WebKit-https.git
8 years agoUnreviewed GTK gardening.
zandobersek@gmail.com [Mon, 1 Oct 2012 15:56:33 +0000 (15:56 +0000)]
Unreviewed GTK gardening.

Adding platform-specific baselines for fast/canvas/canvas-render-layer.html
as required after r129934.
Skipping a calendar picker test that times out, the feature is not yet supported
on the GTK port.
Adding a flaky failure expectation for fast/dom/inline-event-attributes-release.html
that started occurring after r130000.

* platform/gtk/TestExpectations:
* platform/gtk/fast/canvas/canvas-render-layer-expected.png: Added.
* platform/gtk/fast/canvas/canvas-render-layer-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130051 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Rely on the general implementation for detecting the Apache server path.
rakuco@webkit.org [Mon, 1 Oct 2012 15:26:38 +0000 (15:26 +0000)]
[GTK] Rely on the general implementation for detecting the Apache server path.
https://bugs.webkit.org/show_bug.cgi?id=98033

Reviewed by Martin Robinson.

The implemenentation of _path_to_apache() in base.py already
checks for the same paths gtk.py tries; the only difference is
that all of the are tried regardless of the current distro.

Doing so is more portable, and lets us remove some distro-checking
from the code.

* Scripts/webkitpy/layout_tests/port/gtk.py:
(GtkPort._path_to_apache): Remove.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130050 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed. Chromium build fix.
yurys@chromium.org [Mon, 1 Oct 2012 15:26:21 +0000 (15:26 +0000)]
Unreviewed. Chromium build fix.

* bindings/v8/V8DOMMap.h: included Node.h as reportMemoryUsage now uses Node definition.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130049 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: provide memory instrumentation for HashMap
yurys@chromium.org [Mon, 1 Oct 2012 14:46:25 +0000 (14:46 +0000)]
Web Inspector: provide memory instrumentation for HashMap
https://bugs.webkit.org/show_bug.cgi?id=98005

Reviewed by Pavel Feldman.

Source/WebCore:

Updated all call sites of MemoryInstrumentation::addHashMap to use generic
method of reporting memory footprint instead.

* bindings/v8/ScopedDOMDataStore.cpp:
* bindings/v8/V8Binding.cpp:
(WebCore::StringCache::reportMemoryUsage):
* bindings/v8/V8DOMMap.h:
* bindings/v8/V8PerIsolateData.cpp:
(WebCore::V8PerIsolateData::reportMemoryUsage):
* css/CSSImageGeneratorValue.cpp:
(WebCore::CSSImageGeneratorValue::reportBaseClassMemoryUsage):
* css/PropertySetCSSStyleDeclaration.cpp:
(WebCore::PropertySetCSSStyleDeclaration::reportMemoryUsage):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::collectMatchingRulesForList):
* css/StyleSheetContents.cpp:
(WebCore::StyleSheetContents::reportMemoryUsage):
* dom/Document.cpp:
(WebCore::Document::reportMemoryUsage):
* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::reportMemoryUsage):
* loader/cache/CachedResourceLoader.cpp:
(WebCore::CachedResourceLoader::reportMemoryUsage):
* loader/cache/MemoryCache.cpp:
(WebCore::MemoryCache::reportMemoryUsage):
* platform/network/ResourceRequestBase.cpp:
(WebCore::ResourceRequestBase::reportMemoryUsage):
* platform/network/ResourceResponseBase.cpp:
(WebCore::ResourceResponseBase::reportMemoryUsage):
* rendering/style/StyleRareNonInheritedData.cpp:

Source/WebKit/chromium:

Added unit tests for HashMap memory instrumentation.

* tests/MemoryInstrumentationTest.cpp:
(WTF):

Source/WTF:

Extracted HashMap memory instrumentation into its own file. The map's content
elements will be automatically traversed if their types are supported
by the memory instrumentation. No need to call special method for hash map
fields any more.

* GNUmakefile.list.am:
* WTF.gypi:
* WTF.pro:
* WTF.vcproj/WTF.vcproj:
* wtf/MemoryInstrumentation.h:
(MemoryInstrumentation):
(WTF):
* wtf/MemoryInstrumentationHashMap.h: Added.
(WTF):
(WTF::SequenceMemoryInstrumentationTraits::reportMemoryUsage):
(WTF::reportMemoryUsage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130048 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoPut implementation details of StyleBuilder.cpp into anonymous namespace
yurys@chromium.org [Mon, 1 Oct 2012 14:25:19 +0000 (14:25 +0000)]
Put implementation details of StyleBuilder.cpp into anonymous namespace
https://bugs.webkit.org/show_bug.cgi?id=98028

Reviewed by Pavel Feldman.

All types that are declared and used only inside StyleBuilder were moved
into anonymous namespace to avoid name conflicts with the rest of WebCore.

* css/StyleBuilder.cpp:
(WebCore::StyleBuilder::StyleBuilder): renamed BorderImageType::Image into BorderImageType::BorderImage
as otherwise there is an ambiguity at placess where setPropertyHandler is called.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130047 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Save original uri for downloaded files
commit-queue@webkit.org [Mon, 1 Oct 2012 14:18:58 +0000 (14:18 +0000)]
[GTK] Save original uri for downloaded files
https://bugs.webkit.org/show_bug.cgi?id=95188

Patch by Claudio Saavedra <csaavedra@igalia.com> on 2012-10-01
Reviewed by Carlos Garcia Campos.

gvfs stores metadata locally, and this information can later be
used by file management applications. Based on a patch by
Alexander Larsson <alexl@redhat.com>.

* webkit/webkitdownload.cpp:
(webkit_download_open_stream_for_uri): Save the download-uri as
file metadata.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130046 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt][WK2] Device pixel ratio lost upon relaunch of the web process
abecsi@webkit.org [Mon, 1 Oct 2012 14:12:55 +0000 (14:12 +0000)]
[Qt][WK2] Device pixel ratio lost upon relaunch of the web process
https://bugs.webkit.org/show_bug.cgi?id=97908

Reviewed by Jocelyn Turcotte.

Use setIntrinsicDeviceScaleFactor instead of setCustomDeviceScaleFactor
when setting the device pixel ratio to make the setting permanent.

* UIProcess/API/qt/qquickwebview.cpp:
(QQuickWebViewPrivate::didRelaunchProcess):
(QQuickWebViewExperimental::setDevicePixelRatio):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130045 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Remove EFL-specific expectations for fast/regions/webkit-named-flow-*
dominik.rottsches@intel.com [Mon, 1 Oct 2012 14:07:57 +0000 (14:07 +0000)]
[EFL] Remove EFL-specific expectations for fast/regions/webkit-named-flow-*
https://bugs.webkit.org/show_bug.cgi?id=98029

Unreviewed, EFL gardening.

Removing redundant platform-specific test results for webkit-named-flow-* cases.
Fixes bot redness after API rename in r130041.

* platform/efl/fast/regions/webkit-named-flow-collection-crash-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-collection-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-event-add-to-flow-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-event-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-event-no-regions-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-event-remove-from-dom-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-event-remove-from-flow-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-event-target-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-event-to-null-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-existing-flow-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-first-empty-region-index-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-flow-added-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-get-content-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-get-regions-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-invalid-name-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-modified-flow-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-name-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-non-existing-flow-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-overset-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-removed-flow-expected.txt: Removed.
* platform/efl/fast/regions/webkit-named-flow-same-object-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130044 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[BlackBerry] Remove suppressing of VKB hiding when processing changes.
mifenton@rim.com [Mon, 1 Oct 2012 14:04:09 +0000 (14:04 +0000)]
[BlackBerry] Remove suppressing of VKB hiding when processing changes.
https://bugs.webkit.org/show_bug.cgi?id=98023

Reviewed by Rob Buis.

PR 215881.

Allow JS to blur the input field while processing key handling.

Reviewed Internally by Chris Hutten-Czapski

* WebKitSupport/InputHandler.cpp:
(BlackBerry::WebKit::InputHandler::notifyClientOfKeyboardVisibilityChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130043 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed gardening -- clean up expectations for passing tests.
caseq@chromium.org [Mon, 1 Oct 2012 12:52:06 +0000 (12:52 +0000)]
Unreviewed gardening -- clean up expectations for passing tests.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130042 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[CSS Regions] Remove the deprecated API Document.webkitGetFlowByName
commit-queue@webkit.org [Mon, 1 Oct 2012 12:46:56 +0000 (12:46 +0000)]
[CSS Regions] Remove the deprecated API Document.webkitGetFlowByName
https://bugs.webkit.org/show_bug.cgi?id=97657

Patch by Andrei Bucur <abucur@adobe.com> on 2012-10-01
Reviewed by Andreas Kling.

Source/WebCore:

The Document.getFlowByName() API has been deprecated in favor of the NamedFlowCollection.namedItem(DOMString).
Link to spec: http://www.w3.org/TR/css3-regions/#the-namedflow-interface

Tests: The old tests have been adapted to use the new API.

* dom/Document.cpp:
(WebCore):
* dom/Document.h:
(Document):
* dom/Document.idl:

LayoutTests:

To minimize the code change impact, I've created a new method in helper.js getFlowByName() that emulates the
behaviour of Document.webkitGetFlowByName(). All the occurrences of Document.webkitGetFlowByName in the tests/expected results
have been replaced with getFlowByName.

* fast/regions/get-regions-by-content-horiz-bt.html:
* fast/regions/get-regions-by-content-horiz-tb.html:
* fast/regions/get-regions-by-content-vert-lr.html:
* fast/regions/get-regions-by-content-vert-rl.html:
* fast/regions/get-regions-by-content.html:
* fast/regions/get-regions-by-content2.html:
* fast/regions/resources/helper.js:
(getFlowByName):
* fast/regions/webkit-named-flow-event-add-to-flow.html:
* fast/regions/webkit-named-flow-event-no-regions.html:
* fast/regions/webkit-named-flow-event-remove-from-dom.html:
* fast/regions/webkit-named-flow-event-remove-from-flow.html:
* fast/regions/webkit-named-flow-event-target-expected.txt:
* fast/regions/webkit-named-flow-event-target.html:
* fast/regions/webkit-named-flow-event-to-null.html:
* fast/regions/webkit-named-flow-event.html:
* fast/regions/webkit-named-flow-existing-flow.html:
* fast/regions/webkit-named-flow-first-empty-region-index.html:
* fast/regions/webkit-named-flow-flow-added.html:
* fast/regions/webkit-named-flow-get-content-expected.txt:
* fast/regions/webkit-named-flow-get-content.html:
* fast/regions/webkit-named-flow-get-regions.html:
* fast/regions/webkit-named-flow-invalid-name.html:
* fast/regions/webkit-named-flow-modified-flow.html:
* fast/regions/webkit-named-flow-name.html:
* fast/regions/webkit-named-flow-non-existing-flow.html:
* fast/regions/webkit-named-flow-overset.html:
* fast/regions/webkit-named-flow-removed-flow.html:
* fast/regions/webkit-named-flow-same-object.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130041 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL][WK2] fast/history/multiple-classes-visited.html should be skipped
commit-queue@webkit.org [Mon, 1 Oct 2012 12:40:03 +0000 (12:40 +0000)]
[EFL][WK2] fast/history/multiple-classes-visited.html should be skipped
https://bugs.webkit.org/show_bug.cgi?id=98017

Unreviewed EFL gardening.

fast/history/multiple-classes-visited.html is globally skipped for WK2
but marked as explicit PASS for EFL port. Unfortunately, this test is
failing for EFL port as well so we need to remove the explicit PASS
expectation.

Patch by Christophe Dumez <christophe.dumez@intel.com> on 2012-10-01

* platform/efl-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130040 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoCheck that __cplusplus is defined before comparing its value
commit-queue@webkit.org [Mon, 1 Oct 2012 12:28:45 +0000 (12:28 +0000)]
Check that __cplusplus is defined before comparing its value
https://bugs.webkit.org/show_bug.cgi?id=98015

Patch by Alberto Garcia <agarcia@igalia.com> on 2012-10-01
Reviewed by Xan Lopez.

If __cplusplus is not defined is interpreted as having the value
0, but it produces a compilation warning with -Wundef.

This is the case with some API tests that are written in C
(JSNode.c, JSNodeList.c, minidom.c).

* wtf/Compiler.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130039 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Generate baseline for fast/canvas/canvas-render-layer.html
commit-queue@webkit.org [Mon, 1 Oct 2012 12:24:39 +0000 (12:24 +0000)]
[EFL] Generate baseline for fast/canvas/canvas-render-layer.html
https://bugs.webkit.org/show_bug.cgi?id=98018

Unreviewed EFL gardening.

Generate EFL baseline for fast/canvas/canvas-render-layer.html
test.

Patch by Christophe Dumez <christophe.dumez@intel.com> on 2012-10-01

* platform/efl/fast/canvas/canvas-render-layer-expected.png: Added.
* platform/efl/fast/canvas/canvas-render-layer-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130038 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] fast/replaced/no-focus-ring-* tests should be skipped
commit-queue@webkit.org [Mon, 1 Oct 2012 11:56:07 +0000 (11:56 +0000)]
[EFL] fast/replaced/no-focus-ring-* tests should be skipped
https://bugs.webkit.org/show_bug.cgi?id=98012

Unreviewed EFL gardening.

Skip again fast/replaced/no-focus-ring-* tests as they are
failing on the build bots due to X not running.

Patch by Christophe Dumez <christophe.dumez@intel.com> on 2012-10-01

* platform/efl-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130037 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] WebKitDownload: use more of GOwnPtr/GRefPtr
commit-queue@webkit.org [Mon, 1 Oct 2012 11:30:40 +0000 (11:30 +0000)]
[GTK] WebKitDownload: use more of GOwnPtr/GRefPtr
https://bugs.webkit.org/show_bug.cgi?id=98009

Patch by Claudio Saavedra <csaavedra@igalia.com> on 2012-10-01
Reviewed by Carlos Garcia Campos.

Use more GOwnPtr/GRefPtr in WebKitDownload

* webkit/webkitdownload.cpp:
(webkit_download_open_stream_for_uri): Use GRefPtr
for a GFile and GOwnPtr for GError.
(webkit_download_set_destination_uri): Ditto.
(webkit_download_received_data): Use GOwnPtr for GError.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130036 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Fix viewport QML auto tests after the viewport changes
jocelyn.turcotte@digia.com [Mon, 1 Oct 2012 11:20:20 +0000 (11:20 +0000)]
[Qt] Fix viewport QML auto tests after the viewport changes
https://bugs.webkit.org/show_bug.cgi?id=97781

Reviewed by Simon Hausmann.

Since the viewport controller has to hold any QML-side viewport updates
while it applies and render the new requested position, this has the
effect of blocking all those tests if the view is not visible (thus
disabling the tile rendering completely in the web process).
The viewport should be unlocked after the first rendered frame for
static contents.

Fix those tests by forcing the QQuickView to be visible and by making
them wait for the first frame rather than the earlier end of provisional load.

* UIProcess/API/qt/tests/qmltests/WebView/tst_doubleTapToZoom.qml:
* UIProcess/API/qt/tests/qmltests/WebView/tst_fitToView.qml:
* UIProcess/API/qt/tests/qmltests/WebView/tst_wheelEventHandling.qml:
* UIProcess/API/qt/tests/qmltests/common/TestWebView.qml:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130035 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Add a loadVisuallyCommitted() signal
jocelyn.turcotte@digia.com [Mon, 1 Oct 2012 11:19:57 +0000 (11:19 +0000)]
[Qt] Add a loadVisuallyCommitted() signal
https://bugs.webkit.org/show_bug.cgi?id=97780

Reviewed by Kenneth Rohde Christiansen.

This signal is emitted when the first DidRenderFrame message is received from
the web process after loadCommitted.

* UIProcess/API/qt/qquickwebview.cpp:
(QQuickWebViewPrivate::QQuickWebViewPrivate):
(QQuickWebViewPrivate::loadDidCommit):
(QQuickWebViewPrivate::setNeedsDisplay):
* UIProcess/API/qt/qquickwebview_p.h:
* UIProcess/API/qt/qquickwebview_p_p.h:
(QQuickWebViewPrivate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130034 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Move Qt-only related methods from WebPageProxy.cpp to WebPageProxyQt.cpp
jocelyn.turcotte@digia.com [Mon, 1 Oct 2012 11:19:35 +0000 (11:19 +0000)]
[Qt] Move Qt-only related methods from WebPageProxy.cpp to WebPageProxyQt.cpp

Reviewed by Simon Hausmann.

* UIProcess/WebPageProxy.cpp:
* UIProcess/qt/WebPageProxyQt.cpp:
(WebKit::WebPageProxy::didFindZoomableArea):
(WebKit):
(WebKit::WebPageProxy::findZoomableAreaForPoint):
(WebKit::WebPageProxy::didReceiveMessageFromNavigatorQtObject):
(WebKit::WebPageProxy::authenticationRequiredRequest):
(WebKit::WebPageProxy::proxyAuthenticationRequiredRequest):
(WebKit::WebPageProxy::certificateVerificationRequest):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130033 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMake sure that the history position is applied correctly when using delegatesScrolling
jocelyn.turcotte@digia.com [Mon, 1 Oct 2012 11:19:18 +0000 (11:19 +0000)]
Make sure that the history position is applied correctly when using delegatesScrolling
https://bugs.webkit.org/show_bug.cgi?id=97778

Reviewed by Kenneth Rohde Christiansen.

The position is applied asynchronously and the UI process is the one holding the current
state. For this reason we can't rely in WebCore on ScrollView::scrollPosition holding
the current position in that case.

* page/Page.cpp:
(WebCore::Page::setPageScaleFactor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130032 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Decide when to apply a scrolled position to the viewport based on the rect cover...
jocelyn.turcotte@digia.com [Mon, 1 Oct 2012 11:19:01 +0000 (11:19 +0000)]
[Qt] Decide when to apply a scrolled position to the viewport based on the rect covered by the tiles
https://bugs.webkit.org/show_bug.cgi?id=97777

Reviewed by Kenneth Rohde Christiansen.

Source/WebCore:

* platform/graphics/TiledBackingStore.cpp:
(WebCore::TiledBackingStore::createTiles):
* platform/graphics/TiledBackingStore.h:
(WebCore::TiledBackingStore::coverRect):
(WebCore::TiledBackingStore::setCoverRect):
(TiledBackingStore):

Source/WebKit2:

This patch achieves two things:
- Apply a short scroll position request directly if we already have rendered tiles
  at this position.
- Since our visible rect request is asynchronous, the next frame might not always
  be the one covering this position. This will wait for a frame that does to apply it.

* UIProcess/API/qt/raw/qrawwebview_p_p.h:
(QRawWebViewPrivate::didRenderFrame):
* UIProcess/CoordinatedGraphics/LayerTreeCoordinatorProxy.cpp:
(WebKit::LayerTreeCoordinatorProxy::didRenderFrame):
* UIProcess/CoordinatedGraphics/LayerTreeCoordinatorProxy.h:
(LayerTreeCoordinatorProxy):
* UIProcess/CoordinatedGraphics/LayerTreeCoordinatorProxy.messages.in:
* UIProcess/PageClient.h:
(PageClient):
* UIProcess/PageViewportController.cpp:
(WebKit::PageViewportController::didCommitLoad):
(WebKit::PageViewportController::didRenderFrame):
(WebKit::PageViewportController::pageDidRequestScroll):
* UIProcess/PageViewportController.h:
(PageViewportController):
* UIProcess/WebPageProxy.h:
(WebPageProxy):
* UIProcess/qt/QtPageClient.cpp:
(WebKit::QtPageClient::didRenderFrame):
* UIProcess/qt/QtPageClient.h:
(QtPageClient):
* UIProcess/qt/WebPageProxyQt.cpp:
(WebKit::WebPageProxy::didRenderFrame):
* WebProcess/WebPage/CoordinatedGraphics/CoordinatedGraphicsLayer.h:
(WebCore::CoordinatedGraphicsLayer::coverRect):
* WebProcess/WebPage/CoordinatedGraphics/LayerTreeCoordinator.cpp:
(WebKit::LayerTreeCoordinator::flushPendingLayerChanges):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130031 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Delay viewport position, scale and contents size updates until tiles are rendered
jocelyn.turcotte@digia.com [Mon, 1 Oct 2012 11:18:20 +0000 (11:18 +0000)]
[Qt] Delay viewport position, scale and contents size updates until tiles are rendered
https://bugs.webkit.org/show_bug.cgi?id=97775

Reviewed by Kenneth Rohde Christiansen.

A few events from the web process can cause the viewport position or scale to be
adjusted in the UI process:
- Scroll position request
- Viewport attributes (initialScale, minimumScale)
- Contents size change

We previously applied those updates directly to the viewport, which would then in turn
produce a corresponding visible rect request to the web process to render the contents
at the new position/scale. This could leave the viewport showing the old content, either
by scaling the tiles, or by showing checkerboard if no contents was rendered at this
position yet. When the web process was done rendering new tiles, the proper tiles
were then shown to the user.
Since many of these updates happen during page load, all producing an itterative adjustment
to the viewport, this could show sharpness and position jitter until the final rendering
was done.

This patch makes those updates go through the PageViewportController first and then to the
QQuickWebView rather than the other way around, and keep them pending until the
LayerTreeHostProxy reports that the new tiles are ready to be shown. Since the rendering
is blocked until the page first layout is done, this apply the modifications to
the viewport only once when the first frame is ready to be shown.

* UIProcess/API/qt/qquickwebview.cpp:
* UIProcess/API/qt/qquickwebview_p_p.h:
(QQuickWebViewFlickablePrivate):
* UIProcess/API/qt/raw/qrawwebview_p_p.h:
(QRawWebViewPrivate::didRenderFrame):
* UIProcess/CoordinatedGraphics/LayerTreeCoordinatorProxy.cpp:
(WebKit::LayerTreeCoordinatorProxy::didRenderFrame):
* UIProcess/CoordinatedGraphics/LayerTreeCoordinatorProxy.h:
(LayerTreeCoordinatorProxy):
* UIProcess/CoordinatedGraphics/LayerTreeCoordinatorProxy.messages.in:
* UIProcess/PageClient.h:
(PageClient):
* UIProcess/PageViewportController.cpp:
(WebKit::PageViewportController::didCommitLoad):
(WebKit::PageViewportController::didChangeContentsSize):
(WebKit::PageViewportController::didRenderFrame):
(WebKit::PageViewportController::initialViewportReady):
(WebKit::PageViewportController::pageDidRequestScroll):
(WebKit::PageViewportController::didChangeContentsVisibility):
(WebKit::PageViewportController::resumeContent):
(WebKit::PageViewportController::applyScaleAfterRenderingContents):
(WebKit):
(WebKit::PageViewportController::applyPositionAfterRenderingContents):
(WebKit::PageViewportController::updateMinimumScaleToFit):
* UIProcess/PageViewportController.h:
(PageViewportController):
* UIProcess/PageViewportControllerClient.h:
(PageViewportControllerClient):
* UIProcess/WebPageProxy.h:
(WebPageProxy):
* UIProcess/qt/PageViewportControllerClientQt.cpp:
(WebKit::PageViewportControllerClientQt::didChangeContentsSize):
* UIProcess/qt/PageViewportControllerClientQt.h:
(PageViewportControllerClientQt):
* UIProcess/qt/QtPageClient.cpp:
(WebKit::QtPageClient::didRenderFrame):
(WebKit):
(WebKit::QtPageClient::didChangeContentsSize):
* UIProcess/qt/QtPageClient.h:
(QtPageClient):
* UIProcess/qt/QtWebPageLoadClient.cpp:
(WebKit::QtWebPageLoadClient::didCommitLoad):
* UIProcess/qt/WebPageProxyQt.cpp:
(WebKit::WebPageProxy::didRenderFrame):
(WebKit):
* WebProcess/WebPage/CoordinatedGraphics/LayerTreeCoordinator.cpp:
(WebKit::LayerTreeCoordinator::flushPendingLayerChanges):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130030 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Qt] Wait for the UI process before re-enabling rendering during page load
jocelyn.turcotte@digia.com [Mon, 1 Oct 2012 11:17:39 +0000 (11:17 +0000)]
[Qt] Wait for the UI process before re-enabling rendering during page load
https://bugs.webkit.org/show_bug.cgi?id=97773

Reviewed by Kenneth Rohde Christiansen.

During page navigation, WebCore might ask asynchronously the UI process to
scroll to an anchor or restored history position and the UI process will
return the corresponding visible rect to be rendered by the web process.

To avoid rendering tiles for the invalidated area by the new page layout at
the wrong position, we should do an extra message round-trip to the UI
process before resuming the rendering.
Assuming that all messages are handled in order by both the web and UI process,
sending the round-trip request in the web process once we sent all scroll
requests, contents size and viewport attributes updates and then handling
the round-trip response, we make sure that final visible rect request have
been handled already.

* UIProcess/API/qt/raw/qrawwebview.cpp:
(QRawWebViewPrivate::pageTransitionViewportReady):
* UIProcess/API/qt/raw/qrawwebview_p_p.h:
(QRawWebViewPrivate):
* UIProcess/PageClient.h:
(PageClient):
* UIProcess/PageViewportController.cpp:
(WebKit::PageViewportController::pageTransitionViewportReady):
(WebKit):
* UIProcess/PageViewportController.h:
(PageViewportController):
* UIProcess/WebPageProxy.h:
(WebPageProxy):
* UIProcess/WebPageProxy.messages.in:
* UIProcess/qt/QtPageClient.cpp:
(WebKit::QtPageClient::pageTransitionViewportReady):
(WebKit):
* UIProcess/qt/QtPageClient.h:
(QtPageClient):
* UIProcess/qt/WebPageProxyQt.cpp:
(WebKit::WebPageProxy::commitPageTransitionViewport):
(WebKit):
(WebKit::WebPageProxy::pageTransitionViewportReady):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::WebFrameLoaderClient):
(WebKit::WebFrameLoaderClient::dispatchDidLayout):
(WebKit::WebFrameLoaderClient::frameLoadCompleted):
(WebKit::WebFrameLoaderClient::provisionalLoadStarted):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:
(WebFrameLoaderClient):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::didStartPageTransition):
(WebKit):
(WebKit::WebPage::didCompletePageTransition):
* WebProcess/WebPage/WebPage.h:
(WebPage):
* WebProcess/WebPage/WebPage.messages.in:
* WebProcess/WebPage/qt/WebPageQt.cpp:
(WebKit::WebPage::commitPageTransitionViewport):
(WebKit):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130029 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWebPrintOperationGtk destructor should be virtual
commit-queue@webkit.org [Mon, 1 Oct 2012 11:12:15 +0000 (11:12 +0000)]
WebPrintOperationGtk destructor should be virtual
https://bugs.webkit.org/show_bug.cgi?id=98002

Patch by Alberto Garcia <agarcia@igalia.com> on 2012-10-01
Reviewed by Carlos Garcia Campos.

WebPrintOperationGtk is an abstract class which can be deleted
using a pointer to itself, yet its destructor is not virtual.
Fixes -Wdelete-non-virtual-dtor.

* WebProcess/WebPage/gtk/WebPrintOperationGtk.h:
(WebPrintOperationGtk):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130028 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed. Fix make distcheck.
carlosgc@webkit.org [Mon, 1 Oct 2012 11:00:04 +0000 (11:00 +0000)]
Unreviewed. Fix make distcheck.

Source/WebCore:

* GNUmakefile.list.am: ClipPathOperation.h was moved.

Source/WTF:

* GNUmakefile.list.am: Add missing header files to compilation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130027 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMicrodata: names.item() must return null for out of range indexes.
arko@motorola.com [Mon, 1 Oct 2012 10:45:12 +0000 (10:45 +0000)]
Microdata: names.item() must return null for out of range indexes.
https://bugs.webkit.org/show_bug.cgi?id=97898

Reviewed by Kentaro Hara.

Source/WebCore:

DOMStringList.item() must return null for an invalid index.
Spec: http://www.w3.org/TR/DOM-Level-3-Core/core.html#DOMStringList
Removed [IsIndex] extended IDL attribute from item() method's index
parameter in DOMStringList interface.
Firefox and Opera's behavior is consistent with the spec. Both
returns null for invalid index.

Test: fast/dom/MicroData/names-item-out-of-range-index.html

* dom/DOMStringList.idl:

LayoutTests:

Added test case to ensure names.item() behavior for out of range indexes.

* fast/dom/MicroData/names-item-out-of-range-index-expected.txt: Added.
* fast/dom/MicroData/names-item-out-of-range-index.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130026 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, update the URL of the Szeged SVN mirror.
ossy@webkit.org [Mon, 1 Oct 2012 10:36:09 +0000 (10:36 +0000)]
Unreviewed, update the URL of the Szeged SVN mirror.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130025 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Forms] Multiple fields week input UI
yosin@chromium.org [Mon, 1 Oct 2012 10:23:21 +0000 (10:23 +0000)]
[Forms] Multiple fields week input UI
https://bugs.webkit.org/show_bug.cgi?id=97877

Reviewed by Kent Tamura.

Source/WebCore:

This patch introduces multiple fields "week" input UI in DRT. We'll
enable this feature once we add tests.

Note: This patch affects ports which enable both ENABLE_INPUT_TYPE_WEEK
and ENABLE_INPUT_MULTIPLE_FIELDS_UI.

No new tests. To reduce size of this patch, other patches add tests
for multiple fields week input UI.

Note: Actual outputs of two tests
  - fast/forms/week/week-input-visible-string.html
  - fast/forms/week/week-stepup-stepdown-from-renderer.html
are different.

* css/thml.css:
(input::-webkit-datetime-edit-week-field): Added for field appearance.
(input::-webkit-datetime-edit-week-field:focus): Added to remove focus ring.
* html/WeekInputType.cpp:
(WebCore::WeekInputType::formatDateTimeFieldsState): Added to format numeric value to string value as specified in HTML5 specification.
(WebCore::WeekInputType::setupLayoutParameters):  Added to set layout of multiple fields.
* html/WeekInputType.h: Changed to include BaseMultipleFieldsDateAndTimeInputType.h and introduce BaseWeekInputType typedef.
(WebCore::WeekInputType::WeekInputType): Changed base class name to BaseWeekInputType.
(WeekInputType): Changed to add declarations for formatDateTimeFieldsState() and setupLayoutParameters().
* html/shadow/DateTimeEditElement.cpp:
(WebCore::DateTimeEditBuilder::visitField): Changed to support week field.

LayoutTests:

This patch adds Chromium port specific expectations for "week" input
type tests for multiple fields week input UI.

Note: This patch affects ports which enable both ENABLE_INPUT_TYPE_WEEK
and ENABLE_INPUT_MULTIPLE_FIELDS_UI.

* platform/chromium/fast/forms/week/week-input-visible-string-expected.txt: Added. Multiple fields week input UI doesn't have selection.
* platform/chromium/fast/forms/week/week-stepup-stepdown-from-renderer-expected.txt: Added. On multiple fields week input UI, step down/up decrement/increment a field rather than whole value.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130024 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed gardening -- removed TestExpectation entries for passing tests.
caseq@chromium.org [Mon, 1 Oct 2012 09:52:42 +0000 (09:52 +0000)]
Unreviewed gardening -- removed TestExpectation entries for passing tests.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130023 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed gardening -- rebaselined meter element tests on chromium-mac.
caseq@chromium.org [Mon, 1 Oct 2012 09:43:29 +0000 (09:43 +0000)]
Unreviewed gardening -- rebaselined meter element tests on chromium-mac.

* fast/dom/HTMLMeterElement/meter-element-repaint-on-update-value-expected.png: Renamed from LayoutTests/platform/efl/fast/dom/HTMLMeterElement/meter-element-repaint-on-update-value-expected.png.
* platform/chromium-mac-snowleopard/fast/dom/HTMLMeterElement/meter-styles-expected.png:
* platform/chromium-mac/fast/dom/HTMLMeterElement/meter-element-expected.png:
* platform/chromium-mac/fast/dom/HTMLMeterElement/meter-element-repaint-on-update-value-expected.png:
* platform/chromium-mac/fast/dom/HTMLMeterElement/meter-styles-expected.png:
* platform/chromium/TestExpectations:
* platform/gtk/fast/dom/HTMLMeterElement/meter-element-repaint-on-update-value-expected.png: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130022 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: do not use InspectorInstrumentation::hasFrontends() check when collect...
pfeldman@chromium.org [Mon, 1 Oct 2012 09:41:19 +0000 (09:41 +0000)]
Web Inspector: do not use InspectorInstrumentation::hasFrontends() check when collecting stacks
https://bugs.webkit.org/show_bug.cgi?id=96730

Reviewed by Vsevolod Vlasov.

- Introduced InspectorInstrumentation::console|timeline|runtime|canvasAgentEnabled
- Using it all over the place instead of the hasFrontend (the latter is now only used once to guard hot path)
- Introduced explicit "enabled" state of the console and runtime agents

* bindings/js/JSHTMLCanvasElementCustom.cpp:
(WebCore::JSHTMLCanvasElement::getContext):
* bindings/js/JSMainThreadExecState.h:
(WebCore::JSMainThreadExecState::instrumentFunctionCall):
* bindings/js/ScheduledAction.cpp:
(WebCore::ScheduledAction::create):
* bindings/js/ScriptCallStackFactory.cpp:
(WebCore::createScriptCallStackForConsole):
* bindings/js/ScriptCallStackFactory.h:
(WebCore):
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateCallWith):
* bindings/scripts/CodeGeneratorV8.pm:
(GenerateCallWith):
* bindings/scripts/test/JS/JSTestObj.cpp:
(WebCore::jsTestObjWithScriptArgumentsAndCallStackAttribute):
(WebCore::setJSTestObjWithScriptArgumentsAndCallStackAttribute):
(WebCore::jsTestObjPrototypeFunctionWithScriptArgumentsAndCallStack):
* bindings/scripts/test/V8/V8TestObj.cpp:
(WebCore::TestObjV8Internal::withScriptArgumentsAndCallStackAttributeAttrGetter):
(WebCore::TestObjV8Internal::withScriptArgumentsAndCallStackAttributeAttrSetter):
(WebCore::TestObjV8Internal::withScriptArgumentsAndCallStackCallback):
* bindings/v8/ScriptCallStackFactory.cpp:
(WebCore::createScriptCallStackForConsole):
* bindings/v8/ScriptCallStackFactory.h:
(WebCore):
* bindings/v8/ScriptController.cpp:
(WebCore::ScriptController::callFunctionWithInstrumentation):
* bindings/v8/V8DOMWindowShell.cpp:
(WebCore::V8DOMWindowShell::setIsolatedWorldSecurityOrigin):
* bindings/v8/V8WorkerContextEventListener.cpp:
(WebCore::V8WorkerContextEventListener::callListenerFunction):
* bindings/v8/custom/V8DOMWindowCustom.cpp:
(WebCore::WindowSetTimeoutImpl):
* bindings/v8/custom/V8HTMLCanvasElementCustom.cpp:
(WebCore::V8HTMLCanvasElement::getContextCallback):
* bindings/v8/custom/V8WorkerContextCustom.cpp:
(WebCore::SetTimeoutOrInterval):
* inspector/Inspector.json:
* inspector/InspectorConsoleAgent.cpp:
(WebCore):
(WebCore::InspectorConsoleAgent::InspectorConsoleAgent):
(WebCore::InspectorConsoleAgent::enable):
(WebCore::InspectorConsoleAgent::disable):
(WebCore::InspectorConsoleAgent::clearMessages):
(WebCore::InspectorConsoleAgent::clearFrontend):
(WebCore::InspectorConsoleAgent::addConsoleMessage):
* inspector/InspectorConsoleAgent.h:
(WebCore::InspectorConsoleAgent::enabled):
(InspectorConsoleAgent):
* inspector/InspectorController.cpp:
(WebCore::InspectorController::connectFrontend):
(WebCore::InspectorController::disconnectFrontend):
* inspector/InspectorInstrumentation.cpp:
(WebCore):
(WebCore::InspectorInstrumentation::canvasAgentEnabled):
(WebCore::InspectorInstrumentation::consoleAgentEnabled):
(WebCore::InspectorInstrumentation::runtimeAgentEnabled):
(WebCore::InspectorInstrumentation::timelineAgentEnabled):
* inspector/InspectorInstrumentation.h:
(InspectorInstrumentation):
(WebCore::InspectorInstrumentation::canvasAgentEnabled):
(WebCore::InspectorInstrumentation::consoleAgentEnabled):
(WebCore::InspectorInstrumentation::runtimeAgentEnabled):
(WebCore::InspectorInstrumentation::timelineAgentEnabled):
* inspector/InspectorRuntimeAgent.cpp:
(WebCore::InspectorRuntimeAgent::InspectorRuntimeAgent):
* inspector/InspectorRuntimeAgent.h:
(WebCore::InspectorRuntimeAgent::enabled):
(WebCore::InspectorRuntimeAgent::enable):
(WebCore::InspectorRuntimeAgent::disable):
(InspectorRuntimeAgent):
* inspector/PageRuntimeAgent.cpp:
(PageRuntimeAgentState):
(WebCore::PageRuntimeAgent::clearFrontend):
(WebCore::PageRuntimeAgent::restore):
(WebCore):
(WebCore::PageRuntimeAgent::enable):
(WebCore::PageRuntimeAgent::disable):
(WebCore::PageRuntimeAgent::didClearWindowObject):
(WebCore::PageRuntimeAgent::didCreateIsolatedContext):
(WebCore::PageRuntimeAgent::reportExecutionContextCreation):
* inspector/PageRuntimeAgent.h:
(PageRuntimeAgent):
* inspector/WorkerRuntimeAgent.cpp:
* inspector/WorkerRuntimeAgent.h:
* inspector/front-end/RuntimeModel.js:
(WebInspector.RuntimeModel.prototype._didLoadCachedResources):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::postMessage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130021 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed. Create WebSocket watch list, and add myself to that list.
yutak@chromium.org [Mon, 1 Oct 2012 09:37:04 +0000 (09:37 +0000)]
Unreviewed. Create WebSocket watch list, and add myself to that list.

* Scripts/webkitpy/common/config/watchlist:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130020 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WebSocket] Setting wrong value to binaryType should not raise exception
bashi@chromium.org [Mon, 1 Oct 2012 09:29:31 +0000 (09:29 +0000)]
[WebSocket] Setting wrong value to binaryType should not raise exception
https://bugs.webkit.org/show_bug.cgi?id=97999

Reviewed by Yuta Kitamura.

Source/WebCore:

Don't raise exception when binaryType is the wrong value.
Instead, show an error message to console.

No new tests. Updated existing test.

* Modules/websockets/WebSocket.cpp:
(WebCore::WebSocket::setBinaryType): See the description.
* Modules/websockets/WebSocket.h:
(WebSocket): Removed ExceptionCode argument of setBinaryType().
* Modules/websockets/WebSocket.idl:
Removed "setter raises(DOMException)" and "[TreatReturnedNullStringAs=Undefined]".
They are no longer needed.

LayoutTests:

Update binary-type.html.

* http/tests/websocket/tests/hybi/binary-type-expected.txt:
* http/tests/websocket/tests/hybi/binary-type.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130019 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Forms] Adding DateTimeWeekFieldElement for multiple fields "week" input UI
yosin@chromium.org [Mon, 1 Oct 2012 09:18:42 +0000 (09:18 +0000)]
[Forms] Adding DateTimeWeekFieldElement for multiple fields "week" input UI
https://bugs.webkit.org/show_bug.cgi?id=97992

Reviewed by Kent Tamura.

This patch introduces DateTimeWeekFieldElement class for implementing
multiple fields "week" input UI.

No new tests. This patch doesn't change behavior.

* html/shadow/DateTimeFieldElements.cpp:
(WebCore::DateTimeWeekFieldElement::DateTimeWeekFieldElement): Added.
(WebCore::DateTimeWeekFieldElement::create): Added.
(WebCore::DateTimeWeekFieldElement::populateDateTimeFieldsState): Added.
(WebCore::DateTimeWeekFieldElement::setValueAsDate): Added.
(WebCore::DateTimeWeekFieldElement::setValueAsDateTimeFieldsState): Added.
* html/shadow/DateTimeFieldElements.h:
(DateTimeWeekFieldElement): Added.
* platform/DateComponents.h:
(WebCore::DateComponents): Added declarations of static const member variables, DateComponents::maximumWeekNumber and minimumWeekNumber.
* platform/DateComponents.cpp: Added definitions of DateComponents::maximumWeekNumber and minimumWeekNumber.
(WebCore::DateComponents::maxWeekNumberInYear): Changed to use maximumWeekNumber.
(WebCore::DateComponents::parseWeek): Changed to use minimumWeekNumber.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130018 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed gardening. Added expectations for 4 month-multiple-fields tests for chromi...
caseq@chromium.org [Mon, 1 Oct 2012 09:05:05 +0000 (09:05 +0000)]
Unreviewed gardening. Added expectations for 4 month-multiple-fields tests for chromium mac & win.
https://bugs.webkit.org/show_bug.cgi?id=97888

* platform/chromium-mac-snowleopard/fast/forms/month-multiple-fields/month-multiple-fields-appearance-basic-expected.png: Added.
* platform/chromium-mac-snowleopard/fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-classes-expected.png: Added.
* platform/chromium-mac-snowleopard/fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-elements-expected.png: Added.
* platform/chromium-mac-snowleopard/fast/forms/month-multiple-fields/month-multiple-fields-appearance-style-expected.png: Added.
* platform/chromium-mac/fast/forms/month-multiple-fields/month-multiple-fields-appearance-basic-expected.png: Added.
* platform/chromium-mac/fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-classes-expected.png: Added.
* platform/chromium-mac/fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-elements-expected.png: Added.
* platform/chromium-mac/fast/forms/month-multiple-fields/month-multiple-fields-appearance-style-expected.png: Added.
* platform/chromium-win/fast/forms/month-multiple-fields/month-multiple-fields-appearance-basic-expected.png: Added.
* platform/chromium-win/fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-classes-expected.png: Added.
* platform/chromium-win/fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-elements-expected.png: Added.
* platform/chromium-win/fast/forms/month-multiple-fields/month-multiple-fields-appearance-style-expected.png: Added.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130017 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: Separate CSSStyleModelResourceBinding into resource and content binding.
vsevik@chromium.org [Mon, 1 Oct 2012 08:46:41 +0000 (08:46 +0000)]
Web Inspector: Separate CSSStyleModelResourceBinding into resource and content binding.
https://bugs.webkit.org/show_bug.cgi?id=97994

Reviewed by Pavel Feldman.

Extracted StyleContentBinding from CSSStyleModelResourceBinding.
Now CSSStyleModelResourceBinding is responsible for mapping between stylesheets and resources.
StyleContentBinding is now responsible for synchronization between stylesheet content and uiSourceCode content.

* inspector/front-end/CSSStyleModel.js:
(WebInspector.CSSStyleModelResourceBinding):
(WebInspector.CSSStyleModelResourceBinding.prototype.requestStyleSheetIdForResource):
(WebInspector.CSSStyleModelResourceBinding.prototype.requestResourceURLForStyleSheetId):
* inspector/front-end/StyleSource.js:
(WebInspector.StyleSource.prototype._commitIncrementalEdit):
* inspector/front-end/StylesSourceMapping.js:
(WebInspector.StyleContentBinding):
(WebInspector.StyleContentBinding.prototype.setStyleContent.callback):
(WebInspector.StyleContentBinding.prototype.setStyleContent):
(WebInspector.StyleContentBinding.prototype._innerSetContent.callback):
(WebInspector.StyleContentBinding.prototype._innerSetContent):
(WebInspector.StyleContentBinding.prototype._styleSheetChanged.callback):
(WebInspector.StyleContentBinding.prototype._styleSheetChanged):
(WebInspector.StyleContentBinding.prototype._innerStyleSheetChanged):
* inspector/front-end/inspector.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130016 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: [Device Metrics] Remove the gutter overlay moving its functionality...
apavlov@chromium.org [Mon, 1 Oct 2012 08:38:35 +0000 (08:38 +0000)]
Web Inspector: [Device Metrics] Remove the gutter overlay moving its functionality into the InspectorOverlay
https://bugs.webkit.org/show_bug.cgi?id=97799

Reviewed by Pavel Feldman.

Source/WebCore:

Re-applying r129746 with test flakiness fixed.

In order to reduce the amount of port-specific code, the gutter overlay painted in the device metrics emulation mode
has been replaced by the respective functionality in the HTML-based InspectorOverlay in WebCore. The InspectorOverlay
now covers the entire WebView rather than the FrameView only.

* inspector/InspectorController.cpp:
(WebCore::InspectorController::webViewResized):
(WebCore):
* inspector/InspectorController.h:
(WebCore):
(InspectorController):
* inspector/InspectorOverlay.cpp:
(WebCore::InspectorOverlay::InspectorOverlay):
(WebCore::InspectorOverlay::paint):
(WebCore::InspectorOverlay::resize):
(WebCore):
(WebCore::InspectorOverlay::update):
(WebCore::InspectorOverlay::drawGutter):
(WebCore::InspectorOverlay::reset):
* inspector/InspectorOverlay.h:
(InspectorOverlay):
* inspector/InspectorOverlayPage.html: Introduce the gutter painting functionality previously found in the Chromium's
DeviceMetricsSupport class, which used to implement WebPageOverlay.

Source/WebKit/chromium:

- Dispatch the webViewResized() event on InspectorController, which is necessary to update the InspectorOverlay.
- Remove the gutter overlay painting code.

* src/WebDevToolsAgentImpl.cpp:
(WebKit::DeviceMetricsSupport::DeviceMetricsSupport):
(WebKit::DeviceMetricsSupport::~DeviceMetricsSupport):
(WebKit::WebDevToolsAgentImpl::attach):
(WebKit::WebDevToolsAgentImpl::webViewResized):
(WebKit::WebDevToolsAgentImpl::overrideDeviceMetrics):
* src/WebDevToolsAgentImpl.h:
(WebDevToolsAgentImpl):
* src/WebDevToolsAgentPrivate.h:
(WebDevToolsAgentPrivate):
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::resize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130015 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove unused sys/mman.h include
commit-queue@webkit.org [Mon, 1 Oct 2012 08:35:56 +0000 (08:35 +0000)]
Remove unused sys/mman.h include
https://bugs.webkit.org/show_bug.cgi?id=97995

Patch by Jonathan Liu <net147@gmail.com> on 2012-10-01
Reviewed by Kentaro Hara.

The sys/mman.h is not used and removing it improves portability as not
all systems have sys/mman.h.

* jit/ExecutableAllocatorFixedVMPool.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130014 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed. Fix GTK+ build after r129908.
carlosgc@webkit.org [Mon, 1 Oct 2012 08:33:04 +0000 (08:33 +0000)]
Unreviewed. Fix GTK+ build after r129908.

* GNUmakefile.list.am: Add new files to compilation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130013 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Unskip fast/innerHTML/innerHTML-iframe.html after r130003.
rakuco@webkit.org [Mon, 1 Oct 2012 08:26:40 +0000 (08:26 +0000)]
[EFL] Unskip fast/innerHTML/innerHTML-iframe.html after r130003.

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130012 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove overzealous assert in SVGElement::localAttributeToPropertyMap
pdr@google.com [Mon, 1 Oct 2012 08:01:29 +0000 (08:01 +0000)]
Remove overzealous assert in SVGElement::localAttributeToPropertyMap
https://bugs.webkit.org/show_bug.cgi?id=97291

Reviewed by Nikolas Zimmermann.

This patch removes an assert where we did not expect
SVGElement::localAttributeToPropertyMap where we did not to be called. This
function turns out to be useful and this patch removes that assert.

If we encounter a non-SVG tag during SVG parsing (e.g. <svg><price></svg>) we return a
vanilla SVGElement instance from SVGElementFactory::createSVGElement. Previously,
trying to animate this would ASSERT because it was not possible to determine the
animated type. After this patch, an empty localAttributeToPropertyMap is used so
that the animated type returned from SVGAnimateElement::determineAnimatedPropertyType
is AnimatedUnknown.

This patch simply removes an ASSERT so no test is provided.

* svg/SVGElement.cpp:
(WebCore::SVGElement::localAttributeToPropertyMap):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130011 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd appearance tests for multiple fields month input UI
yosin@chromium.org [Mon, 1 Oct 2012 07:58:59 +0000 (07:58 +0000)]
Add appearance tests for multiple fields month input UI
https://bugs.webkit.org/show_bug.cgi?id=97888

Reviewed by Kent Tamura.

This patch adds tests for multiple fields month input UI and disables
these tests on ports which don't enable both ENABLE_INPUT_TYPE_MONTH
and ENABLE_INPUT_MULTIPLE_FIELDS_UI.

Note: We need to do rebaseline tests to update images for Chromium-Mac
and Chromium-Win.

Note: This patch affects ports which enable both ENABLE_INPUT_TYPE_MONTH
and ENABLE_INPUT_MULTIPLE_FIELDS_UI.

* fast/forms/month-multiple-fields/month-multiple-fields-appearance-basic-expected.txt: Added.
* fast/forms/month-multiple-fields/month-multiple-fields-appearance-basic.html:  Added for checking basic appearance.
* fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-classes-expected.txt: Added.
* fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-classes.html: Added for checking appearance of applying CSS pseudo classes, e.g. :enabled, :invalid, and so on.
* fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-elements-expected.txt: Added.
* fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-elements.html: Added for checking appearance of applying CSS pseudo elements, e.g. :after, :before, and so on.
* fast/forms/month-multiple-fields/month-multiple-fields-appearance-style-expected.txt: Added.
* fast/forms/month-multiple-fields/month-multiple-fields-appearance-style.html: Added for checking appearance of applying CSS styles
* platform/chromium-android/TestExpectations: Changed to disable tests for multiple fields month input UI.
* platform/chromium-linux/fast/forms/month-multiple-fields/month-multiple-fields-appearance-basic-expected.png: Added.
* platform/chromium-linux/fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-classes-expected.png: Added.
* platform/chromium-linux/fast/forms/month-multiple-fields/month-multiple-fields-appearance-pseudo-elements-expected.png: Added.
* platform/chromium-linux/fast/forms/month-multiple-fields/month-multiple-fields-appearance-style-expected.png: Added.
* platform/chromium/TestExpectations: Changed to disable tests for multiple fields month input UI of Mac and Win for rebaseline.
* platform/efl/TestExpectations: Changed to disable tests for multiple fields month input UI.
* platform/gtk/TestExpectations: ditto
* platform/mac/TestExpectations: ditto
* platform/qt/TestExpectations: ditto
* platform/win/TestExpectations: ditto

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130010 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rolling out r130004.
commit-queue@webkit.org [Mon, 1 Oct 2012 07:20:09 +0000 (07:20 +0000)]
Unreviewed, rolling out r130004.
http://trac.webkit.org/changeset/130004
https://bugs.webkit.org/show_bug.cgi?id=97996

Test shadow-dom-modify-chardata.html is failing (Requested by
keishi on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-10-01

Source/WebCore:

* dom/CharacterData.cpp:
(WebCore::CharacterData::dispatchModifiedEvent):

LayoutTests:

* inspector/elements/shadow-dom-modify-chardata-expected.txt: Removed.
* inspector/elements/shadow-dom-modify-chardata.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130009 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoREGRESSION(r127727): Calendar picker is ignoring step
keishi@webkit.org [Mon, 1 Oct 2012 07:15:09 +0000 (07:15 +0000)]
REGRESSION(r127727): Calendar picker is ignoring step
https://bugs.webkit.org/show_bug.cgi?id=97893

Reviewed by Kent Tamura.

Source/WebCore:

There were two mistakes:
- An if-statement to check step attribute validity was wrong, and
- DateTiemChooserParameters.step was milleseconds when it should be number of days.
This will be changing the DateTimeChooserParameters.step to milliseconds so we can handle steps for other input types in the future.

Test: fast/forms/date/calendar-picker-appearance-with-step.html

* Resources/pagepopups/calendarPicker.js:
(CalendarPicker):
* html/shadow/CalendarPickerElement.cpp:
(WebCore::CalendarPickerElement::openPopup): If statement was wrong.

LayoutTests:

* fast/forms/date/calendar-picker-appearance-with-step-expected.txt: Added.
* fast/forms/date/calendar-picker-appearance-with-step.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130008 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSign in front of keyframe selector causes stylesheet parsing to abort
commit-queue@webkit.org [Mon, 1 Oct 2012 06:13:53 +0000 (06:13 +0000)]
Sign in front of keyframe selector causes stylesheet parsing to abort
https://bugs.webkit.org/show_bug.cgi?id=96844

Patch by Glenn Adams <glenn@skynav.com> on 2012-09-30
Reviewed by Simon Fraser.

Source/WebCore:

Allow optional unary operator (+|-) on PERCENTAGE in keyframe selector.

Test: animations/keyframe-selector-negative-percentage.html

* css/CSSGrammar.y:
Add maybe_unary_operator to PERCENTAGE on keyframe selector. Negative keyframe
selector value is already ignored in StyleKeyframe::parseKeyString.

LayoutTests:

Add test case for correct handling of negative percentage in keyframe selector.

* animations/keyframe-selector-negative-percentage-expected.txt: Added.
* animations/keyframe-selector-negative-percentage.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130007 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agohttps://bugs.webkit.org/show_bug.cgi?id=97988
morrita@google.com [Mon, 1 Oct 2012 05:56:48 +0000 (05:56 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=97988
Crash on FrameTree::scopedChildCount()

Reviewed by Kent Tamura.

The series of crash reports says that there are some null pointer
access in scopedChildCount(). This change added a null guard
against Frame::document(), that can return null.

No new tests. This is tied to some specific timing and is hard to reproduce.

* page/FrameTree.cpp:
(WebCore::FrameTree::scopedChildCount):
(WebCore::FrameTree::scopedChild):
(WebCore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130006 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMake multiple fields date/time input UI related files to available all ports
yosin@chromium.org [Mon, 1 Oct 2012 05:50:08 +0000 (05:50 +0000)]
Make multiple fields date/time input UI related files to available all ports
https://bugs.webkit.org/show_bug.cgi?id=97989

Reviewed by Kent Tamura.

This patch adds multiple fields date/time input UI related files for
ports not using WebCore.gyp and simplifies include directive in
MonthInputType.h and TimeInputType.h.

Added files are:
  - html/BaseMultipleFieldsDateAndTimeInputType.{cpp,h}
  - html/shadow/DateTimeEditElement.{cpp,h}
  - html/shadow/DateTimeFieldElement.{cpp,h}
  - html/shadow/DateTimeFieldElements.{cpp,h}
  - html/shadow/DateTimeNumericFieldElement.{cpp,h}
  - html/shadow/DateTimeSymbolicFieldElement.{cpp,h}

No new tests. This patch doesn't change behavior.

* CMakeLists.txt: Changed to add multiple fields date/time input UI related files.
* GNUmakefile.list.am: ditto
* Target.pri: ditto
* WebCore.vcproj/WebCore.vcproj: ditto
* WebCore.xcodeproj/project.pbxproj: ditto
* html/MonthInputType.h: Changed to simplify include directive for base class.
* html/TimeInputType.h: ditto

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130005 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: Modifications in a shadow tree don't update the Elements panel.
keishi@webkit.org [Mon, 1 Oct 2012 03:57:15 +0000 (03:57 +0000)]
Web Inspector: Modifications in a shadow tree don't update the Elements panel.
https://bugs.webkit.org/show_bug.cgi?id=97056

Reviewed by Pavel Feldman.

Source/WebCore:

Send characterDataModified event for shadow dom nodes too so they update the elements panel.

Test: inspector/elements/shadow-dom-modify-chardata.html

* dom/CharacterData.cpp:
(WebCore::CharacterData::dispatchModifiedEvent):

LayoutTests:

* inspector/elements/shadow-dom-modify-chardata-expected.txt: Added.
* inspector/elements/shadow-dom-modify-chardata.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130004 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoTrivial test fix after r127534.
commit-queue@webkit.org [Mon, 1 Oct 2012 02:45:45 +0000 (02:45 +0000)]
Trivial test fix after r127534.
https://bugs.webkit.org/show_bug.cgi?id=95813

Patch by Sergey Glazunov <serg.glazunov@gmail.com> on 2012-09-30
Reviewed by Kentaro Hara.

* fast/innerHTML/innerHTML-iframe-expected.txt:
* fast/innerHTML/innerHTML-iframe.html:
* platform/chromium/TestExpectations:
* platform/qt/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130003 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed. Adding myself as a committer.
tzik@chromium.org [Mon, 1 Oct 2012 02:44:06 +0000 (02:44 +0000)]
Unreviewed. Adding myself as a committer.

* Scripts/webkitpy/common/config/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130002 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[chromium] Add a test to verify that a plugin does not receive touch-events unless...
commit-queue@webkit.org [Mon, 1 Oct 2012 01:57:07 +0000 (01:57 +0000)]
[chromium] Add a test to verify that a plugin does not receive touch-events unless it explicitly requests for them
https://bugs.webkit.org/show_bug.cgi?id=97975

Patch by Sadrul Habib Chowdhury <sadrul@chromium.org> on 2012-09-30
Reviewed by Adam Barth.

Tools:

TestWebPlugin can now be configured to accept touch events or not. Use this
configurability to verify that a plugin does not receive touch events unless
it explicitly requests for them.

* DumpRenderTree/chromium/TestWebPlugin.cpp:
(TestWebPlugin::TestWebPlugin):
(TestWebPlugin::initialize):
(TestWebPlugin::parseBoolean):
* DumpRenderTree/chromium/TestWebPlugin.h:
(TestWebPlugin):

LayoutTests:

* platform/chromium/plugins/touch-events-expected.txt: Added.
* platform/chromium/plugins/touch-events.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130001 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSplit EventTargetData out of NodeRareData to reduce memory use.
kling@webkit.org [Mon, 1 Oct 2012 01:30:18 +0000 (01:30 +0000)]
Split EventTargetData out of NodeRareData to reduce memory use.
<http://webkit.org/b/97987>
<rdar://problem/12403258>

Reviewed by Anders Carlsson.

Move EventTargetData to its own Node-flag/hashmap instead of piggybacking on NodeRareData.
This reduces memory consumption by 1.06MB on Membuster3. Note that NodeRareData shrinks by
one pointer as well.

* dom/Node.cpp:
(WebCore::Node::~Node):
(WebCore::eventTargetDataMap):
(WebCore::Node::eventTargetData):
(WebCore::Node::ensureEventTargetData):
(WebCore::Node::clearEventTargetData):
(WebCore::Node::handleLocalEvents):
* dom/Node.h:
(WebCore::Node::hasEventTargetData):
(WebCore::Node::setHasEventTargetData):
* dom/NodeRareData.h:
(NodeRareData):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@130000 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoClean up HasTrivialConstructor/Destructor
mhahnenberg@apple.com [Mon, 1 Oct 2012 00:28:47 +0000 (00:28 +0000)]
Clean up HasTrivialConstructor/Destructor
https://bugs.webkit.org/show_bug.cgi?id=97754

Reviewed by Sam Weinig.

Mixed up the HasTrivialDestructor/Constructor case for VS2010 and later in the previous patch.

* wtf/TypeTraits.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129999 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago444kB below CSSParser::parseDeprecatedGradient() on Membuster3.
kling@webkit.org [Sun, 30 Sep 2012 21:06:32 +0000 (21:06 +0000)]
444kB below CSSParser::parseDeprecatedGradient() on Membuster3.
<http://webkit.org/b/97981>
<rdar://problem/12403058>

Reviewed by Anders Carlsson.

Slap an inline capacity of 2 on the Vector<CSSGradientColorStop> in CSSGradientValue.
This covers the majority of gradient values, and reduces memory consumption by ~250kB on Membuster3.

* css/CSSGradientValue.h:
(WebCore::CSSGradientValue::stopCount):
* css/CSSParser.cpp:
(WebCore::CSSParser::parseLinearGradient):
(WebCore::CSSParser::parseGradientColorStops):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129996 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove FIXME comments refering to non-existent code in JSDOMBinding.cpp
mkwst@chromium.org [Sun, 30 Sep 2012 19:35:52 +0000 (19:35 +0000)]
Remove FIXME comments refering to non-existent code in JSDOMBinding.cpp
https://bugs.webkit.org/show_bug.cgi?id=97977

Reviewed by Adam Barth.

I did a quick grep through the code to determine where these FIXME
comments were suggesting that code should be merged. So far as I can
tell, 'immediatelyReportUnsafeAccessTo' only exists in these comments.

Just cleanup, no functional change.

* bindings/js/JSDOMBinding.cpp:
(WebCore::shouldAllowAccessToFrame):
(WebCore::shouldAllowAccessToDOMWindow):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129995 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WK2][GTK] Add API to get the favicon for a WebKitWebView
mario@webkit.org [Sun, 30 Sep 2012 18:38:48 +0000 (18:38 +0000)]
[WK2][GTK] Add API to get the favicon for a WebKitWebView
https://bugs.webkit.org/show_bug.cgi?id=96477

Reviewed by Carlos Garcia Campos.

Provide a new simple API to synchronously try to get the favicon
associated with a WebView, if any, and to keep track of changes on
it, through a new GObject property.

* UIProcess/API/gtk/WebKitWebView.cpp:
(_WebKitWebViewPrivate):
(webkitWebViewIconReadyCallback): Callback to handle the
'icon-ready' signal coming from WebKitFaviconDatabase.
(webkitWebViewWatchForChangesInFavicon): Connects to the
'icon-ready' signal from WebKitFaviconDatabase, to keep track of
changes in favicons, that must be related to the current view.
(webkitWebViewDisconnectFaviconDatabaseSignalHandlers):
Disconnects the handler for 'icon-ready' if needed.
(webkitWebViewGetProperty): Updated for the new "favicon" property .
(webkitWebViewFinalize): Disconnect the new signal handler.
(webkit_web_view_class_init): Definition of the new property.
(webkitWebViewEmitLoadChanged): Make sure we will be watching for
changes in the favicon from WEBKIT_LOAD_STARTED on.
(webkit_web_view_get_favicon): New API funtcion, returning the
current favicon for the WebView, if any, or NULL otherwise.
* UIProcess/API/gtk/WebKitWebView.h:

Internally expose a way to try to get the favicon associated to a
page URL synchronously, through WebKitFaviconDatabase.

* UIProcess/API/gtk/WebKitFaviconDatabase.cpp:
(webkitFaviconDatabaseGetFaviconSync): New internal function, it
will return either 0 or a valid pointer to a cairo_surface_t.
* UIProcess/API/gtk/WebKitFaviconDatabasePrivate.h:

Add unit tests for checking this new API.

* UIProcess/API/gtk/tests/TestWebKitFaviconDatabase.cpp:
(testWebViewFavicon): New unit test.
(beforeAll): Add the test to the test suite.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129994 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[TextureMapper] [WebKit2] Crash in WebCore::BitmapTextureGL::updateContents
commit-queue@webkit.org [Sun, 30 Sep 2012 16:28:36 +0000 (16:28 +0000)]
[TextureMapper] [WebKit2] Crash in WebCore::BitmapTextureGL::updateContents
https://bugs.webkit.org/show_bug.cgi?id=97394

Patch by Martin Robinson <mrobinson@igalia.com> on 2012-09-30
Reviewed by Noam Rosenthal.

When a TextureMapper is destroyed, layers can still contain references
to textures obtained from that TextureMapper's texture pool. Trying to
access an unreffed TextureMapper in the BitmapTexture's destructor
causes a crash.

Instead of storing a raw pointer to a TextureMapper, we can simply store
a reference to the underlying GraphicsContext3D. All TextureMapper
implementations use the current GL context at this moment, so one GC3D
referencing the current context is the same as any other.

* platform/graphics/texmap/TextureMapper.h: Remove the clearTexturePool
method. It's no longer used.
(WebCore::BitmapTexture::applyFilters): Add a TextureMapper* argument.
* platform/graphics/texmap/TextureMapperGL.cpp:
(WebCore::BitmapTextureGL::BitmapTextureGL): Keep a reference to the
GraphicsContext3D instead of the TextureMapper.
(WebCore::BitmapTextureGL::didReset): Use the GC3D reference.
(WebCore::BitmapTextureGL::updateContents): Ditto.
(WebCore::BitmapTextureGL::applyFilters): Accept the TextureMapper as an argument.
(WebCore::BitmapTextureGL::initializeStencil): Use the GC3D reference.
(WebCore::BitmapTextureGL::clearIfNeeded): Ditto.
(WebCore::BitmapTextureGL::createFboIfNeeded): Ditto.
(WebCore::BitmapTextureGL::bind): Accept the TextureMapper as an argument.
(WebCore::BitmapTextureGL::~BitmapTextureGL): Use the GC3D reference.
(WebCore::TextureMapperGL::~TextureMapperGL): Remove the call to clearTexturePool
as it's no longer necessary.
(WebCore::TextureMapperGL::bindSurface): Ditto.
* platform/graphics/texmap/TextureMapperGL.h:
(BitmapTextureGL): Keep a GC3D reference instead of a TextureMapper pointer.
* platform/graphics/texmap/TextureMapperImageBuffer.cpp:
(WebCore::BitmapTextureImageBuffer::applyFilters): Add a TextureMapper argument.
* platform/graphics/texmap/TextureMapperImageBuffer.h:
(BitmapTextureImageBuffer):
* platform/graphics/texmap/TextureMapperLayer.cpp:
(WebCore::applyFilters): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129993 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WK2][GTK] Fix issues with WebKitFaviconDatabase in debug builds
mario@webkit.org [Sun, 30 Sep 2012 14:56:08 +0000 (14:56 +0000)]
[WK2][GTK] Fix issues with WebKitFaviconDatabase in debug builds
https://bugs.webkit.org/show_bug.cgi?id=97966

Reviewed by Carlos Garcia Campos.

Fix failing ASSERTs detected when running the unit tests from
TestWebKitFaviconDatabase in debug builds.

* UIProcess/API/gtk/WebKitFaviconDatabase.cpp:
(webkitFaviconDatabaseDispose): Close IconDatabase here, which is
the right place to do it according to GObject documentation.
(webkitFaviconDatabaseFinalize): Just destroy the private data
structure and chain up to parent class's finalize method.
(webkit_favicon_database_class_init): Override dispose method.

* UIProcess/API/gtk/tests/TestWebKitFaviconDatabase.cpp:
(webkitFaviconDatabaseFinalizedCallback): Delete files when the
WebKitFaviconDatabase is being finalized, since by that time the
IconDatabase from WebCore will be already closed.
(afterAll): Add a weak reference to WebKitFaviconDatabase and
provide a GWeakNotify callback (webkitFaviconDatabaseFinalizedCallback).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129992 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Chromium] Rebaseline for DateTimeNumericFieldElement should use Localizer functions.
tkent@chromium.org [Sun, 30 Sep 2012 13:46:19 +0000 (13:46 +0000)]
[Chromium] Rebaseline for DateTimeNumericFieldElement should use Localizer functions.
https://bugs.webkit.org/show_bug.cgi?id=97318

* platform/chromium/TestExpectation:
* platform/chromium-linux/fast/forms/time-multiple-fields/time-multiple-fields-localization-expected.txt:
* platform/chromium-win/fast/forms/time-multiple-fields/time-multiple-fields-localization-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129991 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Add proper bug links to the protocol handler crashes in efl-wk2.
rakuco@webkit.org [Sun, 30 Sep 2012 11:47:49 +0000 (11:47 +0000)]
[EFL] Add proper bug links to the protocol handler crashes in efl-wk2.

* platform/efl-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129990 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Unskip a few seemingly passing tests with no associated bug.
rakuco@webkit.org [Sun, 30 Sep 2012 11:19:45 +0000 (11:19 +0000)]
[EFL] Unskip a few seemingly passing tests with no associated bug.

* platform/efl-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129989 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAsyncFileSystem::openFileSystem should have FileSystemType as a parameter.
commit-queue@webkit.org [Sun, 30 Sep 2012 02:29:57 +0000 (02:29 +0000)]
AsyncFileSystem::openFileSystem should have FileSystemType as a parameter.
https://bugs.webkit.org/show_bug.cgi?id=97963

Patch by Dongwoo Joshua Im <dw.im@samsung.com> on 2012-09-29
Reviewed by Gyuyoung Kim.

FileSystemType is an important information to maintain the file system,
and AsyncFileSystem::openFileSystem need to get the type as a parameter.
And, there are "FIXME" comments about that in WebCore source codes.

No new functionality, no new tests.

* Modules/filesystem/LocalFileSystem.cpp: Add FileSystemType as a parameter of AsyncFileSystem::openFileSystem.
(WebCore::openFileSystem):
(WebCore::LocalFileSystem::readFileSystem):
(WebCore::LocalFileSystem::requestFileSystem):
* platform/AsyncFileSystem.cpp: ditto.
(WebCore::AsyncFileSystem::openFileSystem):
* platform/AsyncFileSystem.h: ditto.
(AsyncFileSystem):
* platform/blackberry/AsyncFileSystemBlackBerry.cpp: ditto.
(WebCore::AsyncFileSystem::openFileSystem):
* platform/gtk/AsyncFileSystemGtk.cpp: ditto.
(WebCore::AsyncFileSystem::openFileSystem):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129988 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Unskip test that seems to be passing now.
rakuco@webkit.org [Sat, 29 Sep 2012 22:20:50 +0000 (22:20 +0000)]
[EFL] Unskip test that seems to be passing now.

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129987 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Unskip {horizontal,vertical}-scrollbar-in-rtl.html
rakuco@webkit.org [Sat, 29 Sep 2012 22:13:24 +0000 (22:13 +0000)]
[EFL] Unskip {horizontal,vertical}-scrollbar-in-rtl.html

The expectations in fast/dom are actually Mac-specific (cf.
r73063), and most other ports use their own platform-specific
results due to the differences in behavior of the Home/End keys.

The results being checked in here are in line with the other
ports, so everything seems to be OK.

* platform/efl/TestExpectations:
* platform/efl/fast/dom/horizontal-scrollbar-in-rtl-expected.txt: Added.
* platform/efl/fast/dom/vertical-scrollbar-in-rtl-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129986 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rolling out r129982.
eae@chromium.org [Sat, 29 Sep 2012 20:55:11 +0000 (20:55 +0000)]
Unreviewed, rolling out r129982.
http://trac.webkit.org/changeset/129982
https://bugs.webkit.org/show_bug.cgi?id=97971

hit assert in fast/overflow/overflow-height-float-not-removed-
crash3.html (Requested by eae on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-09-29

* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129985 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rolling out r129965.
commit-queue@webkit.org [Sat, 29 Sep 2012 20:29:34 +0000 (20:29 +0000)]
Unreviewed, rolling out r129965.
http://trac.webkit.org/changeset/129965
https://bugs.webkit.org/show_bug.cgi?id=97970

Causes ASSERTs in workers (Requested by abarth on #webkit).

Patch by Sheriff Bot <webkit.review.bot@gmail.com> on 2012-09-29

* bindings/v8/DOMData.cpp:
(WebCore::DOMData::getCurrentStore):
* bindings/v8/ScopedPersistent.h:
* bindings/v8/ScriptController.cpp:
(WebCore::ScriptController::resetIsolatedWorlds):
(WebCore::ScriptController::evaluateInIsolatedWorld):
(WebCore::ScriptController::currentWorldContext):
* bindings/v8/V8Binding.cpp:
(WebCore::perContextDataForCurrentWorld):
* bindings/v8/V8DOMWindowShell.cpp:
(WebCore::setIsolatedWorldField):
(WebCore::V8DOMWindowShell::enteredIsolatedWorldContext):
(WebCore::V8DOMWindowShell::destroyIsolatedShell):
(WebCore):
(WebCore::isolatedContextWeakCallback):
(WebCore::V8DOMWindowShell::disposeContext):
(WebCore::V8DOMWindowShell::initializeIfNeeded):
(WebCore::V8DOMWindowShell::setIsolatedWorldSecurityOrigin):
* bindings/v8/V8DOMWindowShell.h:
(V8DOMWindowShell):
(WebCore::V8DOMWindowShell::getEntered):
* bindings/v8/V8DOMWrapper.h:
(WebCore::V8DOMWrapper::getCachedWrapper):
* bindings/v8/WorldContextHandle.cpp:
(WebCore::WorldContextHandle::WorldContextHandle):
* bindings/v8/custom/V8DocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8HTMLDocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8SVGDocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8XMLHttpRequestConstructor.cpp:
(WebCore::V8XMLHttpRequest::constructorCallback):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129984 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed build fix for chromium/clank.
eae@chromium.org [Sat, 29 Sep 2012 18:37:59 +0000 (18:37 +0000)]
Unreviewed build fix for chromium/clank.

* platform/FractionalLayoutUnit.h:
(WebCore::FractionalLayoutUnit::FractionalLayoutUnit):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129983 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoEnable SATURATED_LAYOUT_ARITHMETIC for chromium
eae@chromium.org [Sat, 29 Sep 2012 17:43:42 +0000 (17:43 +0000)]
Enable SATURATED_LAYOUT_ARITHMETIC for chromium
https://bugs.webkit.org/show_bug.cgi?id=95053

Reviewed by Abhishek Arya.

Enable the SATURATED_LAYOUT_ARITHMETIC flag for the chromium port.
This changes the behavior of FractionalLayoutUnit to clamp to the
max or min value instead of overflowing.

This may very well impact performance so the current plan is to enable
it for a couple of hours to a day to collect performance data and then
disable it again until we've had a chance to review the perf data.

* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129982 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Rebaseline test after r129695.
rakuco@webkit.org [Sat, 29 Sep 2012 16:03:02 +0000 (16:03 +0000)]
[EFL] Rebaseline test after r129695.

* platform/efl/TestExpectations:
* platform/efl/mathml/presentation/tables-expected.png:
* platform/efl/mathml/presentation/tables-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129981 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Revert r129978.
rakuco@webkit.org [Sat, 29 Sep 2012 15:47:21 +0000 (15:47 +0000)]
[EFL] Revert r129978.

Skip media/unsupported-rtsp.html again, as it still fails on the
bots. I have created a proper bug for the issue now.

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129980 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Unskip fast/css/device-aspect-ratio.html
rakuco@webkit.org [Sat, 29 Sep 2012 15:17:47 +0000 (15:17 +0000)]
[EFL] Unskip fast/css/device-aspect-ratio.html

This seems to be another test with no bug report that is now
passing.

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129979 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Unskip media/unsupported-rtsp.html
rakuco@webkit.org [Sat, 29 Sep 2012 14:54:18 +0000 (14:54 +0000)]
[EFL] Unskip media/unsupported-rtsp.html

The test seems to be consistently passing at least locally, let's
see if the bots agree.

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129978 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix build warning : -Wparentheses.
commit-queue@webkit.org [Sat, 29 Sep 2012 14:40:24 +0000 (14:40 +0000)]
Fix build warning : -Wparentheses.
https://bugs.webkit.org/show_bug.cgi?id=97961

Patch by Byungwoo Lee <bw80.lee@samsung.com> on 2012-09-29
Reviewed by Kentaro Hara.

Explicit braces are added around the && statement to fix -Wparentheses warning.

* mathml/MathMLElement.h:
(WebCore::toMathMLElement):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129977 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Fix build error : Link webkit2 library to ewk2UnitTestInjectedBundleSample.
commit-queue@webkit.org [Sat, 29 Sep 2012 12:51:30 +0000 (12:51 +0000)]
[EFL] Fix build error : Link webkit2 library to ewk2UnitTestInjectedBundleSample.
https://bugs.webkit.org/show_bug.cgi?id=97622

Patch by Byungwoo Lee <bw80.lee@samsung.com> on 2012-09-29
Reviewed by Gyuyoung Kim.

There is no dependency between ewk2UnitTestInjectedBundleSample and
webkit2 library.

Because of this, on the first build, compiler tries to build
injected_bundle_sample.cpp before the forwarding headers are created.
This makes build error that the header file 'WebKit2/WKBaseSoup.h'
cannot be found.

To prevent this, webkit2 library is linked to ewk2UnitTestInjectedBundleSample.

* PlatformEfl.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129976 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoScroll offset of flex items lost during relayout
kenneth@webkit.org [Sat, 29 Sep 2012 10:22:22 +0000 (10:22 +0000)]
Scroll offset of flex items lost during relayout
https://bugs.webkit.org/show_bug.cgi?id=97706

Reviewed by Tony Chang.

Source/WebCore:

Test: fast/flexbox/overflow-keep-scrollpos.html

Flex box does a second pass layout of the flex children.

We layout the child without scrollbars (to get the size
used for flexing), then we relayout the child at its final size.

We must not apply the scroll position during the first pass,
as it will be clamped to 0 (no scrolling possible).

* rendering/RenderBlock.h:
(RenderBlock):

Make updateScrollInfoAfterLayout public

* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::layoutBlock):

Delay applying scroll info until we clamp the size of the child
and get scrollbars back again. For this to work we use
RenderBlock::updateScrollInfoAfterLayout instead of the non-guarded
RenderLayer::updateScrollInfoAfterLayout.

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::updateScrollInfoAfterLayout):

Add workaround for now to keep passing css3/flexbox/child-overflow.html

Basically do not postpone applying scroll changes for RenderBlocks
with opposite writing mode, as they need to have their content
overflow in the right direction.

LayoutTests:

Add a new test for testing that flex items scroll offsets are
not lost during relayout.

* css3/flexbox/overflow-keep-scrollpos-expected.txt: Added.
* css3/flexbox/overflow-keep-scrollpos.html: Added.
* css3/flexbox/child-overflow.html: Fix minor errors.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129975 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSlightly improve clarity of the patch in bug 78595.
dglazkov@chromium.org [Sat, 29 Sep 2012 09:21:11 +0000 (09:21 +0000)]
Slightly improve clarity of the patch in bug 78595.
https://bugs.webkit.org/show_bug.cgi?id=97944

Reviewed by Andreas Kling.

Since all types of relations, except SubSelector are effectively ignoring the calculated value of pseudoId, make the code reflect that a bit more clearly.

No change in behavior, covered by test in bug 78595.

* css/SelectorChecker.cpp:
(WebCore::SelectorChecker::checkSelector): Added ignoreDynamicPseudo value that's given to all callsites that should ignore the result.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129974 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove LocalizedDate*.*
tkent@chromium.org [Sat, 29 Sep 2012 08:49:57 +0000 (08:49 +0000)]
Remove LocalizedDate*.*
https://bugs.webkit.org/show_bug.cgi?id=97957

Reviewed by Kentaro Hara.

The functions declared in LocalizedDate.h are replaced with member
functions of Localizer. LocalizedDate.h and its implementations
are not needed any more.

No new tests. This should not change any behavior.

* GNUmakefile.list.am: Remove LocalizedDate.h and LocalizedDateNode.cpp
* WebCore.vcproj/WebCore.vcproj: ditto.
* WebCore.xcodeproj/project.pbxproj: ditto.
* CMakeLists.txt: Remove LocalizedDateNone.cpp
* Target.pri: ditto.
* WebCore.gyp/WebCore.gyp: Remove LocalizedDate*.*. Update a comment.
* WebCore.gypi: Remove LocalizedDate*.*.

* html/BaseDateAndTimeInputType.cpp: Use Localizer functions.
(WebCore::BaseDateAndTimeInputType::localizeValue):
(WebCore::BaseDateAndTimeInputType::convertFromVisibleValue):
* html/DateInputType.cpp: ditto.
(WebCore::DateInputType::fixedPlaceholder):

* platform/text/Localizer.h:
(Localizer): Move comments from LocalziedDate.h.
* platform/text/mac/LocaleMac.mm:
Remove unnecessary include of LocalizedDate.h.

* platform/text/LocalizedDate.h: Removed.
* platform/text/LocalizedDateICU.cpp: Removed.
* platform/text/LocalizedDateNone.cpp: Removed.
* platform/text/LocalizedDateWin.cpp: Removed.
* platform/text/mac/LocalizedDateMac.cpp: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129973 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WK2][GTK][EFL] Share WebKit2-GTK plugin process implementation with EFL port
commit-queue@webkit.org [Sat, 29 Sep 2012 06:31:08 +0000 (06:31 +0000)]
[WK2][GTK][EFL] Share WebKit2-GTK plugin process implementation with EFL port
https://bugs.webkit.org/show_bug.cgi?id=91844

Patch by Mariusz Grzegorczyk <mariusz.g@samsung.com> on 2012-09-28
Reviewed by Simon Hausmann.

.:

Turn on Netscape Plugin API by default  for WebKit2-Efl.

* Source/cmake/OptionsEfl.cmake:

Source/WebKit2:

Implement windowless plugin and functions needed to launch PluginProcess.

* GNUmakefile.am: Change files and folders related to plugin process from gtk to unix.
* GNUmakefile.list.am: Change files and folders related to plugin process from gtk to unix.
* Platform/Logging.cpp: Add Plugins channel for LOG.
(WebKit):
(WebKit::getChannelFromName):
(WebKit::initializeLogChannelsIfNecessary):
* Platform/Logging.h:
(WebKit):
* PlatformEfl.cmake: Add files needed by plugin process.
* PluginProcess/unix/PluginControllerProxyUnix.cpp: Renamed from Source/WebKit2/PluginProcess/gtk/PluginControllerProxyGtk.cpp.
(WebKit):
(WebKit::PluginControllerProxy::platformInitialize): Add macros for gtk/efl differents.
(WebKit::PluginControllerProxy::platformDestroy):
(WebKit::PluginControllerProxy::platformGeometryDidChange):
* PluginProcess/unix/PluginProcessMainUnix.cpp: Renamed from Source/WebKit2/PluginProcess/gtk/PluginProcessMainGtk.cpp.
(WebKit):
(WebKit::webkitXError):
(WebKit::PluginProcessMainUnix):
* PluginProcess/unix/PluginProcessMainUnix.h: Renamed from Source/WebKit2/PluginProcess/gtk/PluginProcessMainGtk.h.
(WebKit):
* PluginProcess/unix/PluginProcessUnix.cpp: Renamed from Source/WebKit2/PluginProcess/gtk/PluginProcessGtk.cpp.
(WebKit):
(WebKit::PluginProcess::platformInitialize):
* UIProcess/Plugins/unix/PluginProcessProxyUnix.cpp: Renamed from Source/WebKit2/UIProcess/Plugins/gtk/PluginProcessProxyGtk.cpp.
(WebKit):
(WebKit::PluginProcessProxy::platformInitializePluginProcess):
(WebKit::PluginProcessProxy::scanPlugin):
* UIProcess/efl/WebPageProxyEfl.cpp: Add stubs.
(WebKit::WebPageProxy::createPluginContainer):
(WebKit):
(WebKit::WebPageProxy::windowedPluginGeometryDidChange):
* WebProcess/Plugins/Netscape/unix/PluginProxyUnix.cpp: Renamed from Source/WebKit2/WebProcess/Plugins/Netscape/gtk/PluginProxyGtk.cpp.
(WebKit):
(WebKit::PluginProxy::needsBackingStore):
* WebProcess/Plugins/Netscape/x11/NetscapePluginX11.cpp: Plugin's windowless implementation.
(WebKit::getPluginDisplay):
(WebKit::x11Screen):
(WebKit::displayDepth):
(WebKit::rootWindowID):
(WebKit::NetscapePlugin::x11HostDisplay):
(WebKit::NetscapePlugin::platformPaint):
* config.h:
* unix/PluginMainUnix.cpp: Renamed from Source/WebKit2/gtk/PluginMainGtk.cpp.
(main):

Tools:

Turn on Netscape Plugin API option by default.

* Scripts/webkitperl/FeatureList.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129972 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[V8] The concept of forceNewObject is unneeded (dom-traverse gets 3% faster)
abarth@webkit.org [Sat, 29 Sep 2012 02:50:28 +0000 (02:50 +0000)]
[V8] The concept of forceNewObject is unneeded (dom-traverse gets 3% faster)
https://bugs.webkit.org/show_bug.cgi?id=97943

Reviewed by Kentaro Hara.

We don't need the concept of forceNewObject. It doesn't do anything
useful and it makes the bindings slower.

* bindings/scripts/CodeGeneratorV8.pm:
(GenerateHeader):
(GenerateToV8Converters):
(NativeToJSValue):
* bindings/v8/custom/V8DocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8HTMLDocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8HTMLElementCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8NodeCustom.cpp:
(WebCore::toV8Slow):
* bindings/v8/custom/V8SVGDocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8SVGElementCustom.cpp:
(WebCore::toV8):
* dom/make_names.pl:
(printWrapperFactoryCppFile):
(printWrapperFactoryHeaderFile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129971 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAllow any kind of v8::Handle in invokeCallback instead of requiring a v8::Persistent.
commit-queue@webkit.org [Sat, 29 Sep 2012 02:30:16 +0000 (02:30 +0000)]
Allow any kind of v8::Handle in invokeCallback instead of requiring a v8::Persistent.
https://bugs.webkit.org/show_bug.cgi?id=97956

Patch by Elliott Sprehn <esprehn@chromium.org> on 2012-09-28
Reviewed by Adam Barth.

Use v8::Handle instead of v8::Persistent for the callback argument on invokeCallback.
There doesn't seem to be any reason for requiring a Persistent, and it makes it harder
to use this API.

This is factored out of http://wkbug.com/93661

No new tests, this is functionally equivalent.

* bindings/v8/V8Callback.cpp:
(WebCore::invokeCallback):
* bindings/v8/V8Callback.h:
(WebCore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129970 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove support for method overloading from bridge code
andersca@apple.com [Sat, 29 Sep 2012 02:18:06 +0000 (02:18 +0000)]
Remove support for method overloading from bridge code
https://bugs.webkit.org/show_bug.cgi?id=97959

Reviewed by Dan Bernstein.

Source/WebCore:

The method overloading handling was only in place for the (now removed) Java bridge.
Replace MethodList everywhere with a single Method pointer.

* GNUmakefile.am:
* GNUmakefile.list.am:
* WebCore.exp.in:
* bridge/c/c_class.cpp:
(JSC::Bindings::CClass::methodNamed):
* bridge/c/c_class.h:
(CClass):
* bridge/c/c_instance.cpp:
(JSC::Bindings::CRuntimeMethod::create):
(JSC::Bindings::CRuntimeMethod::CRuntimeMethod):
(JSC::Bindings::CInstance::getMethod):
(JSC::Bindings::CInstance::invokeMethod):
* bridge/jsc/BridgeJSC.h:
(Bindings):
(Class):
* bridge/objc/objc_class.h:
(ObjcClass):
* bridge/objc/objc_class.mm:
(JSC::Bindings::ObjcClass::methodNamed):
* bridge/objc/objc_instance.mm:
(ObjCRuntimeMethod::create):
(ObjCRuntimeMethod::ObjCRuntimeMethod):
(ObjcInstance::invokeMethod):
* bridge/qt/qt_class.cpp:
(JSC::Bindings::QtClass::methodNamed):
* bridge/qt/qt_class.h:
(QtClass):
* bridge/qt/qt_instance.cpp:
(JSC::Bindings::QtInstance::getMethod):
* bridge/runtime_method.cpp:
(JSC::RuntimeMethod::RuntimeMethod):
(JSC::RuntimeMethod::lengthGetter):
(JSC::callRuntimeMethod):
* bridge/runtime_method.h:
(JSC::RuntimeMethod::create):
(JSC::RuntimeMethod::method):
(RuntimeMethod):
* bridge/runtime_object.cpp:
(JSC::Bindings::RuntimeObject::getOwnPropertySlot):
(JSC::Bindings::RuntimeObject::getOwnPropertyDescriptor):

Source/WebKit/mac:

Update for WebCore changes; MethodList has been replaced with a single Method* everywhere.

* Plugins/Hosted/ProxyInstance.h:
(ProxyInstance):
* Plugins/Hosted/ProxyInstance.mm:
(ProxyClass):
(WebKit::ProxyClass::methodNamed):
(WebKit::ProxyRuntimeMethod::create):
(WebKit::ProxyRuntimeMethod::ProxyRuntimeMethod):
(WebKit::ProxyInstance::getMethod):
(WebKit::ProxyInstance::invokeMethod):
(WebKit::ProxyInstance::methodNamed):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129969 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix compilation of V8DependentRetained and JSDependentRetained.
commit-queue@webkit.org [Sat, 29 Sep 2012 02:16:19 +0000 (02:16 +0000)]
Fix compilation of V8DependentRetained and JSDependentRetained.
https://bugs.webkit.org/show_bug.cgi?id=97955

Patch by Elliott Sprehn <esprehn@chromium.org> on 2012-09-28
Reviewed by Kentaro Hara.

Fix bad usage of putDirect and removeDirect from JSDependentRetained and fix
incorrect assumptions about how weak handles work in V8. This is refactored
out of the patch on http://wkbug.com/93661

No tests needed, this just fixes the compile and wrong usage of ScopedPersistent.

* bindings/js/JSDependentRetained.h:
(WebCore::JSDependentRetained::JSDependentRetained):
(WebCore::JSDependentRetained::retain):
(WebCore::JSDependentRetained::release):
(JSDependentRetained):
* bindings/v8/V8DependentRetained.h:
(WebCore::V8DependentRetained::V8DependentRetained):
(WebCore::V8DependentRetained::retain):
(WebCore::V8DependentRetained::weakCallback):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129968 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix chromium build after http://trac.webkit.org/changeset/129964.
ojan@chromium.org [Sat, 29 Sep 2012 02:00:46 +0000 (02:00 +0000)]
Fix chromium build after trac.webkit.org/changeset/129964.
* WebCore.gypi:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129967 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoCrash re-entering Document layout with frame flattening enabled
simon.fraser@apple.com [Sat, 29 Sep 2012 01:35:35 +0000 (01:35 +0000)]
Crash re-entering Document layout with frame flattening enabled
https://bugs.webkit.org/show_bug.cgi?id=97841

Reviewed by Brady Eidson.

When creating a CachedFrame, clearTimers on the Frame later; it has to
be done after documentWillSuspendForPageCache(), because the style
changes that HTMLPlugInImageElement::documentWillSuspendForPageCache()
do can schedule a layout on the FrameView, and we don't want this layout
timer to fire while the page is in the page cache.

Add an assertion in FrameView::layout() that the document is not in
the page cache. Without the above change, this would assert in
the plugins/frameset-with-plugin-frame.html test.

* history/CachedFrame.cpp:
(WebCore::CachedFrame::CachedFrame):
* page/FrameView.cpp:
(WebCore::FrameView::layout):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129966 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove V8DOMWindowShell::getEntered
commit-queue@webkit.org [Sat, 29 Sep 2012 00:45:07 +0000 (00:45 +0000)]
Remove V8DOMWindowShell::getEntered
https://bugs.webkit.org/show_bug.cgi?id=96637

Patch by Dan Carney <dcarney@google.com> on 2012-09-28
Reviewed by Adam Barth.

V8DOMWindowShell::getEntered was refactored so that the window shell
no longer has to be kept alive by a v8 context but rather a smaller
object.

No new tests. No change in functionality.

* bindings/v8/DOMData.cpp:
(WebCore::DOMData::getCurrentStore):
* bindings/v8/ScopedPersistent.h:
(WebCore::ScopedPersistent::leakHandle):
(ScopedPersistent):
* bindings/v8/ScriptController.cpp:
(WebCore::ScriptController::resetIsolatedWorlds):
(WebCore::ScriptController::evaluateInIsolatedWorld):
(WebCore::ScriptController::currentWorldContext):
* bindings/v8/V8Binding.cpp:
(WebCore::perContextDataForCurrentWorld):
* bindings/v8/V8DOMWindowShell.cpp:
(WebCore::setIsolatedWorldField):
(WebCore::V8DOMWindowShell::toIsolatedContextData):
(WebCore::isolatedContextWeakCallback):
(WebCore::V8DOMWindowShell::disposeContext):
(WebCore::V8DOMWindowShell::clearIsolatedShell):
(WebCore):
(WebCore::V8DOMWindowShell::initializeIfNeeded):
(WebCore::V8DOMWindowShell::setIsolatedWorldSecurityOrigin):
* bindings/v8/V8DOMWindowShell.h:
(V8DOMWindowShell):
(IsolatedContextData):
(WebCore::V8DOMWindowShell::IsolatedContextData::create):
(WebCore::V8DOMWindowShell::IsolatedContextData::world):
(WebCore::V8DOMWindowShell::IsolatedContextData::perContextData):
(WebCore::V8DOMWindowShell::IsolatedContextData::setSecurityOrigin):
(WebCore::V8DOMWindowShell::IsolatedContextData::securityOrigin):
(WebCore::V8DOMWindowShell::IsolatedContextData::IsolatedContextData):
(WebCore::V8DOMWindowShell::enteredIsolatedContext):
(WebCore::V8DOMWindowShell::enteredIsolatedContextData):
* bindings/v8/V8DOMWrapper.h:
(WebCore::V8DOMWrapper::getCachedWrapper):
* bindings/v8/WorldContextHandle.cpp:
(WebCore::WorldContextHandle::WorldContextHandle):
* bindings/v8/custom/V8DocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8HTMLDocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8SVGDocumentCustom.cpp:
(WebCore::toV8):
* bindings/v8/custom/V8XMLHttpRequestConstructor.cpp:
(WebCore::V8XMLHttpRequest::constructorCallback):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129965 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove Java bridge
andersca@apple.com [Sat, 29 Sep 2012 00:36:58 +0000 (00:36 +0000)]
Remove Java bridge
https://bugs.webkit.org/show_bug.cgi?id=97954

Reviewed by Sam Weinig.

The Java bridge is not used by any port; Mac now has a NPAPI Java plug-in.

Source/WebCore:

* WebCore.exp.in:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/ScriptController.cpp:
(WebCore::ScriptController::ScriptController):
* bindings/js/ScriptController.h:
(ScriptController):
* bindings/js/ScriptControllerMac.mm:
(WebCore::ScriptController::createScriptInstanceForWidget):
* bridge/jni/JNIUtility.cpp: Removed.
* bridge/jni/JNIUtility.h: Removed.
* bridge/jni/JavaType.h: Removed.
* bridge/jni/JobjectWrapper.cpp: Removed.
* bridge/jni/JobjectWrapper.h: Removed.
* bridge/jni/jni_jsobject.h: Removed.
* bridge/jni/jni_jsobject.mm: Removed.
* bridge/jni/jni_objc.mm: Removed.
* bridge/jni/jsc/JNIUtilityPrivate.cpp: Removed.
* bridge/jni/jsc/JNIUtilityPrivate.h: Removed.
* bridge/jni/jsc/JavaArrayJSC.cpp: Removed.
* bridge/jni/jsc/JavaArrayJSC.h: Removed.
* bridge/jni/jsc/JavaClassJSC.cpp: Removed.
* bridge/jni/jsc/JavaClassJSC.h: Removed.
* bridge/jni/jsc/JavaFieldJSC.cpp: Removed.
* bridge/jni/jsc/JavaFieldJSC.h: Removed.
* bridge/jni/jsc/JavaInstanceJSC.cpp: Removed.
* bridge/jni/jsc/JavaInstanceJSC.h: Removed.
* bridge/jni/jsc/JavaMethodJSC.cpp: Removed.
* bridge/jni/jsc/JavaMethodJSC.h: Removed.
* bridge/jni/jsc/JavaRuntimeObject.cpp: Removed.
* bridge/jni/jsc/JavaRuntimeObject.h: Removed.
* bridge/jni/jsc/JavaStringJSC.h: Removed.
* bridge/runtime_root.h:
* loader/FrameLoaderClient.h:
(FrameLoaderClient):

Source/WebKit/mac:

* WebCoreSupport/WebFrameLoaderClient.h:
* WebCoreSupport/WebFrameLoaderClient.mm:
Make sure that we try to load the NPAPI Java plug-in.

Source/WebKit2:

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:
(WebFrameLoaderClient):

Source/WTF:

* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129964 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove unused parameter in SelectorChecker::checkScrollbarPseudoClass.
dglazkov@chromium.org [Sat, 29 Sep 2012 00:04:33 +0000 (00:04 +0000)]
Remove unused parameter in SelectorChecker::checkScrollbarPseudoClass.
https://bugs.webkit.org/show_bug.cgi?id=97941

Reviewed by Kentaro Hara.

The last parameter in checkScrollbarPseudoClass was unused, so I removed it.

No change in behavior, just refactoring.

* css/SelectorChecker.cpp:
(WebCore::SelectorChecker::checkOneSelector): Changed the callsite to accommodate the change.
(WebCore::SelectorChecker::checkScrollbarPseudoClass): Removed last param.
* css/SelectorChecker.h: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129963 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoREGRESSION(r122215) - CachedImage::likelyToBeUsedSoon crashes on accessing a deleted...
hclam@chromium.org [Fri, 28 Sep 2012 23:58:22 +0000 (23:58 +0000)]
REGRESSION(r122215) - CachedImage::likelyToBeUsedSoon crashes on accessing a deleted CachedImageClient
https://bugs.webkit.org/show_bug.cgi?id=97749

Reviewed by James Robinson.

.:

Added a manual test to demonstrate drag image and crashing.

* ManualTests/drag-image-no-crash.html: Added.

Source/WebCore:

All implementations of Clipboard set themselves as clients to CachedImage
through the JS API setDrageImage() but they do not detach during destruction.
This causes memory corruption when CachedImage tries to access a deleted client
when MemoryCache prunes and calls CachedImage::likelyToUsedSoon().

Manual test added: ManualTests/drag-image-no-crash.html

* platform/chromium/ClipboardChromium.cpp:
(WebCore::ClipboardChromium::~ClipboardChromium):
* platform/gtk/ClipboardGtk.cpp:
(WebCore::ClipboardGtk::~ClipboardGtk):
* platform/mac/ClipboardMac.mm:
(WebCore::ClipboardMac::~ClipboardMac):
* platform/win/ClipboardWin.cpp:
(WebCore::ClipboardWin::~ClipboardWin):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129962 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove Instance::setDidExecuteFunction
andersca@apple.com [Fri, 28 Sep 2012 23:41:58 +0000 (23:41 +0000)]
Remove Instance::setDidExecuteFunction
https://bugs.webkit.org/show_bug.cgi?id=97952

Reviewed by Alexey Proskuryakov.

Instance::setDidExecuteFunction was added over 8 years ago to fix a bug where Objective-C
DOM calls weren't updating the document correctly. Nowadays we correctly invalidate the DOM tree
when these calls are made so we don't need an extra step to do so.

* bindings/js/ScriptControllerMac.mm:
(WebCore):
(WebCore::ScriptController::initJavaJSBindings):
* bindings/objc/WebScriptObject.mm:
(-[WebScriptObject callWebScriptMethod:withArguments:]):
(-[WebScriptObject evaluateWebScript:]):
(-[WebScriptObject setValue:forKey:]):
(-[WebScriptObject valueForKey:]):
(-[WebScriptObject removeWebScriptKey:]):
(-[WebScriptObject hasWebScriptKey:]):
(-[WebScriptObject stringRepresentation]):
(-[WebScriptObject webScriptValueAtIndex:]):
(-[WebScriptObject setWebScriptValueAtIndex:value:]):
* bridge/jsc/BridgeJSC.cpp:
* bridge/jsc/BridgeJSC.h:
(Instance):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129961 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd draggable property to WebDraggableRegion.
jianli@chromium.org [Fri, 28 Sep 2012 23:38:48 +0000 (23:38 +0000)]
Add draggable property to WebDraggableRegion.
https://bugs.webkit.org/show_bug.cgi?id=97950

Reviewed by Adam Barth.

This needs to be done before we can land the patch to use new syntax for
draggable regions. This is because Chromium side code cannot enable
the changes to support new syntax side by side when this property is
not present.

* public/WebDraggableRegion.h:
(WebDraggableRegion):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129960 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago<rdar://problem/12398537> WebKit2 lacks API for disabling document.cookie like -...
mitz@apple.com [Fri, 28 Sep 2012 23:14:53 +0000 (23:14 +0000)]
<rdar://problem/12398537> WebKit2 lacks API for disabling document.cookie like -[WebView _setCookieEnabled:]
https://bugs.webkit.org/show_bug.cgi?id=97939

Reviewed by Sam Weinig.

* Shared/WebPreferencesStore.h:
(WebKit): Defined CookieEnabled key with a default value of true.
* UIProcess/API/C/WKPreferences.cpp:
(WKPreferencesSetCookieEnabled): Added this setter.
(WKPreferencesGetCookieEnabled): Added this getter.
* UIProcess/API/C/WKPreferencesPrivate.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updatePreferences): Added a call to Settings::setCookieEnabled to push the
preference into Settings.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129959 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoImprove saturation arithmetic support in FractionalLayoutUnit
eae@chromium.org [Fri, 28 Sep 2012 23:12:56 +0000 (23:12 +0000)]
Improve saturation arithmetic support in FractionalLayoutUnit
https://bugs.webkit.org/show_bug.cgi?id=97938

Reviewed by Levi Weintraub.

Source/WebCore:

Fix bug in FractionalLayoutUnit::setValue where greater than or
equals is used instead of greater than to clamp the values.
Add SATURATED_LAYOUT_ARITHMETIC support to round preventing it
from overflowing when adding the fraction.

Test: fast/sub-pixel/large-sizes.html

* platform/FractionalLayoutUnit.h:
(WebCore::FractionalLayoutUnit::round):
(WebCore::FractionalLayoutUnit::setValue):

LayoutTests:

Update large sizes test to better cover values near limit.

* fast/sub-pixel/large-sizes-expected.txt:
* fast/sub-pixel/large-sizes.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129958 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix pluralization of Info.plist key.
weinig@apple.com [Fri, 28 Sep 2012 23:08:09 +0000 (23:08 +0000)]
Fix pluralization of Info.plist key.

Rubber-stamped by Anders Carlsson.

* WebProcessService/Info.plist:
* WebProcessServiceForWebKitDevelopment/Info.plist:
It's _MultipleInstances, not _MultipleInstance. :(

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129957 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoTextureMapperGL destructor crashes
commit-queue@webkit.org [Fri, 28 Sep 2012 23:07:39 +0000 (23:07 +0000)]
TextureMapperGL destructor crashes
https://bugs.webkit.org/show_bug.cgi?id=97942

Patch by Luiz Agostini <luiz.agostini@nokia.com> on 2012-09-28
Reviewed by Noam Rosenthal.

BitmapTextureGL destructor uses a TextureMapperGL instance. The problem is that
BitmapTextureGL objects are destroyed on TextureMapper destructor and at that
time TextureMapperGL specific methods and data are not available any more.

This patch creates a new protected method TextureMapper::clearTexturePool()
that is called in TextureMapperGL's destructor.

* platform/graphics/texmap/TextureMapper.h:
(WebCore::TextureMapper::clearTexturePool):
* platform/graphics/texmap/TextureMapperGL.cpp:
(WebCore::TextureMapperGL::~TextureMapperGL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129956 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoREGRESSION(r124168): Null crash in RenderLayer::createScrollbar
jchaffraix@webkit.org [Fri, 28 Sep 2012 22:42:57 +0000 (22:42 +0000)]
REGRESSION(r124168): Null crash in RenderLayer::createScrollbar
https://bugs.webkit.org/show_bug.cgi?id=96863

Reviewed by Abhishek Arya.

Source/WebCore:

After r124168, we synchronously create any overflow:scroll scrollbar on the first style change - we used to wait
until layout was called. The issue is that the logic in RenderLayer assumes that our node is completely attached
when the style change is dispatched. The crash occured because the 'content' image code path in
RenderObject::createObject triggered a style change too early.

Test: scrollbars/scrollbar-content-crash.html

* rendering/RenderObject.cpp:
(WebCore::RenderObject::createObject):
We need a style associated with the new RenderImage to call setImageResource but we don't need to trigger a
style change.

LayoutTests:

* scrollbars/scrollbar-content-crash-expected.txt: Added.
* scrollbars/scrollbar-content-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129955 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWontFix frame flattening crash test.
ojan@chromium.org [Fri, 28 Sep 2012 22:05:42 +0000 (22:05 +0000)]
WontFix frame flattening crash test.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129954 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoperf/nested-combined-selectors.html is flaky on Lion as well.
ojan@chromium.org [Fri, 28 Sep 2012 21:51:36 +0000 (21:51 +0000)]
perf/nested-combined-selectors.html is flaky on Lion as well.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129953 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd two resources needed for the test added in r129944.
simon.fraser@apple.com [Fri, 28 Sep 2012 21:41:18 +0000 (21:41 +0000)]
Add two resources needed for the test added in r129944.

* plugins/resources/plugin-frame.html: Added.
* plugins/resources/plugin-page.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129952 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMark this test as Missing since it has no png on any platform.
ojan@chromium.org [Fri, 28 Sep 2012 21:32:23 +0000 (21:32 +0000)]
Mark this test as Missing since it has no png on any platform.
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129951 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd canvas to set of elements that do not allow style sharing in order
schenney@chromium.org [Fri, 28 Sep 2012 21:23:25 +0000 (21:23 +0000)]
Add canvas to set of elements that do not allow style sharing in order
to provoke RenderLayer creation
https://bugs.webkit.org/show_bug.cgi?id=97013

Unreviewed expectations update.

* platform/chromium-mac/fast/canvas/canvas-render-layer-expected.png: Added.
* platform/chromium-mac/fast/canvas/canvas-render-layer-expected.txt: Added.
* platform/chromium-mac/platform/chromium/virtual/gpu/fast/canvas/canvas-render-layer-expected.png: Added.
* platform/chromium-mac/platform/chromium/virtual/gpu/fast/canvas/canvas-render-layer-expected.txt: Added.
* platform/chromium-win/fast/canvas/canvas-render-layer-expected.png: Added.
* platform/chromium-win/fast/canvas/canvas-render-layer-expected.txt: Added.
* platform/chromium-win/platform/chromium/virtual/gpu/fast/canvas/canvas-render-layer-expected.png: Added.
* platform/chromium-win/platform/chromium/virtual/gpu/fast/canvas/canvas-render-layer-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@129950 268f45cc-cd09-0410-ab3c-d52691b4dbfc