Simplify WKCACFLayerRenderer's API

createRenderer/destroyRenderer are now hidden behind setHostWindow.
WKCACFLayerRendererClient::animationsStarted has been removed, as it
was never called. (The work it was supposed to do was already being
accomplished by WKCACFLayerRenderer::render telling each layer that
animations are starting.)

Fixes <http://webkit.org/b/52587> WKCACFLayerRenderer is hard to use

Reviewed by Chris Marrin.

Source/WebCore:

* platform/graphics/win/MediaPlayerPrivateFullscreenWindow.cpp:
(WebCore::MediaPlayerPrivateFullscreenWindow::createWindow): Moved the
call to setHostWindow here from the WM_CREATE handler. The WM_CREATE
handler was causing the Direct3D device to be created, but MSDN says
you shouldn't create a device while handling WM_CREATE. Removed
no-longer-needed call to createRenderer (setHostWindow does this now)
and the never-needed call to setNeedsDisplay (we never draw into the
root layer; this was just creating an unnecessary backing store the
size of the screen!).
(WebCore::MediaPlayerPrivateFullscreenWindow::wndProc): Moved WM_CREATE
code, as described above. Removed call to destroyRenderer when handling
WM_DESTROY; setHostWindow does this now. Fixed up our WM_PAINT handler
to do a synchronous paint and to clear our dirty region, while I was in
here.

* platform/graphics/win/WKCACFLayerRenderer.cpp:
(WebCore::WKCACFLayerRenderer::~WKCACFLayerRenderer): Changed to call
setHostWindow instead of destroyRenderer; the former calls the latter
if needed.
(WebCore::WKCACFLayerRenderer::setHostWindow): Moved here from the
header file. Destroys our old renderer (i.e., IDirect3DDevice9) if
we're losing our window, or creates a renderer if we're gaining a
window.
(WebCore::WKCACFLayerRenderer::createRenderer): Updated for WKSI function rename.
(WebCore::WKCACFLayerRenderer::destroyRenderer): Changed to clear the
D3D device from our context before releasing the device.

* platform/graphics/win/WKCACFLayerRenderer.h: Removed
WKCACFLayerRendererClient::animationsStarted. Removed setNeedsDisplay.
Make createRenderer, destroyRenderer, and renderSoon private.

WebKit/win:

Update for WKCACFLayerRenderer changes

* WebView.cpp:
(WebView::WebViewWndProc): Removed call to layerRendererBecameVisible
when handling WM_SHOWWINDOW. All this did was try to create the
renderer, but it had already been created in setAcceleratedCompositing,
so wasn't needed.
(WebView::setAcceleratedCompositing): Removed call to createRenderer;
setHostWindow does this for us now.

* WebView.h: Removed animationsStarted and layerRendererBecameVisible.

WebKitLibraries:

Make it possible to both set and clear a wkCACFContext's D3D device

* win/include/WebKitSystemInterface/WebKitSystemInterface.h:
* win/lib/WebKitSystemInterface.lib:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75988 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Paint directly into a GraphicsLayer when using accelerated compositing

Before this patch, we were painting into our normal backing store
HBITMAP, wrapping it in a CGImage, and handing it off to Core
Animation. This had at least two disadvantages:
   1) The bitmap could be modified while Core Animation was using it.
   2) It introduced extra complexity.

When entering accelerated compositing mode, WebView now creates a
GraphicsLayer to draw into. This GraphicsLayer sits between the root
layer (owned by WKCACFLayerRenderer) and the RenderView's layer. When
WebView invalidates, it just calls setNeedsDisplayInRect on its
GraphicsLayer. When WebView paints, it just tells its
WKCACFLayerRenderer to paint, which will call back to WebView to draw
into the GraphicsLayer if it has a dirty region. This is very similar
to the current implementation of LayerBackedDrawingArea in WebKit2.

Fixes <http://webkit.org/b/52582> WebView should paint directly into a
GraphicsLayer when in accelerated compositing mode

Reviewed by Simon Fraser and Chris Marrin.

WebCore:

Remove special-cased support for providing content for the root layer

Clients will just have to provide content through the normal
GraphicsLayer channels now!

* platform/graphics/win/WKCACFLayerRenderer.cpp:
(WebCore::WKCACFLayerRenderer::WKCACFLayerRenderer): Removed
initialization of m_backingStoreDirty, which has itself been removed.
(WebCore::WKCACFLayerRenderer::setNeedsDisplay): Moved code to schedule
a sync from here to syncCompositingStateSoon. We only need to call
renderSoon if we don't call syncCompositingStateSoon; the latter
function calls the former.
(WebCore::WKCACFLayerRenderer::paint): Removed code to handle
m_backingStoreDirty. We don't want to know anything about clients'
backing stores.
(WebCore::WKCACFLayerRenderer::syncCompositingStateSoon): Added. Code
came from setNeedsDisplay.

* platform/graphics/win/WKCACFLayerRenderer.h:
Removed setRootContents[AndDisplay], setBackingStoreDirty, and
m_backingStoreDirty. Made paint() public so that clients can force a
synchronous render (e.g., when handling WM_PAINT).

WebKit/win:

* WebCoreSupport/WebChromeClient.cpp:
(WebChromeClient::attachRootGraphicsLayer):
(WebChromeClient::scheduleCompositingLayerSync):
Updated for WebView changes.

* WebView.cpp:
(WebView::repaint): When using accelerated compositing, just invalidate
our backing layer. This matches the behavior of LayerBackedDrawingArea.
(WebView::deleteBackingStore): Removed accelerated compositing code.
The WKCACFLayerRenderer doesn't need to know about our backing store
anymore; we don't use it at all when in accelerated compositing mode.
(WebView::addToDirtyRegion): When using accelerated compositing, just
invalidate our backing layer.
(WebView::scrollBackingStore): Added an assertion that this isn't
called in accelerated compositing mode.
(WebView::sizeChanged): Update our backing layer's size, too, and
invalidate it.
(WebView::updateBackingStore): Added an assertion that this isn't
called in accelerated compositing mode.
(WebView::paint): If we're in accelerated compositing mode, sync our
compositing state. If we're *still* in accelerated compositing mode,
just tell our WKCACFLayerRenderer to paint and clear our dirty region.
(The later changes in this function are just un-indenting code that
used to be inside an if.)

(WebView::paintIntoBackingStore):
(WebView::paintIntoWindow):
Added assertions that these aren't called in accelerated compositing
mode.

(WebView::WebViewWndProc): Updated WM_XP_THEMECHANGED handling for
removal of setRootLayerNeedsDisplay.
(WebView::setRootChildLayer): Changed to take a GraphicsLayer. We now
set the layer as a child of our own backing layer.
(WebView::scheduleCompositingLayerSync): Just call through to
WKCACFLayerRenderer.
(WebView::setAcceleratedCompositing): Create our backing layer and set
it as the child of WKCACFLayerRenderer's root layer.
(WebView::notifyAnimationStarted): Added. We never expect this
GraphicsLayerClient override to be called, as we don't use animations
on our backing layer.
(WebView::notifySyncRequired): Added. Just schedule a sync.
(WebView::paintContents): Added. Just clip and paint!

(WebView::showDebugBorders):
(WebView::showRepaintCounter):
Added. These just call through to Settings.

(WebView::syncCompositingState): Changed to first update layout, then
sync state for our backing layer, then sync WebCore's state. This
matches LayerBackedDrawingArea.

* WebView.h: WebView now implements the GraphicsLayerClient interface.
Removed setRootLayerNeedsDisplay; it's been replaced by calling
setNeedsDisplay on our backing layer and calling
syncCompositingStateSoon on WKCACFLayerRenderer as needed. Removed
updateRootLayerContents; that function was used to pass our backing
store to Core Animation, which we no longer do. Added m_backingLayer.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75987 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove contexts from WKCACFContextFlusher before destroying them

We aren't really using WKCACFContextFlusher for anything useful at the
moment, but that will probably change in the near future.

I couldn't come up with a way to test this because it isn't possible to
resize a window in DumpRenderTree.

Fixes <http://webkit.org/b/52573> REGRESSION (r75262): Crash beneath
WKCACFContextFlusher::flushAllContexts when resizing window on page
that uses accelerated compositing

Reviewed by Simon Fraser.

* platform/graphics/win/WKCACFLayerRenderer.cpp:
(WebCore::WKCACFLayerRenderer::~WKCACFLayerRenderer): Remove our
context from WKCACFContextFlusher. (This code was erroneously removed
from WKCACFLayerRenderer::destroyRenderer in r75262. This is a more
sensible place for it.)

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75986 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Don't access the CACFLayerRef's sublayers directly from PlatformCALayer

There might be a secret extra sublayer (the tile parent layer) that
PlatformCALayer doesn't know about. When PlatformCALayer would
encounter this, it would try to use the tile parent layer's
PlatformCALayer wrapper, which was null, and then would crash. We now
ask PlatformCALayerWinInternal for the sublayer list, since that class
knows about the tile parent layer and can exclude it from the sublayer
list.

Covered by compositing/tiling/huge-layer-resize.html.

Fixes <http://webkit.org/b/52597> Crash beneath
PlatformCALayer::adoptSublayers when switching out of tiling mode
(null-dereference of a PlatformCALayer)

Reviewed by Darin Adler and Chris Marrin.

LayoutTests:

Make compositing/tiling/huge-layer-resize.html faster, more reliable,
and more crashy (when there's a WebKit bug)

This test was trying to cause a layout/paint to happen by returning to
the event loop for a certain amount of time via setTimeout. But this
didn't always result in a layout/paint (at least on Windows). We now
force the layout/paint explicitly, which also lets us speed up the test
by removing the setTimeouts.

* compositing/tiling/huge-layer-resize.html:
(testOnLoad): Changed to use recordLayerTree, which forces a
layout/paint, instead of hoping that setTimeout will do the trick.
(recordLayerTree): Forces a layout/paint, then dumps the layer tree.

Source/WebCore:

* platform/graphics/ca/win/PlatformCALayerWin.cpp:
(PlatformCALayer::adoptSublayers):
(printLayer):
Changed to use PlatformCALayerWinInternal::getSublayers.

* platform/graphics/ca/win/PlatformCALayerWinInternal.cpp:
(PlatformCALayerWinInternal::getSublayers): Added. Retrieves the list
of PlatformCALayers that represent our sublayers. Significantly, this
code knows about the tile parent layer and can thus exclude it.

* platform/graphics/ca/win/PlatformCALayerWinInternal.h: Added
getSublayers.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75985 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17 William Siegrist <wsiegrist@apple.com>

Explicitly set mode during file upload to ensure web server can read the files.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75984 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Nico Weber  <thakis@chromium.org>

        Unreviewed.

        Roll chromium DEPS forward to pick up the skia roll to skia r692.
        https://bugs.webkit.org/show_bug.cgi?id=52600

        * DEPS:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75983 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Naoki Takano  <takano.naoki@gmail.com>

        Reviewed by Kent Tamura.

        [Chromium] Fix popup menu re-positioning when the menu is opened upward, above the corresponding form field.
        https://bugs.webkit.org/show_bug.cgi?id=51382
        http://crbug.com/60427

        Calculate correct location of popup window whenever the items in the window change.

        No new tests, because this fix is for Chromium project and hard to test only in WebKit project

        * platform/chromium/PopupMenuChromium.cpp:
        (WebCore::PopupContainer::layoutAndCalculateWidgetRect): New Function to layout and calculate popup widget rect.
        (WebCore::PopupContainer::showPopup): Move widgetRect calculation logic to calculateWidgetRect().
        (WebCore::PopupContainer::refresh): Add parameter focusRect to take the location and the size of focus text input field to calculate correct popup window location.
        * platform/chromium/PopupMenuChromium.h: Append new input parameter for refresh().
2011-01-17  Naoki Takano  <takano.naoki@gmail.com>

        Reviewed by Kent Tamura.

        [Chromium] Fix popup menu re-positioning when the menu is opened upward, above the corresponding form field.
        https://bugs.webkit.org/show_bug.cgi?id=51382
        http://crbug.com/60427

        Calculate correct location of popup window whenever the items in the window change.

        No new tests, because this fix is for Chromium project and hard to test only in WebKit project

       * WebKit/chromium/src/WebViewImpl.cpp:
       (WebKit::WebViewImpl::refreshAutoFillPopup): Change the logic in refreshAutoFilPopup() to check both the location and the size of popup window.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75982 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Windows build fix. Unreviewed.

* Shared/PrintInfo.cpp:
Use #include, not #import.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75981 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Actually wrote a FIXME promised for a review comment.

        * UIProcess/API/mac/WKView.mm:
        (-[WKView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75980 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Anders Carlsson and Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=52595
        Make basic printing work in WebKit2

        * Shared/WebPreferencesStore.h:
        * UIProcess/API/C/WKPreferences.cpp:
        (WKPreferencesSetShouldPrintBackgrounds):
        (WKPreferencesGetShouldPrintBackgrounds):
        * UIProcess/API/C/WKPreferences.h:
        Implement a preference for printing backgrounds.

        * Shared/PrintInfo.cpp: Added.
        (WebKit::PrintInfo::encode):
        (WebKit::PrintInfo::decode):
        * Shared/PrintInfo.h: Added.
        * Shared/mac/PrintInfoMac.mm: Added.
        (WebKit::PrintInfo::PrintInfo):
        A new class for passing page setup information down to the web process.

        * WebKit2.xcodeproj/project.pbxproj:
        * GNUmakefile.am:
        * WebKit2.pro:
        * win/WebKit2.vcproj:
        Added PrintInfo.

        * Scripts/webkit2/messages.py: Added PrintInfo.h to a list of structs.

        * UIProcess/API/mac/WKView.mm:
        (WKViewData): Cache page count and geometry.
        (WebFrameWrapper): Added a wrapper class for adding a WebFrameProxy to an NSDictionary.
        (-[WKView drawRect:]): Added logging.
        (setFrameBeingPrinted): Store a frame into NSPrintOperation's NSPrintInfo, so that all
        methods invoked for this operation would know what's being printed.
        (frameBeingPrinted): Retrieve this data.
        (-[WKView printOperationWithPrintInfo:forFrame:]): Create a print operation that's appropriate
        for the frame.
        (-[WKView knowsPageRange:]): Ask the web process.
        (-[WKView _recursiveDisplayRectIfNeededIgnoringOpacity:isVisibleRect:rectIsVisibleRectForView:topView:]):
        Take over printing, because I couldn't make drawRect: work correctly with WebKit custom fit
        to page implementation.
        (-[WKView _provideTotalScaleFactorForPrintOperation:]): Help AppKit clip correctly.
        (-[WKView rectForPage:]): Return a pre-calculated rect.

        * UIProcess/WebPageProxy.h: Added m_isInPrintingMode to track whether the web page is in
        printing mode. The UI process controls that.

        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::WebPageProxy): Initialize m_isInPrintingMode.
        (WebKit::WebPageProxy::beginPrinting): Change printing mode in web process if the value changes.
        (WebKit::WebPageProxy::endPrinting): Ditto.
        (WebKit::WebPageProxy::computePagesForPrinting): Ask the web process about output page geometry.
        (WebKit::WebPageProxy::drawRectToPDF): Ask the web process about printed page data.

        * WebProcess/WebPage/WebPage.cpp:
        (WebKit::WebPage::close): Delete m_printContext while there is still a Frame pointer.
        (WebKit::WebPage::updatePreferences): Pass through ShouldPrintBackgrounds.
        (WebKit::WebPage::beginPrinting): Switch into printing mode (or update parameters if we're
        already in printing mode).
        (WebKit::WebPage::endPrinting): Delete m_printContext, implicitly calling end().
        (WebKit::WebPage::computePagesForPrinting): Make sure that we're in printing mode, and
        copmute page rects.
        (WebKit::WebPage::drawRectToPDF): Added a CG specific function to create a PDF document out
        of a page rect. It should really work for onscreen content, but we only need it for printing.

        * WebProcess/WebPage/WebPage.h:

        * WebProcess/WebPage/WebPage.messages.in: Added BeginPrinting, EndPrinting,
        ComputePagesForPrinting and DrawRectToPDF.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75979 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix the 32-bit build.

* WebCore.exp.in:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75976 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Use of invalid hash map key in CSSFontFaceSource::getFontData() with 0-sized remote font
https://bugs.webkit.org/show_bug.cgi?id=52598

Reviewed by Darin Adler.

Source/WebCore:

Test: fast/css/font-face-zero-hash-key.html

* css/CSSFontFaceSource.cpp:
(WebCore::CSSFontFaceSource::getFontData): Add 1 to the font size to avoid a 0 hash key.
* css/CSSSegmentedFontFace.cpp:
(WebCore::CSSSegmentedFontFace::getFontData): Ditto.

LayoutTests:

* fast/css/font-face-zero-hash-key-expected.txt: Added.
* fast/css/font-face-zero-hash-key.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75975 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<webkit.org/b/52596> Add missing DOMDocument/DOMDocumentFragment headers to Xcode project

Reviewed by Dan Bernstein.

WebCore:

This fixes two issues:

Add missing DOMDocumentFragmentPrivate.h and
DOMDocumentPrivate.h files to the project.  These files were
never added to the the project although their *Internal.h
counterparts were added in r16548 and r17390.

Add missing DOMDocumentFragmentInternal.h to the Headers
section.  It was moved from Headers to "Copy Generated Headers"
in r31045, but it should have an entry in both sections.

* WebCore.xcodeproj/project.pbxproj:

WebKit/mac:

* MigrateHeaders.make: Copy DOMDocumentFragmentPrivate.h and
DOMDocumentPrivate.h to the PrivateHeaders directory.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75974 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Helder Correia  <helder@sencha.com>

        Reviewed by Andreas Kling.

        [Qt] Incorrect shadow alpha with pattern fillStyle
        https://bugs.webkit.org/show_bug.cgi?id=52559

        The shadow color opacity needs to be set on the shadow painter.
        This is related to bug 52556.

        * fast/canvas/canvas-fillPath-pattern-shadow-expected.txt: Added.
        * fast/canvas/canvas-fillPath-pattern-shadow.html: Added.
        * fast/canvas/script-tests/canvas-fillPath-pattern-shadow.js: Added.
2011-01-17  Helder Correia  <helder@sencha.com>

        Reviewed by Andreas Kling.

        [Qt] Incorrect shadow alpha with pattern fillStyle
        https://bugs.webkit.org/show_bug.cgi?id=52559

        The shadow color opacity needs to be set on the shadow painter.
        This is related to bug 52556.

        Test: fast/canvas/canvas-fillPath-pattern-shadow.html

        * platform/graphics/qt/GraphicsContextQt.cpp:
        (WebCore::GraphicsContext::fillPath):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75973 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Andreas Kling.

        [GTK] Port a crash fix from libsoup upstream to the libsoup cache code
        https://bugs.webkit.org/show_bug.cgi?id=52586

        No new tests. This is just a merge from upstream.

        * platform/network/soup/cache/soup-http-input-stream.c:
        (webkit_soup_http_input_stream_got_chunk): Properly handle unsigned numbers.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75972 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Fix some headers with missing or misspelled #ifndef guards
        https://bugs.webkit.org/show_bug.cgi?id=52545

        * wtf/RefPtrHashMap.h:
2011-01-17  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Fix some headers with missing or misspelled #ifndef guards
        https://bugs.webkit.org/show_bug.cgi?id=52545

        No new tests because no new functionality.

        * ForwardingHeaders/runtime/InitializeThreading.h:
        * editing/SmartReplace.h:
        * loader/CrossOriginAccessControl.h:
        * loader/NetscapePlugInStreamLoader.h:
        * platform/chromium/ClipboardUtilitiesChromium.h:
        * platform/graphics/cairo/DrawErrorUnderline.h:
        * platform/graphics/cairo/GraphicsContextPlatformPrivateCairo.h:
        * platform/graphics/cg/GraphicsContextPlatformPrivateCG.h:
        * platform/graphics/cg/PDFDocumentImage.h:
        * platform/win/WebCoreTextRenderer.h:
2011-01-17  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Fix some headers with missing or misspelled #ifndef guards
        https://bugs.webkit.org/show_bug.cgi?id=52545

        * WebKitStatistics.h:
2011-01-17  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Fix some headers with missing or misspelled #ifndef guards
        https://bugs.webkit.org/show_bug.cgi?id=52545

        * Shared/UserMessageCoders.h:
        * WebProcess/InjectedBundle/DOM/InjectedBundleNodeHandle.h:
        * WebProcess/InjectedBundle/InjectedBundleUserMessageCoders.h:
        * WebProcess/WebProcessMain.h:
2011-01-17  Tony Gentilcore  <tonyg@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Fix some headers with missing or misspelled #ifndef guards
        https://bugs.webkit.org/show_bug.cgi?id=52545

        * DumpRenderTree/chromium/TestEventPrinter.h:
        * DumpRenderTree/chromium/WebPreferences.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75971 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  David Kilzer  <ddkilzer@apple.com>

        <http://webkit.org/b/52524> fast/dom/Range/range-clone-contents.html fails only on Windows

        Reviewed by Alexey Proskuryakov.

        * platform/win/Skipped: Removed
        fast/dom/Range/range-clone-contents.html.  It appears the
        Windows buildbots need a clean build.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75970 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: follow-up fix for r52574: do not reuse clear() for navigation.

        * inspector/front-end/NetworkPanel.js:
        (WebInspector.NetworkPanel.prototype.clear):
        (WebInspector.NetworkPanel.prototype.mainResourceChanged):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75969 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Build fix.

        * Platform/Logging.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75968 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Sergio Villar Senin  <svillar@igalia.com>

        Reviewed by Martin Robinson.

        [Gtk] No need to content sniff 304 Not Modified responses
        https://bugs.webkit.org/show_bug.cgi?id=52570

        Makes no sense to wait for the outcome of content sniffing when WebCore
        is validating resources. If we get a 304 Not Modified it means that we can
        safely use the cached version of the resource we're asking for.

        No new tests because it does not change functionality, it just
        calls didReceiveResponse sooner for 304 Not Modified responses.

        * platform/network/soup/ResourceHandleSoup.cpp:
        (WebCore::gotHeadersCallback):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75967 268f45cc-cd09-0410-ab3c-d52691b4dbfc

REGRESSION: Assertion failure in FrameLoader::continueLoadAfterWillSubmitForm() when
navigating back to an unreachable URL
https://bugs.webkit.org/show_bug.cgi?id=52388

Reviewed by Anders Carlsson.

Source/WebCore:

Test: http/tests/navigation/go-back-to-error-page.html

* history/PageCache.cpp:
(WebCore::PageCache::canCachePageContainingThisFrame):
Do not cache any error pages (which we can recognize as having substitute data and/or an
unreachableURL).

Tools:

Add a third parameter (unreachableURL) to queueLoadHTMLString.

* DumpRenderTree/LayoutTestController.cpp:
(queueLoadHTMLStringCallback):
(LayoutTestController::queueLoadHTMLString):
(LayoutTestController::queueLoadAlternateHTMLString):
* DumpRenderTree/LayoutTestController.h:

* DumpRenderTree/WorkQueueItem.h:
(LoadHTMLStringItem::LoadHTMLStringItem):

* DumpRenderTree/chromium/LayoutTestController.cpp:
(WorkItemLoadHTMLString::WorkItemLoadHTMLString):
(WorkItemLoadHTMLString::run):
(LayoutTestController::queueLoadHTMLString):

* DumpRenderTree/gtk/WorkQueueItemGtk.cpp:
(LoadHTMLStringItem::invoke):
* DumpRenderTree/mac/WorkQueueItemMac.mm:
(LoadHTMLStringItem::invoke):
* DumpRenderTree/win/WorkQueueItemWin.cpp:
(LoadHTMLStringItem::invoke):

LayoutTests:

Add a test that loads an alternate HTML String for an "unreachable" URL, loads another
page, and then goes back to the "unreachable" URL, which is now reachable.

This tests two things:
1) No assertion failure when going back to an unreachable URL.
2) There is no page cached for the unreachable URL.

* http/tests/navigation/go-back-to-error-page-expected.txt: Added.
* http/tests/navigation/go-back-to-error-page.html: Added.
* http/tests/navigation/resources/page-to-go-back-from.html: Added.
* http/tests/navigation/resources/page-treated-as-unreachable.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75966 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=52592
        Add a logging channel for WKView

        * Platform/Logging.cpp:
        * Platform/Logging.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75965 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: fixing typo in r75952.

        * inspector/front-end/NetworkPanel.js:
        (WebInspector.NetworkDataGridNode.prototype._refreshNameCell):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75964 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Drag and drop support: refactoring of image from link and image from selection
https://bugs.webkit.org/show_bug.cgi?id=52496

Reviewed by Alexey Proskuryakov.

Source/WebCore:

This work cleans up the Mac code and makes it more similar to the Windows implementation,
avoiding the use of an NSView when the FrameView can be used.
The refactoring is a necessary step towards the complete support of drag and drop
in WebKit2.

* page/mac/FrameMac.mm:
(WebCore::Frame::imageFromRect): Modified to use FrameView instead of NSView
to generate the image for drag.

Source/WebKit/mac:

This work cleans up the Mac code and makes it more similar to the Windows implementation,
avoiding the use of an NSView when the FrameView can be used.
The refactoring is a necessary step towards the complete support of drag and drop
in WebKit2.

* WebCoreSupport/WebDragClient.mm:
(WebDragClient::createDragImageForLink): Added.
* WebView/WebHTMLView.mm: Removed dragImageFromLink and dragImageFromURL.
* WebView/WebHTMLViewPrivate.h: Removed dragImageFromLink and dragImageFromURL.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75963 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Factor bitmap context drawing code out into a shared function
        https://bugs.webkit.org/show_bug.cgi?id=52589

        * Platform/cg/CGUtilities.h: Added.
        * Platform/cg/CGUtilities.cpp: Added.
        (WebKit::paintBitmapContext):
        New function that will paint a bitmap context into a CGContext.

        * Shared/cg/ShareableBitmapCG.cpp:
        (WebKit::ShareableBitmap::paint):
        Use paintBitmapContext.

        * UIProcess/mac/BackingStoreMac.mm:
        (WebKit::BackingStore::paint):
        (WebKit::BackingStore::scroll):
        Use paintBitmapContext.

        * WebKit2.xcodeproj/project.pbxproj:
        * win/WebKit2.vcproj:
        * win/WebKit2Common.vsprops:
        Add CGUtilities.{cpp|h}.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75962 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Update xcodeproj svn:ignore to include xcuserdata.

Rubber-stamped by Mark Rowe.

Examples:

* NetscapeCocoaPlugin/NetscapeCocoaPlugin.xcodeproj: Modified property svn:ignore.
* NetscapeCoreAnimationMoviePlugin/NetscapeCoreAnimationMoviePlugin.xcodeproj: Modified property svn:ignore.
* NetscapeCoreAnimationPlugin/NetscapeCoreAnimationPlugin.xcodeproj: Modified property svn:ignore.
* NetscapeInputMethodPlugin/NetscapeInputMethodPlugin.xcodeproj: Modified property svn:ignore.

Source/JavaScriptCore:

* JavaScriptCore.xcodeproj: Modified property svn:ignore.

Source/JavaScriptGlue:

* JavaScriptGlue.xcodeproj: Modified property svn:ignore.

Source/ThirdParty/ANGLE:

* ANGLE.xcodeproj: Modified property svn:ignore.
* src/build_angle.xcodeproj: Modified property svn:ignore.

Source/WebCore:

* WebCore.xcodeproj: Modified property svn:ignore.
* manual-tests/NPN_Invoke/NPN_Invoke.xcodeproj: Modified property svn:ignore.

Source/WebKit:

* WebKit.xcodeproj: Modified property svn:ignore.

Source/WebKit2:

* WebKit2.xcodeproj: Modified property svn:ignore.

Tools:

* DumpRenderTree/DumpRenderTree.xcodeproj: Modified property svn:ignore.
* MiniBrowser/MiniBrowser.xcodeproj: Modified property svn:ignore.
* TestWebKitAPI/TestWebKitAPI.xcodeproj: Modified property svn:ignore.
* WebKitLauncher/WebKitLauncher.xcodeproj: Modified property svn:ignore.
* WebKitTestRunner/WebKitTestRunner.xcodeproj: Modified property svn:ignore.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75961 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Fixing a typo from a bad merge.

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

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75960 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by John Sullivan.

        https://bugs.webkit.org/show_bug.cgi?id=52590
        Should pass a frame into -[WKView canPrintHeadersAndFooters]

        * UIProcess/API/mac/WKView.h:
        * UIProcess/API/mac/WKView.mm: (-[WKView canChangeFrameLayout:]): Also, renamed the method
        to canChangeFrameLayout.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75959 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Leopard build fix.

        * page/PrintContext.cpp: GCC complained about shortening a double value to float.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75958 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove a few more empty directories that I missed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75957 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Remove a large number of empty directories.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75956 268f45cc-cd09-0410-ab3c-d52691b4dbfc

    Reviewed by Dan Bernstein.

        https://bugs.webkit.org/show_bug.cgi?id=52495

        No change in behavior, so no tests.

        * WebCore.exp.in: Export additional methods, as I'm going to use more code from PrintContext.

        * page/PrintContext.cpp:
        (WebCore::PrintContext::~PrintContext): No need to clear m_pageRects, the object is being
        destroyed already.
        (WebCore::PrintContext::pageCount): Changed page count from int to size_t.
        (WebCore::PrintContext::pageRect): Ditto.
        (WebCore::PrintContext::computePageRects): Pass allowHorizontalTiling as an argument.
        PrintContext already has code to calculate scale factor, so it makes sense to make methods
        that contain it universal (allowHorizontalTiling is always true for Safari).
        Round page height to an integer, because Mac code does that, and because page height is
        treated as integer almost everywhere else in code.
        (WebCore::PrintContext::begin): Allow calling this function multiple times. There is no need
        to return to screen mode if e.g. "print backgounds" option changes.
        (WebCore::PrintContext::computeAutomaticScaleFactor): Expose scale factor computation, so
        that clients don't have to copy it.
        (WebCore::PrintContext::spoolRect): Add a way to spool a precomputed rect - handy if a request
        comes from code that doesn't have page number.
        (WebCore::PrintContext::pageNumberForElement): Page number int -> size_t.

        * page/PrintContext.h: Added comments and FIXMEs. PrintContext needs cleanup, but that
        depends on deciding how it really needs to work (e.g. whether computePageRects() should
        cause relayout).

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75955 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: skip flaky inspector test (will look at it tomorrow).

        * platform/mac-leopard/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75954 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Implement scrolling support
        https://bugs.webkit.org/show_bug.cgi?id=52502

        * Shared/UpdateInfo.cpp:
        (WebKit::UpdateInfo::encode):
        (WebKit::UpdateInfo::decode):
        * Shared/UpdateInfo.h:
        Add scrollRect and scrollDelta member variables.

        * UIProcess/BackingStore.h:
        Add scroll member function.

        * UIProcess/DrawingAreaProxyImpl.cpp:
        (WebKit::DrawingAreaProxyImpl::incorporateUpdate):
        Repaint the scroll rect. Force a display of the view when the update info contains a scroll rect.

        * UIProcess/mac/BackingStoreMac.mm:
        (WebKit::BackingStore::incorporateUpdate):
        Call scroll.

        (WebKit::BackingStore::scroll):
        Paint the backing store into itself.

        * WebProcess/WebPage/DrawingAreaImpl.cpp:
        (WebKit::DrawingAreaImpl::display):
        Pass the scroll information in the update info.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75953 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: unify image data source assignment, add image url
        to the image view properties list.
        https://bugs.webkit.org/show_bug.cgi?id=52584

        * English.lproj/localizedStrings.js:
        * inspector/front-end/ImageView.js:
        (WebInspector.ImageView.prototype._createContentIfNeeded.onImageLoad):
        (WebInspector.ImageView.prototype._createContentIfNeeded):
        * inspector/front-end/NetworkPanel.js:
        (WebInspector.NetworkDataGridNode.prototype._refreshNameCell):
        * inspector/front-end/Resource.js:
        (WebInspector.Resource.prototype.populateImageSource):
        (WebInspector.Resource.prototype._contentURL):
        * inspector/front-end/ResourcesPanel.js:
        (WebInspector.FrameResourceTreeElement.prototype.onattach):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75952 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix up the `make` build on Macs

Rubberstamped by Adam Roben.

* Makefile: Point to the correct Makefile.shared

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75951 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Andrey Kosyakov  <caseq@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: [Extensions API] webInspector.resources.onFinished is not fired for redirected resources
        Make resource start/finish hanlding more consistent.
        https://bugs.webkit.org/show_bug.cgi?id=52452

        * inspector/front-end/AuditLauncherView.js: Ignore WebSocket resources when displaying progress indicator
        (WebInspector.AuditLauncherView.prototype._resetResourceCount):
        (WebInspector.AuditLauncherView.prototype.resourceStarted):
        (WebInspector.AuditLauncherView.prototype.resourceFinished):
        * inspector/front-end/NetworkManager.js:
        (WebInspector.NetworkManager): Factor out resource start/finish logic to _startResource()/_finishResource()
        (WebInspector.NetworkManager.prototype.identifierForInitialRequest):
        (WebInspector.NetworkManager.prototype.willSendRequest):
        (WebInspector.NetworkManager.prototype.markResourceAsCached):
        (WebInspector.NetworkManager.prototype.didReceiveResponse):
        (WebInspector.NetworkManager.prototype.didReceiveContentLength):
        (WebInspector.NetworkManager.prototype.didFinishLoading):
        (WebInspector.NetworkManager.prototype.didFailLoading):
        (WebInspector.NetworkManager.prototype.didLoadResourceFromMemoryCache):
        (WebInspector.NetworkManager.prototype.didCreateWebSocket):
        (WebInspector.NetworkManager.prototype.willSendWebSocketHandshakeRequest):
        (WebInspector.NetworkManager.prototype.didReceiveWebSocketHandshakeResponse):
        (WebInspector.NetworkManager.prototype.didCloseWebSocket):
        (WebInspector.NetworkManager.prototype._appendRedirect):
        (WebInspector.NetworkManager.prototype._startResource):
        (WebInspector.NetworkManager.prototype._finishResource):

2011-01-17  Andrey Kosyakov  <caseq@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: [Extensions API] webInspector.resources.onFinished is not fired for redirected resources
        Added test for finished event being sent for redirected resources.
        https://bugs.webkit.org/show_bug.cgi?id=52452

        * http/tests/inspector/extensions-resources-redirect-expected.txt:
        * http/tests/inspector/extensions-resources-redirect.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75950 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: restore dom and network state upon frontend reuse.
        https://bugs.webkit.org/show_bug.cgi?id=52574

        * inspector/Inspector.idl:
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::restoreInspectorStateFromCookie):
        (WebCore::InspectorController::populateScriptObjects):
        (WebCore::InspectorController::pushDataCollectedOffline):
        (WebCore::InspectorController::didCommitLoad):
        * inspector/InspectorController.h:
        * inspector/front-end/NetworkManager.js:
        (WebInspector.NetworkManager.prototype.reset):
        * inspector/front-end/NetworkPanel.js:
        (WebInspector.NetworkPanel.prototype.clear):
        (WebInspector.NetworkPanel.prototype.mainResourceChanged):
        * inspector/front-end/ResourceTreeModel.js:
        (WebInspector.ResourceTreeModel):
        (WebInspector.ResourceTreeModel.prototype.reloadCachedResources):
        * inspector/front-end/ResourcesPanel.js:
        (WebInspector.ResourcesPanel.prototype.clear):
        * inspector/front-end/inspector.js:
        (WebInspector.frontendReused):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75949 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed fix after r75908.

Move WebKit into Source
https://bugs.webkit.org/show_bug.cgi?id=52530

* BuildSlaveSupport/build.webkit.org-config/master.cfg: Add Source to the path for Qt API tests.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75948 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: first timeline-network-resource flakiness.

        * inspector/timeline-network-resource.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75947 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Andrey Kosyakov  <caseq@chromium.org>

        Unreviewed. Fixed malformed reference to WebKitVSPropsRedirectionDir that broke win build.

        * win/WebKit2.vcproj:
        * win/WebKit2GeneratedCommon.vsprops:
        * win/WebKit2WebProcess.vcproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75946 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed buildfix after r75944.

* WebCore.pro:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75945 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Yi Shen  <yi.4.shen@nokia.com>

        Reviewed by Andreas Kling.

        [Qt] Extend the Platform Plugin to support full screen video handler
        https://bugs.webkit.org/show_bug.cgi?id=51249

        Make MediaPlayerPrivateQt support a fullscreen player.

        No new tests because LayoutTests/media/media-fullscreen-inline.html already exists.
        However, this test failed for Qt (QtMediaPlayer) due to durationchange event getting fired twice.
        So, still skip it for Qt.

        * WebCore.pro:
        * features.pri:
        * platform/graphics/qt/MediaPlayerPrivateQt.cpp:
        (WebCore::MediaPlayerPrivateQt::MediaPlayerPrivateQt):
        (WebCore::MediaPlayerPrivateQt::removeVideoItem):
        (WebCore::MediaPlayerPrivateQt::restoreVideoItem):
        * platform/graphics/qt/MediaPlayerPrivateQt.h:
        (WebCore::MediaPlayerPrivateQt::supportsFullscreen):
        (WebCore::MediaPlayerPrivateQt::mediaPlayer):
2011-01-17  Yi Shen  <yi.4.shen@nokia.com>

        Reviewed by Andreas Kling.

        [Qt] Extend the Platform Plugin to support full screen video handler
        https://bugs.webkit.org/show_bug.cgi?id=51249

        Add a new webkit platform plugin interface to support full screen video handler.

        * Api/qwebkitplatformplugin.h:
        (QWebFullScreenVideoHandler::QWebFullScreenVideoHandler):
        (QWebFullScreenVideoHandler::~QWebFullScreenVideoHandler):
        * WebCoreSupport/ChromeClientQt.cpp:
        (WebCore::ChromeClientQt::ChromeClientQt):
        (WebCore::ChromeClientQt::~ChromeClientQt):
        (WebCore::ChromeClientQt::fullScreenVideo):
        (WebCore::ChromeClientQt::supportsFullscreenForNode):
        (WebCore::ChromeClientQt::requiresFullscreenForVideoPlayback):
        (WebCore::ChromeClientQt::enterFullscreenForNode):
        (WebCore::ChromeClientQt::exitFullscreenForNode):
        * WebCoreSupport/ChromeClientQt.h:
        * WebCoreSupport/FullScreenVideoQt.cpp: Added.
        (WebCore::FullScreenVideoQt::FullScreenVideoQt):
        (WebCore::FullScreenVideoQt::~FullScreenVideoQt):
        (WebCore::FullScreenVideoQt::enterFullScreenForNode):
        (WebCore::FullScreenVideoQt::exitFullScreenForNode):
        (WebCore::FullScreenVideoQt::aboutToClose):
        (WebCore::FullScreenVideoQt::mediaPlayer):
        (WebCore::FullScreenVideoQt::mediaPlayerForNode):
        (WebCore::FullScreenVideoQt::requiresFullScreenForVideoPlayback):
        * WebCoreSupport/FullScreenVideoQt.h: Added.
        (WebCore::FullScreenVideoQt::isValid):
        * WebCoreSupport/QtPlatformPlugin.cpp:
        (WebCore::QtPlatformPlugin::createFullScreenVideoHandler):
        * WebCoreSupport/QtPlatformPlugin.h:
        * examples/platformplugin/WebPlugin.cpp:
        (FullScreenVideoWidget::FullScreenVideoWidget):
        (FullScreenVideoWidget::event):
        (FullScreenVideoWidget::keyPressEvent):
        (FullScreenVideoHandler::FullScreenVideoHandler):
        (FullScreenVideoHandler::~FullScreenVideoHandler):
        (FullScreenVideoHandler::requiresFullScreenForVideoPlayback):
        (FullScreenVideoHandler::enterFullScreen):
        (FullScreenVideoHandler::exitFullScreen):
        (WebPlugin::supportsExtension):
        (WebPlugin::createExtension):
        * examples/platformplugin/WebPlugin.h:
        (FullScreenVideoWidget::~FullScreenVideoWidget):
        * examples/platformplugin/platformplugin.pro:
        * examples/platformplugin/qwebkitplatformplugin.h:
        (QWebFullScreenVideoHandler::QWebFullScreenVideoHandler):
        (QWebFullScreenVideoHandler::~QWebFullScreenVideoHandler):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75944 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: fix flaky debugger tests.
        https://bugs.webkit.org/show_bug.cgi?id=51320

        * http/tests/inspector/debugger-test2.js:
        (initialize_DebuggerTest.InspectorTest.setBreakpoint):
        * inspector/debugger-breakpoints-not-activated-on-reload.html:
        * inspector/debugger-pause-on-breakpoint-expected.txt:
        * inspector/debugger-pause-on-breakpoint.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75943 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed.

Remove accidentally committed debug printing after r75908.

* Scripts/webkitdirs.pm:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75942 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Unreviewed.

Add Qt specific expected results for passing CSS tests.
Compared to Mac png files and Mac expected files.

* platform/qt/Skipped: Unskip passing tests.
* platform/qt/css1/[...]: Added.
* platform/qt/css2.1/[...]: Added.
* platform/qt/css3/[...]: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75941 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Anthony Ricaud  <rik@webkit.org>

        Reviewed by Kent Tamura.

        [HTML5] Revert display:none on datalist
        https://bugs.webkit.org/show_bug.cgi?id=52214

        * fast/forms/datalist-expected.txt:
        * fast/forms/datalist-nonoption-child-expected.txt:
        * fast/forms/datalist.html:
2011-01-17  Anthony Ricaud  <rik@webkit.org>

        Reviewed by Kent Tamura.

        [HTML5] Revert display:none on datalist
        https://bugs.webkit.org/show_bug.cgi?id=52214

        * css/html.css:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75940 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: simplify debugger enabling routine.
        https://bugs.webkit.org/show_bug.cgi?id=52472

        * inspector/Inspector.idl:
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::restoreDebugger):
        (WebCore::InspectorController::showAndEnableDebugger):
        (WebCore::InspectorController::enableDebugger):
        * inspector/InspectorController.h:
        * inspector/front-end/DebuggerModel.js:
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel.prototype.show):
        (WebInspector.ScriptsPanel.prototype._toggleDebugging):
2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: simplify debugger enabling routine.
        https://bugs.webkit.org/show_bug.cgi?id=52472

        * WebInspector/WebInspector.mm:
        (-[WebInspector startDebuggingJavaScript:]):
2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: simplify debugger enabling routine.
        https://bugs.webkit.org/show_bug.cgi?id=52472

        * WebInspector.cpp:
        (WebInspector::toggleDebuggingJavaScript):
2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: simplify debugger enabling routine.
        https://bugs.webkit.org/show_bug.cgi?id=52472

        * WebProcess/WebPage/WebInspector.cpp:
        (WebKit::WebInspector::startJavaScriptDebugging):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75939 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Chromium: updated DevTools tests.

        * src/js/Tests.js:
        (.TestSuite.prototype._waitForScriptPause):
        (.TestSuite.prototype._waitUntilScriptsAreParsed.waitForAllScripts):
        (.TestSuite.prototype._waitUntilScriptsAreParsed):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75938 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: refactoring: encapsulate lazy initialization of SourceFrame.
        https://bugs.webkit.org/show_bug.cgi?id=51738

        * http/tests/inspector/debugger-test2.js:
        (initialize_DebuggerTest.InspectorTest.showScriptSource):
        (initialize_DebuggerTest.InspectorTest._showScriptSource):
        * inspector/debugger-pause-in-eval-script-expected.txt:
        * inspector/debugger-pause-in-eval-script.html:
2011-01-17  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: refactoring: encapsulate lazy initialization of SourceFrame.
        https://bugs.webkit.org/show_bug.cgi?id=51738

        Extract content loading logic from SourceView and ScriptView to ContentProvider implementations.
        Pass ContentProvider in SourceFrame constructor to allow SourceFrame manage it's lazy initialization.

        * inspector/front-end/ScriptView.js:
        (WebInspector.ScriptView):
        (WebInspector.SourceFrameContentProviderForScript):
        (WebInspector.SourceFrameContentProviderForScript.prototype.requestContent.didRequestSource):
        (WebInspector.SourceFrameContentProviderForScript.prototype.requestContent):
        (WebInspector.SourceFrameContentProviderForScript.prototype.scripts):
        * inspector/front-end/ScriptsPanel.js:
        (WebInspector.ScriptsPanel.prototype._addScript):
        (WebInspector.ScriptsPanel.prototype.sourceFrameForScript):
        (WebInspector.ScriptsPanel.prototype._sourceFrameForResource):
        * inspector/front-end/SourceFrame.js:
        (WebInspector.SourceFrame):
        (WebInspector.SourceFrame.prototype.set visible):
        (WebInspector.SourceFrame.prototype._createTextViewer):
        (WebInspector.SourceFrame.prototype._breakpointAdded):
        (WebInspector.SourceFrame.prototype._breakpoints):
        (WebInspector.SourceFrame.prototype._sourceIDForLine):
        (WebInspector.SourceFrame.prototype._sourceIDSet):
        (WebInspector.SourceFrameContentProvider):
        (WebInspector.SourceFrameContentProvider.prototype.requestContent):
        (WebInspector.SourceFrameContentProvider.prototype.scripts):
        * inspector/front-end/SourceView.js:
        (WebInspector.SourceView):
        (WebInspector.SourceView.prototype.show):
        (WebInspector.SourceView.prototype.resize):
        (WebInspector.SourceView.prototype.performSearch.didFindSearchMatches):
        (WebInspector.SourceView.prototype.performSearch):
        (WebInspector.SourceView.prototype.revealLine):
        (WebInspector.SourceView.prototype.highlightLine):
        (WebInspector.SourceView.prototype._jumpToSearchResult):
        (WebInspector.SourceFrameContentProviderForResource):
        (WebInspector.SourceFrameContentProviderForResource.prototype.requestContent):
        (WebInspector.SourceFrameContentProviderForResource.prototype.scripts):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75937 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  John Knottenbelt  <jknotten@chromium.org>

        Unreviewed.

        Add myself to committers list
        https://bugs.webkit.org/show_bug.cgi?id=52566

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

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75936 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Balazs Kelemen  <kbalazs@webkit.org>

        Reviewed by Andreas Kling.

        [Qt][WK2] Crash due to double destruction of QSharedMemory
        https://bugs.webkit.org/show_bug.cgi?id=52569

        Avoid deleting the QSharedMemory twice.
        * Platform/qt/SharedMemoryQt.cpp:
        (WebKit::SharedMemory::~SharedMemory):
        * Shared/qt/CleanupHandler.cpp:
        Renamed m_inDeleteObjects to m_hasStartedDeleting and
        added a getter for it.
        (WebKit::CleanupHandler::CleanupHandler):
        (WebKit::CleanupHandler::deleteObjects):
        * Shared/qt/CleanupHandler.h:
        (WebKit::CleanupHandler::unmark):
        (WebKit::CleanupHandler::hasStartedDeleting):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75935 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  John Knottenbelt  <jknotten@chromium.org>

        Reviewed by Jeremy Orlow.

        GeolocationController should call stopUpdating on destruction
        https://bugs.webkit.org/show_bug.cgi?id=52216

        fast/dom/window-close-crash.html tests that we do not fail the
        assertion in the mock GeolocationClient that the GeolocationClient
        is not updating when the GeolocationController is destroyed.

        * fast/dom/Geolocation/resources/window-close-popup.html: Added.
        * fast/dom/Geolocation/script-tests/window-close-crash.js: Added.
        (gotPosition):
        (waitForWindowToClose):
        (failedToCreateWatch):
        * fast/dom/Geolocation/window-close-crash-expected.txt: Added.
        * fast/dom/Geolocation/window-close-crash.html: Added.
        * platform/gtk/Skipped:
        * platform/mac-wk2/Skipped:
        * platform/qt-wk2/Skipped:
2011-01-17  John Knottenbelt  <jknotten@chromium.org>

        Reviewed by Jeremy Orlow.

        GeolocationController should call stopUpdating on destruction
        https://bugs.webkit.org/show_bug.cgi?id=52216

        Test: fast/dom/Geolocation/window-close-crash.html

        * page/GeolocationController.cpp:
        (WebCore::GeolocationController::~GeolocationController):
2011-01-17  John Knottenbelt  <jknotten@chromium.org>

        Reviewed by Jeremy Orlow.

        GeolocationController should call stopUpdating on destruction
        https://bugs.webkit.org/show_bug.cgi?id=52216

        fast/dom/Geolocation/window-close-crash.html requires that a
        Geolocation watch be started in a secondary window. Consequently,
        we need to allow geolocation permission and provide a mock
        geolocation position for the secondary window's
        GeolocationClientMock.

        * DumpRenderTree/chromium/LayoutTestController.cpp:
        (LayoutTestController::setGeolocationPermission):
        (LayoutTestController::setMockGeolocationPosition):
        (LayoutTestController::setMockGeolocationError):
        * DumpRenderTree/chromium/TestShell.h:
        (TestShell::windowList):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75934 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Pavel Feldman  <pfeldman@chromium.org>

        Not reviewed: Qt build fix.

        * inspector/InspectorInstrumentation.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75933 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: make WebCore use InspectorInstrumentation
        for instrumentation calls.
        https://bugs.webkit.org/show_bug.cgi?id=52532

        This change makes WebCore classes issue instrumentation signals
        by means of InspectorInstrumentation interface. It covered migration
        for Document, FrameLoader, Database, DOMStorage, etc. It fixed
        instrumentation handling for Console as well. This all is a part
        of story described in the bug 52510.

        * dom/Document.cpp:
        (WebCore::Document::finishedParsing):
        * dom/Document.h:
        * dom/ScriptExecutionContext.h:
        * inspector/InspectorBrowserDebuggerAgent.cpp:
        (WebCore::InspectorBrowserDebuggerAgent::InspectorBrowserDebuggerAgent):
        (WebCore::InspectorBrowserDebuggerAgent::setDOMBreakpoint):
        (WebCore::InspectorBrowserDebuggerAgent::removeDOMBreakpoint):
        (WebCore::InspectorBrowserDebuggerAgent::willInsertDOMNode):
        (WebCore::InspectorBrowserDebuggerAgent::willRemoveDOMNode):
        (WebCore::InspectorBrowserDebuggerAgent::willModifyDOMAttr):
        (WebCore::InspectorBrowserDebuggerAgent::descriptionForDOMEvent):
        (WebCore::InspectorBrowserDebuggerAgent::pauseOnNativeEventIfNeeded):
        (WebCore::InspectorBrowserDebuggerAgent::willSendXMLHttpRequest):
        * inspector/InspectorBrowserDebuggerAgent.h:
        (WebCore::InspectorBrowserDebuggerAgent::create):
        * inspector/InspectorConsoleAgent.cpp:
        (WebCore::InspectorConsoleAgent::stopTiming):
        (WebCore::InspectorConsoleAgent::count):
        * inspector/InspectorConsoleAgent.h:
        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::InspectorController):
        (WebCore::InspectorController::~InspectorController):
        (WebCore::InspectorController::handleMousePress):
        (WebCore::InspectorController::didClearWindowObjectInWorld):
        (WebCore::PostWorkerNotificationToFrontendTask::performTask):
        * inspector/InspectorController.h:
        * inspector/InspectorDOMAgent.h:
        * inspector/InspectorDOMStorageAgent.h:
        * inspector/InspectorDatabaseAgent.cpp:
        * inspector/InspectorDatabaseAgent.h:
        * inspector/InspectorFrontendHost.cpp:
        * inspector/InspectorFrontendHost.h:
        * inspector/InspectorInstrumentation.cpp:
        (WebCore::InspectorInstrumentation::didClearWindowObjectInWorldImpl):
        (WebCore::InspectorInstrumentation::inspectedPageDestroyedImpl):
        (WebCore::InspectorInstrumentation::mouseDidMoveOverElementImpl):
        (WebCore::InspectorInstrumentation::handleMousePressImpl):
        (WebCore::InspectorInstrumentation::mainResourceFiredLoadEventImpl):
        (WebCore::InspectorInstrumentation::mainResourceFiredDOMContentEventImpl):
        (WebCore::InspectorInstrumentation::frameDetachedFromParentImpl):
        (WebCore::InspectorInstrumentation::didCommitLoadImpl):
        (WebCore::InspectorInstrumentation::addMessageToConsoleImpl):
        (WebCore::InspectorInstrumentation::consoleCountImpl):
        (WebCore::InspectorInstrumentation::startConsoleTimingImpl):
        (WebCore::InspectorInstrumentation::stopConsoleTimingImpl):
        (WebCore::InspectorInstrumentation::consoleMarkTimelineImpl):
        (WebCore::InspectorInstrumentation::addStartProfilingMessageToConsoleImpl):
        (WebCore::InspectorInstrumentation::didOpenDatabaseImpl):
        (WebCore::InspectorInstrumentation::didUseDOMStorageImpl):
        (WebCore::InspectorInstrumentation::didCreateWorkerImpl):
        (WebCore::InspectorInstrumentation::didDestroyWorkerImpl):
        (WebCore::InspectorInstrumentation::didCreateWebSocketImpl):
        (WebCore::InspectorInstrumentation::willSendWebSocketHandshakeRequestImpl):
        (WebCore::InspectorInstrumentation::didReceiveWebSocketHandshakeResponseImpl):
        (WebCore::InspectorInstrumentation::didCloseWebSocketImpl):
        (WebCore::InspectorInstrumentation::networkStateChangedImpl):
        (WebCore::InspectorInstrumentation::updateApplicationCacheStatusImpl):
        * inspector/InspectorInstrumentation.h:
        (WebCore::InspectorInstrumentation::inspectorControllerCreated):
        (WebCore::InspectorInstrumentation::inspectorControllerDeleted):
        (WebCore::InspectorInstrumentation::didClearWindowObjectInWorld):
        (WebCore::InspectorInstrumentation::inspectedPageDestroyed):
        (WebCore::InspectorInstrumentation::willInsertDOMNode):
        (WebCore::InspectorInstrumentation::didInsertDOMNode):
        (WebCore::InspectorInstrumentation::willRemoveDOMNode):
        (WebCore::InspectorInstrumentation::willModifyDOMAttr):
        (WebCore::InspectorInstrumentation::didModifyDOMAttr):
        (WebCore::InspectorInstrumentation::mouseDidMoveOverElement):
        (WebCore::InspectorInstrumentation::handleMousePress):
        (WebCore::InspectorInstrumentation::characterDataModified):
        (WebCore::InspectorInstrumentation::willSendXMLHttpRequest):
        (WebCore::InspectorInstrumentation::didScheduleResourceRequest):
        (WebCore::InspectorInstrumentation::didInstallTimer):
        (WebCore::InspectorInstrumentation::didRemoveTimer):
        (WebCore::InspectorInstrumentation::willCallFunction):
        (WebCore::InspectorInstrumentation::willChangeXHRReadyState):
        (WebCore::InspectorInstrumentation::willDispatchEvent):
        (WebCore::InspectorInstrumentation::willDispatchEventOnWindow):
        (WebCore::InspectorInstrumentation::willEvaluateScript):
        (WebCore::InspectorInstrumentation::willFireTimer):
        (WebCore::InspectorInstrumentation::willLayout):
        (WebCore::InspectorInstrumentation::willLoadXHR):
        (WebCore::InspectorInstrumentation::willPaint):
        (WebCore::InspectorInstrumentation::willRecalculateStyle):
        (WebCore::InspectorInstrumentation::identifierForInitialRequest):
        (WebCore::InspectorInstrumentation::willSendRequest):
        (WebCore::InspectorInstrumentation::markResourceAsCached):
        (WebCore::InspectorInstrumentation::didLoadResourceFromMemoryCache):
        (WebCore::InspectorInstrumentation::willReceiveResourceData):
        (WebCore::InspectorInstrumentation::willReceiveResourceResponse):
        (WebCore::InspectorInstrumentation::didReceiveContentLength):
        (WebCore::InspectorInstrumentation::didFinishLoading):
        (WebCore::InspectorInstrumentation::didFailLoading):
        (WebCore::InspectorInstrumentation::resourceRetrievedByXMLHttpRequest):
        (WebCore::InspectorInstrumentation::scriptImported):
        (WebCore::InspectorInstrumentation::mainResourceFiredLoadEvent):
        (WebCore::InspectorInstrumentation::mainResourceFiredDOMContentEvent):
        (WebCore::InspectorInstrumentation::frameDetachedFromParent):
        (WebCore::InspectorInstrumentation::didCommitLoad):
        (WebCore::InspectorInstrumentation::willWriteHTML):
        (WebCore::InspectorInstrumentation::didOpenDatabase):
        (WebCore::InspectorInstrumentation::didUseDOMStorage):
        (WebCore::InspectorInstrumentation::didCreateWorker):
        (WebCore::InspectorInstrumentation::didDestroyWorker):
        (WebCore::InspectorInstrumentation::didCreateWebSocket):
        (WebCore::InspectorInstrumentation::willSendWebSocketHandshakeRequest):
        (WebCore::InspectorInstrumentation::didReceiveWebSocketHandshakeResponse):
        (WebCore::InspectorInstrumentation::didCloseWebSocket):
        (WebCore::InspectorInstrumentation::networkStateChanged):
        (WebCore::InspectorInstrumentation::updateApplicationCacheStatus):
        (WebCore::InspectorInstrumentation::addMessageToConsole):
        (WebCore::InspectorInstrumentation::consoleCount):
        (WebCore::InspectorInstrumentation::startConsoleTiming):
        (WebCore::InspectorInstrumentation::stopConsoleTiming):
        (WebCore::InspectorInstrumentation::consoleMarkTimeline):
        (WebCore::InspectorInstrumentation::addStartProfilingMessageToConsole):
        (WebCore::InspectorInstrumentation::inspectorControllerForContext):
        (WebCore::InspectorInstrumentation::inspectorControllerForFrame):
        (WebCore::InspectorInstrumentation::inspectorControllerForPage):
        (WebCore::InspectorInstrumentation::inspectorControllerWithFrontendForContext):
        (WebCore::InspectorInstrumentation::inspectorControllerWithFrontendForDocument):
        (WebCore::InspectorInstrumentation::inspectorControllerWithFrontendForFrame):
        (WebCore::InspectorInstrumentation::inspectorControllerWithFrontendForPage):
        * inspector/InspectorState.cpp:
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::detachFromParent):
        (WebCore::FrameLoader::dispatchDidClearWindowObjectInWorld):
        (WebCore::FrameLoader::dispatchDidCommitLoad):
        * page/Chrome.cpp:
        (WebCore::Chrome::mouseDidMoveOverElement):
        * page/Console.cpp:
        (WebCore::Console::addMessage):
        (WebCore::Console::count):
        (WebCore::Console::markTimeline):
        (WebCore::Console::profile):
        (WebCore::Console::time):
        (WebCore::Console::timeEnd):
        (WebCore::Console::group):
        (WebCore::Console::groupCollapsed):
        (WebCore::Console::groupEnd):
        * page/DOMWindow.cpp:
        (WebCore::DOMWindow::sessionStorage):
        (WebCore::DOMWindow::localStorage):
        (WebCore::DOMWindow::dispatchLoadEvent):
        * page/EventHandler.cpp:
        (WebCore::EventHandler::handleMousePressEvent):
        * page/Page.cpp:
        (WebCore::Page::~Page):
        * storage/Database.cpp:
        (WebCore::Database::openDatabase):
        * workers/AbstractWorker.cpp:
        (WebCore::AbstractWorker::onDestroyWorker):
        * workers/SharedWorker.cpp:
        (WebCore::SharedWorker::create):
        * workers/Worker.cpp:
        (WebCore::Worker::create):
        * workers/WorkerMessagingProxy.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75932 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Adam Barth  <abarth@webkit.org>

        Update path to reflect new WebKit directory location.

        * tests/WebFrameTest.cpp:
        (WebKit::WebFrameTest::registerMockedURLLoad):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75931 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-14  Ilya Tikhonovsky  <loislo@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: Enable inspector's tests for chromium Linux.
        https://bugs.webkit.org/show_bug.cgi?id=52445

        * platform/chromium/inspector/timeline-network-resource-expected.txt:
        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75930 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Gyuyoung Kim  <gyuyoung.kim@samsung.com>

        Unreviewed, rolling out r75923.

        GTK guy rolls out the r75914.

        * platform/efl/ScrollViewEfl.cpp:
        (WebCore::ScrollView::platformInit):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75929 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, build fix for 2 GTK tests html file paths.

        * tests/testmimehandling.c:
        (main):
        * tests/testwebview.c:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75928 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Helder Correia  <helder@sencha.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] fast/canvas/canvas-fillPath-gradient-shadow.html does not pass
        https://bugs.webkit.org/show_bug.cgi?id=52556

        The shadow color opacity needs to be set on the shadow painter.
        Additionally, the gradient brush should be transformed.

        * platform/qt/Skipped: Remove test from skipped list.
2011-01-17  Helder Correia  <helder@sencha.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] fast/canvas/canvas-fillPath-gradient-shadow.html does not pass
        https://bugs.webkit.org/show_bug.cgi?id=52556

        The shadow color opacity needs to be set on the shadow painter.
        Additionally, the gradient brush should be transformed.

        * platform/graphics/qt/GraphicsContextQt.cpp:
        (WebCore::GraphicsContext::fillPath):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75927 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  No'am Rosenthal  <noam.rosenthal@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Background image rendering is slow
        https://bugs.webkit.org/show_bug.cgi?id=50527

        When tiling a scaled pixmap in Image::drawPattern, scale the tile
        first and only then draw it to the target. Do so only when drawing
        more than one tile.

        Tests in fast/backgrounds/size cover this.

        * platform/graphics/qt/ImageQt.cpp:
        (WebCore::Image::drawPattern):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75926 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Laszlo Gombos  <laszlo.1.gombos@nokia.com>

        Reviewed by Simon Hausmann.

        [Qt] [Symbian] Fix building NPAPI support
        https://bugs.webkit.org/show_bug.cgi?id=51981

        Make sure that npapi.h is always included outside of the extern "C"
        linkage declaration block.

        No new tests as there is no new functionality.

        * bridge/npruntime.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75925 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, rolling out r75914.
        http://trac.webkit.org/changeset/75914
        https://bugs.webkit.org/show_bug.cgi?id=49177

        multiple crashes on GTK

        * platform/gtk/Skipped:
2011-01-17  Philippe Normand  <pnormand@igalia.com>

        Unreviewed, rolling out r75914.
        http://trac.webkit.org/changeset/75914
        https://bugs.webkit.org/show_bug.cgi?id=49177

        multiple crashes on GTK

        * platform/ScrollView.cpp:
        (WebCore::ScrollView::removeChild):
        (WebCore::ScrollView::wheelEvent):
        * platform/gtk/MainFrameScrollbarGtk.cpp:
        (MainFrameScrollbarGtk::attachAdjustment):
        (MainFrameScrollbarGtk::gtkValueChanged):
        * platform/gtk/ScrollViewGtk.cpp:
        (WebCore::ScrollView::platformInit):
        (WebCore::ScrollView::platformAddChild):
        (WebCore::ScrollView::platformRemoveChild):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75924 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-17  Gyuyoung Kim  <gyuyoung.kim@samsung.com>

        Unreviewed build fix.

        Bug 49177's patch enabled platformInit() in super class(ScrollView).
        But, it didn't removed the function in ScrollViewEfl.cpp.

        * platform/efl/ScrollViewEfl.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75923 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Kent Tamura  <tkent@chromium.org>

        Unreviewed, Chromium-linux build fix.

        * gyp_webkit: Fix a relative path to the top directory.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75922 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Adam Barth  <abarth@webkit.org>

        Update more include paths to reflect WebKit move.

        * DumpRenderTree/chromium/EventSender.cpp:
        (EventSender::keyDown):
        (EventSender::needsShiftModifier):
        * DumpRenderTree/chromium/WebThemeEngineDRTMac.h:
        * DumpRenderTree/chromium/WebThemeEngineDRTMac.mm:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75918 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Kent Tamura  <tkent@chromium.org>

        Unreviewed, Chromium build fix.

        * Scripts/update-webkit-chromium: Webkit/chromium -> Source/WebKit/chromium

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75917 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<rdar://problem/8871903> REGRESSION (r75897): Crash with 0-by-0 iframe in scaled WebView

Reviewed by Sam Weinig.

Avoid use of FloatRect::enclosingBoundingBox(), which can stretch an empty
quad to a non-empty rect.

* rendering/RenderWidget.cpp:
(WebCore::RenderWidget::setWidget):
(WebCore::RenderWidget::updateWidgetPosition):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75916 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Adam Barth  <abarth@webkit.org>

        Roll Chromium DEPS to pick up the changes reflecting the new location
        of WebKit.

        * DEPS:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75915 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] fast/events/scroll-after-click-on-tab-index has been failing on the bots
        https://bugs.webkit.org/show_bug.cgi?id=49177

        * platform/gtk/Skipped: Unskip this test, as it's now passing.
2011-01-16  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] fast/events/scroll-after-click-on-tab-index has been failing on the bots
        https://bugs.webkit.org/show_bug.cgi?id=49177

        * platform/ScrollView.cpp: Move the platform guards around a little. GTK+
        now shares the implementation of platformAddChild and platformRemoveChild,
        but has its own implementation of removeChild, which knows how to inform
        main frame scrollbars that they no longer control the WebCore scrollbar.
        * platform/gtk/MainFrameScrollbarGtk.cpp:
        (MainFrameScrollbarGtk::attachAdjustment): If we are attaching the same adjustment
        that we already have, bail out early. Apply the value changed signal handler
        after configuring the adjustment. We don't want our reset of the adjustment to
        stomp on WebCore values.
        (MainFrameScrollbarGtk::gtkValueChanged): Do not adjust the value if the WebCore
        state already matches ours. This prevents some unnecessary recursion
        * platform/gtk/ScrollViewGtk.cpp:
        (WebCore::ScrollView::removeChild): Added, special cases main frame scrollbars
        which need their adjustments detached.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75914 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-13  Yuzo Fujishima  <yuzo@google.com>

        Reviewed by Antti Koivisto.

        Fix for Bug 52427 - Inconsistent use of m_cache in CachedResourceLoader
        https://bugs.webkit.org/show_bug.cgi?id=52427

        In constructor/destructor of CachedResourceLoader, m_cache has been
        used to call MemoryCache::addCachedResourceLoader/removeCachedResourceLoader
        while cache() is used everywhere else.

        Actually addCachedResourceLoader/removeCachedResourceLoader need not be called at all.
        Remove the call sites and make MemoryCache non-friend of CachedResourceLoader.

        No new tests because the behavior remains the same.

        * loader/cache/CachedResourceLoader.cpp:
        (WebCore::CachedResourceLoader::CachedResourceLoader):
        (WebCore::CachedResourceLoader::~CachedResourceLoader):
        * loader/cache/CachedResourceLoader.h:
        * loader/cache/MemoryCache.cpp:
        * loader/cache/MemoryCache.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75912 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Martin Robinson  <mrobinson@igalia.com>

        Build fix for GTK+ after source directory movement.

        * GNUmakefile.am: Correct include paths.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75911 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Adam Barth  <abarth@webkit.org>

        Teach svn-apply about WebKit move.

        * Scripts/VCSUtils.pm:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75910 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Move WebKit into Source

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75909 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move WebKit into Source
        https://bugs.webkit.org/show_bug.cgi?id=52530

        * WebKit.grd:
        * WebKit.gyp:
2011-01-16  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move WebKit into Source
        https://bugs.webkit.org/show_bug.cgi?id=52530

        * Api/DerivedSources.pro:
        * declarative/declarative.pro:
        * docs/docs.pri:
        * tests/tests.pri:
2011-01-16  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move WebKit into Source
        https://bugs.webkit.org/show_bug.cgi?id=52530

        * GNUmakefile.am:
2011-01-16  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move WebKit into Source
        https://bugs.webkit.org/show_bug.cgi?id=52530

        * GNUmakefile.am:
2011-01-16  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move WebKit into Source
        https://bugs.webkit.org/show_bug.cgi?id=52530

        * Makefile:
        * scripts/generate-webkitversion.pl:
2011-01-16  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move WebKit into Source
        https://bugs.webkit.org/show_bug.cgi?id=52530

        * WebKit.vcproj/Interfaces.vcproj:
        * WebKit.vcproj/WebKit.sln:
        * WebKit.vcproj/WebKit.vcproj:
        * WebKit.vcproj/WebKitGUID.vcproj:
2011-01-16  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move WebKit into Source
        https://bugs.webkit.org/show_bug.cgi?id=52530

        * Android.mk:
        * CMakeLists.txt:
        * DerivedSources.pro:
        * GNUmakefile.am:
        * Makefile:
        * WebKit.pri:
        * WebKit.pro:
        * configure.ac:
        * wscript:
2011-01-16  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move WebKit into Source
        https://bugs.webkit.org/show_bug.cgi?id=52530

        * DumpRenderTree/mac/InternalHeaders/WebKit/WebTypesInternal.h:
        * DumpRenderTree/qt/DumpRenderTreeQt.cpp:
        * DumpRenderTree/qt/DumpRenderTreeQt.h:
        * DumpRenderTree/qt/GCControllerQt.cpp:
        * DumpRenderTree/qt/LayoutTestControllerQt.cpp:
        * DumpRenderTree/qt/PlainTextControllerQt.cpp:
        * DumpRenderTree/qt/TextInputControllerQt.cpp:
        * DumpRenderTree/wscript:
        * QtTestBrowser/launcherwindow.h:
        * Scripts/build-webkit:
        * Scripts/webkitdirs.pm:
2011-01-16  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move WebKit into Source
        https://bugs.webkit.org/show_bug.cgi?id=52530

        * JavaScriptCore.gyp/JavaScriptCore.gyp:
2011-01-16  Adam Barth  <abarth@webkit.org>

        Rubber-stamped by Eric Seidel.

        Move WebKit into Source
        https://bugs.webkit.org/show_bug.cgi?id=52530

        * WebCore.gyp/WebCore.gyp:
        * WebCore.pri:
        * WebCore.pro:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75908 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Assertion failure (!inSetWindow) with in-process plug-in in plugins/destroy-on-setwindow.html
https://bugs.webkit.org/show_bug.cgi?id=52550

Reviewed by Simon Fraser.

* Plugins/WebNetscapePluginView.mm:
(-[WebNetscapePluginView setWindowIfNecessary]): Removed the assertion, making sure
that inSetWindow remains YES until we exit the top-level setWindowIfNecessary.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75907 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein.

        frame-removed-during-resize.html test crashes (shows up as image-map-2.html crash)
        https://bugs.webkit.org/show_bug.cgi?id=52549

        Fix regression from r75900; m_widget->setFrameRect() can run script that
        clears m_widget, so null-check it before calling setBoundsSize().

        Tested by fast/replaced/frame-removed-during-resize.html

        * rendering/RenderWidget.cpp:
        (WebCore::RenderWidget::setWidgetGeometry):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75906 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Simon Fraser  <simon.fraser@apple.com>

        Keep Leopard build happy.

        * platform/mac/WidgetMac.mm:
        (WebCore::Widget::setBoundsSize):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75900 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Oliver Hunt  <oliver@apple.com>

        Reviewed by Sam Weinig.

        [jsfunfuzz] Parser doesn't correctly validate for-loop syntax
        https://bugs.webkit.org/show_bug.cgi?id=52516

        Add test for the case where a syntax error immediately follows
        multiple declarations in a for-loop initialiser.

        * fast/js/parser-syntax-check-expected.txt:
        * fast/js/script-tests/parser-syntax-check.js:
2011-01-16  Oliver Hunt  <oliver@apple.com>

        Reviewed by Sam Weinig.

        [jsfunfuzz] Parser doesn't correctly validate for-loop syntax
        https://bugs.webkit.org/show_bug.cgi?id=52516

        Ensure that we always check for a semicolon after encountering
        multiple declarations in the initialiser portion of a for-loop.

        * parser/JSParser.cpp:
        (JSC::JSParser::parseForStatement):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75899 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein.

        Issues with iframes and plugins when the WebView is scaled.
        <rdar://problem/6213380>

        When _scaleWebView has been called on a WebView, iframes
        in WebKit1 render and hit-test incorrectly, and plug-ins don't scale up.
        This is caused by AppKit NSViews not playing nicely with the scale
        applied through style.

        Work around most of these issues by adjusting the bounds size
        of widgets to allow iframe contents to paint with the correct scale,
        and fix various places in the code where we relied on coordinate
        transforms via NSViews (which ignore CSS transforms).

        * WebCore.exp.in:
        * platform/Widget.cpp:
        (WebCore::Widget::setBoundsSize):
        * platform/Widget.h:
        * platform/mac/WidgetMac.mm:
        (WebCore::Widget::setBoundsSize):
        (WebCore::Widget::paint):
        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::shouldPropagateCompositingToEnclosingIFrame):
        * rendering/RenderWidget.cpp:
        (WebCore::RenderWidget::setWidgetGeometry):
        (WebCore::RenderWidget::setWidget):
        (WebCore::RenderWidget::updateWidgetPosition):
        * rendering/RenderWidget.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75897 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Oliver Hunt  <oliver@apple.com>

        Reviewed by Geoffrey Garen.

        Strict mode restrictions on arguments and eval usage aren't complete
        https://bugs.webkit.org/show_bug.cgi?id=52528

        Add a helper function to aid parse failure tests by testing both direct
        eval/global code, and nested code (so that we test the syntax checker
        mode as well)

        * fast/js/basic-strict-mode-expected.txt:
        * fast/js/script-tests/basic-strict-mode.js:
        (shouldBeSyntaxError):
2011-01-16  Oliver Hunt  <oliver@apple.com>

        Reviewed by Geoffrey Garen.

        Strict mode restrictions on arguments and eval usage aren't complete
        https://bugs.webkit.org/show_bug.cgi?id=52528

        Fix a few bugs in strict mode where we incorrect allow mutation of
        arguments and eval in the parser.

        Alas the "optimisation" used by the syntax checker for validating
        binary and unary expressions was too aggressive: we do actually need
        a stack for operations and operands although it needn't be as complete
        as that used for the full AST builder.

        Also disallow assignment to arguments in all cases as allowing arguments
        to be assignable is always an error in strict mode, regardless of context.

        * parser/ASTBuilder.h:
        (JSC::ASTBuilder::BinaryExprContext::BinaryExprContext):
        (JSC::ASTBuilder::UnaryExprContext::UnaryExprContext):
        * parser/JSParser.cpp:
        (JSC::JSParser::parseAssignmentExpression):
        (JSC::JSParser::parseBinaryExpression):
        (JSC::JSParser::parseUnaryExpression):
        * parser/SyntaxChecker.h:
        (JSC::SyntaxChecker::BinaryExprContext::BinaryExprContext):
        (JSC::SyntaxChecker::BinaryExprContext::~BinaryExprContext):
        (JSC::SyntaxChecker::UnaryExprContext::UnaryExprContext):
        (JSC::SyntaxChecker::UnaryExprContext::~UnaryExprContext):
        (JSC::SyntaxChecker::appendBinaryExpressionInfo):
        (JSC::SyntaxChecker::operatorStackPop):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75896 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Robert Hogan  <robert@webkit.org>

        Unreviewed, re-skip test after r75894.

        [Qt] plugins/keyboard-events.html fails after r72717

        Re-skip plugins/keyboard-events.html.
        Using XKeysymToKeycode(QX11Info::display(), ..) doesn't work on
        a headless build, such as the buildbot. So need to solve this a
        different way.

        https://bugs.webkit.org/show_bug.cgi?id=50050

        * platform/qt/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75895 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Robert Hogan  <robert@webkit.org>

        Reviewed by Andreas Kling.

        [Qt] plugins/keyboard-events.html fails after r72717
        https://bugs.webkit.org/show_bug.cgi?id=50050

        * platform/qt/Skipped: Unskip plugins/keyboard-events.html
2011-01-16  Robert Hogan  <robert@webkit.org>

        Reviewed by Andreas Kling.

        [Qt] plugins/keyboard-events.html fails after r72717
        https://bugs.webkit.org/show_bug.cgi?id=50050

        * plugins/qt/PluginViewQt.cpp:
        (WebCore::setXKeyEventSpecificFields): map event text to keycode

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75894 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Fix for <rdar://problem/8871204>

Reviewed by Kevin Decker.

Don't try to save elasticity state. Just rely on
NSScrollElasticityAutomatic to restore the correct state.
* WebCoreSupport/WebFrameLoaderClient.h:
* WebCoreSupport/WebFrameLoaderClient.mm:
(WebFrameLoaderClient::WebFrameLoaderClient):
(WebFrameLoaderClient::dispatchDidFirstLayout):
(WebFrameLoaderClient::provisionalLoadStarted):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75893 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-16  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Dan Bernstein.

        RenderView needs to take transforms on its layer into account
        https://bugs.webkit.org/show_bug.cgi?id=52536

        The RenderView's coordinate mapping methods failed to
        take into account a transform on the RenderView's layer.

        No tests because it's not possible to get a transform
        on the RenderView's layer through content.

        * rendering/RenderView.cpp:
        (WebCore::RenderView::mapLocalToContainer):
        (WebCore::RenderView::mapAbsoluteToLocalPoint):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75892 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Rolled back in r75886.
        https://bugs.webkit.org/show_bug.cgi?id=52527

        r75886 broke the GTK Linux bot because Linux was -- quite surprisingly --
        set up to use the constants for embedded devices.

        * jit/ExecutableAllocatorFixedVMPool.cpp:
        (JSC::maybeModifyVMPoolSize): Separated Linux constants from embedded
        constants.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75891 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r75886.
        http://trac.webkit.org/changeset/75886
        https://bugs.webkit.org/show_bug.cgi?id=52526

        "Broke GTK+ 64bit" (Requested by xan_ on #webkit).

        * jit/ExecutableAllocatorFixedVMPool.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Skip fast/dom/Range/range-clone-contents.html on Windows

Not reviewed.

Filed: <https://bugs.webkit.org/show_bug.cgi?id=52524>

* platform/win/Skipped: Skip test on Windows. The test passes
on all other bots.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<rdar://problem/8870429> Shrink the executable pool on embedded devices

Reviewed by Sam Weinig.

* jit/ExecutableAllocatorFixedVMPool.cpp: Dropped the pool size from 32MB
to 16MB.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r75708.
        http://trac.webkit.org/changeset/75708
        https://bugs.webkit.org/show_bug.cgi?id=52521

        Breaks Qt build if mobility is not installed. (Requested by
        benjaminp on #webkit).

        * WebCore.pro:
        * platform/graphics/qt/MediaPlayerPrivateQt.cpp:
        (WebCore::MediaPlayerPrivateQt::MediaPlayerPrivateQt):
        * platform/graphics/qt/MediaPlayerPrivateQt.h:
        (WebCore::MediaPlayerPrivateQt::supportsFullscreen):
2011-01-15  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r75708.
        http://trac.webkit.org/changeset/75708
        https://bugs.webkit.org/show_bug.cgi?id=52521

        Breaks Qt build if mobility is not installed. (Requested by
        benjaminp on #webkit).

        * Api/qwebkitplatformplugin.h:
        * WebCoreSupport/ChromeClientQt.cpp:
        (WebCore::ChromeClientQt::ChromeClientQt):
        (WebCore::ChromeClientQt::~ChromeClientQt):
        * WebCoreSupport/ChromeClientQt.h:
        * WebCoreSupport/FullScreenVideoQt.cpp: Removed.
        * WebCoreSupport/FullScreenVideoQt.h: Removed.
        * WebCoreSupport/QtPlatformPlugin.cpp:
        * WebCoreSupport/QtPlatformPlugin.h:
        * examples/platformplugin/WebPlugin.cpp:
        (WebPlugin::supportsExtension):
        (WebPlugin::createExtension):
        * examples/platformplugin/WebPlugin.h:
        * examples/platformplugin/platformplugin.pro:
        * examples/platformplugin/qwebkitplatformplugin.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75885 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Oliver Hunt  <oliver@apple.com>

        Reviewed by Maciej Stachowiak.

        Incorrect behavior changing attributes of an accessor
        https://bugs.webkit.org/show_bug.cgi?id=52515

        defineProperty doesn't correctly handle changing attributes of an accessor
        property.  This is because we don't pass the full descriptor to the
        putDescriptor helper function, which means we have insufficient information
        to do the right thing. Once that's passed the correct behavior is relatively
        simple to implement.

        * runtime/JSObject.cpp:
        (JSC::putDescriptor):
        (JSC::JSObject::defineOwnProperty):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75884 268f45cc-cd09-0410-ab3c-d52691b4dbfc

<webkit.org/b/52512> REGRESSION(r73818): range.cloneContents() ignores end offset

Reviewed by Adele Peterson.

WebCore:

The fix for Bug 50710 in r73799 introduced an off-by-one error
when copying nodes to a local NodeVector for processing.  A fix
was attempted for Bug 50854 in r73818, but instead of stopping
at the end offset, it iterates through all the sibling nodes
because the loop variable (i) is never incremented.  To clean
this up, revert back to the code in r73799 and fix the
off-by-one error.

Test: fast/dom/Range/range-clone-contents.html

* dom/Range.cpp:
(WebCore::Range::processContents): Fix the loop that copies
nodes to a local NodeVector by restoring the code from r73799
and fixing the off-by-one error.

LayoutTests:

* fast/dom/Range/range-clone-contents-expected.txt: Added.
* fast/dom/Range/range-clone-contents.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75882 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Adam Barth  <abarth@webkit.org>

        Teach svn-apply about WebKit2 move.

        * Scripts/VCSUtils.pm:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75881 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Move WebKit2 into Source
https://bugs.webkit.org/show_bug.cgi?id=52438

Rubber-stamped by Eric Seidel.

.:

Update reference to WebKit2.

* DerivedSources.pro:
* GNUmakefile.am:
* Makefile:
* Source/cmake/WebKitPackaging.cmake:
* WebKit.pro:
* configure.ac:

Source/WebCore:

* WebCore.pro:

Source/WebKit2:

* DerivedSources.pro:
* GNUmakefile.am:
* WebKit2.pro:
* WebProcess.pro:
* win/WebKit2.vcproj:
* win/WebKit2WebProcess.vcproj:

Tools:

* MiniBrowser/qt/MiniBrowser.pro:
* Scripts/build-webkit:
* Scripts/do-file-rename:
* Scripts/do-webcore-rename:
* Scripts/test-webkitpy:
* Scripts/webkitdirs.pm:
* Scripts/webkitpy/style/checker.py:
* WebKitTestRunner/DerivedSources.pro:
* WebKitTestRunner/InjectedBundle/qt/InjectedBundle.pro:
* WebKitTestRunner/qt/WebKitTestRunner.pro:

WebKit/win:

Update reference to WebKit2.

* WebKit.vcproj/WebKit.sln:

Websites/bugs.webkit.org:

Remove WebKit2 from the list of top-level source directories.

* PrettyPatch/PrettyPatch.rb:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75880 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Joone Hur  <joone.hur@collabora.co.uk>

        Reviewed by Martin Robinson.

        [GTK] Linux build with FileSystem API enabled fails
        https://bugs.webkit.org/show_bug.cgi?id=43878

        This patch allows WebKitGtk+ to build with FileSystem API option.

        No new tests because no new functionality.

        * GNUmakefile.am: Included AsyncFileSystem.h,cpp and Excluded duplicated JSFileException.h,cpp.
        * bindings/js/JSDirectoryEntryCustom.cpp: Included ExceptionCode.h.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75879 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Robert Hogan  <robert@webkit.org>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] plugins/no-mime-with-valid-extension.html fails

        Check extension of resource against installed
        plugins when determining mime-type.

        https://bugs.webkit.org/show_bug.cgi?id=52288
        https://bugs.webkit.org/show_bug.cgi?id=50657
        * platform/qt/Skipped:
2011-01-15  Robert Hogan  <robert@webkit.org>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] plugins/no-mime-with-valid-extension.html fails
        Check extension of resource against installed
        plugins when determining mime-type.

        https://bugs.webkit.org/show_bug.cgi?id=52288
        https://bugs.webkit.org/show_bug.cgi?id=50657

        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (WebCore::FrameLoaderClientQt::objectContentType):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75878 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Stephen White  <senorblanco@chromium.org>

        Unreviewed; updated chromium test expectations.

        * platform/chromium-gpu/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75877 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Stephen White  <senorblanco@chromium.org>

        Unreviewed; updated chromium test expectations.

        * platform/chromium-gpu/test_expectations.txt:
        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75876 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Jochen Eisinger  <jochen@chromium.org>

        Reviewed by David Levin.

        [chromium] remove obsolete createView method
        https://bugs.webkit.org/show_bug.cgi?id=52437

        * DumpRenderTree/chromium/WebViewHost.cpp:
        (WebViewHost::createView):
        * DumpRenderTree/chromium/WebViewHost.h:
2011-01-15  Jochen Eisinger  <jochen@chromium.org>

        Reviewed by David Levin.

        [chromium] remove obsolete createView method
        https://bugs.webkit.org/show_bug.cgi?id=52437

        * public/WebViewClient.h:
        (WebKit::WebViewClient::createView):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75875 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2011-01-15  Tony Chang  <tony@chromium.org>

        Unreviewed, add new test to GTK+ Skipped list.
        https://bugs.webkit.org/show_bug.cgi?id=52508

        * platform/gtk/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@75874 268f45cc-cd09-0410-ab3c-d52691b4dbfc