Timeline
Apr 16, 2006:
- 11:50 PM Changeset in webkit [13893] by
-
- 8 edits4 adds in trunk
LayoutTests:
Reviewed by Darin.
- test for http://bugzilla.opendarwin.org/show_bug.cgi?id=8408 Paint the highlight behind selected list markers
- editing/selection/extend-by-word-002-expected.checksum: Updated.
- editing/selection/extend-by-word-002-expected.png: Updated.
- fast/lists/markers-in-selection-expected.checksum: Added.
- fast/lists/markers-in-selection-expected.png: Added.
- fast/lists/markers-in-selection-expected.txt: Added.
- fast/lists/markers-in-selection.html: Added.
WebCore:
Reviewed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=8408 Paint the highlight behind selected list markers
Test: fast/lists/markers-in-selection.html
- rendering/RenderObject.h: Added selectionColorImageOverlayAlpha constant - the maximum opacity of the selection color when painted over images.
- rendering/render_list.cpp: (WebCore::RenderListMarker::RenderListMarker): (WebCore::RenderListMarker::paint): Paint the selection highlight if selected: over the marker for image markers, under the marker for all other markers. (WebCore::RenderListMarker::setSelectionState): Added. (WebCore::RenderListMarker::selectionRect): Added. (WebCore::RenderListMarker::selectionColor): Added. Ensures that the selection color is transparent for image markers.
- rendering/render_list.h: (WebCore::RenderListMarker::selectionState): (WebCore::RenderListMarker::canBeSelectionLeaf):
- rendering/render_replaced.cpp (WebCore::RenderReplaced::selectionColor): Changed to use the selectionColorImageOverlayAlpha constant.
- 11:15 PM Changeset in webkit [13892] by
-
- 5 edits4 adds in trunk
LayoutTests:
Reviewed by Darin.
- test for http://bugzilla.opendarwin.org/show_bug.cgi?id=8420 iExploder(#12): Assertion failure in RenderContainer::removeChildNode
- fast/forms/button-inner-block-reuse-expected.checksum: Added.
- fast/forms/button-inner-block-reuse-expected.png: Added.
- fast/forms/button-inner-block-reuse-expected.txt: Added.
- fast/forms/button-inner-block-reuse.html: Added.
WebCore:
Reviewed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=8420 iExploder(#12): Assertion failure in RenderContainer::removeChildNode
Test: fast/forms/button-inner-block-reuse.html
Buttons have a distinguished anonymous child that holds all their other
descendants. Descendants ended up in a sibling anonymous block as the
initial anonymous child was being reused to hold the initial part of an
inline that got split.
- rendering/RenderInline.cpp: (WebCore::RenderInline::splitFlow): Check if the anonymous block's parent allows us to reuse it.
- rendering/RenderObject.h: (WebCore::RenderObject::allowsReusingAnonymousChild): Added. Returns true.
- rendering/render_button.h: (WebCore::RenderButton::allowsReusingAnonymousChild): Added. Returns false.
- 11:01 PM Changeset in webkit [13891] by
-
- 3 edits4 adds in trunk
LayoutTests:
Reviewed by Justin.
- test for http://bugzilla.opendarwin.org/show_bug.cgi?id=8394 Editable region does not accept dropped text if there is no selection
- editing/pasteboard/drop-text-without-selection-expected.checksum: Added.
- editing/pasteboard/drop-text-without-selection-expected.png: Added.
- editing/pasteboard/drop-text-without-selection-expected.txt: Added.
- editing/pasteboard/drop-text-without-selection.html: Added.
WebCore:
Reviewed by Justin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=8394 Editable region does not accept dropped text if there is no selection
Test: editing/pasteboard/drop-text-without-selection.html
- bridge/mac/WebCoreFrameBridge.mm: (-[WebCoreFrameBridge documentFragmentWithText:]): Changed to allow creating a fragment regardless of the selection.
- 10:46 PM Changeset in webkit [13890] by
-
- 11 edits1 add in trunk
WebCore:
Reviewed by Darin.
- WebCore part of fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=8324 REGRESSION: textarea :focus not applied immediately
- bridge/mac/WebCoreFrameBridge.h:
- kwq/KWQComboBox.mm: (-[KWQPopUpButton becomeFirstResponder]): Added call to formControlIsBecomingFirstResponder:. (-[KWQPopUpButton resignFirstResponder]): Cleaned up.
- kwq/KWQListBox.mm: Ditto. (-[KWQTableView becomeFirstResponder]): (-[KWQTableView resignFirstResponder]):
- kwq/KWQSlider.mm: (-[KWQSlider becomeFirstResponder]): Added call to formControlIsBecomingFirstResponder: and cleaned up. (-[KWQSlider resignFirstResponder]): Cleaned up.
- platform/mac/WebCoreTextArea.mm: (-[WebCoreTextView becomeFirstResponder]): Added call to formControlIsBecomingFirstResponder:. (-[WebCoreTextView resignFirstResponder]): Cleaned up.
- platform/mac/WebCoreTextField.mm: (-[KWQTextFieldController setHasFocus:]): Added call to formControlIsBecomingFirstResponder:.
- manual-tests/textarea-focus.html: Added.
WebKit:
Reviewed by Darin.
- WebKit part of fix for http://bugzilla.opendarwin.org/show_bug.cgi?id=8324 REGRESSION: textarea :focus not applied immediately
- WebCoreSupport/WebFrameBridge.m: (-[WebFrameBridge formControlIsBecomingFirstResponder:]): Added. Calls the ancestor WebHTMLView's _formControlIsBecomingFirstResponder:
- WebView/WebHTMLView.m: (-[WebHTMLView _updateFocusState:]): If a descendant is becoming first responder, enable focused appearance. (-[WebHTMLView _formControlIsBecomingFirstResponder:]): Added. Calls _updateFocusState, causing the frame to display with focus attributes.
- WebView/WebHTMLViewInternal.h:
- 10:34 PM Changeset in webkit [13889] by
-
- 7 edits1 add in trunk/WebCore
Reviewed by Adele and Justin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=8298 REGRESSION: Crash occurs when attempting to drag selection into Depart/Return input fields at http://www.travelocity.com/
- remove the mutation event listener that's installed all the time, since it slows things down a bit
Calling SelectionController::nodeWillBeRemoved from Document::notifyBeforeNodeRemoval
fixes the crash, which was happening because the call that was removing the text
node, removeChildren, does not send a "node removed" mutation event (it sends a
"subtree modified" mutation event instead). So this change alone fixes the crash.
But I also changed setInnerText to not blow away the text node each time the value
is changed, and that makes the test case behave even better -- you don't even lose
the selection; it works as it did with the NSTextField-based text field.
- manual-tests/input-empty-on-focus.html: Added.
- page/Frame.h: Tweaked a few comments and functions related to selection.
- page/Frame.cpp: (WebCore::Frame::dragCaret): Made non-const.
- dom/Document.cpp: (WebCore::Document::notifyBeforeNodeRemoval): Call nodeWillBeRemoved on the two selection controllers before removing a node from the document.
- editing/SelectionController.h: Tweak formatting. Remove MutationListener class and m_mutationListener field.
- editing/SelectionController.cpp: (WebCore::SelectionController::SelectionController): Remove code to set up the mutation event listener. (WebCore::SelectionController::setSelection): Remove code to maintain the mutation event listener.
- html/HTMLElement.cpp: (WebCore::HTMLElement::setInnerHTML): In cases where the container has only a single child use replaceChild, and in cases where the HTML being inserted also has only a single child and both are text nodes use setData. It's common to use setInnerHTML to set something that's just text. (WebCore::HTMLElement::setInnerText): Same as above, but simpler since the thing we're replacing with is always text.
- 8:10 PM Changeset in webkit [13888] by
-
- 3 edits in trunk/WebCore
Reviewed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=8417 make-css-file-arrays.pl hangs when run on Linux
- css/make-css-file-arrays.pl: Remove the "-" parameter from the invocation of cpp, which means "send output to stdout". It's optional on Mac OS X, and is causing a hang on Linux.
- rendering/RenderArena.cpp: Added a missing include of <assert.h>.
- 2:36 PM Changeset in webkit [13887] by
-
- 3 edits in trunk/WebKitSite
- Small textual change to my contact info;
- Added bugreporting URL for non ADC members.
- 2:22 PM Changeset in webkit [13886] by
-
- 2 edits in trunk/WebKitSite
Added info on CIA, trac, build bot and webkit-unassigned
- 2:29 AM Changeset in webkit [13885] by
-
- 2 edits in trunk/WebKitTools
Reviewed by Darin.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=8412 Restore color profile after a crash
- DumpRenderTree/DumpRenderTree.m: (crashHandler): Added a handler for signals that are raised on a crash. Keeping separate from the existing handler for manual interruption (SIGINT/SIGHUP/SIGTERM), because I'm trying to find a way to prevent the Crash Reporter dialog from appearing on crash, and this code may change significantly.
(main): Install crashHandler.
(setDefaultColorProfileToRGB): Properly convert CFStringRef to a char buffer
(CFStringGetCStringPtr should NEVER EVER be used!).
Apr 15, 2006:
- 8:05 PM Changeset in webkit [13884] by
-
- 2 edits in trunk/WebKitTools
Reviewed by Eric.
- Scripts/install-win-extras: Try to fix the Windows build by fixing the URLs here. Seems xmlsoft.org's HTTP no longer has what we need.
- 7:28 PM Changeset in webkit [13883] by
-
- 2 edits in trunk/WebCore
- removed references to a couple files that are obsolete but were still in the Windows project file
- WebCore.vcproj/WebCore/WebCore.vcproj: Remove kjs_views.h and .cpp.
- 7:20 PM Changeset in webkit [13882] by
-
- 8 edits1 move in branches/treecode/JavaScriptCore
Reviewed by Darin.
- rename InterpreterState to Opcode
- JavaScriptCore.xcodeproj/project.pbxproj:
- kjs/InterpreterState.h: Removed.
- kjs/Opcode.h: Added. (KJS::):
- kjs/TreeCode.cpp: (KJS::): (KJS::isEvaluateOpcode): (KJS::isEvaluateListOpcode): (KJS::isExecuteState): (KJS::pushEvaluate): (KJS::pushEvaluateList): (KJS::pushExecute): (KJS::InterpreterImp::runInterpreterLoop):
- kjs/grammar.y:
- kjs/internal.cpp: (KJS::printUnwindBarrier): (KJS::InterpreterImp::printStateStack): (KJS::InterpreterImp::unwindToNextBarrier):
- kjs/internal.h: (KJS::InterpreterImp::State::State):
- kjs/nodes.cpp: (Node::Node):
- kjs/nodes.h: (KJS::Node::opcode): (KJS::ExprNode::ExprNode): (KJS::StatementNode::StatementNode): (KJS::JumpNode::JumpNode): (KJS::ForExprEndNode::ForExprEndNode):
- 6:26 PM Changeset in webkit [13881] by
-
- 1 edit2 adds in trunk/LayoutTests
- more results for tests landed without pixel-test results
- fast/text/whitespace/pre-newline-box-test-expected.checksum: Added.
- fast/text/whitespace/pre-newline-box-test-expected.png: Added.
- 12:27 PM Changeset in webkit [13880] by
-
- 2 edits in trunk/WebCore
Reviewed by Maciej.
- fix http://bugzilla.opendarwin.org/show_bug.cgi?id=8405 REGRESSION: Web Inspector's Style pane is blank
- bindings/js/kjs_window.cpp: (KJS::Window::isSafeScript): Use isEmpty() instead of isNull() for checking the domain to determine if the document in a local file.
- 11:13 AM Changeset in webkit [13879] by
-
- 1 edit7 adds in trunk/LayoutTests
- added some results for some tests people added without expected results
- svg/custom/class-selector-expected.txt: Added.
- svg/custom/class-selector-expected.checksum: Added.
- svg/custom/class-selector-expected.png: Added.
- editing/selection/extend-by-sentence-001-expected.checksum: Added.
- editing/selection/extend-by-sentence-001-expected.png: Added.
- fast/dom/Element/null-offset-parent-expected.checksum: Added.
- fast/dom/Element/null-offset-parent-expected.png: Added.
- 11:10 AM Changeset in webkit [13878] by
-
- 2 edits in trunk/LayoutTests
- updated a layout test result for Hyatt's prefix change
- fast/inspector/style-expected.txt: Some background properties now have the "-webkit" prefix.