Timeline
05/23/06:
- 18:21 Changeset [14555] by
-
S60WebKit:
- Corrected build.bat usage statement
- Removed some irrelevant config files
- 18:08 Changeset [14554] by
-
Correct reviewer from me to Tim.
- 17:59 Changeset [14553] by
-
Reviewed by Maciej.
- more loader refactoring to simplify things and remove knowledge of WebView from WebSubresourceLoader
- WebCoreSupport/WebSubresourceLoader.m: (+[WebSubresourceLoader startLoadingResource:withRequest:customHeaders:referrer:forDataSource:]): (-[WebSubresourceLoader didReceiveResponse:]):
- WebView/WebDataSource.m: (-[WebDataSource _loadIcon]): (-[WebDataSource _startLoading]):
- WebView/WebFrame.m: (-[WebFrame _loadRequest:archive:]): (-[WebFrame _loadItem:withLoadType:]): (-[WebFrame _loadURL:referrer:loadType:target:triggeringEvent:form:formValues:]): (-[WebFrame _postWithURL:referrer:target:data:contentType:triggeringEvent:form:formValues:]): (-[WebFrame _addExtraFieldsToRequest:mainResource:alwaysFromRequest:]):
- WebView/WebFrameInternal.h:
- WebView/WebFramePrivate.h:
- WebView/WebLoader.h:
- WebView/WebLoader.m:
- WebView/WebMainResourceLoader.m: (-[WebMainResourceLoader initWithDataSource:]): (-[WebMainResourceLoader didReceiveResponse:]):
- 17:24 Changeset [14552] by
-
Reviewed by Timothy.
- index.html: Link to Nokia's S60 WebKit.
- 16:32 Changeset [14551] by
-
Reviewed by Eric.
Remove the recent idl files from the WebCore target to prevent them
from being copied into the framework resources.
- WebCore.xcodeproj/project.pbxproj:
- 16:31 Changeset [14550] by
-
Reviewed by Justin Garcia and Levi Weintraub.
- fixed Bug 9072: REGRESSION: Misspelling marker incorrect in editing/deleting/delete-and-undo.html
- dom/Document.cpp: (WebCore::Document::removeMarkers): One more time to get this right, ugh. There is still one case where we don't want to advance the iterator, and that's when we remove a marker and do not insert a replacement marker. So I undid my previous patch, and added it++ to each of the two cases that insert a replacement marker. Now the layout tests pass even with the pixel tests enabled.
- 15:48 Changeset [14549] by
-
S60WebKit ... migrated!
- 15:11 Changeset [14548] by
-
Implement CookieJar on Win32.
- 14:37 Changeset [14547] by
-
Tag for WebKit 418.8 (part 2 of 2.)
- 14:37 Changeset [14546] by
-
Tag for WebKit 418.8 (part 1 of 2.)
- 14:36 Changeset [14545] by
-
Versioning
- 14:36 Changeset [14544] by
-
Merged fix from TOT to Safari-2-0-branch
2006-05-23 Tim Omernick < timo@apple.com>
Reviewed by John Sullivan.
< rdar://problem/4558301> REGRESSION (420+): After clearing history and closing bookmark view window,
attempting to select Safari's menu bar results in a crash
- Misc/WebNSWindowExtras.m: (replacementPostWindowNeedsDisplay): My fix yesterday for 4557117 was not quite good enough. Now that we cancel the display timer for a window *before* it deallocates, we need to make sure that while a window deallocates, no new display timers are scheduled for that window. This is actually possible, as 4558301 demonstrates. Luckily, NSWindow sets a handy "windowDying" flag when it deallocates, so we can just check that flag and bail out of the throttle hack if it is set. This should fix the last of the crashes involving display timers scheduled for deallocated or deallocating windows.
- 14:21 Changeset [14543] by
-
Fix post form submission on Win32.
- 13:45 Changeset [14542] by
-
Reviewed by Maciej.
- DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj: Newer Xcode removed some obsolete cruft
- DumpRenderTree/TextInputController.m: (-[TextInputController textInput]): added (id) cast to make newer compiler happy
- 13:31 Changeset [14541] by
-
Fix glyph width access on win32 to not depend on the UChar.
- 12:27 Changeset [14540] by
-
Reviewed by John Sullivan.
< rdar://problem/4558301> REGRESSION (420+): After clearing history and closing bookmark view window,
attempting to select Safari's menu bar results in a crash
- Misc/WebNSWindowExtras.m: (replacementPostWindowNeedsDisplay): My fix yesterday for 4557117 was not quite good enough. Now that we cancel the display timer for a window *before* it deallocates, we need to make sure that while a window deallocates, no new display timers are scheduled for that window. This is actually possible, as 4558301 demonstrates. Luckily, NSWindow sets a handy "windowDying" flag when it deallocates, so we can just check that flag and bail out of the throttle hack if it is set. This should fix the last of the crashes involving display timers scheduled for deallocated or deallocating windows.
- 12:18 Changeset [14539] by
-
S60WebKit: adding patentlicense.txt
- 12:15 Changeset [14538] by
-
Versioning
- 10:51 Changeset [14537] by
-
Reviewed by Hyatt.
Cleaned up changes to markup by adding the non-virtual
nodeNamePreservingCase accessor function.
- dom/Element.cpp: (WebCore::Element::nodeNamePreservingCase):
- dom/Element.h:
- editing/markup.cpp: (WebCore::startMarkup): (WebCore::endMarkup):
- 10:47 Changeset [14536] by
-
New tag.
- 10:35 Changeset [14535] by
-
Reviewed by Anders Carlsson.
- dom/Document.cpp: (WebCore::Document::removeMarkers): Fix to my previous checkin (which broke a layout test). Advance the iterator position every time through the loop; formerly it was not advanced in the case where a marker was actually removed, causing an infinite loop in certain cases. Also modified a comment that was confusing the situation by falsely claiming that iterating over a just-inserted node would always be OK.
- 09:27 Changeset [14534] by
-
WebCore:
2006-05-23 Anders Carlsson < acarlsson@apple.com>
Reviewed by Maciej.
http://bugzilla.opendarwin.org/show_bug.cgi?id=9017
Weather2.css incorrectly treated as JavaScript file @ www.live.com
- dom/Element.cpp: (WebCore::Element::getAttribute): (WebCore::Element::setAttribute): (WebCore::Element::removeAttribute): (WebCore::Element::getAttributeNode): (WebCore::Element::hasAttribute): Use new getAttributeItem method.
- dom/NamedAttrMap.cpp: (WebCore::NamedAttrMap::getNamedItem): (WebCore::NamedAttrMap::removeNamedItem): Use new getAttributeItem method.
(WebCore::NamedAttrMap::getAttributeItem):
- dom/NamedAttrMap.h: New method which returns the attribute based on the nodeName.
LayoutTests:
2006-05-23 Anders Carlsson < acarlsson@apple.com>
Reviewed by Maciej.
Update tests for http://bugzilla.opendarwin.org/show_bug.cgi?id=9017
Weather2.css incorrectly treated as JavaScript file @ www.live.com
- dom/svg/level3/xpath/XPathNSResolver_lookupNamespaceURI_nist_dmstc-expected.txt: This fails in a different way now.
- dom/xhtml/level3/core/nodelookupnamespaceuri17-expected.txt:
- dom/xhtml/level3/core/nodelookupnamespaceuri20-expected.txt:
- dom/xhtml/level3/core/nodelookupprefix17-expected.txt:
- dom/xhtml/level3/core/nodelookupprefix20-expected.txt:
- dom/xhtml/level3/core/noderemovechild28-expected.txt:
- dom/xhtml/level3/core/noderemovechild29-expected.txt: These now pass.
- fast/dom/attribute-namespaces-get-set-expected.txt: Added.
- fast/dom/attribute-namespaces-get-set.html: Added. Add new tests.
- 08:40 Changeset [14533] by
-
Reviewed by Eric.
- remove knowledge of WebView from WebLoader in preparation for moving the code down
- WebView/WebDataSource.m: (-[WebDataSource _defersCallbacks]): (-[WebDataSource _identifierForInitialRequest:]): (-[WebDataSource _willSendRequest:forResource:redirectResponse:]): (-[WebDataSource _didReceiveAuthenticationChallenge:forResource:]): (-[WebDataSource _didCancelAuthenticationChallenge:forResource:]): (-[WebDataSource _didReceiveResponse:forResource:]): (-[WebDataSource _didReceiveData:contentLength:forResource:]): (-[WebDataSource _didFinishLoadingForResource:]): (-[WebDataSource _didFailLoadingWithError:forResource:]): (-[WebDataSource _downloadWithLoadingConnection:request:response:proxy:]): (-[WebDataSource _privateBrowsingEnabled]):
- WebView/WebDataSourceInternal.h:
- WebView/WebLoader.h:
- WebView/WebLoader.m: (-[WebLoader releaseResources]): (-[WebLoader setDataSource:]): (-[WebLoader willSendRequest:redirectResponse:]): (-[WebLoader didReceiveAuthenticationChallenge:]): (-[WebLoader didCancelAuthenticationChallenge:]): (-[WebLoader didReceiveResponse:]): (-[WebLoader didReceiveData:lengthReceived:]): (-[WebLoader signalFinish]): (-[WebLoader didFinishLoading]): (-[WebLoader didFailWithError:]): (-[WebLoader willCacheResponse:]): (-[WebLoader cancelWithError:]):
- WebView/WebMainResourceLoader.m: (-[WebMainResourceLoader continueAfterContentPolicy:response:]):
- WebView/WebView.m: (-[WebView _incrementProgressForIdentifier:response:]): (-[WebView _incrementProgressForIdentifier:data:]): (-[WebView _completeProgressForIdentifier:]):
- WebView/WebViewInternal.h:
05/22/06:
- 18:36 Changeset [14532] by
-
2006-05-22 Eric Seidel < eric@eseidel.com>
Rubber-stamped by adele.
- bindings/js/kjs_html.h: simple style cleanup.
- 16:55 Changeset [14531] by
-
Tag for 521.10
- 16:46 Changeset [14530] by
-
Versioning.
- 16:30 Changeset [14529] by
-
Build fix when doing 'xcodebuild install'.
- 15:18 Changeset [14528] by
-
Build fix.
- 15:10 Changeset [14527] by
-
Reviewed by levi
Fixed and reenabled this test.
- editing/selection/drag-to-contenteditable-iframe-expected.checksum:
- editing/selection/drag-to-contenteditable-iframe-expected.png:
- editing/selection/drag-to-contenteditable-iframe-expected.txt:
- editing/selection/drag-to-contenteditable-iframe.html: Added.
- editing/selection/drag-to-contenteditable-iframe.html-disabled: Removed.
- 14:51 Changeset [14526] by
-
Tag for WebKit 418.7 (part 2 of 2.)
- 14:51 Changeset [14525] by
-
Tag for WebKit 418.7 (part 1 of 2.)
- 14:51 Changeset [14524] by
-
Versioning
- 14:41 Changeset [14523] by
-
Merged fix from TOT to Safari-2-0-branch
2006-05-22 Tim Omernick < timo@apple.com>
Reviewed by Geoff.
< rdar://problem/4557117> TOT REGRESSION: Repro crash in cancelPendingWindowDisplay --> _timerRelease when
opening file from file open dialog
Note that I could not reproduce this situation at all on any of my machines; I had to debug this on Geoff's
machine.
- Misc/WebNSWindowExtras.m: (replacementDealloc): Cancel display timer before dealloc, not after. Cancelling the timer can cause a message to be sent to the window; best to do that before the window deallocates. (replacementFinalize): ditto
- 14:39 Changeset [14522] by
-
Removes svn:executable from MathExtras.h. Caused a build verification failure.
- 14:20 Changeset [14521] by
-
Reviewed by Anders Carlsson.
- fixed Bug 9038: Rework Document::TextMarker mechanism to not use deprecated data structures
Changed Document::markersForNode and Document::MarkerMap to use Vector rather than
DeprecatedValueList and DeprecatedValueListIterator.
- dom/Document.h:
- dom/Document.cpp: (WebCore::Document::addMarker): (WebCore::Document::copyMarkers): (WebCore::Document::removeMarkers): (WebCore::Document::markersForNode): (WebCore::Document::repaintMarkers): (WebCore::Document::shiftMarkers):
- rendering/InlineTextBox.cpp: (WebCore::InlineTextBox::paintAllMarkersOfType):
- 14:00 Changeset [14520] by
-
64-bit build fix, and removes the workaround for < rdar://problem/4294625>.
- 13:55 Changeset [14519] by
-
Reviewed by Geoff.
< rdar://problem/4557117> TOT REGRESSION: Repro crash in cancelPendingWindowDisplay --> _timerRelease when
opening file from file open dialog
Note that I could not reproduce this situation at all on any of my machines; I had to debug this on Geoff's
machine.
- Misc/WebNSWindowExtras.m: (replacementDealloc): Cancel display timer before dealloc, not after. Cancelling the timer can cause a message to be sent to the window; best to do that before the window deallocates. (replacementFinalize): ditto
- 13:21 Changeset [14518] by
-
2006-05-22 Steve Falkenburg < sfalken@apple.com>
Reviewed by adele.
Fix build.
- Spinneret/Spinneret.sln:
- 13:16 Changeset [14517] by
-
2006-05-22 Steve Falkenburg < sfalken@apple.com>
Reviewed by adele.
Fix build.
- WebKitWin.vcproj/WebKitWin.sln:
- 11:34 Changeset [14516] by
-
LayoutTests:
Reviewed by eric.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=8393 <br>s created by createMarkup aren't valid xhtml Changes in case from innerHTML
- editing/execCommand/createLink-expected.txt:
- editing/execCommand/unlink-expected.txt:
- editing/input/attributed-substring-from-range-lines-expected.txt:
- fast/dom/Range/range-exceptions-expected.txt:
- fast/dom/Range/resources/range-exceptions.js:
- fast/innerHTML/005-expected.txt:
- fast/innerHTML/innerHTML-custom-tag-expected.txt:
- fast/innerHTML/innerHTML-cast.html: Added.
- fast/innerHTML/innerHTML-cast-expected.txt: Added.
- fast/xsl/xslt-processer-expected.txt:
WebCore:
Reviewed by Eric.
- Fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=8393 <br>s created by createMarkup aren't valid xhtml Fixed uppered output from innerHTML
- editing/markup.cpp: (WebCore::startMarkup): (WebCore::endMarkup):
- 11:20 Changeset [14515] by
-
Reviewed by Eric, Kevin and Geoff.
Merge open source build fixes. < rdar://problem/4555500>
- kjs/collector.cpp: look at the rsp register in x86_64 (KJS::Collector::markOtherThreadConservatively):
- wtf/Platform.h: add x86_64 to the platform list
- 01:04 Changeset [14514] by
-
Fix Mac bustage.
05/21/06:
- 23:19 Changeset [14513] by
-
Fix win32 bustage.
- 23:18 Changeset [14512] by
-
Fix win32 bustage.
- 16:42 Changeset [14511] by
-
Reviewed by Maciej.
- Fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=9016 crash loading live.com in TreeShared::ref because accessing RenderTextField::text() can destroy the RenderTextField
- rendering/RenderTextField.cpp: (WebCore::RenderTextField::text): Reverting recent change to user innerText instead of textContent. For textareas, we need a better solution than textContent that also takes newlines into account, but this should fix this crasher.
- 00:51 Changeset [14510] by
-
2006-05-21 Anders Carlsson < acarlsson@apple.com>
Fix build.
- ForwardingHeaders/wtf/MathExtras.h: Added.
- 00:18 Changeset [14509] by
-
Add mathextras to javascrtipcore project.
- 00:05 Changeset [14508] by
-
Fix mac bustage. Minor snafu in constructor.
05/20/06:
- 23:58 Changeset [14507] by
-
Switch on Cairo-based text rendering on win32.
- 23:55 Changeset [14506] by
-
Add mathextras.h to wtf to give win32 roundf/lroundf support.
- 19:09 Changeset [14505] by
-
Reviewed by Beth.
- fix assertion failure on layout tests
- WebView/WebDataSource.m: (-[WebDataSource _mainReceivedError:complete:]): Don't do anything if this data source is no longer connected to a frame. Used to be this couldn't happen because the WebView would have been nil, but we no longer go through the WebView.
- 15:42 Changeset [14504] by
-
Reviewed by Anders.
Bug 9018: REGRESSION: resizing the top area of the inspector does not grow as expected
http://bugzilla.opendarwin.org/show_bug.cgi?id=9018
Use clientHeight now that offsetHeight includes the border and padding.
clientHeight should have been used initially.
- WebInspector/WebInspector.m: (-resizeTopArea):
- 14:22 Changeset [14503] by
-
Fix Cairo double buffering.
- 13:19 Changeset [14502] by
-
JavaScriptCore:
2006-05-20 Anders Carlsson < acarlsson@apple.com>
Reviewed by Geoff.
http://bugzilla.opendarwin.org/show_bug.cgi?id=8993
Support function declaration in case statements
- kjs/grammar.y: Get rid of StatementList and use SourceElements instead.
- kjs/nodes.cpp: (CaseClauseNode::evalStatements): (CaseClauseNode::processVarDecls): (CaseClauseNode::processFuncDecl): (ClauseListNode::processFuncDecl): (CaseBlockNode::processFuncDecl): (SwitchNode::processFuncDecl):
- kjs/nodes.h: (KJS::CaseClauseNode::CaseClauseNode): (KJS::ClauseListNode::ClauseListNode): (KJS::ClauseListNode::getClause): (KJS::ClauseListNode::getNext): (KJS::ClauseListNode::releaseNext): (KJS::SwitchNode::SwitchNode): Add processFuncDecl for the relevant nodes.
- kjs/nodes2string.cpp: (CaseClauseNode::streamTo): next got renamed to source.
LayoutTests:
2006-05-20 Anders Carlsson < acarlsson@apple.com>
Reviewed by Geoff.
- fast/js/function-declarations-in-switch-statement-expected.txt: Added.
- fast/js/function-declarations-in-switch-statement.html: Added.
- fast/js/resources/function-declarations-in-switch-statement.js: Added.
- 01:57 Changeset [14501] by
-
Reviewed by Darin.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=8991 REGRESSION: missing or broken CSS custom cursors are displayed as a missing image icon
- page/FrameView.cpp: (WebCore::selectCursor): Fallback to CURSOR_AUTO if the image is an error image.
- manual-tests/custom-cursors.html: Added.
- 01:47 Changeset [14500] by
-
2006-05-20 Mitz Pettel < opendarwin.org@mitzpettel.com>
Reviewed and landed by ap.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9009 REGRESSION: ToT crash in WebCore at Zap2it
Test: fast/table/empty-section-crash.html
- rendering/RenderTableSection.cpp: (WebCore::RenderTableSection::paint): Return immediately if the section has 0 rows or 0 columns.
05/19/06:
- 22:52 Changeset [14499] by
-
Fix Win32 bustage caused by image document. supportsType is not implemented yet for ImageCairo, so we'll return false rather than true to keep every document from being an image document.
- 17:21 Changeset [14498] by
-
LayoutTests:
Reviewed by justin.
Tests for < http://bugzilla.opendarwin.org/show_bug.cgi?id=8931>
Drag caret not painted for editable sub-frames
- editing/resources/select-and-drag-iframe.html: Added.
- editing/selection/drag-in-iframe-expected.checksum: Added.
- editing/selection/drag-in-iframe-expected.png: Added.
- editing/selection/drag-in-iframe-expected.txt: Added.
- editing/selection/drag-in-iframe.html: Added.
WebCore:
Reviewed by justin.
< http://bugzilla.opendarwin.org/show_bug.cgi?id=8931>
Drag caret not painted for editable sub-frames
Moved the drag caret out of Frame and into Page.
Only the Frame that contains the drag caret will paint it.
- editing/SelectionController.h:
- page/Frame.cpp: (WebCore::Frame::dragCaret): (WebCore::Frame::setDragCaret): (WebCore::Frame::paintDragCaret):
- page/FramePrivate.h:
- page/Page.cpp: (WebCore::Page::dragCaret): (WebCore::Page::setDragCaret):
- page/Page.h:
- 16:16 Changeset [14497] by
-
Reviewed by Adele.
- html/HTMLButtonElement.idl: added support for HTMLButtonElement.click()
- 14:18 Changeset [14496] by
- 12:35 Changeset [14495] by
-
LayoutTests:
- Test for overriding non-autogenerated global constructors.
- fast/dom/constructors-overriding-expected.txt: Added.
- fast/dom/constructors-overriding.html: Added.
WebCore:
Reviewed by Eric.
- Fixed non-autogenerated global constructors to match autogenerated ones and FF. (Found this bug while @ the GOOG.)
(1) They're no longer read-only, so they can be overridden.
(2) They now have the default object prototype, so they can do things
like 'toString' and 'valueOf', necessary for general functionality,
including my layout test.
(3) Their prototype properties are now enumerable and not read-only.
- bindings/js/JSDOMParser.cpp: (KJS::DOMParserConstructorImp::DOMParserConstructorImp):
- bindings/js/JSXMLHttpRequest.cpp: (KJS::JSXMLHttpRequestConstructorImp::JSXMLHttpRequestConstructorImp):
- bindings/js/JSXMLHttpRequest.h:
- bindings/js/JSXMLSerializer.cpp: (KJS::XMLSerializerConstructorImp::XMLSerializerConstructorImp):
- bindings/js/JSXSLTProcessor.cpp: (KJS::XSLTProcessorConstructorImp::XSLTProcessorConstructorImp):
- bindings/js/kjs_html.cpp: (KJS::OptionConstructorImp::OptionConstructorImp): (KJS::ImageConstructorImp::ImageConstructorImp):
- bindings/js/kjs_window.cpp:
- 12:07 Changeset [14494] by
-
Tag for JavaScriptCore 418.2 (part 2 of 2.)
- 12:07 Changeset [14493] by
-
Tag for JavaScriptCore 418.2 (part 1 of 2.)
- 12:06 Changeset [14492] by
-
Tag for WebCore 418.8 (part 2 of 2.)
- 12:06 Changeset [14491] by
-
Tag for WebCore 418.8 (part 1 of 2.)
- 12:05 Changeset [14490] by
-
Versioning
- 12:01 Changeset [14489] by
-
JavaScriptCore:
Reviewed by Maciej.
- Fixed < rdar://problem/4543569> 10.4.7 regression: Crash occurs when attempting to view image in slideshow mode at http://d.smugmug.com/gallery/581716 ( KJS::IfNode::execute (KJS::ExecState*) + 312)
- Major fix: On alternate threads, DOMObjects remain in the ScriptInterpreter's cache because they're not collected. So, they need an opportunity to mark their children.
- kjs/collector.cpp: (KJS::Collector::collect):
- kjs/internal.cpp: (InterpreterImp::mark):
- kjs/internal.h:
- kjs/interpreter.h: (KJS::Interpreter::mark):
WebCore:
Reviewed by Maciej.
- Fixed < rdar://problem/4543569> 10.4.7 regression: Crash occurs when attempting to view image in slideshow mode at http://d.smugmug.com/gallery/581716 ( KJS::IfNode::execute (KJS::ExecState*) + 312)
- Minor fix: DOMParserConstructor needed to inherit from DOMObject to prevent collection on alternate threads.
- Major fix: On alternate threads, DOMObjects remain in the ScriptInterpreter's cache because they're not collected. So, they need an opportunity to mark their children.
- khtml/ecma/domparser.h:
- khtml/ecma/kjs_binding.cpp: (KJS::ScriptInterpreter::mark):
- khtml/ecma/kjs_binding.h:
- 11:42 Changeset [14488] by
-
Reviewed by justin.
When the paste performance fix was merged over, uses of isCandidate
were changed to isRenderedContent. isRenderedContent on the branch
is incorrect in a few ways.
- khtml/editing/visible_position.cpp: (khtml::VisiblePosition::init):
- 11:17 Changeset [14487] by
-
Tag for WebKit 418.6 (part 2 of 2.)
- 11:17 Changeset [14486] by
-
Tag for WebKit 418.6 (part 1 of 2.)
- 11:16 Changeset [14485] by
-
Versioning
- 10:55 Changeset [14484] by
-
Merged fix from TOT to Safari-2-0-branch
2006-05-18 Tim Omernick < timo@apple.com>
Reviewed by John Sullivan.
< rdar://problem/4551938> More options needed for window display throttle
We decided to play it safe and leave our window flushing behavior unchanged from the previous
public release. By default, CoreGraphics deferred updates are once again OFF, and the window
display throttle is OFF.
Individual applications should set the WebKitThrottleWindowDisplayPreferenceKey and
WebKitEnableDeferredUpdatesPreferenceKey defaults to suit their needs.
Old behavior (like 10.4.6): WebKitThrottleWindowDisplayPreferenceKey=0 (or unset),
WebKitEnableDeferredUpdatesPreferenceKey (or unset).
Tear-free scrolling/animations: WebKitThrottleWindowDisplayPreferenceKey=0 (or unset),
WebKitEnableDeferredUpdatesPreferenceKey=1. While this configuration fixes the tearing issues
caused by over-flushing, some applications will experience performance problems as over-flushing
with CG deferred updates enabled will cause the app to block.
Tear-free scrolling/animations, high performance: WebKitThrottleWindowDisplayPreferenceKey=1,
WebKitEnableDeferredUpdatesPreferenceKey=1. This is the riskiest configuration in that it
enables the window display throttle "feature", potentially breaking applications' assumptions
about when displays occur. However, it provides the "best of both worlds", in that updates
are tear-free, and performance impact should me minimal.
- WebView/WebPreferenceKeysPrivate.h: Declared WebKitThrottleWindowDisplayPreferenceKey and WebKitEnableDeferredUpdatesPreferenceKey.
- WebView/WebFrameView.m: (-[WebFrameView initWithFrame:]): Turn off CG deferred updates if WebKitEnableDeferredUpdatesPreferenceKey is NO or has no value. Added some comments.
- 03:53 Changeset [14483] by
-
2006-05-19 Anders Carlsson < acarlsson@apple.com>
Reviewed by Eric.
http://bugzilla.opendarwin.org/show_bug.cgi?id=8983
Autogenerate another 22 HTML classes
Already covered by existing DOM tests.
- DerivedSources.make:
- WebCore.vcproj/WebCore/WebCore.vcproj:
- WebCore.xcodeproj/project.pbxproj: Add new IDL files and generated sources.
- bindings/js/JSCanvasRenderingContext2DCustom.cpp: (WebCore::JSCanvasRenderingContext2D::drawImage): (WebCore::JSCanvasRenderingContext2D::drawImageFromRect): (WebCore::JSCanvasRenderingContext2D::createPattern): Use JSHTMLImageElement::info in inherits.
- bindings/js/JSHTMLElementWrapperFactory.cpp: (WebCore::createJSWrapper): Add macros for the functions and the code to populate the hash set.
- bindings/js/kjs_html.cpp: (KJS::): (KJS::JSHTMLElement::classInfo): (KJS::JSHTMLElement::accessors): (KJS::HTMLElementFunction::callAsFunction):
- bindings/js/kjs_html.h: (KJS::JSHTMLElement::): Delete old cruft.
- bindings/scripts/CodeGeneratorJS.pm: Add support for creating a JS object from a HTMLCollection.
- html/HTMLAreaElement.idl: Added.
- html/HTMLBRElement.idl: Added.
- html/HTMLBaseFontElement.idl: Added.
- html/HTMLBlockquoteElement.idl: Added.
- html/HTMLBodyElement.cpp: (WebCore::HTMLBodyElement::scrollLeft): (WebCore::HTMLBodyElement::setScrollLeft): (WebCore::HTMLBodyElement::scrollTop): (WebCore::HTMLBodyElement::setScrollTop): (WebCore::HTMLBodyElement::scrollHeight): (WebCore::HTMLBodyElement::scrollWidth):
- html/HTMLBodyElement.h: Add new functions that used to be implemented in kjs_html.cpp
- html/HTMLBodyElement.idl: Added.
- html/HTMLFieldSetElement.idl: Added.
- html/HTMLFontElement.idl: Added.
- html/HTMLHRElement.idl: Added.
- html/HTMLHeadingElement.idl: Added.
- html/HTMLImageElement.idl: Added.
- html/HTMLIsIndexElement.idl: Added.
- html/HTMLLIElement.idl: Added.
- html/HTMLLabelElement.idl: Added.
- html/HTMLLegendElement.idl: Added.
- html/HTMLMapElement.idl: Added.
- html/HTMLMenuElement.idl: Added.
- html/HTMLModElement.idl: Added.
- html/HTMLParagraphElement.idl: Added.
- html/HTMLParamElement.idl: Added.
- html/HTMLPreElement.idl: Added.
- html/HTMLQuoteElement.idl: Added.
- html/HTMLScriptElement.idl: Added.