Context Navigation

FrameLoader::loadURL and FrameLoader::detachChildren went private
per changes in http://trac.webkit.org/changeset/42055. Follow
Mac's behaviour and use FrameLoader::loadURLIntoChildFrame for
loadURL and FrameLoader::detachParent for detachChildrem.

WebCoreSupport/FrameLoaderClientGtk.cpp: (WebKit::FrameLoaderClient::createFrame):
webkit/webkitwebview.cpp:

Mar 27, 2009:

10:11 PM Changeset in webkit [42068] by oliver@apple.com

2 edits in trunk/JavaScriptCore

Fix non-AllInOneFile builds.

9:04 PM Changeset in webkit [42067] by mrowe@apple.com

3 edits
1 add
1 delete in trunk/LayoutTests

Update text-content-type-with-binary-extension.html to use a .sitx file of random data rather
than a .m4a to ensure that WebKit doesn't handle the resource as a full-frame media element if the network
layer sniffs it as audio/mp4 rather than application/octet-stream.

Rubber-stamped by Dan Bernstein.

http/tests/loading/resources/.htaccess:
http/tests/loading/resources/binaryData.m4a: Removed.
http/tests/loading/resources/binaryData.sitx: Added.
http/tests/loading/text-content-type-with-binary-extension.html:

9:04 PM Changeset in webkit [42066] by mrowe@apple.com

1 edit in trunk/LayoutTests/platform/mac-snowleopard/Skipped

Fix syntax of Skipped file.

8:50 PM Changeset in webkit [42065] by oliver@apple.com

16 edits
3 adds in trunk

Improve performance of Function.prototype.call
<https://bugs.webkit.org/show_bug.cgi?id=24907>

Reviewed by Gavin Barraclough

Optimistically assume that expression.call(..) is going to be a call to
Function.prototype.call, and handle it specially to attempt to reduce the
degree of VM reentrancy.

When everything goes right this removes the vm reentry improving .call()
by around a factor of 10.

7:54 PM Changeset in webkit [42064] by mrowe@apple.com

2 edits in trunk/LayoutTests

Skip two plug-in tests that are failing due to <rdar://problem/6733642>.

Rubber-stamped by Oliver Hunt.

platform/mac-snowleopard/Skipped:

7:46 PM Changeset in webkit [42063] by mrowe@apple.com

3 edits
2 copies
4 moves
1 add in trunk/LayoutTests

Land updated results for two SVG tests involving Geeza Pro.

Rubber-stamped by Dan Bernstein.

platform/mac-leopard/svg/W3C-SVG-1.1/text-intro-05-t-expected.checksum: Renamed from LayoutTests/platform/mac/svg/W3C-SVG-1.1/text-intro-05-t-expected.checksum.
platform/mac-leopard/svg/W3C-SVG-1.1/text-intro-05-t-expected.png: Renamed from LayoutTests/platform/mac/svg/W3C-SVG-1.1/text-intro-05-t-expected.png.
platform/mac-leopard/svg/W3C-SVG-1.1/text-intro-05-t-expected.txt: Copied from LayoutTests/platform/mac/svg/W3C-SVG-1.1/text-intro-05-t-expected.txt.
platform/mac-leopard/svg/text/text-intro-05-t-expected.checksum: Renamed from LayoutTests/platform/mac/svg/text/text-intro-05-t-expected.checksum.
platform/mac-leopard/svg/text/text-intro-05-t-expected.png: Renamed from LayoutTests/platform/mac/svg/text/text-intro-05-t-expected.png.
platform/mac-leopard/svg/text/text-intro-05-t-expected.txt: Copied from LayoutTests/platform/mac/svg/text/text-intro-05-t-expected.txt.
platform/mac/svg/W3C-SVG-1.1/text-intro-05-t-expected.txt:
platform/mac/svg/text/text-intro-05-t-expected.txt:

7:40 PM Changeset in webkit [42062] by mrowe@apple.com

4 edits in trunk/WebCore

Fix two SVG test failures in 64-bit.

Reviewed by Sam Weinig.

getSubStringLength's arguments were declared as unsigned in the IDL and as signed long in the implementation.
This led to a value of -1 in JavaScript being converted to MAX_UINT in the bindings. In 32-bit this was
identical to -1 when interpeted as signed long, but in 64-bit it was still equal to MAX_UINT.

The solution for this is to use the IsIndex attribute on arguments that the SVG spec declares as "unsigned long"
but requires an exception be thrown when a negative value is passed. This results in the JS bindings handling the
check for a negative value and lets the implementation treat the arguments purely as unsigned values.

svg/SVGTextContentElement.cpp:

(WebCore::SVGTextContentElement::getNumberOfChars):
(WebCore::SVGTextContentElement::getSubStringLength):
(WebCore::SVGTextContentElement::getStartPositionOfChar):
(WebCore::SVGTextContentElement::getEndPositionOfChar):
(WebCore::SVGTextContentElement::getExtentOfChar):
(WebCore::SVGTextContentElement::getRotationOfChar):
(WebCore::SVGTextContentElement::getCharNumAtPosition):
(WebCore::SVGTextContentElement::selectSubString):

svg/SVGTextContentElement.h:
svg/SVGTextContentElement.idl:

5:49 PM Changeset in webkit [42061] by Darin Adler

3 edits in trunk/WebCore

2009-03-27 Darin Adler <Darin Adler>

Reviewed by Sam Weinig.

Get mailto code out of FrameLoader.

html/HTMLFormElement.cpp: (WebCore::transferMailtoPostFormDataToURL): Added. Factored out the mailto logic so HTMLFormElement::submit isn't full of ugly bits. This includes the part of the logic that involves transformeing the URL that was previously inside FrameLoader. (WebCore::HTMLFormElement::submit): Call transferMailtoPostFormDataToURL and also release the data for slightly less refcount churn.

loader/FrameLoader.cpp: (WebCore::FrameLoader::submitForm): Removed all the special casing for mailto, since HTMLFormElement now properly prepares both the URL and the form data.

5:41 PM Changeset in webkit [42060] by mitz@apple.com

5 edits in trunk/WebCore

Reviewed by Mark Rowe.

fix <rdar://problem/6724514> A bit of OpenTypeUtilities cleanup

platform/graphics/chromium/FontCustomPlatformData.cpp: (WebCore::EOTStream::EOTStream): Changed to use EOTHeader. (WebCore::createFontCustomPlatformData): Ditto.
platform/graphics/opentype/OpenTypeUtilities.cpp: (WebCore::EOTHeader::EOTHeader): Added. Initializes the buffer size to the size of an EOTPrefix. (WebCore::EOTHeader::updateEOTSize): Added. Updates the size field in the prefix. (WebCore::EOTHeader::appendBigEndianString): Changed the static appendBigEndianStringToEOTHeader() into this member function. (WebCore::EOTHeader::appendPaddingShort): Added. (WebCore::getEOTHeader): Changed to use EOTHeader.
platform/graphics/opentype/OpenTypeUtilities.h: (WebCore::EOTHeader::size): (WebCore::EOTHeader::data): (WebCore::EOTHeader::prefix):
platform/graphics/win/FontCustomPlatformData.cpp: (WebCore::EOTStream::EOTStream): Changed to use EOTHeader. (WebCore::createFontCustomPlatformData): Ditto.

5:06 PM Changeset in webkit [42059] by Darin Adler

2 edits in trunk/WebKit

2009-03-27 Darin Adler <Darin Adler>

StringsNotToBeLocalized.txt: Updated for recent changes.

5:06 PM Changeset in webkit [42058] by Darin Adler

2 edits in trunk/WebKitTools

2009-03-27 Darin Adler <Darin Adler>

Reviewed by Sam Weinig.

Scripts/extract-localizable-strings: Fixed version check for perl 5.10 compatibility.

4:55 PM Changeset in webkit [42057] by Darin Adler

19 edits in trunk/LayoutTests

2009-03-27 Darin Adler <Darin Adler>

Reviewed by Sam Weinig.

Made our mailto form submission tests work more reliably by using the new
layoutTestController.waitForPolicyDelegate function.

fast/encoding/mailto-always-utf-8.html: Updated test to use waitForPolicyDelegate rather than relying on the form submission happening immediately.
fast/forms/mailto/advanced-get.html: Ditto.
fast/forms/mailto/advanced-put.html: Ditto.
fast/forms/mailto/get-multiple-items-text-plain.html: Ditto.
fast/forms/mailto/get-multiple-items-x-www-form-urlencoded.html: Ditto.
fast/forms/mailto/get-multiple-items.html: Ditto.
fast/forms/mailto/get-non-ascii-always-utf-8.html: Ditto.
fast/forms/mailto/get-non-ascii-text-plain-latin-1.html: Ditto.
fast/forms/mailto/get-non-ascii-text-plain.html: Ditto.
fast/forms/mailto/get-non-ascii.html: Ditto.
fast/forms/mailto/get-overwrite-query.html: Ditto.
fast/forms/mailto/post-append-query.html: Ditto.
fast/forms/mailto/post-multiple-items-multipart-form-data.html: Ditto.
fast/forms/mailto/post-multiple-items-text-plain.html: Ditto.
fast/forms/mailto/post-multiple-items-x-www-form-urlencoded.html: Ditto.
fast/forms/mailto/post-multiple-items.html: Ditto.
fast/forms/mailto/post-text-plain-with-accept-charset.html: Ditto.
fast/forms/mailto/post-text-plain.html: Ditto.

4:55 PM Changeset in webkit [42056] by Darin Adler

15 edits in trunk/WebKitTools

2009-03-27 Darin Adler <Darin Adler>

Reviewed by Sam Weinig.

Added waitForPolicyDelegate, a more-reliable way to make our mailto form submission
tests work.

DumpRenderTree/AccessibilityController.cpp: (getFocusedElementCallback): Use static_cast instead of reinterpret_cast because there's no reason to use reinterpret_cast just to get from void* to a specific type. (getRootElementCallback): Ditto.
DumpRenderTree/AccessibilityUIElement.cpp: (toAXElement): Ditto.
DumpRenderTree/GCController.cpp: (collectCallback): Ditto. (collectOnAlternateThreadCallback): Ditto. (getJSObjectCountCallback): Ditto.
DumpRenderTree/LayoutTestController.cpp: (many functions): Ditto. (waitForPolicyDelegateCallback): Added. (LayoutTestController::staticFunctions): Sorted functions by name. Added waitForPolicyDelegate.

DumpRenderTree/LayoutTestController.h: Sorted functions by name. Added waitForPolicyDelegate.

DumpRenderTree/mac/DumpRenderTree.mm: (resetWebViewToConsistentStateBeforeTesting): Added code to reset the state of the policy delegate. (runTest): Added an additional call to resetWebViewToConsistentStateBeforeTesting just before loading an empty page. This prevents extra policy delegate calls from being logged.

DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::waitForPolicyDelegate): Added. Puts us into wait mode, and then tells the policy delegate to notify when done, then points the web view at the policy delegate.

DumpRenderTree/mac/PolicyDelegate.h: Added setControllerToNotifyDone: method.

DumpRenderTree/mac/PolicyDelegate.mm: (-[PolicyDelegate webView:decidePolicyForNavigationAction:request:frame:decisionListener:]): Added code to do notifyDone if controllerToNotifyDone is set. (-[PolicyDelegate setControllerToNotifyDone:]): Added.

DumpRenderTree/win/DumpRenderTree.cpp: (resetWebViewToConsistentStateBeforeTesting): Added code to reset the state of the policy delegate. Also moved the call to resetUndoManager in here. (runTest): Removed the call to resetUndoManager (see above). Added an additional call to resetWebViewToConsistentStateBeforeTesting to match the Mac code.

DumpRenderTree/win/LayoutTestControllerWin.cpp: (LayoutTestController::setCustomPolicyDelegate): 0, not NULL.

DumpRenderTree/win/PolicyDelegate.cpp: (PolicyDelegate::PolicyDelegate): Initialize m_controllerToNotifyDone to 0. (PolicyDelegate::decidePolicyForNavigationAction): Added code to do notifyDone if m_controllerToNotifyDone is set.

DumpRenderTree/win/PolicyDelegate.h: Added setControllerToNotifyDone and m_controllerToNotifyDone.

4:37 PM Changeset in webkit [42055] by Darin Adler

6 edits in trunk/WebCore

2009-03-27 Darin Adler <Darin Adler>

Reviewed by Sam Weinig.

To prepare for work on a form loading fix, remove some unused functions,
consolidate some functions, and make many more functions private in
FrameLoader.

WebCore.base.exp: Remove some unused entry points, update another.

dom/Document.cpp: (WebCore::Document::detach): Clear m_frame directly instead of using clearFramePointer.
dom/Document.h: Ditto.

loader/FrameLoader.cpp: (WebCore::FrameLoader::urlSelected): Consolidate the two functions of this name into a single function. Also changed the event argument to be PassRefPtr<Event>. (WebCore::FrameLoader::submitForm): Changed the event argument to be PassRefPtr<Event>. (WebCore::FrameLoader::scheduleLocationChange): Call completeURL explicitly so we can remove the version of changeLocation that does completeURL for us. (WebCore::FrameLoader::redirectionTimerFired): Ditto. (WebCore::FrameLoader::loadURLIntoChildFrame): Set m_provisionalHistoryItem directly so we don't need a setProvisionalHistoryItem function. (WebCore::FrameLoader::canCachePageContainingThisFrame): Get at m_quickRedirectComing directly so we don't need a isQuickRedirectComing function. (WebCore::FrameLoader::logCanCacheFrameDecision): Ditto. (WebCore::FrameLoader::loadFrameRequestWithFormAndValues): Changed the event argument to be PassRefPtr<Event>. (WebCore::FrameLoader::loadURL): Move the logic of the continueLoadWithData function here because this is the only caller. If we want to make the resulting function smaller we should refactor some other way. Also streamlined the user-chosen encoding logic, but did not change what it does. (WebCore::FrameLoader::finishedLoadingDocument): Changed the event argument to be PassRefPtr<Event>. (WebCore::FrameLoader::loadPostRequest): Ditto. (WebCore::FrameLoader::receivedMainResourceError): Moved the logic from the didNotOpenURL function here, since this was the only caller.

loader/FrameLoader.h: Removed include of FormState.h and used a forward declaration instead. Removed unneeded forward declation of Element. Moved many functions into the private section, and removed some other unused or uneeded functions.

4:23 PM Changeset in webkit [42054] by darin@chromium.org

4 edits
2 adds in trunk

2009-03-27 Xiaomei Ji <xji@chromium.org>

Reviewed by Dan Bernstein.

Fix https://bugs.webkit.org/show_bug.cgi?id=24303
Using keyboard select RTL text, Highlights goes to opposite direction from Firefox and IE.

Test: editing/selection/extend-selection.html

editing/SelectionController.cpp: (WebCore::SelectionController::directionOfEnclosingBlock): Added (WebCore::SelectionController::modifyExtendingRight): Added. Currenctly implemented for character and word granularity, all other granularities are treated as "forward". (WebCore::SelectionController::modifyExtendingForward): Renamed modifyExtendingRightForward() to this. (WebCore::SelectionController::modifyExtendingLeft): Added. Currenctly implemented for character and word granularity, all other granularities are treated as "backward". (WebCore::SelectionController::modifyExtendingBackward): Renamed modifyExtendingLeftBackward() to this. (WebCore::SelectionController::modify): Change to call either the left/right or backward/forward methods depending on the 'dir' argument for extends.
editing/SelectionController.h:

4:15 PM Changeset in webkit [42053] by weinig@apple.com

3 edits in trunk/WebKitTools

2009-03-27 Sam Weinig <sam@webkit.org>

Reviewed by Gavin Barraclough.

Fix for https://bugs.webkit.org/show_bug.cgi?id=24765
prepare-ChangeLog and svn-create-patch doesn't work with svn 1.6

Scripts/prepare-ChangeLog:
Scripts/svn-create-patch:

3:34 PM Changeset in webkit [42052] by dimich@chromium.org

2 edits in trunk/WebCore

2009-03-27 John Abd-El-Malek <jam@google.com>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=24836
Fix navigator.plugins.refresh(false) in Chromium.

plugins/chromium/PluginDataChromium.cpp: (WebCore::PluginData::refresh):

3:31 PM Changeset in webkit [42051] by timothy@apple.com

4 edits in trunk/WebCore

Make scrollLeft, scrollTop, scrollWidth, and scrollHeight
virtual to fix the broken layout tests from my last commit.

Reviewed by Mark Rowe.

dom/Element.cpp: (WebCore::Element::scrollLeft): (WebCore::Element::scrollTop): (WebCore::Element::scrollWidth): (WebCore::Element::scrollHeight):
dom/Element.h:
html/HTMLBodyElement.h:

2:34 PM Changeset in webkit [42050] by mrowe@apple.com

2 edits in trunk/WebCore

Fix <https://bugs.webkit.org/show_bug.cgi?id=24876>.
Bug 24876: fast/forms/select-max-length.html times out in debug builds due to HTMLSelectElement::setLength being O(N²⁾

Reviewed by Sam Weinig.

html/HTMLSelectElement.cpp:

(WebCore::HTMLSelectElement::setLength): Repeatedly calling remove to remove elements causes us to recalculate the list
items after each node is removed, leading to O(N^{2) behaviour. By inlining the batch removal in to setLength we can avoid

this gratuitous recalcuation.}

1:45 PM Changeset in webkit [42049] by krit@webkit.org

2 edits in trunk/WebCore

2009-03-27 Dirk Schulze <krit@webkit.org>

Reviewed by Eric Seidel.

We don't support images for GtkCursors yet. We need a placeholder to avoid crashes.
I took crosshair to have a common behavior with firefox.

[GTK] SVG - Crash on setting the cursor icon
https://bugs.webkit.org/show_bug.cgi?id=24565

platform/gtk/CursorGtk.cpp: (WebCore::Cursor::Cursor):

1:36 PM Changeset in webkit [42048] by timothy@apple.com

7 edits in trunk

Remove redundant attributes and functions from IDL files that have the
same attributes and functions defined in the super class.

Reviewed by Simon Fraser.

WebCore.xcodeproj/project.pbxproj: Remove DOMHTMLBodyElementPrivate.h.
bindings/objc/PublicDOMInterfaces.h: Remove focus and blur from DOMHTMLElement.
html/HTMLBodyElement.idl: Remove scrollLeft, scrollTop, scrollWidth and scrollHeight.
html/HTMLElement.idl: Remove blur and focus.

11:44 AM Changeset in webkit [42047] by timothy@apple.com

2 edits in trunk/WebKitTools/DumpRenderTree/mac

Fix the Mac DumpRenderTree build.

11:37 AM Changeset in webkit [42046] by Adam Roben

6 edits in trunk

Don't include substitute data URLs in global history redirect chains

WebCore:

Don't include substitute data URLs in global history redirect chains

<rdar://6690169>

Reviewed by Darin Adler.

This might be testable if:

support were added to DRT for providing substitute data on failed loads
support were added to DRT for dumping redirect chains

I tried doing the above and was still unable to make a test.

loader/FrameLoader.cpp: (WebCore::FrameLoader::updateHistoryForStandardLoad): (WebCore::FrameLoader::updateHistoryForRedirectWithLockedBackForwardList): Call updateGlobalHistoryRedirectLinks only if the load succeeded (i.e., there is no unreachableURL). Previous the FrameLoaderClient implementations were unconditionally calling updateGlobalHistoryRedirectLinks from within updateGlobalHistory.

WebKit/mac:

Don't include substitute data URLs in global history redirect chains

<rdar://6690169>

Reviewed by Darin Adler.

WebCoreSupport/WebFrameLoaderClient.mm: (WebFrameLoaderClient::updateGlobalHistory): Don't call updateGlobalHistoryRedirectLinks. FrameLoader calls this for us now. (WebFrameLoaderClient::updateGlobalHistoryRedirectLinks): Added an assertion to help catch cases where we might be adding a substitute data URL into a redirect chain.

WebKit/win:

Don't include substitute data URLs in global history redirect chains

<rdar://6690169>

Reviewed by Darin Adler.

WebCoreSupport/WebFrameLoaderClient.cpp: (WebFrameLoaderClient::updateGlobalHistory): Don't call updateGlobalHistoryRedirectLinks. FrameLoader calls this for us now. (WebFrameLoaderClient::updateGlobalHistoryRedirectLinks): Added an assertion to help catch cases where we might be adding a substitute data URL into a redirect chain.

11:27 AM Changeset in webkit [42045] by Darin Adler

2 edits in trunk/WebKit/mac

2009-03-27 Darin Adler <Darin Adler>

Reviewed by Adam Roben.

<rdar://problem/6541923> REGRESSION (r38629): Tab cycle in empty tab is broken

WebView/WebHTMLView.mm: (-[WebHTMLView _wantsKeyDownForEvent:]): Only return YES when we have a Frame.

11:07 AM Changeset in webkit [42044] by pkasting@chromium.org

5 edits
1 delete in trunk/WebCore

2009-03-27 Peter Kasting <pkasting@google.com>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=24861
Chromium port did not render decorations for <input type="search">.
This copies RenderThemeWin.*.

css/themeChromiumWin.css: Removed.
rendering/RenderThemeChromiumLinux.cpp: (WebCore::RenderThemeChromiumLinux::extraDefaultStyleSheet): (WebCore::RenderThemeChromiumLinux::adjustSearchFieldCancelButtonStyle): (WebCore::RenderThemeChromiumLinux::paintSearchFieldCancelButton): (WebCore::RenderThemeChromiumLinux::adjustSearchFieldDecorationStyle): (WebCore::RenderThemeChromiumLinux::adjustSearchFieldResultsDecorationStyle): (WebCore::RenderThemeChromiumLinux::paintSearchFieldResultsDecoration): (WebCore::RenderThemeChromiumLinux::adjustSearchFieldResultsButtonStyle): (WebCore::RenderThemeChromiumLinux::paintSearchFieldResultsButton):
rendering/RenderThemeChromiumLinux.h: (WebCore::RenderThemeChromiumLinux::paintSearchField):
rendering/RenderThemeChromiumWin.cpp: (WebCore::): (WebCore::RenderThemeChromiumWin::extraDefaultStyleSheet): (WebCore::RenderThemeChromiumWin::adjustSearchFieldCancelButtonStyle): (WebCore::RenderThemeChromiumWin::paintSearchFieldCancelButton): (WebCore::RenderThemeChromiumWin::adjustSearchFieldDecorationStyle): (WebCore::RenderThemeChromiumWin::adjustSearchFieldResultsDecorationStyle): (WebCore::RenderThemeChromiumWin::paintSearchFieldResultsDecoration): (WebCore::RenderThemeChromiumWin::adjustSearchFieldResultsButtonStyle): (WebCore::RenderThemeChromiumWin::paintSearchFieldResultsButton):
rendering/RenderThemeChromiumWin.h: (WebCore::RenderThemeChromiumWin::paintSearchField):

11:02 AM Changeset in webkit [42043] by hyatt@apple.com

7 edits
4 adds in trunk

WebCore:

2009-03-27 David Hyatt <hyatt@apple.com>

Reviewed by Simon Fraser

If an object has a self-painting layer, don't count it as part of a block's visual overflow.
This fix has only been made for block-level children. The inline-level case is still broken
(and covered by an existing bug).

Added fast/block/positioning/negative-rel-position.html

rendering/InlineFlowBox.cpp: (WebCore::InlineFlowBox::placeBoxesVertically):
rendering/RenderBlock.cpp: (WebCore::RenderBlock::layoutBlockChildren): (WebCore::RenderBlock::lowestPosition): (WebCore::RenderBlock::rightmostPosition): (WebCore::RenderBlock::leftmostPosition):

LayoutTests:

2009-03-27 David Hyatt <hyatt@apple.com>

Reviewed by Simon Fraser

Test case for relative positioning bug.

fast/block/positioning/negative-rel-position.html: Added.
platform/mac/fast/block/positioning/negative-rel-position-expected.checksum: Added.
platform/mac/fast/block/positioning/negative-rel-position-expected.png: Added.
platform/mac/fast/block/positioning/negative-rel-position-expected.txt: Added.

11:00 AM Changeset in webkit [42042] by andersca@apple.com

2 edits in trunk/WebCore

2009-03-27 Anders Carlsson <andersca@apple.com>

Reviewed by Darin Adler.

<rdar://problem/6642221>
REGRESSION: With the Movie widget, movie trailer fails to load if movie had been previously viewed

Add a dashboard specific workaround so that we always create renderers for object elements, even if display:none is set.

I've filed <rdar://problem/6731022> against the Movies widget.

html/HTMLObjectElement.cpp: (WebCore::HTMLObjectElement::rendererIsNeeded):

10:53 AM Changeset in webkit [42041] by Darin Adler

5 edits in trunk/WebKit/mac

2009-03-27 Darin Adler <Darin Adler>

Reviewed by John Sullivan and Anders Carlsson.

<rdar://problem/5987442> Pasteboard not exposed to WebEditingDelegate for WebViewInsertActionPasted (needed for system services)

Added SPI to tell which pasteboard is currently being inserted.
I chose to put it on WebView to be forward-looking since we're migrating things
from WebHTMLView to WebView in the future.

WebView/WebHTMLView.mm: (-[WebHTMLView _pasteWithPasteboard:allowPlainText:]): Add calls to _setInsertionPasteboard. (-[WebHTMLView _pasteAsPlainTextWithPasteboard:]): Ditto.
WebView/WebView.mm: (-[WebViewPrivate dealloc]): Assert the pasteboard is nil. (-[WebViewPrivate finalize]): Ditto. (-[WebView _insertionPasteboard]): Return the pastebaord. (-[WebView _setInsertionPasteboard:]): Set the pasteboard.
WebView/WebViewInternal.h: Added _setInsertionPasteboard.
WebView/WebViewPrivate.h: Added _insertionPasteboard.

10:50 AM Changeset in webkit [42040] by timothy@apple.com

8 edits
1 add in trunk

Expose new DOM methods as public Objective-C API.

WebCore:

2009-03-25 Timothy Hatcher <timothy@apple.com>

Expose new DOM methods as public Objective-C API.

<rdar://problem/5837350> Expose new DOM classes and methods
as public API (match the additions to the JavaScript DOM)

Reviewed by NOBODY (OOPS!).

WebCore.xcodeproj/project.pbxproj:
bindings/objc/DOMHTML.h:
bindings/objc/DOMPrivate.h:
bindings/objc/PublicDOMInterfaces.h:

WebKit/mac:

2009-03-25 Timothy Hatcher <timothy@apple.com>

Expose new DOM methods as public Objective-C API.

<rdar://problem/5837350> Expose new DOM classes and methods
as public API (match the additions to the JavaScript DOM)

Reviewed by NOBODY (OOPS!).

MigrateHeaders.make:

10:45 AM Changeset in webkit [42039] by Simon Hausmann

2 edits in trunk/WebCore

2009-03-27 Zack Rusin <zack@kde.org>

Reviewed by Simon Hausmann.

https://bugs.webkit.org/show_bug.cgi?id=24280

Fix propagation of fill rules when rendering paths in the Qt build.

10:40 AM Changeset in webkit [42038] by Simon Hausmann

3 edits in trunk/WebCore

2009-03-27 Zack Rusin <zack@kde.org>

Reviewed by Tor Arne Vestbø.

https://bugs.webkit.org/show_bug.cgi?id=24275

Fix text field theming in the Qt build with the KDE 4 Oxygen
style by adjusting the size vertically and horizontally to
set padding on the element equal to the width of the style painted border.

9:23 AM Changeset in webkit [42037] by Simon Hausmann

3 edits in trunk/JavaScriptCore

2009-03-27 Laszlo Gombos <Laszlo Gombos>

Reviewed by Darin Adler.

Bug 24884: Include strings.h for strcasecmp()
https://bugs.webkit.org/show_bug.cgi?id=24884

6:05 AM Changeset in webkit [42036] by Simon Hausmann

2 edits in trunk/WebCore

2009-03-27 Laszlo Gombos <Laszlo Gombos>

Reviewed by Simon Hausmann.

https://bugs.webkit.org/show_bug.cgi?id=24852
Build fix for GNU mode RVCT compilation

5:59 AM Changeset in webkit [42035] by Simon Hausmann

2 edits in trunk/WebKit/qt

2009-03-27 Erik L. Bunce <elbunce@xendom.com>

Reviewed by Simon Hausmann.

https://bugs.webkit.org/show_bug.cgi?id=24869

Fixes an incorrect case in tst_qwebpage.

5:50 AM Changeset in webkit [42034] by Simon Hausmann

2 edits in trunk/WebCore

2009-03-27 Erik L. Bunce <elbunce@xendom.com>

Reviewed by Simon Hausmann.

https://bugs.webkit.org/show_bug.cgi?id=24868

Make text of the writing direction items more user friendly.

5:47 AM Changeset in webkit [42033] by Simon Hausmann

2 edits in trunk/WebCore

2009-03-27 Erik L. Bunce <elbunce@xendom.com>

Reviewed by Simon Hausmann.

https://bugs.webkit.org/show_bug.cgi?id=24274

Fixed the Qt 4.4 Mac build with netscape plugins enabled.

4:54 AM Changeset in webkit [42032] by Simon Hausmann

4 edits in trunk/WebKit/qt

2009-03-27 Erik L. Bunce <elbunce@xendom.com>

Reviewed by Simon Hausmann.

https://bugs.webkit.org/show_bug.cgi?id=24746

Add new editing related QWebPage::WebActions

Clipboard: PasteAndMatchStyle
Formatting: RemoveFormat, ToggleStrikethrough, ToggleSubscript, and ToggleSuperscript
List Editing: InsertUnorderedList, InsertOrderedList, Indent, Outdent
Paragraph Justification: AlignCenter, AlignJustified, AlignLeft, AlignRight

Improved selection and editing tests.

Note: See TracTimeline for information about the timeline view.

Download in other formats:

RSS Feed