WebKit-https.git
9 years ago2011-02-01 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Wed, 2 Feb 2011 01:09:34 +0000 (01:09 +0000)]
2011-02-01  Anders Carlsson  <andersca@apple.com>

        Reviewed by Dan Bernstein.

        Add a layer backed NSView for accelerated content
        https://bugs.webkit.org/show_bug.cgi?id=53555

        Rename the current _layerHostingView to _oldLayerHostingView and
        add a new _layerHostingView to be used with the new drawing area.

        * UIProcess/API/mac/WKView.mm:
        (-[WKView hitTest:]):
        (-[WKView _startAcceleratedCompositing:]):
        (-[WKView _stopAcceleratedCompositing]):
        (-[WKView _enterAcceleratedCompositingMode:]):
        (-[WKView _exitAcceleratedCompositingMode]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77338 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Wed, 2 Feb 2011 01:04:53 +0000 (01:04 +0000)]
2011-02-01  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Eric Seidel.

        [GTK] GObject DOM bindings do no support the CallWith attribute
        https://bugs.webkit.org/show_bug.cgi?id=53331

        Disable building GObject DOM bindings for IndexedDB because we do not support
        the CallWith attribute at this time.

        * bindings/gobject/GNUmakefile.am: Disable building bindings for the IndexedDB API.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77337 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Darin Adler <darin@apple.com>
darin@apple.com [Wed, 2 Feb 2011 01:01:05 +0000 (01:01 +0000)]
2011-02-01  Darin Adler  <darin@apple.com>

        Reviewed by Brady Eidson.

        Fix a couple loose ends from the back/forward tree encode/decode work
        https://bugs.webkit.org/show_bug.cgi?id=53537

        * history/HistoryItem.cpp:
        (WebCore::HistoryItem::encodeBackForwardTreeNode): Remove extra copy of
        original URL string; no need to encode it twice.
        (WebCore::HistoryItem::decodeBackForwardTree): Ditto.
        * history/HistoryItem.h: Removed declaration for function that is no
        longer defined nor used.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77336 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Tony Chang <tony@chromium.org>
tony@chromium.org [Wed, 2 Feb 2011 00:55:40 +0000 (00:55 +0000)]
2011-02-01  Tony Chang  <tony@chromium.org>

        Reviewed by Kent Tamura.

        [chromium] disable arm uninitialized variable warnings
        https://bugs.webkit.org/show_bug.cgi?id=53553

        We just got another error:
        third_party/WebKit/Source/WebCore/css/CSSPrimitiveValue.cpp:123:error:
        'colorTransparent.unstatic.4879' may be used uninitialized in this
        function

        * WebCore.gyp/WebCore.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77335 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Wed, 2 Feb 2011 00:48:26 +0000 (00:48 +0000)]
2011-02-01  Anders Carlsson  <andersca@apple.com>

        Reviewed by Dan Bernstein.

        Pass enter/exit accelerated compositing mode to the WKView
        https://bugs.webkit.org/show_bug.cgi?id=53552

        * UIProcess/API/mac/PageClientImpl.h:
        * UIProcess/API/mac/PageClientImpl.mm:
        (WebKit::PageClientImpl::enterAcceleratedCompositingMode):
        (WebKit::PageClientImpl::exitAcceleratedCompositingMode):
        * UIProcess/API/mac/WKView.mm:
        (-[WKView _enterAcceleratedCompositingMode:]):
        (-[WKView _exitAcceleratedCompositingMode]):
        * UIProcess/API/mac/WKViewInternal.h:
        * UIProcess/DrawingAreaProxyImpl.cpp:
        (WebKit::DrawingAreaProxyImpl::enterAcceleratedCompositingMode):
        (WebKit::DrawingAreaProxyImpl::exitAcceleratedCompositingMode):
        * UIProcess/PageClient.h:
        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::enterAcceleratedCompositingMode):
        (WebKit::WebPageProxy::exitAcceleratedCompositingMode):
        * UIProcess/WebPageProxy.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77334 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 chris reiss <christopher.reiss@nokia.com>
commit-queue@webkit.org [Wed, 2 Feb 2011 00:22:52 +0000 (00:22 +0000)]
2011-02-01  chris reiss  <christopher.reiss@nokia.com>

        Reviewed by Adam Barth.

        Self-replicating code makes Safari hang and eventually crash
        https://bugs.webkit.org/show_bug.cgi?id=15123

        * fast/dom/Document/document-close-iframe-load-expected.txt: Added.
        * fast/dom/Document/document-close-iframe-load.html: Added.
        * fast/dom/Document/document-close-nested-iframe-load-expected.txt: Added.
        * fast/dom/Document/document-close-nested-iframe-load.html: Added.
        * fast/dom/Document/document-write-recursion-expected.txt: Added.
        * fast/dom/Document/document-write-recursion.html: Added.
2011-02-01  chris reiss  <christopher.reiss@nokia.com>

        Reviewed by Adam Barth.

        Self-replicating code makes Safari hang and eventually crash
        https://bugs.webkit.org/show_bug.cgi?id=15123

        Here we are replicating the Firefox safeguard against
        recursive document.write( ) 's.

        See  https://bug197052.bugzilla.mozilla.org/attachment.cgi?id=293907 in bug
        https://bugzilla.mozilla.org/show_bug.cgi?id=197052 .   Firefox does two things -
            a) imposes a recursion limit of 20 on document.write( ) and
            b) once that limit is passed, panics all the way the call stack (rather than just returning one level.)
        To see why this is necessary, consider the script :

        <script>
           var t = document.body.innerHTML;
           document.write(t);
        </script>

        This will create a tree both broad and deep as the script keeps appending itself to the text.   If
        we just return one level after the recursion limit is reached, we still allow millions of copies to
        duplicate (and execute).

        The recursion is fortunately depth-first, so as soon as we cross this limit, we panic up the callstack
        to prevent this situation.    (IE apparently does the same thing, with a lower recursion limit.)

        Test: fast/dom/Document/document-write-recursion.html
        Test: fast/dom/Document/document-close-iframe-load.html
        Test: fast/dom/Document/document-close-nested-iframe-load.html

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77333 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Wed, 2 Feb 2011 00:12:04 +0000 (00:12 +0000)]
2011-02-01  Dimitri Glazkov  <dglazkov@chromium.org>

        Updated text expectations to avoid conflict.

        * platform/chromium/test_expectations.txt: Tweaked.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77332 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Wed, 2 Feb 2011 00:05:19 +0000 (00:05 +0000)]
2011-02-01  Ojan Vafai  <ojan@chromium.org>

        Increment version number to cache-bust code-review.js.

        * PrettyPatch/PrettyPatch.rb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77331 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-31 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Wed, 2 Feb 2011 00:00:36 +0000 (00:00 +0000)]
2011-01-31  Ojan Vafai  <ojan@chromium.org>

        Reviewed by Adam Barth.

        Store draft comments in localStorage
        https://bugs.webkit.org/show_bug.cgi?id=52866

        * code-review.js:
        * code-review-test.html

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77330 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Johnny Ding <jnd@chromium.org>
jnd@chromium.org [Tue, 1 Feb 2011 23:52:27 +0000 (23:52 +0000)]
2011-02-01  Johnny Ding  <jnd@chromium.org>

        Reviewed by Darin Adler.

        Don't set user gesture in HTMLAnchorElement's click handler because the click handler can be triggered by untrusted event.
        https://bugs.webkit.org/show_bug.cgi?id=53424

        * fast/events/popup-blocked-from-untrusted-click-event-on-anchor-expected.txt: Added.
        * fast/events/popup-blocked-from-untrusted-click-event-on-anchor.html: Added.
2011-02-01  Johnny Ding  <jnd@chromium.org>

        Reviewed by Darin Adler.

        Don't set user gesture in HTMLAnchorElement's click handler because the click handler can be triggered by untrusted event.
        https://bugs.webkit.org/show_bug.cgi?id=53424

        Test: fast/events/popup-blocked-from-untrusted-click-event-on-anchor.html

        * html/HTMLAnchorElement.cpp:
        (WebCore::handleLinkClick):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77329 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed Qt buildfix after r77286.
ossy@webkit.org [Tue, 1 Feb 2011 23:50:58 +0000 (23:50 +0000)]
Unreviewed Qt buildfix after r77286.

https://bugs.webkit.org/show_bug.cgi?id=53520
Remove the physical terminology from IntRect and FloatRect.

Source/WebCore:

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

Source/WebKit/qt:

* WebCoreSupport/PopupMenuQt.cpp:
(WebCore::PopupMenuQt::show):

Source/WebKit2:

* UIProcess/qt/ChunkedUpdateDrawingAreaProxyQt.cpp:
(WebKit::ChunkedUpdateDrawingAreaProxy::drawUpdateChunkIntoBackingStore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77328 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Zhenyao Mo <zmo@google.com>
zmo@google.com [Tue, 1 Feb 2011 23:46:41 +0000 (23:46 +0000)]
2011-02-01  Zhenyao Mo  <zmo@google.com>

        Unreviewed, test expectations update.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77327 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Sheriff Bot <webkit.review.bot@gmail.com>
paroga@webkit.org [Tue, 1 Feb 2011 23:32:46 +0000 (23:32 +0000)]
2011-02-01  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r77297.
        http://trac.webkit.org/changeset/77297
        https://bugs.webkit.org/show_bug.cgi?id=53538

        caused leopard crashes (Requested by paroga on #webkit).

        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
        * wtf/text/AtomicString.cpp:
        (WTF::AtomicString::fromUTF8):
        * wtf/unicode/UTF8.cpp:
        (WTF::Unicode::calculateStringHashFromUTF8):
        * wtf/unicode/UTF8.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77324 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Tue, 1 Feb 2011 23:26:15 +0000 (23:26 +0000)]
2011-02-01  Dimitri Glazkov  <dglazkov@chromium.org>

        Updated baselines after switching input-derived parts of HTMLMediaElement
        to use shadow DOM.
        https://bugs.webkit.org/show_bug.cgi?id=53314

        * platform/chromium-linux/media/video-zoom-controls-expected.checksum: Updated.
        * platform/chromium-linux/media/video-zoom-controls-expected.png: Updated.
        * platform/chromium-mac/media/video-zoom-controls-expected.checksum: Updated.
        * platform/chromium-mac/media/video-zoom-controls-expected.png: Updated.
        * platform/chromium-mac/media/video-zoom-controls-expected.txt: Updated.
        * platform/chromium-win/media/video-zoom-controls-expected.checksum: Updated.
        * platform/chromium-win/media/video-zoom-controls-expected.png: Updated.
        * platform/chromium-win/media/video-zoom-controls-expected.txt: Updated.
        * platform/chromium/test_expectations.txt: Expectation of failure removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77323 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix Mac production builds.
weinig@apple.com [Tue, 1 Feb 2011 23:20:09 +0000 (23:20 +0000)]
Fix Mac production builds.

../JavaScriptCore:

* JavaScriptCore.xcodeproj/project.pbxproj:

../WebCore:

* DerivedSources.make:
* WebCore.xcodeproj/project.pbxproj:
* platform/mac/ScrollAnimatorMac.h:
* platform/mac/ScrollbarThemeMac.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77322 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Abhishek Arya <inferno@chromium.org>
inferno@chromium.org [Tue, 1 Feb 2011 23:05:34 +0000 (23:05 +0000)]
2011-02-01  Abhishek Arya  <inferno@chromium.org>

        Unreviewed.

        Test not crashing on chromium, was crashing earlier because of
        some build issues. Also, rebaselining chromium mac since it matches
        rest of chromium and is not same as mac leopard.
        https://bugs.webkit.org/show_bug.cgi?id=53516

        * platform/chromium-mac/fast/css/stylesheet-candidate-nodes-crash-expected.txt: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77321 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Darin Fisher <darin@chromium.org>
darin@chromium.org [Tue, 1 Feb 2011 23:04:18 +0000 (23:04 +0000)]
2011-02-01  Darin Fisher  <darin@chromium.org>

        Reviewed by David Levin.

        [chromium] Remove unnecessary WebCore:: prefixing in a few places and fix-up some indentation.
        https://bugs.webkit.org/show_bug.cgi?id=53523

        * src/WebDevToolsAgentImpl.cpp:
        (WebKit::WebDevToolsAgentImpl::attach):
        (WebKit::WebDevToolsAgentImpl::inspectNode):
        (WebKit::WebDevToolsAgentImpl::inspectorController):
        (WebKit::WebDevToolsAgentImpl::mainFrame):
        (WebKit::WebDevToolsAgentImpl::identifierForInitialRequest):
        (WebKit::WebDevToolsAgentImpl::sendMessageToFrontend):
        (WebKit::WebDevToolsAgentImpl::updateInspectorStateCookie):
        (WebKit::WebDevToolsAgent::interruptAndDispatch):
        (WebKit::WebDevToolsAgent::processPendingMessages):
        * src/WebPopupMenuImpl.cpp:
        (WebKit::WebPopupMenuImpl::resize):
        * src/WebScrollbarImpl.cpp:
        (WebKit::WebScrollbarImpl::setLocation):
        (WebKit::WebScrollbarImpl::setValue):
        (WebKit::WebScrollbarImpl::onMouseWheel):
        (WebKit::WebScrollbarImpl::onKeyDown):
        (WebKit::WebScrollbarImpl::scrollPosition):
        (WebKit::WebScrollbarImpl::setScrollOffset):
        (WebKit::WebScrollbarImpl::invalidateScrollbarRect):
        (WebKit::WebScrollbarImpl::getTickmarks):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77320 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix Qt bustage.
hyatt@apple.com [Tue, 1 Feb 2011 22:56:36 +0000 (22:56 +0000)]
Fix Qt bustage.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77317 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix Qt bustage.
hyatt@apple.com [Tue, 1 Feb 2011 22:52:02 +0000 (22:52 +0000)]
Fix Qt bustage.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77316 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Tue, 1 Feb 2011 22:51:45 +0000 (22:51 +0000)]
2011-02-01  Anders Carlsson  <andersca@apple.com>

        Reviewed by Adam Roben.

        Send messages when entering/exiting accelerated compositing
        https://bugs.webkit.org/show_bug.cgi?id=53534

        * Shared/LayerTreeContext.h: Added.
        * Shared/mac/LayerTreeContextMac.mm: Added.
        Add new LayerTreeContext which holds all information needed to enter
        accelerated compositing mode.

        * UIProcess/DrawingAreaProxy.h:
        (WebKit::DrawingAreaProxy::enterAcceleratedCompositingMode):
        (WebKit::DrawingAreaProxy::exitAcceleratedCompositingMode):
        Add empty stubs.

        * UIProcess/DrawingAreaProxy.messages.in:
        Add new messgaes.

        * UIProcess/DrawingAreaProxyImpl.cpp:
        (WebKit::DrawingAreaProxyImpl::enterAcceleratedCompositingMode):
        (WebKit::DrawingAreaProxyImpl::exitAcceleratedCompositingMode):
        * UIProcess/DrawingAreaProxyImpl.h:
        Add empty stubs.

        * WebKit2.xcodeproj/project.pbxproj:
        Add new files.

        * WebProcess/WebPage/mac/LayerTreeHostMac.h:
        * WebProcess/WebPage/mac/LayerTreeHostMac.mm:
        (WebKit::LayerTreeHostMac::LayerTreeHostMac):
        (WebKit::LayerTreeHostMac::invalidate):
        Send messages.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77315 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Darin Adler <darin@apple.com>
darin@apple.com [Tue, 1 Feb 2011 22:49:01 +0000 (22:49 +0000)]
2011-02-01  Darin Adler  <darin@apple.com>

        Reviewed by Chris Fleizach.

        REGRESSION: Removing focus from area element causes unwanted scrolling
        https://bugs.webkit.org/show_bug.cgi?id=50169

        Test: fast/images/imagemap-scroll.html

        * html/HTMLAreaElement.cpp:
        (WebCore::HTMLAreaElement::setFocus): Added override. Calls the new
        RenderImage::areaElementFocusChanged function.
        (WebCore::HTMLAreaElement::updateFocusAppearance): Removed the code
        here that calls setNeedsLayout on the image's renderer. This was an
        attempt to cause repaint of the renderer, but this function does not
        need to do that. Also changed this to use the imageElement function
        to avoid repeating code.

        * html/HTMLAreaElement.h: Updated for above changes.

        * rendering/RenderImage.cpp:
        (WebCore::RenderImage::paint): Updated for name change.
        (WebCore::RenderImage::paintAreaElementFocusRing): Renamed this from
        paintFocusRing, because it only paints area focus rings, and should
        not be confused with paintFocusRing functions in other classes. Also
        removed the unused style argument. Removed the code that used an
        HTMLCollection to see if the focused area element is for this image
        and instead just call imageElement on the area element.
        (WebCore::RenderImage::areaElementFocusChanged): Added. Calls repaint.

        * rendering/RenderImage.h: Added a public areaElementFocusChanged
        function for HTMLAreaElement to call. Made the paintFocusRing function
        private, renamed it to paintAreaElementFocusRing, and removed its
        unused style argument.
2011-02-01  Darin Adler  <darin@apple.com>

        Reviewed by Chris Fleizach.

        REGRESSION: Removing focus from area element causes unwanted scrolling
        https://bugs.webkit.org/show_bug.cgi?id=50169

        * fast/images/imagemap-scroll-expected.txt: Added.
        * fast/images/imagemap-scroll.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77313 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix Qt bustage.
hyatt@apple.com [Tue, 1 Feb 2011 22:48:39 +0000 (22:48 +0000)]
Fix Qt bustage.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77312 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Zhenyao Mo <zmo@google.com>
zmo@google.com [Tue, 1 Feb 2011 22:42:47 +0000 (22:42 +0000)]
2011-02-01  Zhenyao Mo  <zmo@google.com>

        Reviewed by Kenneth Russell.

        getUniform should support SAMPLER_2D or SAMPLER_CUBE
        https://bugs.webkit.org/show_bug.cgi?id=52190

        * fast/canvas/webgl/gl-object-get-calls-expected.txt:
        * fast/canvas/webgl/gl-object-get-calls.html: Test getUniform with sampler2D and samplerCube uniforms.
        * fast/canvas/webgl/resources/noopUniformShader.vert: Added.
        * fast/canvas/webgl/resources/samplerUniformShader.frag: Added.
2011-02-01  Patrick Gansterer  <paroga@webkit.org>

        Unreviewed WinCE build fix for r77286.

        * platform/graphics/wince/GraphicsContextWinCE.cpp:
        (WebCore::TransparentLayerDC::TransparentLayerDC):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77311 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Tue, 1 Feb 2011 22:39:30 +0000 (22:39 +0000)]
2011-02-01  Patrick Gansterer  <paroga@webkit.org>

        Unreviewed WinCE build fix for r77286.

        * platform/graphics/wince/GraphicsContextWinCE.cpp:
        (WebCore::TransparentLayerDC::TransparentLayerDC):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77309 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoAX: AXPosition of AXScrollArea is wrong
cfleizach@apple.com [Tue, 1 Feb 2011 22:37:29 +0000 (22:37 +0000)]
AX: AXPosition of AXScrollArea is wrong
https://bugs.webkit.org/show_bug.cgi?id=53511

Reviewed by Darin Adler.

Source/WebCore:

AccessibilityScrollView needed to return a valid documentFrameView() object.
At the same time, the code from document() should be consolidated in
AccessibilityObject, so all objects can use it.

Test: platform/mac/accessibility/webkit-scrollarea-position.html

* accessibility/AccessibilityObject.cpp:
(WebCore::AccessibilityObject::document):
* accessibility/AccessibilityObject.h:
* accessibility/AccessibilityScrollView.cpp:
(WebCore::AccessibilityScrollView::accessibilityHitTest):
(WebCore::AccessibilityScrollView::documentFrameView):
* accessibility/AccessibilityScrollView.h:

LayoutTests:

* platform/mac/accessibility/webkit-scrollarea-position-expected.txt: Added.
* platform/mac/accessibility/webkit-scrollarea-position.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77308 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago Reviewed by Dan Bernstein and Darin Adler.
ap@apple.com [Tue, 1 Feb 2011 22:30:11 +0000 (22:30 +0000)]
    Reviewed by Dan Bernstein and Darin Adler.

        <rdar://problem/8830594> Printed page is sometimes shifted down, making footer invisible

        * UIProcess/API/mac/WKPrintingView.mm: (-[WKPrintingView locationOfPrintRect:]): Instead of
        letting AppKit compute content position on page, make the view provide it. We can do this
        because WK(Printing)Views are always printed on their own, not as part of other views.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77306 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Zhenyao Mo <zmo@google.com>
zmo@google.com [Tue, 1 Feb 2011 22:29:33 +0000 (22:29 +0000)]
2011-02-01  Zhenyao Mo  <zmo@google.com>

        Reviewed by Darin Adler.

        Fix the incorrect usage of RetainPtr cases in GraphicsContext3DCG.cpp
        https://bugs.webkit.org/show_bug.cgi?id=53531

        With this fix, running WebGL conformance tests should no longer crash randomly.

        * platform/graphics/cg/GraphicsContext3DCG.cpp:
        (WebCore::GraphicsContext3D::getImageData):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77305 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Tue, 1 Feb 2011 22:14:53 +0000 (22:14 +0000)]
2011-02-01  Dimitri Glazkov  <dglazkov@chromium.org>

        One more Chromium build fix after r77286.

        * platform/chromium/ScrollbarThemeChromiumMac.mm:
        (WebCore::ScrollbarThemeChromiumMac::paint): Changed to not use topLeft().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77302 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoTry to fix the windows build.
weinig@apple.com [Tue, 1 Feb 2011 22:12:10 +0000 (22:12 +0000)]
Try to fix the windows build.

* DerivedSources.make:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77301 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix Qt bustage.
hyatt@apple.com [Tue, 1 Feb 2011 22:11:30 +0000 (22:11 +0000)]
Fix Qt bustage.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77300 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix the build for Beth.
weinig@apple.com [Tue, 1 Feb 2011 22:04:11 +0000 (22:04 +0000)]
Fix the build for Beth.

* platform/mac/ScrollAnimatorMac.mm:
(-[ScrollbarPainterControllerDelegate inLiveResizeForScrollerImpPair:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77298 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Tue, 1 Feb 2011 22:03:04 +0000 (22:03 +0000)]
2011-02-01  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Darin Adler.

        Avoid strlen() in AtomicString::fromUTF8
        https://bugs.webkit.org/show_bug.cgi?id=50516

        Add an overload to calculateStringHashFromUTF8 to get
        strlen() of the input data with only one call.

        This change shows about 3% performance win on the xml-parser benchmark.

        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
        * wtf/text/AtomicString.cpp:
        (WTF::AtomicString::fromUTF8):
        * wtf/unicode/UTF8.cpp:
        (WTF::Unicode::calculateStringHashAndLengthFromUTF8Internal):
        (WTF::Unicode::calculateStringHashFromUTF8):
        (WTF::Unicode::calculateStringHashAndLengthFromUTF8):
        * wtf/unicode/UTF8.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77297 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Tue, 1 Feb 2011 21:57:49 +0000 (21:57 +0000)]
2011-02-01  Anders Carlsson  <andersca@apple.com>

        Reviewed by Adam Roben.

        Make LayerTreeHost ref counted
        https://bugs.webkit.org/show_bug.cgi?id=53530

        LayerTreeHostMac will soon be able to trigger layout, which could cause
        the page to leave accelerated compositing mode and free the LayerTreeHostMac object.

        Making LayerTreeHost ref counted will let LayerTreeHostMac protect itself while triggering layout.

        * WebProcess/WebPage/DrawingAreaImpl.cpp:
        (WebKit::DrawingAreaImpl::setRootCompositingLayer):
        * WebProcess/WebPage/DrawingAreaImpl.h:
        * WebProcess/WebPage/LayerTreeHost.cpp:
        (WebKit::LayerTreeHost::create):
        * WebProcess/WebPage/LayerTreeHost.h:
        * WebProcess/WebPage/mac/LayerTreeHostMac.h:
        * WebProcess/WebPage/mac/LayerTreeHostMac.mm:
        (WebKit::LayerTreeHostMac::create):
        (WebKit::LayerTreeHostMac::LayerTreeHostMac):
        (WebKit::LayerTreeHostMac::~LayerTreeHostMac):
        (WebKit::LayerTreeHostMac::invalidate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77293 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix chromium bustage.
hyatt@apple.com [Tue, 1 Feb 2011 21:56:15 +0000 (21:56 +0000)]
Fix chromium bustage.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77292 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix Qt build bustage.
hyatt@apple.com [Tue, 1 Feb 2011 21:51:16 +0000 (21:51 +0000)]
Fix Qt build bustage.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77291 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Tue, 1 Feb 2011 21:47:57 +0000 (21:47 +0000)]
2011-02-01  Dimitri Glazkov  <dglazkov@chromium.org>

        Added yet another expectations of flakiness.
        https://bugs.webkit.org/show_bug.cgi?id=53527

        * platform/chromium/test_expectations.txt: Flakiness abound.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77288 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoPart 2 for <rdar://problem/8492788>
weinig@apple.com [Tue, 1 Feb 2011 21:45:38 +0000 (21:45 +0000)]
Part 2 for <rdar://problem/8492788>
Adopt WKScrollbarPainterController

Reviewed by Beth Dakin.

Use header detection to define scrollbar painting controller #define.

Source/JavaScriptCore:

* DerivedSources.make:
* JavaScriptCore.xcodeproj/project.pbxproj:

Source/WebCore:

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

Source/WebKit/mac:

* WebCoreSupport/WebSystemInterface.mm:
(InitWebCoreSystemInterface):

Source/WebKit2:

* WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:
(InitWebCoreSystemInterface):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77287 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=53520
hyatt@apple.com [Tue, 1 Feb 2011 21:39:47 +0000 (21:39 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=53520

Reviewed by Oliver Hunt.

Remove the physical terminology from IntRect and FloatRect.

Now that we have flipped RenderBlocks for vertical-rl and horizontal-bt writing modes,
we need to update our terminology to be more accurate.

I'm borrowing a page from AppKit here (which also supports flipped NSViews) and
renaming right() and bottom() to maxX() and maxY().  These terms remain accurate
even for flipped rectangles.

../WebCore:

* accessibility/AccessibilityRenderObject.cpp:
(WebCore::AccessibilityRenderObject::boundsForVisiblePositionRange):
* accessibility/mac/AccessibilityObjectWrapper.mm:
(-[AccessibilityObjectWrapper position]):
* dom/ClientRect.h:
(WebCore::ClientRect::right):
(WebCore::ClientRect::bottom):
* html/HTMLCanvasElement.cpp:
(WebCore::HTMLCanvasElement::convertLogicalToDevice):
* html/canvas/CanvasRenderingContext2D.cpp:
(WebCore::normalizeRect):
* inspector/InspectorAgent.cpp:
(WebCore::InspectorAgent::drawElementTitle):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::adjustWindowRect):
* page/DragController.cpp:
(WebCore::dragLocForSelectionDrag):
* page/EventHandler.cpp:
(WebCore::EventHandler::sendContextMenuEventForKey):
* page/PrintContext.cpp:
(WebCore::PrintContext::computePageRectsWithPageSizeInternal):
(WebCore::PrintContext::pageNumberForElement):
* page/SpatialNavigation.cpp:
(WebCore::end):
(WebCore::areRectsFullyAligned):
(WebCore::areRectsMoreThanFullScreenApart):
(WebCore::below):
(WebCore::rightOf):
(WebCore::isRectInDirection):
(WebCore::entryAndExitPointsForDirection):
(WebCore::virtualRectForDirection):
* page/WindowFeatures.cpp:
(WebCore::WindowFeatures::WindowFeatures):
* platform/ScrollView.cpp:
(WebCore::ScrollView::wheelEvent):
* platform/Scrollbar.cpp:
(WebCore::Scrollbar::setFrameRect):
* platform/ScrollbarThemeComposite.cpp:
(WebCore::ScrollbarThemeComposite::splitTrack):
* platform/chromium/ScrollbarThemeChromium.cpp:
(WebCore::ScrollbarThemeChromium::paintTickmarks):
* platform/graphics/FloatQuad.h:
(WebCore::FloatQuad::FloatQuad):
* platform/graphics/FloatRect.cpp:
(WebCore::FloatRect::intersects):
(WebCore::FloatRect::contains):
(WebCore::FloatRect::intersect):
(WebCore::FloatRect::unite):
(WebCore::enclosingIntRect):
* platform/graphics/FloatRect.h:
(WebCore::FloatRect::maxX):
(WebCore::FloatRect::maxY):
(WebCore::FloatRect::contains):
* platform/graphics/IntRect.cpp:
(WebCore::IntRect::intersects):
(WebCore::IntRect::contains):
(WebCore::IntRect::intersect):
(WebCore::IntRect::unite):
* platform/graphics/IntRect.h:
(WebCore::IntRect::maxX):
(WebCore::IntRect::maxY):
(WebCore::IntRect::shiftXEdgeTo):
(WebCore::IntRect::shiftMaxXEdgeTo):
(WebCore::IntRect::shiftYEdgeTo):
(WebCore::IntRect::shiftMaxYEdgeTo):
(WebCore::IntRect::contains):
* platform/graphics/WidthIterator.cpp:
(WebCore::WidthIterator::advance):
* platform/graphics/cg/GraphicsContextCG.cpp:
(WebCore::GraphicsContext::drawRect):
(WebCore::GraphicsContext::fillPath):
(WebCore::GraphicsContext::fillRect):
* platform/graphics/cg/ImageBufferCG.cpp:
(WebCore::getImageData):
(WebCore::putImageData):
* platform/graphics/cg/ImageCG.cpp:
(WebCore::BitmapImage::draw):
* platform/graphics/filters/FilterEffect.cpp:
(WebCore::FilterEffect::copyImageBytes):
* platform/graphics/mac/ComplexTextController.cpp:
(WebCore::ComplexTextController::adjustGlyphsAndAdvances):
* platform/graphics/mac/SimpleFontDataMac.mm:
(WebCore::SimpleFontData::platformBoundsForGlyph):
* platform/graphics/transforms/AffineTransform.cpp:
(WebCore::AffineTransform::mapRect):
* platform/graphics/win/FontCGWin.cpp:
(WebCore::drawGDIGlyphs):
* platform/graphics/win/MediaPlayerPrivateQuickTimeWin.cpp:
(WebCore::MediaPlayerPrivate::paint):
* platform/gtk/RenderThemeGtk.cpp:
(WebCore::centerRectVerticallyInParentInputElement):
* platform/mac/WidgetMac.mm:
(WebCore::Widget::paint):
* rendering/InlineFlowBox.cpp:
(WebCore::InlineFlowBox::addBoxShadowVisualOverflow):
(WebCore::InlineFlowBox::addTextBoxVisualOverflow):
* rendering/InlineTextBox.cpp:
(WebCore::InlineTextBox::selectionRect):
(WebCore::InlineTextBox::paint):
(WebCore::InlineTextBox::positionForOffset):
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::addOverflowFromChildren):
(WebCore::RenderBlock::paintChildren):
(WebCore::RenderBlock::paintEllipsisBoxes):
(WebCore::RenderBlock::inlineSelectionGaps):
(WebCore::RenderBlock::adjustPointToColumnContents):
(WebCore::RenderBlock::flipForWritingModeIncludingColumns):
(WebCore::RenderBlock::adjustForColumns):
* rendering/RenderBlock.h:
(WebCore::RenderBlock::FloatingObject::right):
(WebCore::RenderBlock::FloatingObject::bottom):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::reflectedRect):
(WebCore::RenderBox::localCaretRect):
(WebCore::RenderBox::addShadowOverflow):
(WebCore::RenderBox::addLayoutOverflow):
(WebCore::RenderBox::visualOverflowRectForPropagation):
(WebCore::RenderBox::layoutOverflowRectForPropagation):
(WebCore::RenderBox::flipForWritingMode):
* rendering/RenderFrameSet.cpp:
(WebCore::RenderFrameSet::paintColumnBorder):
(WebCore::RenderFrameSet::paintRowBorder):
* rendering/RenderInline.cpp:
(WebCore::RenderInline::paintOutlineForLine):
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::getRectToExpose):
(WebCore::cornerRect):
(WebCore::RenderLayer::positionOverflowControls):
(WebCore::RenderLayer::overflowBottom):
(WebCore::RenderLayer::overflowRight):
(WebCore::RenderLayer::paintResizer):
* rendering/RenderLineBoxList.cpp:
(WebCore::RenderLineBoxList::rangeIntersectsRect):
(WebCore::RenderLineBoxList::paint):
* rendering/RenderListItem.cpp:
(WebCore::RenderListItem::positionListMarker):
* rendering/RenderListMarker.cpp:
(WebCore::RenderListMarker::paint):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::repaintAfterLayoutIfNeeded):
* rendering/RenderOverflow.h:
(WebCore::RenderOverflow::RenderOverflow):
(WebCore::RenderOverflow::addLayoutOverflow):
(WebCore::RenderOverflow::addVisualOverflow):
(WebCore::RenderOverflow::setLayoutOverflow):
(WebCore::RenderOverflow::setVisualOverflow):
(WebCore::RenderOverflow::resetLayoutOverflow):
* rendering/RenderReplaced.cpp:
(WebCore::RenderReplaced::shouldPaint):
* rendering/RenderScrollbarTheme.cpp:
(WebCore::RenderScrollbarTheme::constrainTrackRectToTrackPieces):
* rendering/RenderTable.cpp:
(WebCore::RenderTable::paint):
* rendering/RenderTableCell.cpp:
(WebCore::RenderTableCell::paint):
* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::paintObject):
* rendering/RenderText.cpp:
(WebCore::RenderText::absoluteQuads):
* rendering/RenderTextControlSingleLine.cpp:
(WebCore::RenderTextControlSingleLine::forwardEvent):
* rendering/RenderThemeMac.mm:
(WebCore::RenderThemeMac::paintMenuListButtonGradients):
(WebCore::RenderThemeMac::paintMenuListButton):
(WebCore::RenderThemeMac::paintSliderTrack):
* rendering/RenderView.cpp:
(WebCore::RenderView::computeRectForRepaint):
(WebCore::RenderView::docBottom):
(WebCore::RenderView::docRight):
* rendering/RootInlineBox.cpp:
(WebCore::RootInlineBox::paddedLayoutOverflowRect):
* rendering/svg/RenderSVGInlineText.cpp:
(WebCore::RenderSVGInlineText::localCaretRect):

../WebKit/mac:

* WebView/WebHTMLView.mm:
(setNeedsDisplayInRect):

../WebKit2:

* Platform/Region.cpp:
(WebKit::Region::Shape::Shape):
* UIProcess/API/mac/WKPrintingView.mm:
(pageDidComputePageRects):
* UIProcess/FindIndicator.cpp:
(WebKit::FindIndicator::draw):
* UIProcess/TiledDrawingAreaProxy.cpp:
(WebKit::TiledDrawingAreaProxy::invalidate):
(WebKit::TiledDrawingAreaProxy::paint):
(WebKit::TiledDrawingAreaProxy::createTiles):
* UIProcess/mac/ChunkedUpdateDrawingAreaProxyMac.mm:
(WebKit::ChunkedUpdateDrawingAreaProxy::drawUpdateChunkIntoBackingStore):
* WebProcess/Plugins/Netscape/NetscapePlugin.cpp:
(WebKit::NetscapePlugin::callSetWindow):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77286 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Abhishek Arya <inferno@chromium.org>
inferno@chromium.org [Tue, 1 Feb 2011 21:38:41 +0000 (21:38 +0000)]
2011-02-01  Abhishek Arya  <inferno@chromium.org>

        Unreviewed.

        Rebaseline my test for QT, Leopard from r77262 due
        to minor comment error differences.

        * platform/mac-leopard/fast/css/stylesheet-candidate-nodes-crash-expected.txt: Added.
        * platform/qt/fast/css/stylesheet-candidate-nodes-crash-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77285 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Tue, 1 Feb 2011 21:34:41 +0000 (21:34 +0000)]
2011-02-01  Anders Carlsson  <andersca@apple.com>

        Reviewed by Adam Roben.

        Move flushPendingLayerChanges to LayerTreeHostMac
        https://bugs.webkit.org/show_bug.cgi?id=53525

        In preparation for adding more Mac specific code to flushPendingLayerChanges.

        * WebProcess/WebPage/LayerTreeHost.cpp:
        * WebProcess/WebPage/LayerTreeHost.h:
        * WebProcess/WebPage/mac/LayerTreeHostMac.h:
        * WebProcess/WebPage/mac/LayerTreeHostMac.mm:
        (WebKit::LayerTreeHostMac::flushPendingLayerChanges):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77282 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Xiaomei Ji <xji@chromium.org>
xji@chromium.org [Tue, 1 Feb 2011 21:28:36 +0000 (21:28 +0000)]
2011-02-01  Xiaomei Ji  <xji@chromium.org>

        Unreviewed.

        Rebaseline after r77267.

        * platform/chromium-linux/fast/dom/34176-expected.checksum: Added.
        * platform/chromium-linux/fast/dom/34176-expected.png: Added.
        * platform/chromium-mac/fast/dom/34176-expected.checksum: Added.
        * platform/chromium-mac/fast/dom/34176-expected.png: Added.
        * platform/chromium-win/fast/dom/34176-expected.checksum: Added.
        * platform/chromium-win/fast/dom/34176-expected.png: Added.
        * platform/chromium-win/fast/dom/34176-expected.txt: Added.
        * platform/win/fast/dom/34176-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77281 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Tue, 1 Feb 2011 21:15:11 +0000 (21:15 +0000)]
2011-02-01  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by James Robinson.

        [Chromium] run-chromium-webkit-unit-tests looks for binaries in the wrong place
        https://bugs.webkit.org/show_bug.cgi?id=53522

        * Scripts/run-chromium-webkit-unit-tests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77279 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Xiaomei Ji <xji@chromium.org>
xji@chromium.org [Tue, 1 Feb 2011 20:50:27 +0000 (20:50 +0000)]
2011-02-01  Xiaomei Ji  <xji@chromium.org>

        Unreviewed.

        * fast/dom/34176.html: Removed properties svn:executable and svn:mime-type.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77278 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Geoffrey Garen <ggaren@apple.com>
ggaren@apple.com [Tue, 1 Feb 2011 20:44:22 +0000 (20:44 +0000)]
2011-02-01  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        REGRESSION(77082): GC-related crashes seen: on WebKit2 bot; on GTK 32bit
        bot; loading trac pages; typing in search field
        https://bugs.webkit.org/show_bug.cgi?id=53519

        The crashes were all caused by failure to run an object's destructor.

        * runtime/CollectorHeapIterator.h:
        (JSC::ObjectIterator::ObjectIterator): Don't skip forward upon
        construction. The iterator class used to do that when it was designed
        for prior-to-beginning initialization. I forgot to remove this line
        of code when I changed the iterator to normal initialization.

        Skipping forward upon construction was causing the heap to skip running
        the destructor for the very first object in a block when destroying the
        block. This usually did not crash, since block destruction is rare and
        most objects have pretty trivial destructors. However, in the rare case
        when the heap would destroy a block whose first object was a global
        object or a DOM node, BOOM.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77277 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Tue, 1 Feb 2011 20:39:37 +0000 (20:39 +0000)]
2011-02-01  Dimitri Glazkov  <dglazkov@chromium.org>

        Remove expectations of failure for passing tests.

        * platform/chromium/test_expectations.txt: Fewer fails.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77276 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 1 Feb 2011 20:35:41 +0000 (20:35 +0000)]
2011-02-01  Adam Barth  <abarth@webkit.org>

        Reviewed by Mihai Parparita.

        Using Control-C to cancel webkit-patch or other python tools causes
        unpleasant traceback console spew
        https://bugs.webkit.org/show_bug.cgi?id=53462

        Catch the exception and don't print the stack trace.

        * Scripts/webkit-patch:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77275 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix for <rdar://problem/8492788> Adopt WKScrollbarPainterController
bdakin@apple.com [Tue, 1 Feb 2011 20:35:25 +0000 (20:35 +0000)]
Fix for <rdar://problem/8492788> Adopt WKScrollbarPainterController

Reviewed by Sam Weinig.

Source/WebCore:

Lots of new WebCoreSystemInterface functions to export.
* WebCore.exp.in:
* platform/mac/WebCoreSystemInterface.h:
* platform/mac/WebCoreSystemInterface.mm:

Let the scrollAnimator know when the mouse has
moved anywhere inside the page, and when the mouse
has moved in or out of the window.
* page/EventHandler.cpp:
(WebCore::EventHandler::mouseMoved):
(WebCore::EventHandler::updateMouseEventTargetNode):

Let the scrollAnimator know when the window has become
active or inactive.
* page/FocusController.cpp:
(WebCore::FocusController::setActive):

Let the scrollAnimator know when all of these things
are happening.
* page/FrameView.cpp:
(WebCore::FrameView::setContentsSize):
(WebCore::FrameView::didMoveOnscreen):
(WebCore::FrameView::willMoveOffscreen):
(WebCore::FrameView::currentMousePosition):
(WebCore::FrameView::contentsResized):

New functions called through WebKit2 that allow the
scrollAnimator to know when a live resize starts and ends.
(WebCore::FrameView::willStartLiveResize):
(WebCore::FrameView::willEndLiveResize):
* page/FrameView.h:

New functions on ScrollAnimator that pass information
to the WKPainterController when we're using one.
* platform/ScrollAnimator.h:
(WebCore::ScrollAnimator::scrollableArea):
(WebCore::ScrollAnimator::contentAreaWillPaint):
(WebCore::ScrollAnimator::mouseEnteredContentArea):
(WebCore::ScrollAnimator::mouseExitedContentArea):
(WebCore::ScrollAnimator::mouseMovedInContentArea):
(WebCore::ScrollAnimator::willStartLiveResize):
(WebCore::ScrollAnimator::contentsResized):
(WebCore::ScrollAnimator::willEndLiveResize):
(WebCore::ScrollAnimator::contentAreaDidShow):
(WebCore::ScrollAnimator::contentAreaDidHide):
(WebCore::ScrollAnimatorMac::ScrollAnimatorMac):
(WebCore::ScrollAnimatorMac::scrollbarPainterDelegate):
(WebCore::ScrollAnimatorMac::setPainterForPainterController):
(WebCore::ScrollAnimatorMac::removePainterFromPainterController):
(WebCore::ScrollAnimatorMac::notityPositionChanged):
(WebCore::ScrollAnimatorMac::contentAreaWillPaint):
(WebCore::ScrollAnimatorMac::mouseEnteredContentArea):
(WebCore::ScrollAnimatorMac::mouseExitedContentArea):
(WebCore::ScrollAnimatorMac::mouseMovedInContentArea):
(WebCore::ScrollAnimatorMac::willStartLiveResize):
(WebCore::ScrollAnimatorMac::contentsResized):
(WebCore::ScrollAnimatorMac::willEndLiveResize):
(WebCore::ScrollAnimatorMac::contentAreaDidShow):
(WebCore::ScrollAnimatorMac::contentAreaDidHide):

Let the scrollAnimator know when this is happening.
* platform/ScrollView.cpp:
(WebCore::ScrollView::paint):

New function lets the scrollAnimator get the current
mouse position.
* platform/ScrollView.h:
(WebCore::ScrollView::currentMousePosition):

New function that returns the scrollAnimator when needed.
* platform/ScrollableArea.h:
(WebCore::ScrollableArea::scrollAnimator):

Keep track of if we're in a live resize using a new memeber
variable.
* platform/mac/ScrollAnimatorMac.h:
(WebCore::ScrollAnimatorMac::inLiveResize):
* platform/mac/ScrollAnimatorMac.mm:
(WebCore::view):

New delegates for the WKPainter and WKPainterController
(-[ScrollbarPainterControllerDelegate initWithScrollAnimator:WebCore::]):
(-[ScrollbarPainterControllerDelegate contentAreaRectForScrollerImpPair:]):
(-[ScrollbarPainterControllerDelegate inLiveResizeForScrollerImpPair:]):
(-[ScrollbarPainterControllerDelegate mouseLocationInContentAreaForScrollerImpPair:]):
(-[ScrollbarPainterControllerDelegate scrollerImpPair:convertContentPoint:toScrollerImp:]):
(-[ScrollbarPainterControllerDelegate scrollerImpPair:setContentAreaNeedsDisplayInRect:]):
(-[ScrollbarPainterControllerDelegate scrollerImpPair:updateScrollerStyleForNewRecommendedScrollerStyle:]):
(-[ScrollKnobAnimation initWithScrollbarPainter:forScrollAnimator:WebCore::animateKnobAlphaTo:duration:]):
(-[ScrollKnobAnimation setCurrentProgress:]):
(-[ScrollbarPainterDelegate initWithScrollAnimator:WebCore::]):
(-[ScrollbarPainterDelegate convertRectToBacking:]):
(-[ScrollbarPainterDelegate convertRectFromBacking:]):
(-[ScrollbarPainterDelegate layer]):
(-[ScrollbarPainterDelegate setUpAnimation:scrollerPainter:animateKnobAlphaTo:duration:]):
(-[ScrollbarPainterDelegate scrollerImp:animateKnobAlphaTo:duration:]):
(-[ScrollbarPainterDelegate scrollerImp:animateTrackAlphaTo:duration:]):
(-[ScrollbarPainterDelegate scrollerImp:overlayScrollerStateChangedTo:]):

Get the WKScrollbarPainterRefs to synch up with the
WKScrollbarPainterControllerRefs when appropriate
* platform/mac/ScrollbarThemeMac.h:
* platform/mac/ScrollbarThemeMac.mm:
(WebCore::ScrollbarThemeMac::registerScrollbar):
(WebCore::ScrollbarThemeMac::unregisterScrollbar):
(WebCore::ScrollbarThemeMac::setNewPainterForScrollbar):
(WebCore::ScrollbarThemeMac::usesOverlayScrollbars):

Implement ScrollableArea's virtual function contentsSize() for access
through the scrollAnimator.
* rendering/RenderLayer.h:
(WebCore::RenderLayer::contentsSize):

Source/WebKit/mac:

* WebCoreSupport/WebSystemInterface.mm:
(InitWebCoreSystemInterface):

Source/WebKit2:

Implement NSView's viewWillStartLiveResize and
viewDidEndLiveResize and propagate the messages down
to WebCore.
* UIProcess/API/mac/WKView.mm:
(-[WKView viewWillStartLiveResize]):
(-[WKView viewDidEndLiveResize]):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::viewWillStartLiveResize):
(WebKit::WebPageProxy::viewWillEndLiveResize):
* UIProcess/WebPageProxy.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::viewWillStartLiveResize):
(WebKit::WebPage::viewWillEndLiveResize):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:

A bunch of new WebKitSystemInterface functions.
* WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:
(InitWebCoreSystemInterface):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77274 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Carol Szabo <carol.szabo@nokia.com>
carol.szabo@nokia.com [Tue, 1 Feb 2011 20:28:08 +0000 (20:28 +0000)]
2011-02-01  Carol Szabo  <carol.szabo@nokia.com>

        Reviewed by David Hyatt.

        layoutTestController.counterValueForElementById does not return the correct value
        https://bugs.webkit.org/show_bug.cgi?id=53037

        * fast/css/counters/deep-before-expected.txt: Added.
        * fast/css/counters/deep-before.html: Added.
        This tests whether DumpRenderTree properly retrieves counters and
        whether counters are properly set when the before pseudo element's
        renderer is not a direct child of the generating element's renderer.
2011-02-01  Carol Szabo  <carol.szabo@nokia.com>

        Reviewed by David Hyatt.

        layoutTestController.counterValueForElementById does not return the correct value
        https://bugs.webkit.org/show_bug.cgi?id=53037

        Test: fast/css/counters/deep-before.html

        * rendering/RenderTreeAsText.cpp:
        (WebCore::counterValueForElement):
        Modified to use the newly available RenderObject::beforePseudoElement()
        and RenderObject::afterPseudoElement() instead of the old imperfect
        algorithm to find the before and after pseudo elements.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77273 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Anton Muhin <antonm@chromium.org>
antonm@chromium.org [Tue, 1 Feb 2011 20:27:01 +0000 (20:27 +0000)]
2011-02-01  Anton Muhin  <antonm@chromium.org>

        Reviewed by Adam Barth.

        Allow access for security origin same as this.
        https://bugs.webkit.org/show_bug.cgi?id=53440

        Hard to test as newly added path currently is never hit.

        * page/SecurityOrigin.cpp:
        (WebCore::SecurityOrigin::canAccess): allow access if this == other

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77272 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Tue, 1 Feb 2011 20:25:05 +0000 (20:25 +0000)]
2011-02-01  Dimitri Glazkov  <dglazkov@chromium.org>

        Added more flakiness expectations from bots.

        * platform/chromium/test_expectations.txt: Another one bites or doesn't bite the dust.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77271 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-31 Oliver Hunt <oliver@apple.com>
oliver@apple.com [Tue, 1 Feb 2011 20:17:21 +0000 (20:17 +0000)]
2011-01-31  Oliver Hunt  <oliver@apple.com>

        Reviewed by Geoffrey Garen.

        Update JSObject storage for new marking API
        https://bugs.webkit.org/show_bug.cgi?id=53467

        JSObject no longer uses EncodedJSValue for its property storage.
        This produces a stream of mechanical changes to PropertySlot and
        anonymous storage APIs.

        * JavaScriptCore.exp:
        * runtime/ArrayPrototype.cpp:
        (JSC::ArrayPrototype::ArrayPrototype):
        * runtime/BooleanConstructor.cpp:
        (JSC::constructBoolean):
        (JSC::constructBooleanFromImmediateBoolean):
        * runtime/BooleanObject.cpp:
        (JSC::BooleanObject::BooleanObject):
        * runtime/BooleanObject.h:
        * runtime/BooleanPrototype.cpp:
        (JSC::BooleanPrototype::BooleanPrototype):
        * runtime/DateInstance.cpp:
        (JSC::DateInstance::DateInstance):
        * runtime/DatePrototype.cpp:
        (JSC::DatePrototype::DatePrototype):
        * runtime/JSActivation.cpp:
        (JSC::JSActivation::getOwnPropertySlot):
        * runtime/JSArray.cpp:
        (JSC::JSArray::getOwnPropertySlot):
        * runtime/JSFunction.cpp:
        (JSC::JSFunction::getOwnPropertySlot):
        * runtime/JSGlobalObject.h:
        (JSC::JSGlobalObject::JSGlobalObject):
        * runtime/JSObject.cpp:
        (JSC::JSObject::fillGetterPropertySlot):
        * runtime/JSObject.h:
        (JSC::JSObject::getDirectLocation):
        (JSC::JSObject::offsetForLocation):
        (JSC::JSObject::putAnonymousValue):
        (JSC::JSObject::clearAnonymousValue):
        (JSC::JSObject::getAnonymousValue):
        (JSC::JSObject::putThisToAnonymousValue):
        (JSC::JSObject::locationForOffset):
        (JSC::JSObject::inlineGetOwnPropertySlot):
        * runtime/JSObjectWithGlobalObject.cpp:
        (JSC::JSObjectWithGlobalObject::JSObjectWithGlobalObject):
        * runtime/JSWrapperObject.h:
        (JSC::JSWrapperObject::JSWrapperObject):
        (JSC::JSWrapperObject::setInternalValue):
        * runtime/Lookup.cpp:
        (JSC::setUpStaticFunctionSlot):
        * runtime/NumberConstructor.cpp:
        (JSC::constructWithNumberConstructor):
        * runtime/NumberObject.cpp:
        (JSC::NumberObject::NumberObject):
        (JSC::constructNumber):
        * runtime/NumberObject.h:
        * runtime/NumberPrototype.cpp:
        (JSC::NumberPrototype::NumberPrototype):
        * runtime/PropertySlot.h:
        (JSC::PropertySlot::getValue):
        (JSC::PropertySlot::setValue):
        (JSC::PropertySlot::setRegisterSlot):
        * runtime/StringObject.cpp:
        (JSC::StringObject::StringObject):
        * runtime/StringPrototype.cpp:
        (JSC::StringPrototype::StringPrototype):
        * runtime/WriteBarrier.h:
        (JSC::WriteBarrierBase::setWithoutWriteBarrier):
2011-01-31  Oliver Hunt  <oliver@apple.com>

        Reviewed by Geoffrey Garen.

        Update JSObject storage for new marking API
        https://bugs.webkit.org/show_bug.cgi?id=53467

        Update WebCore to handle new anonymous slot behaviour.

        * bindings/js/JSDOMWindowShell.cpp:
        (WebCore::JSDOMWindowShell::setWindow):
        * bindings/js/WorkerScriptController.cpp:
        (WebCore::WorkerScriptController::initScript):
        * bindings/scripts/CodeGeneratorJS.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77269 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Tue, 1 Feb 2011 20:05:07 +0000 (20:05 +0000)]
2011-02-01  Dimitri Glazkov  <dglazkov@chromium.org>

        https://bugs.webkit.org/show_bug.cgi?id=53516
        Test introduced in r77262 is crashing.

        * platform/chromium/test_expectations.txt: Added expectation to crash
            while the fix is worked on.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77268 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Xiaomei Ji <xji@chromium.org>
xji@chromium.org [Tue, 1 Feb 2011 20:02:06 +0000 (20:02 +0000)]
2011-02-01  Xiaomei Ji  <xji@chromium.org>

        Reviewed by David Hyatt.

        Fix a text rendering problem when enclosing block is RTL and text runs
        are in different directionality.
        https://bugs.webkit.org/show_bug.cgi?id=34176

        The problem happens in the following example scenario (ABC represents
        Hebrew characters):
        <div dir=rtl>this is a <span><span>test <span>ABC</span></span></span></div>

        The line consists of 3 text runs -- TextRun1 TextRun2 TextRun3. In which
        TextRun1 and TextRun2's bidi level are 2, and TextRun3's bidi level is 1.
        TextRun2 and TextRun3's least common ancestor is not a sibling of TextRun1.

        The visual bidi run order of the text runs is TextRun3 TextRun1 TextRun2.

        Inside RenderBlock::constructLine(), when RenderBlock::createLineBoxes()
        creates InlineFlowBox for TextRun2, it should check an InlineFlowBox for
        the run's render object's ancestor (not only its parent) has already
        been constructed or has something following it on the line, in which
        case, create a new box for TextRun2 instead of sharing the same box with
        TextRun3.

        In other words, the following 2 div should render the same results
        (ABC represents Hebrew characters).
        <div dir=rtl>this is a <span><span>test <span>ABC</span></span></span></div>
        <div dir=rtl>this is a <span>Test <span>ABC</span></span></div>

        Test: fast/dom/34176.html

        * rendering/RenderBlockLineLayout.cpp:
        (WebCore::parentIsConstructedOrHaveNext):
        (WebCore::RenderBlock::createLineBoxes):
2011-02-01  Xiaomei Ji  <xji@chromium.org>

        Reviewed by David Hyatt.

        Fix a text rendering problem when enclosing block is RTL and text runs
        are in different directionality.
        https://bugs.webkit.org/show_bug.cgi?id=34176

        The problem happens in the following example scenario (ABC represents
        Hebrew characters):
        <div dir=rtl>this is a <span><span>test <span>ABC</span></span></span></div>

        The line consists of 3 text runs -- TextRun1 TextRun2 TextRun3. In which
        TextRun1 and TextRun2's bidi level are 2, and TextRun3's bidi level is 1.
        TextRun2 and TextRun3's least common ancestor is not a sibling of TextRun1.

        The visual bidi run order of the text runs is TextRun3 TextRun1 TextRun2.

        Inside RenderBlock::constructLine(), when RenderBlock::createLineBoxes()
        creates InlineFlowBox for TextRun2, it should check an InlineFlowBox for
        the run's render object's ancestor (not only its parent) has already
        been constructed or has something following it on the line, in which
        case, create a new box for TextRun2 instead of sharing the same box with
        TextRun3.

        In other words, the following 2 div should render the same results
        (ABC represents Hebrew characters).
        <div dir=rtl>this is a <span><span>test <span>ABC</span></span></span></div>
        <div dir=rtl>this is a <span>Test <span>ABC</span></span></div>

        * fast/dom/34176.html: Added.
        * platform/mac/fast/dom/34176-expected.checksum: Added.
        * platform/mac/fast/dom/34176-expected.png: Added.
        * platform/mac/fast/dom/34176-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77267 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUpdate printing layout test for the ceiling->floor change.
hyatt@apple.com [Tue, 1 Feb 2011 19:39:52 +0000 (19:39 +0000)]
Update printing layout test for the ceiling->floor change.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77266 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Scott Cameron <sccameron@rim.com>
dbates@webkit.org [Tue, 1 Feb 2011 19:22:06 +0000 (19:22 +0000)]
2011-02-01  Scott Cameron  <sccameron@rim.com>

        Reviewed by Daniel Bates.

        Use Windows format for MinGW HTTPD path.
        https://bugs.webkit.org/show_bug.cgi?id=53503

        * Scripts/webkitperl/httpd.pm:
          - Use single-quotes around MySys value for $httpdPath in getHTTPDPath()
            so that we don't have to escape the space characters in the path. Also,
            changed path to Windows-style path and removed FIXME comment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77264 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Tue, 1 Feb 2011 19:21:16 +0000 (19:21 +0000)]
2011-02-01  Mihai Parparita  <mihaip@chromium.org>

        Unreviewed. Include actual values in fast/canvas/webgl/gl-teximage.html
        (removed by r77129) in the case of failure.

        * fast/canvas/webgl/gl-teximage.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77263 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-31 Abhishek Arya <inferno@chromium.org>
inferno@chromium.org [Tue, 1 Feb 2011 19:16:35 +0000 (19:16 +0000)]
2011-01-31  Abhishek Arya  <inferno@chromium.org>

        Reviewed by Dan Bernstein.

        Tests that a removed svg style element is not accessed again in document's
        style recalc code and does not result in crash.
        https://bugs.webkit.org/show_bug.cgi?id=53441

        * fast/css/stylesheet-candidate-nodes-crash-expected.txt: Added.
        * fast/css/stylesheet-candidate-nodes-crash.xhtml: Added.
2011-02-01  Abhishek Arya  <inferno@chromium.org>

        Reviewed by Dan Bernstein.

        Do not add a node in the document's stylesheet candidate node list if the
        node is already removed from document.
        https://bugs.webkit.org/show_bug.cgi?id=53441

        Test: fast/css/stylesheet-candidate-nodes-crash.xhtml

        * dom/Document.cpp:
        (WebCore::Document::addStyleSheetCandidateNode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77262 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Tue, 1 Feb 2011 19:09:00 +0000 (19:09 +0000)]
2011-02-01  Dimitri Glazkov  <dglazkov@chromium.org>

        More flakiness updates.

        * platform/chromium/test_expectations.txt: Moar flakiness recorded.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77261 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Daniel Bates <dbates@rim.com>
dbates@webkit.org [Tue, 1 Feb 2011 18:50:17 +0000 (18:50 +0000)]
2011-02-01  Daniel Bates  <dbates@rim.com>

        Reviewed by Antonio Gomes.

        Modify RandomNumberSeed.h to use USE(MERSENNE_TWISTER_19937)
        https://bugs.webkit.org/show_bug.cgi?id=53506

        Currently, use of the Mersenne Twister pseudorandom number generator
        is hardcoded to the Windows CE port. With the passing of bug #53253,
        we can generalize support for this PRNG to all ports that use srand(3)
        and rand(3), including Windows CE.

        * wtf/RandomNumberSeed.h:
        (WTF::initializeRandomNumberGenerator):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77260 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoLand Webpage.cpp for previous patch. Change was lost after conflict resolution from...
hyatt@apple.com [Tue, 1 Feb 2011 18:49:59 +0000 (18:49 +0000)]
Land Webpage.cpp for previous patch.  Change was lost after conflict resolution from a merge.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77259 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoBack out accidental commit of Connection.cpp
hyatt@apple.com [Tue, 1 Feb 2011 18:46:24 +0000 (18:46 +0000)]
Back out accidental commit of Connection.cpp

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77258 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agohttps://bugs.webkit.org/show_bug.cgi?id=46422, make printing and pagination work
hyatt@apple.com [Tue, 1 Feb 2011 18:44:10 +0000 (18:44 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=46422, make printing and pagination work
with vertical text.

Reviewed by Darin Adler.

../WebCore:

Change printing functions to check writing-mode and properly swap width and height
as needed.

Fix the setScrollOrigin function so that the origin doesn't cause
scroll spasming during printing (this is only partially successful, but it's better
than it was).

Rewrite computePageRects to handle both RTL documents properly as well as vertical
text documents properly.

* WebCore.exp.in:
* page/FrameView.cpp:
(WebCore::FrameView::adjustViewSize):
(WebCore::FrameView::forceLayoutForPagination):
* page/PrintContext.cpp:
(WebCore::PrintContext::computePageRects):
(WebCore::PrintContext::computePageRectsWithPageSizeInternal):
(WebCore::PrintContext::computeAutomaticScaleFactor):
(WebCore::PrintContext::spoolPage):
(WebCore::PrintContext::spoolRect):
* page/PrintContext.h:
* page/mac/WebCoreFrameView.h:
* platform/ScrollView.cpp:
(WebCore::ScrollView::wheelEvent):
* platform/ScrollView.h:
* platform/mac/ScrollViewMac.mm:
(WebCore::ScrollView::platformSetScrollOrigin):
* rendering/RenderView.cpp:
(WebCore::RenderView::layout):

../WebKit/mac:

Change printing functions to check writing-mode and properly swap width and height
as needed.

* WebView/WebDynamicScrollBarsView.mm:
(-[WebDynamicScrollBarsView setScrollOrigin:updatePositionAtAll:immediately:]):
* WebView/WebFrame.mm:
(-[WebFrame _computePageRectsWithPrintScaleFactor:pageSize:]):
* WebView/WebFrameInternal.h:
* WebView/WebHTMLView.mm:
(-[WebHTMLView _web_setPrintingModeRecursive]):
(-[WebHTMLView _web_clearPrintingModeRecursive]):
(-[WebHTMLView _web_setPrintingModeRecursiveAndAdjustViewSize]):
(-[WebHTMLView _beginPrintModeWithMinimumPageWidth:height:maximumPageWidth:]):
(-[WebHTMLView _beginPrintModeWithPageWidth:height:shrinkToFit:]):
(-[WebHTMLView _endPrintMode]):
(-[WebHTMLView _beginScreenPaginationModeWithPageSize:shrinkToFit:]):
(-[WebHTMLView _endScreenPaginationMode]):
(-[WebHTMLView layoutToMinimumPageWidth:height:maximumPageWidth:adjustingViewSize:]):
(-[WebHTMLView _setPrinting:minimumPageLogicalWidth:logicalHeight:maximumPageLogicalWidth:adjustViewSize:paginateScreenContent:]):
(-[WebHTMLView adjustPageHeightNew:top:bottom:limit:]):
(-[WebHTMLView _scaleFactorForPrintOperation:]):
(-[WebHTMLView setPageWidthForPrinting:]):
(-[WebHTMLView knowsPageRange:]):

../WebKit2:

Change printing functions to check writing-mode and properly swap width and height
as needed.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::computePagesForPrinting):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77257 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Tue, 1 Feb 2011 18:29:47 +0000 (18:29 +0000)]
2011-02-01  Mihai Parparita  <mihaip@chromium.org>

        Unreviewed roll of Chromium deps to r73306. This picks up a fix to get
        the WebKit root dir correctly after the move to Source/. This allows
        tests like http/tests/security/frame-loading-via-document-write.html
        (which use LayoutTestController.pathToLocalResource) to pass with the
        Chromium DRT.

        * DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77256 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Tue, 1 Feb 2011 17:57:31 +0000 (17:57 +0000)]
2011-02-01  Dimitri Glazkov  <dglazkov@chromium.org>

        Updated flaky test expectations according to build bot results.

        * platform/chromium/test_expectations.txt: Flakiness recorded.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77255 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Mikhail Naganov <mnaganov@chromium.org>
mnaganov@chromium.org [Tue, 1 Feb 2011 17:22:36 +0000 (17:22 +0000)]
2011-02-01  Mikhail Naganov  <mnaganov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: Fix profiles reset to avoid clearing heap profiles in Chromium.

        https://bugs.webkit.org/show_bug.cgi?id=53500

        * inspector/InspectorProfilerAgent.cpp:
        (WebCore::InspectorProfilerAgent::resetFrontendProfiles):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77254 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Balazs Kelemen <kbalazs@webkit.org>
kbalazs@webkit.org [Tue, 1 Feb 2011 17:09:26 +0000 (17:09 +0000)]
2011-02-01  Balazs Kelemen  <kbalazs@webkit.org>

        Reviewed by Andreas Kling.

        [Qt][WK2] Add a way to use shared process model in MiniBrowser
        https://bugs.webkit.org/show_bug.cgi?id=53090

        * MiniBrowser/qt/BrowserView.cpp:
        (BrowserView::BrowserView): Removed the m_context member.
        From now the context is guaranteed to be non-null and we
        don't need to store that in the object.
        * MiniBrowser/qt/BrowserView.h:
        * MiniBrowser/qt/BrowserWindow.cpp:
        Added static bool to determine that new windows need to be
        created with their own context or not. Use the same context
        and web process by default to be inilne with the other ports.
        (BrowserWindow::BrowserWindow):
        (BrowserWindow::newWindow):
        * MiniBrowser/qt/BrowserWindow.h:
        * MiniBrowser/qt/main.cpp:
        (main): Added command line switch to be able to use the
        non-shared process model. Simplify the handling of the command line
        switches a little bit.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77253 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Mikhail Naganov <mnaganov@chromium.org>
mnaganov@chromium.org [Tue, 1 Feb 2011 16:59:38 +0000 (16:59 +0000)]
2011-02-01  Mikhail Naganov  <mnaganov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: [Chromium] Landing detailed heap snapshots, part 1.

        https://bugs.webkit.org/show_bug.cgi?id=53173

        Adding code for accessing heap snapshot data and
        performing graph calculations.

        * English.lproj/localizedStrings.js:
        * inspector/front-end/HeapSnapshot.js:
        (WebInspector.HeapSnapshotArraySlice): Helper class to avoid array contents copying.
        (WebInspector.HeapSnapshotEdge): Wrapper for accessing graph edge properties.
        (WebInspector.HeapSnapshotEdgeIterator):
        (WebInspector.HeapSnapshotNode): Wrapper for accessing graph node properties.
        (WebInspector.HeapSnapshotNodeIterator):
        (WebInspector.HeapSnapshot): Wrapper for the heap snapshot.
        (WebInspector.HeapSnapshotFilteredOrderedIterator):
        (WebInspector.HeapSnapshotEdgesProvider):
        (WebInspector.HeapSnapshotNodesProvider):
        (WebInspector.HeapSnapshotPathFinder):
        * inspector/front-end/HeapSnapshotView.js:
        (WebInspector.HeapSnapshotView.prototype._convertSnapshot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77252 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix linker warnings in Release_LTCG builds
aroben@apple.com [Tue, 1 Feb 2011 15:46:53 +0000 (15:46 +0000)]
Fix linker warnings in Release_LTCG builds

* WebCore.vcproj/WebCore.vcproj: Exclude EventNames.cpp and EventTarget.cpp from all
configurations, since they get pulled in via DOMAllInOne.cpp.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77251 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Add Window menu and Toggle FullScreen action to MiniBrowser.
zoltan@webkit.org [Tue, 1 Feb 2011 14:57:38 +0000 (14:57 +0000)]
[Qt] Add Window menu and Toggle FullScreen action to MiniBrowser.
https://bugs.webkit.org/show_bug.cgi?id=53491

Reviewed by Andreas Kling.

* MiniBrowser/qt/BrowserWindow.cpp:
(BrowserWindow::BrowserWindow):
(BrowserWindow::toggleFullScreenMode):
* MiniBrowser/qt/BrowserWindow.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77250 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Alexander Pavlov <apavlov@chromium.org>
apavlov@chromium.org [Tue, 1 Feb 2011 14:53:15 +0000 (14:53 +0000)]
2011-02-01  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: [Chromium] Wrongly labelled context-menu item for links in Web Inspector's side-pane
        https://bugs.webkit.org/show_bug.cgi?id=53482

        * English.lproj/localizedStrings.js:
        * inspector/front-end/ElementsPanel.js:
        (WebInspector.ElementsPanel.prototype.populateHrefContextMenu):
        * inspector/front-end/inspector.js:
        (WebInspector.resourceForURL):
        (WebInspector.openLinkExternallyLabel):
2011-02-01  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: [Chromium] Wrongly labelled context-menu item for links in Web Inspector's side-pane
        https://bugs.webkit.org/show_bug.cgi?id=53482

        * src/js/DevTools.js:
        (WebInspector.openLinkExternallyLabel): Override to customize a context menu item label

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77249 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Dave Tapuska <dtapuska@rim.com>
commit-queue@webkit.org [Tue, 1 Feb 2011 14:40:03 +0000 (14:40 +0000)]
2011-02-01  Dave Tapuska  <dtapuska@rim.com>

        Reviewed by Gavin Barraclough.

        MacroAssemblerARM would generate code that did 32bit loads
        on addresses that were not aligned. More specifically it would
        generate a ldr r8,[r1, #7] which isn't valid on ARMv5 and lower.
        The intended instruction really is ldrb r8,[r1, #7]; ensure we
        call load8 instead of load32.

        https://bugs.webkit.org/show_bug.cgi?id=46095

        * assembler/MacroAssemblerARM.h:
        (JSC::MacroAssemblerARM::set32Test32):
        (JSC::MacroAssemblerARM::set32Test8):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77248 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoUnreviewed.
ossy@webkit.org [Tue, 1 Feb 2011 14:38:44 +0000 (14:38 +0000)]
Unreviewed.

fast/loader/onload-policy-ignore-for-frame.html fails on Qt bot
https://bugs.webkit.org/show_bug.cgi?id=53490

Add Qt specific expected result.

* platform/qt/fast/loader/onload-policy-ignore-for-frame-expected.txt: Copied from LayoutTests/fast/loader/onload-policy-ignore-for-frame-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77247 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Anton Muhin <antonm@chromium.org>
antonm@chromium.org [Tue, 1 Feb 2011 14:35:10 +0000 (14:35 +0000)]
2011-02-01  Anton Muhin  <antonm@chromium.org>

        Reviewed by Adam Barth.

        Propagate parent document security origin to newly create Document XML response
        https://bugs.webkit.org/show_bug.cgi?id=53444

        Covered by the existing tests.

        * xml/XMLHttpRequest.cpp:
        (WebCore::XMLHttpRequest::responseXML):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77246 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago[Qt] Add Toggle Frame Flattening, Auto Load Images and Disable JavaScript actions...
zoltan@webkit.org [Tue, 1 Feb 2011 13:32:27 +0000 (13:32 +0000)]
[Qt] Add Toggle Frame Flattening, Auto Load Images and Disable JavaScript actions to MiniBrowser.
https://bugs.webkit.org/show_bug.cgi?id=53489

Reviewed by Andreas Kling.

Add Toggle Frame Flattening action to Develop menu. Add Auto Load Images and Disable JavaScript actions
to Settings menu.

* MiniBrowser/qt/BrowserWindow.cpp:
(BrowserWindow::BrowserWindow):
(BrowserWindow::toggleFrameFlattening):
(BrowserWindow::toggleDisableJavaScript):
(BrowserWindow::toggleAutoLoadImages):
* MiniBrowser/qt/BrowserWindow.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77245 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Gyuyoung Kim <gyuyoung.kim@samsung.com>
gyuyoung.kim@samsung.com [Tue, 1 Feb 2011 12:42:44 +0000 (12:42 +0000)]
2011-02-01  Gyuyoung Kim  <gyuyoung.kim@samsung.com>

        Reviewed by Antonio Gomes.

        [EFL] Add an API to get proxy address.
        https://bugs.webkit.org/show_bug.cgi?id=52990

        Gets the proxy URI from network backend. This API returns a proxy URI
        which is set up by ewk_settings_proxy_uri_set(). But, it only implements
        this API for libsoup backend.

        * ewk/ewk_settings.cpp:
        (ewk_settings_proxy_uri_get):
        * ewk/ewk_settings.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77243 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Darin Fisher <darin@chromium.org>
darin@chromium.org [Tue, 1 Feb 2011 12:42:16 +0000 (12:42 +0000)]
2011-02-01  Darin Fisher  <darin@chromium.org>

        Reviewed by Eric Seidel.

        Fix some Visual Studio compiler warnings.
        https://bugs.webkit.org/show_bug.cgi?id=53476

        * wtf/MathExtras.h:
        (clampToInteger):
        (clampToPositiveInteger):
        * wtf/ThreadingWin.cpp:
        (WTF::absoluteTimeToWaitTimeoutInterval):
2011-02-01  Darin Fisher  <darin@chromium.org>

        Reviewed by Eric Seidel.

        Fix some Visual Studio compiler warnings.
        https://bugs.webkit.org/show_bug.cgi?id=53476

        * src/GraphicsContext3DChromium.cpp:
        (WebCore::GraphicsContext3DInternal::paintRenderingResultsToCanvas):
        * src/WebFrameImpl.cpp:
        (WebKit::WebFrameImpl::forms):
        * src/WebMediaPlayerClientImpl.cpp:
        (WebKit::WebMediaPlayerClientImpl::repaint):
        * src/WebScrollbarImpl.cpp:
        (WebKit::WebScrollbarImpl::setValue):
        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::doPixelReadbackToCanvas):
        (WebKit::WebViewImpl::setZoomLevel):
        (WebKit::WebViewImplScrollbarPaintInterface::paint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77242 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Yury Semikhatsky <yurys@chromium.org>
yurys@chromium.org [Tue, 1 Feb 2011 12:16:33 +0000 (12:16 +0000)]
2011-02-01  Yury Semikhatsky  <yurys@chromium.org>

        Unreviewed. Rollout r77230 which caused many layout tests
        crashes on Chromium Debug bots.

        Async event handlers should not fire within a modal dialog
        https://bugs.webkit.org/show_bug.cgi?id=53202

        * dom/Document.cpp:
        (WebCore::Document::Document):
        * dom/EventQueue.cpp:
        (WebCore::EventQueue::EventQueue):
        (WebCore::EventQueue::enqueueEvent):
        (WebCore::EventQueue::pendingEventTimerFired):
        * dom/EventQueue.h:
2011-02-01  Yury Semikhatsky  <yurys@chromium.org>

        Unreviewed. Rollout r77230 which caused many layout tests
        crashes on Chromium Debug bots.

        Async event handlers should not fire within a modal dialog
        https://bugs.webkit.org/show_bug.cgi?id=53202

        * fast/events/scroll-event-during-modal-dialog-expected.txt: Removed.
        * fast/events/scroll-event-during-modal-dialog.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77241 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoLightElement changes does not require relayout.
zherczeg@webkit.org [Tue, 1 Feb 2011 10:48:02 +0000 (10:48 +0000)]
LightElement changes does not require relayout.
https://bugs.webkit.org/show_bug.cgi?id=53232

Reviewed by Dirk Schulze.

When an attribute of a LightElement changes, it
send an update message to the lighting filters
to update its corresponding LightSource objects,
and repaint the filters.

Duplicated 'id' attributes removed from svg-filter-animation.svg.

Existing dynamic-update tests covers this feature.

5x speedup on manual-tests/svg-filter-animation.svg

* manual-tests/svg-filter-animation.svg:
* platform/graphics/filters/DistantLightSource.h:
* platform/graphics/filters/FEDiffuseLighting.cpp:
(WebCore::FEDiffuseLighting::setLightingColor):
(WebCore::FEDiffuseLighting::setSurfaceScale):
(WebCore::FEDiffuseLighting::setDiffuseConstant):
(WebCore::FEDiffuseLighting::setKernelUnitLengthX):
(WebCore::FEDiffuseLighting::setKernelUnitLengthY):
* platform/graphics/filters/FEDiffuseLighting.h:
* platform/graphics/filters/LightSource.cpp:
(WebCore::PointLightSource::setX):
(WebCore::PointLightSource::setY):
(WebCore::PointLightSource::setZ):
(WebCore::SpotLightSource::setX):
(WebCore::SpotLightSource::setY):
(WebCore::SpotLightSource::setZ):
(WebCore::SpotLightSource::setPointsAtX):
(WebCore::SpotLightSource::setPointsAtY):
(WebCore::SpotLightSource::setPointsAtZ):
(WebCore::SpotLightSource::setSpecularExponent):
(WebCore::SpotLightSource::setLimitingConeAngle):
(WebCore::DistantLightSource::setAzimuth):
(WebCore::DistantLightSource::setElevation):
(WebCore::LightSource::setAzimuth):
(WebCore::LightSource::setElevation):
(WebCore::LightSource::setX):
(WebCore::LightSource::setY):
(WebCore::LightSource::setZ):
(WebCore::LightSource::setPointsAtX):
(WebCore::LightSource::setPointsAtY):
(WebCore::LightSource::setPointsAtZ):
(WebCore::LightSource::setSpecularExponent):
(WebCore::LightSource::setLimitingConeAngle):
* platform/graphics/filters/LightSource.h:
* platform/graphics/filters/PointLightSource.h:
* platform/graphics/filters/SpotLightSource.h:
* rendering/svg/RenderSVGResourceFilter.cpp:
(WebCore::RenderSVGResourceFilter::primitiveAttributeChanged):
* svg/SVGFEDiffuseLightingElement.cpp:
(WebCore::SVGFEDiffuseLightingElement::setFilterEffectAttribute):
(WebCore::SVGFEDiffuseLightingElement::lightElementAttributeChanged):
(WebCore::SVGFEDiffuseLightingElement::build):
(WebCore::SVGFEDiffuseLightingElement::findLightElement):
(WebCore::SVGFEDiffuseLightingElement::findLight):
* svg/SVGFEDiffuseLightingElement.h:
* svg/SVGFELightElement.cpp:
(WebCore::SVGFELightElement::svgAttributeChanged):
* svg/SVGFilterPrimitiveStandardAttributes.cpp:
(WebCore::SVGFilterPrimitiveStandardAttributes::setFilterEffectAttribute):
* svg/SVGFilterPrimitiveStandardAttributes.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77240 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Roland Steiner <rolandsteiner@chromium.org>
rolandsteiner@chromium.org [Tue, 1 Feb 2011 10:41:19 +0000 (10:41 +0000)]
2011-02-01  Roland Steiner  <rolandsteiner@chromium.org>

        Reviewed by Dimitri Glazkov.

        Bug 53289 - DOM: Move DocumentOrderedMap from Document into separate files
        https://bugs.webkit.org/show_bug.cgi?id=53289

        Moving the nested class DocumentOrderedMap from Document into separate files,
        updating code where necessary.

        No new tests. (refactoring)

        * Android.mk:
        * CMakeLists.txt:
        * GNUMakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * dom/Document.cpp:
        (WebCore::Document::getElementById):
        (WebCore::Document::getImageMap):
        * dom/Document.h:
        * dom/DocumentOrderedMap.cpp: Added.
        (WebCore::keyMatchesId):
        (WebCore::keyMatchesMapName):
        (WebCore::keyMatchesLowercasedMapName):
        (WebCore::DocumentOrderedMap::clear):
        (WebCore::DocumentOrderedMap::add):
        (WebCore::DocumentOrderedMap::remove):
        (WebCore::DocumentOrderedMap::get):
        (WebCore::DocumentOrderedMap::getElementById):
        (WebCore::DocumentOrderedMap::getElementByMapName):
        (WebCore::DocumentOrderedMap::getElementByLowercasedMapName):
        * dom/DocumentOrderedMap.h: Added.
        (WebCore::DocumentOrderedMap::contains):
        (WebCore::DocumentOrderedMap::containsMultiple):
        * dom/DOMAllInOne.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77239 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-01-28 Commit Queue <commit-queue@webkit.org>
commit-queue@webkit.org [Tue, 1 Feb 2011 10:32:17 +0000 (10:32 +0000)]
2010-01-28  Commit Queue  <commit-queue@webkit.org>

        Unreviewed.  Test commit for commit-queue@webkit.org.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77238 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2010-01-28 Commit Queue <commit-queue@webkit.org>
commit-queue@webkit.org [Tue, 1 Feb 2011 10:27:49 +0000 (10:27 +0000)]
2010-01-28  Commit Queue  <commit-queue@webkit.org>

        Unreviewed.  Test commit for commit-queue@webkit.org.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77237 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Hans Wennborg <hans@chromium.org>
hans@chromium.org [Tue, 1 Feb 2011 10:18:28 +0000 (10:18 +0000)]
2011-02-01  Hans Wennborg  <hans@chromium.org>

        Reviewed by Jeremy Orlow.

        Chromium: remove WebIDBCallbacks::onSucces()
        https://bugs.webkit.org/show_bug.cgi?id=53415

        Remove un-used function.

        * public/WebIDBCallbacks.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77236 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Mario Sanchez Prada <msanchez@igalia.com>
mario@webkit.org [Tue, 1 Feb 2011 09:58:30 +0000 (09:58 +0000)]
2011-02-01  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Martin Robinson.

        [Gtk] atk_text_set_caret_offset fails for list items
        https://bugs.webkit.org/show_bug.cgi?id=53388

        Allow using text ranges across list items.

        * accessibility/gtk/AccessibilityObjectAtk.cpp:
        (WebCore::AccessibilityObject::allowsTextRanges): Add list items
        to the list of accessibility objects supporting text ranges.
2011-02-01  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Martin Robinson.

        [Gtk] atk_text_set_caret_offset fails for list items
        https://bugs.webkit.org/show_bug.cgi?id=53388

        Update unit test to check the fix for this bug.

        * tests/testatk.c:
        (testWebkitAtkCaretOffsets): Ensure setting the caret in a list
        item through the AtkText interface is possible.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77235 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Mario Sanchez Prada <msanchez@igalia.com>
mario@webkit.org [Tue, 1 Feb 2011 09:54:23 +0000 (09:54 +0000)]
2011-02-01  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] character range extents is off when the end of a wrapped line is included
        https://bugs.webkit.org/show_bug.cgi?id=53323

        Fixed wrong calculation getting the range extents.

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (webkit_accessible_text_get_range_extents): Removed '+1' since the
        requested interval shouldn't include the last character.
2011-02-01  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] character range extents is off when the end of a wrapped line is included
        https://bugs.webkit.org/show_bug.cgi?id=53323

        Update unit test to check the fix for this bug.

        * tests/testatk.c:
        (testWebkitAtkGetExtents): Check that the extent for a full line
        are the same height than for a partial section of the same line.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77234 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Mario Sanchez Prada <msanchez@igalia.com>
mario@webkit.org [Tue, 1 Feb 2011 09:49:25 +0000 (09:49 +0000)]
2011-02-01  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Caret Offset is one off at the end of wrapped lines
        https://bugs.webkit.org/show_bug.cgi?id=53300

        Consider linebreaks as special cases.

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (objectAndOffsetUnignored): In order to avoid getting wrong values
        when around linebreaks, we need to workaround this by explicitly
        avoiding those '\n' text nodes from affecting the result of
        calling to TextIterator:rangeLength().
2011-02-01  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Caret Offset is one off at the end of wrapped lines
        https://bugs.webkit.org/show_bug.cgi?id=53300

        Update unit test to check the fix for this bug.

        * tests/testatk.c:
        (testWebkitAtkCaretOffsetsAndExtranousWhiteSpaces): Set and get
        the caret offset at the edge of the line.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77233 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Yury Semikhatsky <yurys@chromium.org>
yurys@chromium.org [Tue, 1 Feb 2011 09:37:23 +0000 (09:37 +0000)]
2011-02-01  Yury Semikhatsky  <yurys@chromium.org>

        Unreviewed. Fix Chromium Debug compilation broken by r77228.

        [Chromium] Autofill should work with HTML5 form elements
        https://bugs.webkit.org/show_bug.cgi?id=51809

        * src/WebInputElement.cpp:
        (WebKit::toWebInputElement):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77232 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Roland Steiner <rolandsteiner@chromium.org>
rolandsteiner@chromium.org [Tue, 1 Feb 2011 09:32:28 +0000 (09:32 +0000)]
2011-02-01  Roland Steiner  <rolandsteiner@chromium.org>

        Unreviewed, rolling out r77229.
        http://trac.webkit.org/changeset/77229
        https://bugs.webkit.org/show_bug.cgi?id=53289

        revert mysterious build breakage

        * Android.mk:
        * CMakeLists.txt:
        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * dom/DOMAllInOne.cpp:
        * dom/Document.cpp:
        (WebCore::Document::DocumentOrderedMap::clear):
        (WebCore::Document::DocumentOrderedMap::add):
        (WebCore::Document::DocumentOrderedMap::remove):
        (WebCore::Document::DocumentOrderedMap::get):
        (WebCore::keyMatchesId):
        (WebCore::Document::getElementById):
        (WebCore::keyMatchesMapName):
        (WebCore::keyMatchesLowercasedMapName):
        (WebCore::Document::getImageMap):
        * dom/Document.h:
        (WebCore::Document::DocumentOrderedMap::contains):
        (WebCore::Document::DocumentOrderedMap::containsMultiple):
        * dom/DocumentOrderedMap.cpp: Removed.
        * dom/DocumentOrderedMap.h: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77231 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Mihai Parparita <mihaip@chromium.org>
mihaip@chromium.org [Tue, 1 Feb 2011 09:20:59 +0000 (09:20 +0000)]
2011-02-01  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by James Robinson.

        Async event handlers should not fire within a modal dialog
        https://bugs.webkit.org/show_bug.cgi?id=53202

        Add tests for a scroll event triggered right before a modal dialog is
        shown.

        * fast/events/scroll-event-during-modal-dialog-expected.txt: Added.
        * fast/events/scroll-event-during-modal-dialog.html: Added.
2011-02-01  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by James Robinson.

        Async event handlers should not fire within a modal dialog
        https://bugs.webkit.org/show_bug.cgi?id=53202

        Asychronous events that use EventQueue would currently fire while a
        modal dialog (e.g. window.alert()) was up. Change EventQueue to use a
        SuspendableTimer (which automatically gets suspended while dialogs are
        up and in other cases where JS execution is not allowed).

        Test: fast/events/scroll-event-during-modal-dialog.html

        * dom/Document.cpp:
        (WebCore::Document::Document):
        * dom/EventQueue.cpp:
        (WebCore::EventQueueTimer::EventQueueTimer):
        (WebCore::EventQueueTimer::fired):
        (WebCore::EventQueue::EventQueue):
        (WebCore::EventQueue::enqueueEvent):
        (WebCore::EventQueue::pendingEventTimerFired):
        * dom/EventQueue.h:
        (WebCore::EventQueue::create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77230 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-28 Roland Steiner <rolandsteiner@chromium.org>
rolandsteiner@chromium.org [Tue, 1 Feb 2011 09:11:02 +0000 (09:11 +0000)]
2011-01-28  Roland Steiner  <rolandsteiner@chromium.org>

        Reviewed by NOBODY (OOPS!).

        Bug 53289 - DOM: Move DocumentOrderedMap from Document into separate files
        https://bugs.webkit.org/show_bug.cgi?id=53289

        Moving the nested class DocumentOrderedMap from Document into separate files,
        updating code where necessary.

        No new tests. (refactoring)

        * Android.mk:
        * CMakeLists.txt:
        * GNUMakefile.am
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * dom/Document.cpp:
        (WebCore::Document::getElementById):
        (WebCore::Document::getImageMap):
        * dom/Document.h:
        * dom/DocumentOrderedMap.cpp: Added.
        (WebCore::keyMatchesId):
        (WebCore::keyMatchesMapName):
        (WebCore::keyMatchesLowercasedMapName):
        (WebCore::DocumentOrderedMap::clear):
        (WebCore::DocumentOrderedMap::add):
        (WebCore::DocumentOrderedMap::remove):
        (WebCore::DocumentOrderedMap::get):
        (WebCore::DocumentOrderedMap::getElementById):
        (WebCore::DocumentOrderedMap::getElementByMapName):
        (WebCore::DocumentOrderedMap::getElementByLowercasedMapName):
        * dom/DocumentOrderedMap.h: Added.
        (WebCore::DocumentOrderedMap::contains):
        (WebCore::DocumentOrderedMap::containsMultiple):
        * dom/DOMAllInOne.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77229 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-02-01 Naoki Takano <takano.naoki@gmail.com>
commit-queue@webkit.org [Tue, 1 Feb 2011 08:54:25 +0000 (08:54 +0000)]
2011-02-01  Naoki Takano  <takano.naoki@gmail.com>

        Reviewed by Darin Fisher.

        [Chromium] Autofill should work with HTML5 form elements
        https://bugs.webkit.org/show_bug.cgi?id=51809
        http://crbug.com/65654

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

        * html/InputType.h: Insert comment for canSetSuggestedValue().
        * html/TextFieldInputType.cpp:
        (WebCore::TextFieldInputType::canSetSuggestedValue): Implemented to return always true for that all text filed inputs can be completed.
        * html/TextFieldInputType.h: Declare canSetSuggestedValue().
        * html/TextInputType.cpp: Delete canSetSuggestedValue() not to return true anymore.
        * html/TextInputType.h: Delete canSetSuggestedValue() not to return true anymore.
2011-02-01  Naoki Takano  <takano.naoki@gmail.com>

        Reviewed by Darin Fisher.

        [Chromium] Autofill should work with HTML5 form elements
        https://bugs.webkit.org/show_bug.cgi?id=51809
        http://crbug.com/65654

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

        * public/WebInputElement.h: Added toWebInputElement() declarations and implemented const version of toWebInputElement().
        * src/WebInputElement.cpp:
        (WebKit::WebInputElement::toWebInputElement): Implemented a convert function to cast no-const WebElement* to no-const WebInputElement*. Because Chrome needs safe cast for autofill completion.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77228 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoREGRESSION (r65062): Safari loops forever under WebCore::plainTextToMallocAllocatedBu...
tkent@chromium.org [Tue, 1 Feb 2011 08:05:08 +0000 (08:05 +0000)]
REGRESSION (r65062): Safari loops forever under WebCore::plainTextToMallocAllocatedBuffer()
https://bugs.webkit.org/show_bug.cgi?id=53272

Reviewed by Dan Bernstein.

Source/WebCore:

* editing/TextIterator.cpp:
(WebCore::TextIterator::handleTextBox): Pass the appropriate renderer to emitText().

LayoutTests:

* fast/dom/inner-text-first-letter.html:
  - Add a new test case reproducing the bug.
  - Remove unnecessary text output.
* fast/dom/inner-text-first-letter-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77224 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago Reviewed by Maciej Stachowiak.
ap@apple.com [Tue, 1 Feb 2011 05:53:05 +0000 (05:53 +0000)]
    Reviewed by Maciej Stachowiak.

        https://bugs.webkit.org/show_bug.cgi?id=53466
        Move WebKit2 to printing via API methods

        Also fixes <rdar://problem/8933724> REGRESSION: Wrong pages are printed when not printing from page 1

        Also fixes horizontal tiling (no bug filed).

        The main idea here is that we don't force AppKit to use a different scaling factor when
        a Web view can't be resized to exact page size, and scale when drawing instead.

        * UIProcess/API/mac/WKPrintingView.mm:
        (-[WKPrintingView _firstPrintedPageNumber]): Factored out into a separate method.
        (-[WKPrintingView _lastPrintedPageNumber]): Ditto.
        (pageDidDrawToPDF): Avoid crashing if data is null.
        (-[WKPrintingView _preparePDFDataForPrintingOnSecondaryThread]): Use -_firstPrintedPageNumber
        and -_lastPrintedPageNumber.
        (pageDidComputePageRects): Set frame size to what it would have been for a normal NSView
        that's been asked to resize.
        (-[WKPrintingView _pageForRect:]): Check x(), because several pages can be at the same y()
        when tiling horizontally.
        (-[WKPrintingView _drawPDFDocument:page:atPoint:]): Point to draw at and frame size are
        now using different units, translate as appropriate.
        (-[WKPrintingView _drawPreview:]): Scale the rect to WebCore units.
        (-[WKPrintingView drawRect:]): Correctly calculate page number when the first page to print
        isn't page 1.
        (-[WKPrintingView rectForPage:]): Scale the rect to AppKit units.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77220 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-31 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 1 Feb 2011 05:51:57 +0000 (05:51 +0000)]
2011-01-31  Ryosuke Niwa  <rniwa@webkit.org>

        Unreviewed Chromium test expectation updates.

        Removed media/controls-strict.html and svg/custom/clip-path-referencing-use2.svg (Mac)
        since it has been passing since r77171 and r77172-r77173 respectively.

        Also removed http/tests/media/reload-after-dialog.html because it has been passing for a while.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77219 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-31 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Tue, 1 Feb 2011 05:40:38 +0000 (05:40 +0000)]
2011-01-31  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Adam Barth.

        Remove obsolete comment after r41871
        https://bugs.webkit.org/show_bug.cgi?id=53406

        * dom/Document.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77216 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-31 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 1 Feb 2011 05:33:08 +0000 (05:33 +0000)]
2011-01-31  Simon Fraser  <simon.fraser@apple.com>

        Fix according to reviewer comments: can just use Color::black now.

        * platform/graphics/ShadowBlur.cpp:
        (WebCore::ShadowBlur::drawInsetShadow):
        (WebCore::ShadowBlur::drawRectShadowWithoutTiling):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77215 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-31 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 1 Feb 2011 05:19:58 +0000 (05:19 +0000)]
2011-01-31  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Sam Weinig.

        Clean up ShadowBlur
        https://bugs.webkit.org/show_bug.cgi?id=53472

        Some minor ShadowBlur cleanup.

        * platform/graphics/ShadowBlur.h:
        * platform/graphics/ShadowBlur.cpp:
        (WebCore::ShadowBlur::ShadowBlur): Use m_blurRadius rather than the radius
        paramter.
        (WebCore::ShadowBlur::adjustBlurRadius): Renamed from adjustBlurDistance.
        (WebCore::ShadowBlur::calculateLayerBoundingRect): Rename layerFloatRect to
        layerRect. Make frameSize a float.
        (WebCore::ShadowBlur::beginShadowLayer): This now takes a precomputed
        layerRect rather than calling calculateLayerBoundingRect() to compute
        it itself, since we were calling calculateLayerBoundingRect() twice.
        (WebCore::ShadowBlur::drawRectShadow): Optimize to call calculateLayerBoundingRect()
        only once. The shadowRect variable was unused, so two return paths could be
        collapsed into one.
        (WebCore::ShadowBlur::drawInsetShadow): Call calculateLayerBoundingRect() before
        beginShadowLayer() now.
        (WebCore::ShadowBlur::drawRectShadowWithoutTiling): The layerRect gets passed in.
        We always used alpha=1, so no need to pass that in.
        (WebCore::ShadowBlur::drawRectShadowWithTiling): We always used alpha=1, so no need to
        pass that in. Move shadowRect down to first use.
        ShadowBlur::clipBounds() was unused.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77213 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-31 No'am Rosenthal <noam.rosenthal@nokia.com>
noam.rosenthal@nokia.com [Tue, 1 Feb 2011 04:34:42 +0000 (04:34 +0000)]
2011-01-31  No'am Rosenthal  <noam.rosenthal@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] QWebElements example from QtWebKit Bridge documentation does not work at all
        https://bugs.webkit.org/show_bug.cgi?id=46748

        This problem disappears when we register QWebElement using qRegisterMetaType, which we now do in QtInstance.
        Added a regression test to tst_QWebFrame.

        * bridge/qt/qt_instance.cpp:
        (JSC::Bindings::QtInstance::QtInstance):
2011-01-31  No'am Rosenthal  <noam.rosenthal@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] QWebElements example from QtWebKit Bridge documentation does not work at all
        https://bugs.webkit.org/show_bug.cgi?id=46748

        Problem was that we need to register QWebElement's variant-type to Qt's meta type system at some point.
        This wasn't caught by previous tests inside tst_QWebFrame because it only appears when there's also a QWebElement property.
        Added a regression test to tst_QWebFrame.

        * tests/qwebframe/tst_qwebframe.cpp:
        (MyWebElementSlotOnlyObject::doSomethingWithWebElement):
        (MyWebElementSlotOnlyObject::tagName):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77212 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-27 MORITA Hajime <morrita@google.com>
morrita@google.com [Tue, 1 Feb 2011 04:05:52 +0000 (04:05 +0000)]
2011-01-27  MORITA Hajime  <morrita@google.com>

        Reviewed by Dimitri Glazkov.

        Convert <progress> shadow DOM to a DOM-based shadow.
        https://bugs.webkit.org/show_bug.cgi?id=50660

        * Removed RenderProgress::m_valuePart, moved the shadow node
          to the shadow root of HTMLProgressElement.
        * Removed hard-coded pseudo ID for -webkit-progress-bar-value.
          ProgressBarValueElement is defined only for overriding
          shadowPseudoId().

        No new tests. No behavioral change.

        * css/CSSSelector.cpp:
        (WebCore::CSSSelector::pseudoId):
        (WebCore::nameToPseudoTypeMap):
        (WebCore::CSSSelector::extractPseudoType):
        * css/CSSSelector.h:
        * html/HTMLProgressElement.cpp:
        (WebCore::ProgressBarValueElement::ProgressBarValueElement):
        (WebCore::ProgressBarValueElement::shadowPseudoId):
        (WebCore::ProgressBarValueElement::create):
        (WebCore::ProgressBarValueElement::detach):
        (WebCore::HTMLProgressElement::parseMappedAttribute):
        (WebCore::HTMLProgressElement::attach):
        (WebCore::HTMLProgressElement::valuePart):
        (WebCore::HTMLProgressElement::didElementStateChange):
        (WebCore::HTMLProgressElement::createShadowSubtreeIfNeeded):
        * html/HTMLProgressElement.h:
        * rendering/RenderProgress.cpp:
        (WebCore::RenderProgress::~RenderProgress):
        (WebCore::RenderProgress::updateFromElement):
        (WebCore::RenderProgress::layoutParts):
        (WebCore::RenderProgress::shouldHaveParts):
        (WebCore::RenderProgress::valuePart):
        * rendering/RenderProgress.h:
        * rendering/style/RenderStyleConstants.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77211 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years ago2011-01-31 Charlie Reis <creis@chromium.org>
commit-queue@webkit.org [Tue, 1 Feb 2011 04:04:16 +0000 (04:04 +0000)]
2011-01-31  Charlie Reis  <creis@chromium.org>

        Reviewed by Mihai Parparita.

        Add sanity check to help diagnose bug 52819
        https://bugs.webkit.org/show_bug.cgi?id=53402

        Crash early if the children of fromItem look invalid.

        * loader/HistoryController.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77210 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9 years agoFix the change log entry order.
rniwa@webkit.org [Tue, 1 Feb 2011 03:56:28 +0000 (03:56 +0000)]
Fix the change log entry order.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@77209 268f45cc-cd09-0410-ab3c-d52691b4dbfc