Timeline


and

12/10/09:

23:54 QtWebKit/CodeCamp200912 edited by zecke@selfish.org
Start documenting day 3 (diff)
23:51 QtWebKit/CodeCamp200912 edited by zecke@selfish.org
(diff)
19:42 Changeset [51978] by abarth@webkit.org
  • 5 edits in trunk/JavaScriptCore

2009-12-10 Adam Barth <abarth@webkit.org>

No review, rolling out r51975.
http://trac.webkit.org/changeset/51975

  • jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION):
  • runtime/JSString.h: (JSC::JSString::JSString): (JSC::JSString::appendStringInConstruct):
  • runtime/Operations.cpp: (JSC::jsAddSlowCase):
  • runtime/Operations.h: (JSC::jsString): (JSC::jsAdd):
19:12 Changeset [51977] by abarth@webkit.org
  • 3 edits
    3 adds in trunk

2009-12-10 Adam Barth <abarth@webkit.org>

Reviewed by Darin Adler.

Mixed content shouldn't trigger for plug-ins without URLs
https://bugs.webkit.org/show_bug.cgi?id=32384

  • http/tests/security/mixedContent/empty-url-plugin-in-frame-expected.txt: Added.
  • http/tests/security/mixedContent/empty-url-plugin-in-frame.html: Added.
  • http/tests/security/mixedContent/resources/frame-with-empty-url-plugin.html: Added.

2009-12-10 Adam Barth <abarth@webkit.org>

Reviewed by Darin Adler.

Mixed content shouldn't trigger for plug-ins without URLs
https://bugs.webkit.org/show_bug.cgi?id=32384

These plug-ins cannot be controlled by active network attackers, so
there's no reason to trigger a mixed content warning.

Test: http/tests/security/mixedContent/empty-url-plugin-in-frame.html

  • loader/FrameLoader.cpp: (WebCore::FrameLoader::isMixedContent):
18:22 Changeset [51976] by oliver@apple.com
  • 5 edits in trunk

Incorrect caching of prototype lookup with dictionary base
https://bugs.webkit.org/show_bug.cgi?id=32402

Reviewed by Gavin Barraclough

Make sure we don't add cached prototype lookup to the proto_list
lookup chain if the top level object is a dictionary.

18:07 Changeset [51975] by barraclough@apple.com
  • 5 edits in trunk/JavaScriptCore

https://bugs.webkit.org/show_bug.cgi?id=32400
Switch remaining cases of string addition to use ropes.

Reviewed by Oliver Hunt.

~1% progression on Sunspidey.

  • jit/JITStubs.cpp:

(JSC::DEFINE_STUB_FUNCTION):

  • runtime/JSString.h:

(JSC::JSString::JSString):
(JSC::JSString::appendStringInConstruct):

  • runtime/Operations.cpp:

(JSC::jsAddSlowCase):

  • runtime/Operations.h:

(JSC::jsString):
(JSC::jsAdd):

16:58 Changeset [51974] by eroman@chromium.org
  • 2 edits in trunk/WebKit/chromium

Unreviewed, build fix for chromium.

  • src/FrameLoaderClientImpl.cpp: (WebKit::FrameLoaderClientImpl::dispatchWillSendRequest):
16:41 Changeset [51973] by ap@apple.com
  • 3 edits
    3 adds in trunk

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=32140
REGRESSION(r50072): Mailman administrative functionality is broken

Test: http/tests/misc/percent-sign-in-form-field-name.html

  • platform/network/FormDataBuilder.cpp: (WebCore::appendQuotedString): Don't encode percent sign, we never had a good reason to do that.
16:20 Changeset [51972] by eric@webkit.org
  • 3 edits in trunk/WebKitTools

2009-12-10 Eric Seidel <eric@webkit.org>

No review, just updating unit tests to match recent checkins.

  • Scripts/modules/mock_bugzillatool.py:
    • Add missing red_core_builders_names method causing exception.
  • Scripts/modules/workqueue_unittest.py:
    • processutils is dead, use executive.py instead.
16:06 Changeset [51971] by eric@webkit.org
  • 17 edits in trunk

2009-12-10 Kent Hansen <kent.hansen@nokia.com>

Reviewed by Geoffrey Garen.

Remove JSObject::getPropertyAttributes() and all usage of it.
https://bugs.webkit.org/show_bug.cgi?id=31933

getOwnPropertyDescriptor() should be used instead.

  • JavaScriptCore.exp:
  • JavaScriptCore.order:
  • JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
  • debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::getOwnPropertyDescriptor):
  • debugger/DebuggerActivation.h:
  • runtime/JSObject.cpp: (JSC::JSObject::propertyIsEnumerable):
  • runtime/JSObject.h:
  • runtime/JSVariableObject.cpp:
  • runtime/JSVariableObject.h:

2009-12-10 Kent Hansen <kent.hansen@nokia.com>

Reviewed by Geoffrey Garen.

Remove getPropertyAttributes() from JavaScript bindings.
https://bugs.webkit.org/show_bug.cgi?id=31933

The functionality is provided by getOwnPropertyDescriptor().

  • WebCore.order:
  • bindings/js/JSDOMWindowCustom.cpp:
  • bindings/js/JSDOMWindowShell.cpp:
  • bindings/js/JSDOMWindowShell.h:
  • bindings/scripts/CodeGeneratorJS.pm:
  • page/DOMWindow.idl:
15:55 Changeset [51970] by eric@webkit.org
  • 23 edits
    3 adds in trunk

2009-12-10 Kenneth Russell <kbr@google.com>

Reviewed by Oliver Hunt.

Changed WebGLRenderingContext to synthesize GL errors rather than
raising JavaScript exceptions. Removed internal getError() calls
after each graphics call. The GraphicsContext3D maintains the
synthetic exceptions because only it has complete information
about certain conditions requiring them to be raised.

Based on idea from Ilmari Heikkinen, added create3DDebugContext()
to webgl-test.js and changed the WebGL layout tests expecting
error conditions to use it. Updated expected.txt files, which now
implicitly test the OpenGL error as it is part of the exception's
message.

Added new targeted test covering aspects of synthetic errors as
well as regression tests for bugs uncovered during its development.

Test: fast/canvas/webgl/error-reporting.html

  • fast/canvas/webgl/drawArraysOutOfBounds-expected.txt:
  • fast/canvas/webgl/drawArraysOutOfBounds.html:
  • fast/canvas/webgl/drawElementssOutOfBounds-expected.txt:
  • fast/canvas/webgl/drawElementssOutOfBounds.html:
  • fast/canvas/webgl/error-reporting-expected.txt: Added.
  • fast/canvas/webgl/error-reporting.html: Added.
  • fast/canvas/webgl/getActiveTest-expected.txt:
  • fast/canvas/webgl/incorrect-context-object-behaviour-expected.txt:
  • fast/canvas/webgl/invalidPassedParams-expected.txt:
  • fast/canvas/webgl/invalidPassedParams.html:
  • fast/canvas/webgl/null-object-behaviour-expected.txt:
  • fast/canvas/webgl/resources/webgl-test.js: (create3DDebugContext.wrap.getError): (create3DDebugContext):
  • fast/canvas/webgl/script-tests/error-reporting.js: Added.
  • fast/canvas/webgl/script-tests/getActiveTest.js:
  • fast/canvas/webgl/script-tests/incorrect-context-object-behaviour.js:
  • fast/canvas/webgl/script-tests/null-object-behaviour.js:
  • fast/canvas/webgl/script-tests/uniform-location.js:
  • fast/canvas/webgl/uniform-location-expected.txt:

2009-12-10 Kenneth Russell <kbr@google.com>

Reviewed by Oliver Hunt.

Changed WebGLRenderingContext to synthesize GL errors rather than
raising JavaScript exceptions. Removed internal getError() calls
after each graphics call. The GraphicsContext3D maintains the
synthetic exceptions because only it has complete information
about certain conditions requiring them to be raised.

Based on idea from Ilmari Heikkinen, added create3DDebugContext()
to webgl-test.js and changed the WebGL layout tests expecting
error conditions to use it. Updated expected.txt files, which now
implicitly test the OpenGL error as it is part of the exception's
message.

Added new targeted test covering aspects of synthetic errors as
well as regression tests for bugs uncovered during its development.

Test: fast/canvas/webgl/error-reporting.html

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::sizeInBytes): (WebCore::WebGLRenderingContext::activeTexture): (WebCore::WebGLRenderingContext::attachShader): (WebCore::WebGLRenderingContext::bindAttribLocation): (WebCore::WebGLRenderingContext::bindBuffer): (WebCore::WebGLRenderingContext::bindFramebuffer): (WebCore::WebGLRenderingContext::bindRenderbuffer): (WebCore::WebGLRenderingContext::bindTexture): (WebCore::WebGLRenderingContext::bufferData): (WebCore::WebGLRenderingContext::bufferSubData): (WebCore::WebGLRenderingContext::compileShader): (WebCore::WebGLRenderingContext::createShader): (WebCore::WebGLRenderingContext::detachShader): (WebCore::WebGLRenderingContext::disableVertexAttribArray): (WebCore::WebGLRenderingContext::drawArrays): (WebCore::WebGLRenderingContext::drawElements): (WebCore::WebGLRenderingContext::enableVertexAttribArray): (WebCore::WebGLRenderingContext::framebufferRenderbuffer): (WebCore::WebGLRenderingContext::framebufferTexture2D): (WebCore::WebGLRenderingContext::getActiveAttrib): (WebCore::WebGLRenderingContext::getActiveUniform): (WebCore::WebGLRenderingContext::getBufferParameter): (WebCore::WebGLRenderingContext::getFramebufferAttachmentParameter): (WebCore::WebGLRenderingContext::getParameter): (WebCore::WebGLRenderingContext::getProgramParameter): (WebCore::WebGLRenderingContext::getProgramInfoLog): (WebCore::WebGLRenderingContext::getRenderbufferParameter): (WebCore::WebGLRenderingContext::getShaderParameter): (WebCore::WebGLRenderingContext::getShaderInfoLog): (WebCore::WebGLRenderingContext::getShaderSource): (WebCore::WebGLRenderingContext::getTexParameter): (WebCore::WebGLRenderingContext::getUniform): (WebCore::WebGLRenderingContext::getUniformLocation): (WebCore::WebGLRenderingContext::getVertexAttrib): (WebCore::WebGLRenderingContext::linkProgram): (WebCore::WebGLRenderingContext::shaderSource): (WebCore::WebGLRenderingContext::texImage2D): (WebCore::WebGLRenderingContext::texSubImage2D): (WebCore::WebGLRenderingContext::uniform1f): (WebCore::WebGLRenderingContext::uniform1fv): (WebCore::WebGLRenderingContext::uniform1i): (WebCore::WebGLRenderingContext::uniform1iv): (WebCore::WebGLRenderingContext::uniform2f): (WebCore::WebGLRenderingContext::uniform2fv): (WebCore::WebGLRenderingContext::uniform2i): (WebCore::WebGLRenderingContext::uniform2iv): (WebCore::WebGLRenderingContext::uniform3f): (WebCore::WebGLRenderingContext::uniform3fv): (WebCore::WebGLRenderingContext::uniform3i): (WebCore::WebGLRenderingContext::uniform3iv): (WebCore::WebGLRenderingContext::uniform4f): (WebCore::WebGLRenderingContext::uniform4fv): (WebCore::WebGLRenderingContext::uniform4i): (WebCore::WebGLRenderingContext::uniform4iv): (WebCore::WebGLRenderingContext::uniformMatrix2fv): (WebCore::WebGLRenderingContext::uniformMatrix3fv): (WebCore::WebGLRenderingContext::uniformMatrix4fv): (WebCore::WebGLRenderingContext::useProgram): (WebCore::WebGLRenderingContext::validateProgram): (WebCore::WebGLRenderingContext::vertexAttribPointer):
  • html/canvas/WebGLRenderingContext.h: (WebCore::WebGLRenderingContext::cleanupAfterGraphicsCall):
  • platform/graphics/GraphicsContext3D.h:
  • platform/graphics/mac/GraphicsContext3DMac.cpp: (WebCore::GraphicsContext3D::getActiveAttrib): (WebCore::GraphicsContext3D::getActiveUniform): (WebCore::GraphicsContext3D::getError): (WebCore::GraphicsContext3D::synthesizeGLError):

2009-12-10 Kenneth Russell <kbr@google.com>

Reviewed by Oliver Hunt.

Changed WebGLRenderingContext to synthesize GL errors rather than
raising JavaScript exceptions. Removed internal getError() calls
after each graphics call. The GraphicsContext3D maintains the
synthetic exceptions because only it has complete information
about certain conditions requiring them to be raised.

Based on idea from Ilmari Heikkinen, added create3DDebugContext()
to webgl-test.js and changed the WebGL layout tests expecting
error conditions to use it. Updated expected.txt files, which now
implicitly test the OpenGL error as it is part of the exception's
message.

Added new targeted test covering aspects of synthetic errors as
well as regression tests for bugs uncovered during its development.

Test: fast/canvas/webgl/error-reporting.html

  • src/GraphicsContext3D.cpp: (WebCore::GraphicsContext3DInternal::reshape): (WebCore::GraphicsContext3DInternal::getError): (WebCore::GraphicsContext3DInternal::synthesizeGLError): (WebCore::GraphicsContext3D::getActiveAttrib): (WebCore::GraphicsContext3D::getActiveUniform): (WebCore::GraphicsContext3D::getError): (WebCore::GraphicsContext3D::synthesizeGLError):
15:47 Changeset [51969] by eric@webkit.org
  • 4 edits in trunk/WebKitTools

2009-12-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Move run_command onto Executive to make code which uses run_command testable
https://bugs.webkit.org/show_bug.cgi?id=32396

  • Scripts/modules/executive.py:
    • Move run_command and error handlers onto Executive.
  • Scripts/modules/scm.py:
  • Scripts/modules/scm_unittest.py:
15:40 Changeset [51968] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-12-10 Stephen White <senorblanco@chromium.org>

Reviewed by Dimitri Glazkov.

Fix for assert on Chrome/skia with SVG Filters enabled.
https://bugs.webkit.org/show_bug.cgi?id=32394

Covered by LayoutTests/svg/W3C-SVG-1.1/filters-tile-01-b.svg and others.

  • platform/graphics/skia/ImageBufferSkia.cpp: (WebCore::ImageBuffer::platformTransformColorSpace):
15:33 Changeset [51967] by eric@webkit.org
  • 3 edits in trunk/WebKit/chromium

2009-12-10 Mike Belshe <mike@belshe.com>

Reviewed by Darin Fisher.

Fix FrameLoader to use the new ResourceTypes properly.

https://bugs.webkit.org/show_bug.cgi?id=32336

  • public/WebURLRequest.h: (WebKit::WebURLRequest::):
  • src/FrameLoaderClientImpl.cpp: (WebKit::setTargetTypeFromLoader): (WebKit::FrameLoaderClientImpl::dispatchWillSendRequest):
15:28 Changeset [51966] by hclam@chromium.org
  • 1 edit
    5 adds in trunk/LayoutTests

Layout test for video element on HTTP
https://bugs.webkit.org/show_bug.cgi?id=32333

Reviewed by Eric Carlson.

Added a new layout test that tests video element on HTTP protocol.
This test load the same video file twice. The first load will cache
the video file if caching is supported. The second load ensure
the cache works correctly.

  • http/tests/media/resources/test.mp4: Copied from LayoutTests/media/content/test.mp4.
  • http/tests/media/resources/test.ogv: Copied from LayoutTests/media/content/test.ogv.

The above two files are copied from LayoutTests/media/content because the HTTP
server is rooted at LayoutTests/http/tests. And we don't want to serve these
files with cgi because that will ruin the HTTP headers.

  • http/tests/media/video-load-twice-expected.txt: Added.
  • http/tests/media/video-load-twice.html: Added.
14:59 Changeset [51965] by enrica@apple.com
  • 7 edits
    2 adds in trunk

REGRESSION(4.0.4-42a12): With 2 highlighted lines of text in gmail/hotmail selecting Bold selects other 2 edit buttons automatically.
<rdar://problem/7442065>
https://bugs.webkit.org/show_bug.cgi?id=32285

Reviewed by Darin Adler.

WebCore:

When examining the styles of the nodes after the first in a range selection, we take into
cosideration differences in style of txt nodes only.

Test: editing/execCommand/queryCommandState-02.html

  • editing/Editor.cpp:

(WebCore::Editor::selectionHasStyle):

LayoutTests:

  • editing/execCommand/queryCommandState-02-expected.txt: Added.
  • editing/execCommand/queryCommandState-02.html: Added.
  • editing/execCommand/script-tests/toggle-compound-styles.js: Fixed typo in text.
  • platform/mac/editing/execCommand/toggle-compound-styles-expected.txt: Fixed typo in text.
  • platform/qt/editing/execCommand/toggle-compound-styles-expected.txt: Fixed typo in text.
  • platform/win/editing/execCommand/toggle-compound-styles-expected.txt: Fixed typo in text.
14:13 Changeset [51964] by barraclough@apple.com
  • 7 edits in trunk/JavaScriptCore

https://bugs.webkit.org/show_bug.cgi?id=32367
Add support for short Ropes (up to 3 entries) inline within JSString.
(rather than externally allocating an object to hold the rope).
Switch jsAdd of (JSString* + JSString*) to now make use of Ropes.

Reviewed by Oliver Hunt & Mark Rowe.

~1% progression on Sunspidey.

  • interpreter/Interpreter.cpp:

(JSC::Interpreter::privateExecute):

  • jit/JITOpcodes.cpp:

(JSC::JIT::privateCompileCTIMachineTrampolines):

  • jit/JITStubs.cpp:

(JSC::DEFINE_STUB_FUNCTION):

  • runtime/JSString.cpp:

(JSC::JSString::resolveRope):
(JSC::JSString::toBoolean):
(JSC::JSString::getStringPropertyDescriptor):

  • runtime/JSString.h:

(JSC::JSString::Rope::Fiber::deref):
(JSC::JSString::Rope::Fiber::ref):
(JSC::JSString::Rope::Fiber::refAndGetLength):
(JSC::JSString::Rope::append):
(JSC::JSString::JSString):
(JSC::JSString::~JSString):
(JSC::JSString::value):
(JSC::JSString::tryGetValue):
(JSC::JSString::length):
(JSC::JSString::canGetIndex):
(JSC::JSString::appendStringInConstruct):
(JSC::JSString::appendValueInConstructAndIncrementLength):
(JSC::JSString::isRope):
(JSC::JSString::string):
(JSC::JSString::ropeLength):
(JSC::JSString::getStringPropertySlot):

  • runtime/Operations.h:

(JSC::jsString):
(JSC::jsAdd):
(JSC::resolveBase):

14:10 Changeset [51963] by dglazkov@chromium.org
  • 2 edits in trunk/LayoutTests

2009-12-10 Dimitri Glazkov <dglazkov@chromium.org>

Unreviewed, fixing layout test mishap.

Add a newline to the layout test.

  • http/tests/security/isolatedWorld/events-expected.txt: Added a newline at the end of the file.
14:06 Changeset [51962] by oliver@apple.com
  • 5 edits
    3 adds in trunk

Crash in XMLTokenizer::popCurrentNode if window.close() is called during parsing
https://bugs.webkit.org/show_bug.cgi?id=31576

Reviewed by Alexey Proskuryakov.

Add a RefCounted wrapper object around xmlParserCtxtPtr so we can
maintain it's lifetime more effectively.

Test: fast/parser/xhtml-close-while-parsing.xhtml

13:59 BuildingQtOnOSX edited by piet.webkit@yahoo.com
Require qt4-mac-dev (v4.6) instead of qt4-mac (v4.5.3). See … (diff)
13:54 Changeset [51961] by pfeldman@chromium.org
  • 4 edits in trunk/WebCore

2009-12-10 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: debugger shortcuts don't work when
Search field or Console drawer has focus.

https://bugs.webkit.org/show_bug.cgi?id=32392

13:42 Changeset [51960] by dglazkov@chromium.org
  • 17 edits
    2 copies
    2 adds in trunk

WebCore:

2009-12-10 Dimitri Glazkov <dglazkov@chromium.org>

Reviewed by Adam Barth.

[V8] Events created in isolated worlds may fire in main world.
https://bugs.webkit.org/show_bug.cgi?id=32386

Test: http/tests/security/isolatedWorld/events.html

  • WebCore.gypi: Added WorldContextHandle.
  • bindings/v8/ScriptEventListener.cpp: (WebCore::createAttributeEventListener): Added WorldContextHandle params.
  • bindings/v8/SharedPersistent.h: Fixed a few style/include issues.
  • bindings/v8/V8AbstractEventListener.cpp: (WebCore::V8AbstractEventListener::V8AbstractEventListener): Added WorldContextHandle params. (WebCore::V8AbstractEventListener::handleEvent): Adjusted context retrieval to use WorldContextHandle. (WebCore::V8AbstractEventListener::invokeEventHandler): Ditto.
  • bindings/v8/V8AbstractEventListener.h: (WebCore::V8AbstractEventListener::worldContext): Added WorldContextHandle params.
  • bindings/v8/V8EventListenerList.h: (WebCore::V8EventListenerList::findOrCreateWrapper): Ditto.
  • bindings/v8/V8LazyEventListener.cpp: (WebCore::V8LazyEventListener::V8LazyEventListener): Ditto. (WebCore::V8LazyEventListener::prepareListenerObject): Adjusted context retrieval to use WorldContextHandle.
  • bindings/v8/V8LazyEventListener.h: (WebCore::V8LazyEventListener::create): Added WorldContextHandle params.
  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::context): Refactored to use mainWorldContext(); (WebCore::V8Proxy::mainWorldContext): Added. (WebCore::toV8Context): Changed to use WorldContextHandle.
  • bindings/v8/V8Proxy.h: Added mainWorldContext decl.
  • bindings/v8/V8Utilities.cpp: (WebCore::reportException): Added an extra check to avoid crashes during frame teardown.
  • bindings/v8/V8WorkerContextEventListener.cpp: (WebCore::V8WorkerContextEventListener::V8WorkerContextEventListener): Added WorldContextHandle params.
  • bindings/v8/V8WorkerContextEventListener.h: (WebCore::V8WorkerContextEventListener::create): Added WorldContextHandle params.
  • bindings/v8/WorldContextHandle.cpp: Added.
  • bindings/v8/WorldContextHandle.h: Added.
  • bindings/v8/custom/V8CustomEventListener.cpp: (WebCore::V8EventListener::V8EventListener): Added WorldContextHandle params.
  • bindings/v8/custom/V8CustomEventListener.h: (WebCore::V8EventListener::create): Added WorldContextHandle params.

LayoutTests:

2009-12-10 Dimitri Glazkov <dglazkov@chromium.org>

Reviewed by Adam Barth.

[V8] Events created in isolated worlds may fire in main world.
https://bugs.webkit.org/show_bug.cgi?id=32386

This test is relevant to both JSC and V8 isolated world implementation,
ensuring that events are fired in the right worlds.

  • http/tests/security/isolatedWorld/events.html: Added.
  • http/tests/security/isolatedWorld/events-expected.txt: Added.
13:17 Changeset [51959] by eric@webkit.org
  • 6 edits in trunk/WebKitTools

2009-12-09 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

bugzilla-tool needs a command to list patches needing cq+
https://bugs.webkit.org/show_bug.cgi?id=32351

  • Scripts/modules/bugzilla.py:
    • Parse attacher_email from attachment xml.
  • Scripts/modules/bugzilla_unittest.py:
    • Test new attacher_email parsing.
  • Scripts/modules/commands/queries.py:
    • Add PatchesToCommitQueue
  • Scripts/modules/commands/queries_unittest.py:
    • Tests for PatchesToCommitQueue
  • Scripts/modules/mock_bugzillatool.py:
    • Add necessary mock methods for running PatchesToCommitQueue
13:04 Changeset [51958] by jhoneycutt@apple.com
  • 4 edits in trunk/WebKit/mac

Mac build fix. Unreviewed.

Re-adds code that was mistakenly removed from my last patch.

  • WebCoreSupport/WebPluginHalterClient.mm:

Add necessary #import.

  • WebView/WebDelegateImplementationCaching.h:

Declare a new overload of CallUIDelegateReturningBoolean.

  • WebView/WebDelegateImplementationCaching.mm:

(CallDelegateReturningBoolean):
Add a new overload with different arguments.
(CallUIDelegateReturningBoolean):
Ditto.

12:24 Changeset [51957] by kov@webkit.org
  • 2 edits in trunk/WebKit/gtk

Reviewed by Xan Lopez.

Add a missing null-check, that is causing some crash reports.

  • WebCoreSupport/FrameLoaderClientGtk.cpp: (WebKit::FrameLoaderClient::committedLoad):
11:57 Changeset [51956] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-10 Adam Barth <abarth@webkit.org>

Unreviewed. Turns out every StepSequence command needs a --quiet
option.

  • Scripts/modules/stepsequence.py:
11:51 Changeset [51955] by andersca@apple.com
  • 4 edits in trunk/JavaScriptCore

Fix three more things found by compiling with clang++.

Reviewed by Geoffrey Garen.

  • runtime/Structure.h:

(JSC::StructureTransitionTable::reifySingleTransition):
Add the 'std' qualifier to the call to make_pair.

  • wtf/DateMath.cpp:

(WTF::initializeDates):
Incrementing a bool is deprecated according to the C++ specification.

  • wtf/PtrAndFlags.h:

(WTF::PtrAndFlags::PtrAndFlags):
Name lookup should not be done in dependent bases, so explicitly qualify the call to set.

11:33 Changeset [51954] by bweinstein@apple.com
  • 1 edit in trunk/WebKit/win/Interfaces/WebKit.idl

Touch WebKit.idl to try and fix the Windows build.

11:23 Changeset [51953] by jhoneycutt@apple.com
  • 16 edits in trunk

Pass more information about a plug-in to the PluginHalterDelegate

Reviewed by Adam Roben.

WebCore:

  • loader/EmptyClients.h:

Remove this unused class.

  • page/HaltablePlugin.h:

Add new functions to return the plug-in's name and whether it is
windowed.

  • page/PluginHalter.cpp:

(WebCore::PluginHalter::timerFired):
Pass new arguments to the client.

  • page/PluginHalterClient.h:

Add new parameters.

  • plugins/PluginView.cpp:

(WebCore::PluginView::pluginName):
Return the name from the PluginPackage.

  • plugins/PluginView.h:

(WebCore::PluginView::isWindowed):

WebKit/mac:

  • Plugins/WebBaseNetscapePluginView.mm:

(WebHaltablePlugin::isWindowed):
Return false - the Mac doesn't really have windowed plug-ins.
(WebHaltablePlugin::pluginName):
Return the name from the plug-in package.

  • WebCoreSupport/WebPluginHalterClient.h:

Update for new parameters.

  • WebCoreSupport/WebPluginHalterClient.mm:

(WebPluginHalterClient::shouldHaltPlugin):
Ditto; pass them when making the delegate call.

  • WebView/WebUIDelegatePrivate.h:

Update for new parameters.

WebKit/win:

  • Interfaces/IWebPluginHalterDelegate.idl:

Add new parameters.

  • WebCoreSupport/WebPluginHalterClient.cpp:

(WebPluginHalterClient::shouldHaltPlugin):
Update for new parameters. Pass them when making the delegate call.

  • WebCoreSupport/WebPluginHalterClient.h:

Update for new parameters.

10:58 Changeset [51952] by bweinstein@apple.com
  • 3 edits in trunk/WebCore

Fixes <http://webkit.org/b/31738>.
Web Inspector: Console Scope Bar should be on top of console when Console is a Panel.

Reviewed by Pavel Feldman.

When the console is set to be shown as a full panel, move the scope bar to the top
where it looks best. When it is only shown as a drawer, move it to the bottom, where
it originally was. This takes up less space as a drawer, and makes it more discoverable
and usable when it is the full panel.

  • inspector/front-end/ConsolePanel.js:

(WebInspector.ConsolePanel.prototype.show): Move filter bar to top.
(WebInspector.ConsolePanel.prototype.hide): Move filter bar back to bottom.

  • inspector/front-end/inspector.css:
09:54 Changeset [51951] by abarth@webkit.org
  • 1 edit
    3 adds in trunk/LayoutTests

2009-12-10 Patrik Persson <patrik.j.persson@ericsson.com>

Reviewed by Adam Barth.

Test case to verify that sandboxing 'about:blank' does not make
the containing frame sandboxed.

https://bugs.webkit.org/show_bug.cgi?id=21288#c28

  • fast/frames/resources/sandboxed-iframe-about-blank.html: Added.
  • fast/frames/sandboxed-iframe-about-blank-expected.txt: Added.
  • fast/frames/sandboxed-iframe-about-blank.html: Added.
09:13 Changeset [51950] by senorblanco@chromium.org
  • 2 edits in trunk/WebCore

Fix for alpha blending in SVG Filters on Chromium/skia.
https://bugs.webkit.org/show_bug.cgi?id=32378

Reviewed by Dirk Schulze.

Covered by LayoutTests/svg/filters/feGaussianBlur.svg and others.

  • platform/graphics/skia/ImageBufferSkia.cpp:

(WebCore::ImageBuffer::platformTransformColorSpace):

08:44 Changeset [51949] by pfeldman@chromium.org
  • 2 edits in trunk/WebKit/chromium

2009-12-10 Ilya Tikhonovsky <loislo@google.com>

Reviewed by Pavel Feldman.

Chromium: Expose devtools agent API for inspector layout tests.

https://bugs.webkit.org/show_bug.cgi?id=32379

  • public/WebDevToolsAgent.h: (WebKit::WebDevToolsAgent::setTimelineProfilingEnabled): (WebKit::WebDevToolsAgent::evaluateInWebInspector):
06:29 Changeset [51948] by kov@webkit.org
  • 8 edits
    2 adds in trunk

Reviewed by Xan Lopez.

[GTK] Should provide an API to control the IconDatabase
https://bugs.webkit.org/show_bug.cgi?id=32334

First step towards a full IconDatabase API. This provides
notification for pages with favicons, and lets clients handle
them.

  • WebCoreSupport/FrameLoaderClientGtk.cpp: (WebKit::FrameLoaderClient::registerForIconNotification): (WebKit::FrameLoaderClient::dispatchDidReceiveIcon):
  • tests/resources/blank.ico: Added.
  • tests/testwebview.c: Added. (server_callback): (idle_quit_loop_cb): (icon_uri_changed_cb): (icon_loaded_cb): (test_webkit_web_view_icon_uri): (main):
  • webkit/webkitprivate.cpp: (closeIconDatabaseOnExit): (webkit_init):
  • webkit/webkitprivate.h:
  • webkit/webkitwebview.cpp: (webkit_web_view_get_property): (webkit_web_view_finalize): (webkit_web_view_class_init): (webkit_web_view_get_icon_uri):
  • webkit/webkitwebview.h:
05:55 Changeset [51947] by eric@webkit.org
  • 4 edits in trunk

2009-12-10 Eric Z. Ayers <zundel@google.com>

Reviewed by Pavel Feldman.

DumpRenderTree has been patched so that the timeline-paint.html
unit test now runs under GTK+.

https://bugs.webkit.org/show_bug.cgi?id=31729

  • platform/gtk/Skipped:

2009-12-10 Eric Z. Ayers <zundel@google.com>

Reviewed by Pavel Feldman.

Implements displayWebView() to force an invalidation and repaint.
This fixes a problem running the timeline-paint.html unit test.

https://bugs.webkit.org/show_bug.cgi?id=31729

  • DumpRenderTree/gtk/DumpRenderTree.cpp: (displayWebView):
04:53 Changeset [51946] by pfeldman@chromium.org
  • 10 edits in trunk/WebCore

2009-12-10 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Add context menu actions for DOM tree.

https://bugs.webkit.org/show_bug.cgi?id=32348

  • English.lproj/localizedStrings.js:
  • inspector/front-end/ContextMenu.js: (WebInspector.ContextMenu): (WebInspector.ContextMenu.prototype.show): (WebInspector.ContextMenu.prototype.appendItem): (WebInspector.ContextMenu.prototype.appendSeparator): (WebInspector.ContextMenu.prototype._itemSelected): (WebInspector.contextMenuItemSelected): (WebInspector.contextMenuCleared):
  • inspector/front-end/ElementsTreeOutline.js: (WebInspector.ElementsTreeOutline.prototype._onmouseout): (WebInspector.ElementsTreeOutline.prototype.populateContextMenu): (WebInspector.ElementsTreeElement.prototype.ondblclick): (WebInspector.ElementsTreeElement.prototype._startEditingFromEvent): (WebInspector.ElementsTreeElement.prototype._populateTagContextMenu): (WebInspector.ElementsTreeElement.prototype._populateTextContextMenu): (WebInspector.ElementsTreeElement.prototype._startEditing): (WebInspector.ElementsTreeElement.prototype._addNewAttribute):
  • inspector/front-end/ObjectPropertiesSection.js: (WebInspector.ObjectPropertyTreeElement.prototype.ondblclick):
  • inspector/front-end/ResourceView.js: (WebInspector.ResourceView.prototype._toggleURLdecoding):
  • inspector/front-end/ResourcesPanel.js: (WebInspector.ResourceSidebarTreeElement.prototype.ondblclick):
  • inspector/front-end/StylesSidebarPane.js: (WebInspector.StylePropertyTreeElement.prototype):
  • inspector/front-end/inspector.js: (WebInspector.contextMenuEventFired):
  • inspector/front-end/treeoutline.js: (TreeElement.treeElementDoubleClicked):
04:05 Changeset [51945] by zimmermann@webkit.org
  • 20 edits in trunk/WebCore

2009-12-09 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Oliver Hunt.

Filters contain some leaks in untested code
https://bugs.webkit.org/show_bug.cgi?id=32325

Fix obvious leak in SVGFE*Lighting classes. Implement the create() idiom for
all classes in svg/graphics, that were missing it. The lighting filters aren't
implemented so far, but the associated FilterEffect objects are build, which created
these leaks.

This removes the SVG related failures in the leaks bot.

04:04 Changeset [51944] by zimmermann@webkit.org
  • 3 edits in trunk/LayoutTests

2009-12-10 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Skip two tests on gtk/qt that generate new results.

  • platform/gtk/Skipped:
  • platform/qt/Skipped:
02:13 QtBackLog edited by zecke@selfish.org
(diff)
02:13 QtBackLog edited by zecke@selfish.org
(diff)
01:21 Changeset [51943] by eric@webkit.org
  • 3 edits
    2 adds in trunk

2009-12-10 Kenneth Russell <kbr@google.com>

Reviewed by Oliver Hunt.

[Chromium] SporeViewer demo doesn't work in Chromium
https://bugs.webkit.org/show_bug.cgi?id=32364

Test: fast/canvas/webgl/bug-32364.html

  • fast/canvas/webgl/bug-32364-expected.txt: Added.
  • fast/canvas/webgl/bug-32364.html: Added.

2009-12-10 Kenneth Russell <kbr@google.com>

Reviewed by Oliver Hunt.

[Chromium] SporeViewer demo doesn't work in Chromium
https://bugs.webkit.org/show_bug.cgi?id=32364

Test: fast/canvas/webgl/bug-32364.html

  • bindings/v8/custom/V8WebGLRenderingContextCustom.cpp: (WebCore::CALLBACK_FUNC_DECL): (WebCore::vertexAttribAndUniformHelperf): (WebCore::uniformHelperi):
00:33 Changeset [51942] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-10 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Convert more commands to StepSequences
https://bugs.webkit.org/show_bug.cgi?id=32362

We should eventually convert all the commands, but I'm starting with
the easy ones.

  • Scripts/modules/commands/download.py:
00:21 Changeset [51941] by xan@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-10 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

[GTK] editing/selection/shrink-selection-after-shift-pagedown.html failing
https://bugs.webkit.org/show_bug.cgi?id=31103

Give focus to the webviews when we create them, since some tests expect this.

  • DumpRenderTree/gtk/DumpRenderTree.cpp: (main):
00:18 Changeset [51940] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

bugzilla-tool "builders are red" error should tell you which builders
https://bugs.webkit.org/show_bug.cgi?id=32211

  • Scripts/modules/buildsteps.py:

12/09/09:

23:47 Changeset [51939] by mjs@apple.com
  • 4 edits
    3 adds in trunk

2009-12-09 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver Hunt.

Google reader gets stuck in the "Loading..." state and does not complete
https://bugs.webkit.org/show_bug.cgi?id=32256
<rdar://problem/7456388>

  • jit/JITArithmetic.cpp: (JSC::JIT::emitSlow_op_jless): Fix some backward branches.

2009-12-09 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver Hunt.

Test for: Google reader gets stuck in the "Loading..." state and does not complete
https://bugs.webkit.org/show_bug.cgi?id=32256

  • fast/js/codegen-jless-expected.txt: Added.
  • fast/js/codegen-jless.html: Added.
  • fast/js/script-tests/codegen-jless.js: Added.
23:29 Changeset [51938] by ossy@webkit.org
  • 2 edits in trunk/LayoutTests

[Qt] Unreviewed. Skip a timeouter flakey test.

  • platform/qt/Skipped: fast/frames/sandboxed-iframe-scripting.html added.
23:22 Changeset [51937] by yurys@chromium.org
  • 2 edits in trunk/WebKit/chromium

2009-12-09 Yury Semikhatsky <yurys@chromium.org>

Unreviewed, build fix.

Change forward declaration of WebContextMenuData from class to
struct to match its definition. The bug was introduced by
http://trac.webkit.org/changeset/51874

  • src/ContextMenuClientImpl.h:
22:07 Changeset [51936] by bdakin@apple.com
  • 5 edits
    8 adds in trunk

WebCore: Fix for https://bugs.webkit.org/show_bug.cgi?id=32346 SVG property
-webkit-shadow should apply shadow on the result after compositing
-and corresponding-
<rdar://problem/7389404>

Reviewed by Oliver Hunt.

Set a transparency layer when setting a shadow to apply the shadow
to the composite.

  • rendering/SVGRenderSupport.cpp:

(WebCore::SVGRenderBase::prepareToRenderSVGContent):
(WebCore::SVGRenderBase::finishRenderSVGContent):

LayoutTests: Tests for https://bugs.webkit.org/show_bug.cgi?id=32346 SVG
property -webkit-shadow should apply shadow on the result after
compositing
-and corresponding-
<rdar://problem/7389404>

Reviewed by Oliver Hunt.

New tests:

  • platform/mac/svg/css/composite-shadow-example-expected.checksum: Added.
  • platform/mac/svg/css/composite-shadow-example-expected.png: Added.
  • platform/mac/svg/css/composite-shadow-example-expected.txt: Added.
  • platform/mac/svg/css/composite-shadow-with-opacity-expected.checksum: Added.
  • platform/mac/svg/css/composite-shadow-with-opacity-expected.png: Added.
  • platform/mac/svg/css/composite-shadow-with-opacity-expected.txt: Added.
  • svg/css/composite-shadow-example.html: Added.
  • svg/css/composite-shadow-with-opacity.html: Added.

New and improved results:

  • platform/mac/svg/css/group-with-shadow-expected.checksum:
  • platform/mac/svg/css/group-with-shadow-expected.png:
19:32 Changeset [51935] by ap@apple.com
  • 5 edits in trunk

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=32332
WebSocket events should be dispatched synchronously

Updated websocket/tests/simple to test for the new behavior.

When Web Sockets API says that events should be queued for async dispatch, it means something
different. We should keep this in mind when dealing with other HTML5-related specs.

The model for HTML5 is that code running in response to network events (e.g. WebSocket or
XMLHttpRequest algorithms) runs in a separate thread of execution, and thus needs to post
async events as its only way to communicate with client code. As long as network events are
queued themselves (as they are in WebKit), there is no need to queue JS events for async
dispatch.

  • websockets/WebSocket.cpp: (WebCore::WebSocket::didConnect): (WebCore::WebSocket::didReceiveMessage): (WebCore::WebSocket::didClose):
19:30 Changeset [51934] by ap@apple.com
  • 3 edits in trunk/WebCore

Reviewed by Oliver Hunt.

https://bugs.webkit.org/show_bug.cgi?id=32355
Assertion failure when opening a WebSocket connection

I couldn't make a reliable test for this. Once the test from bug 32299 is landed, it
will provide partial coverage, as I was frequently seeing the assertion failure with it.

  • platform/network/SocketStreamHandleBase.cpp: (WebCore::SocketStreamHandleBase::send): It's not an error if zero if returned from platformSend() - it just means that nothing could be pushed down to the network layer, and all data was queued for later.
  • platform/network/cf/SocketStreamHandleCFNet.cpp: (WebCore::SocketStreamHandle::readStreamCallback): This will no longer happen with SocketStream, but a client can potentially destroy the handle from any callback, so we need to check that this didn't happen. (WebCore::SocketStreamHandle::writeStreamCallback): Ditto.
17:44 Changeset [51933] by barraclough@apple.com
  • 6 edits in trunk

JavaScriptCore: https://bugs.webkit.org/show_bug.cgi?id=32228
Make destruction of ropes non-recursive to prevent stack exhaustion.
Also, pass a UString& into initializeFiber rather than a Ustring::Rep*,
since the Rep is not being ref counted this could result in usage of a
Rep with refcount zero (where the Rep comes from a temporary UString
returned from a function).

Reviewed by Oliver Hunt.

  • runtime/JSString.cpp:

(JSC::JSString::Rope::destructNonRecursive):
(JSC::JSString::Rope::~Rope):

  • runtime/JSString.h:

(JSC::JSString::Rope::initializeFiber):

  • runtime/Operations.h:

(JSC::concatenateStrings):

LayoutTests: https://bugs.webkit.org/show_bug.cgi?id=32228
Reenabling tests.

Reviewed by Oliver Hunt.

  • platform/win/Skipped:
16:58 Changeset [51932] by eric@webkit.org
  • 5 edits in trunk

2009-12-09 Marwan Al Jubeh <marwan.aljubeh@gmail.com>

Reviewed by Adam Roben.

Fixes: https://bugs.webkit.org/show_bug.cgi?id=31228
Set the WebKitOutputDir, WebKitLibrariesDir and Cygwin environment variables automatically
in Windows as part of running update_webkit.

  • building/build.html:
    • removed the reference to forgetting to set environment variables as a common source of errors on Windows. This is because this patch would make it unnecessary for the user to set these variables manually.

2009-12-09 Marwan Al Jubeh <marwan.aljubeh@gmail.com>

Reviewed by Adam Roben.

Fixes: https://bugs.webkit.org/show_bug.cgi?id=31228
Set the WebKitOutputDir, WebKitLibrariesDir and Cygwin environment variables automatically
in Windows as part of running update_webkit.

  • Scripts/update-webkit:
    • Run setupAppleWinEnv() on Apple's Windows port.
  • Scripts/webkitdirs.pm:
    • Added functions that return the source directory, libraries directory and default build directory on Windows.
    • Added isWindowsNT() which tests if the current Windows version is from the Windows NT family.
    • Implemented setupAppleWinEnv() which sets the environment variables WebKitOutputDir, WebKitLibrariesDir and Cygwin to their desired values.
16:23 Changeset [51931] by weinig@apple.com
  • 3 edits in trunk/WebCore

Roll out 51919 and 51920. They were incorrect and unnecessary right now.

  • platform/mac/WebCoreObjCExtras.mm:

(WebCoreObjCScheduleDeallocateOnMainThread):

  • platform/network/mac/NetworkStateNotifierMac.cpp:

(WebCore::NetworkStateNotifier::NetworkStateNotifier):

15:32 Changeset [51930] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-12-09 Steve Block <steveblock@google.com>

Reviewed by Darin Fisher.

Adds wtf/StdLibExtras.h include for DEFINE_STATIC_LOCAL in V8 ScriptState.
https://bugs.webkit.org/show_bug.cgi?id=32330

Build fix only, no new tests.

  • bindings/v8/ScriptState.cpp: Modified. Adds wtf/StdLibExtras.h include.
15:08 Changeset [51929] by eric@webkit.org
  • 2 edits in trunk/WebKitSite

2009-12-09 Chris Jerdonek <chris.jerdonek@gmail.com>

Reviewed by Darin Adler.

Added the "using std::foo" rule to the coding style guidelines.

https://bugs.webkit.org/show_bug.cgi?id=32301

Added to the coding style guidelines the "using std::foo" rule
checked by the check-webkit-style script. Also clarified the
"using" statement guidelines that apply to implementation files.

  • coding/coding-style.html:
15:01 Changeset [51928] by eric@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-12-09 Zoltan Herczeg <zherczeg@inf.u-szeged.hu>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=31930

Update to r51457. ASSERTs changed to COMPILE_ASSERTs.
The speedup is 25%.

  • runtime/JSGlobalData.cpp: (JSC::VPtrSet::VPtrSet):
14:54 Changeset [51927] by eric@webkit.org
  • 3 edits in trunk/WebCore

2009-12-09 Steve Block <steveblock@google.com>

Reviewed by Darin Adler.

Adds ENABLE(INSPECTOR) guards around script binding methods that use types
defined only when INSPECTOR is enabled.
https://bugs.webkit.org/show_bug.cgi?id=32328

Build fix only, no new tests.

  • bindings/js/ScriptValue.cpp: Modified. Added ENABLE(INSPECTOR) guard to ScriptValue::quarantineValue.
  • bindings/v8/ScriptObject.cpp: Modified. Added ENABLE(INSPECTOR) guard to some overloads of ScriptGlobalObject::set.
14:47 Changeset [51926] by eric@webkit.org
  • 10 edits
    2 adds in trunk

2009-12-09 Steve Block <steveblock@google.com>

Reviewed by Adam Barth.

Adds Android Makefiles for building with V8.
https://bugs.webkit.org/show_bug.cgi?id=32278

  • Android.mk: Modified. Includes Makefiles for V8.

2009-12-09 Steve Block <steveblock@google.com>

Reviewed by Adam Barth.

Updates Android Makefiles with latest additions.
https://bugs.webkit.org/show_bug.cgi?id=32278

  • Android.mk: Modified.
  • Android.v8.wtf.mk: Modified.

2009-12-09 Steve Block <steveblock@google.com>

Reviewed by Adam Barth.

Adds Android Makefiles for building with V8.
Also updates existing Android Makefiles with latest additions.
https://bugs.webkit.org/show_bug.cgi?id=32278

Build fix only, no new tests.

  • Android.derived.jscbindings.mk: Modified.
  • Android.derived.mk: Modified.
  • Android.derived.v8bindings.mk: Added.
  • Android.jscbindings.mk: Modified.
  • Android.mk: Modified.
  • Android.v8bindings.mk: Added.
14:35 Changeset [51925] by steveblock@google.com
  • 2 edits in trunk/WebCore

Bug 32338 - [Chromium] Fix Chromium builder by including missing GeolocationServiceChromium
https://bugs.webkit.org/show_bug.cgi?id=32338

Patch by Jonathan Dixon <joth@chromium.org> on 2009-12-09
Reviewed by Dimitri Glazkov.

Fix build break: re-include file dropped out by merge in http://trac.webkit.org/changeset/51681/trunk/WebCore/WebCore.gypi

  • WebCore.gypi: Add missing GeolocationServiceChromium.cpp
14:32 Changeset [51924] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-12-09 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Brady Eidson.

Fixed a typo in http://trac.webkit.org/changeset/51644 that broke
the QWebPage autotest.

No new test required.

  • loader/FrameLoader.cpp: (WebCore::FrameLoader::navigateToDifferentDocument):
14:25 Changeset [51923] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-09 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

exception thrown when running apply-patches
https://bugs.webkit.org/show_bug.cgi?id=32344

The update step now takes a port option. Once we finish the Steps
refactoring, we won't have to worry about this kind of bug again.

  • Scripts/modules/commands/download.py:
14:18 Changeset [51922] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-12-09 Robert Hogan <robert@roberthogan.net>

Reviewed by Eric Seidel.

Exclude JSSVG*.cpp generated files from --minimal and --no-svg builds.

https://bugs.webkit.org/show_bug.cgi?id=32286

  • WebCore.pro:
14:11 Changeset [51921] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-12-09 Philippe Normand <pnormand@igalia.com>

Reviewed by Eric Carlson.

Ogg mimetypes are incorrect
https://bugs.webkit.org/show_bug.cgi?id=27113

The ogg extension is handled by audio/ogg instead of
application/ogg. See
http://wiki.xiph.org/MIME_Types_and_File_Extensions

  • platform/MIMETypeRegistry.cpp: (WebCore::TypeExtensionPair::):
13:57 Changeset [51920] by weinig@apple.com
  • 2 edits in trunk/WebCore

Use the current run loop instead of the main runloop for the NetworkStateNotifier.

Reviewed by Anders Carlsson.

  • platform/network/mac/NetworkStateNotifierMac.cpp:

(WebCore::NetworkStateNotifier::NetworkStateNotifier):

13:55 Changeset [51919] by weinig@apple.com
  • 2 edits in trunk/WebCore

Use isMainThread() helper instead of pthread_main_np() != 0.

Reviewed by Anders Carlsson.

  • platform/mac/WebCoreObjCExtras.mm:

(WebCoreObjCScheduleDeallocateOnMainThread):

13:37 Changeset [51918] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Skip a failing test that needs DRT support to override standard
preferences.

  • platform/gtk/Skipped:
13:36 Changeset [51917] by darin@chromium.org
  • 5 edits in trunk/WebKit/chromium

2009-12-09 Darin Fisher <darin@chromium.org>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=32324
[Chromium] Suppress WebFrameClient callbacks from a detached frame.

This change means that we no longer need the ClientHandle class.
FrameLoaderClient::detachedFromParent3() is called on each frame
in the frame tree from within frameDetached().

Test: http/tests/loading/gmail-assert-on-load.html

  • src/FrameLoaderClientImpl.cpp: (WebKit::FrameLoaderClientImpl::detachedFromParent3):
  • src/WebFrameImpl.cpp: (WebKit::WebFrameImpl::create): (WebKit::WebFrameImpl::WebFrameImpl): (WebKit::WebFrameImpl::createChildFrame):
  • src/WebFrameImpl.h: (WebKit::WebFrameImpl::client): (WebKit::WebFrameImpl::dropClient):
  • src/WebViewImpl.cpp: (WebKit::WebViewImpl::close):
13:31 Changeset [51916] by weinig@apple.com
  • 11 edits in trunk/WebCore

Add some #ifdefs to allow us to experiment with a single NSView mac WebKit.

Reviewed by Anders Carlsson.

  • page/Chrome.h:
  • page/EventHandler.h:
  • page/mac/ChromeMac.mm:
  • page/mac/DragControllerMac.mm:

(WebCore::DragController::dragOperation):

  • page/mac/EventHandlerMac.mm:

(WebCore::EventHandler::eventLoopHandleMouseDragged):
(WebCore::EventHandler::eventActivatedView):
(WebCore::EventHandler::passMousePressEventToSubframe):
(WebCore::EventHandler::passMouseMoveEventToSubframe):
(WebCore::EventHandler::passMouseReleaseEventToSubframe):
(WebCore::EventHandler::passWheelEventToWidget):
(WebCore::EventHandler::focusDocumentView):
(WebCore::EventHandler::passWidgetMouseDownEventToWidget):
(WebCore::EventHandler::createDraggingClipboard):
(WebCore::isKeyboardOptionTab):
(WebCore::EventHandler::invertSenseOfTabsToLinks):
(WebCore::EventHandler::tabsToAllControls):
(WebCore::EventHandler::needsKeyboardEventDisambiguationQuirks):
(WebCore::EventHandler::accessKeyModifiers):

  • platform/ScrollView.cpp:

(WebCore::ScrollView::wheelEvent):

  • platform/Widget.cpp:
  • platform/Widget.h:
  • platform/mac/ScrollViewMac.mm:
  • platform/mac/WidgetMac.mm:

(WebCore::Widget::Widget):
(WebCore::Widget::~Widget):
(WebCore::Widget::show):
(WebCore::Widget::hide):
(WebCore::Widget::setCursor):
(WebCore::Widget::paint):
(WebCore::Widget::setFocus):
(WebCore::Widget::setIsSelected):
(WebCore::Widget::frameRect):
(WebCore::Widget::setFrameRect):

12:36 Changeset [51915] by eric@webkit.org
  • 4 edits in trunk/WebCore

2009-12-09 Michael Nordman <michaeln@google.com>

Reviewed by Dimitri Glazkov.

Chromium build fix.

No new tests.

  • bindings/v8/DOMData.cpp
  • bindings/v8/V8DOMWrapper.cpp
  • bindings/v8/custom/V8WebGLRenderingContextCustom.cpp: (WebCore::vertexAttribAndUniformHelperf): (WebCore::uniformHelperi): (WebCore::uniformMatrixHelper):
12:31 Changeset [51914] by weinig@apple.com
  • 2 edits in trunk/JavaScriptCore

Fix a bug found while trying to compile JavaScriptCore with clang++.

Reviewed by Gavin Barraclough.

  • yarr/RegexPattern.h:

(JSC::Yarr::PatternTerm::PatternTerm): Don't self assign here. Use false instead.

12:26 Changeset [51913] by bfulgham@webkit.org
  • 2 edits in trunk/WebKit/win

Revert incorrect commit-box update r51911.
It mistakenly set the selectAll implementation in the
deslectAll method.

  • WebFrame.cpp:

(WebFrame::deselectAll):

12:14 Changeset [51912] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-12-09 Søren Gjesse <sgjesse@chromium.org>

Reviewed by Pavel Feldman.

[V8] Isolated world reuse failure
https://bugs.webkit.org/show_bug.cgi?id=32306

Fixed bug introduced in r51407: (http://trac.webkit.org/changeset/51407) which caused layout test
LayoutTests/http/tests/security/isolatedWorld/world-reuse.html to fail.

  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::setInjectedScriptContextDebugId): (WebCore::V8Proxy::contextDebugId):
12:06 Changeset [51911] by eric@webkit.org
  • 2 edits in trunk/WebKit/win

2009-12-09 Brent Fulgham <bfulgham@webkit.org>

Reviewed by Darin Adler.

Provide an implementation for 'selectAll'
https://bugs.webkit.org/show_bug.cgi?id=32296

  • WebFrame.cpp: (WebFrame::selectAll): Implement "SelectAll" command.
11:59 Changeset [51910] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-12-09 Patrick Scott <phanna@email.unc.edu>

Fix the build with ENABLE_ORIENTATION_EVENTS
https://bugs.webkit.org/show_bug.cgi?id=32321

  • page/Frame.cpp: (WebCore::Frame::sendOrientationChangeEvent):
11:52 Changeset [51909] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-12-09 Avi Drissman <avi@chromium.org>

Reviewed by Darin Fisher.

Chromium tickmarks in scrollbar now UX approved.
https://bugs.webkit.org/show_bug.cgi?id=32069

  • platform/chromium/ScrollbarThemeChromiumMac.mm: (WebCore::ScrollbarThemeChromiumMac::paint):
11:00 Changeset [51908] by andersca@apple.com
  • 2 edits in trunk/JavaScriptCore

Attempt to fix the Windows build.

Reviewed by Sam Weinig.

  • wtf/FastMalloc.h:
10:53 Changeset [51907] by bfulgham@webkit.org
  • 2 edits in trunk/WebKit/win

Provide an implementation for 'selectAll'
https://bugs.webkit.org/show_bug.cgi?id=32296

Reviewed by Darin Adler.

  • WebFrame.cpp:

(WebFrame::selectAll): Implement "SelectAll" command.

10:44 Changeset [51906] by cmarrin@apple.com
  • 9 edits in trunk

Delay load DLLs for accelerated compositing
https://bugs.webkit.org/show_bug.cgi?id=31856


Add logic to turn off accelerated compositing if d3d9 or QuartzCore
DLLs are not present.

10:40 Changeset [51905] by andersca@apple.com
  • 4 edits in trunk/JavaScriptCore

Fix some things found while trying to compile JavaScriptCore with clang++.

Reviewed by Sam Weinig.

  • wtf/FastMalloc.h:

Add correct exception specifications for the allocation/deallocation operators.

  • wtf/Vector.h:
  • wtf/VectorTraits.h:

Fix a bunch of struct/class mismatches.

10:22 Changeset [51904] by darin@chromium.org
  • 2 edits in trunk/WebKit/chromium

2009-12-08 Stuart Morgan <stuartmorgan@chromium.org>

Reviewed by Darin Fisher.

Include clickCount when doing MouseEvent -> WebMouseEvent conversion
in Chromium API.

https://bugs.webkit.org/show_bug.cgi?id=32289

  • src/WebInputEventConversion.cpp: (WebKit::WebMouseEventBuilder::WebMouseEventBuilder):
10:11 Changeset [51903] by abarth@webkit.org
  • 4 edits in trunk/WebKitTools

2009-12-09 Adam Barth <abarth@webkit.org>

Unreviewed. Fix ScriptError includes. I don't understand why these
didn't throw during unit testing...

  • Scripts/modules/commands/download.py:
  • Scripts/modules/logging_unittest.py:
  • Scripts/modules/workqueue.py:
09:43 Changeset [51902] by japhet@chromium.org
  • 13 edits
    2 moves in trunk/WebCore

2009-12-09 Nate Chapin <japhet@chromium.org>

Reviewed by Darin Adler.

Rename dom/ClassNames to SpaceSplitString and update build files.

https://bugs.webkit.org/show_bug.cgi?id=32250

  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::matchRules):
  • dom/ClassNames.cpp: Removed.
  • dom/ClassNames.h: Removed.
  • dom/ClassNodeList.h:
  • dom/NamedMappedAttrMap.h: (WebCore::NamedMappedAttrMap::classNames):
  • dom/SpaceSplitString.cpp: Copied from WebCore/dom/ClassNames.cpp. (WebCore::SpaceSplitStringData::createVector): (WebCore::SpaceSplitStringData::containsAll):
  • dom/SpaceSplitString.h: Copied from WebCore/dom/ClassNames.h. (WebCore::SpaceSplitStringData::SpaceSplitStringData): (WebCore::SpaceSplitString::SpaceSplitString): (WebCore::SpaceSplitString::set): (WebCore::SpaceSplitString::containsAll):
  • dom/StyledElement.h: (WebCore::StyledElement::classNames):
  • html/HTMLAnchorElement.cpp: (WebCore::HTMLAnchorElement::setRel):
09:28 Changeset [51901] by senorblanco@chromium.org
  • 8 edits in trunk/WebCore

Enable SVG filters in Chromium build.
https://bugs.webkit.org/show_bug.cgi?id=32323

Reviewed by Darin Fisher.

Covered by SVG filter layout tests.

  • WebCore.gyp/WebCore.gyp:
  • WebCore.gypi:
  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/v8/DerivedSourcesAllInOne.cpp:
  • bindings/v8/V8DOMWrapper.cpp:
  • bindings/v8/V8Index.cpp:
  • bindings/v8/V8Index.h:
08:59 QtWebKit/CodeCamp200912 edited by zecke@selfish.org
mention the performance talks (diff)
08:50 bbandix_drt_tools.pdf attached to QtWebKit/CodeCamp200912 by abecsi@inf.u-szeged.hu
07:26 Changeset [51900] by yurys@chromium.org
  • 2 edits in trunk/WebKit/chromium

2009-12-09 Yury Semikhatsky <yurys@chromium.org>

Unreviewed, build fix.

Add return statement to default implementation of hostIdentifier method
introduced in http://trac.webkit.org/changeset/51897

  • public/WebDevToolsAgentClient.h: (WebKit::WebDevToolsAgentClient::hostIdentifier):
07:05 Changeset [51899] by yurys@chromium.org
  • 3 edits in trunk/WebKit/chromium

2009-12-09 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Add declaration of static methods that allow to dispatch
DevTools messages on the IO thread and accept WebDevToolsMessageData
argument.

https://bugs.webkit.org/show_bug.cgi?id=32314

  • public/WebDevToolsAgent.h:
  • public/WebDevToolsAgentClient.h:
06:49 WkMem.pdf attached to QtWebKit/CodeCamp200912 by zherczeg@inf.u-szeged.hu
Meamory measurement using freya (valgrind)
05:27 Changeset [51898] by ossy@webkit.org
  • 2 edits in trunk/LayoutTests

[Qt] Skip a failing test introduced in r51877.

  • platform/qt/Skipped: fast/loader/input-element-page-cache-crash.html added.
05:19 Changeset [51897] by yurys@chromium.org
  • 6 edits
    1 copy in trunk/WebKit/chromium

2009-12-09 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Introduce a class for devtools message data

https://bugs.webkit.org/show_bug.cgi?id=32314

  • WebKit.gyp:
  • public/WebDevToolsAgent.h:
  • public/WebDevToolsAgentClient.h: (WebKit::WebDevToolsAgentClient::sendMessageToFrontend): (WebKit::WebDevToolsAgentClient::forceRepaint): (WebKit::WebDevToolsAgentClient::hostIdentifier): (WebKit::WebDevToolsAgentClient::runtimeFeatureStateChanged):
  • public/WebDevToolsFrontend.h:
  • public/WebDevToolsFrontendClient.h: (WebKit::WebDevToolsFrontendClient::sendMessageToAgent):
  • public/WebDevToolsMessageData.h: Added.
05:15 ARM-JIT.pdf attached to QtWebKit/CodeCamp200912 by zherczeg@inf.u-szeged.hu
ARM-JIT in nutshell
05:04 Changeset [51896] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

2009-12-09 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Fix typo in the Skipped list s/.htm/.html/ in the last test I added.

  • platform/mac-snowleopard/Skipped:
03:34 QtWebKit/CodeCamp200912 edited by vestbo@webkit.org
(diff)
02:09 QtWebKit/CodeCamp200912 edited by akiss@inf.u-szeged.hu
(diff)
02:05 akiss-uszintro.pdf attached to QtWebKit/CodeCamp200912 by akiss@inf.u-szeged.hu
Intro to #u-szeged
01:54 WebKit Team edited by xan@webkit.org
(diff)
01:39 Changeset [51895] by abarth@webkit.org
  • 4 edits
    1 delete in trunk/WebKitTools

2009-12-09 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Remove unused PatchCollection class
https://bugs.webkit.org/show_bug.cgi?id=32312

It's dead code.

  • Scripts/modules/patchcollection.py:
  • Scripts/modules/patchcollection_unittest.py: Removed.
01:29 Changeset [51894] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-09 Adam Barth <abarth@webkit.org>

Unreviewed "build" fix. CheckStyle needs a --no-upate option.

  • Scripts/modules/commands/download.py:
01:24 Changeset [51893] by abarth@webkit.org
  • 4 edits
    1 add in trunk/WebKitTools

2009-12-09 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Convert Build to use Sequence
https://bugs.webkit.org/show_bug.cgi?id=32310

So much prettier.

  • Scripts/modules/buildsteps.py:
  • Scripts/modules/commands/download.py:
  • Scripts/modules/landingsequence.py:
  • Scripts/modules/stepsequence.py: Added.
01:17 Changeset [51892] by oliver@apple.com
  • 3 edits in trunk/WebCore

Build fix

00:26 Changeset [51891] by abarth@webkit.org
  • 1 edit
    1 add in trunk/WebKitTools

2009-12-09 Adam Barth <abarth@webkit.org>

Add missing file.

  • Scripts/modules/executive.py: Added.
00:24 Changeset [51890] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-09 Adam Barth <abarth@webkit.org>

Unreviewed "build" fix.

  • Scripts/modules/landingsequence.py:
00:20 Changeset [51889] by abarth@webkit.org
  • 12 edits
    1 delete in trunk/WebKitTools

2009-12-09 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Implement abstract Steps
https://bugs.webkit.org/show_bug.cgi?id=32212

This is a fairly disruptive change that refactors how we build
commands. Instead of using a landing sequence, we can now assemble a
sequence of steps directly. We still use the landing sequence in the
interim, but this will be removed soon.

  • Scripts/bugzilla-tool:
  • Scripts/modules/buildsteps.py:
  • Scripts/modules/commands/download.py:
  • Scripts/modules/commands/early_warning_system.py:
  • Scripts/modules/commands/queues.py:
  • Scripts/modules/commands/queues_unittest.py:
  • Scripts/modules/landingsequence.py:
  • Scripts/modules/mock_bugzillatool.py:
  • Scripts/modules/processutils.py: Removed.
  • Scripts/modules/scm.py:
  • Scripts/modules/scm_unittest.py:
  • Scripts/modules/webkitport.py:

12/08/09:

23:51 Changeset [51888] by eric@webkit.org
  • 11 edits in trunk/WebKitTools

2009-12-08 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

run_command and ScriptError should move into processutils.py
https://bugs.webkit.org/show_bug.cgi?id=32305

Turns out there are a zillion callers to run_command.

  • Scripts/modules/commands/download.py:
  • Scripts/modules/commands/early_warning_system.py:
  • Scripts/modules/commands/queues.py:
  • Scripts/modules/landingsequence.py:
  • Scripts/modules/logging_unittest.py:
  • Scripts/modules/processutils.py:
  • Scripts/modules/scm.py:
  • Scripts/modules/scm_unittest.py:
  • Scripts/modules/workqueue.py:
  • Scripts/modules/workqueue_unittest.py:
19:43 Changeset [51887] by mjs@apple.com
  • 1 edit in trunk/JavaScriptCore/ChangeLog

Remove stray conflict marker.

19:39 Changeset [51886] by ukai@chromium.org
  • 7 edits
    3 adds in trunk

2009-12-08 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Darin Adler.

WebSocket allow space (U+0020) in sub protocol name.
https://bugs.webkit.org/show_bug.cgi?id=32266

Skipped sub-protocol-with-space.html for now until new pywebsocket is
landed.

  • websocket/tests/bad-sub-protocol-expected.txt:
  • websocket/tests/script-tests/bad-sub-protocol.js:
  • websocket/tests/script-tests/sub-protocol-with-space.js: Added.
  • websocket/tests/sub-protocol-with-space-expected.txt: Added.
  • websocket/tests/sub-protocol-with-space.html: Added.

2009-12-08 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Darin Adler.

WebSocket allow space (U+0020) in sub protocol name.
https://bugs.webkit.org/show_bug.cgi?id=32266

Test: websocket/tests/sub-protocol-with-space.html

  • websockets/WebSocket.cpp: (WebCore::isValidProtocolString):
19:20 Changeset [51885] by ukai@chromium.org
  • 1 edit
    3 adds in trunk/LayoutTests

2009-12-08 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Darin Adler.

Add tests that WebSocket-Protocol: is ignored if protocol was not specified.
https://bugs.webkit.org/show_bug.cgi?id=32267

  • websocket/tests/script-tests/websocket-protocol-ignored.js: Added.
  • websocket/tests/websocket-protocol-ignored-expected.txt: Added.
  • websocket/tests/websocket-protocol-ignored.html: Added.
18:38 Changeset [51884] by eric@webkit.org
  • 14 edits
    7 adds in trunk

2009-12-08 Peterson Trethewey <petersont@google.com>

Reviewed by Oliver Hunt.

Implement WebGLUniformLocation and change API to use it.
https://bugs.webkit.org/show_bug.cgi?id=31173

  • fast/canvas/webgl/resources/structUniformShader.vert: Added.
  • fast/canvas/webgl/script-tests/uniform-location.js: Added.
  • fast/canvas/webgl/uniform-location-expected.txt: Added.
  • fast/canvas/webgl/uniform-location.html: Added.

2009-12-08 Peterson Trethewey <petersont@google.com>

Reviewed by Oliver Hunt.

Implement WebGLUniformLocation and change API to use it.
https://bugs.webkit.org/show_bug.cgi?id=31173

Test: fast/canvas/webgl/uniform-location.html

  • DerivedSources.make:
  • WebCore.gypi:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSWebGLRenderingContextCustom.cpp: (WebCore::JSWebGLRenderingContext::getProgramParameter): (WebCore::JSWebGLRenderingContext::getUniform): (WebCore::functionForUniform): (WebCore::dataFunctionf): (WebCore::dataFunctioni): (WebCore::dataFunctionMatrix):
  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/v8/DerivedSourcesAllInOne.cpp:
  • bindings/v8/V8Index.cpp:
  • bindings/v8/V8Index.h:
  • bindings/v8/custom/V8WebGLRenderingContextCustom.cpp: (WebCore::toWebGLUniformLocation): (WebCore::): (WebCore::CALLBACK_FUNC_DECL): (WebCore::isFunctionToCallForAttribute): (WebCore::vertexAttribAndUniformHelperf): (WebCore::uniformHelperi): (WebCore::uniformMatrixHelper):
  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::getUniform): (WebCore::WebGLRenderingContext::getUniformLocation): (WebCore::WebGLRenderingContext::uniform1f): (WebCore::WebGLRenderingContext::uniform1fv): (WebCore::WebGLRenderingContext::uniform1i): (WebCore::WebGLRenderingContext::uniform1iv): (WebCore::WebGLRenderingContext::uniform2f): (WebCore::WebGLRenderingContext::uniform2fv): (WebCore::WebGLRenderingContext::uniform2i): (WebCore::WebGLRenderingContext::uniform2iv): (WebCore::WebGLRenderingContext::uniform3f): (WebCore::WebGLRenderingContext::uniform3fv): (WebCore::WebGLRenderingContext::uniform3i): (WebCore::WebGLRenderingContext::uniform3iv): (WebCore::WebGLRenderingContext::uniform4f): (WebCore::WebGLRenderingContext::uniform4fv): (WebCore::WebGLRenderingContext::uniform4i): (WebCore::WebGLRenderingContext::uniform4iv): (WebCore::WebGLRenderingContext::uniformMatrix2fv): (WebCore::WebGLRenderingContext::uniformMatrix3fv): (WebCore::WebGLRenderingContext::uniformMatrix4fv):
  • html/canvas/WebGLRenderingContext.h:
  • html/canvas/WebGLRenderingContext.idl:
  • html/canvas/WebGLUniformLocation.cpp: Added. (WebCore::WebGLUniformLocation::create): (WebCore::WebGLUniformLocation::WebGLUniformLocation):
  • html/canvas/WebGLUniformLocation.h: Added. (WebCore::WebGLUniformLocation::~WebGLUniformLocation): (WebCore::WebGLUniformLocation::program): (WebCore::WebGLUniformLocation::location):
  • html/canvas/WebGLUniformLocation.idl: Added.
17:43 Changeset [51883] by agl@chromium.org
  • 3 edits
    2 adds in trunk

2009-12-08 Adam Langley <agl@google.com>

Reviewed by Darin Adler.

Fix assertion failure in WebCore::RenderBlock::startDelayUpdateScrollInfo

startDelayUpdateScrollInfo calls a function that can end up calling
startDelayUpdateScrollInfo again. However, it's static state is
inconsistent when this happens leading to an assertion failure (or
probably a memory leak if assertions are off).

Thanks to Robert Swiecki for the test case.

https://bugs.webkit.org/show_bug.cgi?id=32172
http://code.google.com/p/chromium/issues/detail?id=28880

Test: fast/css/recursive-delay-update-scroll.html

  • rendering/RenderBlock.cpp: (WebCore::RenderBlock::finishDelayUpdateScrollInfo):

2009-12-08 Adam Langley <agl@google.com>

Reviewed by Darin Adler.

Fix assertion failure in WebCore::RenderBlock::startDelayUpdateScrollInfo

startDelayUpdateScrollInfo calls a function that can end up calling
startDelayUpdateScrollInfo again. However, it's static state is
inconsistent when this happens leading to an assertion failure (or
probably a memory leak if assertions are off).

Thanks to Robert Swiecki for the test case.

https://bugs.webkit.org/show_bug.cgi?id=32172
http://code.google.com/p/chromium/issues/detail?id=28880

  • fast/css/recursive-delay-update-scroll.html: Added.
  • fast/css/recursive-delay-update-scroll-expected.txt: Added.
17:29 Changeset [51882] by mjs@apple.com
  • 9 edits
    1 copy in trunk/JavaScriptCore

move code generation portions of Nodes.cpp to bytecompiler directory
https://bugs.webkit.org/show_bug.cgi?id=32284

Reviewed by Darin Adler.

  • bytecompiler/NodesCodegen.cpp: Copied from parser/Nodes.cpp. Removed parts that

are not about codegen.

  • parser/Nodes.cpp: Removed everything that is about codegen.

Update build systems:

16:59 Changeset [51881] by sfalken@apple.com
  • 2 edits in trunk/WebCore

WebCore source level debugging on Windows not reliable in some cases
https://bugs.webkit.org/show_bug.cgi?id=32297

Reviewed by Adam Roben.

Customized pdb file name/location for WebCore static library.

  • WebCore.vcproj/WebCoreCommon.vsprops: Add ProgramDataBaseFileName.
16:52 Changeset [51880] by kevino@webkit.org
  • 14 edits
    1 add in trunk

Reviewed by Kevin Ollivier.

[wx] Mac plugins support.

https://bugs.webkit.org/show_bug.cgi?id=32236

16:40 Changeset [51879] by simon.fraser@apple.com
  • 2 edits in trunk/WebKit/mac

2009-12-08 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dan Bernstein.

<rdar://problem/7295070> WebKit video fullscreen keeps playing after closing the window

Fix a leak of the QTMovieView when exiting fullscreen video, and remove the
old workaround.

  • WebView/WebVideoFullscreenController.mm: (-[WebVideoFullscreenController windowDidLoad]): Set the movie view as the contentView directly. (-[WebVideoFullscreenController setMediaElement:WebCore::]): Cast the contentView to a movie view. (-[WebVideoFullscreenController windowDidExitFullscreen]): Remove the old workaround.
16:14 Changeset [51878] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Skip another failing media test, that I already mentioned in the bug report.

  • platform/mac-snowleopard/Skipped:
16:08 Changeset [51877] by beidson@apple.com
  • 3 edits
    2 adds in trunk

Navigating to a cached page can result in accessing a destroyed HTMLInputElement.
<rdar://problem/6856662> and https://webkit.org/b/32293

Reviewed by Darin Adler.

WebCore:

Test: fast/loader/input-element-page-cache-crash.html

  • html/HTMLInputElement.cpp:

(WebCore::HTMLInputElement::parseMappedAttribute): Make sure to unregister for the activation

callback after the new m_autocomplete setting has been stored so the unregistration actually
takes place.

LayoutTests:

  • fast/loader/input-element-page-cache-crash-expected.txt: Added.
  • fast/loader/input-element-page-cache-crash.html: Added.
15:54 Changeset [51876] by levin@chromium.org
  • 3 edits in trunk/WebKitTools

[check-webkit-style] False positive for DEFINE_OPCODE(op_jtrue) {
https://bugs.webkit.org/show_bug.cgi?id=32193

Reviewed by Adam Barth.

  • Scripts/modules/cpp_style.py: For the function { rule, if there is space at the beginning of line, ignore lines which then have a macro.
  • Scripts/modules/cpp_style_unittest.py: Add tests to verify the new behavior.
15:27 Changeset [51875] by dimich@chromium.org
  • 14 edits
    5 deletes in trunk

Revert and reopen "Add asserts to RefCounted to make sure ref/deref happens on the right thread."
It may have caused massive increase of reported leaks on the bots.
https://bugs.webkit.org/show_bug.cgi?id=31639

Rubber-stamped by David Levin.

JavaScriptCore:

(JSC::Structure::Structure):

  • wtf/RefCounted.h:

(WTF::RefCountedBase::ref):
(WTF::RefCountedBase::hasOneRef):
(WTF::RefCountedBase::refCount):
(WTF::RefCountedBase::derefBase):

  • wtf/ThreadVerifier.h: Removed.

JavaScriptGlue:

  • ForwardingHeaders/wtf/ThreadVerifier.h: Removed.

WebCore:

  • ForwardingHeaders/wtf/ThreadVerifier.h: Removed.
  • loader/icon/IconRecord.cpp:

(WebCore::IconRecord::IconRecord):

  • platform/SharedBuffer.cpp:

(WebCore::SharedBuffer::SharedBuffer):

  • platform/text/StringImpl.cpp:

(WebCore::StringImpl::StringImpl):

WebKit/mac:

  • ForwardingHeaders/wtf/ThreadVerifier.h: Removed.

WebKitTools:

  • DumpRenderTree/ForwardingHeaders/wtf/ThreadVerifier.h: Removed.
14:45 WebInspector edited by timothy@apple.com
(diff)
14:43 Changeset [51874] by pfeldman@chromium.org
  • 14 edits
    1 copy in trunk

2009-12-08 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Darin Fisher.

Chromium: support custom WebCore context menu items in Chromium port.

https://bugs.webkit.org/show_bug.cgi?id=32277

WebCore:

  • platform/ContextMenu.h:
  • platform/ContextMenuItem.h: (WebCore::PlatformMenuItemDescription::PlatformMenuItemDescription):
  • platform/chromium/ContextMenuChromium.cpp: (WebCore::ContextMenu::ContextMenu): (WebCore::ContextMenu::itemCount): (WebCore::ContextMenu::insertItem): (WebCore::ContextMenu::appendItem): (WebCore::ContextMenu::itemWithAction): (WebCore::ContextMenu::itemAtIndex): (WebCore::ContextMenu::platformDescription):
  • platform/chromium/ContextMenuItemChromium.cpp: (WebCore::ContextMenuItem::ContextMenuItem): (WebCore::ContextMenuItem::releasePlatformDescription): (WebCore::ContextMenuItem::type): (WebCore::ContextMenuItem::action): (WebCore::ContextMenuItem::title): (WebCore::ContextMenuItem::checked): (WebCore::ContextMenuItem::enabled): (WebCore::ContextMenuItem::setType): (WebCore::ContextMenuItem::setAction): (WebCore::ContextMenuItem::setTitle): (WebCore::ContextMenuItem::setChecked): (WebCore::ContextMenuItem::setEnabled):

WebKit/chromium:

  • public/WebContextMenuData.h:
  • public/WebPopupMenuInfo.h: (WebKit::WebPopupMenuInfo::Item::):
  • public/WebView.h:
  • src/ContextMenuClientImpl.cpp: (WebKit::ContextMenuClientImpl::getCustomMenuFromDefaultItems):
  • src/WebViewImpl.cpp: (WebKit::WebViewImpl::executeCustomContextMenuAction):
  • src/WebViewImpl.h:
14:42 Changeset [51873] by zimmermann@webkit.org
  • 3 edits in trunk/LayoutTests

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Skip fast/repaint/block-selection-gap-in-table-cell.html (generates new result) to turn Qt/Gtk builds green again.

  • platform/gtk/Skipped:
  • platform/qt/Skipped:
13:58 Changeset [51872] by agl@chromium.org
  • 2 edits in trunk/WebCore

2009-12-08 Adam Langley <agl@google.com>

Reviewed by Dimitri Glazkov.

Chromium Linux: set default scrollbar colours.

This is so that layout tests have sane defaults for the scrollbar
colours, even if the API user doesn't set any.

This is a fix due to r51827.

https://bugs.webkit.org/show_bug.cgi?id=32287

This is very well covered by existing layout tests.

  • rendering/RenderThemeChromiumLinux.cpp:
13:35 Changeset [51871] by kov@webkit.org
  • 2 edits in trunk/JavaScriptCore

Reviewed by Darin Adler.

Make WebKit build correctly on FreeBSD, IA64, and Alpha.
Based on work by Petr Salinger <Petr.Salinger@seznam.cz>,
and Colin Watson <cjwatson@ubuntu.com>.

  • wtf/Platform.h:
13:14 Changeset [51870] by bdakin@apple.com
  • 2 edits in trunk/WebCore

Fix for <rdar://problem/7397808> Crash occurs at
RenderObject::containingBlock() as I type in the Google Search
field (during page load)

Reviewed by Sam Weinig.

  • editing/Editor.cpp:

(WebCore::Editor::insideVisibleArea): ownerRenderer() needs to be
null checked.

12:25 Changeset [51869] by dimich@chromium.org
  • 14 edits
    5 adds in trunk

Add asserts to RefCounted to make sure ref/deref happens on the right thread.
https://bugs.webkit.org/show_bug.cgi?id=31639

Reviewed by Darin Adler.

JavaScriptCore:

  • runtime/Structure.cpp:

(JSC::Structure::Structure): Disable thread verification on this class since it uses addressOfCount().

  • wtf/RefCounted.h:

(WTF::RefCountedBase::ref): Add ASSERT.
(WTF::RefCountedBase::hasOneRef): Ditto.
(WTF::RefCountedBase::refCount): Ditto.
(WTF::RefCountedBase::derefBase): Ditto.
(WTF::RefCountedBase::disableThreadVerification): delegate to ThreadVerifier method.

  • wtf/ThreadVerifier.h: Added.

(WTF::ThreadVerifier::ThreadVerifier): New Debug-only class to verify that ref/deref of RefCounted is done on the same thread.
(WTF::ThreadVerifier::activate): Activates checks. Called when ref count becomes above 2.
(WTF::ThreadVerifier::deactivate): Deactivates checks. Called when ref count drops below 2.
(WTF::ThreadVerifier::disableThreadVerification): used on objects that should not be checked (StringImpl etc)
(WTF::ThreadVerifier::verifyThread):

JavaScriptGlue:

  • ForwardingHeaders/wtf/ThreadVerifier.h: Added.

WebCore:

  • loader/icon/IconRecord.cpp:

(WebCore::IconRecord::IconRecord): Disable thread verification for this class because of its cross-thread use in IconDatabase.

  • platform/SharedBuffer.cpp:

(WebCore::SharedBuffer::SharedBuffer): Ditto.

  • platform/text/StringImpl.cpp:

(WebCore::StringImpl::StringImpl): Disable thread verification for this class, add FIXME to enforce proper usage via crossThreadString.

  • ForwardingHeaders/wtf/ThreadVerifier.h: Added.

WebKit/mac:

  • ForwardingHeaders/wtf/ThreadVerifier.h: Added.

WebKitTools:

  • DumpRenderTree/ForwardingHeaders/wtf/ThreadVerifier.h: Added.
11:16 Changeset [51868] by ajwong@chromium.org
  • 2 edits in trunk/WebKit/chromium

Not reviewed: Chromium build fix.

r51859 changed the case for some of the ResourceRequest::TargetType
enums.

  • src/FrameLoaderClientImpl.cpp:

(WebKit::determineTargetTypeFromLoader):

10:59 Changeset [51867] by aroben@apple.com
  • 2 edits in trunk/WebKit/win

Windows build fix

  • Interfaces/WebKit.idl: Touched this to force Interfaces.vcproj to rebuild.
10:55 Changeset [51866] by levin@chromium.org
  • 3 edits in trunk/WebKitTools

[check-webkit-style] False positive for camel case of JSC op codes
https://bugs.webkit.org/show_bug.cgi?id=32192

Reviewed by Adam Barth.

  • Scripts/modules/cpp_style.py: Added an exception for the JSC op code functions and const_iterator as well since I noticed a false positive there when testing the fix.
  • Scripts/modules/cpp_style_unittest.py: Added tests for these changes.
10:44 Changeset [51865] by abarth@webkit.org
  • 7 edits
    2 adds in trunk

2009-12-08 John Gregg <johnnyg@google.com>

Reviewed by Adam Barth.

https://bugs.webkit.org/show_bug.cgi?id=31886

Test for accessing the notificationCenter of a window after it's closed.

  • fast/notifications/notification-after-close-expected.txt: Added.
  • fast/notifications/notification-after-close.html: Added.

2009-12-08 John Gregg <johnnyg@google.com>

Reviewed by Adam Barth.

Inform the NotificationCenter when its window goes away in the same
manner as other DOMWindow fields, and prevent invalid operations on
it after that happens.

As part of this, change the V8 bindings for notifications to call
through the NotificationCenter rather than doing those operations
itself.

https://bugs.webkit.org/show_bug.cgi?id=31886

Test: fast/notifications/notification-after-close.html

  • bindings/v8/custom/V8NotificationCenterCustom.cpp: (WebCore::CALLBACK_FUNC_DECL):
  • notifications/Notification.cpp: (WebCore::Notification::Notification):
  • notifications/NotificationCenter.cpp: (WebCore::NotificationCenter::checkPermission): (WebCore::NotificationCenter::requestPermission):
  • notifications/NotificationCenter.h: (WebCore::NotificationCenter::createHTMLNotification): (WebCore::NotificationCenter::createNotification): (WebCore::NotificationCenter::disconnectFrame):
  • page/DOMWindow.cpp: (WebCore::DOMWindow::clear):
10:44 CommitQueue edited by abarth@webkit.org
(diff)
09:37 Changeset [51864] by mitz@apple.com
  • 4 edits
    4 adds in trunk

WebCore: Fixed block selection gap repainting in table cells

Reviewed by John Sullivan.

Test: fast/repaint/block-selection-gap-in-table-cell.html

This recently-introduced regression exposed the fact that
RenderTableCell did not override offsetFromContainer() to adjust for
the fact that table cells’ coordinates are relative to the table
section, not the table row. With this fixed, RenderTableCell no longer
needs to override mapLocalToContainer() and mapAbsoluteToLocalPoint(),
since the base class implementations of those use offsetFromContainer().

  • rendering/RenderTableCell.cpp:

(WebCore::RenderTableCell::offsetFromContainer):

  • rendering/RenderTableCell.h:

LayoutTests: Added a test of block selection gap repainting in table cells

Reviewed by John Sullivan.

  • fast/repaint/block-selection-gap-in-table-cell.html: Added.
  • platform/mac/fast/repaint/block-selection-gap-in-table-cell-expected.checksum: Added.
  • platform/mac/fast/repaint/block-selection-gap-in-table-cell-expected.png: Added.
  • platform/mac/fast/repaint/block-selection-gap-in-table-cell-expected.txt: Added.
09:35 Changeset [51863] by zimmermann@webkit.org
  • 2 edits in trunk/WebCore

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Sane way to force a rebuild on all win slaves, according to Adam Roben.

  • WebCorePrefix.h:
09:32 Changeset [51862] by mitz@apple.com
  • 1 edit in trunk/WebCore/ChangeLog

Correct Bugzilla link

09:31 Changeset [51861] by mitz@apple.com
  • 2 edits in trunk/WebCore

Build fix

  • loader/loader.cpp:

(WebCore::cachedResourceTypeToTargetType):

09:08 Changeset [51860] by zimmermann@webkit.org
  • 2 edits in trunk/WebCore

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Touch svgtags.in to eventually fix the win debug build.

  • svg/svgtags.in:
09:04 Changeset [51859] by eric@webkit.org
  • 4 edits in trunk/WebCore

2009-12-08 Mike Belshe <mike@belshe.com>

Reviewed by Darin Fisher.

https://bugs.webkit.org/show_bug.cgi?id=32152
Update the ResourceRequest::RequestType. This previously
was specific to Chromium. Moved into ResourceRequestBase, enabling
more specificity about the type (which is otherwise only known to the
loader), and also making this information available to all platforms.
Any platform with a network layer which can utilize this information
may want to use it for prioritization.

Note to Chromium glue: TargetIsSubResource renamed to TargetIsSubresource.

  • loader/loader.cpp: (WebCore::Loader::Loader): (WebCore::CachedResourceTypeToTargetType): (WebCore::Loader::Host::servePendingRequests):
  • platform/network/ResourceRequestBase.h: (WebCore::ResourceRequestBase::): (WebCore::ResourceRequestBase::targetType): (WebCore::ResourceRequestBase::setTargetType): (WebCore::ResourceRequestBase::ResourceRequestBase):
  • platform/network/chromium/ResourceRequest.h: (WebCore::ResourceRequest::ResourceRequest):
08:55 Changeset [51858] by eric@webkit.org
  • 4 edits
    7 adds in trunk

2009-12-08 Steve Block <steveblock@google.com>

Reviewed by Adam Barth.

[Android] Adds Makefiles for Android port.
https://bugs.webkit.org/show_bug.cgi?id=31325

  • Android.mk: Added.

2009-12-08 Steve Block <steveblock@google.com>

Reviewed by Adam Barth.

[Android] Adds Makefiles for Android port.
https://bugs.webkit.org/show_bug.cgi?id=31325

  • Android.mk: Added.
  • Android.v8.wtf.mk: Added.

2009-12-08 Steve Block <steveblock@google.com>

Reviewed by Adam Barth.

[Android] Adds Makefiles for Android port.
https://bugs.webkit.org/show_bug.cgi?id=31325

Build system change only. No tests possible.

  • Android.derived.jscbindings.mk: Added.
  • Android.derived.mk: Added.
  • Android.jscbindings.mk: Added.
  • Android.mk: Added.
  • WebCorePrefix.h: Modified. Sets up some flags and adds a header required for building on Android.
08:55 Changeset [51857] by sullivan@apple.com
  • 2 edits in trunk/WebKit/win

Split two-clause assertions into two separate assertions.

Reviewed by Dan Bernstein

  • DOMHTMLClasses.cpp:

(DOMHTMLInputElement::isTextField):
Split a two-clause assertion into two separate assertions.
(DOMHTMLInputElement::rectOnScreen):
Added the two assertions here that all other functions in this group shared.
(DOMHTMLInputElement::selectedRange):
Split a two-clause assertion into two separate assertions.
(DOMHTMLInputElement::setAutofilled):
Split a two-clause assertion into two separate assertions.
(DOMHTMLInputElement::isAutofilled):
Split a two-clause assertion into two separate assertions.

08:36 Changeset [51856] by christian@webkit.org
  • 4 edits in trunk

2009-12-08 Christian Dywan <christian@twotoasts.de>

Reviewed by Xan Lopez.

  • configure.ac: Require only libSoup 2.27.91 but check for 2.29.3 and define HAVE_LIBSOUP_2_29_3 in that case.
08:28 Changeset [51855] by zimmermann@webkit.org
  • 10 edits in trunk

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Rubber-stamped by Maciej Stachowiak.

Turn on (SVG) Filters for Win.
https://bugs.webkit.org/show_bug.cgi?id=32224

08:17 Changeset [51854] by christian@webkit.org
  • 2 edits in trunk/WebKit/gtk

2009-12-08 Christian Dywan <christian@twotoasts.de>

Reviewed by Gustavo Noronha Silva.

Point out the meaning of "Content-Disposition" when implementing the
mime-type-policy-decision-requested signal.

  • webkit/webkitwebview.cpp: (webkit_web_view_class_init):
07:58 Changeset [51853] by kov@webkit.org
  • 2 edits in trunk

Rubber-stamped by Xan Lopez.

Late post-release version bump.

  • configure.ac:
07:33 Changeset [51852] by eric@webkit.org
  • 5 edits in trunk/LayoutTests

2009-12-08 Kent Tamura <tkent@chromium.org>

Reviewed by Darin Adler.

These test results should not have platform-dependent pixel values.
https://bugs.webkit.org/show_bug.cgi?id=32265

Change is75PercentOf() so that it doesn't show pixel values when
the check is passed.

  • fast/replaced/table-percent-height-expected.txt:
  • fast/replaced/table-percent-height-text-controls-expected.txt:
  • fast/replaced/table-percent-height-text-controls.html:
  • fast/replaced/table-percent-height.html:
07:20 Changeset [51851] by eric@webkit.org
  • 3 edits
    2 adds in trunk

2009-12-08 Carol Szabo <carol.szabo@nokia.com>

Reviewed by Darin Adler.

CSS Counter Nesting still does not work according to the spec.
https://bugs.webkit.org/show_bug.cgi?id=31723

  • fast/css/counters/nesting-expected.txt: Added.
  • fast/css/counters/nesting.html: Added. This test tests compliance with the CSS2.1 counter scoping and nesting rules.

2009-12-08 Carol Szabo <carol.szabo@nokia.com>

Reviewed by Darin Adler.

CSS Counter Nesting still does not work according to the spec.
https://bugs.webkit.org/show_bug.cgi?id=31723

Test: fast/css/counters/nesting.html

  • rendering/RenderCounter.cpp: (WebCore::findPlaceForCounter): Replaced the faulty counter insertion algorithm with one that works.
07:08 Changeset [51850] by sullivan@apple.com
  • 7 edits in trunk

Add isAutofilled getter to match existing setter.

Reviewed by Ada Chan.

WebCore:

  • bindings/objc/DOMHTML.mm:

(-[DOMHTMLInputElement _isAutofilled]):
Implemented new cover function.

  • bindings/objc/DOMPrivate.h:

Declared new cover function.

WebKit/win:

  • DOMHTMLClasses.cpp:

(DOMHTMLInputElement::isAutofilled):
Implemented new cover function.

  • DOMHTMLClasses.h:

Declared new cover function.

  • Interfaces/DOMPrivate.idl:

Declared new interface.

07:02 Changeset [51849] by ossy@webkit.org
  • 2 edits in trunk/LayoutTests

[Qt] Put test into skiplist because of missing layoutTestController.evaluateInWebInspector().

  • platform/qt/Skipped: inspector/cookie-resource-match.html added.
06:58 Changeset [51848] by eric@webkit.org
  • 4 edits
    1 add in trunk

2009-12-08 Dominik Röttsches <dominik.roettsches@access-company.com>

Reviewed by Gustavo Noronha Silva.

[Gtk] Create a TextBreakIterator implementation based on GLib (without ICU)
https://bugs.webkit.org/show_bug.cgi?id=31469

Removing hybrid configuration for --with-unicode-backend=glib
ICU not required anymore.

  • autotools/webkit.m4:

2009-12-08 Dominik Röttsches <dominik.roettsches@access-company.com>

Reviewed by Gustavo Noronha Silva.

[Gtk] Create a TextBreakIterator implementation based on GLib (without ICU)
https://bugs.webkit.org/show_bug.cgi?id=31469

Added a TextBreakIterator implementation based on GLib and pango,
which allows compiling WebCore without ICU.

  • GNUmakefile.am:
  • platform/text/gtk/TextBreakIteratorGtk.cpp: Added. (WebCore::): (WebCore::setUpIterator): (WebCore::characterBreakIterator): (WebCore::cursorMovementIterator): (WebCore::wordBreakIterator): (WebCore::lineBreakIterator): (WebCore::sentenceBreakIterator): (WebCore::textBreakFirst): (WebCore::textBreakLast): (WebCore::textBreakNext): (WebCore::textBreakPrevious): (WebCore::textBreakPreceding): (WebCore::textBreakFollowing): (WebCore::textBreakCurrent): (WebCore::isTextBreak):
06:35 Changeset [51847] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Update Gtk's custom prototype-inheritance-2-expected.txt, after enabling filters
(As gtk builds with geolocation, unlike the other platforms, it needs its own results)

This turns the Gtk slave to green again.

  • platform/gtk/fast/dom/prototype-inheritance-2-expected.txt:
06:09 QtWebKit/CodeCamp200912 edited by tapani.mikola@nokia.com
(diff)
06:02 QtWebKit edited by tapani.mikola@nokia.com
(diff)
05:56 QtWebKit/CodeCamp200912 created by tapani.mikola@nokia.com
05:21 Changeset [51846] by eric@webkit.org
  • 11 edits
    2 adds in trunk

2009-12-08 Alexander Pavlov <apavlov@chromium.org>

Reviewed by Pavel Feldman.

Change the way cookies are retrieved in the WebInspector frontend.

Moved the cookie filtering from the native code into the frontend so that
all cookies can be retrieved regardless of the associated domain
(required for certain audits to run.)
https://bugs.webkit.org/show_bug.cgi?id=32160

  • inspector/cookie-resource-match-expected.txt: Added.
  • inspector/cookie-resource-match.html: Added.

2009-12-08 Alexander Pavlov <apavlov@chromium.org>

Reviewed by Pavel Feldman.

Change the way cookies are retrieved in the WebInspector frontend.

Moved the cookie filtering from the native code into the frontend so that
all cookies can be retrieved regardless of the associated domain
(required for certain audits to run.)
https://bugs.webkit.org/show_bug.cgi?id=32160

Test: inspector/cookie-resource-match.html

  • inspector/InspectorBackend.cpp: (WebCore::InspectorBackend::getCookies):
  • inspector/InspectorBackend.h:
  • inspector/InspectorBackend.idl:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::getCookies):
  • inspector/InspectorController.h:
  • inspector/front-end/CookieItemsView.js: (WebInspector.CookieItemsView.prototype.update.callback): (WebInspector.CookieItemsView.prototype.update): (WebInspector.CookieItemsView.prototype._cookiesForDomain):
  • inspector/front-end/DOMAgent.js: (WebInspector.Cookies.getCookiesAsync): (WebInspector.Cookies.cookieMatchesResourceURL): (WebInspector.Cookies.cookieDomainMatchesResourceDomain):
  • inspector/front-end/Resource.js: (WebInspector.Resource): (WebInspector.Resource.prototype.get documentURL): (WebInspector.Resource.prototype.set documentURL):
  • inspector/front-end/inspector.js: (WebInspector.addResource):
04:55 Changeset [51845] by zimmermann@webkit.org
  • 2 edits in trunk/WebCore

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Touch svgtags.in, in order to force SVGNames.* regeneration. Praying for gtk bot to be fixed :-)

  • svg/svgtags.in:
04:30 Changeset [51844] by zimmermann@webkit.org
  • 2 edits in trunk/WebCore

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Trying to force a rebuild on the Gtk slave. If that doesn't work, someone must force a rebuild from scratch.

  • bindings/scripts/CodeGeneratorJS.pm:
03:51 Changeset [51843] by zimmermann@webkit.org
  • 6 edits in trunk

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Rubber-stamped by Maciej Stachowiak.

Turn on (SVG) Filters for Gtk.
https://bugs.webkit.org/show_bug.cgi?id=32224

03:21 Changeset [51842] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Skip some tests that fail for a longer period on the SnowLeopard bot. Filed two bugs, notifying the authors.

  • platform/mac-snowleopard/Skipped:
02:58 Changeset [51841] by zimmermann@webkit.org
  • 4 edits in trunk/LayoutTests

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Oops forgot to update three results -> Qt slave green again.

  • platform/qt/svg/filters/feGaussianBlur-expected.txt:
  • platform/qt/svg/filters/feMerge-wrong-input-expected.txt:
  • platform/qt/svg/filters/feOffset-expected.txt:
02:47 Changeset [51840] by zimmermann@webkit.org
  • 11 edits in trunk/LayoutTests

2009-12-08 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Now that the Qt slave has been rebuilt, reenable all tests that failed
because of the build problems (Qt slave didn't rebuild all files with ENABLE_FILTERS set).

Update some Qt specfic filter test results.

  • platform/qt/Skipped:
  • platform/qt/svg/filters/feComposite-expected.txt:
  • platform/qt/svg/filters/feMerge-expected.txt:
  • platform/qt/svg/filters/feTile-expected.txt:
  • platform/qt/svg/filters/shadow-on-filter-expected.txt:
  • platform/qt/svg/filters/shadow-on-rect-with-filter-expected.txt:
  • platform/qt/svg/filters/sourceAlpha-expected.txt:
  • platform/qt/svg/filters/subRegion-in-userSpace-expected.txt:
  • platform/qt/svg/filters/subRegion-one-effect-expected.txt:
  • platform/qt/svg/filters/subRegion-two-effects-expected.txt:
02:19 Changeset [51839] by pfeldman@chromium.org
  • 22 edits
    2 copies in trunk/WebCore

2009-12-06 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: provide custom context menu in the front-end window.

https://bugs.webkit.org/show_bug.cgi?id=32200

  • English.lproj/localizedStrings.js:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSInspectorFrontendHostCustom.cpp: (WebCore::JSInspectorFrontendHost::showContextMenu):
  • inspector/InspectorFrontend.cpp: (WebCore::InspectorFrontend::contextMenuItemSelected): (WebCore::InspectorFrontend::contextMenuCleared):
  • inspector/InspectorFrontend.h:
  • inspector/InspectorFrontendHost.cpp: (WebCore::InspectorFrontendHost::InspectorFrontendHost): (WebCore::InspectorFrontendHost::~InspectorFrontendHost): (WebCore::InspectorFrontendHost::showContextMenu): (WebCore::InspectorFrontendHost::contextMenuItemSelected): (WebCore::InspectorFrontendHost::contextMenuCleared):
  • inspector/InspectorFrontendHost.h: (WebCore::InspectorFrontendHost::MenuSelectionHandler::create): (WebCore::InspectorFrontendHost::MenuSelectionHandler::~MenuSelectionHandler): (WebCore::InspectorFrontendHost::MenuSelectionHandler::disconnect): (WebCore::InspectorFrontendHost::MenuSelectionHandler::contextMenuItemSelected): (WebCore::InspectorFrontendHost::MenuSelectionHandler::contextMenuCleared): (WebCore::InspectorFrontendHost::MenuSelectionHandler::MenuSelectionHandler):
  • inspector/InspectorFrontendHost.idl:
  • inspector/front-end/ContextMenu.js: Added.
  • inspector/front-end/WebKit.qrc:
  • inspector/front-end/inspector.html:
  • inspector/front-end/inspector.js: (WebInspector.loaded): (WebInspector.documentContextMenu):
  • page/ContextMenuController.cpp: (WebCore::ContextMenuController::ContextMenuController): (WebCore::ContextMenuController::clearContextMenu): (WebCore::ContextMenuController::handleContextMenuEvent): (WebCore::ContextMenuController::showContextMenu): (WebCore::ContextMenuController::createContextMenu): (WebCore::ContextMenuController::contextMenuItemSelected):
  • page/ContextMenuController.h:
  • page/ContextMenuSelectionHandler.h: Added. (WebCore::ContextMenuSelectionHandler::ContextMenuSelectionHandler): (WebCore::ContextMenuSelectionHandler::~ContextMenuSelectionHandler):
  • platform/ContextMenu.cpp: (WebCore::ContextMenu::checkOrEnableIfNeeded):
  • platform/ContextMenuItem.h:
02:04 Changeset [51838] by ossy@webkit.org
  • 2 edits in trunk/WebCore

Rubber-stamped by Kenneth Rohde Christiansen.

Patch by Andras Becsi <abecsi@inf.u-szeged.hu> on 2009-12-08
[Qt] Build fix of duplicated platform/graphics/filters/FEGaussianBlur.cpp in WebCore.pro after enabled svg filters.

No new tests needed.

  • WebCore.pro:
00:32 Changeset [51837] by ukai@chromium.org
  • 1 edit
    10 adds in trunk/LayoutTests

2009-12-08 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Darin Adler.

Add tests that WebSocket protocol argument works.
https://bugs.webkit.org/show_bug.cgi?id=3178

  • websocket/tests/bad-sub-protocol-expected.txt: Added.
  • websocket/tests/bad-sub-protocol.html: Added.
  • websocket/tests/handshake-fail-by-sub-protocol-mismatch-expected.txt: Added.
  • websocket/tests/handshake-fail-by-sub-protocol-mismatch.html: Added.
  • websocket/tests/protocol-test_wsh.py: Added.
  • websocket/tests/script-tests/bad-sub-protocol.js: Added.
  • websocket/tests/script-tests/handshake-fail-by-sub-protocol-mismatch.js: Added.
  • websocket/tests/script-tests/sub-protocol.js: Added.
  • websocket/tests/sub-protocol-expected.txt: Added.
  • websocket/tests/sub-protocol.html: Added.

12/07/09:

22:00 Changeset [51836] by ossy@webkit.org
  • 2 edits in trunk/LayoutTests

[Qt] Unreviewed fix to make QtBuildBot green.
Skip failing tests caused by r51828 (Turn on (SVG) Filters for Qt.)
https://bugs.webkit.org/show_bug.cgi?id=32224
http://build.webkit.org/results/Qt%20Linux%20Release/r51828%20(4620)/results.html

  • platform/qt/Skipped:
21:58 Changeset [51835] by hamaji@chromium.org
  • 3 edits in trunk/WebKitTools

2009-12-07 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by David Levin.

[check-webkit-style] S_OK is a fine identifier
https://bugs.webkit.org/show_bug.cgi?id=32225

  • Scripts/modules/cpp_style.py:
  • Scripts/modules/cpp_style_unittest.py:
21:48 Changeset [51834] by hamaji@chromium.org
  • 5 edits in trunk/LayoutTests

2009-12-07 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

editing/selection/last-empty-inline is failing in Windows
https://bugs.webkit.org/show_bug.cgi?id=31985

  • editing/selection/last-empty-inline-expected.txt:
  • editing/selection/last-empty-inline.html:
  • platform/qt/Skipped:
  • platform/win/Skipped:
21:29 Changeset [51833] by hamaji@chromium.org
  • 3 edits in trunk/WebCore

2009-12-07 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

Add a function to show render tree for debugging
https://bugs.webkit.org/show_bug.cgi?id=31288

No new tests because this just adds a function for debugging.

  • rendering/RenderObject.cpp: (WebCore::RenderObject::showRenderObject): (WebCore::RenderObject::showRenderTreeAndMark): (showRenderTree):
  • rendering/RenderObject.h:
20:55 Changeset [51832] by eric@webkit.org
  • 2 edits in trunk/WebKit/chromium

2009-12-07 Zhe Su <suzhe@chromium.org>

Reviewed by Darin Fisher.

[Chromium] accesskey is not supported on Chromium Mac.
https://bugs.webkit.org/show_bug.cgi?id=32213

This change fixes (partially) the accesskey issue by decoupling the
code for handling accesskeys and system keys. Because on Mac,
the modifiers of accesskeys are ctrl+alt, which are not marked as
system keys.
In order to fully fix this issue, some changes in Chromium code is
also required.

  • src/WebViewImpl.cpp: (WebKit::WebViewImpl::keyEvent): (WebKit::WebViewImpl::charEvent):
20:33 Changeset [51831] by eric@webkit.org
  • 2 edits in trunk/WebCore

2009-12-07 Victor Wang <victorw@chromium.org>

Reviewed by Dimitri Glazkov.

Implement NamedPropertyEnumerator and IndexedPropertyEnumerator for
V8 NPObject. This should make the object enumerable and fix the
enumeration issue in layout test plugins/netscape-enumerate.html.

Also fix some existing style issues per webkit style guidelines.

https://bugs.webkit.org/show_bug.cgi?id=32254

Test: plugins/netscape-enumerate.html

  • bindings/v8/V8NPObject.cpp: (npObjectInvokeImpl): (npObjectInvokeDefaultHandler): (npObjectGetProperty): (npObjectPropertyEnumerator): (npObjectNamedPropertyEnumerator): (npObjectIndexedPropertyEnumerator): (createV8ObjectForNPObject):
20:19 Changeset [51830] by ukai@chromium.org
  • 5 edits
    3 deletes in trunk

2009-12-07 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Alexey Proskuryakov.

Fragments now make WebSocket URL parsing fail.
https://bugs.webkit.org/show_bug.cgi?id=32144

  • websocket/tests/script-tests/url-parsing.js:
  • websocket/tests/script-tests/url-with-fragment.js: Removed.
  • websocket/tests/url-parsing-expected.txt:
  • websocket/tests/url-with-fragment-expected.txt: Removed.
  • websocket/tests/url-with-fragment.html: Removed.

2009-12-07 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Alexey Proskuryakov.

Fragments now make WebSocket URL parsing fail.
https://bugs.webkit.org/show_bug.cgi?id=32144

  • websockets/WebSocket.cpp: (WebCore::WebSocket::connect):
20:03 Changeset [51829] by ukai@chromium.org
  • 5 edits in trunk

2009-12-07 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Darin Adler.

Fix wrong length parsing in WebSocket.
https://bugs.webkit.org/show_bug.cgi?id=32203

These two tests assumed wrong length encoding in frame: it parsed
length from bytes with 8th bit on. But spec says length is encoded
as a series of 7-bit bytes stored in octests with the 8th bit on
*but the last byte*.
These tests encodes a frame that has 129 length, so it must be
\x81\0x01 instead of \x81\0x81.

  • websocket/tests/frame-length-longer-than-buffer_wsh.py:
  • websocket/tests/frame-length-skip_wsh.py:

2009-12-07 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Darin Adler.

Fix wrong length parsing in WebSocket.
https://bugs.webkit.org/show_bug.cgi?id=32203

  • websockets/WebSocketChannel.cpp: (WebCore::WebSocketChannel::didReceiveData):
19:44 Changeset [51828] by zimmermann@webkit.org
  • 6 edits in trunk

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Rubber-stamped by Maciej Stachowiak.

Turn on (SVG) Filters for Qt.
https://bugs.webkit.org/show_bug.cgi?id=32224

18:54 Changeset [51827] by eric@webkit.org
  • 8 edits in trunk

2009-12-07 Evan Martin <evan@chromium.org>

Reviewed by Eric Seidel.

Chromium: theme scrollbars to match GTK theme.
Add functions to RenderThemeChromiumLinux to change the scrollbar
color.

Since the change is to the Chromium WebKit API layer, testing will
be in Chromium's test shell.
http://bugs.webkit.org/show_bug.cgi?id=32048

Patch from Markus Gutschke <markus@chromium.org>.

  • platform/chromium/ScrollbarThemeChromiumLinux.cpp: (WebCore::saturateAndBrighten): (WebCore::outlineColor): (WebCore::ScrollbarThemeChromiumLinux::paintTrackPiece): (WebCore::ScrollbarThemeChromiumLinux::paintThumb):
  • rendering/RenderThemeChromiumLinux.cpp: (WebCore::RenderThemeChromiumLinux::setScrollbarColors):
  • rendering/RenderThemeChromiumLinux.h: (WebCore::RenderThemeChromiumLinux::thumbInactiveColor): (WebCore::RenderThemeChromiumLinux::thumbActiveColor): (WebCore::RenderThemeChromiumLinux::trackColor):

2009-12-07 Evan Martin <evan@chromium.org>

Reviewed by Eric Seidel.

Chromium: theme scrollbars to match GTK theme.
Add API to set the colors.

Since the change is to the Chromium WebKit API layer, testing will
be in Chromium's test shell.
http://bugs.webkit.org/show_bug.cgi?id=32048

Patch from Markus Gutschke <markus@chromium.org>.

  • public/WebView.h:
  • src/WebViewImpl.cpp: (WebKit::WebViewImpl::setScrollbarColors):
  • src/WebViewImpl.h:
18:50 Changeset [51826] by hamaji@chromium.org
  • 2 edits in trunk/BugsSite

2009-12-07 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by David Kilzer.

Bugzilla should show images in git patches
https://bugs.webkit.org/show_bug.cgi?id=31395

Attempt to go r51748 again using --directory option of git-apply.

  • PrettyPatch/PrettyPatch.rb:
18:44 Changeset [51825] by kov@webkit.org
  • 10 edits in trunk

WebCore

Reviewed by Xan Lopez.

[GTK] Doesn't respect Content-Disposition for downloads, and provides no way for apps to do that
https://bugs.webkit.org/show_bug.cgi?id=32247

Allow creating a ResourceResponse from a SoupMessage.

Covered by API test.

  • platform/network/soup/ResourceResponse.h: (WebCore::ResourceResponse::ResourceResponse):
  • platform/network/soup/ResourceResponseSoup.cpp: (WebCore::ResourceResponse::updateFromSoupMessage):

WebKit/gtk

Reviewed by Xan Lopez.

[GTK] Doesn't respect Content-Disposition for downloads, and provides no way for apps to do that
https://bugs.webkit.org/show_bug.cgi?id=32247

Provide a way for applications to respect Content-Disposition, by
fetching the WebKitNetworkResponse from the frame during
mime-type-policy-decision-requested.

  • WebCoreSupport/FrameLoaderClientGtk.cpp: (WebKit::FrameLoaderClient::dispatchDecidePolicyForMIMEType): Download content when Content-Disposition is attachment.
  • tests/testmimehandling.c: (server_callback): (mime_type_policy_decision_requested_cb):
  • tests/testwebframe.c: (test_webkit_web_frame_response): (main):
  • webkit/webkitprivate.cpp: (WebKit::core):
  • webkit/webkitwebframe.cpp: (webkit_web_frame_get_network_response):
  • webkit/webkitwebframe.h:
18:40 Changeset [51824] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-07 Yuzo Fujishima <yuzo@google.com>

Reviewed by Alexey Proskuryakov.

Make run-webkit-websocketserver log to stderr rather than to a file.
https://bugs.webkit.org/show_bug.cgi?id=32234

  • Scripts/run-webkit-websocketserver:
18:29 Changeset [51823] by zimmermann@webkit.org
  • 3 edits
    3 deletes in trunk

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Reverting problematic patch, causing errors.

Revert r51789 (Avoid zero division during SVGPaintServerPattern::setup()). Crashes on all Windows slaves.
Reopen bug https://bugs.webkit.org/show_bug.cgi?id=29912.

  • svg/graphics/SVGPaintServerPattern.cpp: (WebCore::SVGPaintServerPattern::setup):
18:26 Changeset [51822] by tkent@chromium.org
  • 6 edits
    9 adds in trunk

2009-12-07 Kent Tamura <tkent@chromium.org>

Reviewed by Darin Adler.

Add tests for HTMLInputElement::stepUp() and stepDown().
https://bugs.webkit.org/show_bug.cgi?id=27451

  • fast/forms/input-step-number-expected.txt: Added.
  • fast/forms/input-step-number.html: Added.
  • fast/forms/input-step-range-expected.txt: Added.
  • fast/forms/input-step-range.html: Added.
  • fast/forms/input-step-unsupported-expected.txt: Added.
  • fast/forms/input-step-unsupported.html: Added.
  • fast/forms/script-tests/input-step-number.js: Added.
  • fast/forms/script-tests/input-step-range.js: Added.
  • fast/forms/script-tests/input-step-unsupported.js: Added.

2009-12-07 Kent Tamura <tkent@chromium.org>

Reviewed by Darin Adler.

Add support for HTMLInputElement::stepUp() and stepDown() for
type=number and type=range.
https://bugs.webkit.org/show_bug.cgi?id=27451

Our implementation of stepUp() and stepDown() rounds the resultant
value to conform to the step value.
Change the number-string conversion method for RenderSlider to be
consistent with type=number.

Tests: fast/forms/input-step-number.html

fast/forms/input-step-range.html
fast/forms/input-step-unsupported.html

  • html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::rangeUnderflow): Unify the code for NUMBER and RANGE. (WebCore::HTMLInputElement::rangeOverflow): Unify the code for NUMBER and RANGE. (WebCore::HTMLInputElement::minimum): Renamed from rangeMinimum(), and support for NUMBER. (WebCore::HTMLInputElement::maximum): Renamed from rangeMaximum(), and support for NUMBER. (WebCore::HTMLInputElement::stepBase): (WebCore::HTMLInputElement::stepMismatch): Use stepBase(). (WebCore::HTMLInputElement::applyStepForNumberOrRange): (WebCore::HTMLInputElement::stepUp): (WebCore::HTMLInputElement::stepDown): (WebCore::HTMLInputElement::formStringFromDouble):
  • html/HTMLInputElement.h: (WebCore::HTMLInputElement::stepUp): (WebCore::HTMLInputElement::stepDown):
  • html/HTMLInputElement.idl: Add stepUp() and stepDown().
  • rendering/RenderSlider.cpp: (WebCore::SliderRange::SliderRange): Sync with rangeMinimum()/rangeMaximum() renaming. (WebCore::RenderSlider::updateFromElement): Use formStringFromDouble(). (WebCore::RenderSlider::setValueForPosition): Use formStringFromDouble().
18:21 Changeset [51821] by ajwong@chromium.org
  • 2 edits in trunk/WebCore

Not reviewed: Chromium build fix try 2 (dumb error).

  • bindings/v8/ScriptValue.h:

(WebCore::ScriptValue::getString):

17:55 Changeset [51820] by ajwong@chromium.org
  • 3 edits in trunk/WebCore

Not reviewed: Chromium build fix.

Fix Chromium build break caused by an added dependency from

ScriptControllerBase to JSDOMWindowShell, and an API change in
ScriptValue.

  • bindings/ScriptControllerBase.cpp:

(WebCore::ScriptController::executeIfJavaScriptURL):

  • bindings/v8/ScriptValue.h:

(WebCore::ScriptValue::getString):

17:31 Changeset [51819] by enrica@apple.com
  • 3 edits
    2 adds in trunk

WebCore: REGRESSION(4.0.4-42a12): Indent does not work for twice modified with background text in gmail.
<rdar://problem/7442387>
https://bugs.webkit.org/show_bug.cgi?id=32233

Reviewed by Darin Adler.

Fixes problem in re-creating the correct hierarchy under the new block.
Added extensive testing to cover all the cases.

Test: editing/execCommand/indent-with-style2.html

  • editing/CompositeEditCommand.cpp:

(WebCore::CompositeEditCommand::cloneParagraphUnderNewElement):

LayoutTests: REGRESSION(4.0.4-42a12): Indent does not work for twice modified with background text in gmail.
<rdar://problem/7442387>
https://bugs.webkit.org/show_bug.cgi?id=32233

Reviewed by Darin Adler.

  • editing/execCommand/indent-with-style2-expected.txt: Added.
  • editing/execCommand/indent-with-style2.html: Added.
17:22 Changeset [51818] by eric@webkit.org
  • 3 edits in trunk/WebKit/chromium

2009-12-07 Finnur Thorarinsson <finnur.webkit@gmail.com>

Reviewed by Darin Fisher.

A patch for Chromium to restrict the scope of the Gmail focus fix,
where we set the Selection to 0,0 for content-editable fields and
also make sure we set the selection end-state for find to select the
text found (when not focusing a link we found).
WebKit bug: https://bugs.webkit.org/show_bug.cgi?id=32248

  • src/WebFrameImpl.cpp: (WebKit::WebFrameImpl::setFindEndstateFocusAndSelection):
  • src/WebViewImpl.cpp: (WebKit::WebViewImpl::setFocus):
17:15 Changeset [51817] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

Not reviewed. Add some missing files that depend on SVG Filters support, only mac supports it for now.
Still one more crash affecting the win slaves, unrelated to the filters patch though.

17:03 Changeset [51816] by aroben@apple.com
  • 2 edits in trunk/WebKitLibraries

Windows build fix for checkouts with a space in the path

  • win/tools/scripts/auto-version.sh: Quote the output file's directory

before passing it to mkdir.

16:59 Changeset [51815] by ossy@webkit.org
  • 2 edits in trunk/LayoutTests

[Qt] Unreviewed. Skip a timeouter flakey test.

  • platform/qt/Skipped: fast/frames/sandboxed-iframe-plugins.html added.
16:51 Changeset [51814] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

Not reviewed. Skip yet another test, what a journey...

16:50 Changeset [51813] by dimich@chromium.org
  • 15 edits in trunk

Remove ENABLE_SHARED_SCRIPT flags
https://bugs.webkit.org/show_bug.cgi?id=32245
This patch was obtained by "git revert" command and then un-reverting of ChangeLog files.

Rubber-stamped by Darin Adler.

  • configure.ac:

JavaScriptCore:

  • Configurations/FeatureDefines.xcconfig:
  • wtf/Platform.h:

WebCore:

  • Configurations/FeatureDefines.xcconfig:
  • GNUmakefile.am:
  • WebCore.pro:

WebKit/chromium:

  • features.gypi:

WebKit/mac:

  • Configurations/FeatureDefines.xcconfig:

WebKitTools:

  • Scripts/build-webkit:
16:40 Changeset [51812] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

Not reviewed. Adjust one layout test to the results of the Leopard Intel slave.

16:37 Changeset [51811] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

Not reviewed. Turn Qt slave green again, after landing filters.

16:10 Changeset [51810] by zimmermann@webkit.org
  • 2 edits in trunk/WebCore

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Force JS binding regeneration, eventually fixing the broken tests on the bots.

  • bindings/scripts/CodeGeneratorJS.pm:
16:04 Changeset [51809] by mitz@apple.com
  • 5 edits in trunk/LayoutTests

Corrected expected results

  • platform/mac/fast/repaint/block-selection-gap-stale-cache-2-expected.checksum:
  • platform/mac/fast/repaint/block-selection-gap-stale-cache-2-expected.png:
  • platform/mac/fast/repaint/block-selection-gap-stale-cache-expected.checksum:
  • platform/mac/fast/repaint/block-selection-gap-stale-cache-expected.png:
16:02 Changeset [51808] by mitz@apple.com
  • 1 edit
    1 move in trunk/LayoutTests

Rubber-stamped by Simon Fraser.

Re-enabled this test now that <rdar://problem/7450481> is fixed.

  • fast/repaint/block-selection-gap-in-composited-layer.html: Copied from LayoutTests/fast/repaint/block-selection-gap-in-composited-layer.html.disabled.
  • fast/repaint/block-selection-gap-in-composited-layer.html.disabled: Removed.
16:00 Changeset [51807] by bweinstein@apple.com
  • 1 edit in trunk/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def

Add new exported symbols to finally fix the build.

15:53 Changeset [51806] by barraclough@apple.com
  • 2 edits in trunk/WebKit/win
  • WebView.cpp:

(WebView::stringByEvaluatingJavaScriptFromString):

Reviewed by NOBODY (Windows build fix part III).

15:47 Changeset [51805] by barraclough@apple.com
  • 2 edits in trunk/WebKit/win
  • WebView.cpp:

(WebView::stringByEvaluatingJavaScriptFromString):

Reviewed by NOBODY (Windows build fix part II).

15:30 Changeset [51804] by simon.fraser@apple.com
  • 11 edits in trunk

2009-12-07 Simon Fraser <simon.fraser@apple.com>

Reviewed by Darin Adler.

<rdar://problem/7450481> One compositing test keeps DRT in "compositing mode", breaks repaint tests

The counter that WebView used to keep track of the number of enclosed WebHTMLViews using
accelerated compositing was hard to manage, and maintained incorrectly in a number of cases.
This caused one compositing test make DumpRenderTree think that all subsequent tests
were compositing too.

Replace this counter with notifications, which are only fired if a client (DRT) requests them. The
notification informs the client that a WebHTMLView entered compositing mode (or an already-
compositing WebHTML was added); it does not say when a view becomes uncomposited, or all
compositing subviews were removed, since this is tricky to get right.

Change -[WebView _isUsingAcceleratedCompositing] to manually walk the frames, and
return YES if any document view is composited.

  • WebKit.exp:
  • WebView/WebHTMLView.mm: (-[WebHTMLView close]): (-[WebHTMLView viewDidMoveToSuperview]): (-[WebHTMLView attachRootLayer:]): (-[WebHTMLView detachRootLayer]):
  • WebView/WebView.mm: (+[WebView automaticallyNotifiesObserversForKey:]): (-[WebView _postsAcceleratedCompositingNotifications]): (-[WebView _setPostsAcceleratedCompositingNotifications:]): (-[WebView _isUsingAcceleratedCompositing]):
  • WebView/WebViewData.h:
  • WebView/WebViewInternal.h:
  • WebView/WebViewPrivate.h:
15:26 Changeset [51803] by barraclough@apple.com
  • 2 edits in trunk/JavaScriptCore

Reviewed by NOBODY (Windows build fixage part I).

15:15 Changeset [51802] by zimmermann@webkit.org
  • 2 edits in trunk/WebCore

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Try to fix mac builds by touching svgtags.in - I only tried from-scratch builds.

  • svg/svgtags.in:
15:14 Changeset [51801] by barraclough@apple.com
  • 42 edits in trunk

https://bugs.webkit.org/show_bug.cgi?id=32184
Handle out-of-memory conditions with JSC Ropes with a JS exception, rather than crashing.
Switch from using fastMalloc to tryFastMalloc, pass an ExecState to record the exception on.

Reviewed by Oliver Hunt.

JavaScriptCore:

  • API/JSCallbackObjectFunctions.h:

(JSC::::toString):

  • API/JSValueRef.cpp:

(JSValueIsStrictEqual):

(JSC::BytecodeGenerator::emitEqualityOp):

  • debugger/DebuggerCallFrame.cpp:

(JSC::DebuggerCallFrame::functionName):
(JSC::DebuggerCallFrame::calculatedFunctionName):

  • interpreter/Interpreter.cpp:

(JSC::Interpreter::callEval):
(JSC::Interpreter::privateExecute):

  • jit/JITStubs.cpp:

(JSC::DEFINE_STUB_FUNCTION):

  • profiler/ProfileGenerator.cpp:

(JSC::ProfileGenerator::addParentForConsoleStart):

  • profiler/Profiler.cpp:

(JSC::Profiler::willExecute):
(JSC::Profiler::didExecute):
(JSC::Profiler::createCallIdentifier):
(JSC::createCallIdentifierFromFunctionImp):

  • profiler/Profiler.h:
  • runtime/ArrayPrototype.cpp:

(JSC::arrayProtoFuncIndexOf):
(JSC::arrayProtoFuncLastIndexOf):

  • runtime/DateConstructor.cpp:

(JSC::constructDate):

  • runtime/FunctionPrototype.cpp:

(JSC::functionProtoFuncToString):

  • runtime/InternalFunction.cpp:

(JSC::InternalFunction::name):
(JSC::InternalFunction::displayName):
(JSC::InternalFunction::calculatedDisplayName):

  • runtime/InternalFunction.h:
  • runtime/JSCell.cpp:

(JSC::JSCell::getString):

  • runtime/JSCell.h:

(JSC::JSValue::getString):

  • runtime/JSONObject.cpp:

(JSC::gap):
(JSC::Stringifier::Stringifier):
(JSC::Stringifier::appendStringifiedValue):

  • runtime/JSObject.cpp:

(JSC::JSObject::putDirectFunction):
(JSC::JSObject::putDirectFunctionWithoutTransition):
(JSC::JSObject::defineOwnProperty):

  • runtime/JSObject.h:
  • runtime/JSPropertyNameIterator.cpp:

(JSC::JSPropertyNameIterator::get):

  • runtime/JSString.cpp:

(JSC::JSString::Rope::~Rope):
(JSC::JSString::resolveRope):
(JSC::JSString::getPrimitiveNumber):
(JSC::JSString::toNumber):
(JSC::JSString::toString):
(JSC::JSString::toThisString):
(JSC::JSString::getStringPropertyDescriptor):

  • runtime/JSString.h:

(JSC::JSString::Rope::createOrNull):
(JSC::JSString::Rope::operator new):
(JSC::JSString::value):
(JSC::JSString::tryGetValue):
(JSC::JSString::getIndex):
(JSC::JSString::getStringPropertySlot):
(JSC::JSValue::toString):

  • runtime/JSValue.h:
  • runtime/NativeErrorConstructor.cpp:

(JSC::NativeErrorConstructor::NativeErrorConstructor):

  • runtime/Operations.cpp:

(JSC::JSValue::strictEqualSlowCase):

  • runtime/Operations.h:

(JSC::JSValue::equalSlowCaseInline):
(JSC::JSValue::strictEqualSlowCaseInline):
(JSC::JSValue::strictEqual):
(JSC::jsLess):
(JSC::jsLessEq):
(JSC::jsAdd):
(JSC::concatenateStrings):

  • runtime/PropertyDescriptor.cpp:

(JSC::PropertyDescriptor::equalTo):

  • runtime/PropertyDescriptor.h:
  • runtime/StringPrototype.cpp:

(JSC::stringProtoFuncReplace):
(JSC::stringProtoFuncToLowerCase):
(JSC::stringProtoFuncToUpperCase):

WebCore:

  • bindings/ScriptControllerBase.cpp:

(WebCore::ScriptController::executeIfJavaScriptURL):

  • bindings/js/JSCanvasRenderingContext2DCustom.cpp:

(WebCore::toHTMLCanvasStyle):
(WebCore::JSCanvasRenderingContext2D::setFillColor):
(WebCore::JSCanvasRenderingContext2D::setStrokeColor):
(WebCore::JSCanvasRenderingContext2D::setShadow):

  • bindings/js/ScriptCallStack.cpp:

(WebCore::ScriptCallStack::ScriptCallStack):
(WebCore::ScriptCallStack::initialize):

  • bindings/js/ScriptValue.cpp:

(WebCore::ScriptValue::getString):

  • bindings/js/ScriptValue.h:
  • bindings/js/SerializedScriptValue.cpp:

(WebCore::SerializingTreeWalker::convertIfTerminal):

  • bindings/objc/WebScriptObject.mm:

(+[WebScriptObject _convertValueToObjcValue:originRootObject:rootObject:]):

  • page/Console.cpp:

(WebCore::Console::addMessage):

WebKit/mac:

  • WebView/WebView.mm:

(aeDescFromJSValue):

15:03 Changeset [51800] by zimmermann@webkit.org
  • 249 edits
    2 adds in trunk

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Holger Hans Peter Freyther.

Turn on (SVG) Filters support, by default.
https://bugs.webkit.org/show_bug.cgi?id=32224

Only enable filters on mac for now, skipped all affected tests on the other platforms.
Will enable them one after another, to minimize breakage.

  • Configurations/FeatureDefines.xcconfig: Enable FILTERS build flag.
14:50 Changeset [51799] by eric@webkit.org
  • 5 edits in trunk

2009-12-07 Kent Tamura <tkent@chromium.org>

Reviewed by Darin Adler.

Some changes to ValidityState-typeMismatch-url.html.
https://bugs.webkit.org/show_bug.cgi?id=29913

  • Add more test cases Username, IPv6 address, mailto:, file:, tel:, fragment, and IDN.
  • Handle difference between KURL and KURLGoogle
  • fast/forms/ValidityState-typeMismatch-url-expected.txt:
  • fast/forms/script-tests/ValidityState-typeMismatch-url.js:

2009-12-07 Kent Tamura <tkent@chromium.org>

Reviewed by Darin Adler.

Fix a bug that KURL rejects a host name which has an IPv4 address
inside an IPv6 form like [::127.0.0.1].
https://bugs.webkit.org/show_bug.cgi?id=29913

  • platform/KURL.cpp: (WebCore::characterClassTable): Add IPv6Char to '.'.
14:43 Changeset [51798] by darin@chromium.org
  • 5 edits in trunk/LayoutTests

2009-12-07 Darin Fisher <darin@chromium.org>

Reviewed by Darin Adler.

http/tests/cache/subresource-failover-to-network.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=32038

Rewrite test to avoid dumpResourceLoadCallbacks, which do not
interleave predictably with the output of dumpFrameLoadCallbacks.
Other things besides resource loading can delay onload.

The newly written test uses a cookie to determine if the test has
passed.

Unskip this test for platform/win.

  • http/tests/cache/resources/subresource-failover-to-network.cgi:
  • http/tests/cache/subresource-failover-to-network-expected.txt:
  • http/tests/cache/subresource-failover-to-network.html:
  • platform/win/Skipped:
14:35 Changeset [51797] by ap@apple.com
  • 1 edit
    2 adds in trunk/LayoutTests

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=32244
Test sending and receiving small WebSocket messages of different lengths

  • websocket/tests/frame-lengths-expected.txt: Added.
  • websocket/tests/frame-lengths.html: Added.
14:27 Changeset [51796] by beidson@apple.com
  • 2 edits in trunk/WebCore

<rdar://problem/7443417> - Closing a window opened with "New windows open with: Same Page" results in crash in ::~HistoryItem

Reviewed by Sam Weinig.

The HistoryItem copy constructor has one very specific use case - to support the feature of a WebView copying another WebView's
BackForwardList. That constructor wasn't initializing its m_document pointer, leading to this crash in the destructor.

  • history/HistoryItem.cpp:

(WebCore::HistoryItem::~HistoryItem): Change the setDocument(0) to an ASSERT that the document is already 0. A HistoryItem cannot

outlive its Document and Documents will always clear the back-pointer in their ::detach() method.

(WebCore::HistoryItem::HistoryItem): Missed the m_document initializer in the copy constructor. Also remove an invalid assertion

about the original items m_cachedPage object (which is irrelevant to the copy).

13:54 Changeset [51795] by dimich@chromium.org
  • 1 edit in trunk/WebCore/ChangeLog

Not reviewed - fix typo in ChangeLog.

13:52 Changeset [51794] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-07 Yaar Schnitman <yaar@chromium.org>

Reviewed by Darin Fisher.

Chromium buildbots are not red when they fail
https://bugs.webkit.org/show_bug.cgi?id=32235

  • Scripts/build-webkit:
13:51 Changeset [51793] by dimich@chromium.org
  • 8 edits
    9 deletes in trunk/WebCore

Remove partial SharedScript implementation.
https://bugs.webkit.org/show_bug.cgi?id=32237
The patch was obtained by running "git revert" command and then un-reverting WebCore/ChangeLog.

Reviewed by Daring Adler.

  • DerivedSources.make:
  • SharedScript/SharedScriptContext.cpp: Removed.
  • SharedScript/SharedScriptContext.h: Removed.
  • SharedScript/SharedScriptContext.idl: Removed.
  • SharedScript/SharedScriptController.h: Removed.
  • SharedScript/WebKitSharedScript.cpp: Removed.
  • SharedScript/WebKitSharedScript.h: Removed.
  • SharedScript/WebKitSharedScript.idl: Removed.
  • SharedScript/WebKitSharedScriptRepository.cpp: Removed.
  • SharedScript/WebKitSharedScriptRepository.h: Removed.
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • dom/Document.cpp:

(WebCore::Document::detach):

  • dom/EventTarget.cpp:
  • dom/EventTarget.h:
  • dom/ScriptExecutionContext.h:
13:25 Changeset [51792] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-07 Yaar Schnitman <yaar@chromium.org>

Reviewed by Darin Fisher.

Typo in chromium linux builder
https://bugs.webkit.org/show_bug.cgi?id=32238

  • Scripts/webkitdirs.pm:
13:06 Changeset [51791] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Add two tests generating new results to gtks collection of tests doing so.

  • platform/gtk/Skipped:
13:05 Changeset [51790] by ap@apple.com
  • 2 edits in trunk/WebCore

Reviewed by Nikolas Zimmermann.

https://bugs.webkit.org/show_bug.cgi?id=32226
Random crashes in WebSocket tests on Leopard Debug builbot

  • platform/network/SocketStreamHandleBase.cpp: (WebCore::SocketStreamHandleBase::close): Make sure the object still exists when setting m_state.
12:32 Changeset [51789] by krit@webkit.org
  • 3 edits
    3 adds in trunk

2009-12-07 Shiki Okasaka <shiki@google.com>

Reviewed by Dirk Schulze.

Avoid zero division during SVGPaintServerPattern::setup()
https://bugs.webkit.org/show_bug.cgi?id=29912

Fix zero division bugs in SVGPaintServerPattern::setup() that occurred
if the tile of a pattern was bigger than the pattern and the pattern
size was < 0.5, and if the attribute overflow was set to visible.

Test: svg/custom/small-pattern.html

  • svg/graphics/SVGPaintServerPattern.cpp: (WebCore::SVGPaintServerPattern::setup):

2009-12-07 Dirk Schulze <krit@webkit.org>

Reviewed by Adam Barth.

If the tile of a pattern is bigger than the pattern and the
pattern size is < 0.5, the SVGPaintServer produced a division
by zero. This also only happens, if the attribute overflow is
set to visible.
This is a test with a pattern size of 0.1 and a tile size of 1.

  • svg/custom/script-tests/small-pattern.js: Added.
  • svg/custom/small-pattern-expected.txt: Added.
  • svg/custom/small-pattern.html: Added.
12:17 Changeset [51788] by eric@webkit.org
  • 6 edits in trunk

2009-12-07 Yaar Schnitman <yaar@chromium.org>

Reviewed by Darin Fisher.

fixes to build-webkit --chromium
https://bugs.webkit.org/show_bug.cgi?id=32179

  • DEPS:
  • WebKit.gyp:
  • gyp_webkit:

2009-12-07 Yaar Schnitman <yaar@chromium.org>

Reviewed by Darin Fisher.

fixes to build-webkit --chromium
https://bugs.webkit.org/show_bug.cgi?id=32179

  • Scripts/webkitdirs.pm:
11:46 Changeset [51787] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-07 Eric Seidel <eric@webkit.org>

No review, just adding two recently approved committers.

  • Scripts/modules/committers.py:
11:36 Changeset [51786] by ap@apple.com
  • 2 edits in trunk/WebKitTools
  • Scripts/run-webkit-websocketserver: Added property svn:executable.
11:36 Changeset [51785] by eric@webkit.org
  • 3 edits
    2 adds in trunk

2009-12-07 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

[v8] WebCore::WebGLArrayBufferInternal::byteLengthAttrGetter NULL pointer
https://bugs.webkit.org/show_bug.cgi?id=31889

Fixed bug in handling of zero-argument constructor call.

Test: fast/canvas/webgl/bug-31889.html

  • fast/canvas/webgl/bug-31889-expected.txt: Added.
  • fast/canvas/webgl/bug-31889.html: Added.

2009-12-07 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

[v8] WebCore::WebGLArrayBufferInternal::byteLengthAttrGetter NULL pointer
https://bugs.webkit.org/show_bug.cgi?id=31889

Fixed bug in handling of zero-argument constructor call.

Test: fast/canvas/webgl/bug-31889.html

  • bindings/v8/custom/V8WebGLArrayBufferCustom.cpp: (WebCore::CALLBACK_FUNC_DECL):
11:26 Changeset [51784] by kov@webkit.org
  • 1 edit
    2 adds in trunk/LayoutTests

Unreviewed. Landing files I forgot to add to my working copy before
committing r51734.

11:22 Changeset [51783] by eric@webkit.org
  • 4 edits in trunk

2009-12-07 Gyuyoung Kim <gyuyoung@gmail.com>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=32024
[GTK] WebKit does not compile without JAVASCRIPT_DEBUGGER

Fix build errors when javascript-debugger is disabled on GTK

  • GNUmakefile.am:

2009-12-07 Gyuyoung Kim <gyuyoung@gmail.com>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=32024
[GTK] WebKit does not compile without JAVASCRIPT_DEBUGGER

Fix build errors when javascript-debugger is disabled on GTK

  • webkit/webkitwebinspector.cpp: (webkit_web_inspector_set_property): (webkit_web_inspector_get_property):
11:09 Changeset [51782] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Don't skip websocket/tests/url-with-credential.html, it's one of earlier tests that is
to blame.

  • platform/mac/Skipped:
11:06 Changeset [51781] by bweinstein@apple.com
  • 2 edits in trunk/LayoutTests

Rubber-stamped by Steve Falkenburg.

Add one more fast/js test to the Skipped list - also due to:
https://bugs.webkit.org/show_bug.cgi?id=32228.

  • platform/win/Skipped:
10:51 Changeset [51780] by zimmermann@webkit.org
  • 3 edits in trunk/LayoutTests

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Cleaning up Skipped files in order to make more bots green, as I plan to land changes, that affect all bots (svg filters support).

Remove non-existant animations/3d directory in qt&win Skipped lists.
Skip one more test on windows (see bug #32228, already in radar, just guessing - not yet sure if it's a random crash issue though)
Skip 19 (!) tests on the Qt slave, failing since a longer time. Qt developers: Please investigate.

  • platform/qt/Skipped: Also remove some duplicates I spotted.
  • platform/win/Skipped:
10:34 Changeset [51779] by eric@webkit.org
  • 1 edit
    1 add in trunk/WebKitTools

2009-12-07 Yuzo Fujishima <yuzo@google.com>

Reviewed by Alexey Proskuryakov.

Add run-webkit-websocketserver

https://bugs.webkit.org/show_bug.cgi?id=31390

  • Scripts/run-webkit-websocketserver: Added.
10:23 Changeset [51778] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Disable two flaky SnowLeopard tests, filed bug and notified author.

Tests "plugins/document-open.html" and "plugins/mouse-events.html" are flaky on SnowLeopard
https://bugs.webkit.org/show_bug.cgi?id=32229

  • platform/mac-snowleopard/Skipped:
10:23 Changeset [51777] by sfalken@apple.com
  • 4 edits in tags/Safari-532.5.1

Merge r51773.

10:22 Changeset [51776] by mitz@apple.com
  • 7 edits
    8 adds in trunk

WebCore: Fixed <rdar://problem/7437820> Weird selection artifacts

Reviewed by Darin Adler.

Tests: fast/repaint/block-selection-gap-stale-cache-2.html

fast/repaint/block-selection-gap-stale-cache.html

Instead of caching the block selection gaps’ bounds in the RenderView at setSelection()
time, cache them in each RenderLayer at paint time. This prevents the cache from getting
stale due to layout changes and overflow scrolling.

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::selectionGapRectsForRepaint): Account for overflow scroll.
(WebCore::RenderBlock::paintSelection): Update the enclosing layer’s selection gaps bounds.

  • rendering/RenderLayer.cpp:

(WebCore::RenderLayer::addBlockSelectionGapsBounds): Added. Updates the selection gaps
bounds to include the given rect.
(WebCore::RenderLayer::clearBlockSelectionGapsBounds): Added. Recursively clears the cached
selection gaps bounds.
(WebCore::RenderLayer::repaintBlockSelectionGaps): Added. Recursively invalidates the
selection gaps bounds.

  • rendering/RenderLayer.h:
  • rendering/RenderView.cpp:

(WebCore::RenderView::setSelection): Clear the layer-level selection gaps bounds instead
of the view-level cache.
(WebCore::RenderView::clearSelection): Changed to call repaintBlockSelectionGaps().

  • rendering/RenderView.h:

LayoutTests: Tests for <rdar://problem/7437820> Weird selection artifacts

Reviewed by Darin Adler.

  • fast/repaint/block-selection-gap-stale-cache-2.html: Added.
  • fast/repaint/block-selection-gap-stale-cache.html: Added.
  • platform/mac/fast/repaint/block-selection-gap-stale-cache-2-expected.checksum: Added.
  • platform/mac/fast/repaint/block-selection-gap-stale-cache-2-expected.png: Added.
  • platform/mac/fast/repaint/block-selection-gap-stale-cache-2-expected.txt: Added.
  • platform/mac/fast/repaint/block-selection-gap-stale-cache-expected.checksum: Added.
  • platform/mac/fast/repaint/block-selection-gap-stale-cache-expected.png: Added.
  • platform/mac/fast/repaint/block-selection-gap-stale-cache-expected.txt: Added.
10:09 Changeset [51775] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed.

REGRESSION: fast/js/excessive-comma-usage.html and fast/js/math-transforms.html fail on win
https://bugs.webkit.org/show_bug.cgi?id=32228

Adding above two tests to the Skipped list on win, attempting to turn the win slave green again.

  • platform/win/Skipped:
10:06 Changeset [51774] by sfalken@apple.com
  • 1 copy in tags/Safari-532.5.1

New tag.

10:03 Changeset [51773] by sfalken@apple.com
  • 4 edits in trunk

Build fix. Be flexible about which version of ICU is used on Windows.

JavaScriptCore:

  • JavaScriptCore.vcproj/jsc/jscCommon.vsprops: Add optional xcopy commands to copy ICU 4.2.

WebKitTools:

  • DumpRenderTree/win/DumpRenderTree.vcproj: Add optional xcopy commands to copy ICU 4.2.
09:43 Changeset [51772] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Remove non-existant files in gtk/Skipped list, and exclude a printing test, without results. Should turn the gtk slave green again.

  • platform/gtk/Skipped:
09:21 Changeset [51771] by zimmermann@webkit.org
  • 2 edits in trunk/LayoutTests

2009-12-07 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Attempt to get all mac build slaves to green again. Disable crashing test, and notify the author of it to fix it.

New test "websocket/tests/url-with-credential.html" crashes in debug builds
https://bugs.webkit.org/show_bug.cgi?id=32226

  • platform/mac/Skipped:
09:10 Changeset [51770] by zimmermann@webkit.org
  • 1 edit
    1 add in trunk/LayoutTests

Not reviewed. Add missing layout test result.

09:07 Changeset [51769] by kov@webkit.org
  • 2 edits in trunk/WebCore

2009-12-07 Philippe Normand <pnormand@igalia.com>

Reviewed by Gustavo Noronha.

Fixed whitelist comment and removed useless (tags, sdp) mimetypes
from it. Also, GStreamer doesn't handle directly SMIL.

  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp: (WebCore::mimeTypeCache):
07:38 Changeset [51768] by eric@webkit.org
  • 4 edits in trunk

2009-12-07 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Xan Lopez.

https://bugs.webkit.org/show_bug.cgi?id=25415
[GTK][ATK] Please implement support for get_text_at_offset

Eliminate the segfaults which occur when accessing the text interface now
implemented by text controls.

  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (getPangoLayoutForAtk):

2009-12-07 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Xan Lopez.

https://bugs.webkit.org/show_bug.cgi?id=25415
[GTK][ATK] Please implement support for get_text_at_offset

Eliminate the segfaults which occur when accessing the text interface now
implemented by text controls.

  • tests/testatk.c (test_webkit_atk_get_text_at_offset_textarea): (test_webkit_atk_get_text_at_offset_text_input): (main):
07:31 Changeset [51767] by eric@webkit.org
  • 2 edits in trunk/WebKit/qt

2009-12-07 Benjamin Poulain <benjamin.poulain@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Improve the autotests of QtWebkit
https://bugs.webkit.org/show_bug.cgi?id=32216

Remove the calls to qWait() of the autotest of QWebView

  • tests/qwebview/tst_qwebview.cpp: (tst_QWebView::reusePage):
07:02 Changeset [51766] by eric@webkit.org
  • 3 edits in trunk/WebKit/chromium

2009-12-07 Mikhail Naganov <mnaganov@chromium.org>

Reviewed by Pavel Feldman.

[Chromium] Add declarations for DevTools message dispatching on IO thread
(see http://codereview.chromium.org/460018)

https://bugs.webkit.org/show_bug.cgi?id=32217

  • public/WebDevToolsAgent.h:
  • public/WebDevToolsAgentClient.h:
06:52 Changeset [51765] by mjs@apple.com
  • 2 edits in trunk/JavaScriptCore

2009-12-07 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver Hunt.

op_loop_if_less JIT codegen is broken for 64-bit
https://bugs.webkit.org/show_bug.cgi?id=32221

  • jit/JITOpcodes.cpp: (JSC::JIT::emit_op_loop_if_false): Fix codegen in this version - test was backwards.
06:35 Changeset [51764] by kov@webkit.org
  • 5 edits
    6 adds in trunk

Reviewed by Xan Lopez.

Build the new API test.

[GTK] REGRESSION: webkit thinks it can render PDFs
https://bugs.webkit.org/show_bug.cgi?id=32183

  • GNUmakefile.am:

WebCore

Reviewed by Xan Lopez.

Turn the MediaPlayer supported types blacklist into a whitelist

[GTK] REGRESSION: webkit thinks it can render PDFs
https://bugs.webkit.org/show_bug.cgi?id=32183

Covered by API test.

  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp: (WebCore::mimeTypeCache):

WebKit/gtk

Reviewed by Xan Lopez.

New test to make sure we do not regress this feature anymore. All
test files were created by me.

[GTK] REGRESSION: webkit thinks it can render PDFs
https://bugs.webkit.org/show_bug.cgi?id=32183

  • tests/resources/test.html: Added.
  • tests/resources/test.ogg: Added.
  • tests/resources/test.pdf: Added.
  • tests/resources/test.txt: Added.
  • tests/testmimehandling.c: Added. (server_callback): (idle_quit_loop_cb): (mime_type_policy_decision_requested_cb): (test_mime_type): (test_mime_pdf): (test_mime_html): (test_mime_text): (test_mime_ogg): (main):
06:34 Changeset [51763] by eric@webkit.org
  • 2 edits in trunk/WebKit/qt

2009-12-07 Benjamin Poulain <benjamin.poulain@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Improve the autotests of QtWebkit
https://bugs.webkit.org/show_bug.cgi?id=32216

Refactor tst_qwebframe to remove qWait() and use
the function waitForSignal() from util.h

  • tests/qwebframe/tst_qwebframe.cpp:
06:27 Changeset [51762] by eric@webkit.org
  • 3 edits
    3 adds in trunk

2009-12-07 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Xan Lopez.

https://bugs.webkit.org/show_bug.cgi?id=25524
[Gtk] Expose the title attribute to assistive technologies

  • platform/gtk/accessibility/title-and-alt.html: Added.
  • platform/gtk/accessibility/title-and-alt-expected.txt: Added.

2009-12-07 Joanmarie Diggs <joanmarie.diggs@gmail.com>

Reviewed by Xan Lopez.

https://bugs.webkit.org/show_bug.cgi?id=25524
[Gtk] Expose the title attribute to assistive technologies

Expose 'alt' attribute from images as accessible name.
Expose the 'title' core HTML attribute as accessible description.
This is a modified version of the original fix submitted by Mario Sanchez Prada,
adjusted so that it doesn't impact other platforms.

  • accessibility/gtk/AccessibilityObjectWrapperAtk.cpp: (webkit_accessible_get_name): (webkit_accessible_get_description):
06:13 Changeset [51761] by eric@webkit.org
  • 3 edits in trunk/WebKit/qt

2009-12-07 Benjamin Poulain <benjamin.poulain@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Improve the autotests of QtWebkit
https://bugs.webkit.org/show_bug.cgi?id=32216

Refactor the test of QGraphicsWebView:
-make waitForSignal() available to all the tests.
-remove QTest::qWait()

  • tests/qgraphicswebview/tst_qgraphicswebview.cpp: (tst_QGraphicsWebView::crashOnViewlessWebPages):
  • tests/util.h: (waitForSignal):
05:50 Changeset [51760] by oliver@apple.com
  • 5 edits in trunk

Object.create fails if properties on the descriptor are getters
https://bugs.webkit.org/show_bug.cgi?id=32219

Reviewed by Maciej Stachowiak.

Correctly initialise the PropertySlots with the descriptor object.

05:44 Changeset [51759] by girish@forwardbias.in
  • 2 edits in trunk/WebKit/qt

[Qt] Plugins: Force windowless mode when there is no native window handle

Reviewed by Simon Hausmann.

Inject wmode=opaque while instantiating the plugin for the case when the
webpage is not backed by a native window handle.

https://bugs.webkit.org/show_bug.cgi?id=32059

05:29 Changeset [51758] by vestbo@webkit.org
  • 4 edits in trunk/WebCore

[Qt] Fix support for generic font families

Reviewed by Simon Hausmann.

By not propagating the resolved font family in FontCacheQt and
only concidering the font description we ended up passing the
generic font families to Qt directly. Since most systems don't
have a mapping for these font families we ended up using the
default font in most cases.

  • platform/graphics/qt/FontCacheQt.cpp:
  • platform/graphics/qt/FontPlatformData.h:
  • platform/graphics/qt/FontPlatformDataQt.cpp:
05:06 Changeset [51757] by oliver@apple.com
  • 9 edits
    1 add in trunk

Object.getOwnPropertyDescriptor() allows cross-frame access
https://bugs.webkit.org/show_bug.cgi?id=32119

Reviewed by Maciej Stachowiak.

Make all implementations of getOwnPropertyDescriptor that have
cross domain restrictions simply fail immediately on cross domain
access, rather than trying to mimic the behaviour of normal
property access.

Test: http/tests/security/cross-frame-access-getOwnPropertyDescriptor.html

04:36 Changeset [51756] by steveblock@google.com
  • 9 edits
    3 adds in trunk/WebCore

Moves JSC-specific functions from jni_utility and moves them to new jsc/jni_utility_private files.
https://bugs.webkit.org/show_bug.cgi?id=32157

Reviewed by Adam Barth.

Build fix only, no new tests.

  • WebCore.xcodeproj/project.pbxproj: Modified. Add jni_utility_private.[cpp|h].
  • bridge/jni/jni_instance.cpp: Modified. Include jni_utility_private.h for JSC.
  • bridge/jni/jni_jsobject.mm: Modified. Include jni_utility_private.h for JSC.
  • bridge/jni/jni_objc.mm: Modified. Include jni_utility_private.h for JSC.
  • bridge/jni/jni_runtime.cpp: Modified. Include jni_utility_private.h for JSC.
  • bridge/jni/jni_utility.cpp: Modified. Removed convertValueToJValue and convertArrayInstanceToJavaArray.
  • bridge/jni/jni_utility.h: Modified. Removed convertValueToJValue and dispatchJNICall.
  • bridge/jni/jsc: Added.
  • bridge/jni/jsc/jni_utility_private.cpp: Added.

(JSC::Bindings::convertArrayInstanceToJavaArray):
(JSC::Bindings::convertValueToJValue):

  • bridge/jni/jsc/jni_utility_private.h: Added. Header for convertValueToJValue and dispatchJNICall.
  • platform/android/TemporaryLinkStubs.cpp: Modified. Include jni_utility_private.h for JSC.
04:04 Changeset [51755] by krit@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-07 Dirk Schulze <krit@webkit.org>

Not reviewed, adding myself to the reviewers list.

  • Scripts/modules/committers.py:
00:51 QtWebKitMediaElementSupport edited by vestbo@webkit.org
(diff)
00:50 QtWebKitMediaElementSupport edited by vestbo@webkit.org
(diff)
00:46 QtWebKitMediaElementSupport created by vestbo@webkit.org
00:46 QtWebKit edited by vestbo@webkit.org
(diff)

12/06/09:

23:53 Changeset [51754] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-06 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Add an API for uploading results files to StatusBot
https://bugs.webkit.org/show_bug.cgi?id=32210

Add Content-Type: plain/text which was forgotten
from the previous commit when I landed with land-patches
instead of land-diff.

  • QueueStatusServer/queue_status.py:
23:45 Changeset [51753] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-06 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Add an API for uploading results files to StatusBot
https://bugs.webkit.org/show_bug.cgi?id=32210

  • Scripts/modules/statusbot.py:
23:42 Changeset [51752] by eric@webkit.org
  • 3 edits in trunk/WebKitTools

2009-12-06 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Status Server needs a way to handle uploaded results
https://bugs.webkit.org/show_bug.cgi?id=32209

  • QueueStatusServer/queue_status.py: Add a ShowResults (results/*) command
  • QueueStatusServer/update_status.html: Add file upload.
23:06 Changeset [51751] by hamaji@chromium.org
  • 2 edits in trunk/BugsSite

2009-12-06 Shinichiro Hamaji <hamaji@chromium.org>

Unreviewed. Revert r51748.

Bugzilla should show images in git patches
https://bugs.webkit.org/show_bug.cgi?id=31395

  • PrettyPatch/PrettyPatch.rb:
22:42 Changeset [51750] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-06 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] BuildAttachment should not check the builders
https://bugs.webkit.org/show_bug.cgi?id=32207

This is code that got copied here by accident when the class was created.

  • Scripts/modules/commands/download.py:
22:42 Changeset [51749] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-06 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Make LandingSequence.update aware of ports
https://bugs.webkit.org/show_bug.cgi?id=32208

This is required to make build work on the Chromium port because
Chromium has a custom update-webkit.

  • Scripts/modules/landingsequence.py:
22:42 Changeset [51748] by hamaji@chromium.org
  • 2 edits in trunk/BugsSite

2009-12-06 Shinichiro Hamaji <shinichiro.hamaji@gmail.com>

Reviewed by David Kilzer.

Bugzilla should show images in git patches
https://bugs.webkit.org/show_bug.cgi?id=31395

Show images in git patches using git-apply.

  • PrettyPatch/PrettyPatch.rb:
21:20 Changeset [51747] by abarth@webkit.org
  • 4 edits in trunk/WebKitTools

2009-12-06 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Add early warning system commands to bugzilla-tool.

  • Scripts/bugzilla-tool:
  • Scripts/modules/landingsequence.py:
  • Scripts/modules/webkitport.py:
21:11 Changeset [51746] by abarth@webkit.org
  • 2 edits
    1 add in trunk/WebKitTools

2009-12-06 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Implement Qt EarlyWarningSystem and Chromium EarlyWarningSystem
https://bugs.webkit.org/show_bug.cgi?id=32205

  • Scripts/modules/commands/early_warning_system.py: Added.
  • Scripts/modules/commands/queues.py:
19:27 Changeset [51745] by mitz@apple.com
  • 2 edits in trunk/WebKitTools

REGRESSION (r51728): update-webkit fails when the current directory is
not inside a Subversion working copy
https://bugs.webkit.org/show_bug.cgi?id=32204

Reviewed by Adele Peterson.

  • Scripts/update-webkit: Invoke isSVN() in the correct working

directory.

18:45 Changeset [51744] by eric@webkit.org
  • 5 edits in trunk/LayoutTests

2009-12-06 Kent Tamura <tkent@chromium.org>

Reviewed by Darin Adler.

Improve portability and readability of table-percent-height* tests.
https://bugs.webkit.org/show_bug.cgi?id=31858

The purpose of these tests is to confirm that an element with
height:N% inside a table cell has non-zero height. We don't need
to check concrete pixel sizes of form controls.

This updates the tests so that they check heights are non-zero and
75% heights are same as corresponding 100% heights or actually 75%.

  • fast/replaced/table-percent-height-expected.txt:
  • fast/replaced/table-percent-height-text-controls-expected.txt:
  • fast/replaced/table-percent-height-text-controls.html:
  • fast/replaced/table-percent-height.html:
18:19 Changeset [51743] by ukai@chromium.org
  • 6 edits
    8 adds in trunk/LayoutTests

2009-12-06 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Alexey Proskuryakov.

Add tests for cross-origin WebSocket usage
https://bugs.webkit.org/show_bug.cgi?id=32055

  • websocket/tests/cross-origin-expected.txt: Added.
  • websocket/tests/cross-origin.html: Added.
  • websocket/tests/fixed-origin_wsh.py: Added.
  • websocket/tests/handshake-fail-by-cross-origin-expected.txt: Added.
  • websocket/tests/handshake-fail-by-cross-origin.html: Added.
  • websocket/tests/origin-test_wsh.py: Added.
  • websocket/tests/script-tests/cross-origin.js: Added.
  • websocket/tests/script-tests/frame-length-longer-than-buffer.js:
  • websocket/tests/script-tests/frame-length-skip.js:
  • websocket/tests/script-tests/handshake-error.js:
  • websocket/tests/script-tests/handshake-fail-by-cross-origin.js: Added.
  • websocket/tests/script-tests/simple.js:
  • websocket/tests/simple-expected.txt:
17:09 Changeset [51742] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-12-06 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Rename AbstractTryQueue to AbstractReviewQueue
https://bugs.webkit.org/show_bug.cgi?id=32202

  • Scripts/modules/commands/queues.py:
09:40 Changeset [51741] by eric@webkit.org
  • 3 edits in trunk/WebKit/haiku

2009-12-06 Maxime Simon <simon.maxime@gmail.com>

Reviewed by Adam Barth.

Build fix. Add in FrameLoaderClientHaiku the declaration of three functions introduced in r51644.

  • WebCoreSupport/FrameLoaderClientHaiku.cpp: (WebCore::FrameLoaderClientHaiku::dispatchDidPushStateWithinPage): (WebCore::FrameLoaderClientHaiku::dispatchDidReplaceStateWithinPage): (WebCore::FrameLoaderClientHaiku::dispatchDidPopStateWithinPage):
  • WebCoreSupport/FrameLoaderClientHaiku.h:
03:37 Changeset [51740] by oliver@apple.com
  • 2 edits in trunk/WebCore

texImage2D pixel junk for transparency
https://bugs.webkit.org/show_bug.cgi?id=32188

Reviewed by Maciej Stachowiak

Use kCGBlendModeCopy when drawing an image to the intermediate context
used to create a GL texture.

No test as we don't currently have any mechanism to retrieve pixel data
from the webgl context.

02:30 Changeset [51739] by mjs@apple.com
  • 2 edits in trunk/JavaScriptCore

Not reviewed, build fix.

Actually tested 64-bit *and* 32-bit build this time.

  • jit/JITOpcodes.cpp:

(JSC::JIT::emit_op_loop_if_false):

02:09 Changeset [51738] by mjs@apple.com
  • 2 edits in trunk/JavaScriptCore

Not reviewed, build fix.

Really really fix 64-bit build for prior patch (actually tested this time).

  • jit/JITOpcodes.cpp:

(JSC::JIT::emit_op_loop_if_false):
(JSC::JIT::emitSlow_op_loop_if_false):

01:57 Changeset [51737] by mjs@apple.com
  • 2 edits in trunk/JavaScriptCore

Not reviewed, build fix.

Really fix 64-bit build for prior patch.

  • jit/JITArithmetic.cpp:

(JSC::JIT::emitSlow_op_jless):

01:51 Changeset [51736] by mjs@apple.com
  • 2 edits in trunk/JavaScriptCore

Not reviewed, build fix.

Fix 64-bit build for prior patch.

  • jit/JITOpcodes.cpp:

(JSC::JIT::emitSlow_op_loop_if_less):

01:42 Changeset [51735] by mjs@apple.com
  • 17 edits in trunk

2009-12-05 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver Hunt.

conway benchmark spends half it's time in op_less (jump fusion fails)
https://bugs.webkit.org/show_bug.cgi?id=32190

<1% speedup on SunSpider and V8
2x speedup on "conway" benchmark


Two optimizations:

1) Improve codegen for logical operators &&,
and ! in a condition context


When generating code for combinations of &&,
and !, in a

condition context (i.e. in an if statement or loop condition), we
used to produce a value, and then separately jump based on its
truthiness. Now we pass the false and true targets in, and let the
logical operators generate jumps directly. This helps in four
ways:

a) Individual clauses of a short-circuit logical operator can now
jump directly to the then or else clause of an if statement (or to
the top or exit of a loop) instead of jumping to a jump.


b) It used to be that jump fusion with the condition of the first
clause of a logical operator was inhibited, because the register
was ref'd to be used later, in the actual condition jump; this no
longer happens since a jump straight to the final target is
generated directly.

c) It used to be that jump fusion with the condition of the second
clause of a logical operator was inhibited, because there was a
jump target right after the second clause and before the actual
condition jump. But now it's no longer necessary for the first
clause to jump there so jump fusion is not blocked.

d) We avoid generating excess mov statements in some cases.


As a concrete example this source:


if (!((x < q && y < q)
(t < q && z < q))) {

...

}


Used to generate this bytecode:


[ 34] less r1, r-15, r-19
[ 38] jfalse r1, 7(->45)
[ 41] less r1, r-16, r-19
[ 45] jtrue r1, 14(->59)
[ 48] less r1, r-17, r-19
[ 52] jfalse r1, 7(->59)
[ 55] less r1, r-18, r-19
[ 59] jtrue r1, 17(->76)


And now generates this bytecode (also taking advantage of the second optimization below):


[ 34] jnless r-15, r-19, 8(->42)
[ 38] jless r-16, r-19, 26(->64)
[ 42] jnless r-17, r-19, 8(->50)
[ 46] jless r-18, r-19, 18(->64)


Note the jump fusion and the fact that there's less jump
indirection - three of the four jumps go straight to the target
clause instead of indirecting through another jump.


2) Implement jless opcode to take advantage of the above, since we'll now often generate
a less followed by a jtrue where fusion is not forbidden.


  • parser/Nodes.h: (JSC::ExpressionNode::hasConditionContextCodegen): Helper function to determine whether a node supports special conditional codegen. Return false as this is the default. (JSC::ExpressionNode::emitBytecodeInConditionContext): Assert not reached - only really defined for nodes that do have conditional codegen. (JSC::UnaryOpNode::expr): Add const version. (JSC::LogicalNotNode::hasConditionContextCodegen): Returne true only if subexpression supports it. (JSC::LogicalOpNode::hasConditionContextCodegen): Return true.
  • parser/Nodes.cpp: (JSC::LogicalNotNode::emitBytecodeInConditionContext): Implemented - just swap the true and false targets for the child node. (JSC::LogicalOpNode::emitBytecodeInConditionContext): Implemented - handle jumps directly, improving codegen quality. Also handles further nested conditional codegen. (JSC::ConditionalNode::emitBytecode): Use condition context codegen when available. (JSC::IfNode::emitBytecode): ditto (JSC::IfElseNode::emitBytecode): ditto (JSC::DoWhileNode::emitBytecode): ditto (JSC::WhileNode::emitBytecode): ditto (JSC::ForNode::emitBytecode): ditto
  • bytecode/Opcode.h:
  • Added loop_if_false opcode - needed now that falsey jumps can be backwards.
  • Added jless opcode to take advantage of new fusion opportunities.
  • bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): Handle above.
  • bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitJumpIfTrue): Add peephole for less + jtrue ==> jless. (JSC::BytecodeGenerator::emitJumpIfFalse): Add handling of backwrds falsey jumps.
  • bytecompiler/BytecodeGenerator.h: (JSC::BytecodeGenerator::emitNodeInConditionContext): Wrapper to handle tracking of overly deep expressions etc.
  • interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): Implement the two new opcodes (loop_if_false, jless).
  • jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): Implement JIT support for the two new opcodes. (JSC::JIT::privateCompileSlowCases): ditto
  • jit/JIT.h:
  • jit/JITArithmetic.cpp: (JSC::JIT::emit_op_jless): (JSC::JIT::emitSlow_op_jless): ditto (JSC::JIT::emitBinaryDoubleOp): ditto
  • jit/JITOpcodes.cpp: (JSC::JIT::emitSlow_op_loop_if_less): ditto (JSC::JIT::emit_op_loop_if_false): ditto (JSC::JIT::emitSlow_op_loop_if_false): ditto
  • jit/JITStubs.cpp:
  • jit/JITStubs.h: (JSC::):

2009-12-05 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver Hunt.

conway benchmark spends half it's time in op_less (jump fusion fails)
https://bugs.webkit.org/show_bug.cgi?id=32190

  • fast/js/codegen-loops-logical-nodes-expected.txt:
  • fast/js/script-tests/codegen-loops-logical-nodes.js: Update to test some newly sensitive cases of codegen that were not already covered.
Note: See TracTimeline for information about the timeline view.