Timeline
Nov 23, 2009:
- 8:25 PM Changeset in webkit [51336] by
-
- 19 edits in trunk
2009-11-23 Laszlo Gombos <Laszlo Gombos>
Reviewed by Kenneth Rohde Christiansen.
Include "config.h" to meet Coding Style Guidelines
https://bugs.webkit.org/show_bug.cgi?id=31792
- wtf/unicode/UTF8.cpp:
- wtf/unicode/glib/UnicodeGLib.cpp:
- wtf/unicode/wince/UnicodeWince.cpp:
2009-11-23 Laszlo Gombos <Laszlo Gombos>
Reviewed by Kenneth Rohde Christiansen.
Include "config.h" to meet Coding Style Guidelines
https://bugs.webkit.org/show_bug.cgi?id=31792
- GEN_DOMObject.cpp:
- WebScriptObject.cpp:
2009-11-23 Laszlo Gombos <Laszlo Gombos>
Reviewed by Kenneth Rohde Christiansen.
Include "config.h" to meet Coding Style Guidelines
https://bugs.webkit.org/show_bug.cgi?id=31792
- WebKitSupport/DragClientWx.cpp:
2009-11-23 Laszlo Gombos <Laszlo Gombos>
Reviewed by Kenneth Rohde Christiansen.
Include "config.h" to meet Coding Style Guidelines
https://bugs.webkit.org/show_bug.cgi?id=31792
No new tests as there is no new functionality.
- platform/graphics/win/IntPointWin.cpp:
- platform/graphics/win/IntRectWin.cpp:
- platform/graphics/win/IntSizeWin.cpp:
- platform/network/chromium/ResourceRequest.cpp:
- platform/win/PlatformMouseEventWin.cpp:
2009-11-23 Laszlo Gombos <Laszlo Gombos>
Reviewed by Kenneth Rohde Christiansen.
Include "config.h" to meet Coding Style Guidelines
https://bugs.webkit.org/show_bug.cgi?id=31792
- DumpRenderTree/pthreads/JavaScriptThreadingPthreads.cpp:
- DumpRenderTree/qt/jsobjects.cpp:
- DumpRenderTree/win/DRTDesktopNotificationPresenter.cpp:
- 6:33 PM Changeset in webkit [51335] by
-
- 5 edits in trunk
WebCore: * bindings/v8/custom/V8WebGLRenderingContextCustom.cpp:
(WebCore::CALLBACK_FUNC_DECL):
- platform/graphics/GraphicsContext3D.h:
Reviewed by NOBODY (Chromium build fix).
WebKit/chromium: * src/GraphicsContext3D.cpp:
(WebCore::GraphicsContext3D::createShader):
Reviewed by NOBODY (Chromium build fix).
- 6:30 PM Changeset in webkit [51334] by
-
- 4 edits in trunk/JavaScriptCore
Streamlined some Math functions where we expect or know the result not
to be representable as an int.
Reviewed by Oliver Hunt.
SunSpider says 0.6% faster.
- runtime/JSNumberCell.h:
(JSC::JSValue::JSValue):
- runtime/JSValue.h:
(JSC::JSValue::):
(JSC::jsDoubleNumber):
(JSC::JSValue::JSValue): Added a function for making a numeric JSValue
and skipping the "can I encode this as an int?" check, avoiding the
overhead of int <-> double roundtripping and double <-> double comparison
and branching.
- runtime/MathObject.cpp:
(JSC::mathProtoFuncACos):
(JSC::mathProtoFuncASin):
(JSC::mathProtoFuncATan):
(JSC::mathProtoFuncATan2):
(JSC::mathProtoFuncCos):
(JSC::mathProtoFuncExp):
(JSC::mathProtoFuncLog):
(JSC::mathProtoFuncRandom):
(JSC::mathProtoFuncSin):
(JSC::mathProtoFuncSqrt):
(JSC::mathProtoFuncTan): For these functions, which we expect or know
to produce results not representable as ints, call jsDoubleNumber instead
of jsNumber.
- 6:24 PM Changeset in webkit [51333] by
-
- 3 edits in trunk/JavaScriptCore
Unreviewed. Unbreak the regression tests after r51329.
- API/JSBase.cpp:
(JSEvaluateScript): Null-check clientData before dereferencing it.
- API/JSObjectRef.cpp:
(JSObjectCallAsFunction): Ditto.
- 5:31 PM Changeset in webkit [51332] by
-
- 1 edit in trunk/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def
Add exported functions to fix the Windows build.
- 5:07 PM Changeset in webkit [51331] by
-
- 4 edits in trunk/WebCore
Prevent ResourceHandleMac's version of ResourceHandle::receivedCredential from stripping
identity and certificate information from a WebCore::Credential when receivedCredential
needs to modify the credential's persistence.
Reviewed by Alexey Proskuryakov.
- 4:55 PM Changeset in webkit [51330] by
-
- 10 edits in trunk/WebCore
Part 2/3 of <rdar://problem/7377477> REGRESSION: Many web pages fail to render after interesting script runs in isolated world
Reviewed by Geoff Garen.
Some clients of the JavaScriptCore API expect to be able to make callbacks over the JSC API,
and for this to automagically cause execution to take place in the world associated with the
global object associated with the ExecState (JSContextRef) passed. However this is not how
things work - the world must be explicitly set within WebCore.
Making this work just for API calls to evaluate & call will be a far from perfect solution,
since direct (non-API) use of JSC still relies on WebCore setting the current world correctly.
A better solution would be to make this all work automagically all throughout WebCore, but this
will require more refactoring.
Add references from the JSDOMWindowShell and the JSDOMGlobalObject to the world that owns them,
so that we can get to the world from the lexical global object of an ExecState. In the long-term
we should switch over to using this approach for all cases we want to get a world from an exec state.
- bindings/js/JSDOMBinding.cpp:
(WebCore::WebCoreJSClientData::beginningExecution):
(WebCore::WebCoreJSClientData::completedExecution):
- bindings/js/JSDOMBinding.h:
- bindings/js/JSDOMGlobalObject.h:
(WebCore::JSDOMGlobalObject::world):
(WebCore::JSDOMGlobalObject::JSDOMGlobalObjectData::JSDOMGlobalObjectData):
- bindings/js/JSDOMWindowBase.cpp:
(WebCore::JSDOMWindowBase::JSDOMWindowBaseData::JSDOMWindowBaseData):
- bindings/js/JSDOMWindowBase.h:
- bindings/js/JSDOMWindowShell.cpp:
(WebCore::JSDOMWindowShell::JSDOMWindowShell):
- bindings/js/JSDOMWindowShell.h:
(WebCore::JSDOMWindowShell::world):
- bindings/js/JSWorkerContextBase.cpp:
(WebCore::JSWorkerContextBase::JSWorkerContextBase):
- bindings/js/ScriptController.cpp:
(WebCore::ScriptController::initScript):
- 4:54 PM Changeset in webkit [51329] by
-
- 6 edits in trunk/JavaScriptCore
Part 1/3 of <rdar://problem/7377477> REGRESSION: Many web pages fail to render after interesting script runs in isolated world
Reviewed by Geoff Garen.
Some clients of the JavaScriptCore API expect to be able to make callbacks over the JSC API,
and for this to automagically cause execution to take place in the world associated with the
global object associated with the ExecState (JSContextRef) passed. However this is not how
things work - the world must be explicitly set within WebCore.
Making this work just for API calls to evaluate & call will be a far from perfect solution,
since direct (non-API) use of JSC still relies on WebCore setting the current world correctly.
A better solution would be to make this all work automagically all throughout WebCore, but this
will require more refactoring.
Since the API is in JSC but worlds only exist in WebCore, add callbacks on the JSGlobalData::ClientData
to allow it to update the current world on entry/exit via the JSC API. This is temporary duck
tape, and should be removed once the current world no longer needs to be explicitly tracked.
- API/JSBase.cpp:
(JSEvaluateScript):
- API/JSObjectRef.cpp:
(JSObjectCallAsFunction):
- JavaScriptCore.exp:
- runtime/JSGlobalData.cpp:
(JSC::JSGlobalData::ClientData::beginningExecution):
(JSC::JSGlobalData::ClientData::completedExecution):
- runtime/JSGlobalData.h:
- 4:35 PM Changeset in webkit [51328] by
-
- 1 edit in trunk/WebCore/html/canvas/WebGLRenderingContext.cpp
Forgot one file for https://bugs.webkit.org/show_bug.cgi?id=31239
- 4:33 PM Changeset in webkit [51327] by
-
- 15 edits4 adds in trunk
Add range checks to rendering calls in WebGL
https://bugs.webkit.org/show_bug.cgi?id=31239
I am now tracking the size of the data in each CanvasBuffer object
and keeping track of the buffer size of each active vertex attrib.
In drawArrays and drawElements I make sure no attempt is made to
access elements outside the valid buffer ranges. The test at:
http://cs.helsinki.fi/u/ilmarihe/c3d/functions/drawArraysOutOfBounds.html
no longer crashes.
I also added all the WebGL enumerations to GraphicsContext3D to use them in the validation checks
- 4:30 PM Changeset in webkit [51326] by
-
- 6 edits10 adds in trunk
2009-11-23 Erik Arvidsson <arv@chromium.org>
Reviewed by Eric Seidel.
JSC bindings for HasIndexGetter generates incorrect code (affects
MediaList and CSSStyleDeclaration).
This cleans up the edge cases for indexing out of range for style and
computed style objects to return an empty string according to the spec.
MediaList now returns null when indexed out of range.
https://bugs.webkit.org/show_bug.cgi?id=31683
- fast/dom/CSSStyleDeclaration/css-computed-style-item-expected.txt: Added.
- fast/dom/CSSStyleDeclaration/css-computed-style-item.html: Added.
- fast/dom/CSSStyleDeclaration/css-style-item-expected.txt: Added.
- fast/dom/CSSStyleDeclaration/css-style-item.html: Added.
- fast/dom/CSSStyleDeclaration/script-tests/css-computed-style-item.js: Added.
- fast/dom/CSSStyleDeclaration/script-tests/css-style-item.js: Added.
- fast/dom/StyleSheet/css-medialist-item-expected.txt: Added.
- fast/dom/StyleSheet/css-medialist-item.html: Added.
- fast/dom/StyleSheet/script-tests/TEMPLATE.html: Copied from LayoutTests/fast/backgrounds/repeat/script-tests/TEMPLATE.html.
- fast/dom/StyleSheet/script-tests/css-medialist-item.js: Added.
2009-11-23 Erik Arvidsson <arv@chromium.org>
Reviewed by Eric Seidel.
JSC bindings for HasIndexGetter generates incorrect code (affects
MediaList and CSSStyleDeclaration).
This cleans up the edge cases for indexing out of range for style and
computed style objects to return an empty string according to the spec.
MediaList now returns null when indexed out of range.
https://bugs.webkit.org/show_bug.cgi?id=31683
Tests: fast/dom/CSSStyleDeclaration/css-computed-style-item.html
fast/dom/CSSStyleDeclaration/css-style-item.html
fast/dom/StyleSheet/css-medialist-item.html
- bindings/scripts/CodeGeneratorJS.pm: If IndexGetterReturnsString then
we do not check the length before calling the item function.
- css/CSSComputedStyleDeclaration.cpp: (WebCore::CSSComputedStyleDeclaration::item):
- css/CSSMutableStyleDeclaration.cpp: (WebCore::CSSMutableStyleDeclaration::item):
- css/CSSStyleDeclaration.idl:
- 4:16 PM Changeset in webkit [51325] by
-
- 2 edits in trunk/LayoutTests
Unreviewed. Skip new test added in r51324.
- platform/qt/Skipped: Added editing/selection/trransformed-selection-rects.html
- 3:30 PM Changeset in webkit [51324] by
-
- 15 edits4 adds in trunk
2009-11-23 Simon Fraser <Simon Fraser>
Reviewed by Dan Bernstein.
DocumentMarkers need to be educated about transforms
https://bugs.webkit.org/show_bug.cgi?id=31751
Find highlight is incorrect with transforms
<rdar://problem/6358394>
Allow callers to specify that Frame::selectionTextRects() takes transforms into account
when computing the set of rects that encompass a selection. For transformed elemenets, the
selection rect will be the bounding box of the selected content.
Fix DocumentMarkers to cache rects in absolute coordinates, rather than painting coordinates.
Test: editing/selection/transformed-selection-rects.html
- WebCore.base.exp: Frame::selectionTextRects() has a new parameter.
- dom/Document.cpp: (WebCore::Document::setRenderedRectForMarker):
- dom/Document.h: Pass the marker as a const reference.
- dom/Range.h:
- dom/Range.cpp: (WebCore::Range::textQuads): Add a new method, textQuads(), which returns a list of quads, respecting transforms.
- page/Frame.h:
- page/Frame.cpp: (WebCore::Frame::selectionTextRects): Add a new parameter, respectTransforms, and when that is RespectTransforms, use the quad method to get quads for ranges, and then take their bounding boxes.
- rendering/InlineTextBox.h:
- rendering/InlineTextBox.cpp: (WebCore::InlineTextBox::paintSpellingOrGrammarMarker): (WebCore::InlineTextBox::paintTextMatchMarker): (WebCore::InlineTextBox::computeRectForReplacementMarker): (WebCore::InlineTextBox::paintDocumentMarkers): (WebCore::InlineTextBox::textPos): (WebCore::InlineTextBox::offsetForPosition):
Pass DocumentMarkers as a const references.
Convert the argument to setRenderedRectForMarker() into absolute coordinates.
- rendering/RenderView.cpp: (WebCore::RenderView::selectionBounds):
- 2:55 PM Changeset in webkit [51323] by
-
- 1 edit2 adds in trunk/JavaScriptCore
2009-11-23 Steve Block <steveblock@google.com>
Reviewed by Dmitry Titov.
Adds MainThreadAndroid.cpp with Android-specific WTF threading functions.
https://bugs.webkit.org/show_bug.cgi?id=31807
- wtf/android: Added.
- wtf/android/MainThreadAndroid.cpp: Added. (WTF::timeoutFired): (WTF::initializeMainThreadPlatform): (WTF::scheduleDispatchFunctionsOnMainThread):
- 2:36 PM Changeset in webkit [51322] by
-
- 2 edits in trunk/WebCore
2009-11-23 Dirk Schulze <krit@webkit.org>
Reviewed by Oliver Hunt.
[Cairo] support blurred test-shadow
https://bugs.webkit.org/show_bug.cgi?id=31797
Support for blurred text-shadows on Cairo. This patch
reuses the code of blurred box-shadows, introduced in
bug 26102. For a full textshadow support, a filters enabled
build is needed.
- platform/graphics/cairo/FontCairo.cpp: (WebCore::Font::drawGlyphs):
- 2:25 PM Changeset in webkit [51321] by
-
- 6 edits in trunk/WebCore
Change incorrect calls to the constructor "EventNames()" to the correct accessor
"eventNames()". This saves ~100 AtomicString lookups each time.
https://bugs.webkit.org/show_bug.cgi?id=31811
Reviewed by Geoffrey Garen.
- dom/EventNames.h: Make constructor private to prevent this from happening again.
- history/CachedFrame.cpp:
(WebCore::CachedFrameBase::restore): EventNames() --> eventNames()
- html/HTMLFormControlElement.cpp:
(WebCore::HTMLFormControlElement::checkValidity): EventNames() --> eventNames()
- loader/FrameLoader.cpp:
(WebCore::FrameLoader::stopLoading): EventNames() --> eventNames()
(WebCore::FrameLoader::pageHidden): EventNames() --> eventNames()
- page/DOMWindow.cpp:
(WebCore::DOMWindow::dispatchAllPendingUnloadEvents): EventNames() --> eventNames()
- 2:22 PM Changeset in webkit [51320] by
-
- 2 edits in trunk/WebKitTools
Reviewed by Oliver Hunt.
Implement new required function to pass test we used to pass. This
change is required since r51294.
- DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::setAllowUniversalAccessFromFileURLs):
- 2:10 PM Changeset in webkit [51319] by
-
- 2 edits in trunk/WebCore
2009-11-23 Adam Langley <agl@google.com>
Reviewed by Dmitry Titov.
Chromium Linux: Limit the stroke width and mitre limit.
Limit the stroke width and mitre limit that we'll pass into Skia to
avoid overflowing Skia's uint16_t glyph widths.
http://code.google.com/p/chromium/issues/detail?id=28250
https://bugs.webkit.org/show_bug.cgi?id=31747
- platform/graphics/skia/PlatformContextSkia.cpp: (scalarBound): (PlatformContextSkia::setupPaintForStroking):
- 1:57 PM Changeset in webkit [51318] by
-
- 2 edits in trunk/LayoutTests
Unreviewed. Skip new tests.
- platform/qt/Skipped:
- 1:53 PM Changeset in webkit [51317] by
-
- 3 edits in trunk/WebCore
Reviewed by Oliver Hunt.
https://bugs.webkit.org/show_bug.cgi?id=31812
WebSocket code uses RefPtr::get() where it shouldn't
No change in funcitonality, just coding style correction.
- websockets/WebSocket.cpp: (WebCore::WebSocket::~WebSocket):
- websockets/WebSocketChannel.cpp: (WebCore::WebSocketChannel::connect): (WebCore::WebSocketChannel::send): (WebCore::WebSocketChannel::bufferedAmount): (WebCore::WebSocketChannel::close): (WebCore::WebSocketChannel::disconnect): (WebCore::WebSocketChannel::didOpen): (WebCore::WebSocketChannel::didClose): (WebCore::WebSocketChannel::didReceiveData): (WebCore::WebSocketChannel::didFail):
- 1:48 PM Changeset in webkit [51316] by
-
- 4 edits in trunk
2009-11-23 Jakub Wieczorek <faw217@gmail.com>
Reviewed by Kenneth Rohde Christiansen.
[Qt] DRT: dumpBackForwardList() does not work properly with non-file URLs.
https://bugs.webkit.org/show_bug.cgi?id=31775
LayoutTestController::dumpBackForwardList() should work with local URLs
as well as with normal URLs (in http tests for instance).
Currently it does not output the latter properly.
Unskip a bunch of passing http/navigation tests.
- platform/qt/Skipped:
2009-11-23 Jakub Wieczorek <faw217@gmail.com>
Reviewed by Kenneth Rohde Christiansen.
[Qt] DRT: dumpBackForwardList() does not work properly with non-file URLs.
https://bugs.webkit.org/show_bug.cgi?id=31775
LayoutTestController::dumpBackForwardList() should work with local URLs
as well as with normal URLs (in http tests for instance).
Currently it does not output the latter properly.
Unskip a bunch of passing http/navigation tests.
- DumpRenderTree/qt/DumpRenderTree.cpp: (WebCore::dumpHistoryItem):
- 1:28 PM Changeset in webkit [51315] by
-
- 2 edits in trunk/WebCore
Build fix.
- platform/network/cf/SocketStreamHandleCFNet.cpp: (WebCore::SocketStreamHandle::pacExecutionCallbackMainThread): pacExecutionCallbackMainThread is static, so it can't use member variables directly.
- 1:00 PM Changeset in webkit [51314] by
-
- 5 edits in trunk
Reviewed by Brady Eidson.
https://bugs.webkit.org/show_bug.cgi?id=31748
Make WebSocketHandleCFNet respect proxy auto-configuration files via CFProxySupport
- platform/network/cf/SocketStreamHandle.h: Removed names from some void* arguments, since they didn't carry useful information.
- platform/network/cf/SocketStreamHandleCFNet.cpp: (WebCore::SocketStreamHandle::SocketStreamHandle): When we need an http-style URL, we actually always need https. (WebCore::SocketStreamHandle::scheduleStreams): Factored out from constructor, since streams only get scheduled after PAC is fetched and executed asynchronously. (WebCore::SocketStreamHandle::copyPACExecutionDescription): Return a description for event source. (WebCore::MainThreadPACCallbackInfo::MainThreadPACCallbackInfo): Forward callback to main thread. (WebCore::SocketStreamHandle::pacExecutionCallback): Ditto. (WebCore::SocketStreamHandle::pacExecutionCallbackMainThread): Ditto. To avoid code duplication, we make the call even on Mac. (WebCore::SocketStreamHandle::executePACFileURL): Make an async call to CFNetworkExecuteProxyAutoConfigurationURL. (WebCore::SocketStreamHandle::removePACRunLoopSource): Once PAC execution is done or aborted, we need to get rid of the event source. (WebCore::SocketStreamHandle::chooseProxy): Use stored m_httpsURL.get. (WebCore::SocketStreamHandle::chooseProxyFromArray): Factored out from chooseProxy - the array may come directly from system configuration, or from PAC. (WebCore::SocketStreamHandle::chooseProxy): Tiger version of this function is now completely separate. (WebCore::SocketStreamHandle::~SocketStreamHandle): Run loop source should be destroyed before we get to the destructor. (WebCore::SocketStreamHandle::platformClose): Destroy the run loop source, if PAC execution is still in progress.
- 12:10 PM Changeset in webkit [51313] by
-
- 2 edits in trunk/WebCore
Don't leak the CGImage we create when drawing a sub image.
Reviewed by Geoff Garen.
- 11:12 AM Changeset in webkit [51312] by
-
- 3 edits in trunk/WebCore
2009-11-23 Adam Barth <abarth@webkit.org>
Reviewed by Dimitri Glazkov.
[V8] Don't crash when OOM in creating isolated world
https://bugs.webkit.org/show_bug.cgi?id=31805
We need to add some more null checks to avoid crashing. No new tests
because we don't have a good way to test out-of-memory bugs.
- bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::evaluateInIsolatedWorld): (WebCore::V8Proxy::evaluateInNewContext): (WebCore::V8Proxy::setInjectedScriptContextDebugId):
- bindings/v8/V8Proxy.h:
- 11:07 AM Changeset in webkit [51311] by
-
- 1 edit in trunk/WebKit/win/Interfaces/WebKit.idl
Touch WebKit.idl to force a rebuild.
- 8:15 AM Changeset in webkit [51310] by
-
- 28 edits in trunk/WebCore
2009-11-23 Dirk Schulze <krit@webkit.org>
Reviewed by Nikolas Zimmermann.
This is the implementation of the filterRes attribute. It
helps the SVG developer to set the quality of a filter by
giving the width or height of filter.
This patch also sets the filter resolution to lower values
if a intermediate ImageBuffer size is bigger than the given
maximal size.
The maximal size is set to 5000x5000 by default. This is a
subjectiv decission. Everthing greater than this values gets
sensible slower. Values of 10000x10000 crashed on WebKitGtk.
For mobil devices a maximum size of 100x100 or 200x200 seems
to be reasonable.
The important fact on filter resolution is, that the output
size is still the size given by the <filter> element.
Tests: svg/filters/big-sized-filter-2.svg
svg/filters/big-sized-filter.svg
svg/filters/filterRes.svg
- platform/graphics/FloatRect.cpp: (WebCore::FloatRect::scale): Add the abbility to scale a rect by x and y.
- platform/graphics/FloatRect.h: (WebCore::FloatRect::scale): Add the abbility to scale a rect by x and y.
- platform/graphics/cairo/GraphicsContextCairo.cpp: (WebCore::GraphicsContext::createPlatformShadow): Use scaledSubRegion for calculation.
- platform/graphics/filters/FEBlend.cpp: (WebCore::FEBlend::apply): Use scaledSubRegion for effect intern calculations.
- platform/graphics/filters/FEColorMatrix.cpp: (WebCore::FEColorMatrix::apply): Use scaledSubRegion for effect intern calculations.
- platform/graphics/filters/FEComponentTransfer.cpp: (WebCore::FEComponentTransfer::apply): Use scaledSubRegion for effect intern calculations.
- platform/graphics/filters/FEComposite.cpp: (WebCore::FEComposite::apply): Use scaledSubRegion for effect intern calculations.
- platform/graphics/filters/FEGaussianBlur.cpp: (WebCore::FEGaussianBlur::apply): Use scaledSubRegion for effect intern calculations.
- platform/graphics/filters/Filter.h: Add the abbility to change the quality of a filter output. (WebCore::Filter::filterResolution): (WebCore::Filter::setFilterResolution): (WebCore::Filter::calculateEffectSubRegion): Calculates the correct subRegion as well as the scaledSubRegion. It also searches for the biggest effect size. We have to change the filter resolution, if one intermediate ImageBuffer size doesn't fit in the maximal image size.
- platform/graphics/filters/FilterEffect.cpp: (WebCore::FilterEffect::calculateDrawingIntRect): Use scaledSubRegion to get the right part of a previous effect result. (WebCore::FilterEffect::calculateDrawingRect): Use scaledSubRegion to get the right part of a previous effect result. (WebCore::FilterEffect::getEffectContext): Use scaledSubRegion to create a new intermediate ImageBuffer for the result of the current effect.
- platform/graphics/filters/FilterEffect.h: (WebCore::FilterEffect::scaledSubRegion): The scaled subRegion of a the filter effect. (WebCore::FilterEffect::setScaledSubRegion): (WebCore::FilterEffect::effectBoundaries): The original values of the EffectElement for a second subRegion calculation. (WebCore::FilterEffect::setEffectBoundaries):
- platform/graphics/filters/ImageBufferFilter.cpp: (WebCore::ImageBufferFilter::ImageBufferFilter): Set the scale factor to one.
- platform/graphics/filters/ImageBufferFilter.h: (WebCore::ImageBufferFilter::maxImageSize): (WebCore::ImageBufferFilter::calculateEffectSubRegion):
- platform/graphics/filters/SourceAlpha.cpp: (WebCore::SourceAlpha::calculateEffectRect): Use scaledSubRegion for effect intern calculations.
- platform/graphics/filters/SourceGraphic.cpp: (WebCore::SourceGraphic::calculateEffectRect): Use scaledSubRegion for effect intern calculations.
- svg/SVGFilterElement.cpp: (WebCore::SVGFilterElement::parseMappedAttribute): Parse filterRes attribute. (WebCore::SVGFilterElement::buildFilter): Give SVGResourceFilter the current filterResolution.
- svg/SVGFilterPrimitiveStandardAttributes.cpp: (WebCore::SVGFilterPrimitiveStandardAttributes::setStandardAttributes): Save values to effectBoundaries of the filter effect
- svg/graphics/SVGResourceFilter.cpp: (WebCore::SVGResourceFilter::SVGResourceFilter): (WebCore::shouldProcessFilter): Return signal if a neccessary value is zero. (WebCore::SVGResourceFilter::fitsInMaximumImageSize): Checks if the given size fits into the maximal image size, modifys scale factors if not and return a bool: fits. (WebCore::SVGResourceFilter::prepareFilter): Scale the SourceImage to filterResolution (given by FilterElement or calculated on to big image sizes). Set the scale level to SVGFilter. (WebCore::SVGResourceFilter::applyFilter): Don't apply filters if shouldProcessFilter is wrong.
- svg/graphics/SVGResourceFilter.h: (WebCore::SVGResourceFilter::setFilterResolution): FilterResolution of FilterElement. (WebCore::SVGResourceFilter::setHasFilterResolution): Does FilterElement provides a FilterResolution? (WebCore::SVGResourceFilter::scaleX): Current scale factor for horizontal. (WebCore::SVGResourceFilter::scaleY): Current scale factor for vertical.
- svg/graphics/filters/SVGFEDisplacementMap.cpp: (WebCore::FEDisplacementMap::apply): Use scaledSubRegion for effect intern calculations. Kernel values are scaled to current filter resolution too.
- svg/graphics/filters/SVGFEFlood.cpp: (WebCore::FEFlood::apply): Use scaledSubRegion for effect intern calculations.
- svg/graphics/filters/SVGFEMerge.cpp: (WebCore::FEMerge::apply): Use scaledSubRegion for effect intern calculations. Kernel values are scaled to current filter resolution too.
- svg/graphics/filters/SVGFEMorphology.cpp: (WebCore::FEMorphology::apply): Use scaledSubRegion for effect intern calculations. Kernel values are scaled to current filter resolution too.
- svg/graphics/filters/SVGFEOffset.cpp: (WebCore::FEOffset::apply): Use scaledSubRegion for effect intern calculations.
- svg/graphics/filters/SVGFETile.cpp: (WebCore::FETile::apply): Use scaledSubRegion for effect intern calculations.
- svg/graphics/filters/SVGFilter.cpp: (WebCore::SVGFilter::calculateEffectSubRegion): Calculate subRegion for LayoutTests, scaledSubRegion according to the current filterResolution and get the maximal image size.
- svg/graphics/filters/SVGFilter.h: (WebCore::SVGFilter::effectBoundingBoxMode): Original values of the FilterElement. (WebCore::SVGFilter::filterRegion): Use virtual for clarification. (WebCore::SVGFilter::sourceImageRect): Use virtual for clarification. (WebCore::SVGFilter::maxImageSize): Get the maximal image size.
- 8:05 AM Changeset in webkit [51309] by
-
- 2 edits in trunk/WebKit/mac
Reviewed by Adam Roben.
<rdar://problem/7401503>
Added a workaround for plug-ins not drawing immediately.
- Plugins/WebBaseNetscapePluginView.mm: Added new constant. (-[WebBaseNetscapePluginView _clearSubstituteImage]): Added. (-[WebBaseNetscapePluginView resumeFromHalt]): Call above new method.
- 7:27 AM QtWebKitBackportingFixes edited by
- (diff)
- 7:17 AM Changeset in webkit [51308] by
-
- 4 edits in trunk/WebKit/qt
2009-11-23 David Boddie <dboddie@trolltech.com>
Reviewed by Simon Hausmann.
Updated the QWebElement documentation with links to examples and
external resources.
Fixed the project file for the webelement snippet and tidied up the
markers used for quoting the code.
- Api/qwebelement.cpp:
- docs/webkitsnippets/webelement/main.cpp: (findAll):
- docs/webkitsnippets/webelement/webelement.pro:
- 7:02 AM QtBackLog edited by
- Mention QtPlugin bug (diff)
- 7:00 AM QtWebKitPerformanceToolBackLog: edited by
- (diff)
- 6:37 AM QtWebKitJournal edited by
- (diff)
- 6:31 AM QtWebKitJournal edited by
- (diff)
- 6:29 AM QtWebKitJournal edited by
- (diff)
- 6:29 AM QtWebKitJournal edited by
- (diff)
- 6:25 AM QtWebKitJournal edited by
- (diff)
- 5:04 AM QtWebKitBackportingFixes edited by
- (diff)
- 5:02 AM Changeset in webkit [51307] by
-
- 2 edits in trunk/JavaScriptCore
[Symbian] Fix lastIndexOf() for Symbian
https://bugs.webkit.org/show_bug.cgi?id=31773
Patch by Laszlo Gombos <Laszlo Gombos> on 2009-11-23
Reviewed by Kenneth Rohde Christiansen.
Symbian soft floating point library has problems with operators
comparing NaN to numbers. Without a workaround lastIndexOf()
function does not work.
Patch developed by David Leong.
- runtime/StringPrototype.cpp:
(JSC::stringProtoFuncLastIndexOf):Add an extra test
to check for NaN for Symbian.
- 4:58 AM Changeset in webkit [51306] by
-
- 5 edits in trunk
2009-11-23 Simon Hausmann <simon.hausmann@nokia.com>
Reviewed by Kenneth Rohde Christiansen.
[Qt] Wrong runtime instance objects of wrapped QObjects may be used if
the wrapped object died before the gc removed the instance.
Before using a cached instance, verify that its wrapped QObject is
still alive.
- bridge/qt/qt_instance.cpp: (JSC::Bindings::QtInstance::getQtInstance):
- bridge/qt/qt_instance.h: (JSC::Bindings::QtInstance::hashKey):
2009-11-23 Simon Hausmann <simon.hausmann@nokia.com>
Reviewed by Kenneth Rohde Christiansen.
[Qt] Wrong runtime instance objects of wrapped QObjects may be used if
the wrapped object died before the gc removed the instance.
Added a unit-test to verify that wrapping a QObject with the
same identity as a previously but now dead object works.
- tests/qwebframe/tst_qwebframe.cpp:
- 3:29 AM Changeset in webkit [51305] by
-
- 2 edits in trunk/JavaScriptCore
Android port lacks implementation of atomicIncrement and atomicDecrement
Nov 22, 2009:
- 10:21 PM Changeset in webkit [51304] by
-
- 12 edits2 adds in trunk
2009-11-22 Chris Fleizach <Chris Fleizach>
Reviewed by Oliver Hunt.
ARIA: support aria-flowto
https://bugs.webkit.org/show_bug.cgi?id=31762
- platform/mac/accessibility/aria-flowto-expected.txt: Added.
- platform/mac/accessibility/aria-flowto.html: Added.
2009-11-22 Chris Fleizach <Chris Fleizach>
Reviewed by Oliver Hunt.
ARIA: support aria-flowto
https://bugs.webkit.org/show_bug.cgi?id=31762
Test: platform/mac/accessibility/aria-flowto.html
- accessibility/AccessibilityObject.h: (WebCore::AccessibilityObject::ariaOwnsElements): (WebCore::AccessibilityObject::supportsARIAFlowTo): (WebCore::AccessibilityObject::ariaFlowToElements):
- accessibility/AccessibilityRenderObject.cpp: (WebCore::AccessibilityRenderObject::linkedUIElements): (WebCore::AccessibilityRenderObject::supportsARIAFlowTo): (WebCore::AccessibilityRenderObject::ariaFlowToElements):
- accessibility/AccessibilityRenderObject.h:
- html/HTMLAttributeNames.in:
2009-11-22 Chris Fleizach <Chris Fleizach>
Reviewed by Oliver Hunt.
ARIA: support aria-flowto
https://bugs.webkit.org/show_bug.cgi?id=31762
- DumpRenderTree/AccessibilityUIElement.cpp: (ariaFlowToElementAtIndexCallback): (AccessibilityUIElement::getJSClass):
- DumpRenderTree/AccessibilityUIElement.h:
- DumpRenderTree/gtk/AccessibilityUIElementGtk.cpp: (AccessibilityUIElement::ariaFlowToElementAtIndex):
- DumpRenderTree/mac/AccessibilityUIElementMac.mm: (AccessibilityUIElement::ariaFlowToElementAtIndex):
- DumpRenderTree/win/AccessibilityUIElementWin.cpp: (AccessibilityUIElement::ariaFlowToElementAtIndex):
- 8:27 PM Changeset in webkit [51303] by
-
- 4 edits in trunk/WebKitLibraries
Updated WebKitSystemInterface.
Reviewed by Eric Carlson.
- libWebKitSystemInterfaceLeopard.a:
- libWebKitSystemInterfaceSnowLeopard.a:
- libWebKitSystemInterfaceTiger.a:
- 7:45 PM Changeset in webkit [51302] by
-
- 2 edits in trunk/JavaScriptCore
2009-11-22 Laszlo Gombos <Laszlo Gombos>
Unreviewed.
[Qt] Sort source lists and remove obsolete comments
from the build system.
- JavaScriptCore.pri:
- 5:38 PM HighLevelOverview edited by
- Added information about JavaScriptGlue (diff)
- 5:33 PM Changeset in webkit [51301] by
-
- 1 edit10 adds in trunk/LayoutTests
2009-11-22 Dirk Schulze <krit@webkit.org>
Reviewed by Nikolas Zimmermann.
Tests to the filterRes implementation for SVG Filters
https://bugs.webkit.org/show_bug.cgi?id=6021
big-sized-filter: tests if filter quality is scaled down on a too
big SourceImage (SourceGraphic, SourceAlpha)
big-sized-filter-2: tests if the filter quality is scaled down
on a too big filter effect size.
filterRes: Various tests of SVG Filters with different values for
filterRes attributes. Checks
- that kernel values (if present) are scaled with the filterRes
- subRegion calculation is correct for effects with deeper dependencies.
- platform/mac/svg/filters/big-sized-filter-2-expected.txt: Added.
- platform/mac/svg/filters/big-sized-filter-expected.checksum: Added.
- platform/mac/svg/filters/big-sized-filter-expected.png: Added.
- platform/mac/svg/filters/big-sized-filter-expected.txt: Added.
- platform/mac/svg/filters/filterRes-expected.checksum: Added.
- platform/mac/svg/filters/filterRes-expected.png: Added.
- platform/mac/svg/filters/filterRes-expected.txt: Added.
- svg/filters/big-sized-filter-2.svg: Added.
- svg/filters/big-sized-filter.svg: Added.
- svg/filters/filterRes.svg: Added.
- 5:29 PM Changeset in webkit [51300] by
-
- 3 edits in trunk/WebCore
Not reviewed. Build fix for WML enabled builds.
Adopt WebCore API changes.
- 3:14 PM Changeset in webkit [51299] by
-
- 2 edits in trunk/WebKitTools
[Qt] fast/history/back-forward-reset-after-error-handling.html failing due to WorkQueue not being un-frozen
https://bugs.webkit.org/show_bug.cgi?id=31638
Reviewed by Kenneth Christiansen.
Unfreeze WorkQueue after each test execution.
- DumpRenderTree/qt/DumpRenderTree.cpp:
(WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):
- 2:29 PM Changeset in webkit [51298] by
-
- 5 edits in trunk
2009-11-22 Jakub Wieczorek <faw217@gmail.com>
Reviewed by Adam Barth.
[Qt] DumpRenderTree should explicitly ignore any SSL certificate errors
for localhost and 127.0.0.1.
https://bugs.webkit.org/show_bug.cgi?id=31783
Unskip the http/tests/ssl/verify-ssl-enabled.php test, which is passing now.
- platform/qt/Skipped:
2009-11-22 Jakub Wieczorek <faw217@gmail.com>
Reviewed by Adam Barth.
[Qt] DumpRenderTree should explicitly ignore any SSL certificate errors
for localhost and 127.0.0.1.
https://bugs.webkit.org/show_bug.cgi?id=31783
Unskip the http/tests/ssl/verify-ssl-enabled.php test, which is passing now.
- DumpRenderTree/qt/DumpRenderTree.cpp: (WebCore::NetworkAccessManager::NetworkAccessManager): (WebCore::NetworkAccessManager::sslErrorsEncountered): (WebCore::WebPage::WebPage):
- DumpRenderTree/qt/DumpRenderTree.h:
- 12:54 PM BuildingQtOnLinux edited by
- libxslt is no longer required (diff)
- 12:05 PM Changeset in webkit [51297] by
-
- 3 edits in trunk/WebCore
2009-11-22 Pavel Feldman <pfeldman@chromium.org>
Reviewed by Timothy Hatcher.
Web Inspector: Introduce sidebar background on timeline panel in order
to prevent it from flickering on scroll.
- inspector/front-end/TimelinePanel.js: (WebInspector.TimelinePanel): (WebInspector.TimelinePanel.prototype.setSidebarWidth): (WebInspector.TimelinePanel.prototype._onScroll): (WebInspector.TimelinePanel.prototype._scheduleRefresh):
- inspector/front-end/inspector.css:
- 10:25 AM Changeset in webkit [51296] by
-
- 9 edits2 adds in trunk
2009-11-22 Pavel Feldman <pfeldman@chromium.org>
Reviewed by Timothy Hatcher.
Web Inspector: Reimplement TimelinePanel to make it fast:
- Extract grid and overview into separate files
- Make timeline create only divs for visible rows
- WebCore.gypi:
- WebCore.vcproj/WebCore.vcproj:
- inspector/front-end/AbstractTimelinePanel.js: (WebInspector.AbstractTimelinePanel.prototype.createInterface): (WebInspector.AbstractTimelinePanel.prototype.refresh): (WebInspector.AbstractTimelinePanel.prototype.set calculator):
- inspector/front-end/TimelineGrid.js: Added. (WebInspector.TimelineGrid): (WebInspector.TimelineGrid.prototype.get itemsGraphsElement): (WebInspector.TimelineGrid.prototype.updateDividers): (WebInspector.TimelineGrid.prototype.addEventDivider): (WebInspector.TimelineGrid.prototype.setScrollAndDividerTop):
- inspector/front-end/TimelineOverviewPane.js: Added. (WebInspector.TimelineOverviewPane): (WebInspector.TimelineOverviewPane.prototype._onCheckboxClicked): (WebInspector.TimelineOverviewPane.prototype.update): (WebInspector.TimelineOverviewPane.prototype.setSidebarWidth): (WebInspector.TimelineOverviewPane.prototype.updateMainViewWidth): (WebInspector.TimelineOverviewPane.prototype.reset): (WebInspector.TimelineOverviewPane.prototype._resizeWindow): (WebInspector.TimelineOverviewPane.prototype._windowResizeDragging): (WebInspector.TimelineOverviewPane.prototype._dragWindow): (WebInspector.TimelineOverviewPane.prototype._windowDragging): (WebInspector.TimelineOverviewPane.prototype._resizeWindowLeft): (WebInspector.TimelineOverviewPane.prototype._resizeWindowRight): (WebInspector.TimelineOverviewPane.prototype._setWindowPosition): (WebInspector.TimelineOverviewPane.prototype._endWindowDragging): (WebInspector.TimelineOverviewCalculator): (WebInspector.TimelineOverviewCalculator.prototype.computeBarGraphPercentages): (WebInspector.TimelineOverviewCalculator.prototype.reset): (WebInspector.TimelineOverviewCalculator.prototype.updateBoundaries): (WebInspector.TimelineOverviewCalculator.prototype.get boundarySpan): (WebInspector.TimelineOverviewCalculator.prototype.formatValue): (WebInspector.TimelineCategoryTreeElement): (WebInspector.TimelineCategoryTreeElement.prototype.onattach): (WebInspector.TimelineCategoryGraph): (WebInspector.TimelineCategoryGraph.prototype.get graphElement): (WebInspector.TimelineCategoryGraph.prototype.addChunk): (WebInspector.TimelineCategoryGraph.prototype.clearChunks): (WebInspector.TimelineCategoryGraph.prototype.set dimmed):
- inspector/front-end/TimelinePanel.js: (WebInspector.TimelinePanel): (WebInspector.TimelinePanel.prototype._toggleTimelineButtonClicked): (WebInspector.TimelinePanel.prototype.addRecordToTimeline): (WebInspector.TimelinePanel.prototype._formatRecord): (WebInspector.TimelinePanel.prototype.setSidebarWidth): (WebInspector.TimelinePanel.prototype.updateMainViewWidth): (WebInspector.TimelinePanel.prototype.resize): (WebInspector.TimelinePanel.prototype.reset): (WebInspector.TimelinePanel.prototype.show): (WebInspector.TimelinePanel.prototype._onScroll): (WebInspector.TimelinePanel.prototype._scheduleRefresh): (WebInspector.TimelinePanel.prototype._refresh): (WebInspector.TimelinePanel.prototype._refreshRecords): (WebInspector.TimelinePanel.prototype._adjustScrollPosition): (WebInspector.TimelineCategory): (WebInspector.TimelineCalculator): (WebInspector.TimelineCalculator.prototype.get boundarySpan): (WebInspector.TimelineRecordListRow): (WebInspector.TimelineRecordListRow.prototype.update): (WebInspector.TimelineRecordGraphRow): (WebInspector.TimelineRecordGraphRow.prototype.update):
- inspector/front-end/WebKit.qrc:
- inspector/front-end/inspector.css:
- inspector/front-end/inspector.html:
- 9:22 AM Changeset in webkit [51295] by
-
- 3 edits in trunk
2009-11-22 Chris Evans <cevans@chromium.org>
Reviewed by Adam Barth.
Disable access to file:/// directory listings
https://bugs.webkit.org/show_bug.cgi?id=31329
Updates the test to make sure it is disabling universal access for
file:/// URLs (if the API is supported). This is required to do
effective intra-file:/// access tests.
Note that Safari only passes by virtual of supporting no directory
listings at all.
- fast/xmlhttprequest/resources/xmlhttprequest-nonexistent-file-real.html: Added.
- fast/xmlhttprequest/xmlhttprequest-nonexistent-file-expected.txt:
- fast/xmlhttprequest/xmlhttprequest-nonexistent-file.html: Disable universal access for file:/// URLs (if API available). Run the real test in an iframe without universal access.
2009-11-22 Chris Evans <cevans@chromium.org>
Reviewed by Adam Barth.
Disable access to file:/// directory listings
https://bugs.webkit.org/show_bug.cgi?id=31329
Deny access to directory listings. This needs doing in WebKit for
WebKit clients that do permit top-level navigation to file:///dir.
This matches Firefox, plus the existing Safari implementation (which
does not support any directory access at all). It will fix the
LayoutTest named below for Chromium.
Test: fast/xmlhttprequest/xmlhttprequest-nonexistent-file.html
- page/SecurityOrigin.cpp: Deny access to directory listings. (WebCore::SecurityOrigin::SecurityOrigin): (WebCore::SecurityOrigin::canRequest):
2009-11-22 Chris Evans <cevans@chromium.org>
Reviewed by Adam Barth.
Disable access to file:/// directory listings
https://bugs.webkit.org/show_bug.cgi?id=31329
Implemented setAllowUniversalAccessFromFileURLs to support testing of
file URL security.
- DumpRenderTree/LayoutTestController.cpp: (setAllowUniversalAccessFromFileURLsCallback): (LayoutTestController::staticFunctions):
- DumpRenderTree/LayoutTestController.h:
- DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::setAllowUniversalAccessFromFileURLs):
- DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::setAllowUniversalAccessFromFileURLs):
- DumpRenderTree/win/LayoutTestControllerWin.cpp: (LayoutTestController::setAllowUniversalAccessFromFileURLs):
- DumpRenderTree/wx/LayoutTestControllerWx.cpp: (LayoutTestController::setAllowUniversalAccessFromFileURLs):
- 9:22 AM Changeset in webkit [51294] by
-
- 12 edits1 copy in trunk
2009-11-22 Chris Evans <cevans@chromium.org>
Reviewed by Adam Barth.
Updates the test to make sure it is disabling universal access for
file:/// URLs (if the API is supported). This is required to do
effective intra-file:/// access tests.
Note that Safari only passes by virtual of supporting no directory
listings at all.
- fast/xmlhttprequest/resources/xmlhttprequest-nonexistent-file-real.html: Added.
- fast/xmlhttprequest/xmlhttprequest-nonexistent-file-expected.txt:
- fast/xmlhttprequest/xmlhttprequest-nonexistent-file.html: Disable universal access for file:/// URLs (if API available). Run the real test in an iframe without universal access.
2009-11-22 Chris Evans <cevans@chromium.org>
Reviewed by Adam Barth.
Deny access to directory listings. This needs doing in WebKit for
WebKit clients that do permit top-level navigation to file:///dir.
This matches Firefox, plus the existing Safari implementation (which
does not support any directory access at all). It will fix the
LayoutTest named below for Chromium.
Test: fast/xmlhttprequest/xmlhttprequest-nonexistent-file.html
- page/SecurityOrigin.cpp: Deny access to directory listings. (WebCore::SecurityOrigin::SecurityOrigin): (WebCore::SecurityOrigin::canRequest):
2009-11-22 Chris Evans <cevans@chromium.org>
Reviewed by Adam Barth.
Implemented setAllowUniversalAccessFromFileURLs to support testing of
file URL security.
- DumpRenderTree/LayoutTestController.cpp: (setAllowUniversalAccessFromFileURLsCallback): (LayoutTestController::staticFunctions):
- DumpRenderTree/LayoutTestController.h:
- DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::setAllowUniversalAccessFromFileURLs):
- DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::setAllowUniversalAccessFromFileURLs):
- DumpRenderTree/win/LayoutTestControllerWin.cpp: (LayoutTestController::setAllowUniversalAccessFromFileURLs):
- DumpRenderTree/wx/LayoutTestControllerWx.cpp: (LayoutTestController::setAllowUniversalAccessFromFileURLs):
- 7:16 AM Changeset in webkit [51293] by
-
- 2 edits in trunk/WebKitTools
2009-11-22 Jakub Wieczorek <faw217@gmail.com>
Reviewed by Kenneth Rohde Christiansen.
[Qt] Fix the timeout of fast/frames/frame-navigation.html
https://bugs.webkit.org/show_bug.cgi?id=31638
The test is timeouting, because it uses the WorkQueue to load a document in one
of the child frames and once the loading is finished, the DRT does not dump the
tree. This is because it waits for the QWebFrame::loadFinished() signal from
the main frame, while it should connect to QWebPage::loadFinished().
- DumpRenderTree/qt/DumpRenderTree.cpp: (WebCore::DumpRenderTree::DumpRenderTree):
- 2:04 AM Changeset in webkit [51292] by
-
- 4 edits in trunk/WebKitLibraries
Fixed a leak in WKDrawMediaUIPart().
Reviewed by Cameron Zwarich.
- libWebKitSystemInterfaceLeopard.a:
- libWebKitSystemInterfaceSnowLeopard.a:
- libWebKitSystemInterfaceTiger.a:
