ddkilzer [Fri, 30 Jun 2006 04:24:40 +0000 (04:24 +0000)]
WebKitTools:
Reviewed by Darin.
Bug 9615: Buildbot configuration should be in SVN repository
http://bugzilla.opendarwin.org/show_bug.cgi?id=9615
Import BuildBot configuration files as used by build.webkit.org.
auth.py has been stubbed out so that slave passwords are not disclosed.
* BuildSlaveSupport/build.webkit.org-config/Makefile: Added.
* BuildSlaveSupport/build.webkit.org-config/buildbot.css: Added.
* BuildSlaveSupport/build.webkit.org-config/buildbot.tac: Added.
* BuildSlaveSupport/build.webkit.org-config/master.cfg: Added.
* BuildSlaveSupport/build.webkit.org-config/webkit/__init__.py: Added.
* BuildSlaveSupport/build.webkit.org-config/webkit/auth.py: Added.
* BuildSlaveSupport/build.webkit.org-config/webkit/basesteps.py: Added.
* BuildSlaveSupport/build.webkit.org-config/webkit/builders.py: Added.
* BuildSlaveSupport/build.webkit.org-config/webkit/factories.py: Added.
* BuildSlaveSupport/build.webkit.org-config/webkit/schedulers.py: Added.
* BuildSlaveSupport/build.webkit.org-config/webkit/status.py: Added.
* BuildSlaveSupport/build.webkit.org-config/webkit/steps.py: Added.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15099
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Fri, 30 Jun 2006 04:20:18 +0000 (04:20 +0000)]
WebKitTools:
Reviewed by Darin.
Bug 9614: Nightly builds should notify user if a newer build is available
http://bugzilla.opendarwin.org/show_bug.cgi?id=9614
* BuildSlaveSupport/build-launcher-app: Write revision number to WebKit.app's VERSION file.
* BuildSlaveSupport/build-launcher-dmg: Pass revision number to server-side nightly maintenance
script so it can keep track of the latest revision.
* Scripts/run-webkit-tests: Pass build-dumprendertree's exit status through correctly.
* WebKitLauncher/VERSION: Added. Placeholder for SVN revision number
* WebKitLauncher/WebKitLauncher.xcodeproj/project.pbxproj:
* WebKitLauncher/start.html: Retrieve revision number from VERSION file, and pass it through to
the nightly start page.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15098
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Fri, 30 Jun 2006 02:56:14 +0000 (02:56 +0000)]
Reviewed by Anders.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9622
showModalDialog returnValue ignored, function result is always "undefined"
* manual-tests/modal-dialog.html: Added.
* manual-tests/show-modal-dialog-test.html: Added.
* page/Frame.cpp: (WebCore::Frame::clear): Call KJSProxy::clear with the
clearWindowProperties parameter instead of not calling it at all when
clearWindowProperties is false.
* bindings/js/kjs_proxy.h: Add boolean clearWindowProperties parameter.
* bindings/js/kjs_proxy.cpp: (WebCore::KJSProxy::clear): Pass clearWindowProperties
variable through to Window::clear instead of not calling it at all.
* bindings/js/kjs_window.h: Add boolean clearWindowProperties parameter.
* bindings/js/kjs_window.cpp: (KJS::Window::clear): If clearWindowProperties
is false, do only the returnValue work, not the rest of the work.
* manual-tests/modal-dialog.html: Added.
* manual-tests/show-modal-dialog-test.html: Added.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15097
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
mjs [Thu, 29 Jun 2006 23:32:47 +0000 (23:32 +0000)]
Reviewed by Geoff.
- add headerdoc comments to some of the new JS API headers
* API/JSBase.h:
* API/JSValueRef.h:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15096
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
bdakin [Thu, 29 Jun 2006 23:18:23 +0000 (23:18 +0000)]
Reviewed by Hyatt.
Fix for <rdar://problem/
4450615> Add composite mode to background
styles
-webkit-background-composite will set a composite mode on a
background image.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::):
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSPropertyNames.in: Add -webkit-background-composite
* css/CSSValueKeywords.in: Add all possible values (clear, copy,
source-over, source-in, source-out, source-atop, destination-over,
destination-in, destination-out, destination-atop, xor, plus-
darker, highlight, and plus-lighter)
* css/cssparser.cpp:
(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::parseBackgroundProperty):
* css/cssstyleselector.cpp:
(WebCore::CSSStyleSelector::applyProperty):
(WebCore::CSSStyleSelector::mapBackgroundComposite):
* css/cssstyleselector.h:
* platform/GraphicsContext.cpp:
(WebCore::GraphicsContext::drawTiledImage):
* platform/GraphicsContext.h: drawTiledImage() now takes a
composite operator.
* platform/Image.h: drawTiled() now takes a composite operator.
* platform/cairo/ImageCairo.cpp:
(WebCore::Image::drawTiled): Set the composite operator.
* platform/mac/ImageMac.mm:
(WebCore::Image::drawTiled): Set the composite operator.
* platform/win/TemporaryLinkStubs.cpp:
(Image::drawTiled):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::paintBackgroundExtended): Send the composite
operator in the render style over to the graphics context.
* rendering/render_style.cpp:
(WebCore::BackgroundLayer::BackgroundLayer):
(WebCore::BackgroundLayer::operator=):
(WebCore::BackgroundLayer::operator==):
(WebCore::BackgroundLayer::fillUnsetProperties):
(WebCore::BackgroundLayer::cullEmptyLayers):
* rendering/render_style.h:
(WebCore::BackgroundLayer::backgroundComposite):
(WebCore::BackgroundLayer::isBackgroundCompositeSet):
(WebCore::BackgroundLayer::setBackgroundComposite):
(WebCore::BackgroundLayer::clearBackgroundComposite):
(WebCore::RenderStyle::backgroundComposite):
(WebCore::RenderStyle::initialBackgroundComposite):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15095
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
tomernic [Thu, 29 Jun 2006 19:28:20 +0000 (19:28 +0000)]
Reviewed by John Sullivan.
WebHistoryItem now supports getting and setting arbitrary properties via _transientPropertyForKey: and
-_setTransientProperty:forKey:.
For now, these properties do not persist with the rest of the history data. They are intended to hold transient
per-history-item state, which is something that was until now difficult for a WebKit client app to do.
* History/WebHistoryItemPrivate.h:
* History/WebHistoryItem.m:
(-[WebHistoryItemPrivate dealloc]):
(-[WebHistoryItem _transientPropertyForKey:]):
(-[WebHistoryItem _setTransientProperty:forKey:]):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15094
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Thu, 29 Jun 2006 18:18:42 +0000 (18:18 +0000)]
Reviewed by Harrison.
Smart insert and delete, continuous spell checking and autoscroll
can now be used for any WebView, not just editable ones. All of
these make sense for documents that might contain content editable
areas or our new text fields. Autoscroll is usefull for dragging
for file input controls also.
Added a SPI to toggle WebViews tab key behavior, tabKeyCyclesThroughElements.
WebHTMLView's _interceptEditingKeyEvent now uses WebView's
tabKeyCyclesThroughElements state to determine whether or not
to process tab key events. The idea here is that tabKeyCyclesThroughElements
will be YES when this WebView is being used in a browser, and we
desire the behavior where tab moves to the next element in tab order.
If tabKeyCyclesThroughElements is NO, it is likely that the WebView
is being embedded as the whole view, as in Mail, and tabs should input
tabs as expected in a text editor. Using Option-Tab always cycles
through elements.
* WebView/WebHTMLRepresentation.m:
(-[WebHTMLRepresentation finishedLoadingWithDataSource:]):
* WebView/WebHTMLView.m:
(-[NSArray _interceptEditingKeyEvent:]):
* WebView/WebView.m:
(-[WebViewPrivate init]):
(-[WebView _autoscrollForDraggingInfo:timeDelta:]):
(-[WebView _shouldAutoscrollForDraggingInfo:]):
(-[WebView validateUserInterfaceItem:]):
(-[WebView toggleSmartInsertDelete:]):
(-[WebView toggleContinuousSpellChecking:]):
(-[WebView setTabKeyCyclesThroughElements:]):
(-[WebView tabKeyCyclesThroughElements]):
* WebView/WebViewPrivate.h:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15093
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Thu, 29 Jun 2006 15:28:48 +0000 (15:28 +0000)]
WebCore:
Reviewed by Darin.
http://bugzilla.opendarwin.org/show_bug.cgi?id=9649
The no-svg build broke after r15091
* bindings/objc/DOMInternal.h: Added #if XPATH_SUPPORT/#endif.
* bindings/objc/DOMInternal.mm: Ditto.
(raiseDOMException):
* bindings/objc/DOMXPath.h: Ditto.
* bindings/objc/DOMXPath.mm: Ditto.
* bindings/objc/DOMXPathInternal.h: Ditto.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15092
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
andersca [Thu, 29 Jun 2006 07:54:49 +0000 (07:54 +0000)]
WebCore:
2006-06-29 Anders Carlsson <acarlsson@apple.com>
Reviewed by Tim O.
<rdar://problem/
4599760> Objective-C XPath support
* WebCore.xcodeproj/project.pbxproj:
Add files.
* bindings/objc/DOMInternal.h:
* bindings/objc/DOMInternal.mm:
(raiseDOMException):
Handle XPath exceptions
* bindings/objc/DOMXPath.h: Added.
(-[DOMXPathNSResolver lookupNamespaceURI:]):
* bindings/objc/DOMXPath.mm: Added.
(-[DOMNativeXPathNSResolver dealloc]):
(-[DOMNativeXPathNSResolver finalize]):
(-[DOMNativeXPathNSResolver _xpathNSResolver]):
(-[DOMNativeXPathNSResolver _initWithXPathNSResolver:]):
(+[DOMNativeXPathNSResolver _xpathNSResolverWith:]):
(-[DOMNativeXPathNSResolver lookupNamespaceURI:]):
(-[DOMXPathResult dealloc]):
(-[DOMXPathResult finalize]):
(-[DOMXPathResult resultType]):
(-[DOMXPathResult numberValue]):
(-[DOMXPathResult stringValue]):
(-[DOMXPathResult booleanValue]):
(-[DOMXPathResult singleNodeValue]):
(-[DOMXPathResult invalidIteratorState]):
(-[DOMXPathResult snapshotLength]):
(-[DOMXPathResult iterateNext]):
(-[DOMXPathResult snapshotItem:]):
(-[DOMXPathResult _xpathResult]):
(-[DOMXPathResult _initWithXPathResult:]):
(+[DOMXPathResult _xpathResultWith:]):
(-[DOMXPathExpression evaluate:::]):
(-[DOMXPathExpression _xpathExpression]):
(-[DOMXPathExpression _initWithXPathExpression:]):
(+[DOMXPathExpression _xpathExpressionWith:]):
(-[DOMDocument createExpression::]):
(-[DOMDocument createNSResolver:]):
(-[DOMDocument evaluate:::::]):
* bindings/objc/DOMXPathInternal.h: Added.
Add Objective-C XPath API.
WebKit:
2006-06-29 Anders Carlsson <acarlsson@apple.com>
Reviewed by Tim O.
* WebKit.xcodeproj/project.pbxproj:
Add DOMXPath.h header.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15091
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
hyatt [Thu, 29 Jun 2006 00:32:55 +0000 (00:32 +0000)]
WebCore:
Add support for custom highlighting to paint at the root line box level
(and to inflate the box it uses so that it can go outside the bounds of
the line if necessary).
Reviewed by harrison
* bridge/mac/FrameMac.h:
* bridge/mac/FrameMac.mm:
(WebCore::FrameMac::customHighlightLineRect):
(WebCore::FrameMac::paintCustomHighlight):
* bridge/mac/WebCoreFrameBridge.h:
* rendering/InlineTextBox.cpp:
(WebCore::InlineTextBox::paintCustomHighlight):
* rendering/RootInlineBox.cpp:
(WebCore::RootInlineBox::paintCustomHighlight):
(WebCore::RootInlineBox::paint):
* rendering/RootInlineBox.h:
WebKit:
Fix custom highlighting so that you can paint the entire line (and go
outside the bounds of the line).
Reviewed by harrison
* WebCoreSupport/WebFrameBridge.m:
(-[WebFrameBridge customHighlightRect:forLine:]):
(-[WebFrameBridge paintCustomHighlight:forBox:onLine:behindText:entireLine:]):
* WebKit.xcodeproj/project.pbxproj:
* WebView/WebHTMLViewPrivate.h:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15090
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
mjs [Thu, 29 Jun 2006 00:20:12 +0000 (00:20 +0000)]
Reviewed by Anders.
- fix Frame leak on layout tests
* WebCoreSupport/WebPageBridge.m:
(-[WebPageBridge outerView]): Return WebFrameView for main frame instead
of WebView to avoid reference cycle between WebView and Page.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15089
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
lweintraub [Wed, 28 Jun 2006 23:44:03 +0000 (23:44 +0000)]
LayoutTests:
Reviewed by justin
<rdar://problem/
4592005> FormatBlock inserts in the wrong place, disregarding editable content
<rdar://problem/
4590763> FormatBlock for paragraph style doesn't work
* editing/execCommand/format-block-with-braces-expected.checksum: Added.
* editing/execCommand/format-block-with-braces-expected.png: Added.
* editing/execCommand/format-block-with-braces-expected.txt: Added.
* editing/execCommand/format-block-with-braces.html: Added.
* editing/execCommand/indent-list-item-expected.checksum: Added.
* editing/execCommand/indent-list-item-expected.png: Added.
* editing/execCommand/indent-list-item-expected.txt: Added.
* editing/execCommand/indent-list-item.html: Added.
* editing/execCommand/indent-selection-expected.checksum: Added.
* editing/execCommand/indent-selection-expected.png: Added.
* editing/execCommand/indent-selection-expected.txt: Added.
* editing/execCommand/indent-selection.html: Added.
* editing/execCommand/outdent-selection-expected.checksum: Added.
* editing/execCommand/outdent-selection-expected.png: Added.
* editing/execCommand/outdent-selection-expected.txt: Added.
* editing/execCommand/outdent-selection.html: Added.
WebCore:
Reviewed by justin
<rdar://problem/
4592005> FormatBlock inserts in the wrong place, disregarding editable content
<rdar://problem/
4590763> FormatBlock for paragraph style doesn't work
* dom/Node.cpp:
(WebCore::Node::rootEditableElement): Return inline nodes
* editing/FormatBlockCommand.cpp:
(WebCore::FormatBlockCommand::doApply): Keep from inserting before editable content
* editing/JSEditor.cpp: Support html braces (<>) around tag names
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15088
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
justing [Wed, 28 Jun 2006 23:18:01 +0000 (23:18 +0000)]
Reviewed by levi
svg/custom/text-hit-test.svg was failing
* rendering/RenderContainer.cpp:
(WebCore::RenderContainer::positionForCoordinates): Don't skip child
renderers that have a height of 0 when looking for the closest child
renderer, they may contain content that overflows.
* rendering/RenderText.cpp:
(WebCore::RenderText::positionForCoordinates): Fixed typos.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15087
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
hyatt [Wed, 28 Jun 2006 20:43:18 +0000 (20:43 +0000)]
Make the view source document properly handle <script>. We don't syntax
highlight the JS yet, but it shows up properly now.
Reviewed by john
* WebCore.xcodeproj/project.pbxproj:
* html/HTMLTokenizer.cpp:
(WebCore::HTMLTokenizer::scriptHandler):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15086
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Wed, 28 Jun 2006 20:32:45 +0000 (20:32 +0000)]
JavaScriptCore:
Prefer the Stabs debugging symbols format until DWARF bugs are fixed.
* JavaScriptCore.xcodeproj/project.pbxproj:
JavaScriptGlue:
Prefer the Stabs debugging symbols format until DWARF bugs are fixed.
* JavaScriptGlue.xcodeproj/project.pbxproj:
WebCore:
Prefer the Stabs debugging symbols format until DWARF bugs are fixed.
* WebCore.xcodeproj/project.pbxproj:
WebKit:
Prefer the Stabs debugging symbols format until DWARF bugs are fixed.
* WebKit.xcodeproj/project.pbxproj:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15085
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
justing [Wed, 28 Jun 2006 20:18:11 +0000 (20:18 +0000)]
Reviewed by levi
Fixed EditActions for InsertHorizontalRule and RemoveFormat.
* editing/JSEditor.cpp:
* editing/ReplaceSelectionCommand.cpp:
(WebCore::ReplaceSelectionCommand::ReplaceSelectionCommand):
(WebCore::ReplaceSelectionCommand::editingAction):
* editing/ReplaceSelectionCommand.h:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15084
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
aliceli1 [Wed, 28 Jun 2006 20:09:50 +0000 (20:09 +0000)]
added google spreadsheets and yahoo mail beta, moved google calendar and yahoo preview to alumni list.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15083
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
lweintraub [Wed, 28 Jun 2006 18:58:19 +0000 (18:58 +0000)]
Build Fix
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15082
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
lweintraub [Wed, 28 Jun 2006 18:36:46 +0000 (18:36 +0000)]
Build Fix. Reviewed by NOBODY
* editing/IndentOutdentCommand.cpp:
(WebCore::IndentOutdentCommand::indentRegion):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15081
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
lweintraub [Wed, 28 Jun 2006 18:17:58 +0000 (18:17 +0000)]
LayoutTests:
Reviewed by justin
http://bugzilla.opendarwin.org/show_bug.cgi?id=7568
Bug 7568: Implement Indent/Outdent
* editing/execCommand/indent-list-item-expected.checksum: Added.
* editing/execCommand/indent-list-item-expected.png: Added.
* editing/execCommand/indent-list-item-expected.txt: Added.
* editing/execCommand/indent-list-item.html: Added.
* editing/execCommand/indent-selection-expected.checksum: Added.
* editing/execCommand/indent-selection-expected.png: Added.
* editing/execCommand/indent-selection-expected.txt: Added.
* editing/execCommand/indent-selection.html: Added.
* editing/execCommand/outdent-selection.html: Added.
WebCore:
Reviewed by justin
http://bugzilla.opendarwin.org/show_bug.cgi?id=7568
Bug 7568: Implement Indent/Outdent
* WebCore.vcproj/WebCore/WebCore.vcproj: Added IndentOutdentCommand.h/cpp
* WebCore.xcodeproj/project.pbxproj: Ditto.
* editing/IndentOutdentCommand.cpp: Added.
(WebCore::IndentOutdentCommand::IndentOutdentCommand):
(WebCore::enclosingListOrBlockquote):
(WebCore::IndentOutdentCommand::splitTreeToNode): Splits the DOM tree from a
descendent node to an ending ancestor, duplicating nodes when necessary. Returns
the last node split. Used to insert blockquotes at the topmost level.
(WebCore::IndentOutdentCommand::indentRegion):
(WebCore::IndentOutdentCommand::outdentParagraph):
(WebCore::IndentOutdentCommand::outdentRegion):
(WebCore::IndentOutdentCommand::doApply):
* editing/IndentOutdentCommand.h: Added.
(WebCore::IndentOutdentCommand::):
* editing/JSEditor.cpp: Added the execCommands Indent and Outdent
* editing/htmlediting.cpp:
(WebCore::enclosingNodeWithTag): Finds the enclosing node with any specified tag.
(WebCore::enclosingListChild): Added checks for nill and editable boundaries.
(WebCore::outermostEnclosingListChild): Returns the highest ancestor list child.
(WebCore::highestAncestor): Added.
(WebCore::createElement): Changed the passed in string to a const reference.
* editing/htmlediting.h:
WebKit:
Reviewed by justin
http://bugzilla.opendarwin.org/show_bug.cgi?id=7568
Bug 7568: Implement Indent/Outdent
Added undo action strings and enum values
* English.lproj/Localizable.strings:
* WebCoreSupport/WebFrameBridge.m:
(-[WebFrameBridge nameForUndoAction:]):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15080
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Wed, 28 Jun 2006 15:42:13 +0000 (15:42 +0000)]
LayoutTests:
Reviewed by Darin and Hyatt.
- test for http://bugzilla.opendarwin.org/show_bug.cgi?id=9561
REGRESSION: Content property on :before of button causes hang on click or hover
* fast/forms/button-generated-content-expected.checksum: Added.
* fast/forms/button-generated-content-expected.png: Added.
* fast/forms/button-generated-content-expected.txt: Added.
* fast/forms/button-generated-content.html: Added.
WebCore:
Reviewed by Darin and Hyatt.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9561
REGRESSION: Content property on :before of button causes hang on click or hover
Test: fast/forms/button-generated-content.html
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::setStyle):
* rendering/RenderButton.cpp:
(WebCore::RenderButton::updatePseudoChild): Added. Calls updatePseudoChildForObject()
on the button itself or on its inner div, if it has one.
* rendering/RenderButton.h:
(WebCore::RenderButton::createsAnonymousWrapper): Added. Returns true.
* rendering/RenderContainer.cpp:
(WebCore::RenderContainer::pseudoChild): Added. Returns the child containing
generated content of the requested type if there is one, or otherwise the
child before/after which such generated content should come.
(WebCore::RenderContainer::updatePseudoChild): Removed the child parameter and moved the guts
of this method into updatePseudoChildForObject, which this method calls unless the object
is an anonymous wrapper.
(WebCore::RenderContainer::updatePseudoChildForObject): Added. Does the what
updatePseudoChildForObject did previously, but fetches the pseudo style from
the styledObject parameter.
* rendering/RenderContainer.h:
* rendering/RenderInline.cpp:
(WebCore::RenderInline::setStyle):
(WebCore::RenderInline::addChildToFlow):
(WebCore::RenderInline::splitInlines):
(WebCore::RenderInline::splitFlow):
* rendering/RenderObject.h:
(WebCore::RenderObject::createsAnonymousWrapper): Added. This method replaces
allowsReusingAnonymousChild and returns the opposite of what that method used to
return. Returns true if the object creates an anonymous child to contain all its
other children.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15079
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Wed, 28 Jun 2006 14:39:50 +0000 (14:39 +0000)]
Reviewed by Darin.
Improved test to check offsetX/Y after dispatching instead of before.
* fast/dom/window-xy-properties.html:
* fast/dom/window-xy-properties-expected.txt:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15078
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
beidson [Wed, 28 Jun 2006 07:18:28 +0000 (07:18 +0000)]
WebCore:
Reviewed by Maciej
Hookup the new semi-functional SQLite icon database.
For now, it is living side-by-side with the old DB so one can compare the
two for debugging purposes. Also, it is disabled (in WebKit) by default unless you
compile with ICONDEBUG #defined.
* WebCore.xcodeproj/project.pbxproj: Added file SiteIcon.cpp
* bridge/mac/WebCoreIconDatabaseBridge.h:
* bridge/mac/WebCoreIconDatabaseBridge.mm:
(-[WebCoreIconDatabaseBridge privateBrowsingEnabled]):
(-[WebCoreIconDatabaseBridge iconForPageURL:withSize:]):
(-[WebCoreIconDatabaseBridge iconURLForPageURL:]):
(-[WebCoreIconDatabaseBridge _setIconData:forIconURL:]):
* icon/IconDatabase.cpp:
(WebCore::IconDatabase::isValidDatabase):
(WebCore::IconDatabase::recreateDatabase):
(WebCore::IconDatabase::createPrivateTables):
(WebCore::IconDatabase::deletePrivateTables):
(WebCore::IconDatabase::imageDataForIconID):
(WebCore::IconDatabase::imageDataForIconURL):
(WebCore::IconDatabase::imageDataForPageURL):
(WebCore::IconDatabase::setPrivateBrowsingEnabled):
(WebCore::IconDatabase::iconForPageURL):
(WebCore::IconDatabase::iconURLForURL):
(WebCore::IconDatabase::setIconDataForIconURL):
(WebCore::IconDatabase::establishIconIDForEscapedIconURL):
(WebCore::IconDatabase::setHaveNoIconForIconURL):
(WebCore::IconDatabase::setIconURLForPageURL):
(WebCore::IconDatabase::hasIconForIconURL):
* icon/IconDatabase.h:
(WebCore::SiteIcon::getIconURL):
(WebCore::IconDatabase::getPrivateBrowsingEnabled):
* icon/SiteIcon.cpp: Added.
(SiteIcon::SiteIcon):
(SiteIcon::~SiteIcon):
(SiteIcon::getImage):
(SiteIcon::resetExpiration):
(SiteIcon::getExpiration):
WebKit:
Reviewed by Maciej
Hookup the new semi-functional SQLite icon database.
For now, it is living side-by-side with the old DB so one can compare the
two for debugging purposes. Also, it is disabled (in WebKit) by default unless you
compile with ICONDEBUG #defined.
Note: To repeat that, if you want to try the new DB, #define ICONDEBUG (WebKitPrefix.h is a good place to do it)
* Misc/WebIconDatabase.m:
(-[NSMutableDictionary iconForURL:withSize:cache:]):
(-[NSMutableDictionary iconURLForURL:]):
(-[NSMutableDictionary retainIconForURL:]):
(-[NSMutableDictionary releaseIconForURL:]):
(-[WebIconDatabase _setHaveNoIconForIconURL:]):
(-[WebIconDatabase _setIconURL:forURL:]):
(-[WebIconDatabase _hasIconForIconURL:]):
(-[WebIconDatabase _resetCachedWebPreferences:]):
* Misc/WebIconLoader.m:
(-[WebIconLoader didFinishLoading]):
* WebKit.xcodeproj/project.pbxproj:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15077
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Wed, 28 Jun 2006 03:52:29 +0000 (03:52 +0000)]
LayoutTests:
Reviewed by Darin.
Updated test for http://bugzilla.opendarwin.org/show_bug.cgi?id=9605
* fast/dom/window-xy-properties.html:
* fast/dom/window-xy-properties-expected.txt:
WebCore:
Reviewed by Darin.
http://bugzilla.opendarwin.org/show_bug.cgi?id=9605
REGRESSION: pageX and pageY are both 0 for events created with initMouseEvent
Calculate pageX/pageY and clientX/clientY values when using initMouseEvent.
* dom/MouseEvent.cpp:
(WebCore::MouseEvent::initMouseEvent):
* dom/MouseRelatedEvent.cpp:
(WebCore::MouseRelatedEvent::initCoordinates):
* dom/MouseRelatedEvent.h:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15076
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Wed, 28 Jun 2006 03:32:56 +0000 (03:32 +0000)]
LayoutTests:
Reviewed by Darin.
- test for http://bugzilla.opendarwin.org/show_bug.cgi?id=9317
REGRESSION: crash in HTML tokenizer at Japanese Apple support page
* fast/tokenizer/write-unclosed-script-expected.txt: Added.
* fast/tokenizer/write-unclosed-script.html: Added.
WebCore:
Reviewed by Maciej and Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9317
REGRESSION: crash in HTML tokenizer at Japanese Apple support page
Test: fast/tokenizer/write-unclosed-script.html
* html/HTMLTokenizer.cpp:
(WebCore::HTMLTokenizer::scriptHandler): Changed to reset the inScript flag
and scriptCodeSize before possible script execution, but not afterwards.
This way, if script execution write()s a <script> tag without closing it,
the tokenizer is left in inScript state. Added code to set the requestingScript
flag around the request for the cached script, to let notifiyFinished() know
that it's being called under scriptHandler().
(WebCore::HTMLTokenizer::scriptExecution): Removed code that saved and restored
the inScript flag. This function is always entered now with inScript being false.
(WebCore::HTMLTokenizer::notifyFinished): Changed to use the new requestingScript
state flag instead of the inScript flag, which is always false now when entering
this function.
* html/HTMLTokenizer.h:
Added the requestingScript state bit, used to tell notifyFinished() that it
is being called under scriptHandler() (which happens when the script is already
in cache).
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15075
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
justing [Wed, 28 Jun 2006 02:26:52 +0000 (02:26 +0000)]
Fixed some typos.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15074
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
justing [Wed, 28 Jun 2006 02:23:56 +0000 (02:23 +0000)]
LayoutTests:
Reviewed by levi
* editing/execCommand/remove-formatting-expected.checksum: Added.
* editing/execCommand/remove-formatting-expected.png: Added.
* editing/execCommand/remove-formatting-expected.txt: Added.
* editing/execCommand/remove-formatting.html: Added.
WebCore:
Reviewed by levi
* editing/JSEditor.cpp: Added RemoveFormat.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15073
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
kdecker [Wed, 28 Jun 2006 02:17:43 +0000 (02:17 +0000)]
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15072
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Wed, 28 Jun 2006 00:18:15 +0000 (00:18 +0000)]
Reviewed by Tim O.
<rdar://problem/
4448350> Deprecated ObjC language API used in JavaScriptCore, WebCore, WebKit and WebBrowser
Switch to the new ObjC 2 API, ifdefed the old code around OBJC_API_VERSION so it still works on Tiger.
Removed the use of the old stringWithCString, switched to the new Tiger version that accepts an encoding.
Lots of code style cleanup.
* JavaScriptCore.xcodeproj/project.pbxproj:
* bindings/objc/objc_class.h:
* bindings/objc/objc_class.mm:
(KJS::Bindings::ObjcClass::~ObjcClass):
(KJS::Bindings::_createClassesByIsAIfNecessary):
(KJS::Bindings::ObjcClass::classForIsA):
(KJS::Bindings::ObjcClass::name):
(KJS::Bindings::ObjcClass::methodsNamed):
(KJS::Bindings::ObjcClass::fieldNamed):
(KJS::Bindings::ObjcClass::fallbackObject):
* bindings/objc/objc_header.h:
* bindings/objc/objc_instance.h:
* bindings/objc/objc_instance.mm:
(ObjcInstance::ObjcInstance):
(ObjcInstance::~ObjcInstance):
(ObjcInstance::operator=):
(ObjcInstance::begin):
(ObjcInstance::end):
(ObjcInstance::getClass):
(ObjcInstance::invokeMethod):
(ObjcInstance::invokeDefaultMethod):
(ObjcInstance::setValueOfField):
(ObjcInstance::supportsSetValueOfUndefinedField):
(ObjcInstance::setValueOfUndefinedField):
(ObjcInstance::getValueOfField):
(ObjcInstance::getValueOfUndefinedField):
(ObjcInstance::defaultValue):
(ObjcInstance::stringValue):
(ObjcInstance::numberValue):
(ObjcInstance::booleanValue):
(ObjcInstance::valueOf):
* bindings/objc/objc_runtime.h:
* bindings/objc/objc_runtime.mm:
(ObjcMethod::ObjcMethod):
(ObjcMethod::name):
(ObjcMethod::getMethodSignature):
(ObjcMethod::setJavaScriptName):
(ObjcField::name):
(ObjcField::type):
(ObjcField::valueFromInstance):
(convertValueToObjcObject):
(ObjcField::setValueToInstance):
(ObjcArray::operator=):
(ObjcArray::setValueAt):
(ObjcArray::valueAt):
(ObjcFallbackObjectImp::ObjcFallbackObjectImp):
(ObjcFallbackObjectImp::callAsFunction):
(ObjcFallbackObjectImp::defaultValue):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15071
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
andersca [Tue, 27 Jun 2006 23:59:31 +0000 (23:59 +0000)]
2006-06-28 Anders Carlsson <acarlsson@apple.com>
Reviewed by Geoff.
http://bugzilla.opendarwin.org/show_bug.cgi?id=8636
REGRESSION: JavaScript access to Java applet causes hang (_webViewURL not implemented)
* bindings/jni/jni_objc.mm:
(KJS::Bindings::dispatchJNICall):
Just pass nil as the calling URL. This will cause the Java plugin to use the URL of the page
containing the applet (which is what we used to do).
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15070
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
andersca [Tue, 27 Jun 2006 23:58:52 +0000 (23:58 +0000)]
2006-06-28 Anders Carlsson <acarlsson@apple.com>
Reviewed by Darin.
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
Move FIXME comment here.
* page/DOMWindow.cpp:
(WebCore::DOMWindow::getComputedStyle):
Don't do a null check on the renderer here. Everything's handled by
CSSComputedStyleDeclaration anyway.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15069
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
justing [Tue, 27 Jun 2006 22:56:22 +0000 (22:56 +0000)]
Reviewed by darin
* rendering/RenderContainer.cpp: Fixed #include formatting.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15068
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
justing [Tue, 27 Jun 2006 22:26:22 +0000 (22:26 +0000)]
LayoutTests:
Reviewed by levi
<rdar://problem/
4599735> Clicking out of a table at the extremes of an editable document doesn't appear to work
Added:
* editing/selection/click-before-and-after-table-expected.txt: Added.
* editing/selection/click-before-and-after-table.html: Added.
Updated to reflect fix:
* editing/selection/move-by-line-001-expected.checksum:
* editing/selection/move-by-line-001-expected.png:
WebCore:
Reviewed by levi
<rdar://problem/
4599735> Clicking out of a table at the extremes of an editable document doesn't appear to work
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::positionForCoordinates): In the fall through case,
where we can't find a child block that contains (x, y), pass to RenderFlow
instead of the last visible child block.
* rendering/RenderContainer.cpp:
(WebCore::RenderContainer::positionForCoordinates): Don't pass to the closest
child renderer if this is a table and (x, y) is outside the table. Return
a VisiblePosition before or after the table.
Fixed the code that finds the closest child renderer.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15067
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
hyatt [Tue, 27 Jun 2006 22:04:02 +0000 (22:04 +0000)]
Preserve entity text in view source mode. This patch doesn't highlight the
the entities yet, but it at least prevents their resolution and removal.
Reviewed by darin
* WebCore.xcodeproj/project.pbxproj:
* html/HTMLTokenizer.cpp:
(WebCore::HTMLTokenizer::parseEntity):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15066
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
andersca [Tue, 27 Jun 2006 21:30:58 +0000 (21:30 +0000)]
WebCore:
2006-06-27 Anders Carlsson <acarlsson@apple.com>
Reviewed by Geoff.
http://bugzilla.opendarwin.org/show_bug.cgi?id=9623
REGRESSION: getComputedStyle should do a relayout
* page/DOMWindow.cpp:
(WebCore::DOMWindow::getComputedStyle):
Do a relayout.
LayoutTests:
2006-06-27 Anders Carlsson <acarlsson@apple.com>
Reviewed by Geoff.
http://bugzilla.opendarwin.org/show_bug.cgi?id=9623
REGRESSION: getComputedStyle should do a relayout
* fast/css/getComputedStyle-relayout-expected.txt: Added.
* fast/css/getComputedStyle-relayout.html: Added.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15065
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Tue, 27 Jun 2006 21:13:00 +0000 (21:13 +0000)]
Adding symbols for debug builds.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15064
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Tue, 27 Jun 2006 20:53:05 +0000 (20:53 +0000)]
Reviewed by Darin.
<rdar://problem/
4406785> Add an export file to TOT JavaScriptCore like the Safari-2-0-branch
* JavaScriptCore.exp: Added.
* JavaScriptCore.xcodeproj/project.pbxproj:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15058
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
mjs [Tue, 27 Jun 2006 20:36:20 +0000 (20:36 +0000)]
Reviewed by Adele
- fixed <rdar://problem/
4550473> Reproducible hang on www.digg.com
* dom/NodeList.cpp:
(WebCore::NodeList::recursiveItem): Make NodeList caching also
work for backwards iteration - if the requested index is before
the last cached, but closer to it than to the start of the list,
then search backwards from there.
(WebCore::NodeList::itemForwardsFromCurrent): Split this out as a
helper method.
(WebCore::NodeList::itemBackwardsFromCurrent): New helper, similar
to the above.
* dom/NodeList.h:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15057
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
beidson [Tue, 27 Jun 2006 19:07:34 +0000 (19:07 +0000)]
Reviewed by Levi
Some SQLDatabase cleanup (style stuff) and enhancements
* icon/SQLDatabase.cpp:
(SQLDatabase::open):
(SQLDatabase::executeCommand):
(SQLDatabase::returnsAtLeastOneResult): Accessor to see if this query returns results
(SQLDatabase::lastInsertRowID): Accessor for the rowid of the last insert
* icon/SQLDatabase.h:
(WebCore::SQLStatement::prepareAndStep):
* icon/SQLStatement.cpp:
(WebCore::SQLStatement::bindInt64):
-All of the following used to require the user to manually prepare and step a statement
before asking for the result. Now it will prepare and step itself so you can use a temporary
object to get a quick result
(WebCore::SQLStatement::getColumnName):
(WebCore::SQLStatement::getColumnName16):
(WebCore::SQLStatement::getColumnText):
(WebCore::SQLStatement::getColumnText16):
(WebCore::SQLStatement::getColumnDouble):
(WebCore::SQLStatement::getColumnInt):
(WebCore::SQLStatement::getColumnInt64):
(WebCore::SQLStatement::getColumnBlobAsVector):
(WebCore::SQLStatement::getColumnBlob):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15056
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
adachan [Tue, 27 Jun 2006 18:15:31 +0000 (18:15 +0000)]
2006-06-27 Ada Chan <adachan@apple.com>
Reviewed by sfalken.
Fixed build.
* WebCore.vcproj/WebCore/WebCore.vcproj:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15055
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
bdakin [Tue, 27 Jun 2006 17:56:09 +0000 (17:56 +0000)]
Reviewed by Hyatt.
Fix for <rdar://problem/
4602635> Media queries must be extended to
address CSS pixel scaling factor
* css/MediaFeatureNames.h: Add -webkit-device-pixel-ratio, -webkit-
min-device-pixel-ratio, and -webkit-max-device-pixel-ratio to the
list of feature names.
* css/MediaQueryEvaluator.cpp:
(WebCore::cmpvalue): Added a version of cmpvalue that compares
floats.
(WebCore::device_pixel_ratioMediaFeatureEval):
(WebCore::min_device_pixel_ratioMediaFeatureEval):
(WebCore::max_device_pixel_ratioMediaFeatureEval):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15054
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Tue, 27 Jun 2006 17:44:50 +0000 (17:44 +0000)]
Reviewed by Geoff.
- finish with http://bugzilla.opendarwin.org/show_bug.cgi?id=9606
REGRESSION: some callers are still using clientX/Y to get pageX/Y
* rendering/RenderFrameSet.cpp: (WebCore::RenderFrameSet::userResize):
Change to use pageX/Y, but this really doesn't matter since framesets
can't be scrolled -- it's the frames that are scrolled. So this change
has no user-visible effect, but it's still better this way.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15053
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Tue, 27 Jun 2006 17:22:38 +0000 (17:22 +0000)]
LayoutTests:
Tests for fix for 2/3 of http://bugzilla.opendarwin.org/show_bug.cgi?id=9606
REGRESSION: some callers are still using clientX/Y to get pageX/Y
* fast/events/anchor-image-scrolled-x-y-expected.txt: Added.
* fast/events/anchor-image-scrolled-x-y.html: Added.
* fast/events/input-image-scrolled-x-y-expected.txt: Added.
* fast/events/input-image-scrolled-x-y.html: Added.
WebCore:
Reviewed by Geoff.
- fix 2/3 of http://bugzilla.opendarwin.org/show_bug.cgi?id=9606
REGRESSION: some callers are still using clientX/Y to get pageX/Y
Test: fast/events/anchor-image-scrolled-x-y.html
Test: fast/events/input-image-scrolled-x-y.html
* html/HTMLAnchorElement.cpp: (WebCore::HTMLAnchorElement::defaultEventHandler): Use pageX/Y.
* html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::defaultEventHandler): Ditto.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15052
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
beidson [Tue, 27 Jun 2006 07:35:01 +0000 (07:35 +0000)]
Reviewed by Maciej
Added functionality to WebCore::Image to get multiple representations/sizes into TIFF data
* platform/mac/ImageMac.mm:
(WebCore::Image::getTIFFRepresentation):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15051
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
hyatt [Mon, 26 Jun 2006 23:53:02 +0000 (23:53 +0000)]
Fix for 9538, support syntax highlighted HTML source. Lots of loose
ends still (script/style/comments/doctype/entities/accurate whitespace).
Reviewed by darin
* DerivedSources.make:
* WebCore.xcodeproj/project.pbxproj:
* bridge/mac/WebCoreFrameBridge.h:
* bridge/mac/WebCoreFrameBridge.mm:
(-[WebCoreFrameBridge setInViewSourceMode:]):
(-[WebCoreFrameBridge inViewSourceMode]):
* css/cssstyleselector.cpp:
(WebCore::CSSStyleSelector::loadDefaultStyle):
(WebCore::CSSStyleSelector::matchUARules):
* css/cssstyleselector.h:
* css/view-source.css: Added.
* dom/Document.cpp:
(WebCore::Document::implicitClose):
* dom/xml_tokenizer.h:
(WebCore::Tokenizer::Tokenizer):
(WebCore::Tokenizer::~Tokenizer):
(WebCore::Tokenizer::inViewSourceMode):
(WebCore::Tokenizer::setInViewSourceMode):
* html/HTMLAttributeNames.in:
* html/HTMLFrameElement.cpp:
(WebCore::HTMLFrameElement::init):
(WebCore::HTMLFrameElement::parseMappedAttribute):
(WebCore::HTMLFrameElement::attach):
* html/HTMLFrameElement.h:
(WebCore::HTMLFrameElement::viewSourceMode):
* html/HTMLTokenizer.cpp:
(WebCore::HTMLTokenizer::HTMLTokenizer):
(WebCore::HTMLTokenizer::scriptHandler):
(WebCore::HTMLTokenizer::scriptExecution):
(WebCore::HTMLTokenizer::parseTag):
(WebCore::HTMLTokenizer::continueProcessing):
(WebCore::HTMLTokenizer::write):
(WebCore::HTMLTokenizer::timerFired):
(WebCore::HTMLTokenizer::end):
(WebCore::HTMLTokenizer::processToken):
(WebCore::HTMLTokenizer::notifyFinished):
* html/HTMLTokenizer.h:
* html/HTMLViewSourceDocument.cpp: Added.
(WebCore::HTMLViewSourceDocument::HTMLViewSourceDocument):
(WebCore::HTMLViewSourceDocument::createTokenizer):
(WebCore::HTMLViewSourceDocument::addViewSourceToken):
(WebCore::HTMLViewSourceDocument::addViewSourceAttribute):
(WebCore::HTMLViewSourceDocument::addSpanWithClassName):
* html/HTMLViewSourceDocument.h: Added.
* page/Frame.cpp:
(WebCore::Frame::begin):
(WebCore::Frame::inViewSourceMode):
(WebCore::Frame::setInViewSourceMode):
* page/Frame.h:
* page/FramePrivate.h:
(WebCore::FramePrivate::FramePrivate):
* rendering/RenderPartObject.cpp:
(WebCore::RenderPartObject::updateWidget):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15050
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
hyatt [Mon, 26 Jun 2006 23:48:37 +0000 (23:48 +0000)]
Fix for 9538, support syntax highlighting for HTML source.
Reviewed by darin
* WebKit.xcodeproj/project.pbxproj:
* WebView/WebView.m:
(-[WebView _setInViewSourceMode:]):
(-[WebView _inViewSourceMode]):
* WebView/WebViewPrivate.h:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15049
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
andersca [Mon, 26 Jun 2006 21:26:43 +0000 (21:26 +0000)]
2006-06-26 Anders Carlsson <acarlsson@apple.com>
Reviewed by Darin.
http://bugzilla.opendarwin.org/show_bug.cgi?id=9031
REGRESSION: Crash when closing tabs on newegg.com
* page/Frame.cpp:
(WebCore::Frame::pageDestroyed):
* page/Frame.h:
Add new method which is invoked when the frame's page has been destroyed.
Currently this function disassociates the JS Window object with the frame,
stopping any running JS timers.
* page/Page.cpp:
(WebCore::Page::~Page):
Call pageDestroyed on all frames.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15048
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Mon, 26 Jun 2006 16:38:55 +0000 (16:38 +0000)]
Reviewed and reworked by Darin.
Code cleanup, calculate clientX/clientY in one central place.
* dom/EventTargetNode.h:
* dom/EventTargetNode.cpp: (WebCore::EventTargetNode::dispatchMouseEvent):
Renamed parameters to pageX/Y since they aren't clientX/Y values.
Removed code to compute clientX/Y (was a bit confused in naming before),
because that's now done inside MouseEvent.
* dom/MouseEvent.h:
* dom/MouseEvent.cpp:
(WebCore::MouseEvent::MouseEvent): Removed clientX/Y parameters.
(WebCore::MouseEvent::initMouseEvent): Added comments about problems I noticed.
(WebCore::MouseEvent::which): Fixed comment that confusingly mentioned KHTML,
when it was talking about the DOM standard.
* dom/MouseRelatedEvent.h:
* dom/MouseRelatedEvent.cpp:
(WebCore::contentsX): Added helper function.
(WebCore::contentsY): Ditto.
(WebCore::MouseRelatedEvent::MouseRelatedEvent): Removed clientX/Y parameter,
and added code to compute clientX/Y based on pageX/Y.
(WebCore::MouseRelatedEvent::receivedTarget): Fixed comments.
* dom/WheelEvent.h:
* dom/WheelEvent.cpp: (WebCore::WheelEvent::WheelEvent): Removed clientX/Y
parameter.
* page/FrameView.cpp: (WebCore::FrameView::dispatchDragEvent): Removed the code
to compute coordinates, and renamed the confusingly named clientPos (which was
a page-relative position) to contentsPos.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15047
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Mon, 26 Jun 2006 15:26:50 +0000 (15:26 +0000)]
LayoutTests:
Reviewed by Darin.
* LayoutTests/fast/events/objc-event-api-expected.txt: Updated to
reflect flipped coordinates dumped by the layout tests.
WebKitTools:
Reviewed by Darin.
Bug 9579: LayoutTests/fast/events/objc-event-api.html failing due to screenY output
http://bugzilla.opendarwin.org/show_bug.cgi?id=9579
Report screenY values as (height of zero screen - screenY)
* DumpRenderTree/EventSendingController.m: (-[EventSendingController handleEvent:]):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15046
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Mon, 26 Jun 2006 05:09:12 +0000 (05:09 +0000)]
Reviewed by Geoff.
Bug 9591: [Drosera] breakpoints should be dragable and deletable by dragging off the gutter
http://bugzilla.opendarwin.org/show_bug.cgi?id=9591
Makes breakpoints dragable. If dragged off the gutter they are deleted.
* Drosera/debugger.js:
* Drosera/viewer.css:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15045
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Mon, 26 Jun 2006 04:51:30 +0000 (04:51 +0000)]
LayoutTests:
Reviewed by Darin.
Tests for http://bugzilla.opendarwin.org/show_bug.cgi?id=9185
* fast/forms/input-appearance-height-expected.png: Added.
* fast/forms/input-appearance-height-expected.txt: Added.
* fast/forms/input-appearance-height.html: Added.
Tests the height attribute on all input types
* fast/forms/input-type-change-expected.checksum: Added.
* fast/forms/input-type-change-expected.png: Added.
* fast/forms/input-type-change-expected.txt: Added.
* fast/forms/input-type-change.html: Added.
Tests changing an input type=TEXT to type=IMAGE with
a height and width specified on the element.
* fast/forms/input-type-change2-expected.checksum: Added.
* fast/forms/input-type-change2-expected.png: Added.
* fast/forms/input-type-change2-expected.txt: Added.
* fast/forms/input-type-change2.html: Added.
Tests changing an input type=IMAGE to type=TEXT with
a height and width specified on the element.
WebCore:
Reviewed and tweaked by Darin.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=9185
REGRESSION: UserID field appears with an incorrect height
on americanexpresslogin page
- Height and Width attribute only applies to IMAGE and HIDDEN
types.
* dom/NamedMappedAttrMap.h:
(WebCore::NamedMappedAttrMap::getAttributeItem):
Added to static_cast method.
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::setInputType):
Update the style decl if input type has changed.
(WebCore::HTMLInputElement::mapToEntry):
Only Cache height and width if it is an IMAGE or HIDDEN
(WebCore::HTMLInputElement::parseMappedAttribute):
Check type before adding height and width style.
(WebCore::HTMLInputElement::attach):
Remove code for discarding width.
* html/HTMLInputElement.h:
(WebCore::HTMLInputElement::respectHeightAndWidthAttrs):
New method added to report if this input type uses height & width.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15044
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ggaren [Mon, 26 Jun 2006 03:12:18 +0000 (03:12 +0000)]
Reviewed by Adele.
- Added JSConstructorMake to match JSFunctionMake, along with test code.
[ I checked in the ChangeLog before without the actual files. ]
* API/JSObjectRef.cpp:
(JSConstructorMake):
* API/JSObjectRef.h:
* API/testapi.c:
(myConstructor_callAsConstructor):
(main):
* API/testapi.js:
* ChangeLog:
* JavaScriptCore.xcodeproj/project.pbxproj: Moved testapi.c to the testapi
target -- this was an oversight in my earlier check-in.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15043
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Mon, 26 Jun 2006 02:17:48 +0000 (02:17 +0000)]
Reviewed by Darin.
Bug 9568: assertion failure in Safari after quitting Drosera
http://bugzilla.opendarwin.org/show_bug.cgi?id=9568
Call switchToServerNamed:nil and not removeLister to make sure
the server object is set to nil to prevent further removeListener calls.
* Drosera/DebuggerDocument.m:
(-[DebuggerDocument applicationTerminating:]):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15042
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Mon, 26 Jun 2006 00:28:58 +0000 (00:28 +0000)]
LayoutTests:
Reviewed by Darin.
* fast/forms/textarea-hard-linewrap-empty.html: Submit form without setTimeout().
* fast/forms/textarea-hard-linewrap.html: Ditto.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15041
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sun, 25 Jun 2006 23:55:53 +0000 (23:55 +0000)]
WebCore:
Reviewed by Darin.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=9537
REGRESSION: Reproducible crash in WebCore::RenderTextField::textWithHardLineBreaks()
Test: fast/forms/textarea-hard-linewrap-empty.html
* rendering/RenderTextField.cpp:
(WebCore::RenderTextField::textWithHardLineBreaks): Added null check for
m_div->firstChild()->renderer().
LayoutTests:
Reviewed by Darin.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=9537
REGRESSION: Reproducible crash in WebCore::RenderTextField::textWithHardLineBreaks()
* fast/forms/textarea-hard-linewrap-empty-expected.txt: Added.
* fast/forms/textarea-hard-linewrap-empty.html: Added.
* fast/forms/textarea-hard-linewrap.html: Reduced timeout from 200ms to 0ms and
renamed function from foo() to submitForm().
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15040
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sun, 25 Jun 2006 23:28:14 +0000 (23:28 +0000)]
WebKitTools:
Reviewed by Darin.
* Scripts/svn-apply: Speed up isDirectoryEmptyForRemoval() by returning as soon as we find
that the directory is not empty instead of reading in every single file and directory first,
then checking the count. Avoid warning in removeDirectoriesIfNeeded() if $svnOutput is not
defined.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15039
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sun, 25 Jun 2006 23:17:09 +0000 (23:17 +0000)]
WebCore:
Reviewed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9388
REGRESSION: outlines don't get erased when removed in some cases
Test: fast/repaint/outline-shrinking.html
* rendering/RenderObject.cpp:
(WebCore::RenderObject::setStyle): Always repaint with the old style if the outline is
becoming narrower, even if we are going to relayout.
LayoutTests:
Reviewed by Darin.
- repaint test for http://bugzilla.opendarwin.org/show_bug.cgi?id=9388
REGRESSION: outlines don't get erased when removed in some cases
* fast/repaint/outline-shrinking-expected.checksum: Added.
* fast/repaint/outline-shrinking-expected.png: Added.
* fast/repaint/outline-shrinking-expected.txt: Added.
* fast/repaint/outline-shrinking.html: Added.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15038
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Sun, 25 Jun 2006 20:56:34 +0000 (20:56 +0000)]
Build fix for for GCC 5406
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15034
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Sun, 25 Jun 2006 20:37:47 +0000 (20:37 +0000)]
Rolls out a build fix I made on Friday, this was correctly fixed by adding a long version of operator[] to Vector.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15033
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sun, 25 Jun 2006 20:37:37 +0000 (20:37 +0000)]
WebCore:
Reviewed by Darin.
Fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=8707:
event.clientX and event.clientY should be relative to the viewport, not the canvas
Make sure clientX and clientY take into account the scroll offsets
of the view.
* dom/EventTargetNode.cpp:
(WebCore::EventTargetNode::dispatchMouseEvent):
* dom/MouseEvent.cpp:
(WebCore::MouseEvent::MouseEvent):
* dom/MouseEvent.h:
* dom/MouseRelatedEvent.cpp:
(WebCore::MouseRelatedEvent::MouseRelatedEvent):
(WebCore::MouseRelatedEvent::initCoordinates):
* dom/MouseRelatedEvent.h:
* dom/WheelEvent.cpp:
(WebCore::WheelEvent::WheelEvent):
* page/FrameView.cpp:
(WebCore::FrameView::dispatchDragEvent):
LayoutTests:
Reviewed by Darin.
Test cases for
<http://bugzilla.opendarwin.org/show_bug.cgi?id=8707>
event.clientX and event.clientY should be relative to the viewport, not the canvas
* fast/dom/window-xy-properties-expected.txt: Added.
* fast/dom/window-xy-properties.html: Added.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15032
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Sun, 25 Jun 2006 20:32:37 +0000 (20:32 +0000)]
Leak fix
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15031
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
darin [Sun, 25 Jun 2006 20:01:03 +0000 (20:01 +0000)]
* Scripts/svn-apply: Tweak comments.
* Scripts/svn-create-patch: Ditto.
* Scripts/svn-unapply: Ditto.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15030
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sun, 25 Jun 2006 19:42:14 +0000 (19:42 +0000)]
WebKitTools:
Formatting fix per Bug 9571 Comment #2. Reviewed by NOBODY.
http://bugzilla.opendarwin.org/show_bug.cgi?id=9571#c2
* Scripts/svn-apply: Formatting fix.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15029
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sun, 25 Jun 2006 19:39:22 +0000 (19:39 +0000)]
WebKitTools:
Reviewed by Darin.
http://bugzilla.opendarwin.org/show_bug.cgi?id=9571
Teach svn-apply and svn-unapply to handle directory adds and removes better
* Scripts/svn-apply: Handle directory adds more intelligently. Handle directory removes.
* Scripts/svn-unapply: Handle undoing both directory adds and removes.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15028
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Sun, 25 Jun 2006 18:59:28 +0000 (18:59 +0000)]
bugzilla.opendarwin.org/show_bug.cgi?id=9560
This fix converted the resize corner image to a B/W tiff.
The pixel tests showed 0.0% changes but still said it had differences.
Updating the tests that use CSS3 resize or textarea.
* fast/forms/floating-textfield-relayout-expected.checksum:
* fast/forms/form-element-geometry-expected.checksum:
* fast/forms/form-element-geometry-expected.png:
* fast/forms/negativeLineHeight-expected.checksum:
* fast/forms/negativeLineHeight-expected.png:
* fast/forms/textAreaLineHeight-expected.checksum:
* fast/forms/textAreaLineHeight-expected.png:
* fast/forms/textarea-setinnerhtml-expected.checksum:
* fast/forms/textarea-setinnerhtml-expected.png:
* fast/overflow/overflow-x-y-expected.checksum:
* fast/overflow/overflow-x-y-expected.png:
* fast/table/003-expected.checksum:
* fast/table/003-expected.png:
* tables/mozilla/bugs/bug194024-expected.checksum:
* tables/mozilla/bugs/bug194024-expected.png:
* tables/mozilla/bugs/bug30559-expected.checksum:
* tables/mozilla/bugs/bug30559-expected.png:
* tables/mozilla/bugs/bug30692-expected.checksum:
* tables/mozilla/bugs/bug30692-expected.png:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15027
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Sun, 25 Jun 2006 17:53:03 +0000 (17:53 +0000)]
JavaScriptCore:
Reviewed by Darin.
Bug 9574: Drosera should show inline scripts within the original HTML
http://bugzilla.opendarwin.org/show_bug.cgi?id=9574
Pass the starting line number and error message to the debugger.
* kjs/debugger.cpp:
(Debugger::sourceParsed):
* kjs/debugger.h:
* kjs/function.cpp:
(KJS::GlobalFuncImp::callAsFunction):
* kjs/function_object.cpp:
(FunctionObjectImp::construct):
* kjs/interpreter.cpp:
(KJS::Interpreter::evaluate):
WebCore:
Reviewed by Darin.
Bug 9574: Drosera should show inline scripts within the original HTML
http://bugzilla.opendarwin.org/show_bug.cgi?id=9574
* Pass the starting line number and error message to the debugger.
* Call parsedSource even if there was a script parse error so the
debugger can show the parse error.
* Pass NSURL objects to the ObjC delegate for the script URLs.
* bridge/mac/WebCoreScriptDebugger.h:
* bridge/mac/WebCoreScriptDebugger.mm:
(toNSURL):
(WebCoreScriptDebuggerImp::sourceParsed):
WebKit:
Reviewed by Darin.
Bug 9574: Drosera should show inline scripts within the original HTML
http://bugzilla.opendarwin.org/show_bug.cgi?id=9574
* Adds a new version of the didParseSource delegate callback with base line number.
* Adds a new delegate callback for when a script fails to parse.
* These new callbacks use NSURLs for the url parameter.
* Adds a new script listener callback to notify when the main resource loads.
* Adds a WebScriptErrorDomian and other keys for use with NSError.
* DefaultDelegates/WebDefaultScriptDebugDelegate.m:
(-[WebDefaultScriptDebugDelegate webView:didParseSource:baseLineNumber:fromURL:sourceId:forWebFrame:]):
(-[WebDefaultScriptDebugDelegate webView:failedToParseSource:baseLineNumber:fromURL:withError:forWebFrame:]):
* DefaultDelegates/WebScriptDebugServer.h:
* DefaultDelegates/WebScriptDebugServer.m:
(-[WebScriptDebugServer webView:didLoadMainResourceForDataSource:]):
(-[WebScriptDebugServer webView:didParseSource:baseLineNumber:fromURL:sourceId:forWebFrame:]):
(-[WebScriptDebugServer webView:failedToParseSource:baseLineNumber:fromURL:withError:forWebFrame:]):
* DefaultDelegates/WebScriptDebugServerPrivate.h:
* WebKit.exp:
* WebView/WebDataSource.m:
(-[WebDataSource _setPrimaryLoadComplete:]):
* WebView/WebScriptDebugDelegate.h:
* WebView/WebScriptDebugDelegate.m:
(-[WebScriptCallFrame parsedSource:fromURL:sourceId:startLine:errorLine:errorMessage:]):
WebKitTools:
Reviewed by Darin.
Bug 9574: Drosera should show inline scripts within the original HTML
http://bugzilla.opendarwin.org/show_bug.cgi?id=9574
Refactor the JavaScript code to have a distinction between files
and scripts. Show the script in the context of the HTML file if
it's URL is the same as the frame's main resource. At the time of
the disParseScript callback the main resource might not be completely
loaded, but Drosera needs to show whatever we have at the time. Once
the main resource is finished, update the file source and reload the file.
* Drosera/DebuggerDocument.m:
(-[DebuggerDocument pause]):
(-[DebuggerDocument webView:didLoadMainResourceForDataSource:]):
(-[DebuggerDocument webView:didParseSource:baseLineNumber:fromURL:sourceId:forWebFrame:]):
(-[DebuggerDocument webView:failedToParseSource:baseLineNumber:fromURL:withError:forWebFrame:]):
* Drosera/debugger.css:
* Drosera/debugger.js:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15026
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sun, 25 Jun 2006 06:01:41 +0000 (06:01 +0000)]
WebKitTools:
Reviewed by Darin.
http://bugzilla.opendarwin.org/show_bug.cgi?id=9570
Teach prepare-ChangeLog to operate on a list of files or directories
* Scripts/prepare-ChangeLog: Accept a list of files and/or directories when running.
* Scripts/svn-create-patch: Code refactoring.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15025
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sun, 25 Jun 2006 03:02:01 +0000 (03:02 +0000)]
WebCore:
Build fix. Reviewed by NOBODY.
- Backed out patch for http://bugzilla.opendarwin.org/show_bug.cgi?id=8707
event.clientX and event.clientY should be relative to the viewport, not the canvas
* dom/EventTargetNode.cpp:
(WebCore::EventTargetNode::dispatchMouseEvent):
* dom/MouseEvent.cpp:
(WebCore::MouseEvent::MouseEvent):
* dom/MouseEvent.h:
* dom/MouseRelatedEvent.cpp:
(WebCore::MouseRelatedEvent::MouseRelatedEvent):
(WebCore::MouseRelatedEvent::initCoordinates):
* dom/MouseRelatedEvent.h:
* dom/WheelEvent.cpp:
(WebCore::WheelEvent::WheelEvent):
LayoutTests:
Build fix. Reviewed by NOBODY.
- Backed out patch for http://bugzilla.opendarwin.org/show_bug.cgi?id=8707
event.clientX and event.clientY should be relative to the viewport, not the canvas
* fast/dom/window-xy-properties-expected.txt: Removed.
* fast/dom/window-xy-properties.html: Removed.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15024
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sun, 25 Jun 2006 02:09:48 +0000 (02:09 +0000)]
WebCore:
Reviewed by Darin. Patch by Rob Buis.
Fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=8707:
event.clientX and event.clientY should be relative to the viewport, not the canvas
Make sure clientX and clientY take into account the scroll offsets
of the view.
Test: fast/dom/window-xy-properties.html
* dom/EventTargetNode.cpp:
(WebCore::EventTargetNode::dispatchMouseEvent):
* dom/MouseEvent.cpp:
(WebCore::MouseEvent::MouseEvent):
* dom/MouseEvent.h:
* dom/MouseRelatedEvent.cpp:
(WebCore::MouseRelatedEvent::MouseRelatedEvent):
(WebCore::MouseRelatedEvent::initCoordinates):
* dom/MouseRelatedEvent.h:
* dom/WheelEvent.cpp:
(WebCore::WheelEvent::WheelEvent):
LayoutTests:
Reviewed by Darin. Patch by Rob Buis.
Test cases for
<http://bugzilla.opendarwin.org/show_bug.cgi?id=8707>
event.clientX and event.clientY should be relative to the viewport, not the canvas
* fast/dom/window-xy-properties-expected.txt: Added.
* fast/dom/window-xy-properties.html: Added.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15023
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sun, 25 Jun 2006 01:53:50 +0000 (01:53 +0000)]
LayoutTests:
Reviewed by Darin. Patch by James G. Speth.
- Tests for http://bugzilla.opendarwin.org/show_bug.cgi?id=8843
add a way to build Objective-C test cases in HTML
* plugins/jsobjc-dom-wrappers-expected.txt: Added.
* plugins/jsobjc-dom-wrappers.html: Added.
* plugins/jsobjc-simple-expected.txt: Added.
* plugins/jsobjc-simple.html: Added.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15022
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sat, 24 Jun 2006 21:55:46 +0000 (21:55 +0000)]
LayoutTests:
Reviewed by Darin. Patch by Rob Buis.
Test cases for
<http://bugzilla.opendarwin.org/show_bug.cgi?id=9352>
Webkit doesn't respect stroke-width in SVG for smaller <circle>s
* svg/custom/gradient-stroke-width-expected.checksum: Added.
* svg/custom/gradient-stroke-width-expected.png: Added.
* svg/custom/gradient-stroke-width-expected.txt: Added.
* svg/custom/gradient-stroke-width.svg: Added.
WebCore:
Reviewed by Darin. Patch by Rob Buis.
http://bugzilla.opendarwin.org/show_bug.cgi?id=9352
Webkit doesn't respect stroke-width in SVG for smaller <circle>s
Make sure the stroke path is set before applying gradient transforms.
* kcanvas/device/quartz/KRenderingPaintServerGradientQuartz.mm:
(WebCore::KRenderingPaintServerGradientQuartz::setup):
(WebCore::KRenderingPaintServerGradientQuartz::renderPath):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15021
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sat, 24 Jun 2006 21:08:31 +0000 (21:08 +0000)]
WebKitTools:
Reviewed by Darin. Patch by James G. Speth.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=8843
add a way to build Objective-C test cases in HTML
Tests running in DumpRenderTree can choose to relax some restrictions of the JavaScript/Objective-C bridge
allowing more extensive testing of the Obj-C API. (and by more extensive, I mean this lets scripts do
pretty much whatever they want, including acting as delegates, generating events, instantiating obj-c
objects, etc... )
* DumpRenderTree/ObjCPlugin.h:
* DumpRenderTree/ObjCPlugin.m: this ability is only exposed to scripts running in DumpRenderTree
(+[ObjCPlugin isSelectorExcludedFromWebScript:]):
(+[ObjCPlugin webScriptNameForSelector:]):
(-[ObjCPlugin removeBridgeRestrictions:]): scripts call this to open up obj-c for extensive testing
(+[NSObject setAllowsScriptsFullAccess:]): makes NSObject allow all selectors to be invoked from script
(+[NSObject allowsScriptsFullAccess]):
(+[NSObject isSelectorExcludedFromWebScript:]): when full access is enabled, no selector is excluded
(+[NSObject webScriptNameForSelector:]): always return nil for the default selector mutation
(-[JSObjC classNameOfObject:isSelectorExcludedFromWebScript:]):
(+[JSObjC webScriptNameForSelector:]):
(-[JSObjC invokeDefaultMethodWithArguments:]): shortcut for accessing classes
(-[JSObjC lookUpClass:]): allow scripts to retrieve obj-c classes by name
(-[JSObjC log:]): access to NSLog function
(-[JSObjC retainObject:]): useful obj-c functions that are otherwise hard to reach from javascript
(-[JSObjC classOfObject:]):
(-[JSObjC classNameOfObject:]):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15020
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ap [Sat, 24 Jun 2006 21:07:35 +0000 (21:07 +0000)]
Rubber-stamped by Eric.
Add a -h (do not follow symlinks) option to ln in derived sources build script (without it,
a symlink was created inside the source directory on second build).
JavaScriptCore:
* JavaScriptCore.xcodeproj/project.pbxproj:
WebCore:
* WebCore.xcodeproj/project.pbxproj:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15019
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sat, 24 Jun 2006 20:51:21 +0000 (20:51 +0000)]
WebCore:
Reviewed by Darin. Patch by James G. Speth. Copyrights tweaked by ddkilzer.
Bug 8331: DOMNodeLists returned to Objective-C are not properly wrapped
http://bugzilla.opendarwin.org/show_bug.cgi?id=8331
DOMNodes were the only objects handled by createObjcInstanceForValue.
now it handles all of these: DOMNode, DOMNodeList, DOMNamedNodeMap, DOMStyleSheetList,
DOMStyleSheet, DOMMediaList, DOMCSSRuleList, DOMCSSRule, DOMCSSStyleDeclaration, and DOMCSSValue
* WebCore.xcodeproj/project.pbxproj: added file DOMCSSInternal.h to project
* bindings/js/kjs_css.cpp:
(KJS::): changed ClassInfo for DOMCSSStyleSheet to have DOMStyleSheet as its parent class
* bindings/objc/DOM.mm:
(-[DOMDocumentType :]): moved DOMNamedNodeMap (WebCoreInternal) interface to DOMInternal.h
* bindings/objc/DOMCSS.mm: moved WebCoreInternal interfaces to DOMCSSInternal.h
* bindings/objc/DOMCSSInternal.h: Added file, contains WebCoreInternal interface for CSS classes
* bindings/objc/DOMInternal.h: added DOMNamedNodeMap (WebCoreInternal) interface
* bindings/objc/DOMUtility.mm:
(KJS::ScriptInterpreter::createObjcInstanceForValue): support for more classes
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15018
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sat, 24 Jun 2006 20:01:04 +0000 (20:01 +0000)]
JavaScriptCore:
Reviewed by Timothy.
* Info.plist: Fixed copyright to include 2003-2006.
JavaScriptGlue:
Reviewed by Timothy.
* Info.plist: Fixed copyright to include 2003-2006.
WebCore:
Reviewed by Timothy.
* Info.plist: Fixed copyright to include 2003-2006.
* manual-tests/NPN_Invoke/Info.plist: Added copyright statement.
WebKit:
Reviewed by Timothy.
* Info.plist: Fixed copyright to include 2003-2006.
WebKitTools:
Reviewed by Timothy.
* DrawTest/Info.plist: Added copyright statement.
* Drosera/Info.plist: Ditto.
* DumpRenderTree/TestNetscapePlugIn.subproj/Info.plist: Ditto.
* WebKitLauncher/Info.plist: Ditto.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15017
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sat, 24 Jun 2006 19:00:28 +0000 (19:00 +0000)]
WebKitTools:
Build fix. Reviewed by NOBODY.
- Backed out http://bugzilla.opendarwin.org/show_bug.cgi?id=7802
devenv.com not available in VC++ Express installations
* Scripts/webkitdirs.pm: Backed out previous change.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15016
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sat, 24 Jun 2006 18:45:27 +0000 (18:45 +0000)]
WebKitTools:
Reviewed by Maciej. Patch by Bjoern Graf.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=7802
devenv.com not available in VC++ Express installations
* Scripts/webkitdirs.pm: Make Windows build work with Visual C++ Express.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15015
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ap [Sat, 24 Jun 2006 16:02:10 +0000 (16:02 +0000)]
2006-06-24 Mitz Pettel <opendarwin.org@mitzpettel.com>
Reviewed by Darin, landed by ap.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9547
Resize corner does not track the mouse accurately
Test: fast/css/resize-corner-tracking.html
* page/FrameView.cpp:
(WebCore::FrameView::handleMousePressEvent): Store the offset from the layer's
resizing corner where dragging has started.
(WebCore::FrameView::handleMouseMoveEvent): Pass the stored offset to the
layer's resize method.
* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::resize): Added an offset parameter, which is added to
the mouse location.
For objects that may have intrinsic margins applied to them,
added code to lift the margins into the object's style in order to keep them
constant.
Changed to take the content size as the base size if the box-sizing property
is content-box.
(WebCore::RenderLayer::offsetFromResizeCorner): Added. Returns the offset of the
given point from the corner that tracks the mouse when resizing.
* rendering/RenderLayer.h:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15014
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ap [Sat, 24 Jun 2006 15:52:01 +0000 (15:52 +0000)]
Reviewed by Darin, landed by ap.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9316
REGRESSION: text field width shrinks on first keystroke
Test: fast/forms/floating-textfield-relayout.html
* page/FrameView.cpp:
(WebCore::FrameView::layoutRoot): Added. During subtree relayout, returns
the subtree root. Otherwise returns 0.
(WebCore::FrameView::layoutPending): Made const.
* page/FrameView.h:
* rendering/RenderBox.cpp:
(WebCore::RenderBox::calcWidth): Added early return if this is the root
of a subtree being laid out.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15013
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ap [Sat, 24 Jun 2006 15:37:54 +0000 (15:37 +0000)]
Reviewed by Darin.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=9564
A bunch of fixes to run-webkit-httpd
- Add an -all-interfaces (-a) flag to bind to all interfaces, not just
127.0.0.1. Useful for testing with WinIE running on another machine;
- don't call checkFrameworks() - we do not need a built WebKit here;
- changed tabs to spaces;
- disable HTTP keepalive (since Apache doesn't spawn sub-processes in
interactive mode, they were a hassle when testing with several browsers, as one
had to wait for connection to expire);
- remove httpd.pid when done, so that Apache doesn't complain next time.
* Scripts/run-webkit-httpd:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15012
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ddkilzer [Sat, 24 Jun 2006 15:09:21 +0000 (15:09 +0000)]
LayoutTests:
Reviewed by Hyatt. Patch by Mitz.
- updated test results for http://bugzilla.opendarwin.org/show_bug.cgi?id=9531
Pages that need only a horizontal scroll bar get a vertical one as well
* css1/classification/white_space-expected.checksum:
* css1/classification/white_space-expected.png:
* css1/classification/white_space-expected.txt:
* editing/selection/select-all-004-expected.checksum:
* editing/selection/select-all-004-expected.png:
* editing/selection/select-all-004-expected.txt:
* editing/selection/unrendered-001-expected.checksum:
* editing/selection/unrendered-001-expected.png:
* editing/selection/unrendered-001-expected.txt:
* editing/selection/unrendered-002-expected.checksum:
* editing/selection/unrendered-002-expected.png:
* editing/selection/unrendered-002-expected.txt:
* editing/selection/unrendered-003-expected.checksum:
* editing/selection/unrendered-003-expected.png:
* editing/selection/unrendered-003-expected.txt:
* editing/selection/unrendered-004-expected.checksum:
* editing/selection/unrendered-004-expected.png:
* editing/selection/unrendered-004-expected.txt:
* editing/selection/unrendered-005-expected.checksum:
* editing/selection/unrendered-005-expected.png:
* editing/selection/unrendered-005-expected.txt:
* fast/block/float/019-expected.checksum:
* fast/block/float/019-expected.png:
* fast/block/float/019-expected.txt:
* fast/block/margin-collapse/104-expected.checksum:
* fast/block/margin-collapse/104-expected.png:
* fast/block/margin-collapse/104-expected.txt:
* fast/block/positioning/relative-overflow-block-expected.checksum:
* fast/block/positioning/relative-overflow-block-expected.txt:
* fast/block/positioning/relative-overflow-replaced-expected.checksum:
* fast/block/positioning/relative-overflow-replaced-expected.txt:
* fast/block/positioning/relative-overflow-replaced-float-expected.checksum:
* fast/block/positioning/relative-overflow-replaced-float-expected.txt:
* fast/lists/li-br-expected.checksum:
* fast/lists/li-br-expected.png:
* fast/lists/li-br-expected.txt:
* fast/media/mq-relative-constraints-02-expected.checksum:
* fast/media/mq-relative-constraints-02-expected.png:
* fast/media/mq-relative-constraints-02-expected.txt:
* fast/media/mq-relative-constraints-03-expected.checksum:
* fast/media/mq-relative-constraints-03-expected.png:
* fast/media/mq-relative-constraints-03-expected.txt:
* fast/media/mq-relative-constraints-04-expected.checksum:
* fast/media/mq-relative-constraints-04-expected.png:
* fast/media/mq-relative-constraints-04-expected.txt:
* fast/media/mq-relative-constraints-05-expected.checksum:
* fast/media/mq-relative-constraints-05-expected.png:
* fast/media/mq-relative-constraints-05-expected.txt:
* fast/media/mq-relative-constraints-06-expected.checksum:
* fast/media/mq-relative-constraints-06-expected.png:
* fast/media/mq-relative-constraints-06-expected.txt:
* fast/media/mq-relative-constraints-07-expected.checksum:
* fast/media/mq-relative-constraints-07-expected.png:
* fast/media/mq-relative-constraints-07-expected.txt:
* fast/media/mq-relative-constraints-09-expected.checksum:
* fast/media/mq-relative-constraints-09-expected.png:
* fast/media/mq-relative-constraints-09-expected.txt:
* fast/media/mq-width-absolute-01-expected.checksum:
* fast/media/mq-width-absolute-01-expected.png:
* fast/media/mq-width-absolute-01-expected.txt:
* fast/media/mq-width-absolute-02-expected.checksum:
* fast/media/mq-width-absolute-02-expected.png:
* fast/media/mq-width-absolute-02-expected.txt:
* fast/media/mq-width-absolute-03-expected.checksum:
* fast/media/mq-width-absolute-03-expected.png:
* fast/media/mq-width-absolute-03-expected.txt:
* fast/media/mq-width-absolute-04-expected.checksum:
* fast/media/mq-width-absolute-04-expected.png:
* fast/media/mq-width-absolute-04-expected.txt:
* fast/table/wide-colspan-expected.checksum:
* fast/table/wide-colspan-expected.png:
* fast/table/wide-colspan-expected.txt:
* fast/table/wide-column-expected.checksum:
* fast/table/wide-column-expected.png:
* fast/table/wide-column-expected.txt:
* tables/mozilla/bugs/bug131020-expected.checksum:
* tables/mozilla/bugs/bug131020-expected.png:
* tables/mozilla/bugs/bug131020-expected.txt:
* tables/mozilla/bugs/bug23151-expected.checksum:
* tables/mozilla/bugs/bug23151-expected.png:
* tables/mozilla/bugs/bug23151-expected.txt:
* tables/mozilla/bugs/bug29314-expected.checksum:
* tables/mozilla/bugs/bug29314-expected.png:
* tables/mozilla/bugs/bug29314-expected.txt:
* tables/mozilla/bugs/bug43039-expected.checksum:
* tables/mozilla/bugs/bug43039-expected.png:
* tables/mozilla/bugs/bug43039-expected.txt:
* tables/mozilla/bugs/bug43854-1-expected.checksum:
* tables/mozilla/bugs/bug43854-1-expected.png:
* tables/mozilla/bugs/bug43854-1-expected.txt:
* tables/mozilla/bugs/bug625-expected.checksum:
* tables/mozilla/bugs/bug625-expected.png:
* tables/mozilla/bugs/bug625-expected.txt:
* tables/mozilla/bugs/bug92143-expected.checksum:
* tables/mozilla/bugs/bug92143-expected.png:
* tables/mozilla/bugs/bug92143-expected.txt:
* tables/mozilla/bugs/bug96334-expected.checksum:
* tables/mozilla/bugs/bug96334-expected.png:
* tables/mozilla/bugs/bug96334-expected.txt:
* tables/mozilla/core/misc-expected.checksum:
* tables/mozilla/core/misc-expected.png:
* tables/mozilla/core/misc-expected.txt:
* tables/mozilla/core/nested1-expected.checksum:
* tables/mozilla/core/nested1-expected.png:
* tables/mozilla/core/nested1-expected.txt:
* tables/mozilla_expected_failures/bugs/bug220653-expected.checksum:
* tables/mozilla_expected_failures/bugs/bug220653-expected.png:
* tables/mozilla_expected_failures/bugs/bug220653-expected.txt:
* tables/mozilla_expected_failures/bugs/bug67915-2-expected.checksum:
* tables/mozilla_expected_failures/bugs/bug67915-2-expected.png:
* tables/mozilla_expected_failures/bugs/bug67915-2-expected.txt:
WebCore:
Reviewed by Hyatt. Patch by Mitz.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9531
Pages that need only a horizontal scroll bar get a vertical one as well
The change affects many existing tests.
* page/FrameView.cpp:
(WebCore::FrameView::layout): Mark the body renderer as needing layout if
the view's height has changed and the body stretches to the view's height.
* rendering/RenderBox.cpp:
(WebCore::RenderBox::calcHeight):
* rendering/RenderBox.h:
(WebCore::RenderBox::stretchesToViewHeight): Added. Moved the logic for when
the HTML block needs to stretch to the view's height from calcHeight() here.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15011
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ap [Sat, 24 Jun 2006 15:00:16 +0000 (15:00 +0000)]
Reviewed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=7461
Always encode the path part of an URI as UTF-8
Test: http/tests/uri/utf8-path.html
* platform/KURL.cpp:
(encodeRelativeString): Always set pathEncoding to UTF-8.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15010
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ap [Sat, 24 Jun 2006 14:43:23 +0000 (14:43 +0000)]
2006-06-24 Graham Dennis <Graham.Dennis@gmail.com>
Reviewed by Darin, landed by ap.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9467
TreeWalker/NodeIterator do not restrict the traversal
Test: traversal/stay-within-root.html
* dom/Node.cpp:
(WebCore::Node::traversePreviousNode):
* dom/Node.h:
Add a stayWithin argument to traversePreviousNode just like the other
traversal functions
* dom/NodeIterator.cpp:
(WebCore::NodeIterator::findNextNode):
(WebCore::NodeIterator::findPreviousNode):
Ensure that when moving forward or backward that the root node is not
exited
* dom/TreeWalker.cpp:
(WebCore::TreeWalker::previousNode):
(WebCore::TreeWalker::nextNode):
Ditto
(WebCore::CompositeEditCommand::moveParagraphContentsToNewBlockIfNecessary):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15009
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ap [Sat, 24 Jun 2006 13:50:29 +0000 (13:50 +0000)]
svn-commit.tmp
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15008
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ap [Sat, 24 Jun 2006 13:01:40 +0000 (13:01 +0000)]
2006-06-24 Mitz Pettel <opendarwin.org@mitzpettel.com>
Reviewed and landed by ap.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=9563
Remove unused layoutSuppressed field and inLayout() method
No test added because there is no change in functionality.
* page/FrameView.cpp:
(WebCore::FrameViewPrivate::reset):
(WebCore::FrameView::layout):
* page/FrameView.h:
* rendering/RenderView.cpp:
(WebCore::RenderView::layout):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15007
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ap [Sat, 24 Jun 2006 12:53:57 +0000 (12:53 +0000)]
Reviewed by Darin.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=8177
Javascript search incredibly slow
No change in functionality, thus no test.
Use info sharing for named collections, too. Moved shared HTMLCollection infos from HTMLDocument
to Document. Renamed collection types to use the preferred WebKit style. Replaced direct
HTMLCollection creation with using accessors from Document in a few places.
* bindings/objc/DOMHTML.mm:
(-[DOMHTMLElement children]):
(-[DOMHTMLDocument images]):
(-[DOMHTMLDocument applets]):
(-[DOMHTMLDocument links]):
(-[DOMHTMLDocument forms]):
(-[DOMHTMLDocument anchors]):
(-[DOMHTMLMapElement areas]):
(-[DOMHTMLTableSectionElement rows]):
(-[DOMHTMLTableElement rows]):
(-[DOMHTMLTableElement tBodies]):
(-[DOMHTMLTableRowElement cells]):
* dom/Document.cpp:
(WebCore::Document::images):
(WebCore::Document::applets):
(WebCore::Document::embeds):
(WebCore::Document::objects):
(WebCore::Document::scripts):
(WebCore::Document::links):
(WebCore::Document::forms):
(WebCore::Document::anchors):
(WebCore::Document::all):
(WebCore::Document::windowNamedItems):
(WebCore::Document::documentNamedItems):
(WebCore::Document::nameCollectionInfo):
* dom/Document.h:
(WebCore::Document::collectionInfo):
* html/HTMLCollection.cpp:
(WebCore::HTMLCollection::HTMLCollection):
(WebCore::HTMLCollection::traverseNextItem):
(WebCore::HTMLCollection::checkForNameMatch):
(WebCore::HTMLCollection::updateNameCache):
* html/HTMLCollection.h:
(WebCore::HTMLCollection::):
* html/HTMLDocument.h:
* html/HTMLElement.cpp:
(WebCore::HTMLElement::children):
* html/HTMLMapElement.cpp:
(WebCore::HTMLMapElement::areas):
* html/HTMLNameCollection.h:
* html/HTMLNameCollection.cpp:
(WebCore::HTMLNameCollection::HTMLNameCollection):
(WebCore::HTMLNameCollection::traverseNextItem):
* html/HTMLFormCollection.cpp:
* html/HTMLOptionsCollection.cpp:
(WebCore::HTMLOptionsCollection::HTMLOptionsCollection):
* html/HTMLSelectElement.h:
(WebCore::HTMLSelectElement::collectionInfo):
* html/HTMLTableElement.cpp:
(WebCore::HTMLTableElement::rows):
(WebCore::HTMLTableElement::tBodies):
* html/HTMLTableRowElement.cpp:
(WebCore::HTMLTableRowElement::cells):
* html/HTMLTableSectionElement.cpp:
(WebCore::HTMLTableSectionElement::rows):
* kwq/WebCoreAXObject.mm:
(-[WebCoreAXObject accessibilityAttributeValue:]):
* page/Frame.cpp:
(WebCore::Frame::gotoAnchor):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15006
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ap [Sat, 24 Jun 2006 11:50:43 +0000 (11:50 +0000)]
Reviewed by Darin.
- http://bugzilla.opendarwin.org/show_bug.cgi?id=9418
WebKit will not build when Space exists in path
Enclose search paths in quotes; create symlinks to avoid passing paths with spaces to make.
* JavaScriptCore/JavaScriptCore.xcodeproj/project.pbxproj:
* WebCore/WebCore.xcodeproj/project.pbxproj:
* WebKit/WebKit.xcodeproj/project.pbxproj:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15005
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
tomernic [Sat, 24 Jun 2006 05:19:14 +0000 (05:19 +0000)]
LayoutTests:
Reviewed by Tim Hatcher.
<rdar://problem/
4560947> Leopard REGRESSION: Dragging to TileGame Widget is broken
* fast/events/ondragenter-expected.checksum: Added.
* fast/events/ondragenter-expected.png: Added.
* fast/events/ondragenter-expected.txt: Added.
* fast/events/ondragenter.html: Added.
WebCore:
Reviewed by Tim Hatcher.
<rdar://problem/
4560947> Leopard REGRESSION: Dragging to TileGame Widget is broken
* page/FrameView.cpp:
(WebCore::FrameView::updateDragAndDrop):
Use the passed mouse event, instead of the "current" event. There is no such thing as
a "drag event". Asking for the current event during a drag returns a weird system-
internal event with an invalid mouse location.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15004
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Sat, 24 Jun 2006 05:07:08 +0000 (05:07 +0000)]
Reviewed by Eric.
Manual test added: cursor.html
Bug 9560: should support new CSS3 cursor types to get more resizing directions
http://bugzilla.opendarwin.org/show_bug.cgi?id=9560
Adds support for ew-resize, ns-resize, nesw-resize, nwse-resize,
col-resize and row-resize CSS3 cursor values.
Also converted some of the cursor images to grayscale tiffs,
some were RGB but had no color data.
* Resources/eastResizeCursor.tiff:
* Resources/eastWestResizeCursor.tiff: Added.
* Resources/helpCursor.tiff:
* Resources/moveCursor.tiff:
* Resources/northEastResizeCursor.tiff:
* Resources/northEastSouthWestResizeCursor.tiff: Added.
* Resources/northResizeCursor.tiff:
* Resources/northSouthResizeCursor.tiff: Added.
* Resources/northWestResizeCursor.tiff:
* Resources/northWestSouthEastResizeCursor.tiff: Added.
* Resources/southEastResizeCursor.tiff:
* Resources/southResizeCursor.tiff:
* Resources/southWestResizeCursor.tiff:
* Resources/textAreaResizeCorner.tiff:
* Resources/waitCursor.tiff:
* Resources/westResizeCursor.tiff:
* WebCore.xcodeproj/project.pbxproj:
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue):
* css/CSSValueKeywords.in:
* css/cssparser.cpp:
(WebCore::CSSParser::parseValue):
* manual-tests/cursor.html: Added.
* page/FrameView.cpp:
(WebCore::selectCursor):
* platform/Cursor.h:
* platform/mac/CursorMac.mm:
(WebCore::northSouthResizeCursor):
(WebCore::eastWestResizeCursor):
(WebCore::northEastSouthWestResizeCursor):
(WebCore::northWestSouthEastResizeCursor):
(WebCore::columnResizeCursor):
(WebCore::rowResizeCursor):
* platform/win/CursorWin.cpp:
(WebCore::northSouthResizeCursor):
(WebCore::eastWestResizeCursor):
(WebCore::northEastSouthWestResizeCursor):
(WebCore::northWestSouthEastResizeCursor):
(WebCore::columnResizeCursor):
(WebCore::rowResizeCursor):
* rendering/render_style.h: cursor needs 5 bits now
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15003
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Sat, 24 Jun 2006 01:09:03 +0000 (01:09 +0000)]
Reviewed by Darin.
Adding more operator[] overloads for long and short types.
* wtf/Vector.h:
(WTF::Vector::operator[]):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15002
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
kdecker [Sat, 24 Jun 2006 00:58:34 +0000 (00:58 +0000)]
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15001
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
tomernic [Fri, 23 Jun 2006 23:56:57 +0000 (23:56 +0000)]
Reviewed by John Sullivan.
* Plugins/WebBaseNetscapePluginView.m:
(-[WebBaseNetscapePluginView sendEvent:]):
Fixed a bug I found in the CoreGraphics drawing model that was preventing certain types of
events from being dispatched to the plugin, unless the plugin was being updated. The check
for portState was only required to call -setWindowIfNecessary, not required for the entire
event dispatch.
Also, don't paint the green debug rect unless this is a QuickDraw plugin. Otherwise the
current QD port is not set, and the green rect fills the entire screen. Pretty awesome
looking, but not intended behavior.
(-[WebBaseNetscapePluginView setWindowIfNecessary]):
Improved the logging here to include the NPWindow's width and height.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@15000
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Fri, 23 Jun 2006 23:34:17 +0000 (23:34 +0000)]
Reviewed by Tim O.
Adds back SPI that Mail is still depending on.
* WebView/WebDataSource.m:
(-[WebDataSource _addSubframeArchives:]):
* WebView/WebDataSourcePrivate.h:
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@14999
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
sullivan [Fri, 23 Jun 2006 23:12:00 +0000 (23:12 +0000)]
Reviewed by Tim Omernick
- fixed <rdar://problem/
4597685> stale rects for document markers can be left behind in certain cases
* dom/Document.h:
* dom/Document.cpp:
(WebCore::Document::setRenderedRectForMarker):
changed parameter from IntRect to const IntRect& at Darin's request
(WebCore::Document::invalidateRenderedRectsForMarkersInRect):
new method, tests whether the rendered rect for each marker intersects the
passed rect. If so, sets the rendered rect back to the placeholder rect that
means that the rect for this marker isn't known.
* page/Frame.cpp:
(WebCore::Frame::paint):
call invalidateRenderedRectsForMarkersInRect on the rect we're about to paint.
This ensures that old stale rects won't be preserved if the associated marker
has moved such that it is out of the painted area.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@14998
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
justing [Fri, 23 Jun 2006 22:38:42 +0000 (22:38 +0000)]
Added bug number.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@14997
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
justing [Fri, 23 Jun 2006 22:36:12 +0000 (22:36 +0000)]
LayoutTests:
Reviewed by levi
* editing/deleting/merge-into-empty-block-1-expected.checksum: Added.
* editing/deleting/merge-into-empty-block-1-expected.png: Added.
* editing/deleting/merge-into-empty-block-1-expected.txt: Added.
* editing/deleting/merge-into-empty-block-1.html: Added.
* editing/deleting/merge-into-empty-block-2-expected.checksum: Added.
* editing/deleting/merge-into-empty-block-2-expected.png: Added.
* editing/deleting/merge-into-empty-block-2-expected.txt: Added.
* editing/deleting/merge-into-empty-block-2.html: Added.
WebCore:
Reviewed by levi
* editing/DeleteSelectionCommand.cpp:
(WebCore::DeleteSelectionCommand::mergeParagraphs):
Add a rule for when to merge into empty blocks.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@14996
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
ggaren [Fri, 23 Jun 2006 22:08:03 +0000 (22:08 +0000)]
Reviewed by Adele.
- Added JSConstructorMake to match JSFunctionMake, along with test code.
* API/JSObjectRef.cpp:
(JSConstructorMake):
* API/JSObjectRef.h:
* API/testapi.c:
(myConstructor_callAsConstructor):
(main):
* API/testapi.js:
* ChangeLog:
* JavaScriptCore.xcodeproj/project.pbxproj: Moved testapi.c to the testapi
target -- this was an oversight in my earlier check-in.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@14995
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Fri, 23 Jun 2006 21:36:14 +0000 (21:36 +0000)]
Versoning
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@14993
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
thatcher [Fri, 23 Jun 2006 21:27:59 +0000 (21:27 +0000)]
Reviewed by Tim Hatcher.
<rdar://problem/
4223619> Safari crashed in -[NSFont __isSystemFont]
CFRetain and CFRelease NSFont's when FontPlatformData has a reference.
We only do this when GC is enabled for performance reasons.
No tests possible.
* platform/mac/FontPlatformData.h:
(WebCore::FontPlatformData::FontPlatformData):
(WebCore::FontPlatformData::~FontPlatformData):
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@14992
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
justing [Fri, 23 Jun 2006 21:19:59 +0000 (21:19 +0000)]
LayoutTests:
Reviewed by levi
<http://bugzilla.opendarwin.org/show_bug.cgi?id=8074>
Caret before/after block tables is too small
<rdar://problem/
4598331>
right arrow out of a table shows cursor in a weird place
Demonstrates 9557:
* editing/input/range-for-empty-document-expected.txt:
Added:
* editing/selection/table-caret-1-expected.checksum: Added.
* editing/selection/table-caret-1-expected.png: Added.
* editing/selection/table-caret-1-expected.txt: Added.
* editing/selection/table-caret-1.html: Added.
* editing/selection/table-caret-2-expected.checksum: Added.
* editing/selection/table-caret-2-expected.png: Added.
* editing/selection/table-caret-2-expected.txt: Added.
* editing/selection/table-caret-2.html: Added.
* editing/selection/table-caret-3-expected.checksum: Added.
* editing/selection/table-caret-3-expected.png: Added.
* editing/selection/table-caret-3-expected.txt: Added.
* editing/selection/table-caret-3.html: Added.
WebCore:
Reviewed by levi
<http://bugzilla.opendarwin.org/show_bug.cgi?id=8074>
Caret before/after block tables is too small
<rdar://problem/
4598331>
right arrow out of a table shows cursor in a weird place
* rendering/RenderBox.cpp:
(WebCore::RenderBox::caretRect): Don't propagate to children,
VisiblePositions inside containers don't refer to children.
Don't use the font height for tables.
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@14991
268f45cc-cd09-0410-ab3c-
d52691b4dbfc
git://git.webkit.org / WebKit-https.git / log