgeable):
(WebCore::Cache::evict): Cast to int before using unary '-', to avoid an
MSVC warning.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67998 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoAdd DirectX SDK include path.
sfalken@apple.com [Wed, 22 Sep 2010 00:27:24 +0000 (00:27 +0000)]
Add DirectX SDK include path.

Rubber stamped by Simon Fraser.

* win/WebKit2DirectX.vsprops:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67997 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoUse purgeable memory to keep more dead resources in cache
ggaren@apple.com [Wed, 22 Sep 2010 00:18:03 +0000 (00:18 +0000)]
Use purgeable memory to keep more dead resources in cache
https://bugs.webkit.org/show_bug.cgi?id=44806
<rdar://problem/8350901>

Patch by Pratik Solanki <psolanki@apple.com> on 2010-09-21
Reviewed by Geoffrey Garen and Darin Adler.

This changes the behavior of dead resources in the WebCore cache to be the following if
shouldMakeResourcePurgeableOnEviction() returns true.

1. Dead resources in the cache are kept in non-purgeable memory.
2. When we prune dead resources, instead of freeing them, we mark their memory as purgeable
and keep the resources until the kernel reclaims the purgeable memory.

By leaving the in-cache dead resources in dirty resident memory, we decrease the likelihood
of the kernel claiming that memory and forcing us to refetch the resource (for example when
a user presses back).

And by having an unbounded number of resource objects using purgeable memory, we can use
as much memory as is available on the machine. The trade-off is that the CachedResource
object (and its member variables) are allocated in non-purgeable TC-malloc'd memory so
we would see slightly more memory use due to this.

* loader/Cache.cpp:
(WebCore::Cache::resourceForURL): Adjust sizes appropriately if we made resource memory
non-purgeable.
(WebCore::Cache::pruneDeadResources): When removing dead resources, try first to mark their
memory as purgeable. If not, evict the resource.
(WebCore::Cache::makeResourcePurgeable): Added. Try to mark resource
memory as purgeable. If successful, adjust the sizes so that we don't
factor this resources size in the Cache size calculation.
(WebCore::Cache::evict): Don't decrement size if we already did it in makeResourcePurgeable.
(WebCore::Cache::dumpLRULists): Extra debug logging.
* loader/Cache.h:
(WebCore::Cache::shouldMakeResourcePurgeableOnEviction): Added. Indicates if the new
behaviour is enabled.
* loader/CachedCSSStyleSheet.cpp:
(WebCore::CachedCSSStyleSheet::allClientsRemoved): Do not mark memory as purgeable. The
Cache class takes care of this.
* loader/CachedImage.cpp:
(WebCore::CachedImage::destroyDecodedData): Ditto.
* loader/CachedScript.cpp:
(WebCore::CachedScript::destroyDecodedData): Ditto.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67996 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoMake a new buildbot for Leopard Debug Test
slewis@apple.com [Wed, 22 Sep 2010 00:17:03 +0000 (00:17 +0000)]
Make a new buildbot for Leopard Debug Test

Reviewed by Stephanie Lewis.

* BuildSlaveSupport/build.webkit.org-config/config.json:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67995 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago<rdar://problem/7729077> Extending the selection to sentence boundary after a line...
mitz@apple.com [Wed, 22 Sep 2010 00:13:37 +0000 (00:13 +0000)]
<rdar://problem/7729077> Extending the selection to sentence boundary after a line break may select extra character
https://bugs.webkit.org/show_bug.cgi?id=46232

Reviewed by Darin Adler.

WebCore:

Test: editing/selection/extend-by-sentence-002.html

* editing/visible_units.cpp:
(WebCore::nextBoundary): The text iterator’s range end can be the position after
the line break, in which case the next visible is actually after the first character
of the next sentence. Instead, advance the text iterator past the newline character
and return the beginning of its range, which is guaranteed to still be before the
next sentence.

LayoutTests:

* editing/selection/extend-by-sentence-002-expected.txt: Added.
* editing/selection/extend-by-sentence-002.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67994 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Wed, 22 Sep 2010 00:06:58 +0000 (00:06 +0000)]
2010-09-21  Dirk Pranke  <dpranke@chromium.org>

        Unreviewed, build fix.

        Fix breakage of Chromium Mac DRT port caused by r67905.

        https://bugs.webkit.org/show_bug.cgi?id=46230

        * Scripts/webkitpy/layout_tests/port/webkit.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67993 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Kenneth Russell <kbr@google.com>
kbr@google.com [Tue, 21 Sep 2010 23:59:53 +0000 (23:59 +0000)]
2010-09-21  Kenneth Russell  <kbr@google.com>

        Reviewed by James Robinson.

        [chromium] Roll forward Chromium DEPS to pick up removal of WebGLES2Context dependencies
        https://bugs.webkit.org/show_bug.cgi?id=46231

        Ran build-webkit --chromium --debug on Mac OS X to test the roll-forward.

        * DEPS:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67992 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoForward window focus changes to the plug-in
andersca@apple.com [Tue, 21 Sep 2010 23:39:53 +0000 (23:39 +0000)]
Forward window focus changes to the plug-in
https://bugs.webkit.org/show_bug.cgi?id=46227

Reviewed by Dan Bernstein.

* WebProcess/Plugins/Netscape/mac/NetscapePluginMac.mm:
(WebKit::NetscapePlugin::windowFocusChanged):
Send the NPCocoaEventWindowFocusChanged event.

(WebKit::NetscapePlugin::windowFrameChanged):
(WebKit::NetscapePlugin::windowVisibilityChanged):
Add stubs.

* WebProcess/Plugins/Plugin.h:
Add windowFocusChanged, windowFrameChanged and windowVisibilityChanged.

* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::webPage):
Make webPage a member function instead.

(WebKit::PluginView::setWindowIsFocused):
Call the plug-in member function.

(WebKit::PluginView::initializePlugin):
When the plug-in has been initialized, update its window frame, window visibility
and window focus states.

(WebKit::PluginView::setParent):
Move viewGeometryDidChange to initializePlugin.

* WebProcess/Plugins/PluginView.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::setActive):
Tell all plug-in views about the new window focus state.

(WebKit::WebPage::windowIsFocused):
Return whether the window is focused or not.

* WebProcess/WebPage/WebPage.h:
(WebKit::WebPage::windowIsVisible):
(WebKit::WebPage::windowFrame):
Add getters.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67991 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Oliver Hunt <oliver@apple.com>
oliver@apple.com [Tue, 21 Sep 2010 23:19:53 +0000 (23:19 +0000)]
2010-09-21  Oliver Hunt  <oliver@apple.com>

        Reviewed by Geoffrey Garen.

        Speed up function.apply(..., arguments)
        https://bugs.webkit.org/show_bug.cgi?id=46207

        Add code to do argument copying inline in the case
        where we're using Function.apply to forward our arguments
        directly.

        * jit/JIT.cpp:
        (JSC::JIT::privateCompileSlowCases):
           Splitted op_load_varargs into fast and slow paths, so add the call
           to the slow path generator.
        * jit/JIT.h:
        * jit/JITCall32_64.cpp:
          Remove 32bit specific emit_op_load_varargs as the logic is the
          same for all value representations
        * jit/JITOpcodes.cpp:
        (JSC::JIT::emit_op_load_varargs):
          Copy arguments inline
        (JSC::JIT::emitSlow_op_load_varargs):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67990 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Robert Hogan <robert@webkit.org>
robert@webkit.org [Tue, 21 Sep 2010 22:27:17 +0000 (22:27 +0000)]
2010-09-21  Robert Hogan  <robert@webkit.org>

        Rubber-stamped by Ariya Hidayat.

        [Qt] Build fix against qt trunk

        * platform/graphics/qt/ContextShadow.cpp:
        (WebCore::ContextShadow::beginShadowLayer):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67989 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Jeremy Orlow <jorlow@chromium.org>
jorlow@chromium.org [Tue, 21 Sep 2010 21:04:02 +0000 (21:04 +0000)]
2010-09-21  Jeremy Orlow  <jorlow@chromium.org>

        Reviewed by Nate Chapin.

        [Chromium] Plumb the IDBTransansaction's id
        https://bugs.webkit.org/show_bug.cgi?id=46197

        Next step, add the plumbing to Chromium.  Then we add code to WebKit
        that uses it (and take out the default IDBTransaction param).

        * public/WebIDBCursor.h:
        (WebKit::WebIDBCursor::key):
        (WebKit::WebIDBCursor::value):
        * public/WebIDBFactory.h:
        (WebKit::WebIDBFactory::open):
        * public/WebIDBIndex.h:
        (WebKit::WebIDBIndex::openObjectCursor):
        (WebKit::WebIDBIndex::openCursor):
        (WebKit::WebIDBIndex::getObject):
        (WebKit::WebIDBIndex::get):
        * public/WebIDBObjectStore.h:
        (WebKit::WebIDBObjectStore::get):
        (WebKit::WebIDBObjectStore::put):
        (WebKit::WebIDBObjectStore::remove):
        (WebKit::WebIDBObjectStore::createIndex):
        (WebKit::WebIDBObjectStore::removeIndex):
        (WebKit::WebIDBObjectStore::openCursor):
        * src/WebIDBIndexImpl.cpp:
        (WebKit::WebIDBIndexImpl::openCursor):
        (WebKit::WebIDBIndexImpl::openObjectCursor):
        (WebKit::WebIDBIndexImpl::getObject):
        (WebKit::WebIDBIndexImpl::get):
        * src/WebIDBIndexImpl.h:
        * src/WebIDBObjectStoreImpl.cpp:
        (WebKit::WebIDBObjectStoreImpl::get):
        (WebKit::WebIDBObjectStoreImpl::put):
        (WebKit::WebIDBObjectStoreImpl::remove):
        (WebKit::WebIDBObjectStoreImpl::openCursor):
        * src/WebIDBObjectStoreImpl.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67986 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Tony Chang <tony@chromium.org>
tony@chromium.org [Tue, 21 Sep 2010 21:00:30 +0000 (21:00 +0000)]
2010-09-21  Tony Chang  <tony@chromium.org>

        Unreviewed, QT build fix.

        * platform/qt/ClipboardQt.cpp:
        (WebCore::ClipboardQt::ClipboardQt):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67985 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years agoHandle booleans in injected bundle messages
andersca@apple.com [Tue, 21 Sep 2010 20:51:05 +0000 (20:51 +0000)]
Handle booleans in injected bundle messages
https://bugs.webkit.org/show_bug.cgi?id=46213

Reviewed by John Sullivan.

* Shared/UserMessageCoders.h:
(WebKit::UserMessageEncoder::baseEncode):
(WebKit::UserMessageDecoder::baseDecode):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67984 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 years ago2010-09-21 Andy Estes <aestes@apple.com>
aestes@apple.com [Tue, 21 Sep 2010 20:43:51 +0000 (20:43 +0000)]
2010-09-21  Andy Estes  <aestes@apple.com>

        Reviewed by Darin Adler.

        REGRESSION (r61285): AIM 2.1.296: Code rendered as text in Welcome screen
        https://bugs.webkit.org/show_bug.cgi?id=46134

        AIM clients linked against versions of WebKit prior to the introduction
        of the HTML5 parser contain markup incompatible with the new parser.
        Enable parser quirks in this case to remain compatible with these
        clients.

        * WebView/WebView.mm:
        (shouldUsePreHTML5ParserQuirks): Returns true if the embedding
        application is AIM and was linked against a version of WebKit prior to
        the introduction of the HTML5 parser, or if the
        WebKitPreHTML5ParserQuirks WebPreference is enabled.
        (-[WebView _preferencesChangedNotification:]): Call
        WebCore::Settings::setUsePreHTML5ParserQuirks().
2010-09-21  Andy Estes  <aestes@apple.com>

        Reviewed by Darin Adler.

        REGRESSION (r61285): AIM 2.1.296: Code rendered as text in Welcome screen
        https://bugs.webkit.org/show_bug.cgi?id=46134

        * WebCore.exp.in: Export __ZN7WebCore32applicationIsAOLInstantMessengerEv.
        * platform/mac/RuntimeApplicationChecks.h:
        * platform/mac/RuntimeApplicationChecks.mm:
        (WebCore::applicationIsAOLInstantMessenger): Determine if the embedding
        application is AOL Instant Messenger by checking the bundle identifier.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67983 268f45cc-cd09-0410-ab3c-d52691b4dbfc