Index: WebCore/ChangeLog |
=================================================================== |
--- WebCore/ChangeLog (revision 72934) |
+++ WebCore/ChangeLog (working copy) |
@@ -1,527 +1,3 @@ |
-2010-11-29 James Robinson <jamesr@chromium.org> |
- |
- [chromium] REGRESSION(72470): Crash in skia on some images |
- https://bugs.webkit.org/show_bug.cgi?id=50201 |
- |
- Reverts r72470 |
- |
- * platform/graphics/skia/ImageSkia.cpp: |
- (WebCore::computeResamplingMode): |
- (WebCore::drawResampledBitmap): |
- |
-2010-11-29 Dmitry Titov <dimich@chromium.org> |
- |
- Reviewed by David Levin. |
- |
- Crash when iframe transfers from one page to another and has child frames. |
- https://bugs.webkit.org/show_bug.cgi?id=50200 |
- |
- * page/Frame.cpp: |
- (WebCore::Frame::transferChildFrameToNewDocument): |
- avoid overriding 'didTransfer' which has the result of previous check. |
- |
-2010-11-19 Ryosuke Niwa <rniwa@webkit.org> |
- |
- Reviewed by David Hyatt. |
- |
- up arrow doesn't work with RTL text with word wrapping |
- https://bugs.webkit.org/show_bug.cgi?id=41987 |
- |
- The bug was caused by positionForPoint's not returning the correct affinity when the x coordiate |
- is to the left of the first line, and startOfLine's incorrectly moving position to the previous candidate. |
- |
- Fixed the bug by returning upstream VisiblePosition in positionForPoint if the first inline text box's offset |
- is greater than 0. Also removed the code added by http://trac.webkit.org/changeset/23608 since this changeset |
- does not add any test, and the problem described in the changelog reproduces regardless of the code's presence. |
- |
- Tests: editing/selection/click-left-of-rtl-wrapping-text.html |
- editing/selection/modify-up-on-rtl-wrapping-text.html |
- |
- * editing/visible_units.cpp: |
- (WebCore::startOfLine): |
- * rendering/RenderText.cpp: |
- (WebCore::RenderText::positionForPoint): |
- |
-2010-11-29 Ilya Sherman <isherman@chromium.org> |
- |
- Reviewed by Kent Tamura. |
- |
- Ensure that option elements are rendered with normal font weight on Windows/Linux |
- https://bugs.webkit.org/show_bug.cgi?id=50055 |
- |
- * WebCore.gyp/WebCore.gyp: include themeChromiumSkia.css |
- * css/themeChromiumSkia.css: Added. |
- (option): font-weight: normal !important; |
- * rendering/RenderThemeChromiumSkia.cpp: |
- (WebCore::RenderThemeChromiumSkia::extraDefaultStyleSheet): include themeChromiumSkia.css |
- |
-2010-11-29 Brent Fulgham <bfulgham@webkit.org> |
- |
- Unreviewed build fix. |
- |
- * WebCore.vcproj/WebCoreCairo.vsprops: Provide missing WIN_CAIRO |
- definition so that proper targets are build. |
- * platform/graphics/win/cairo/FontPlatformData.h: |
- (WebCore::FontPlatformData::orientation): Add missing implementation. |
- |
-2010-11-29 Chris Rogers <crogers@google.com> |
- |
- Reviewed by Kenneth Russell. |
- |
- Switch web audio code to use FloatPoint3D instead of Vector3 |
- https://bugs.webkit.org/show_bug.cgi?id=50186 |
- |
- No new tests since audio API is not yet implemented. |
- |
- * platform/audio/Cone.cpp: |
- (WebCore::ConeEffect::gain): |
- * platform/audio/Cone.h: |
- * platform/graphics/FloatPoint3D.h: |
- (WebCore::FloatPoint3D::isZero): |
- (WebCore::operator*): |
- (WebCore::FloatPoint3D::distanceTo): |
- * webaudio/AudioListener.h: |
- (WebCore::AudioListener::setPosition): |
- (WebCore::AudioListener::position): |
- (WebCore::AudioListener::setOrientation): |
- (WebCore::AudioListener::orientation): |
- (WebCore::AudioListener::setUpVector): |
- (WebCore::AudioListener::upVector): |
- (WebCore::AudioListener::setVelocity): |
- (WebCore::AudioListener::velocity): |
- * webaudio/AudioPannerNode.cpp: |
- (WebCore::AudioPannerNode::AudioPannerNode): |
- (WebCore::AudioPannerNode::getAzimuthElevation): |
- (WebCore::AudioPannerNode::dopplerRate): |
- (WebCore::AudioPannerNode::distanceConeGain): |
- * webaudio/AudioPannerNode.h: |
- (WebCore::AudioPannerNode::position): |
- (WebCore::AudioPannerNode::setPosition): |
- (WebCore::AudioPannerNode::orientation): |
- (WebCore::AudioPannerNode::setOrientation): |
- (WebCore::AudioPannerNode::velocity): |
- (WebCore::AudioPannerNode::setVelocity): |
- |
-2010-11-29 Xiaomei Ji <xji@chromium.org> |
- |
- Reviewed by David Hyatt. |
- |
- Right-to-left pages should be scrollable to reveal left overflow. |
- https://bugs.webkit.org/show_bug.cgi?id=23556 |
- |
- For RTL page, save left layout overflow and include it into the document |
- size during layout. Use the left layout overflow when scroll and paint |
- the page. Behavior on LTR page should be untouched since left layout |
- overflow is set as 0 for LTR page. |
- |
- Tests: fast/dom/horizontal-scrollbar-in-rtl-doesnt-fire-onscroll.html |
- fast/dom/horizontal-scrollbar-in-rtl.html |
- fast/dom/horizontal-scrollbar-when-dir-change.html |
- fast/dom/left-overflow-in-ltr.html |
- fast/dom/right-overflow-in-rtl.html |
- |
- * page/FrameView.cpp: |
- (WebCore::FrameView::adjustViewSize): Save negative of left layout overflow as scroll X origin. And includes left layout overflow into document size. |
- * page/mac/WebCoreFrameView.h: Add methods for setting and getting scroll X origin. |
- * platform/ScrollView.cpp: |
- (WebCore::ScrollView::ScrollView): |
- (WebCore::ScrollView::maximumScrollPosition): Minus scroll X origin from maximum horizontal scroll position. |
- (WebCore::ScrollView::minimumScrollPosition): |
- (WebCore::ScrollView::adjustScrollPositionWithinRange): |
- (WebCore::ScrollView::valueChanged): |
- (WebCore::ScrollView::setScrollPosition): |
- (WebCore::ScrollView::updateScrollbars): |
- (WebCore::ScrollView::wheelEvent): |
- * platform/ScrollView.h: |
- * platform/mac/ScrollViewMac.mm: |
- (WebCore::ScrollView::platformSetContentsSize): |
- (WebCore::ScrollView::platformSetScrollPosition): |
- (WebCore::ScrollView::platformSetScrollOriginX): |
- * rendering/RenderBox.cpp: |
- (WebCore::RenderBox::styleDidChange): |
- (WebCore::RenderBox::paintRootBoxDecorations): Include left layout overflow into canvas size. |
- * rendering/RenderView.cpp: |
- (WebCore::RenderView::layout): Save left layout overflow. |
- (WebCore::RenderView::docLeft): |
- (WebCore::RenderView::docWidth): Include left layout overflow into doc width for RTL page. |
- * rendering/RenderView.h: |
- |
-2010-11-29 Simon Fraser <simon.fraser@apple.com> |
- |
- Reviewed by Ojan Vafai. |
- |
- bunch of pixel failures after http://trac.webkit.org/changeset/72839/ |
- https://bugs.webkit.org/show_bug.cgi?id=50188 |
- |
- We need to call updateLayerTransform() for reflections, and |
- table rows as well. |
- |
- * rendering/RenderReplica.cpp: |
- (WebCore::RenderReplica::layout): |
- * rendering/RenderTableRow.cpp: |
- (WebCore::RenderTableRow::layout): |
- * rendering/RenderTableSection.cpp: |
- (WebCore::RenderTableSection::layoutRows): |
- |
-2010-11-29 Xiaomei Ji <xji@chromium.org> |
- |
- Reviewed by David Hyatt. |
- |
- Fix tab overflow problem when alignment is not left. |
- https://bugs.webkit.org/show_bug.cgi?id=25459 |
- |
- I checked WordPad in Windows, TextEdit in Mac, OpenOffice, and Google doc. |
- In all of them, (when the writing mode is horizontal) when the alignment is not |
- left alignment, the length of tab is computed based on the tab's position |
- relative to the left start of line, not based on the tab's position relative |
- to the left edge of the document. |
- |
- WebKit should follow the same. |
- |
- The m_xPos in each TextRun should be relative to the left start of the line, |
- not left start of its containing block (in right alignment and center alignment, |
- left start of the line is not the same as left start of the containing block). |
- So, when the text is drawn, the width of tab is consistent with the width |
- computed in RenderBlock::computeInlineDirectionPositionsForLine(). |
- |
- I did not test vertical writing mode, but I think the same rule should apply. |
- |
- Test: fast/dom/tab-in-right-alignment.html |
- |
- * platform/graphics/TextRun.h: |
- * rendering/InlineTextBox.cpp: |
- (WebCore::InlineTextBox::textPos): |
- * rendering/InlineTextBox.h: |
- |
-2010-11-29 Mikhail Naganov <mnaganov@chromium.org> |
- |
- Reviewed by Pavel Feldman. |
- |
- WebInspector: Request JSON-serialized heap snapshot from JS engine. |
- This simplifies heap snapshots interaction and API. Instead of |
- having objects representing snapshot entities, the whole snapshot |
- is transferred to WebInspector and parsed there. |
- |
- https://bugs.webkit.org/show_bug.cgi?id=49974 |
- |
- * bindings/js/ScriptHeapSnapshot.h: |
- (WebCore::ScriptHeapSnapshot::OutputStream::~OutputStream): |
- (WebCore::ScriptHeapSnapshot::~ScriptHeapSnapshot): |
- (WebCore::ScriptHeapSnapshot::writeJSON): |
- (WebCore::ScriptHeapSnapshot::ScriptHeapSnapshot): |
- * bindings/v8/ScriptHeapSnapshot.cpp: |
- (WebCore::ScriptHeapSnapshot::writeJSON): |
- * bindings/v8/ScriptHeapSnapshot.h: |
- (WebCore::ScriptHeapSnapshot::OutputStream::~OutputStream): |
- * inspector/Inspector.idl: |
- * inspector/InspectorProfilerAgent.cpp: |
- (WebCore::InspectorProfilerAgent::getProfile): |
- * inspector/front-end/HeapSnapshotView.js: |
- (WebInspector.HeapSnapshotEdgesIterator): |
- (WebInspector.HeapSnapshotEdgesIterator.prototype.get done): |
- (WebInspector.HeapSnapshotEdgesIterator.prototype.get isElement): |
- (WebInspector.HeapSnapshotEdgesIterator.prototype.get isHidden): |
- (WebInspector.HeapSnapshotEdgesIterator.prototype.get name): |
- (WebInspector.HeapSnapshotEdgesIterator.prototype.next): |
- (WebInspector.HeapSnapshotEdgesIterator.prototype.get node): |
- (WebInspector.HeapSnapshotEdgesIterator.prototype.get nodeIndex): |
- (WebInspector.HeapSnapshotEdgesIterator.prototype._getNameOrIndex): |
- (WebInspector.HeapSnapshotEdgesIterator.prototype._getType): |
- (WebInspector.HeapSnapshotNodeWrapper): |
- (WebInspector.HeapSnapshotNodeWrapper.prototype.get edges): |
- (WebInspector.HeapSnapshotNodeWrapper.prototype.get edgesCount): |
- (WebInspector.HeapSnapshotNodeWrapper.prototype.get instancesCount): |
- (WebInspector.HeapSnapshotNodeWrapper.prototype.get isHidden): |
- (WebInspector.HeapSnapshotNodeWrapper.prototype.get name): |
- (WebInspector.HeapSnapshotNodeWrapper.prototype.get selfSize): |
- (WebInspector.HeapSnapshotNodeWrapper.prototype._getName): |
- (WebInspector.HeapSnapshotNodeWrapper.prototype._getEdges): |
- (WebInspector.HeapSnapshotNodeWrapper.prototype._getType): |
- (WebInspector.HeapSnapshot): |
- (WebInspector.HeapSnapshot.prototype._init): |
- (WebInspector.HeapSnapshot.prototype.get rootEdges): |
- (WebInspector.HeapSnapshotView.prototype.snapshotLoaded): |
- (WebInspector.HeapSnapshotView.prototype._loadProfile.processLoadedSnapshot): |
- (WebInspector.HeapSnapshotView.prototype._loadProfile): |
- (WebInspector.HeapSnapshotView.prototype._convertSnapshot): |
- (WebInspector.HeapSnapshotView.prototype._prepareProfile.mergeRetainers): |
- (WebInspector.HeapSnapshotView.prototype._prepareProfile): |
- (WebInspector.HeapSnapshotView.prototype._sortData): |
- * inspector/front-end/ProfilesPanel.js: |
- (WebInspector.ProfilesPanel): |
- (WebInspector.ProfilesPanel.prototype.addHeapSnapshotChunk): |
- (WebInspector.ProfilesPanel.prototype.finishHeapSnapshot): |
- * inspector/front-end/inspector.js: |
- (WebInspector.addHeapSnapshotChunk): |
- (WebInspector.finishHeapSnapshot): |
- |
-2010-11-29 Simon Fraser <simon.fraser@apple.com> |
- |
- Reviewed by Dave Hyatt. |
- |
- layer()->currentTransform() is stale when layout changes the box size |
- https://bugs.webkit.org/show_bug.cgi?id=50175 |
- |
- RenderLayer::updateTransform() is fairly expensive, so we |
- have to avoid calling it from RenderBox::applyLayerTransformToRect() |
- every time. This requires that we update the layer's transform |
- during layout, instead of from updateLayerPositions() which is too late. |
- |
- So call updateLayerTransform() from the various layout methods |
- on renderers that can be transformed, and remove it from |
- updateLayerPositions(). |
- |
- Test: fast/overflow/overflow-update-transform.html |
- |
- * rendering/RenderBlock.cpp: |
- (WebCore::RenderBlock::layoutBlock): |
- (WebCore::RenderBlock::layoutOnlyPositionedObjects): |
- * rendering/RenderBox.cpp: |
- (WebCore::RenderBox::applyLayerTransformToRect): |
- (WebCore::RenderBox::updateLayerTransform): |
- * rendering/RenderBox.h: |
- * rendering/RenderEmbeddedObject.cpp: |
- (WebCore::RenderEmbeddedObject::layout): |
- * rendering/RenderFlexibleBox.cpp: |
- (WebCore::RenderFlexibleBox::layoutBlock): |
- * rendering/RenderIFrame.cpp: |
- (WebCore::RenderIFrame::layout): |
- * rendering/RenderLayer.cpp: |
- (WebCore::RenderLayer::updateLayerPositions): |
- * rendering/RenderReplaced.cpp: |
- (WebCore::RenderReplaced::layout): |
- * rendering/RenderTable.cpp: |
- (WebCore::RenderTable::layout): |
- |
-2010-11-29 Chris Rogers <crogers@google.com> |
- |
- Reviewed by Kenneth Russell. |
- |
- Add remaining custom JSC web audio bindings |
- https://bugs.webkit.org/show_bug.cgi?id=50172 |
- |
- No new tests since audio API is not yet implemented. |
- |
- * bindings/js/JSAudioBufferSourceNodeCustom.cpp: Added. |
- (WebCore::JSAudioBufferSourceNode::setBuffer): |
- * bindings/js/JSConvolverNodeCustom.cpp: Added. |
- (WebCore::JSConvolverNode::setBuffer): |
- * bindings/js/JSJavaScriptAudioNodeCustom.cpp: Added. |
- (WebCore::JSJavaScriptAudioNode::markChildren): |
- * webaudio/AudioBufferSourceNode.idl: |
- * webaudio/ConvolverNode.idl: |
- |
-2010-11-29 Yael Aharon <yael.aharon@nokia.com> |
- |
- Reviewed by Darin Adler. |
- |
- HTMLFormElement.elements doesn't include fieldsets |
- https://bugs.webkit.org/show_bug.cgi?id=48193 |
- |
- Add fieldset and keygen to the list of form collection elements, by making them enumeratable. |
- This patch does not add object element to the list. A separate bug will do that. |
- |
- Test: fast/forms/form-collection-elements.html |
- |
- * html/HTMLFieldSetElement.h: |
- (WebCore::HTMLFieldSetElement::isEnumeratable): |
- * html/HTMLKeygenElement.h: |
- (WebCore::HTMLKeygenElement::isEnumeratable): |
- |
-2010-11-29 Xan Lopez <xlopez@igalia.com> |
- |
- Reviewed by Martin Robinson. |
- |
- Get distcheck going. |
- |
- * GNUmakefile.am: |
- |
-2010-11-29 Alexey Proskuryakov <ap@apple.com> |
- |
- Reviewed by Darin Adler. |
- |
- https://bugs.webkit.org/show_bug.cgi?id=50181 |
- CSS style declarations don't GC protect parents |
- |
- Test: fast/dom/StyleSheet/gc-declaration-parent-rule.html |
- |
- * bindings/js/JSCSSStyleDeclarationCustom.cpp: (WebCore::JSCSSStyleDeclaration::markChildren): |
- Protect the parent rule. |
- |
-2010-11-29 Ilya Tikhonovsky <loislo@chromium.org> |
- |
- Reviewed by Pavel Feldman. |
- |
- Web Inspector: UI fix. In Timeline, Paint events "Location" value is displayed |
- in a confusing manner. The format string should be changed from %d x %d |
- pattern to (%d, %d). |
- |
- https://bugs.webkit.org/show_bug.cgi?id=50178 |
- |
- * English.lproj/localizedStrings.js: |
- * inspector/front-end/TimelinePanel.js: |
- (WebInspector.TimelinePanel.FormattedRecord.prototype._generatePopupContent): |
- |
-2010-11-29 Darin Adler <darin@apple.com> |
- |
- Reviewed by Andreas Kling. |
- |
- Remove a couple unneeded overflow checks |
- https://bugs.webkit.org/show_bug.cgi?id=49816 |
- |
- * platform/text/TextCodecUTF16.cpp: |
- (WebCore::TextCodecUTF16::encode): Use an ASSERT instead of |
- an overflow check with CRASH. |
- |
-2010-11-29 Tony Chang <tony@chromium.org> |
- |
- Unreviewed, disable warnings again on chromium linux |
- |
- * WebCore.gyp/WebCore.gyp: |
- |
-2010-11-29 Cosmin Truta <ctruta@chromium.org> |
- |
- Reviewed by Nikolas Zimmermann. |
- |
- getBoundingClientRect: Do not truncate the coordinates to integers |
- https://bugs.webkit.org/show_bug.cgi?id=48110 |
- |
- Remove the float-to-int coordinate conversions inside getBoundingClientRect, |
- to comply with the CSSOM View Module API. |
- |
- Tests: fast/dom/Element/getBoundingClientRect.html |
- fast/dom/Range/getBoundingClientRect.html |
- svg/zoom/page/zoom-getBoundingClientRect.xhtml |
- svg/zoom/page/zoom-zoom-coords.xhtml |
- |
- * dom/ClientRect.cpp: |
- (WebCore::ClientRect::ClientRect): Added overload to allow a FloatRect argument. |
- * dom/ClientRect.h: |
- (WebCore::ClientRect::create): Ditto. |
- * dom/Element.cpp: |
- (Element::getBoundingClientRect): Removed conversion from FloatRect to IntRect. |
- * dom/Range.cpp: |
- (WebCore::Range::getBoundingClientRect): Ditto. |
- * rendering/RenderObject.h: |
- (adjustIntRectForAbsoluteZoom): Removed. |
- (adjustFloatRectForAbsoluteZoom): Added. |
- * rendering/style/RenderStyle.h: |
- (adjustFloatForAbsoluteZoom): Added. |
- |
-2010-11-29 Dimitri Glazkov <dglazkov@chromium.org> |
- |
- Reviewed by Darin Adler. |
- |
- Converge means of querying a parent node into one way, which is Node::parentNode. |
- https://bugs.webkit.org/show_bug.cgi?id=49686 |
- |
- A follow-up to r72259, which explicitly forbids further use of Node::parent by |
- making it private. |
- |
- No change in behavior, so no new tests. |
- |
- * dom/Document.cpp: |
- (WebCore::Document::styleForElementIgnoringPendingStylesheets): Changed to use parentNode. |
- (WebCore::Document::hoveredNodeDetached): Ditto. |
- (WebCore::Document::activeChainNodeDetached): Ditto. |
- * dom/Node.h: Made Node::parent private. |
- * page/SpatialNavigation.cpp: |
- (WebCore::canBeScrolledIntoView): Changed to use parentNode. |
- * xml/XSLTProcessorQt.cpp: |
- (WebCore::XSLTProcessor::transformToString): Ditto. |
- |
-2010-11-29 Tony Chang <tony@chromium.org> |
- |
- Reviewed by James Robinson. |
- |
- [chromium] try to enable WebCore compiler warnings on linux |
- https://bugs.webkit.org/show_bug.cgi?id=50168 |
- |
- * WebCore.gyp/WebCore.gyp: |
- * platform/graphics/chromium/FontLinux.cpp: |
- (WebCore::TextRunWalker::setPadding): |
- (WebCore::TextRunWalker::setGlyphXPositions): |
- (WebCore::glyphIndexForXPositionInScriptRun): |
- |
-2010-11-29 Alexey Proskuryakov <ap@apple.com> |
- |
- Reviewed by Darin Adler. |
- |
- https://bugs.webkit.org/show_bug.cgi?id=50165 |
- CSS style rules don't GC protect parents |
- |
- Tests: fast/dom/StyleSheet/gc-parent-rule.html |
- fast/dom/StyleSheet/gc-parent-stylesheet.html |
- |
- * bindings/js/JSCSSRuleCustom.cpp: (WebCore::JSCSSRule::markChildren): Mark parents. The code |
- is super naive compared to what we have for nodes - but CSSOM has shallow hierarchies, so |
- it should be OK. |
- |
- * css/CSSRule.idl: Added CustomMarkFunction. |
- |
-2010-11-29 Gavin Peters <gavinp@chromium.org> |
- |
- Reviewed by Adam Barth. |
- |
- Web page can prevent WebKit from loading subresources on other |
- pages (cache poisoning) |
- https://bugs.webkit.org/show_bug.cgi?id=35404 |
- |
- Tests: http/tests/misc/unloadable-script.html |
- loader/reload-subresource-when-type-changes.html |
- |
- * loader/cache/MemoryCache.cpp: |
- (WebCore::MemoryCache::requestResource): |
- |
-2010-11-29 Simon Fraser <simon.fraser@apple.com> |
- |
- Reviewed by Dan Bernstein. |
- |
- Crash in WebCore::CSSStyleSelector::loadPendingImages |
- https://bugs.webkit.org/show_bug.cgi?id=50149 |
- |
- Null-check style->boxReflect() when loading pending style images. |
- |
- Test: fast/reflections/pending-reflection-mask-crash.html |
- |
- * css/CSSStyleSelector.cpp: |
- (WebCore::CSSStyleSelector::loadPendingImages): |
- |
-2010-11-29 Pavel Feldman <pfeldman@chromium.org> |
- |
- Reviewed by Yury Semikhatsky. |
- |
- Web Inspector: crash upon inspecting user style sheet. |
- https://bugs.webkit.org/show_bug.cgi?id=50109 |
- |
- * inspector/InspectorStyleSheet.cpp: |
- (WebCore::InspectorStyleSheet::resourceStyleSheetText): |
- |
-2010-11-29 Patrick Gansterer <paroga@webkit.org> |
- |
- Reviewed by Adam Roben. |
- |
- [WINCE] Implement WebCore::directoryName |
- https://bugs.webkit.org/show_bug.cgi?id=50031 |
- |
- Also fix handling of paths without path separator. |
- |
- * platform/wince/FileSystemWinCE.cpp: |
- (WebCore::reverseFindPathSeparator): |
- (WebCore::makeAllDirectories): |
- (WebCore::pathGetFileName): |
- (WebCore::directoryName): |
- |
-2010-11-29 Patrick Gansterer <paroga@webkit.org> |
- |
- Reviewed by Adam Roben. |
- |
- Simplify directoryName in FileSystemWin.cpp |
- https://bugs.webkit.org/show_bug.cgi?id=50028 |
- |
- * platform/win/FileSystemWin.cpp: |
- (WebCore::directoryName): |
- |
2010-11-24 Dimitri Glazkov <dglazkov@chromium.org> |
Reviewed by Darin Adler. |