⚠ Archived content — this site is no longer maintained.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

Timeline



Dec 21, 2005:

11:39 PM Changeset in webkit [11725] by ap
  • 2 edits in trunk/WebKitTools

Reviewed by Darin Adler.

  • Scripts/cvs-create-patch: Handle newly-added directories using chdir.
10:55 PM Changeset in webkit [11724] by darin
  • 3 edits in trunk/LayoutTests
  • fast/text/attributed-substring-from-range-expected.txt: Updated.
  • fast/text/attributed-substring-from-range-lines-expected.txt: Updated.
8:11 PM Changeset in webkit [11723] by darin
  • 14 edits in trunk/WebCore

Reviewed by Justin.

  • khtml/editing/apply_style_command.cpp: (khtml::ApplyStyleCommand::applyBlockStyle): Call new updateLayout member function. (khtml::ApplyStyleCommand::applyInlineStyle): Ditto. (khtml::ApplyStyleCommand::pushDownTextDecorationStyleAtBoundaries): Ditto.
  • khtml/editing/composite_edit_command.cpp: (khtml::CompositeEditCommand::addBlockPlaceholderIfNeeded): Ditto. (khtml::CompositeEditCommand::findBlockPlaceholder): Ditto. (khtml::CompositeEditCommand::moveParagraphContentsToNewBlockIfNecessary): Ditto.
  • khtml/editing/delete_selection_command.cpp: (khtml::DeleteSelectionCommand::fixupWhitespace): Ditto. (khtml::DeleteSelectionCommand::moveNodesAfterNode): Ditto.
  • khtml/editing/edit_command.cpp: (khtml::EditCommand::apply): Ditto. (khtml::EditCommand::unapply): Ditto. (khtml::EditCommand::reapply): Ditto. (khtml::EditCommand::updateLayout): Added. Calls updateLayoutIgnorePendingStylesheets on the document.
  • khtml/editing/edit_command.h: Added updateLayout member function.
  • khtml/editing/insert_line_break_command.cpp: (khtml::InsertLineBreakCommand::doApply): Call new updateLayout member function.
  • khtml/editing/insert_paragraph_separator_command.cpp: (khtml::InsertParagraphSeparatorCommand::doApply): Ditto.
  • khtml/editing/jsediting.cpp: (DOM::JSEditor::execCommand): Call updateLayoutIgnorePendingStylesheets instead of updateLayout. (DOM::JSEditor::queryCommandEnabled): Ditto. (DOM::JSEditor::queryCommandIndeterm): Ditto. (DOM::JSEditor::queryCommandState): Ditto. (DOM::JSEditor::queryCommandValue): Ditto.
  • khtml/editing/markup.cpp: (khtml::createMarkup): Ditto.
  • khtml/editing/replace_selection_command.cpp: (khtml::ReplacementFragment::insertFragmentForTestRendering): Ditto. (khtml::ReplaceSelectionCommand::fixupNodeStyles): Call new updateLayout member function. (khtml::ReplacementFragment::computeStylesUsingTestRendering): Call updateLayoutIgnorePendingStylesheets instead of updateLayout. (khtml::ReplaceSelectionCommand::doApply): Call new updateLayout member function. (khtml::ReplaceSelectionCommand::removeLinePlaceholderIfNeeded): Ditto. (khtml::ReplaceSelectionCommand::completeHTMLReplacement): Ditto.
  • khtml/editing/visible_units.cpp: (khtml::previousLinePosition): Call updateLayoutIgnorePendingStylesheets instead of updateLayout. (khtml::nextLinePosition): Ditto.
  • khtml/html/html_elementimpl.cpp: (HTMLElementImpl::innerText): Ditto.
  • kwq/WebCoreBridge.mm: (-[WebCoreBridge setSelectedDOMRange:affinity:closeTyping:]): Ditto. (-[WebCoreBridge smartDeleteRangeForProposedRange:]): Ditto.
8:06 PM Changeset in webkit [11722] by darin
  • 4 edits in trunk/WebCore

Reviewed by Geoff.

  • khtml/ecma/kjs_events.cpp: (KJS::DOMMouseEvent::getValueProperty): Change everything to just call through instead of doing the work here. (KJS::DOMWheelEvent::getValueProperty): Ditto.
  • khtml/xml/dom2_eventsimpl.h: (DOM::MouseRelatedEventImpl::offsetX): (DOM::MouseRelatedEventImpl::offsetY):
  • khtml/xml/dom2_eventsimpl.cpp: (DOM::MouseRelatedEventImpl::MouseRelatedEventImpl): Initialize new m_pageX, m_pageY, m_layerX, m_layerY, m_offsetX, and m_offsetY. (DOM::MouseRelatedEventImpl::computePositions): Renamed from computeLayerPos since it handles page position andoffset position too now. (DOM::MouseRelatedEventImpl::pageX): Added. Not in header since it's virtual. (DOM::MouseRelatedEventImpl::pageY): Ditto. (DOM::MouseRelatedEventImpl::x): Added, with FIXME since it should change eventually. (DOM::MouseRelatedEventImpl::y): Ditto. (DOM::MouseEventImpl::initMouseEvent): Call computePositions instead of computeLayerPos. (DOM::MouseEventImpl::toElement): Added. (DOM::MouseEventImpl::fromElement): Added.
5:40 PM Changeset in webkit [11721] by thatcher
  • 6 edits in trunk

JavaScriptCore:

  • JavaScriptCore.xcodeproj/project.pbxproj: Set tab width to 8, indent width to 4 and uses tabs to false per file.

WebCore:

  • WebCore.xcodeproj/project.pbxproj: Set tab width to 8, indent width to 4 and uses tabs to false per file.

WebKit:

  • WebKit.xcodeproj/project.pbxproj: Set tab width to 8, indent width to 4 and uses tabs to false per file.
5:32 PM Changeset in webkit [12226] by thatcher
  • 2 edits in trunk/JavaScriptGlue

JavaScriptGlue:

  • JavaScriptGlue.xcodeproj/project.pbxproj: Set tab width to 8, indent width to 4 and uses tabs to false per file.

WebBrowser:

  • WebBrowser.xcodeproj/project.pbxproj: Set tab width to 8, indent width to 4 and uses tabs to false per file.

WebKitSystemInterface:

  • WebKitSystemInterface.xcodeproj/project.pbxproj: Set tab width to 8, indent width to 4 and uses tabs to false per file.
5:25 PM Changeset in webkit [11720] by sullivan
  • 2 edits in trunk/WebCore

Reviewed by Tim Omernick and Darin Adler.

  • fixed HiDPI problem with forms auto-fill menu width

No test cases added; this code is used for "chrome" only.

  • kwq/DOMHTML.mm: (-[DOMHTMLInputElement _rectOnScreen]): convert entire rect to window coordinates, not just origin
4:55 PM Changeset in webkit [11719] by ggaren
  • 4 edits in trunk/JavaScriptCore

Reviewed by Darin.

Removed evil hack for determining if a type is an integer, replaced
with template metaprogramming.

  • JavaScriptCore.xcodeproj/project.pbxproj: Set tab size to 2 for testkjs.cpp
  • kjs/testkjs.cpp: (main): Inserted asserts to test IsInteger. FIXME: Move these to KXMLCore unit tests directory when we create one.
  • kxmlcore/HashTraits.h: (KXMLCore::): Added IsInteger class for querying types.
2:51 PM Changeset in webkit [11718] by harrison
  • 13 edits in trunk/WebCore

Reviewed by Justin).

<rdar://problem/4039777> Pasting particular snippet of HTML containing list items and a link creates too many list items

  • Fixed paste crash by making calling RenderBox::deleteLineBoxWrapper() from RenderObject::remove(), so that the connection is broken before the InlineBox's parent gets deleted.
  • Fixed overzealous style changes when setting the style on a specific range of elements. These specific ranges are derived programmatically, e.g. as a part of pasting, so they must be respected exactly rather than modified by converting the endpoints to VisiblePositions.

Test cases coming soon.

  • khtml/editing/apply_style_command.cpp: (khtml::ApplyStyleCommand::ApplyStyleCommand): (khtml::ApplyStyleCommand::updateStartEnd): (khtml::ApplyStyleCommand::startPosition): (khtml::ApplyStyleCommand::endPosition): (khtml::ApplyStyleCommand::applyBlockStyle): (khtml::ApplyStyleCommand::applyRelativeFontStyleChange): (khtml::ApplyStyleCommand::applyInlineStyle): (khtml::ApplyStyleCommand::removeInlineStyle): (khtml::ApplyStyleCommand::splitTextAtStartIfNeeded): (khtml::ApplyStyleCommand::splitTextAtEndIfNeeded): (khtml::ApplyStyleCommand::splitTextElementAtStartIfNeeded): (khtml::ApplyStyleCommand::splitTextElementAtEndIfNeeded): (khtml::ApplyStyleCommand::mergeStartWithPreviousIfIdentical): (khtml::ApplyStyleCommand::mergeEndWithNextIfIdentical): (khtml::ApplyStyleCommand::joinChildTextNodes):
  • khtml/editing/apply_style_command.h:
  • khtml/editing/composite_edit_command.cpp: (khtml::CompositeEditCommand::applyStyle):
  • khtml/editing/composite_edit_command.h:
  • khtml/editing/insert_line_break_command.cpp: (khtml::InsertLineBreakCommand::doApply):
  • khtml/editing/replace_selection_command.cpp: (khtml::ReplaceSelectionCommand::fixupNodeStyles): (khtml::ReplaceSelectionCommand::completeHTMLReplacement):
  • khtml/rendering/render_box.cpp: (RenderBox::destroy): (RenderBox::deleteLineBoxWrapper):
  • khtml/rendering/render_box.h:
  • khtml/rendering/render_list.cpp: (RenderListMarker::~RenderListMarker): (RenderListMarker::setStyle):
  • khtml/rendering/render_object.cpp: (RenderObject::remove):
  • khtml/rendering/render_object.h:
  • khtml/rendering/render_replaced.cpp: (RenderWidget::destroy):
2:42 PM Changeset in webkit [11717]
  • 212 copies
    17 deletes in tags/WebCore-417~17

This commit was manufactured by cvs2svn to create tag
'WebCore-417~17'.

2:42 PM Changeset in webkit [11716] by thatcher
  • 2 edits in branches/Safari-2-0-branch/WebCore

Versioning

1:57 PM Changeset in webkit [11715] by adele
  • 2 edits in branches/Safari-2-0-branch/WebCore

Reviewed by Tim Hatcher.

Fix for <rdar://problem/4387433> Seed: Radio buttons behave incorrectly in Gmail settings

  • khtml/html/html_formimpl.cpp: (DOM::HTMLFormElementImpl::registerFormElement): If this form element is already checked in the default form, remove it from m_selectedRadioButtons and add it for the new form.
9:43 AM Changeset in webkit [11714] by thatcher
  • 1 add in branches/Safari-1-3-branch/WebCore/ChangeLog

Rolled over to ChangeLog-2005-12-19

8:51 AM Changeset in webkit [11713] by thatcher
  • 1 delete in branches/Safari-1-3-branch/WebCore/ChangeLog

file ChangeLog was added on branch Safari-1-3-branch on 2005-12-21 17:43:18 +0000

8:50 AM Changeset in webkit [11712]
  • 1 copy in branches/Safari-1-3-branch/WebCore/ChangeLog

This commit was manufactured by cvs2svn to create branch
'Safari-1-3-branch'.

8:50 AM Changeset in webkit [11711] by adele
  • 19 edits in trunk

LayoutTests:

Reviewed by Darin.

Updating results for:
​http://bugzilla.opendarwin.org/show_bug.cgi?id=5911
REGRESSION: Page scroll position jumps when clicking on word in editable div

These results all changed because we now select all for contenteditable elements when focus() is called on them.

  • editing/selection/focus_editable_html-expected.checksum:
  • editing/selection/focus_editable_html-expected.png:
  • editing/selection/focus_editable_html-expected.txt:
  • fast/dom/blur-contenteditable-expected.txt:
  • fast/dom/focus-contenteditable-expected.checksum:
  • fast/dom/focus-contenteditable-expected.png:
  • fast/dom/focus-contenteditable-expected.txt:
  • fast/overflow/overflow-focus-ring-expected.checksum:
  • fast/overflow/overflow-focus-ring-expected.png:
  • fast/overflow/overflow-focus-ring-expected.txt:

WebCore:

Reviewed by Darin.

Fixed ​http://bugzilla.opendarwin.org/show_bug.cgi?id=5911
REGRESSION: Page scroll position jumps when clicking on word in editable div

I moved the scrolling code out of setFocusNode, and consolidated some of the focus code to scroll when necessary.

  • khtml/html/html_formimpl.cpp: Removed HTMLButtonElementImpl::blur, HTMLButtonElementImpl::focus, HTMLInputElementImpl::blur, HTMLInputElementImpl::focus. New code in ElementImpl::focus will now handle these cases. This allows tabbing through the elements to go through the same code path as calling focus() on an element. Before, focus() would scroll to reveal for any form elements that had a RenderWidget, but wouldn't scroll for anchor elements, or any of the new form elements. Now the behavior will be more consistent. (DOM::HTMLLabelElementImpl::focus): calls ElementImpl::focus. (DOM::HTMLLegendElementImpl::focus): ditto.
  • khtml/html/html_formimpl.h:
  • khtml/khtml_part.cpp: (KHTMLPart::selectAll): calls new function selectContentsOfNode (KHTMLPart::selectContentsOfNode): factored out code to selectAll for a particular node- which is useful for contenteditable elements.
  • khtml/khtml_part.h: added selectContentsOfNode
  • khtml/xml/dom_docimpl.cpp: (DocumentImpl::setFocusNode): removed scrolling code.
  • khtml/xml/dom_elementimpl.cpp: (ElementImpl::focus): Calls updateLayout in case focus() is called before there's a renderer. Makes a selection for editable elements (right now we select all, but this will change). Doesn't scroll if the renderer is a RenderWidget, since that is handled when the view becomes first responder. This will go away when we convert the rest of our form elements.
  • kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::revealSelection): Made this more like centerSelectionInVisibleArea where we get the right rectangle if the selection is a caret. (KWQKHTMLPart::nextKeyViewInFrame): call ElementImpl::focus() for the node. This will set the selection too, which used to be done here.
2:53 AM Changeset in webkit [11710] by eseidel
  • 1 edit in trunk/LayoutTests/ChangeLog

Bug #: none
Submitted by: eseidel
Reviewed by: none needed.

Fixed ChangeLog typo.

2:44 AM Changeset in webkit [11709] by eseidel
  • 3 edits
    2 adds in trunk/LayoutTests

Bug #: none
Submitted by: eseidel
Reviewed by: mjs

Beef up our XSLT processor tests to include a test which passes
a parameter into the transform. I also took this opportunity
to clean up the output a bit, and in the process found another
bug and file: ​http://bugzilla.opendarwin.org/show_bug.cgi?id=6180

  • fast/xsl/resources/xslt-param.xml: Added.
  • fast/xsl/resources/xslt-param.xsl: Added.
  • fast/xsl/xslt-processer-expected.txt:
  • fast/xsl/xslt-processer.html:

Dec 20, 2005:

6:06 PM Changeset in webkit [11708] by justing
  • 1 edit
    4 adds in trunk/LayoutTests

Reviewed by justin

New layout tests for:
<​http://bugzilla.opendarwin.org/show_bug.cgi?id=4682>
-[WebHTMLView firstRectForCharacterRange:] is using _selectedRange instead of the given range if no marked text

  • editing/input/firstrectforcharacterrange-plain-expected.txt: Added.
  • editing/input/firstrectforcharacterrange-plain.html: Added.
  • editing/input/firstrectforcharacterrange-styled-expected.txt: Added.
  • editing/input/firstrectforcharacterrange-styled.html: Added.
6:05 PM Changeset in webkit [11707] by justing
  • 3 edits in trunk/WebCore

Reviewed by justin

<​http://bugzilla.opendarwin.org/show_bug.cgi?id=4682>
-[WebHTMLView firstRectForCharacterRange:] is using _selectedRange instead of the given range if no marked text

Added layout tests:

  • editing/input/firstrectforcharacterrange-styled
  • editing/input/firstrectforcharacterrange-plain
  • khtml/editing/visible_text.cpp: (khtml::TextIterator::rangeFromLocationAndLength): Return null if the range isn't found, instead of a startless/endless range. Set the end if the requested location+length is out of bounds.
  • kwq/WebCoreBridge.mm: (-[WebCoreBridge convertToDOMRange:]): Handle larged unsigned values before calling rangeWithLocationAndLength, which expects signed ints.
6:04 PM Changeset in webkit [11706] by justing
  • 2 edits in trunk/WebKit

Reviewed by justin

<​http://bugzilla.opendarwin.org/show_bug.cgi?id=4682>
-[WebHTMLView firstRectForCharacterRange:] is using _selectedRange instead of the given range if no marked text

  • WebView.subproj/WebHTMLView.m: (-[WebHTMLView firstRectForCharacterRange:]): Handle some large unsigned values the way NSTextView does. Actually use the range passed in instead of _selectedRange, use of _selectedRange was a workaround that is no longer necessary.
5:41 PM Changeset in webkit [11705]
  • 222 copies
    18 deletes in tags/WebCore-417~16

This commit was manufactured by cvs2svn to create tag
'WebCore-417~16'.

5:41 PM Changeset in webkit [11704] by thatcher
  • 2 edits in branches/Safari-2-0-branch/WebCore

Versioning to WebCore-417.16

5:36 PM Changeset in webkit [11703] by adele
  • 3 edits in branches/Safari-2-0-branch/WebCore

Reviewed by Darin.

Fix for <rdar://problem/4387630> REGRESSION: <select> element's onClick event doesn't fire @ bugweb.apple.com

  • kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::passSubframeEventToSubframe): only pass mouse down for khtmlviews.
5:31 PM Changeset in webkit [11702] by thatcher
  • 1 add in branches/Safari-2-0-branch/WebCore/ChangeLog

Rolled over to ChangeLog-2005-12-19

5:16 PM Changeset in webkit [11701] by sullivan
  • 2 edits in trunk/WebKit

Reviewed by Darin Adler.

This is a problem with a particular font that was installed by Microsoft Office X. Though the
font and/or lower levels of font-handling code in the system are buggy, this bad symptom will
occur for users of Safari and other WebKit clients who happen to have one of these bad fonts.
This adds a workaround to avoid the problem.

  • WebCoreSupport.subproj/WebTextRendererFactory.m: (-[WebTextRendererFactory fontWithFamilies:traits:size:]): When we're going to synthesize bold or italic, yet the font we looked up was apparently a match for the traits, try to look up a font that without the to-be-synthesized traits. This way, instead of applying synthetic bold over Arial Bold, we'll apply synthetic bold over Arial Regular, which is uglier than just using Arial Bold, but far less ugly than using Arial Bold with synthetic bold too.

2005-12-16 Justin Garcia <​justin.garcia@apple.com>

<rdar://problem/4103393> Frequent Safari crash on lexisnexus.com (khtml::Selection::xPosForVerticalArrowNavigation)
<rdar://problem/4330451> CrashTracer: [REGRESSION] 2235 crashes in Safari at com.apple.WebCore: khtml::Selection::xPosForVerticalArrowNavigation const 436

Reviewed by darin

WebCore will crash when a selection that starts or ends in a node
that has been removed from the document is modify()d. This can occur:
(1) in non-editable regions (4103393 and 4330451), (2) in editable
regions (4383146) as the result of arbitrary DOM operations, and (3) in
Mail (4099739) as the result of an editing operation that sets a
bad ending selection.

Crashes of type (1) can occur when the user uses the arrow keys
to interact with a web app, or when the user tries to use
command-shift-arrow to switch tabs (this is a depricated
combo that will work if no one else responds to it). The easiest
way to fix these crashes is to disallow editing'ish selection changes
like moveDown:, selectWord:, pageDown:, etc, when the selection
is in a non-editable region.

Crashes of type (2) will require a more complicated fix (but occur
much less often than type (1)). Crashes of type (3) must be
fixed by tracking down the editing operation that sets bad selections.

Added a layout-test:

  • editing/selection/selection-actions.html
  • WebView.subproj/WebHTMLView.m: (-[WebHTMLView _canAlterCurrentSelection]): (-[WebHTMLView _alterCurrentSelection:direction:granularity:]): (-[WebHTMLView _alterCurrentSelection:verticalDistance:]): (-[WebHTMLView _expandSelectionToGranularity:]):
  • WebView.subproj/WebHTMLViewPrivate.h:
4:48 PM Changeset in webkit [11700] by thatcher
  • 1 delete in branches/Safari-2-0-branch/WebCore/ChangeLog

file ChangeLog was added on branch Safari-2-0-branch on 2005-12-21 01:31:19 +0000

4:48 PM Changeset in webkit [11699]
  • 1 copy in branches/Safari-2-0-branch/WebCore/ChangeLog

This commit was manufactured by cvs2svn to create branch
'Safari-2-0-branch'.

4:48 PM Changeset in webkit [11698] by adele
  • 2 edits in trunk/WebCore

Reviewed by Darin.

Fix for <rdar://problem/4387630> REGRESSION: <select> element's onClick event doesn't fire @ bugweb.apple.com

  • kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::passSubframeEventToSubframe): only pass mouse down for khtmlviews.
4:23 PM Changeset in webkit [11697] by justing
  • 1 edit
    4 adds in trunk/LayoutTests

Reviewed by darin

Layout test for:
<rdar://problem/4103393> Frequent Safari crash on lexisnexus.com (khtml::Selection::xPosForVerticalArrowNavigation)

<rdar://problem/4330451> CrashTracer: [REGRESSION] 2235 crashes in Safari at com.apple.WebCore: khtml::Selection::xPosForVerticalArrowNavigation const 436

  • editing/selection/selection-actions-expected.checksum: Added.
  • editing/selection/selection-actions-expected.png: Added.
  • editing/selection/selection-actions-expected.txt: Added.
  • editing/selection/selection-actions.html: Added.
4:23 PM Changeset in webkit [11696] by justing
  • 3 edits in trunk/WebKit

2005-12-16 Justin Garcia <​justin.garcia@apple.com>

<rdar://problem/4103393> Frequent Safari crash on lexisnexus.com (khtml::Selection::xPosForVerticalArrowNavigation)
<rdar://problem/4330451> CrashTracer: [REGRESSION] 2235 crashes in Safari at com.apple.WebCore: khtml::Selection::xPosForVerticalArrowNavigation const 436

Reviewed by darin

WebCore will crash when a selection that starts or ends in a node
that has been removed from the document is modify()d. This can occur:
(1) in non-editable regions (4103393 and 4330451), (2) in editable
regions (4383146) as the result of arbitrary DOM operations, and (3) in
Mail (4099739) as the result of an editing operation that sets a
bad ending selection.

Crashes of type (1) can occur when the user uses the arrow keys
to interact with a web app, or when the user tries to use
command-shift-arrow to switch tabs (this is a depricated
combo that will work if no one else responds to it). The easiest
way to fix these crashes is to disallow editing'ish selection changes
like moveDown:, selectWord:, pageDown:, etc, when the selection
is in a non-editable region.

Crashes of type (2) will require a more complicated fix (but occur
much less often than type (1)). Crashes of type (3) must be
fixed by tracking down the editing operation that sets bad selections.

Added a layout-test:

  • editing/selection/selection-actions.html
  • WebView.subproj/WebHTMLView.m: (-[WebHTMLView _canAlterCurrentSelection]): (-[WebHTMLView _alterCurrentSelection:direction:granularity:]): (-[WebHTMLView _alterCurrentSelection:verticalDistance:]): (-[WebHTMLView _expandSelectionToGranularity:]):
  • WebView.subproj/WebHTMLViewPrivate.h:
4:00 PM Changeset in webkit [11695] by justing
  • 12 edits in trunk/WebCore

Reviewed by: eric, thatcher

<rdar://problem/4172984> KWQExceptions needs to use @try/@catch instead of relying on NSException.h internals

Needed to undef try/catch because of 4333439. Moved the
declarations of variables that are returned from within a @try
block outside the @try block (because of "might be clobbered by
a longjmp or vfork" warnings). Moved some return statements
inside the @try block to fix volatilization errors with gcc4.

3:05 PM Changeset in webkit [11694] by justing
  • 2 edits in trunk/WebCore

Reviewed by eric, thatcher

<rdar://problem/4172984> KWQExceptions needs to use @try/@catch instead of relying on NSException.h internals

Needed to undef try/catch because of 4333439. Moved the
declarations of variables that are returned from within a @try
block outside the @try block (because of "might be clobbered by
a longjmp or vfork" warnings). Moved some return statements
inside the @try block to fix volatilization errors with gcc4.

  • WebCore.xcodeproj/project.pbxproj:
  • WebCorePrefix.h:
  • kcanvas/device/quartz/KCanvasFilterQuartz.mm: (KCanvasFEBlendQuartz::getCIFilter): (KCanvasFEColorMatrixQuartz::getCIFilter): (KCanvasFECompositeQuartz::getCIFilter): (getPointLightVectors): (getLightVectors): (getNormalMap): (KCanvasFEDiffuseLightingQuartz::getCIFilter): (KCanvasFEFloodQuartz::getCIFilter): (KCanvasFEImageQuartz::getCIFilter): (KCanvasFEMergeQuartz::getCIFilter): (KCanvasFESpecularLightingQuartz::getCIFilter):
  • kwq/KWQExceptions.h:
  • kwq/KWQExceptions.mm: (KWQReportBlockedException):
  • kwq/KWQFileButton.mm: (KWQFileButton::sizeForCharacterWidth): (KWQFileButton::frameGeometry):
  • kwq/KWQKCursor.mm: (+[NSCursor _WebCore_cursorWithName:hotSpot:_WebCore_cursorWithName:hotSpot:]):
  • kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::createPart): (KWQKHTMLPart::nextKeyView): (KWQKHTMLPart::runJavaScriptPrompt): (KWQKHTMLPart::keyEvent): (KWQKHTMLPart::sendContextMenuEvent): (KWQKHTMLPart::fileWrapperForElement): (KWQKHTMLPart::attributedString): (KWQKHTMLPart::imageFromRect):
  • kwq/KWQKHTMLPartBrowserExtension.mm:
  • kwq/KWQLineEdit.mm: (QLineEdit::sizeForCharacterWidth):
  • kwq/KWQListBox.mm: (QListBox::sizeForNumberOfLines):
  • kwq/KWQLoader.mm: (KWQCheckCacheObjectStatus):
  • kwq/KWQWidget.mm: (QWidget::frameGeometry): (QWidget::mapFromGlobal):
2:31 PM Changeset in webkit [11693] by harrison
  • 2 edits in trunk/WebCore

<rdar://problem/4294417> Cannot un-italicize some text after triple clicking it

Reviewed by Justin.

  • khtml/editing/apply_style_command.cpp: (khtml::ApplyStyleCommand::addInlineStyleIfNeeded): Removed check for tab span because ApplyStyleCommand::removeCSSStyle() already makes the same check.
1:36 PM Changeset in webkit [11692] by justing
  • 2 edits in trunk/WebKit

Reviewed by mitz

Fixed more uninitialized variable warnings, and
removed an extra semicolon.

  • WebCoreSupport.subproj/WebTextRenderer.m: (overrideLayoutOperation): (createATSULayoutParameters):
1:18 PM Changeset in webkit [11691] by justing
  • 2 edits in trunk/WebCore

<rdar://problem/4387270> editing/deleting/delete-3800834-fix failing
<​http://bugzilla.opendarwin.org/show_bug.cgi?id=6160> REGRESSION: Crash when running editing/deleting/delete-3800834-fix.html
<​http://bugzilla.opendarwin.org/show_bug.cgi?id=6161> REGRESSION: crash when pressing tab in editable WebHTMLView

Reviewed by darin

Some callers call setAttribute on a floating element. So, using
a RefPtr for the element inside addAttribute can destroy
it.

  • khtml/xml/dom_elementimpl.cpp: (NamedAttrMapImpl::addAttribute):
12:48 PM Changeset in webkit [11690] by mjs
  • 1 edit in trunk/JavaScriptCore/ChangeLog

Fix bug numbers in last two entries.

12:42 PM Changeset in webkit [11689] by mjs
  • 1 edit in trunk/WebKit/ChangeLog

Added bugzilla link.

12:41 PM Changeset in webkit [11688] by mjs
  • 8 edits in trunk/WebKit

Reviewed by Darin.

  • push more of frame lookup and management from WebView to WebFrame, this is in preparation for shifting this to WebCore
  • WebCoreSupport.subproj/WebBridge.m: (-[WebBridge closeWindowSoon]): Adjust for change to WebFrameNamespaces (-[WebBridge runModal]): ditto
  • WebView.subproj/WebControllerSets.h:
  • WebView.subproj/WebControllerSets.m: (+[WebFrameNamespaces addFrame:toNamespace:]): This now operates in terms of WebFrames (expected to be the main frame) not WebViews. (+[WebFrameNamespaces framesInNamespace:]): Ditto.
  • WebView.subproj/WebFrame.m: (-[WebFrame _setFrameNamespace:]): Set self, not WebView. (-[WebFrame _shouldAllowAccessFrom:]): Moved this code above use to avoid prototyping the method. (-[WebFrame _descendantFrameNamed:sourceFrame:]): Ditto. (-[WebFrame _frameInAnyWindowNamed:sourceFrame:]): Copied logic over from WebView. (-[WebFrame findFrameNamed:]): Do it all here, don't call WebView.
  • WebView.subproj/WebFramePrivate.h:
  • WebView.subproj/WebView.m:
  • WebView.subproj/WebViewPrivate.h:
12:40 PM Changeset in webkit [11687] by mjs
  • 2 edits in trunk/WebCore

Reviewed by Darin.

  • khtml/xml/dom_elementimpl.cpp: (StyledElementImpl::createAttributeMap): the new attribute map is going in a RefPtr, don't also ref it manually.
12:27 PM Changeset in webkit [11686] by mjs
  • 1 edit in trunk/WebCore/khtml/xml/dom_elementimpl.cpp

Remove accidentally committed change (which I am about to recommit, so no ChangeLog
comment for the revert).

12:18 PM Changeset in webkit [11685] by mjs
  • 2 edits in trunk/JavaScriptCore

Reviewed by Darin.

  • kxmlcore/AlwaysInline.h:
12:12 PM Changeset in webkit [11684] by mjs
  • 10 edits in trunk

JavaScriptCore:

Reviewed by Darin.

  • fixed a leak in the assignment operator from PassRefPtr to RefPtr
  • kxmlcore/RefPtr.h: (KXMLCore::RefPtr::operator=):
  • fix problem with PassRefPtr that darin spotted - it lacked a copy constructor and therefore was using the default one, which can lead to excess derefs

I fixed this by adding a copy constructor from non-const
reference, and by adding a template pass() function that you have
to use when raw pointer or RefPtr are passed where PassRefPtr is
expected.

  • kjs/identifier.cpp: (KJS::Identifier::add): Changed to have PassRefPtr return type and pass() the results.
  • kjs/identifier.h:
  • kjs/property_map.cpp: (KJS::PropertyMap::addSparseArrayPropertiesToReferenceList): Use pass() where required.
  • kjs/ustring.cpp: (KJS::UString::UString): Use pass() as needed. (KJS::UString::append): ditto (KJS::UString::substr): ditto
  • kjs/ustring.h: (KJS::UString::UString): Use initializer instead of assignment
  • kxmlcore/PassRefPtr.h: (KXMLCore::PassRefPtr::PassRefPtr): Added copy constructor (KXMLCore::pass): new template function to make it convenient to pass a PassRefPtr

WebCore:

Reviewed by Darin.

  • change an assignment to a contructor declaration to build with PassRefPtr leak fix changes
  • ksvg2/svg/SVGTransformableImpl.cpp: (SVGTransformableImpl::parseTransformAttribute):
10:47 AM Changeset in webkit [11683] by ggaren
  • 2 edits in trunk/WebKit

Reviewed by adele.

Fixed build failure due to missing 'b's in my last checkin.

  • WebCoreSupport.subproj/WebBridge.m: (-[WebBridge isStatusbarVisible]): changed 'B' to 'b' (-[WebBridge setStatusbarVisible:]): ditto
10:34 AM Changeset in webkit [11682] by ggaren
  • 5 edits
    2 adds in trunk

JavaScriptCore:

Reviewed by Maciej.

Fixed <rdar://problem/4370397> Missing return statement in
JSMethodNameToObjcMethodName.

JSMethodNameToObjcMethodName had a check for a name being too long, but
the check was missing a return statement.

A lot of this code was confusing and some of it was wrong, so I fixed
it up, added some asserts to catch this type of bug in the future,
changed some comments, and renamed some variables.

The two advantages of the new algorithm are (1) It makes writing past
the end of the buffer virtually impossible because the test on the main
loop is "while (not past end of buffer)" and (2) It's twice as fast
because it doesn't call strlen. (There's no need to call strlen when
we're walking the string ourselves.)

methodsNamed also supports arbitrary-length method names now. Just in
case the AppKit folks start getting REALLY verbose...

  • bindings/objc/objc_class.mm: (KJS::Bindings::ObjcClass::methodsNamed):
  • bindings/objc/objc_utility.h:
  • bindings/objc/objc_utility.mm: (KJS::Bindings::JSMethodNameToObjcMethodName):

LayoutTests:

Layout test for <rdar://problem/4370397> Missing return statement in
JSMethodNameToObjcMethodName.

  • fast/js/objc-big-method-name-expected.txt: Added.
  • fast/js/objc-big-method-name.html: Added.
10:29 AM Changeset in webkit [11681] by adele
  • 2 edits in trunk/WebKit

Reviewed by Darin, committed by Adele.

  • WebCoreSupport.subproj/WebTextRenderer.m: (createATSULayoutParameters): Assign initial values, which will never be used, to substituteRenderer and firstSmallCap, to avoid uninitialized variable warnings.
9:06 AM Changeset in webkit [11680] by ggaren
  • 8 edits
    3 adds in trunk

WebCore:

Reviewed by John.

Fixed <rdar://problem/4310363> JavaScript window.open: Height is 1
pixel short, and related bugs.

There were a few bugs here.
(1) Our code took size arguments and applied them to the window's

content rect. That's incorrect. The Rhino book says the arguments
should apply to the WebView. Other things that occupy the content
rect include the tab bar, the status bar, and the 1 pixel border
between brushed metal and document. All of these used to impinge
on the web page's display area.

The fix is to calculate sizing based on the WebView instead of
the content rect. This means that the webViewContentRect and
setContentRect delegate methods are obsolete and no longer called
by any of our code. (setContentRect was never called in the
first place.)

(2) None of our sizing accounted for scaled resolutions.

The fix is to ask the WebView to scale all coordintes for us.

(3) Our code assumed that all window accoutrements were on by default.

Safari works that way, but other WebKit clients might not.

The fix is always to explicitly set an on/off state.

(a) To facilitate scaling, I added a new bridge method, webView, to
access the webView.

(b) For internal consistency, I changed _Bars to _bars in bridge
methods, and _bars to _Bars in WinArgs data members. (Interestingly,
the different classes in our code are evenly divided on which format to
use.)

Added manual test:

  • manual-tests/window-open-features.html: Added.
  • manual-tests/resources/200x200.png: Added.
  • manual-tests/resources/popup200x200.html: Added.
  • khtml/ecma/kjs_window.cpp: (KJS::showModalDialog): see (b) (KJS::WindowFunc::callAsFunction): see (b)
  • kwq/KWQKHTMLPart.mm: (KWQKHTMLPart::statusbarVisible): see (b)
  • kwq/KWQKHTMLPartBrowserExtension.mm: (KHTMLPartBrowserExtension::createNewWindow): At the top of this method, I just did some formatting cleanup and moved the 'referrer' variable closer to where it's used. The changes in the middle of the method are (3), the bottom, (2).
  • kwq/KWQKPartsBrowserExtension.h: (KParts::WindowArgs::WindowArgs): see (b)
  • kwq/WebCoreBridge.h: see (a)

WebKit:

Reviewed by John.

Part of fix for <rdar://problem/4310363> JavaScript window.open: Height
is 1 pixel short, and related bugs. See WebCore ChageLog.

  • WebCoreSupport.subproj/WebBridge.m: (-[WebBridge webView]): Added method.
2:08 AM Changeset in webkit [11679] by eseidel
  • 2 edits in trunk/WebKit

Submitted by: eseidel
Reviewed by: mjs

Development-only build fix.

  • WebView.subproj/WebFrame.m: (-[WebFramePrivate dealloc]): fixed typo
1:59 AM Changeset in webkit [11678] by mjs
  • 2 edits in trunk/WebKit

Not reviewed.

  • revert accidental commit of this file.
1:30 AM Changeset in webkit [11677] by eseidel
  • 4 edits in trunk/WebCore

Bug #: 6156
Submitted by: eseidel
Reviewed by: mjs

Leaks when running SVG tests
​http://bugzilla.opendarwin.org/show_bug.cgi?id=6156
No additional tests necessary, leaks already caught by other tests.

  • kcanvas/KCanvasFilters.cpp: (KCanvasFEDiffuseLighting::setLightSource): takes ownership (KCanvasFESpecularLighting::setLightSource): takes ownership
  • kcanvas/KCanvasFilters.h: (KCanvasFEDiffuseLighting::KCanvasFEDiffuseLighting): added (KCanvasFEDiffuseLighting::~KCanvasFEDiffuseLighting): added (KCanvasFEDiffuseLighting::lightSource): fixed spacing (KCanvasFESpecularLighting::KCanvasFESpecularLighting): added (KCanvasFESpecularLighting::~KCanvasFESpecularLighting): added (KCanvasFESpecularLighting::lightSource): fixed spacing
  • kcanvas/device/quartz/KRenderingDeviceQuartz.mm: (KRenderingDeviceQuartz::stringForPath): added missing CFRelease
1:16 AM Changeset in webkit [11676] by eseidel
  • 4 edits in trunk/WebCore

Bug #: 5931
Submitted by: eseidel
Reviewed by: darin

Remove additional bit-rotted DEBUG* ifdefs from WebCore.
This removes PARSER_DEBUG, FORMS_DEBUG and CSS_STYLESHEET_DEBUG.
​http://bugzilla.opendarwin.org/show_bug.cgi?id=5931
No tests possible, only removing dead code.

  • khtml/css/css_stylesheetimpl.cpp: (CSSStyleSheetImpl::parseString): (CSSStyleSheetImpl::isLoading):
  • khtml/html/html_formimpl.cpp: (DOM::HTMLFormElementImpl::formData): (DOM::HTMLFormElementImpl::submit): (DOM::HTMLFormElementImpl::reset): (DOM::HTMLGenericFormElementImpl::getForm):
  • khtml/html/htmlparser.cpp: (HTMLParser::processCloseTag): (HTMLParser::createHead):
12:49 AM Changeset in webkit [11675] by eseidel
  • 12 edits in trunk/WebCore

Bug #: 6107
Submitted by: eseidel
Reviewed by: darin

Move Decoder onto Shared<T> and clients onto RefPtr.
​http://bugzilla.opendarwin.org/show_bug.cgi?id=6107
No test cases possible, no functional changes.

  • khtml/ecma/XSLTProcessor.cpp: (KJS::XSLTProcessorProtoFunc::callAsFunction):
  • khtml/ecma/xmlhttprequest.cpp: (KJS::XMLHttpRequest::XMLHttpRequest): (KJS::XMLHttpRequest::~XMLHttpRequest): (KJS::XMLHttpRequest::changeState): (KJS::XMLHttpRequest::abort): (KJS::XMLHttpRequest::slotFinished): (KJS::XMLHttpRequest::slotData):
  • khtml/ecma/xmlhttprequest.h:
  • khtml/khtml_part.cpp: (KHTMLPart::clear): (KHTMLPart::begin): (KHTMLPart::write):
  • khtml/khtmlpart_p.h: (KHTMLPartPrivate::KHTMLPartPrivate):
  • khtml/misc/decoder.cpp: (Decoder::Decoder): (Decoder::~Decoder):
  • khtml/misc/decoder.h:
  • khtml/misc/loader.h:
  • khtml/xml/dom_docimpl.cpp: (DocumentImpl::DocumentImpl): (DocumentImpl::~DocumentImpl): (DocumentImpl::prepareMouseEvent): (DocumentImpl::setDecoder):
  • khtml/xml/dom_docimpl.h: (DOM::DocumentImpl::decoder):
  • khtml/xsl/xslt_processorimpl.cpp: (DOM::XSLTProcessorImpl::createDocumentFromSource):
12:32 AM Changeset in webkit [11674] by mjs
  • 8 edits in trunk/WebKit

Reviewed by Eric.

  • Move handling of frame namespaces down to WebFrame.
  • Put some internal class declarations in the implementation file.
  • WebView.subproj/WebControllerSets.m: (+[WebFrameNamespaces addWebView:toFrameNamespace:]): (+[WebFrameNamespaces webViewsInFrameNamespace:]):
  • WebView.subproj/WebFrame.m: (-[WebFramePrivate dealloc]): (-[WebFrame _setFrameNamespace:]): (-[WebFrame _frameNamespace]):
  • WebView.subproj/WebFrameInternal.h:
  • WebView.subproj/WebFramePrivate.h:
  • WebView.subproj/WebView.m: (-[WebView _close]): (-[WebView _findFrameNamed:sourceFrame:]): (-[WebView setGroupName:]): (-[WebView groupName]):
  • WebView.subproj/WebViewInternal.h:
Note: See TracTimeline for information about the timeline view.