⚠ Archived content — this site is no longer maintained.   Current WebKit documentation is at docs.webkit.org.

Timeline



Jan 17, 2007:

11:57 PM Changeset in webkit [22782] by aliceli1
  • 3 edits in branches/WindowsMerge/WebKitWin

Added stubs to keep the boomer build from failing.

  • WebEditorClient.cpp: (WebEditorClient::didSetSelectionTypesForPasteboard):
  • WebEditorClient.h:
11:47 PM Changeset in webkit [18931] by aliceli1
  • 14 edits in trunk

WebCore:

Reviewed by Harrison.

Fix for <rdar://problem/4894155> REGRESSION: Extra line break is pasted with content into message body after choosing File - Paste

Migration of some editing code from WebHTMView to WebCore::Editor
resulted in not calling pasteboardTypesForSelection, which Mail was
overriding for the special purpose of adding a type to the
pasteboard after WebKit did. This patch adds 2 separate code paths
for Tiger and Leopard. On Tiger we give in and call the WebView's
pasteboardTypesForSelection. On Leopard we call a delegate after
the pasteboard types are set.

  • bridge/EditorClient.h:
  • editing/Editor.h:
  • platform/Pasteboard.h:
  • platform/graphics/svg/SVGImageEmptyClients.h: (WebCore::SVGEmptyEditorClient::didSetSelectionTypesForPasteboard): (WebCore::SVGEmptyEditorClient::pasteboardTypesForSelection):
  • platform/mac/PasteboardMac.mm: (WebCore::Pasteboard::Pasteboard): (WebCore::Pasteboard::clear): (WebCore::Pasteboard::writeSelection): (WebCore::Pasteboard::writeURL): (WebCore::Pasteboard::plainText): (WebCore::Pasteboard::documentFragment):

WebKit:

Reviewed by Harrison.

Fix for <rdar://problem/4894155> REGRESSION: Extra line break is pasted with content into message body after choosing File - Paste

Migration of some editing code from WebHTMView to WebCore::Editor
resulted in not calling pasteboardTypesForSelection, which Mail was
overriding for the special purpose of adding a type to the
pasteboard after WebKit did. This patch adds 2 separate code paths
for Tiger and Leopard. On Tiger we give in and call the WebView's
pasteboardTypesForSelection. On Leopard we call a delegate after
the pasteboard types are set.

  • DefaultDelegates/WebDefaultEditingDelegate.m: (-[WebDefaultEditingDelegate webView:didSetSelectionTypesForPasteboard:]):
  • WebCoreSupport/WebEditorClient.h:
  • WebCoreSupport/WebEditorClient.mm: (WebEditorClient::didSetSelectionTypesForPasteboard): (WebEditorClient::pasteboardTypesForSelection):
  • WebView/WebEditingDelegate.h:

WebKitQt:

Added these stubs to keep the Qt build from failing.

  • WebCoreSupport/EditorClientQt.cpp: (WebCore::EditorClientQt::didSetSelectionTypesForPasteboard):
  • WebCoreSupport/EditorClientQt.h:
10:05 PM Changeset in webkit [18930] by bdash
  • 1 copy in tags/Safari-419

New tag.

9:50 PM Changeset in webkit [18929] by bdash
  • 6 edits in branches/Safari-2-0-branch

Adjust installation paths.

8:32 PM Changeset in webkit [18928] by sullivan
  • 15 edits in trunk

WebCore:

Reviewed by Darin

  • WebCore part of fix for <rdar://problem/4462420> REGRESSION: Mail hangs during Replace All if the replacement string contains the search string


  • page/Frame.h:
  • page/Frame.cpp: (WebCore::Frame::findString): new startInSelection parameter -- if true, search starts looking in selection; if false, search starts looking after the selection


  • page/mac/WebCoreFrameBridge.h:
  • page/mac/WebCoreFrameBridge.mm: (-[WebCoreFrameBridge searchFor:direction:caseSensitive:wrap:startInSelection:]): new startInSelection parameter, passed through to Frame::findString()
  • editing/JSEditor.cpp: pass false for new findString parameter, to match old behavior

WebKit:

Reviewed by Darin

  • WebKit part of fix for <rdar://problem/4462420> REGRESSION: Mail hangs during Replace All if the replacement string contains the search string
  • Misc/WebKitVersionChecks.h: Added extern "C" so this can be used from .mm files. I don't need this change anymore for this fix, but it's still worth fixing now so it doesn't bite anyone later.
  • WebView/WebDocumentPrivate.h: Invented new private protocol WebDocumentIncrementalSearching, that has one method. The one method is just like the one WebDocumentSearching method, but with an additional parameter. We hope to eliminate this dependence on protocols someday, but adding another one as SPI seems like it won't make anything worse.


  • WebView/WebHTMLView.mm: (-[WebHTMLView searchFor:direction:caseSensitive:wrap:]): now calls through to new method that has one additional parameter, passing NO to match old behavior (-[WebHTMLView searchFor:direction:caseSensitive:wrap:startInSelection:]): pass new parameter to bridge


  • WebView/WebPDFView.h: Declare conformance to WebDocumentIncrementalSearching protocol


  • WebView/WebPDFView.mm: (-[WebPDFView searchFor:direction:caseSensitive:wrap:]): now calls through to new method that has one additional parameter, passing NO to match old behavior (-[WebPDFView searchFor:direction:caseSensitive:wrap:startInSelection:]): new method, former guts of searchFor:direction:caseSensitive:wrap: but now handles startInSelection parameter


  • WebView/WebViewPrivate.h: Declare new searchFor:direction:caseSensitive:wrap:startInSelection: method, just like existing method but with one additional parameter
  • WebView/WebView.mm: (-[WebView searchFor:direction:caseSensitive:wrap:]): now calls through to new method that has one additional parameter, passing NO to match old behavior (-[WebView searchFor:direction:caseSensitive:wrap:startInSelection:]): new method, former guts of searchFor:direction:caseSensitive:wrap: but now handles startInSelection parameter
8:00 PM Changeset in webkit [18927] by beidson
  • 1 edit in trunk/WebKit/Misc/WebDownload.m

Removed an internal change I had I wasn't supposed to commit. *sigh*

7:58 PM Changeset in webkit [18926] by beidson
  • 2 edits in trunk/WebKit

Reviewed by Deth Bakin and Brian Dash

Drop Panther Support (?!?) and change the comment explaining some SPI forward decls

  • Misc/WebDownload.m:
6:32 PM Changeset in webkit [18925] by bdakin
  • 2 edits in trunk/WebCore

Adding comment.

6:30 PM Changeset in webkit [18924] by justing
  • 3 edits
    6 adds in trunk

LayoutTests:

Reviewed by darin

<rdar://problem/4920488>
REGRESSION: A crash occurs at WebCore::Range::processContents () when removing formatting from selected text (that contains a link)

  • editing/execCommand/4920488-expected.checksum: Added.
  • editing/execCommand/4920488-expected.png: Added.
  • editing/execCommand/4920488-expected.txt: Added.
  • editing/execCommand/4920488.html: Added.
  • fast/dom/Range/compareBoundaryPoints-1-expected.txt: Added.
  • fast/dom/Range/compareBoundaryPoints-1.html: Added.

WebCore:

Reviewed by darin


<rdar://problem/4920488>
REGRESSION: A crash occurs at WebCore::Range::processContents () when removing formatting from selected text (that contains a link)

  • dom/Range.cpp: (WebCore::Range::commonAncestorContainer): Return null if the nodes don't have a common ancestor. As a side effect, the public commonAncestorContainer now sets an exception if the Range starts in one document and ends in another, or starts or ends in a tree that's been removed from the document. The spec doesn't cover this and there is no precedent in other browsers for this case because they don't allow invalid Ranges (you can create one with WebCore by removing a node that contains the start/end of a Range). (WebCore::Range::compareBoundaryPoints): If the the two positions are in nodes that have no common ancestor, return null. This doesn't change the behavior of the public compareBoundaryPoints: it already throws an exception if the two positions are in nodes that don't have a common ancestor. This follows the spec.
5:49 PM Changeset in webkit [18923] by sullivan
  • 2 edits in trunk/WebCore

The one word was typed by me, but only after debugging for a long time with Darin.

  • fixed <rdar://problem/4912002> REGRESSION: Series of words marked as a single misspelled word
  • editing/TextIterator.cpp: (WebCore::WordAwareIterator::WordAwareIterator): We were creating a TextIterator with the default kind, but WordAwareIterator is designed to work in concert with CharacterIterator, so they need to use the same kind. CharacterIterator uses kind RUNFINDER, so WordAwareIterator needs to do so also.
5:09 PM Changeset in webkit [18922] by adele
  • 2 edits in trunk/WebCore

Reviewed by Brady.

Fix for <rdar://problem/4900895> REGRESSION: onChange doesn't always fire for popups that have had value changed by JS

No test case since you need to open up the menu to fire onChange.

  • html/HTMLSelectElement.cpp: (WebCore::HTMLSelectElement::setSelectedIndex): Keep track of m_lastOnChangeIndex even when we're not firing onChange. Since the selected index has changed, we still want the m_lastOnChangeIndex to be up to date when we compare it later. This matches our behavior in Safari 2.0.
4:15 PM Changeset in webkit [18921] by adele
  • 2 edits in trunk/WebCore

Reviewed by Darin and John.

  • Fix for <rdar://problem/4816550> REGRESSION: autoscrolling doesn't work after double-click
  • page/EventHandler.cpp: (WebCore::EventHandler::handleMousePressEvent): Set swallowEvent bool instead of returning early for double and triple clicks.
3:31 PM Changeset in webkit [22781] by beidson
  • 9 edits in branches/WindowsMerge

Final hookup of HTTP Auth (from the engine perspective)

3:17 PM Changeset in webkit [18920] by eseidel
  • 4 edits in trunk/LayoutTests

2007-01-17 Eric Seidel <​eric@webkit.org>

Reviewed by Brady.


Update results after SVG container hit testing changes.

  • svg/custom/pointer-events-text-expected.checksum:
  • svg/custom/pointer-events-text-expected.png:
  • svg/custom/pointer-events-text-expected.txt:
3:13 PM Changeset in webkit [18919] by darin
  • 3 edits in trunk/WebCore

Reviewed by Hyatt.

  • rendering/RenderObject.h: Removed unused nextEditable and previousEditable.
  • rendering/RenderObject.cpp: Ditto.
2:50 PM Changeset in webkit [18918] by darin
  • 15 edits in trunk

WebCore:

Reviewed by Maciej.

  • WebCore.exp: Added HitTestResult::imageRect, removed HitTestResult::boundingBox.
  • rendering/HitTestResult.h:
  • rendering/HitTestResult.cpp: (WebCore::HitTestResult::imageRect): Added. Returns the content rectangle of the image.
  • rendering/RenderObject.h:
  • rendering/RenderObject.cpp: (WebCore::RenderObject::absolutePosition): Made const. (WebCore::RenderObject::contentBox): Added. (WebCore::RenderObject::absoluteContentBox): Added.
  • rendering/RenderBox.h:
  • rendering/RenderBox.cpp: (WebCore::RenderBox::absolutePosition): Made const.
  • rendering/RenderTableCell.h:
  • rendering/RenderTableCell.cpp: (WebCore::RenderTableCell::absolutePosition): Made const.
  • rendering/RenderView.h:
  • rendering/RenderView.cpp: (WebCore::RenderView::absolutePosition): Made const.

WebKit:

Reviewed by Maciej.

  • Misc/WebElementDictionary.mm: (-[WebElementDictionary _imageRect]): Call HitTestResult::imageRect, not HitTestResult::boundingBox.
2:15 PM Changeset in webkit [18917] by lars
  • 2 edits in trunk/WebKitTools

Fix my last commit to actually work in all cases.

2:05 PM Changeset in webkit [18916] by adele
  • 2 edits in trunk/LayoutTests

Reviewed by Mitz.

Updating this new layout test so that it actually crashes before the last fix.

  • fast/css/pseudostyle-anonymous-text.html:
1:50 PM Changeset in webkit [18915] by adele
  • 3 edits
    2 adds in trunk

LayoutTests:

Reviewed by Beth.

Test for ​http://bugs.webkit.org/show_bug.cgi?id=10139
<rdar://problem/4823028> REGRESSION: Reproducible crash when selecting a popup/button with ::selection pseudostyle

  • fast/css/pseudostyle-anonymous-text-expected.txt: Added.
  • fast/css/pseudostyle-anonymous-text.html: Added.

WebCore:

Reviewed by Beth.

Fix for ​http://bugs.webkit.org/show_bug.cgi?id=10139
<rdar://problem/4823028> REGRESSION: Reproducible crash when selecting a popup/button with ::selection pseudostyle

Test: fast/css/pseudostyle-anonymous-text.html

  • rendering/RenderObject.cpp: (WebCore::RenderObject::getPseudoStyle): Nil check the node.
1:19 PM Changeset in webkit [18914] by lars
  • 2 edits in trunk/WebKitTools

Reviewed by Maciej

Make sure DumpRenderTree exits when run-webkit-tests
is done.

1:01 PM Changeset in webkit [18913] by adele
  • 3 edits
    4 adds in trunk

LayoutTests:

Reviewed by Anders and John.

Test for ​http://bugs.webkit.org/show_bug.cgi?id=11278
<rdar://problem/4826875> REGRESSION: Popup width doesn't change to accommodate dynamic changes to options

  • fast/forms/menulist-width-change-expected.checksum: Added.
  • fast/forms/menulist-width-change-expected.png: Added.
  • fast/forms/menulist-width-change-expected.txt: Added.
  • fast/forms/menulist-width-change.html: Added.

WebCore:

Reviewed by Anders and John.

Fix for ​http://bugs.webkit.org/show_bug.cgi?id=11278
<rdar://problem/4826875> REGRESSION: Popup width doesn't change to accommodate dynamic changes to options

Test: fast/forms/menulist-width-change.html

  • rendering/RenderMenuList.cpp: (WebCore::RenderMenuList::updateFromElement): When the options change, call setNeedsLayoutAndMinMaxRecalc.
11:34 AM Changeset in webkit [18912] by ap
  • 29 edits in trunk

Reviewed by Maciej.

​http://bugs.webkit.org/show_bug.cgi?id=12268
Give object prototypes their own names

JavaScriptCore:

  • kjs/lookup.h: Append "Prototype" to ClassName in KJS_IMPLEMENT_PROTOTYPE.

WebCore:

  • bindings/scripts/CodeGeneratorJS.pm: Append "Proto" to interface name for the prototype's ClassInfo.
  • bindings/js/JSHTMLInputElementBase.cpp: (WebCore::JSHTMLInputElementBasePrototypeFunction::callAsFunction): (WebCore::JSHTMLInputElementBase::JSHTMLInputElementBase):
  • bindings/js/JSHTMLInputElementBase.h:
  • bindings/js/JSXMLHttpRequest.cpp: (KJS::JSXMLHttpRequestConstructorImp::JSXMLHttpRequestConstructorImp): (KJS::JSXMLHttpRequest::JSXMLHttpRequest): (KJS::JSXMLHttpRequestPrototypeFunction::callAsFunction):
  • bindings/js/JSXMLHttpRequest.h:
  • bindings/js/JSXSLTProcessor.cpp: (KJS::JSXSLTProcessor::JSXSLTProcessor): (KJS::XSLTProcessorPrototypeFunction::callAsFunction): (KJS::XSLTProcessorConstructorImp::XSLTProcessorConstructorImp):
  • bindings/js/kjs_css.cpp: (KJS::DOMCSSStyleDeclaration::DOMCSSStyleDeclaration): (KJS::DOMCSSStyleDeclarationPrototypeFunction::callAsFunction): (KJS::DOMMediaList::DOMMediaList): (KJS::KJS::DOMMediaListPrototypeFunction::callAsFunction): (KJS::DOMCSSStyleSheet::DOMCSSStyleSheet): (KJS::DOMCSSStyleSheetPrototypeFunction::callAsFunction): (KJS::DOMCSSValue::DOMCSSValue): (KJS::DOMCSSValuePrototypeFunction::callAsFunction):
  • bindings/js/kjs_css.h:
  • bindings/js/kjs_dom.cpp: (KJS::DOMNode::DOMNode): (KJS::DOMNodePrototypeFunction::callAsFunction): (KJS::DOMEventTargetNode::DOMEventTargetNode): (KJS::DOMEventTargetNodePrototypeFunction::callAsFunction): (KJS::DOMNamedNodeMap::DOMNamedNodeMap): (KJS::DOMNamedNodeMapPrototypeFunction::callAsFunction):
  • bindings/js/kjs_dom.h:
  • bindings/js/kjs_domnode.h:
  • bindings/js/kjs_events.cpp: (KJS::DOMEvent::DOMEvent): (KJS::DOMEventPrototypeFunction::callAsFunction): (KJS::Clipboard::Clipboard): (KJS::ClipboardPrototypeFunction::callAsFunction):
  • bindings/js/kjs_events.h:
  • bindings/js/kjs_html.cpp: (KJS::JSHTMLDocumentPrototypeFunction::callAsFunction): (KJS::): (KJS::JSHTMLDocument::JSHTMLDocument): (KJS::JSHTMLElementPrototypeFunction::callAsFunction): (KJS::JSHTMLElement::JSHTMLElement): (KJS::JSHTMLCollection::JSHTMLCollection): (KJS::JSHTMLCollectionPrototypeFunction::callAsFunction):
  • bindings/js/kjs_html.h:
  • bindings/js/kjs_navigator.cpp:
  • bindings/js/kjs_traversal.cpp: (KJS::DOMNodeFilter::DOMNodeFilter): (KJS::DOMNodeFilterPrototypeFunction::callAsFunction):
  • bindings/js/kjs_window.cpp: (KJS::Window::clear): Changed Proto to Prototype, and ProtoFunc to PrototypeFunction.

LayoutTests:

  • fast/dom/Window/window-properties-expected.txt:
  • fast/dom/prototype-chain-expected.txt:
  • fast/events/event-view-toString-expected.txt:
  • fast/events/mouseclick-target-and-positioning-expected.txt:
  • fast/tokenizer/external-script-document-write-expected.checksum:
  • fast/tokenizer/external-script-document-write-expected.png:
  • fast/tokenizer/external-script-document-write-expected.txt: Updated the results.
11:04 AM Changeset in webkit [18911] by lars
  • 7 edits in trunk/WebCore

Reviewed by Zack and Darin

Fix the Qt build again after the latest Image changes.
Also make some of the methods implemented in ImageCG
crossplatform.

11:02 AM Changeset in webkit [22780] by beidson
  • 15 edits
    8 adds in branches/WindowsMerge/WebKitWin

WebKitWin:

Reviewed by Steve Falkenburg

Next round of plumbing for HTTP Auth, or as I like to call it:
"Implementing platform dependent wrappers for my platform independent

wrappers which wrap platform dependent objects"


*sigh*


Also, not included in this ChangeLog, I changed the eol property to
native on a handful of WebKitWin files who didn't have it set yet


  • Interfaces/IWebURLAuthenticationChallenge.idl: Added some string constants, nuked an unneeded initializer from IWebCredential


  • WebError.cpp: Added QueryInterface for new GUID (WebError::resourceError): Added accessor to WebCore backer
  • WebError.h: Added GUID


  • WebKit.vcproj/WebKit.vcproj:


  • WebURLAuthenticationChallenge.cpp: Added. (WebURLAuthenticationChallenge::WebURLAuthenticationChallenge): (WebURLAuthenticationChallenge::~WebURLAuthenticationChallenge): (WebURLAuthenticationChallenge::createInstance): (WebURLAuthenticationChallenge::QueryInterface): (WebURLAuthenticationChallenge::AddRef): (WebURLAuthenticationChallenge::Release): (WebURLAuthenticationChallenge::initWithProtectionSpace): (WebURLAuthenticationChallenge::initWithAuthenticationChallenge): (WebURLAuthenticationChallenge::error): (WebURLAuthenticationChallenge::failureResponse): (WebURLAuthenticationChallenge::previousFailureCount): (WebURLAuthenticationChallenge::proposedCredential): (WebURLAuthenticationChallenge::protectionSpace): (WebURLAuthenticationChallenge::sender): (WebURLAuthenticationChallenge::authenticationChallenge):
  • WebURLAuthenticationChallenge.h: Added.


  • WebURLAuthenticationChallengeSender.cpp: Added. (WebURLAuthenticationChallengeSender::WebURLAuthenticationChallengeSender): (WebURLAuthenticationChallengeSender::~WebURLAuthenticationChallengeSender): (WebURLAuthenticationChallengeSender::createInstance): (WebURLAuthenticationChallengeSender::QueryInterface): (WebURLAuthenticationChallengeSender::AddRef): (WebURLAuthenticationChallengeSender::Release): (WebURLAuthenticationChallengeSender::cancelAuthenticationChallenge): (WebURLAuthenticationChallengeSender::continueWithoutCredentialForAuthenticationChallenge): (WebURLAuthenticationChallengeSender::useCredential): (WebURLAuthenticationChallengeSender::resourceHandle):
  • WebURLAuthenticationChallengeSender.h: Added.


  • WebURLCredential.cpp: Added. (WebURLCredential::WebURLCredential): (WebURLCredential::~WebURLCredential): (WebURLCredential::createInstance): (WebURLCredential::QueryInterface): (WebURLCredential::AddRef): (WebURLCredential::Release): (WebURLCredential::hasPassword): (WebURLCredential::initWithuser): (WebURLCredential::password): (WebURLCredential::persistence): (WebURLCredential::user): (WebURLCredential::credential):
  • WebURLCredential.h: Added.


  • WebURLProtectionSpace.cpp: Added. (WebURLProtectionSpace::WebURLProtectionSpace): (WebURLProtectionSpace::~WebURLProtectionSpace): (WebURLProtectionSpace::createInstance): (WebURLProtectionSpace::QueryInterface): (WebURLProtectionSpace::AddRef): (WebURLProtectionSpace::Release): (WebURLProtectionSpace::authenticationMethod): (WebURLProtectionSpace::host): (coreScheme): Helper for the init methods (WebURLProtectionSpace::initWithHost): (WebURLProtectionSpace::initWithProxyHost): (WebURLProtectionSpace::isProxy): (WebURLProtectionSpace::port): (WebURLProtectionSpace::protocol): (WebURLProtectionSpace::proxyType): (WebURLProtectionSpace::realm): (WebURLProtectionSpace::receivesCredentialSecurely): (WebURLProtectionSpace::protectionSpace):
  • WebURLProtectionSpace.h: Added.


  • WebURLResponse.cpp: Added QueryInterface for new GUID (WebURLResponse::resourceResponse): Added accessor to WebCore backer
  • WebURLResponse.h: Added GUID
10:45 AM Changeset in webkit [18910] by andersca
  • 5 edits in trunk/WebKit

Reviewed by John Sullivan.


Move all code in WebNetscapePluginEmbeddedView down to WebBaseNetscapePluginView.


  • Plugins/WebBaseNetscapePluginView.h:
  • Plugins/WebBaseNetscapePluginView.mm: (-[WebBaseNetscapePluginView initWithFrame:pluginPackage:URL:baseURL:MIMEType:attributeKeys:attributeValues:loadManually:DOMElement:]): (-[WebBaseNetscapePluginView didStart]): (-[WebBaseNetscapePluginView dataSource]): (-[WebBaseNetscapePluginView dealloc]): (-[WebBaseNetscapePluginView pluginView:receivedResponse:]): (-[WebBaseNetscapePluginView pluginView:receivedData:]): (-[WebBaseNetscapePluginView pluginView:receivedError:]): (-[WebBaseNetscapePluginView pluginViewFinishedLoading:]): (-[WebBaseNetscapePluginView _redeliverStream]):
  • Plugins/WebNetscapePluginEmbeddedView.h:
  • Plugins/WebNetscapePluginEmbeddedView.mm:
9:08 AM Changeset in webkit [18909] by darin
  • 2 edits in trunk/WebCore
  • quick attempt to fix the Qt build
  • platform/graphics/qt/ImageQt.cpp: (WebCore::BitmapImage::drawTiled): Remove obsolete parameters from this stub.
7:27 AM Changeset in webkit [18908] by brmorris
  • 1 copy in S60/branches/s60-RnD-07wk03

New branch for s60/trunk release

6:59 AM Changeset in webkit [18907] by eseidel
  • 3 edits
    4 adds in trunk

2007-01-17 Eric Seidel <​eric@webkit.org>

Reviewed by rwlbuis.

Test: svg/custom/inner-svg-hit-test.svg

  • rendering/RenderSVGContainer.cpp: (WebCore::RenderSVGContainer::nodeAtPoint): never allow containers to be hit.
5:18 AM Changeset in webkit [18906] by lars
  • 17 edits in trunk

Reviewed by Zack

Don't call receivedResponse twice for file: URLs.
Never call back into the ResourceHandleClient once
cancel() has been called on the ResourceHandle.

Remove the inheritance from Shared<XxxClient> in the
client classes.

Answer asynchronously to most of the Policy checking methods
in FrameLoaderClientQt to avoid some crashes in the loader.

Small fixes in DumpRenderTree, so we don't by
accident dump twice for the same test.

Exclude one more test as it currently causes DumpRenderTree to
hang forever.

4:21 AM Changeset in webkit [18905] by eseidel
  • 2 edits in trunk/WebCore

2007-01-17 Eric Seidel <​eric@webkit.org>

No review, build fix only.

  • platform/graphics/cg/ImageCG.cpp: correct drawPatternCombined to drawPattern (WebCore::Image::drawPattern): (WebCore::Image::drawTiled):
4:10 AM Changeset in webkit [18904] by eseidel
  • 13 edits in trunk

2007-01-17 Eric Seidel <​eric@webkit.org>

Reviewed by mitz and rwlbuis.


SVGs do not work as background images:
​http://bugs.webkit.org/show_bug.cgi?id=12096

Tests updated:

  • fast/images/svg-as-background.svg: now passes
  • fast/images/svg-as-tiled-background.svg: now passes
  • platform/graphics/BitmapImage.h: (WebCore::BitmapImage::nativeImageForCurrentFrame): (WebCore::BitmapImage::mayFillWithSolidColor): (WebCore::BitmapImage::solidColor):
  • platform/graphics/Image.h: (WebCore::Image::mayFillWithSolidColor): (WebCore::Image::solidColor): (WebCore::Image::nativeImageForCurrentFrame): (WebCore::Image::startAnimation):
  • platform/graphics/cg/ImageCG.cpp: (WebCore::Image::drawPatternCallback): (WebCore::Image::drawPatternCombined): (WebCore::caculatePatternScale): (WebCore::Image::drawTiled):
  • platform/graphics/cg/PDFDocumentImage.cpp:
  • platform/graphics/cg/PDFDocumentImage.h:
  • platform/graphics/svg/SVGImage.cpp: (WebCore::SVGImage::nativeImageForCurrentFrame):
  • platform/graphics/svg/SVGImage.h:
2:11 AM Changeset in webkit [18903] by mjs
  • 2 edits in trunk/WebKit

Reviewed by Eric.


<rdar://problem/4887781> Seed: Repro Safari crash in -[WebHTMLRepresentation receivedData:withDataSource:] (music.aol.com)

  • WebView/WebDataSource.mm: (-[WebDataSource _receivedData:]): Protect self against destruction partway through this method.
1:58 AM Changeset in webkit [18902] by rwlbuis
  • 2 edits in trunk/WebCore

Reviewed by Eric

Add proper condition for rx attribute negative check.

1:14 AM Changeset in webkit [18901] by lars
  • 25 edits
    1 delete in trunk

Add a message handler to DumpRenderTree that will
suppress all debug output coming from qDebug() statements.
Like this we can get rid of all the noise coming
from the notImplemented() macro when running the
layout tests. You can get it back by adding -v to
DumpRenderTree's command line.

Changed run-webkit-tests slightly, so we by default
don't fail anymore when our output differs from what
is generated on the Mac. Added a --strict option, so
that we can still see this cases and fix them one by one.

change all notImplemented() macros to use qDebug
instead of fprintf(stder, ...)

12:13 AM Changeset in webkit [18900] by beidson
  • 3 edits in trunk/WebCore

WebCore:

Reviewed by Oliver

A few enhancements to the HTTP Auth architecture

  • platform/network/ProtectionSpace.cpp: (WebCore::ProtectionSpace::isProxy): (WebCore::ProtectionSpace::receivesCredentialSecurely):
  • platform/network/ProtectionSpace.h:

Jan 16, 2007:

8:47 PM Changeset in webkit [22779] by aliceli1
  • 3 edits in branches/WindowsMerge/WebKitWin

Added stubs to keep the boomer build from failing.

  • WebEditorClient.cpp: (WebEditorClient::didWriteSelectionToPasteboard):
  • WebEditorClient.h:
8:37 PM Changeset in webkit [18899] by aliceli1
  • 13 edits in trunk

WebCore:

Reviewed by harrison.

Fixed <rdar://problem/4921134> WebKit needs extensible cut/copy to allow additional types to be written to pasteboard

  • WebCore.xcodeproj/project.pbxproj:
  • bridge/EditorClient.h:
  • editing/Editor.cpp: (WebCore::Editor::copy): (WebCore::Editor::didWriteSelectionToPasteboard):
  • editing/Editor.h:
  • platform/graphics/svg/SVGImageEmptyClients.h: (WebCore::SVGEmptyEditorClient::didWriteSelectionToPasteboard):

WebKit:

Reviewed by harrison.

Fixed <rdar://problem/4921134> WebKit needs extensible cut/copy to allow additional types to be written to pasteboard

  • DefaultDelegates/WebDefaultEditingDelegate.m: (-[WebDefaultEditingDelegate webView:didWriteSelectionToPasteboard:]):
  • WebCoreSupport/WebEditorClient.h:
  • WebCoreSupport/WebEditorClient.mm: (WebEditorClient::didWriteSelectionToPasteboard):
  • WebKit.xcodeproj/project.pbxproj:
  • WebView/WebEditingDelegate.h:

WebKitQt:

Added these stubs to keep the Qt build from failing.

  • WebCoreSupport/EditorClientQt.cpp: (WebCore::EditorClientQt::didWriteSelectionToPasteboard):
  • WebCoreSupport/EditorClientQt.h:
7:26 PM Changeset in webkit [22778] by beidson
  • 3 edits in branches/WindowsMerge/WebCore

Further BString improvements/bugfixes

6:34 PM Changeset in webkit [22777] by beidson
  • 3 edits in branches/WindowsMerge/WebCore

Added == and != operators to BString

6:13 PM Changeset in webkit [18898] by zbujtas
  • 2 edits in S60/branches/3.1m/WebCore

2007-01-15 w3liu <​wei.liu@nokia.com>

Reviewed by Zalan Bujtas <​zbujtas@gmail.com>.
DESC: [S60] Some text shown as mess code when openning website www.163.com
​http://bugs.webkit.org/show_bug.cgi?id=12286

WARNING: NO TEST CASES ADDED OR CHANGED

  • khtml/ecma/xmlhttprequest.cpp: (KJS::XMLHttpRequest::XMLHttpRequest): (KJS::XMLHttpRequest::~XMLHttpRequest): (KJS::XMLHttpRequest::abort): (KJS::XMLHttpRequest::slotData):
6:04 PM Changeset in webkit [18897] by zbujtas
  • 2 edits in S60/branches/3.1m/WebCore

2007-01-16 w3liu <​wei.liu@nokia.com>

Reviewed by Zalan Bujtas <​zbujtas@gmail.com>
DESC: TSW Error ID: ELZG-6WUELW [S60] Browser is closed when attempting to cancel reloading saved deck.
​http://bugs.webkit.org/show_bug.cgi?id=12292

WARNING: NO TEST CASES ADDED OR CHANGED

  • bridge/WebCoreBridge.cpp: (CWebCoreBridge::GetPageDataL):
5:59 PM Changeset in webkit [18896] by zbujtas
  • 2 edits in S60/trunk/WebCore

2007-01-16 w3liu <​wei.liu@nokia.com>

Reviewed by Zalan Bujtas (​zbujtas@gmail.com).
DESC: TSW Error ID: ELZG-6WUELW [S60] Browser is closed when attempting to cancel reloading saved deck.
​http://bugs.webkit.org/show_bug.cgi?id=12292

WARNING: NO TEST CASES ADDED OR CHANGED

  • bridge/WebCoreBridge.cpp: (CWebCoreBridge::GetPageDataL):
5:10 PM Changeset in webkit [18895] by justing
  • 6 edits
    6 adds in trunk

LayoutTests:

Reviewed by harrison


<rdar://problem/4747450>
Gmail Editor: Caret doesn't always position itself after list marker
<rdar://problem/4917055>
GMail Editor: A hang can occur when attempting to apply list styling to selected indented text

Added:

  • editing/execCommand/4747450-expected.checksum: Added.
  • editing/execCommand/4747450-expected.png: Added.
  • editing/execCommand/4747450-expected.txt: Added.
  • editing/execCommand/4747450.html: Added.
  • editing/execCommand/4917055-expected.txt: Added.
  • editing/execCommand/4917055.html: Added. Fixed:
  • editing/execCommand/create-list-with-hr-expected.checksum:
  • editing/execCommand/create-list-with-hr-expected.png:
  • editing/execCommand/create-list-with-hr-expected.txt:

WebCore:

Reviewed by harrison


<rdar://problem/4747450>
Gmail Editor: Caret doesn't always position itself after list marker
<rdar://problem/4917055>
GMail Editor: A hang can occur when attempting to apply list styling to selected indented text

  • editing/TextIterator.cpp: (WebCore::TextIterator::rangeFromLocationAndLength): Use VisiblePositions to compute the end of a run for emitted '\n's, because it's often wrong in that case.
4:40 PM Changeset in webkit [18894] by darin
  • 2 edits in trunk/WebCore
  • quick lame attempt to get Qt compiling again
  • platform/qt/ContextMenuQt.cpp: Removed show() function.
4:07 PM Changeset in webkit [18893] by bdash
  • 2 edits in trunk/WebCore

2007-01-17 Nicholas Shanks <​contact@nickshanks.com>

Reviewed by Darin.

  • platform/network/ResourceRequest.h: (WebCore::ResourceRequest::setHTTPAccept): Correct argument name.
3:44 PM Changeset in webkit [22776] by sfalken
  • 1 edit in branches/WindowsMerge/WebKitWin/WebKit.vcproj/VERSION

Bump version for submit

3:34 PM Changeset in webkit [18892] by sfalken
  • 1 copy in tags/Safari-521.32.12b

New tag.

2:02 PM Changeset in webkit [18891] by brmorris
  • 26 edits
    2 adds in S60/trunk

2007-01-16 yaharon <​yael.aharon@nokia.com>

Reviewed by Sachin and Sriram, landed by Bradley.
DESC: Event handling enhancements
​http://bugs.webkit.org/show_bug.cgi?id=12226

2:00 PM Changeset in webkit [22775] by bdakin
  • 7 edits in branches/WindowsMerge

Fix for <rdar://problem/4874565> Draw context menus using CG

1:55 PM Changeset in webkit [18890] by bdakin
  • 1 edit in trunk/WebCore/ChangeLog

Clean up merge conflict spew.

1:40 PM Changeset in webkit [18889] by bdakin
  • 4 edits in trunk/WebCore

Reviewed by Adam.

Get rid of show() and hide() since they don't do anything.

  • page/ContextMenuController.cpp: (WebCore::ContextMenuController::handleContextMenuEvent):
  • platform/ContextMenu.h:
  • platform/mac/ContextMenuMac.mm:
1:33 PM Changeset in webkit [22774] by sfalken
  • 2 edits in branches/WindowsMerge/WebKitWin
1:07 PM Changeset in webkit [18888] by ggaren
  • 3 edits in trunk/JavaScriptCore

Reviewed by Darin Adler.


Added re-entrency checking to GC allocation and collection. It is an error
to allocate or collect from within a collection. We've had at least one
case of each bug in the past.


Added a comment to the API header, explaining that API clients must not
make this mistake, either.


Layout tests and JS tests pass.

  • API/JSObjectRef.h:
  • kjs/collector.cpp: (KJS::GCLock::GCLock): (KJS::GCLock::~GCLock): (KJS::Collector::allocate): (KJS::Collector::collect):
7:38 AM Changeset in webkit [18887] by darin
  • 4 edits in trunk/LayoutTests
  • updated one more test results affected by Rob's coordinates patch
  • svg/W3C-SVG-1.1/coords-units-03-b-expected.checksum: Regenerated.
  • svg/W3C-SVG-1.1/coords-units-03-b-expected.png: Ditto.
  • svg/W3C-SVG-1.1/coords-units-03-b-expected.txt: Ditto.
7:03 AM Changeset in webkit [18886] by lars
  • 24 edits
    7 adds
    2 deletes in trunk

Added the start of a public API to the Qt build
of WebKit. Currently we have QWebPage and QWebFrame
with a few small methods.

Converted the QtLauncher and DumpRenderTree
to use the new public API.

4:25 AM Changeset in webkit [18885] by lars
  • 4 edits in trunk/WebCore

Reviewed by Zack

Looks like ResourceHandle _has_ to do everything
asynchronously if we don't want to run into
lots of trouble in the loader.

Implement support data: URLs.

Remove some unused Qt only code in ResourceHandleClient.h

3:40 AM Changeset in webkit [18884] by eseidel
  • 3 edits in trunk/WebCore

2007-01-16 Eric Seidel <​eric@webkit.org>

Reviewed by mitz.

Refactor drawTiled implementations in preparation for combining them.
Removed unused RoundTile code path.

  • platform/graphics/cg/ImageCG.cpp: (WebCore::drawPattern): (WebCore::BitmapImage::drawTiled): (WebCore::caculatePatternScale):
12:14 AM Changeset in webkit [18883] by darin
  • 3 edits in trunk/WebCore

Reviewed by Mark Rowe.

  • removed unused field from loader
  • loader/loader.cpp: (WebCore::Loader::numRequests): (WebCore::Loader::cancelRequests):
  • loader/loader.h:
Note: See TracTimeline for information about the timeline view.