Timeline


and

07/01/09:

23:54 Changeset [45469] by eric@webkit.org

2009-07-01 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

bugzilla-tool fails for SVN users
https://bugs.webkit.org/show_bug.cgi?id=26914

To fix this I moved svn from -F - to using -m
In order for -m to work I had to move us off of shell=True
To move off of shell=True all call sites for run_command
which take args, need to pass their args as an list instead of a string.

In order for the final bug update to work correctly, I had to
abstract the way that we parse out revision numbers from the commit text.

  • Scripts/bugzilla-tool:
  • Scripts/modules/scm.py:
23:30 Changeset [45468] by simon.fraser@apple.com

2009-07-01 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dan Bernstein.

<video> fails to show on http://camendesign.com/code/video_for_everybody
<rdar://problem/7026010>
https://bugs.webkit.org/show_bug.cgi?id=26919

Fix an issue introduced in r44961. In that revision we changed to only update
compositing layer geometry when all siblings had been laid out (i.e. we pushed
the updates one level down). However, that left out the root layer, so this
fix ensures that the root layer geometry gets updated at the end.

Test: compositing/geometry/root-layer-update.html

  • rendering/RenderLayerBacking.cpp: (WebCore::RenderLayerBacking::updateAfterLayout):
23:26 Changeset [45467] by darin@chromium.org

2009-07-01 John Abd-El-Malek <jam@chromium.org>

Reviewed by Darin Fisher.

Small refactoring of MessagePortChannel so that PlatformMessagePortChannel
may be defined at the WebKit layer.

https://bugs.webkit.org/show_bug.cgi?id=26905

  • dom/MessageChannel.cpp: (WebCore::MessageChannel::MessageChannel):
  • dom/MessagePortChannel.cpp:
  • dom/MessagePortChannel.h:
  • dom/default/PlatformMessagePortChannel.cpp: (WebCore::MessagePortChannel::createChannel): (WebCore::MessagePortChannel::create): (WebCore::MessagePortChannel::MessagePortChannel): (WebCore::MessagePortChannel::~MessagePortChannel):
23:04 Changeset [45466] by xan@webkit.org

2009-07-01 Xan Lopez <xlopez@igalia.com>

Disable accessibility tests until we have the accessibility
controller.

  • platform/gtk/Skipped:
22:13 Changeset [45465] by simon.fraser@apple.com

2009-07-01 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dan Bernstein

<rdar://problem/7026010> <video> fails to show on http://camendesign.com/code/video_for_everybody

First part of fix: when outline width changes, don't to a synchronous
layer update right away, but just set the flag to say that compositing
layers need updating. The synchronous layer update left the layer geometries
out of whack when it happened in the middle of layout (e.g. when inside ::first-letter).

Test: compositing/geometry/outline-change.html

  • rendering/RenderView.cpp: (WebCore::RenderView::setMaximalOutlineSize):
22:13 Changeset [45464] by eric@webkit.org

2009-07-01 Eric Seidel <eric@webkit.org>

Reviewed by Maciej Stachowiak.

prepare-ChangeLog should have a --bug= argument and use it for url autofill
https://bugs.webkit.org/show_bug.cgi?id=26383

prepare-ChangeLog now knows how to grab the bug title from bugs.webkit.org
I also added a bit more template text in an effort to get better ChangeLogs
from casual contributers.

  • Scripts/prepare-ChangeLog:
19:08 Changeset [45463] by eric@webkit.org

2009-07-01 Robert Hogan <robert@roberthogan.net>

Reviewed by NOBODY.


Fix Qt segfault when javascript disabled.
If clients call addToJavaScriptWindowObject even though JavascriptEnabled is false
webkit will segfault on the assert:

ASSERTION FAILED: _rootObject
(../../../WebCore/bridge/runtime.cpp:52
JSC::Bindings::Instance::Instance(WTF::PassRefPtr<JSC::Bindings::RootObject>))

Fix is to ensure JavaScript is enabled when client calls addToJavaScriptWindowObject.

https://bugs.webkit.org/show_bug.cgi?id=26906

  • Api/qwebframe.cpp: (QWebFrame::addToJavaScriptWindowObject):
19:01 Changeset [45462] by eric@webkit.org

2009-07-01 Kwang Yul Seo <skyul@company100.net>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=26842
Build fix when ENABLE_DATABASE is off

Move Database.h into ENABLE(DATABASE) guard so toggling ENABLE_DATABASE
off does not break builds.

  • bindings/js/ScriptObjectQuarantine.cpp:
18:35 Changeset [45461] by abarth@webkit.org

WebCore:

2009-07-01 Daniel Bates <dbates@intudata.com>

Reviewed by Adam Barth.


https://bugs.webkit.org/show_bug.cgi?id=26899


Modified XSSAuditor::decodeURL to only remove null characters so that
it is consistent with the behavior of HTMLTokenizer and prevents
injected scripts that contain control characters.

Tests: http/tests/security/xssAuditor/script-tag-control-char.html

http/tests/security/xssAuditor/script-tag-null-char.html

  • page/XSSAuditor.cpp: (WebCore::XSSAuditor::decodeURL): Modified to only remove null characters.
  • page/XSSAuditor.h: Reverted naming of third argument of method XSSAuditor::decodeURL from allowControlCharacters back to allowNullCharacters.
  • platform/network/ResourceResponseBase.cpp: Reverted back to rev #45003. (WebCore::isControlCharacter):
  • platform/network/ResourceResponseBase.h: Reverted back to rev #45003.

LayoutTests:

2009-07-01 Daniel Bates <dbates@intudata.com>

Reviewed by Adam Barth.


https://bugs.webkit.org/show_bug.cgi?id=26899


Tests that XSSAuditor prevents injected scripts that contain control characters.

  • http/tests/security/xssAuditor/script-tag-control-char-expected.txt: Added.
  • http/tests/security/xssAuditor/script-tag-control-char.html: Added.
  • http/tests/security/xssAuditor/script-tag-null-char-expected.txt: Added.
  • http/tests/security/xssAuditor/script-tag-null-char.html: Added.
18:03 Changeset [45460] by eric@webkit.org

2009-07-01 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Timothy Hatcher.

  • supported keyboard shortcuts compatible with Firebug in Scripts Debugger: F5, Ctrl+/ (win,lin), Command+/ (mac) - continue F10, Ctrl+' (win, lin), Command+' (mac) - step over F11, Ctrl+; (win, lin), Command+; (mac) - step into Shift+F11, Ctrl+Shift+; (win, lin), Command+Shift+; (mac) - step out
  • added call stack navigation shortcuts: Ctrl+. - next call frame Ctrl+, - previouse call frame

https://bugs.webkit.org/show_bug.cgi?id=23849

  • inspector/front-end/CallStackSidebarPane.js: (WebInspector.CallStackSidebarPane): (WebInspector.CallStackSidebarPane.prototype.handleKeyEvent): (WebInspector.CallStackSidebarPane.prototype._selectNextCallFrameOnStack): (WebInspector.CallStackSidebarPane.prototype._selectPreviousCallFrameOnStack): (WebInspector.CallStackSidebarPane.prototype._selectedPlacardByIndex): (WebInspector.CallStackSidebarPane.prototype._selectedCallFrameIndex):
  • inspector/front-end/KeyboardShortcut.js: Added. (WebInspector.KeyboardShortcut): (WebInspector.KeyboardShortcut.makeKey): (WebInspector.KeyboardShortcut.makeKeyFromEvent): (WebInspector.KeyboardShortcut.makeKeyFromCodeAndModifiers_):
  • inspector/front-end/ScriptsPanel.js: (WebInspector.ScriptsPanel): (WebInspector.ScriptsPanel.prototype.handleKeyEvent):
  • inspector/front-end/WebKit.qrc:
  • inspector/front-end/inspector.html:
17:54 Changeset [45459] by agl@chromium.org

2009-07-01 Daniel Erat <derat@google.com>

Reviewed by David Levin.

Chromium Linux: Move the scrollbar thumb on middle-click.

https://bugs.webkit.org/show_bug.cgi?id=26910
http://code.google.com/p/chromium/issues/detail?id=11976

Tested by building Chrome and checking that the scrollbar moves as expected.

  • platform/chromium/ScrollbarThemeChromium.cpp:
  • platform/chromium/ScrollbarThemeChromium.h:
  • platform/chromium/ScrollbarThemeChromiumLinux.cpp:
  • platform/chromium/ScrollbarThemeChromiumLinux.h:
  • platform/chromium/ScrollbarThemeChromiumWin.cpp:
  • platform/chromium/ScrollbarThemeChromiumWin.h:
17:38 Changeset [45458] by beidson@apple.com

WebCore:

2009-07-01 Brady Eidson <beidson@apple.com>

Reviewed by Sam Weinig.

Add a preference/setting to toggle whether content sniffing is enabled for file URLs.

  • WebCore.base.exp:
  • page/Settings.cpp: (WebCore::Settings::setLocalFileContentSniffingEnabled):
  • page/Settings.h: (WebCore::Settings::localFileContentSniffingEnabled):
  • platform/network/mac/ResourceHandleMac.mm: (WebCore::ResourceHandle::start): (-[WebCoreResourceHandleAsDelegate connection:didReceiveResponse:]):

WebKit/mac:

2009-07-01 Brady Eidson <beidson@apple.com>

Reviewed by Sam Weinig.

Add a preference/setting to toggle whether content sniffing is enabled for file URLs.

  • WebView/WebPreferenceKeysPrivate.h:
  • WebView/WebPreferences.mm: (+[WebPreferences initialize]): (-[WebPreferences localFileContentSniffingEnabled]): (-[WebPreferences setLocalFileContentSniffingEnabled:]):
  • WebView/WebPreferencesPrivate.h:
  • WebView/WebView.mm: (-[WebView _preferencesChangedNotification:]):
17:13 Changeset [45457] by weinig@apple.com

2009-07-01 Sam Weinig <sam@webkit.org>

Reviewed by Gavin Barraclough.

Don't reload values in emitBinaryDoubleOp.

SunSpider reports a 0.6% progression.

  • jit/JIT.h:
  • jit/JITArithmetic.cpp: (JSC::JIT::emit_op_jnless): (JSC::JIT::emit_op_jnlesseq): (JSC::JIT::emitBinaryDoubleOp):
16:34 Changeset [45456] by cfleizach@apple.com

Bug 26909: aria-label needs to be supported
https://bugs.webkit.org/show_bug.cgi?id=26909

16:31 Changeset [45455] by eric@webkit.org

2009-07-01 Eric Seidel <eric@webkit.org>

Reviewed by Adam Roben.

prepare-ChangeLog should fail when EMAIL_ADDRESS or REAL_NAME are missing:
https://bugs.webkit.org/show_bug.cgi?id=26692

I also made prepare-ChangeLog sanity-check names and email addresses a little.
Names must contain a space, and email addresses must contain '@'

Also "fixed" $email_address to $emailAddress. This script uses mixed variable name styles.

  • Scripts/prepare-ChangeLog:
16:15 Changeset [45454] by hyatt@apple.com

2009-07-01 David Hyatt <hyatt@apple.com>

Reviewed by Oliver Hunt.

More datagrid columns work. Make sure columns cache both a style for the column header as well
as a style for the column background (behind the cells).

Hook up to CSSStyleSelector methods for obtaining the pseudo styles for columns (stubbed out and
unimplemented).

  • css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::pseudoStyleForDataGridColumn): (WebCore::CSSStyleSelector::pseudoStyleForDataGridColumnHeader):
  • css/CSSStyleSelector.h:
  • html/DataGridColumn.h: (WebCore::DataGridColumn::setColumnList): (WebCore::DataGridColumn::columnStyle): (WebCore::DataGridColumn::setColumnStyle): (WebCore::DataGridColumn::headerStyle): (WebCore::DataGridColumn::setHeaderStyle):
  • rendering/RenderDataGrid.cpp: (WebCore::RenderDataGrid::recalcStyleForColumn): (WebCore::RenderDataGrid::columnStyle): (WebCore::RenderDataGrid::headerStyle): (WebCore::RenderDataGrid::paintColumnHeaders):
  • rendering/RenderDataGrid.h:
16:12 JSC edited by bfulgham@webkit.org
(diff)
16:06 JSC edited by bfulgham@webkit.org
Add information about built-in functions. (diff)
15:33 Changeset [45453] by eric.carlson@apple.com

2009-07-01 Eric Carlson <eric.carlson@apple.com>

Reviewed by Oliver Hunt.


<rdar://problem/7009870> After <audio> movie finishes playing, Pause button fails to change
back to Play button

Grab onto the current time when QTKit "ended" notification fires and use it as
duration from that point on as QuickTime sometimes refuses to play all the way to
the time it reports for duration and buffered. HTMLMediaElement assumes that there
more to play in this situation and tries to restart playback, endlessly.

  • platform/graphics/mac/MediaPlayerPrivateQTKit.h:

Rename m_duration to m_reportedDuration. Declare m_cachedDuration.

  • platform/graphics/mac/MediaPlayerPrivateQTKit.mm: (WebCore::MediaPlayerPrivate::MediaPlayerPrivate):

Rename m_duration to m_reportedDuration. Initialize m_cachedDuration.

(WebCore::MediaPlayerPrivate::duration):

Return m_cachedDuration once it has been set.

(WebCore::MediaPlayerPrivate::updateStates):

Rename m_duration to m_reportedDuration.

(WebCore::MediaPlayerPrivate::didEnd):

Set m_cachedDuration to currentTime().

15:29 JSC edited by bfulgham@webkit.org
(diff)
15:01 Changeset [45452] by cfleizach@apple.com

Bug 26900: AX: Manual spell check with Command-; does not bring up suggestions
https://bugs.webkit.org/show_bug.cgi?id=26900

14:41 Changeset [45451] by mrowe@apple.com

Change how WebKitNightlyEnabler.dylib lies about Safari's identity to LaunchServices.

This makes WebKit.app behave as expected when set as the default web browser on a wider
range of OS versions.

Reviewed by Simon Fraser.

  • WebKitLauncher/WebKitNightlyEnabler.m:

(poseAsWebKitApp):
(enableWebKitNightlyBehaviour):

14:41 Changeset [45450] by mrowe@apple.com

Shuffle some code around in preparation for a change to how poseAsWebKitApp works.

Reviewed by Simon Fraser.

  • WebKitLauncher/WebKitNightlyEnabler.m:

(systemVersion):
(webKitLauncherBundle):
(insideSafari4OnTigerTrampoline):

14:19 Changeset [45449] by weinig@apple.com

2009-07-01 Sam Weinig <sam@webkit.org>

Reviewed by Geoffrey Garen.

Convert op_div to load op1 and op2 up front.

  • jit/JITArithmetic.cpp: (JSC::JIT::emit_op_div):
14:06 Changeset [45448] by levin@chromium.org

2009-07-01 David Levin <levin@chromium.org>

Reviewed by Eric Seidel.

Remove unused code in SVGTransformList and SVGTransformDistance
<https://bugs.webkit.org/show_bug.cgi?id=26891>

  • svg/SVGTransformDistance.cpp: (WebCore::SVGTransformDistance::SVGTransformDistance):
  • svg/SVGTransformList.cpp:
  • svg/SVGTransformList.h:
14:02 Changeset [45447] by weinig@apple.com

2009-07-01 Sam Weinig <sam@webkit.org>

Reviewed by Geoffrey Garen.

Don't emit code in emitBinaryDoubleOp if code is unreachable, observable
via an empty (unlinked) jumplist passed in. This only effects op_jnless
and op_jnlesseq at present.

  • jit/JITArithmetic.cpp: (JSC::JIT::emitSlow_op_jnless): (JSC::JIT::emitSlow_op_jnlesseq): (JSC::JIT::emitBinaryDoubleOp):
13:59 Changeset [45446] by hyatt@apple.com

2009-07-01 David Hyatt <hyatt@apple.com>

Reviewed by Anders Carlsson.

Begin stubbing out functions for layout, painting and style/geometry caching on columns. Not enough
is implemented for this to do much of anything yet, but it should help keep patch sizes down. :)

  • html/DataGridColumn.cpp: (WebCore::DataGridColumn::columnChanged):
  • html/DataGridColumn.h: (WebCore::DataGridColumn::setId): (WebCore::DataGridColumn::setLabel): (WebCore::DataGridColumn::setType): (WebCore::DataGridColumn::setSortable): (WebCore::DataGridColumn::setSortDirection): (WebCore::DataGridColumn::setColumnList): (WebCore::DataGridColumn::style): (WebCore::DataGridColumn::setStyle): (WebCore::DataGridColumn::rect): (WebCore::DataGridColumn::setRect):
  • html/DataGridColumnList.cpp: (WebCore::DataGridColumnList::DataGridColumnList): (WebCore::DataGridColumnList::setDataGridNeedsLayout): (WebCore::DataGridColumnList::add): (WebCore::DataGridColumnList::remove): (WebCore::DataGridColumnList::move): (WebCore::DataGridColumnList::clear): (WebCore::DataGridColumnList::primaryColumnChanged):
  • html/DataGridColumnList.h: (WebCore::DataGridColumnList::create): (WebCore::DataGridColumnList::dataGrid): (WebCore::DataGridColumnList::clearDataGrid):
  • html/HTMLDataGridColElement.cpp: (WebCore::HTMLDataGridColElement::HTMLDataGridColElement): (WebCore::HTMLDataGridColElement::findDataGridAncestor): (WebCore::HTMLDataGridColElement::insertedIntoTree): (WebCore::HTMLDataGridColElement::removedFromTree):
  • html/HTMLDataGridColElement.h: (WebCore::HTMLDataGridColElement::dataGrid):
  • html/HTMLDataGridElement.cpp: (WebCore::HTMLDataGridElement::HTMLDataGridElement): (WebCore::HTMLDataGridElement::~HTMLDataGridElement):
  • rendering/RenderDataGrid.cpp: (WebCore::RenderDataGrid::styleDidChange): (WebCore::RenderDataGrid::recalcStyleForColumns): (WebCore::RenderDataGrid::recalcStyleForColumn): (WebCore::RenderDataGrid::styleForColumn): (WebCore::RenderDataGrid::layout): (WebCore::RenderDataGrid::layoutColumns): (WebCore::RenderDataGrid::paintColumnHeaders): (WebCore::RenderDataGrid::paintColumnHeader):
  • rendering/RenderDataGrid.h: (WebCore::RenderDataGrid::renderName): (WebCore::RenderDataGrid::canHaveChildren):
13:41 JSC created by bfulgham@webkit.org
Initial entry to document the utility.
13:38 Changeset [45445] by abarth@webkit.org

2009-07-01 Daniel Bates <dbates@intudata.com>

Reviewed by Adam Barth.

https://bugs.webkit.org/show_bug.cgi?id=26807


Fixes this address by checking whether frame->document()->decoder() is null.

  • page/XSSAuditor.cpp: (WebCore::XSSAuditor::findInRequest):
12:01 Changeset [45444] by zimmermann@webkit.org

2009-07-01 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

Fixes: https://bugs.webkit.org/show_bug.cgi?id=26885

Correctly reset history length to '0', not '1'. Old relict from early WML days.
Add some new WML testcases covering the use of history length (by onenterforward event handling)

Tests: wml/onenterforward-event.html

wml/onenterforward-inline-event.html
wml/ontimer-event.html


  • wml/WMLPageState.cpp: (WebCore::WMLPageState::reset):

2009-07-01 Nikolas Zimmermann <nikolas.zimmermann@torchmobile.com>

Reviewed by George Staikos.

Fixes: https://bugs.webkit.org/show_bug.cgi?id=26885

Add new WML testcases covering onenterforward/ontimer support.

  • wml/onenterforward-event-expected.txt: Added.
  • wml/onenterforward-event.html: Added.
  • wml/onenterforward-inline-event-expected.txt: Added.
  • wml/onenterforward-inline-event.html: Added.
  • wml/ontimer-event-expected.txt: Added.
  • wml/ontimer-event.html: Added.
  • wml/resources/onenterforward-event.js: Added. (setupTestDocument): (prepareTest): (executeTest):
  • wml/resources/onenterforward-event.wml: Added.
  • wml/resources/onenterforward-inline-event.js: Added. (setupTestDocument): (prepareTest): (executeTest):
  • wml/resources/onenterforward-inline-event.wml: Added.
  • wml/resources/ontimer-event.js: Added. (setupTestDocument): (prepareTest): (executeTest):
  • wml/resources/ontimer-event.wml: Added.
  • wml/resources/ontimer-inline-event.js:
11:41 Changeset [45443] by hyatt@apple.com

WebCore:

2009-07-01 David Hyatt <hyatt@apple.com>

Reviewed by Simon Fraser.

Make sure setting attributes on dcol elements properly updates the corresponding DataGridColumn object.

Added new test in fast/dom/HTMLDataGridElement.

  • html/HTMLDataGridColElement.cpp: (WebCore::HTMLDataGridColElement::parseMappedAttribute):
  • html/HTMLDataGridColElement.h:

LayoutTests:

2009-07-01 David Hyatt <hyatt@apple.com>

Reviewed by Simon Fraser.

Add layout test for being able to set attributes on dcol elements.


  • fast/dom/HTMLDataGridElement/DataGridColumns-dom-attributes-expected.txt: Added.
  • fast/dom/HTMLDataGridElement/DataGridColumns-dom-attributes.html: Added.
11:19 Changeset [45442] by simon.fraser@apple.com

2009-07-01 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dimitri Glazkov.

Add some missing pixel results for compositing tests.

  • platform/mac/compositing/layers-inside-overflow-scroll-expected.checksum: Added.
  • platform/mac/compositing/layers-inside-overflow-scroll-expected.png: Added.
  • platform/mac/compositing/overflow/overflow-scroll-expected.checksum: Added.
  • platform/mac/compositing/overflow/overflow-scroll-expected.png: Added.
11:12 Changeset [45441] by ggaren@apple.com

2009-07-01 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.

Converted op_mod to put { tag, payload } in { regT1, regT0 }, and
tidied up its constant case.


SunSpider reports a 0.2% regression, but a micro-benchmark of op_mod
shows a 12% speedup, and the SunSpider test that uses op_mod most should
benefit a lot from result caching in the end, since it almost always
performs (expression) % constant.

  • jit/JITArithmetic.cpp: (JSC::JIT::emit_op_mod): (JSC::JIT::emitSlow_op_mod):
11:11 Drosera edited by bfulgham@webkit.org
Add link to Web Inspector page. (diff)
09:53 Changeset [45440] by hyatt@apple.com

WebCore:

2009-07-01 David Hyatt <hyatt@apple.com>

Reviewed by Tim Hatcher.

<rdar://problem/6998524> REGRESSION (r44474): Form text field has focus ring, looks focused,
even though the field is not actually focused for keyboard input


Add the concept of whether or not the Page is focused by adding a boolean to the focusController. This allows the
focused frame and focused node to both be cached and changed programmatically even when the Page itself doesn't
actually happen to have focus at that time.

  • WebCore.base.exp:
  • page/FocusController.cpp: (WebCore::FocusController::FocusController): (WebCore::FocusController::setFocusedFrame): (WebCore::FocusController::setFocused): (WebCore::FocusController::setActive):
  • page/FocusController.h: (WebCore::FocusController::isFocused):

WebKit/mac:

2009-07-01 David Hyatt <hyatt@apple.com>

Reviewed by Tim Hatcher.

<rdar://problem/6998524> REGRESSION (r44474): Form text field has focus ring, looks focused,
even though the field is not actually focused for keyboard input


Add the concept of whether or not the Page is focused by adding a boolean to the focusController. This allows the
focused frame and focused node to both be cached and changed programmatically without causing errors when the Page doesn't
have focus.

  • WebView/WebHTMLView.mm: (-[WebHTMLView becomeFirstResponder]): (-[WebHTMLView resignFirstResponder]):

WebKit/win:

2009-07-01 David Hyatt <hyatt@apple.com>

Reviewed by Tim Hatcher.

<rdar://problem/6998524> REGRESSION (r44474): Form text field has focus ring, looks focused,
even though the field is not actually focused for keyboard input


Add the concept of whether or not the Page is focused by adding a boolean to the focusController. This allows the
focused frame and focused node to both be cached and changed programmatically without causing errors when the Page doesn't
have focus.

  • WebView.cpp: (WebViewWndProc):
09:47 Changeset [45439] by zimmermann@webkit.org

Reviewed by George Staikos.

Fixes: https://bugs.webkit.org/show_bug.cgi?id=26884

WMLTestCase.js relies on timing-dependant hacks, making it difficult to predict results on slower machines.
Fix by rewriting and simplifying the testing procedure, which contained unnecessary quirks working around older WML bugs.

09:06 Changeset [45438] by eric.carlson@apple.com

2009-07-01 Eric Carlson <eric.carlson@apple.com>

Reviewed by Simon Fraser.

Update WebKitSystemInterface for <rdar://problem/7014990>

  • libWebKitSystemInterfaceLeopard.a:
  • libWebKitSystemInterfaceSnowLeopard.a:
  • libWebKitSystemInterfaceTiger.a:
08:44 Changeset [45437] by zimmermann@webkit.org

Not reviewed. Ignore *.pyc files to not clutter up 'svn st'.

08:10 Changeset [45436] by zimmermann@webkit.org

Reviewed by Holger Freyther.
Apply same cleanup to WML testcases living in http/tests/wml/. Forgot about them in my last commit.

08:03 Changeset [45435] by zimmermann@webkit.org

Reviewed by Holger Freyther.

Slightly cleanup some WML testcases:
Use "anchorElement.textContent = '...'" instead of "anchorElement.appendChild(document.createTextNode('...'))"

05:55 Changeset [45434] by hausmann@webkit.org

WebCore:

2009-07-01 Jakub Wieczorek <faw217@gmail.com>

Reviewed by Simon Hausmann.

[Qt] Move some API headers from WebCore.pro to headers.pri so that they
get installed when running make install from the build directory.

  • WebCore.pro:

WebKit/qt:

2009-07-01 Jakub Wieczorek <faw217@gmail.com>

Reviewed by Simon Hausmann.

[Qt] Move some API headers from WebCore.pro to headers.pri so that they
get installed when running make install from the build directory.

  • Api/headers.pri:
04:36 Changeset [45433] by hausmann@webkit.org

2009-07-01 Balazs Kelemen <kelemen.balazs.3@stud.u-szeged.hu>

Reviewed by Simon Hausmann.

Fixed robotized QtLauncher to work when there is no index.html in the user's home.

  • QtLauncher/main.cpp: (main):
02:40 Changeset [45432] by hausmann@webkit.org

2009-07-01 Simon Hausmann <simon.hausmann@nokia.com>

Rubber-stamped by Ariya Hidayat.

Ran WebKitTools/Scripts/generate-qt-inspector-resource to update the
qrc file with new png files from the web inspector.

  • inspector/front-end/WebKit.qrc:

06/30/09:

23:51 Changeset [45431] by xan@webkit.org

2009-06-30 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

https://bugs.webkit.org/show_bug.cgi?id=26877
[GTK] DRT fixes to pass new plugin test

Silence compiler warnings.

Newer GCC don't like using or returning a static string where a
'char*' is expected. Ideally we'd change the function signatures
to return or take 'const char*' I suppose, but since we can't do
that just cast the strings to 'char*'.

  • DumpRenderTree/gtk/TestNetscapePlugin/TestNetscapePlugin.cpp: (webkit_test_plugin_get_value):
23:42 Changeset [45430] by xan@webkit.org

2009-06-30 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

https://bugs.webkit.org/show_bug.cgi?id=26877
[GTK] DRT fixes to pass new plugin test

Use the common pluginLog function instead of a local copy.

  • DumpRenderTree/gtk/TestNetscapePlugin/TestNetscapePlugin.cpp: (webkit_test_plugin_new_instance): (webkit_test_plugin_destroy_instance): (webkit_test_plugin_set_window): (webkit_test_plugin_handle_event):
23:41 Changeset [45429] by xan@webkit.org

2009-06-30 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

https://bugs.webkit.org/show_bug.cgi?id=26877
[GTK] DRT fixes to pass new plugin test

Modify our DRT as done in bug #15457 to log logSrc as src in order
to pass test plugins/netscape-plugin-map-data-to-src.html.

  • DumpRenderTree/gtk/TestNetscapePlugin/TestNetscapePlugin.cpp: (webkit_test_plugin_new_instance):
23:10 Changeset [45428] by weinig@apple.com

2009-06-30 Sam Weinig <sam@webkit.org>

Reviewed by Geoffrey Garen.

Converted some more arithmetic ops to put { tag, payload } in
{ regT1, regT0 }.

  • jit/JITArithmetic.cpp: (JSC::JIT::emit_op_mul): (JSC::JIT::emitSlow_op_mul):
23:07 Changeset [45427] by adele@apple.com

2009-06-30 Adele Peterson <adele@apple.com>

Reviewed by Dan Bernstein.

Fix for <rdar://problem/7014389> REGRESSION(4-TOT): Hyperlinks have no tooltips in Mail unless you first click in the message body

Updated these new methods (added as part of the viewless WebKit effort) that call through to WebHTMLView to use _selectedOrMainFrame,
instead of just selectedFrame. When the selection changes, I don't think there is a situation where there's no selected frame, but
it seems better to keep these two uses the same, since if there is a case, this will be more like the old behavior, before there was
a version of _selectionChanged in WebView.

  • WebView/WebView.mm: (-[WebView _setToolTip:]): (-[WebView _selectionChanged]):
22:58 Changeset [45426] by xan@webkit.org

2009-06-30 Xan Lopez <xlopez@igalia.com>

Disable test using eventSender.

  • platform/gtk/Skipped:
22:44 Changeset [45425] by xan@webkit.org

2009-06-30 Xan Lopez <xlopez@igalia.com>

Disable new tests until we enable the accessibility controller.

  • platform/gtk/Skipped:
22:28 Changeset [45424] by ggaren@apple.com

2009-06-30 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.

Converted some more arithmetic ops to put { tag, payload } in
{ regT1, regT0 }, and added a case for subtract constant.


SunSpider says no change. v8 says 0.3% slower.

  • jit/JIT.h:
  • jit/JITArithmetic.cpp: (JSC::JIT::emit_op_add): (JSC::JIT::emitAdd32Constant): (JSC::JIT::emitSlow_op_add): (JSC::JIT::emit_op_sub): (JSC::JIT::emitSub32Constant): (JSC::JIT::emitSlow_op_sub):
22:19 Changeset [45423] by mrowe@apple.com

Land some code that has a hope of compiling.

  • platform/graphics/cg/ImageSourceCG.cpp:

(WebCore::imageSourceOptions):

21:55 Changeset [45422] by barraclough@apple.com

2009-06-30 Gavin Barraclough <barraclough@apple.com>

Reviewed by Sam Weinig.

Remove more uses of addressFor(), load double constants directly from
the constantpool in the CodeBlock, rather than from the register file.

  • jit/JITArithmetic.cpp: (JSC::JIT::emitAdd32Constant): (JSC::JIT::emitBinaryDoubleOp):
21:37 Changeset [45421] by slewis@apple.com

2009-06-30 Stephanie Lewis <slewis@apple.com>

Reviewed by Simon Fraser.

Flip back on block caching. Throwing away block data, while
a perceived memory win, is a performance hit in cases where we
are repainting large images (i.e. backgrounds) frequently
(i.e. a flash video playing on top).
<rdar://problem/6933515> REGRESSION(L-SL): Large JPEG images are
decoded on drawing, slowing down painting of sites that frequently
repaint e.g. because of Flash (pandora.com, Starcraft 2)

  • platform/graphics/cg/ImageSourceCG.cpp: (WebCore::imageSourceOptions):
20:50 Changeset [45420] by ggaren@apple.com

JavaScriptCore:

2009-06-30 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.


Fixed a bug in postfix ops, where we would treat x = x++ and x = x--
as a no-op, even if x were not an int, and the ++/-- could have side-effects.

  • jit/JITArithmetic.cpp: (JSC::JIT::emit_op_post_inc): (JSC::JIT::emitSlow_op_post_inc): (JSC::JIT::emit_op_post_dec): (JSC::JIT::emitSlow_op_post_dec):

LayoutTests:

2009-06-30 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.


Added a test for x=x++ and x=x-- where the ++ / -- have side-effects.

  • fast/js/postfix-syntax-expected.txt:
  • fast/js/resources/postfix-syntax.js: (o.valueOf):
20:24 Changeset [45419] by mjs@apple.com

2009-06-30 Maciej Stachowiak <mjs@apple.com>

Reviewed by Sam Weinig.

  • DerivedSources.make:
  • GNUmakefile.am:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • css/CSSStyleSelector.cpp: (WebCore::loadFullDefaultStyle):
  • css/html.css: Copied from css/html4.css.
  • css/html4.css: Removed.
  • css/themeWin.css:
  • platform/Theme.h:
  • platform/qt/RenderThemeQt.cpp: (WebCore::RenderThemeQt::adjustMenuListButtonStyle):
  • rendering/RenderTheme.h:
19:53 Changeset [45418] by jmalonzo@webkit.org

Fix ChangeLog of r45415

19:40 Changeset [45417] by ggaren@apple.com

Fixed a ChangeLog merge fail.

19:40 Changeset [45416] by ggaren@apple.com

2009-06-30 Gavin Barraclough <barraclough@apple.com>

Reviewed by Sam Weinig.

Start removing cases of addressFor().

  • jit/JIT.h:
  • jit/JITArithmetic.cpp: (JSC::JIT::emitAdd32Constant): (JSC::JIT::emitBinaryDoubleOp): (JSC::JIT::emit_op_div):
  • jit/JITInlineMethods.h: (JSC::JIT::emitLoadDouble): (JSC::JIT::emitLoadInt32ToDouble): (JSC::JIT::emitStoreDouble):
  • jit/JITOpcodes.cpp: (JSC::JIT::emit_op_jfalse): (JSC::JIT::emit_op_jtrue):
19:27 Changeset [45415] by jmalonzo@webkit.org

2009-06-30 Jan Michael Alonzo <jmalonzo@webkit.org>

Reviewed by Jan Alonzo.

[GTK] GtkLauncher depends on gtk+2-2.15 because of gtk_orientable_set_orientation()
https://bugs.webkit.org/show_bug.cgi?id=26835

  • GtkLauncher/main.c: (create_toolbar):
19:16 Changeset [45414] by jmalonzo@webkit.org

2009-06-30 Zan Dobersek <zandobersek@gmail.com>

Reviewed by Gustavo Noronha.

[GTK] Drag and drop support
https://bugs.webkit.org/show_bug.cgi?id=23642

Define DragImageRef as GdkPixbuf and implement essential
functions which manipulate drag images.

  • platform/DragImage.h:
  • platform/gtk/DragImageGtk.cpp: (WebCore::dragImageSize): (WebCore::deleteDragImage): (WebCore::scaleDragImage): (WebCore::createDragImageFromImage):
19:15 Changeset [45413] by jmalonzo@webkit.org

2009-06-30 Maxime Simon <simon.maxime@gmail.com>

Reviewed by Eric Seidel.

StorageNamespace.cpp build issue
https://bugs.webkit.org/show_bug.cgi?id=26859

Moved #endif ENABLE(DOM_STORAGE) position
so that the build doesn't fail.
The namespace WebCore was started after the #if ENABLE(DOM_STORAGE)
but closed after the #endif.

  • storage/StorageNamespace.cpp:
18:01 Changeset [45412] by barraclough@apple.com

2009-06-30 Gavin Barraclough <barraclough@apple.com>

Reviewed by Sam Weinig.

Start removing cases of addressFor().

  • jit/JIT.h:
  • jit/JITArithmetic.cpp: (JSC::JIT::emitAdd32Constant): (JSC::JIT::emitBinaryDoubleOp): (JSC::JIT::emit_op_div):
  • jit/JITInlineMethods.h: (JSC::JIT::emitLoadDouble): (JSC::JIT::emitLoadInt32ToDouble): (JSC::JIT::emitStoreDouble):
  • jit/JITOpcodes.cpp: (JSC::JIT::emit_op_jfalse): (JSC::JIT::emit_op_jtrue):
17:40 Changeset [45411] by simon.fraser@apple.com

2009-06-30 Simon Fraser <simon.fraser@apple.com>

Rubber-stamped by Dan Bernstein.

Update compositing pixel results now that DRT dumps them correctly.

  • compositing/overflow/overflow-scroll.html: Fixed typo in this testcase.
  • platform/mac/compositing/overflow/ancestor-overflow-expected.checksum:
  • platform/mac/compositing/overflow/ancestor-overflow-expected.png:
  • platform/mac/compositing/overflow/overflow-positioning-expected.checksum:
  • platform/mac/compositing/overflow/overflow-positioning-expected.png:
  • platform/mac/compositing/overflow/overflow-scroll-expected.txt:
  • platform/mac/compositing/overflow/parent-overflow-expected.checksum:
  • platform/mac/compositing/overflow/parent-overflow-expected.png:
  • platform/mac/compositing/repaint/content-into-overflow-expected.checksum:
  • platform/mac/compositing/repaint/content-into-overflow-expected.png:
  • platform/mac/compositing/repaint/overflow-into-content-expected.checksum:
  • platform/mac/compositing/repaint/overflow-into-content-expected.png:
17:40 Changeset [45410] by simon.fraser@apple.com

2009-06-30 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dan Bernstein.

Fix drawing of view background in pixel snapshots when using compositing.
We still need to call -displayIfNeeded to ensure that the NSView drawing
has happened.

  • DumpRenderTree/mac/PixelDumpSupportMac.mm: (createBitmapContextFromWebView):
16:29 Changeset [45409] by ggaren@apple.com

2009-06-30 Geoffrey Garen <ggaren@apple.com>

Rolled back in my last patch with regression fixed.

  • jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases):
  • jit/JIT.h:
  • jit/JITOpcodes.cpp: (JSC::JIT::emit_op_loop_if_less): (JSC::JIT::emit_op_loop_if_lesseq): (JSC::JIT::emit_op_resolve_global): (JSC::JIT::emitSlow_op_resolve_global): (JSC::JIT::emit_op_eq): (JSC::JIT::emitSlow_op_eq): (JSC::JIT::emit_op_neq): (JSC::JIT::emitSlow_op_neq):
16:09 Changeset [45408] by eric@webkit.org

2009-06-30 Eric Seidel <eric@webkit.org>

No review, reverting r45396. (George Staikos agrees the change was wrong.)
http://trac.webkit.org/changeset/45396

  • fast/css/getComputedStyle/computed-style-font-family-expected.txt:
16:01 Changeset [45407] by eric.carlson@apple.com

2009-06-30 Eric Carlson <eric.carlson@apple.com>

Reviewed by Oliver Hunt.

<rdar://problem/7020825> HTMLMediaElement should not assume seeking is only possible in
buffered time ranges

  • html/HTMLMediaElement.cpp:

Don't ASSERT if the network state goes to Idle when when the ready state is
HaveEnoughData or higher, it is perfectly legal.

(WebCore::HTMLMediaElement::setReadyState):

Don't clear m_seeking after posting a 'seeking' event.

  • platform/graphics/mac/MediaPlayerPrivateQTKit.mm: (WebCore::MediaPlayerPrivate::seek):

Do nothing when asked to seek to the same time we are already seeking to. Check requested
seek time against maxTimeSeekable, not maxTimeLoaded.

(WebCore::MediaPlayerPrivate::doSeek):

Don't bother stopping a movie that isn't playing. Minor cleanup to make the code more readable.

(WebCore::MediaPlayerPrivate::seekTimerFired):

Check requested seek time against maxTimeSeekable, not maxTimeLoaded.

(WebCore::MediaPlayerPrivate::updateStates):

A streaming movie has as much data as it needs once it reaches "playable", so
set the ready state to HaveFutureData. A movie with metadata doesn't drop back
to "have nothing" when seeking. A streaming movie doesn't use the network when paused.

(WebCore::MediaPlayerPrivate::timeChanged):

It may not be possible to seek to a specific time in a streamed movie but when seeking
in a streaming movie QuickTime sets the time to closest time possible and posts a
timechanged notification, so update m_seekTo so we can detect when the seek completes.

15:43 Changeset [45406] by ggaren@apple.com

2009-06-30 Geoffrey Garen <ggaren@apple.com>

Rolled out my last patch because it was a 2% SunSpider regression.

  • jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases):
  • jit/JIT.h:
  • jit/JITOpcodes.cpp: (JSC::JIT::emit_op_loop_if_less): (JSC::JIT::emit_op_loop_if_lesseq): (JSC::JIT::emit_op_resolve_global): (JSC::JIT::emit_op_eq): (JSC::JIT::emitSlow_op_eq): (JSC::JIT::emit_op_neq): (JSC::JIT::emitSlow_op_neq):
15:43 Changeset [45405] by eric@webkit.org

2009-06-30 Eric Seidel <eric@webkit.org>

Typo fix only, no review.

Fix typo causing failure during land-diff

  • Scripts/bugzilla-tool:
15:27 Changeset [45404] by ggaren@apple.com

2009-06-30 Geoffrey Garen <ggaren@apple.com>

Reviewed by Gavin "Sam Weinig" Barraclough.


Standardized the rest of our opcodes to put { tag, payload } in
{ regT1, regT0 } where possible.

  • jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases):
  • jit/JIT.h:
  • jit/JITOpcodes.cpp: (JSC::JIT::emit_op_loop_if_less): (JSC::JIT::emit_op_loop_if_lesseq): (JSC::JIT::emit_op_resolve_global): (JSC::JIT::emitSlow_op_resolve_global): (JSC::JIT::emit_op_eq): (JSC::JIT::emitSlow_op_eq): (JSC::JIT::emit_op_neq): (JSC::JIT::emitSlow_op_neq):
15:06 Changeset [45403] by sfalken@apple.com

https://bugs.webkit.org/show_bug.cgi?id=26860

Reviewed by Sam Weinig.

Blacklist Yahoo Application State plug-in for versions prior to 1.0.0.6.
Earlier versions cause corruption crashes.

  • plugins/win/PluginPackageWin.cpp: (WebCore::PluginPackage::isPluginBlacklisted):
14:57 Changeset [45402] by dino@apple.com

2009-06-30 Dean Jackson <dino@apple.com>

Reviewed by Simon Fraser.

<rdar://problem/7005207> <video> in canvas broken with ACCEL_COMPOSITING

Fix drawImage of video in CanvasRenderingContext2D when
accelerated compositing is enabled. This is done by
adding a new paint method to HTMLVideoElement that
is only called from canvas, which will create the software
renderer if necessary, but otherwise calls the regular
paint method. Meanwhile, add logic to the software video
renderer so that it can be created without calling repaint
every time it sees a new frame.

  • html/CanvasRenderingContext2D.cpp: (WebCore::CanvasRenderingContext2D::drawImage):
  • html/HTMLVideoElement.cpp: (WebCore::HTMLVideoElement::paintCurrentFrameInContext):
  • html/HTMLVideoElement.h:
  • platform/graphics/MediaPlayer.cpp: (WebCore::MediaPlayer::paintCurrentFrameInContext):
  • platform/graphics/MediaPlayer.h:
  • platform/graphics/MediaPlayerPrivate.h: (WebCore::MediaPlayerPrivateInterface::paintCurrentFrameInContext):
  • platform/graphics/mac/MediaPlayerPrivateQTKit.h:
  • platform/graphics/mac/MediaPlayerPrivateQTKit.mm: (WebCore::MediaPlayerPrivate::createQTMovie): (WebCore::MediaPlayerPrivate::createQTVideoRenderer): (WebCore::MediaPlayerPrivate::currentRenderingMode): (WebCore::MediaPlayerPrivate::setUpVideoRendering): (WebCore::MediaPlayerPrivate::tearDownVideoRendering): (WebCore::MediaPlayerPrivate::paintCurrentFrameInContext):
14:47 Changeset [45401] by eric@webkit.org

2009-06-30 Simon Fraser <simon.fraser@apple.com>

Reviewed by Mark Rowe.

createBitmapContextFromWebView() incorrectly used #ifdef BUILDING_ON_LEOPARD,
and thus failed on SnowLeopard.

  • DumpRenderTree/mac/PixelDumpSupportMac.mm: (createBitmapContextFromWebView):
14:39 Changeset [45400] by barraclough@apple.com

2009-06-30 Gavin Barraclough <barraclough@apple.com>

Reviewed by Geoffrey Garen.

Replace calls to store32(tagFor()) and store32(payloadFor())
with emitStoreInt32(), emitStoreBool(), and emitStoreCell().

  • jit/JIT.h:
  • jit/JITArithmetic.cpp: (JSC::JIT::emit_op_negate): (JSC::JIT::emit_op_lshift): (JSC::JIT::emit_op_rshift): (JSC::JIT::emit_op_bitand): (JSC::JIT::emitBitAnd32Constant): (JSC::JIT::emit_op_bitor): (JSC::JIT::emitBitOr32Constant): (JSC::JIT::emit_op_bitxor): (JSC::JIT::emitBitXor32Constant): (JSC::JIT::emit_op_bitnot): (JSC::JIT::emit_op_post_inc): (JSC::JIT::emit_op_post_dec): (JSC::JIT::emit_op_pre_inc): (JSC::JIT::emit_op_pre_dec): (JSC::JIT::emit_op_add): (JSC::JIT::emitAdd32Constant): (JSC::JIT::emit_op_sub): (JSC::JIT::emitSub32ConstantLeft): (JSC::JIT::emitSub32ConstantRight): (JSC::JIT::emit_op_mul): (JSC::JIT::emitSlow_op_mul): (JSC::JIT::emit_op_div): (JSC::JIT::emit_op_mod):
  • jit/JITCall.cpp: (JSC::JIT::emit_op_load_varargs):
  • jit/JITInlineMethods.h: (JSC::JIT::emitStoreInt32): (JSC::JIT::emitStoreCell): (JSC::JIT::emitStoreBool): (JSC::JIT::emitStore):
  • jit/JITOpcodes.cpp: (JSC::JIT::emit_op_instanceof): (JSC::JIT::emit_op_not): (JSC::JIT::emit_op_eq): (JSC::JIT::emitSlow_op_eq): (JSC::JIT::emit_op_neq): (JSC::JIT::emitSlow_op_neq): (JSC::JIT::compileOpStrictEq): (JSC::JIT::emit_op_eq_null): (JSC::JIT::emit_op_neq_null):
  • jit/JITStubCall.h: (JSC::JITStubCall::call):
14:38 Changeset [45399] by hyatt@apple.com

Remove conflict markers from ChangeLog.

14:37 Changeset [45398] by simon.fraser@apple.com

2009-06-30 Simon Fraser <simon.fraser@apple.com>

Reviewed by Mark Rowe.

createBitmapContextFromWebView() incorrectly used #ifdef BUILDING_ON_LEOPARD,
and thus failed on SnowLeopard.

  • DumpRenderTree/mac/PixelDumpSupportMac.mm: (createBitmapContextFromWebView):
14:37 Changeset [45397] by hyatt@apple.com

WebCore:

<<<<<<< .mine

LayoutTests:

2009-06-30 David Hyatt <hyatt@apple.com>

Reviewed by Beth Dakin.

Add layout test for https://bugs.webkit.org/show_bug.cgi?id=26858.

  • fast/dom/HTMLDataGridElement/DataGridColumns-dom-expected.txt: Added.
  • fast/dom/HTMLDataGridElement/DataGridColumns-dom.html: Added.
14:27 Changeset [45396] by hausmann@webkit.org

2009-06-30 Antonio Gomes <antonio.gomes@openbossa.org>

Reviewed by Simon Hausmann.

Updated LayoutTests/fast/css/getComputedStyle/computed-style-font-family-expected.txt
according to the default font size value set in WebKit/qt/Api/qwebsettings.cpp (in
this case '14').

  • fast/css/getComputedStyle/computed-style-font-family-expected.txt:
14:27 Changeset [45395] by hausmann@webkit.org

2009-06-30 Antonio Gomes <antonio.gomes@openbossa.org>

Reviewed by Simon Hausmann.

Updated LayoutTests/platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt
to some newly cross-platform CSS properties.

  • platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
14:07 Changeset [45394] by levin@chromium.org

2009-06-30 Jeremy Orlow (jorlow@chromium.org)

Reviewed by David Levin.

https://bugs.webkit.org/show_bug.cgi?id=26852

Fix some minor build issues in the v8 bindings.

  • bindings/v8/custom/V8StorageCustom.cpp: (WebCore::V8Custom::v8StorageNamedPropertyEnumerator): (WebCore::storageGetter): (WebCore::storageSetter): (WebCore::storageDeleter):
  • storage/StorageArea.cpp:
13:16 Changeset [45393] by mrowe@apple.com

Versioning.

13:16 Changeset [45392] by mrowe@apple.com

New tag.

13:10 Changeset [45391] by agl@chromium.org

2009-06-30 Adam Langley <agl@google.com>

Reviewed by Eric Seidel.

Chromium Linux: use different fonts for each script run.

https://bugs.webkit.org/show_bug.cgi?id=26853

Previously, when rendering complex text, we picked a single font which
could render all the glyphs needed for the run. However, this meant
that sometimes lines were rendered with, for example, [LATIN, THAI,
LATIN] and we could end up with a different font for the Latin parts
than for lines without Thai in them.

With this patch, we pick a font for each script run.

This change is covered by existing layout tests.

13:02 Changeset [45390] by mrowe@apple.com

Fix up the ChangeLog.

13:02 Changeset [45389] by levin@chromium.org

2009-06-30 David Levin <levin@chromium.org>

Reviewed by Oliver Hunt.

Make media/video-document-types pass again. It broke when the
call to waitUntilDone was accidently made asynchronous in r45378.

  • media/video-document-types.html:
12:57 Changeset [45388] by mrowe@apple.com

Merge r45386.

12:53 Changeset [45387] by simon.fraser@apple.com

2009-06-30 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dave Hyatt, Dan Bernstein.

<rdar://problem/6191676> Redraw issues scrolling overflow:scroll div with compositing

There are two parts to this fix. First, if a compositing layer has overflow clipping
or a mask, then the compositing layer does not need to be sized to encompass all
non-composited children.

Second, when scrolling, we have to back up to the compositing ancestor and have
it reposition descendant compositing layers, because overflow doesn't follow
the z-order tree.

Test: compositing/overflow/overflow-scroll.html

  • page/FrameView.cpp: (WebCore::FrameView::layout): Pass the UpdateCompositingLayers flag when updating layers after layout, so that we can reposition compositing layers if we're not about to do a layer rebuild.
  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::updateLayerPositions): Only update compositing layers if the flag is set.

(WebCore::RenderLayer::scrollToOffset):
Do compositing layer updates from the compositing ancestor in one fell swoop.

  • rendering/RenderLayer.h: Replace the unused FullUpdate flag with one that allows us to specify whether compositing layers should be updated.
  • rendering/RenderLayerBacking.cpp: (WebCore::RenderLayerBacking::updateAfterLayout):
  • rendering/RenderLayerBacking.h: New param that we can pass to updateCompositingDescendantGeometry to control whether the update goes deep.
  • rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::calculateCompositedBounds): We can return earlier if this is not a self-painting layer. If the layer has overflow clip or a mask, then the composited bounds are just the local bounds, excluding descendants.

(WebCore::RenderLayerCompositor::updateCompositingDescendantGeometry):

  • rendering/RenderLayerCompositor.h: Renamed from updateCompositingChildrenGeometry, and added a flag that allows us to do a deep update.
12:52 Changeset [45386] by mrowe@apple.com

<rdar://problem/7006959> 'Save as…’ does not work in Mail

Reviewed by Timothy Hatcher.

  • WebView/WebFrameView.mm:

(-[WebFrameView initWithFrame:]): Call _needMailThreadWorkaroundIfCalledOffMainThread rather than
needMailThreadWorkaround, as the latter is intended to be used at the point which a workaround would
be applied and thus contains a main thread check. Since -initWithFrame: is called on the main thread,
this was causing us to not switch from exception-throwing to logging for the thread violation behavior.

11:27 Changeset [45385] by ap@webkit.org

Reviewed by Dan Bernstein.

https://bugs.webkit.org/show_bug.cgi?id=26764
Uncaught NOT_FOUND_ERR: DOMException 8 loading empty text file

Test: fast/parser/empty-text-resource.html

  • loader/TextDocument.cpp: (WebCore::TextTokenizer::finish): Create document tree if it hasn't been created yet.
11:23 Changeset [45384] by bweinstein@apple.com

2009-06-30 Brian Weinstein <bweinstein@apple.com>

Reviewed by Adam Roben.

Renamed scrollbarUnderPoint to scrollbarAtPoint.

  • page/EventHandler.cpp: (WebCore::EventHandler::hitTestResultAtPoint): (WebCore::EventHandler::handleMousePressEvent): (WebCore::EventHandler::handleMouseMoveEvent):
  • platform/ScrollView.cpp: (WebCore::ScrollView::scrollbarAtPoint):
  • platform/ScrollView.h:
  • platform/chromium/PopupMenuChromium.cpp: (WebCore::PopupListBox::handleMouseDownEvent): (WebCore::PopupListBox::handleMouseMoveEvent):

2009-06-30 Brian Weinstein <bweinstein@apple.com>

Reviewed by Adam Roben.


Renamed scrollbarUnderPoint to scrollbarAtPoint to follow conventions.

  • Api/qwebpage.cpp: (QWebPage::swallowContextMenuEvent):
11:05 Changeset [45383] by oliver@apple.com

<rdar://problem/6960286> CrashTracer: [REGRESSION] 1120 crashes in Safari at com.apple.WebCore • WebCore::JSLazyEventListener::parseCode const + 62

Reviewed by Geoff Garen

<select> elements and other elements the produce popups can keep their popups
live across a page load. If this occurs and the <select> element has an
onchange handler it is possible to get into a state where we try to execute
JS on a page that no longer has an execution context.

10:42 Changeset [45382] by ggaren@apple.com

2009-06-30 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.


Standardized the rest of the property access instructions to put { tag,
payload } in { regT1, regT0 }.

Small v8 speedup, 0.2% SunSpider slowdown.

  • jit/JIT.h:
  • jit/JITInlineMethods.h: (JSC::JIT::emitLoad): (JSC::JIT::emitLoad2):
  • jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_get_by_val): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::emit_op_put_by_val): (JSC::JIT::emitSlow_op_put_by_val): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::emitSlow_op_put_by_id): (JSC::JIT::patchPutByIdReplace):
10:32 Changeset [45381] by levin@chromium.org

WebCore:

2009-06-30 Drew Wilson <atwilson@google.com>

Reviewed by David Levin.

<https://bugs.webkit.org/show_bug.cgi?id=25435>

Added support for sending MessagePorts to/from Workers via postMessage().

  • bindings/js/JSWorkerContextCustom.cpp: (WebCore::JSWorkerContext::messageChannel): Exposes MessageChannel constructor WorkerContext.
  • bindings/js/JSWorkerCustom.cpp:
  • bindings/v8/custom/V8MessageChannelConstructor.cpp: (WebCore::CALLBACK_FUNC_DECL): Now correctly handles being instantiated from within a Worker.
  • platform/CrossThreadCopier.h: (WebCore::GenericWorkerTaskTraits<PassOwnPtr>): Added support for passing PassOwnPtr as argument to GenericWorkerTask
  • workers/GenericWorkerTask.h: (WebCore::):
    • workers/Worker.cpp: (WebCore::Worker::postMessage): Now handles disentangling/entangling MessagePorts sent to/from via postMessage. (WebCore::Worker::dispatchMessage):
    • workers/Worker.h:
    • workers/Worker.idl: Added an optional MessagePort argument to postMessage()
    • workers/WorkerContext.cpp: (WebCore::WorkerContext::hasPendingActivity): Workers now report pending activity whenever there are remotely entangled ports. (WebCore::WorkerContext::postMessage): Now handles disentangling/entangling MessagePorts sent to/from via postMessage. (WebCore::WorkerContext::dispatchMessage):
    • workers/WorkerContext.h:
    • workers/WorkerContext.idl: Added an optional MessagePort argument to postMessage()
    • workers/WorkerContextProxy.h:
    • workers/WorkerMessagingProxy.cpp: Updated messaging infrastructure to transport an optional MessagePort/MessagePortChannel. (WebCore::MessageWorkerContextTask::create): (WebCore::MessageWorkerContextTask::MessageWorkerContextTask): (WebCore::MessageWorkerContextTask::performTask): (WebCore::MessageWorkerTask::create): (WebCore::MessageWorkerTask::MessageWorkerTask): (WebCore::MessageWorkerTask::performTask): (WebCore::WorkerMessagingProxy::postMessageToWorkerObject): (WebCore::WorkerMessagingProxy::postMessageToWorkerContext):
    • workers/WorkerMessagingProxy.h: Added additional postMessage() API that accepts a MessagePort
    • workers/WorkerObjectProxy.h: Added additional postMessage() API that accepts a MessagePort

LayoutTests:

2009-06-30 Drew Wilson <atwilson@google.com>

Reviewed by David Levin.

<https://bugs.webkit.org/show_bug.cgi?id=25435>

Added tests for sending MessagePorts to/from workers.

  • fast/workers/resources/worker-cloneport.js: Added. (onmessage):
  • fast/workers/resources/worker-messageport.js: Added. (onmessage): (pingBack):
  • fast/workers/worker-cloneport-expected.txt: Added.
  • fast/workers/worker-cloneport.html-disabled: Added. Tests for cloning a port while messages are sent from another thread.
  • fast/workers/worker-messageport-expected.txt: Added.
  • fast/workers/worker-messageport.html-disabled: Added. General tests for sending messages to workers via message ports.
  • fast/workers/worker-messageport-gc-expected.txt: Added.
  • fast/workers/worker-messageport-gc.html-disabled: Added. Tests to ensure Workers don't shut down when reachable only via MessagePorts.
10:07 Changeset [45380] by treat@webkit.org

2009-06-30 Joe Ligman <joseph.ligman@nokia.com>

Reviewed by Adam Treat.

Bug 26422: [Qt] QWebPagePrivate::frameAt calculates wrong frame

Added a public method QWebPage::frameAt
Removed QWebPagePrivate::frameAt, which calcuated the wrong frame
Modified QWebPage::swallowContextMenuEvent to use the new frameAt method
New test case for frameAt added to tst_qwebpage.cpp

09:57 Changeset [45379] by agl@chromium.org

2009-06-30 Adam Langley <agl@google.com>

Reviewed by Eric Seidel.

Chromium: Add complex text support on Linux.

https://bugs.webkit.org/show_bug.cgi?id=25068

This patch adds complex text support on Linux using Harfbuzz. It's not
the fastest code possible: some caching of font tables will certainly
be required. However, it's probably the simplest code that works.

This will require checking in new baselines in the Chromium tree for
those layout tests which now pass.

09:30 Changeset [45378] by eric.carlson@apple.com

2009-06-30 Eric Carlson <eric.carlson@apple.com>

Reviewed by Simon Fraser.

<rdar://problem/7019799> Slow loading MediaDocument can fall back to plug-in unnecessarily

  • platform/graphics/mac/MediaPlayerPrivateQTKit.mm: (WebCore::MediaPlayerPrivate::updateStates):

Always fall back to a plug-in if m_hasUnsupportedTracks is set.

(WebCore::MediaPlayerPrivate::disableUnsupportedTracks):

Set m_hasUnsupportedTracks if we disable a track.

2009-06-30 Eric Carlson <eric.carlson@apple.com>

Reviewed by Simon Fraser.


<rdar://problem/7019799> Slow loading MediaDocument can fall back to plug-in unnecessarily


Don't test iframe immediately, switch over from <video> to <embed> may not have
happend when the document 'load' event fires.

  • media/video-document-types.html:
09:25 Changeset [45377] by japhet@chromium.org

2009-06-30 Nate Chapin <japhet@chromium.org>

Reviewed by Dimitri Glazkov.

Fix inverted if/else that's causing a layout test in src.chromium.org to fail.

https://bugs.webkit.org/show_bug.cgi?id=26820

08:49 Changeset [45376] by mitz@apple.com

WebCore:

Reviewed by Dave Hyatt.

  • fix <rdar://problem/6946611> REGRESSION (r30673): Shade10:" D_Snap to Grid" window is clipping

Added an app-specific quirk to revert the behavior of <link> elements
to the way it was before r30673, namely that if the rel attribute
is not "stylesheet" but the type attribute contains "text/css", the
link behaves as a stylesheet link.

  • WebCore.base.exp: Exported setTreatsAnyTextCSSLinkAsStylesheet().
  • html/HTMLLinkElement.cpp: (WebCore::HTMLLinkElement::process): Check

Settings::treatsAnyTextCSSLinkAsStylesheet() and if true, allow a
style sheet link if the type contains "text/css".

  • page/Settings.cpp: (WebCore::Settings::Settings):

Initialize m_treatsAnyTextCSSLinkAsStylesheet.

(WebCore::Settings::setTreatsAnyTextCSSLinkAsStylesheet): Added.

  • page/Settings.h: (WebCore::Settings::treatsAnyTextCSSLinkAsStylesheet): Added.

WebKit/mac:

Reviewed by Dave Hyatt.

  • fix <rdar://problem/6946611> REGRESSION (r30673): Shade10:" D_Snap to Grid" window is clipping
  • Misc/WebKitVersionChecks.h: Added

WEBKIT_FIRST_VERSION_WITHOUT_SHADE10_QUIRK

  • WebView/WebView.mm: (-[WebView _needsLinkElementTextCSSQuirk]): Added. Returns YES if the

client app is a version of Shade 10 earlier than 10.6 and it was
linked against a version of WebKit earlier than 531.2.

(-[WebView _preferencesChangedNotification:]): Added a call to

Settings::setTreatsAnyTextCSSLinkAsStylesheet().

07:14 Changeset [45375] by hausmann@webkit.org

2009-06-30 Jakub Wieczorek <faw217@gmail.com>

Reviewed by Simon Hausmann.

Add QWebFrame::baseUrl() function that exposes the base URL of a frame.
Autotests included.

  • Api/qwebframe.cpp: (QWebFrame::baseUrl):
  • Api/qwebframe.h:
  • tests/qwebframe/tst_qwebframe.cpp:
07:05 Changeset [45374] by staikos@webkit.org

2009-06-30 Kwang Yul Seo <skyul@company100.net>

Reviewed by Eric Seidel.

Override operator new/delete with const std::nothrow_t& as the second
argument.

06:54 QtWebKitContrib edited by treat@kde.org
(diff)
06:51 QtWebKitContrib edited by treat@kde.org
(diff)
05:15 Changeset [45373] by xan@webkit.org

2009-06-30 Adrien Nader <camaradetux@gmail.com>

Reviewed by Xan Lopez.

Fix typo in GNUMakefile.am: it reads WebCOre instead of WebCore.

  • GNUmakefile.am:
05:10 Changeset [45372] by jmalonzo@webkit.org

2009-06-30 Jan Michael Alonzo <jmalonzo@webkit.org>

Reviewed by Xan Lopez.

[Gtk] Turn on ENABLE_DATAGRID so we can run the layout tests.

  • GNUmakefile.am:
03:55 Changeset [45371] by xan@webkit.org

2009-06-30 Xan Lopez <xlopez@igalia.com>

Skip new editing tests.

They are failing because we don't have a dumpEditingCallbacks
implementation.

  • platform/gtk/Skipped:
02:55 Changeset [45370] by ddkilzer@apple.com

2009-06-30 Gabor Loki <loki@inf.u-szeged.hu>

Reviewed by Sam Weinig.

<https://bugs.webkit.org/show_bug.cgi?id=24986>

Remove unnecessary references to AssemblerBuffer.

  • interpreter/Interpreter.cpp:
  • interpreter/Interpreter.h:
02:33 Changeset [45369] by eric@webkit.org

2009-06-30 Eric Seidel <eric@webkit.org>

Reviewed by Maciej Stachowiak.

Include a built copy of BeautifulSoup 3.1.0.1 so that
bugzilla-tool users do not need to install it manually.
https://bugs.webkit.org/show_bug.cgi?id=26833

BeautifulSoup is BSD-licensed (thus WebKit compatible) and available from:
http://www.crummy.com/software/BeautifulSoup

  • Scripts/bugzilla-tool: fixed a typo in a log message.
  • Scripts/modules/BeautifulSoup.py: Added.
  • Scripts/modules/bugzilla.py:
02:03 Changeset [45368] by eric@webkit.org

2009-06-30 Eric Seidel <eric@webkit.org>

Reviewed by Simon Hausmann.

Fix our create_patch handling to not
hang when creating large patches.
https://bugs.webkit.org/show_bug.cgi?id=26834

We're now using StringIO to make an in-memory buffer for the patch
instead of piping from the patch command directly to the upload.
The previous hang was caused by calling wait() when the process had already
filed the stdout buffer.

  • Scripts/modules/scm.py:
00:52 Changeset [45367] by hausmann@webkit.org

2009-06-30 Raju Kunnath <raju.kunnath@nokia.com>

Reviewed by Simon Hausmann.

<https://bugs.webkit.org/show_bug.cgi?id=26752>

[Qt] Windows release build issue with Qt4.5 due to -GL flag on msvc2005 and msvc2008.

  • WebCore.pro: Removed -GL compiler option for win32-msvc2005|win32-msvc2008.
00:48 Changeset [45366] by xan@webkit.org

2009-06-30 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

Initialize member variables in the right order. Fixes compiler
warning.

  • platform/image-decoders/bmp/BMPImageReader.cpp: (WebCore::BMPImageReader::BMPImageReader):
00:24 Changeset [45365] by xan@webkit.org

2009-06-29 Xan Lopez <xlopez@igalia.com>

Reviewed by David Levin.

https://bugs.webkit.org/show_bug.cgi?id=26831
Fix compiler warning in WorkerLoaderProxy.h

Forward declaration of a class within a class is not allowed,
since the compiler can't know if the declaration is legal at that
point. This gives the following compiler warning:

../../WebCore/workers/WorkerLoaderProxy.h:40: warning: declaration
"class WebCore::ScriptExecutionContext::Task" does not declare
anything

Include the full ScriptExecutionHeader header instead.

  • workers/WorkerLoaderProxy.h:

06/29/09:

23:43 Changeset [45364] by xan@webkit.org

2009-06-29 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

Fix a couple of compiler warnings.

  • webkit/webkitwebnavigationaction.cpp:
  • webkit/webkitwebview.cpp: (webkit_web_view_real_move_cursor):
22:56 Changeset [45363] by xan@webkit.org

2009-06-29 Xan Lopez <xlopez@igalia.com>

Reviewed by Jan Alonzo.

Remove unused function.

  • plugins/gtk/PluginPackageGtk.cpp:
22:12 Changeset [45362] by simon.fraser@apple.com

2009-06-29 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dan Bernstein.

Change RenderLayer::updateLayerPositions() to use a bitmask instead of two
boolean arguments. FullUpdate is unused at present, but will be used soon.

  • page/FrameView.cpp: (WebCore::FrameView::layout):
  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::updateLayerPositions): (WebCore::RenderLayer::scrollToOffset):
  • rendering/RenderLayer.h: (WebCore::RenderLayer::):
20:08 Changeset [45361] by weinig@apple.com

2009-06-29 Sam Weinig <sam@webkit.org>

Reviewed by Gavin Barraclough.

Various cleanups.

  • Use fpRegT* instead of X86::xmm*.
  • Use a switch statement in emitBinaryDoubleOp instead of a bunch of if/elses.
  • jit/JITArithmetic.cpp: (JSC::JIT::emitAdd32Constant): (JSC::JIT::emitBinaryDoubleOp): (JSC::JIT::emit_op_div):
17:09 Changeset [45360] by levin@chromium.org

2009-06-29 David Levin <levin@chromium.org>

Reviewed by Oliver Hunt.

Still seeing occasional leaks from UString::sharedBuffer code
https://bugs.webkit.org/show_bug.cgi?id=26420

The problem is that the pointer to the memory allocation isn't visible
by "leaks" due to the lower bits being used as flags. The fix is to
make the pointer visible in memory (in debug only). The downside of
this fix that the memory allocated by sharedBuffer will still look like
a leak in non-debug builds when any flags are set.

  • wtf/PtrAndFlags.h: (WTF::PtrAndFlags::set):
16:44 Changeset [45359] by japhet@chromium.org

2009-06-29 Nate Chapin <japhet@chromium.org>

Reviewed by Dimitri Glazkov.

Fix one more bad merge in V8Proxy.

https://bugs.webkit.org/show_bug.cgi?id=26819

16:21 Changeset [45358] by hyatt@apple.com

2009-06-29 David Hyatt <hyatt@apple.com>

Fix Windows bustage. Missed two configurations for enabling DATAGRID.

  • WebKit.vcproj/WebKit.vcproj:
15:35 Changeset [45357] by weinig@apple.com

JavaScriptCore:

2009-06-29 Sam Weinig <sam@webkit.org>

Reviewed by Mark Rowe.

Remove more unused scons support.

  • SConstruct: Removed.

WebCore:

2009-06-29 Sam Weinig <sam@webkit.org>

Reviewed by Mark Rowe.

Remove more unused scons support.

  • SConstruct: Removed.

WebKitTools:

2009-06-29 Sam Weinig <sam@webkit.org>

Reviewed by Mark Rowe.

Remove more unused scons support.

  • Scripts/build-jsc:
  • Scripts/build-webkit:
  • Scripts/webkitdirs.pm:
15:33 Changeset [45356] by oliver@apple.com

<rdar://problem/7016214> JSON.parse fails to parse valid JSON with most Unicode characters
<https://bugs.webkit.org/show_bug.cgi?id=26802>

Reviewed by Gavin Barraclough.

In the original JSON.parse patch unicode was handled correctly, however in some last
minute "clean up" I oversimplified isSafeStringCharacter. This patch corrects this bug.

14:38 Changeset [45355] by weinig@apple.com

2009-06-29 Sam Weinig <sam@webkit.org>

Reviewed by Dave Hyatt.

Fix some incorrect create functions.

  • html/DataGridColumn.h: (WebCore::DataGridColumn::create):
  • html/DataGridColumnList.h: (WebCore::DataGridColumnList::create):
14:32 Changeset [45354] by hausmann@webkit.org

2009-06-29 Antonio Gomes <antonio.gomes@openbossa.org>

Reviewed by Simon Hausmann.

Updated LayoutTests/platform/qt/fast/css/getComputedStyle/computed-style-expected.txt to some newly cross-platform CSS properties.

The file has not been updated for a while.

14:31 Changeset [45353] by weinig@apple.com

WebCore:

2009-06-29 Sam Weinig <sam@webkit.org>

Reviewed by Dave Hyatt.

Remove initialize method from DataGridDataSource and add
DOMDataGridDataSource.

  • GNUmakefile.am:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • bindings/js/JSDataGridDataSource.cpp:
  • bindings/js/JSDataGridDataSource.h:
  • bindings/js/JSHTMLDataGridElementCustom.cpp: (WebCore::JSHTMLDataGridElement::setDataSource):
  • html/DOMDataGridDataSource.cpp: Added. (WebCore::DOMDataGridDataSource::DOMDataGridDataSource): (WebCore::DOMDataGridDataSource::~DOMDataGridDataSource):
  • html/DOMDataGridDataSource.h: Added. (WebCore::DOMDataGridDataSource::create): (WebCore::DOMDataGridDataSource::isDOMDataGridDataSource): (WebCore::asDOMDataGridDataSource):
  • html/DataGridDataSource.h: (WebCore::DataGridDataSource::isDOMDataGridDataSource): (WebCore::DataGridDataSource::isJSDataGridDataSource):
  • html/HTMLDataGridElement.cpp: (WebCore::HTMLDataGridElement::HTMLDataGridElement): (WebCore::HTMLDataGridElement::setDataSource): (WebCore::HTMLDataGridElement::dataSource):
  • html/HTMLDataGridElement.h:

LayoutTests:

2009-06-29 Sam Weinig <sam@webkit.org>

Reviewed by Dave Hyatt.

Update for the change in DataGridDataSource API.

  • fast/dom/HTMLDataGridElement/DataGridDataSource-basic-expected.txt:
  • fast/dom/HTMLDataGridElement/DataGridDataSource-basic.html:
13:58 Changeset [45352] by japhet@chromium.org

2009-06-29 Nate Chapin <japhet@chromium.org>

Reviewed by Dimitri Glazkov.

Fix a couple of bad merge items from my previous V8Proxy patch.

https://bugs.webkit.org/show_bug.cgi?id=26813

13:42 Changeset [45351] by eric@webkit.org

2009-06-26 Eric Seidel <eric@webkit.org>

Reviewed by David Levin.

Rename land-and-update to land-diff and make it awesome.
https://bugs.webkit.org/show_bug.cgi?id=26734

Renamed land-and-update to land-diff
Made the BUGID argument optional.
Added optional support for building and testing before landing.

land-diff [options] [BUGID] Lands the current working directory diff and updates the bug if provided.

Options:

-r REVIEWER, --reviewer=REVIEWER

Update ChangeLogs to say Reviewed by REVIEWER.

--no-close Leave bug open after landing.
--no-build Commit without building first, implies --no-test.
--no-test Commit without running run-webkit-tests.

  • Scripts/bugzilla-tool:
13:41 Changeset [45350] by cmarrin@apple.com

https://bugs.webkit.org/show_bug.cgi?id=26706

Fixed crash due to dereference of m_toStyle

13:36 Changeset [45349] by alice.liu@apple.com

build fix. accidental omission of end paren

13:22 Changeset [45348] by eric.carlson@apple.com

2009-06-29 Eric Carlson <eric.carlson@apple.com>

Reviewed by Simon Fraser.

<rdar://problem/7014813> Ask media engine if a movie is streamed or downloaded.

  • WebCore.base.exp:

Export _wkQTMovieGetType

  • platform/graphics/MediaPlayer.cpp: (WebCore::MediaPlayer::hasSingleSecurityOrigin):

Move to keep with other non-callback functions.

(WebCore::MediaPlayer::movieLoadType):

New, returns the movie type.

  • platform/graphics/MediaPlayer.h:

Declare MovieLoadType enum and movieLoadType method.

  • platform/graphics/MediaPlayerPrivate.h: (WebCore::MediaPlayerPrivateInterface::movieLoadType):

Default implementation of movieLoadType

  • platform/graphics/mac/MediaPlayerPrivateQTKit.h:
  • platform/graphics/mac/MediaPlayerPrivateQTKit.mm: (WebCore::MediaPlayerPrivate::createQTMovie):

Don't base m_isStreaming on protocol, there are other types of streaming movies.

(WebCore::MediaPlayerPrivate::maxTimeBuffered):

Don't assume all streams are unbuffered.

(WebCore::MediaPlayerPrivate::updateStates):

Update m_isStreaming once we have metadata.

(WebCore::MediaPlayerPrivate::disableUnsupportedTracks):

Tracks that are disabled to begin with shouldn't be included in m_enabledTrackCount.

(WebCore::MediaPlayerPrivate::movieLoadType):

New, return movie type.

  • platform/mac/WebCoreSystemInterface.h:
  • platform/mac/WebCoreSystemInterface.mm:

Add wkQTMovieGetType.

2009-06-29 Eric Carlson <eric.carlson@apple.com>

Reviewed by Simon Fraser.

<rdar://problem/7014813> Ask media engine if a movie is streamed or downloaded.

  • WebCoreSupport/WebSystemInterface.m: (InitWebCoreSystemInterface):

Add QTMovieGetType.

2009-06-29 Eric Carlson <eric.carlson@apple.com>

Reviewed by Simon Fraser.

Update WebKitSystemInterface for <rdar://problem/7014813>

  • WebKitSystemInterface.h:
  • libWebKitSystemInterfaceLeopard.a:
  • libWebKitSystemInterfaceSnowLeopard.a:
  • libWebKitSystemInterfaceTiger.a:
13:15 Changeset [45347] by alice.liu@apple.com

2009-06-29 Alice Liu <alice.liu@apple.com>

Fixed <rdar://problem/6930280> Reproducible crash at USA Today photo gallery

Reviewed by Anders Carlsson.

No test added because the crash requires a flash plugin

  • plugins/win/PluginMessageThrottlerWin.cpp: Rely on the hWnd of the plugin to tell us whether the PluginView has been deleted during execution of its wndProc. (WebCore::PluginMessageThrottlerWin::messageThrottleTimerFired):
13:02 Changeset [45346] by dimich@chromium.org

2009-06-29 Dmitry Titov <dimich@chromium.org>

Reviewed by David Levin.

https://bugs.webkit.org/show_bug.cgi?id=26811

[Chromium] Remove a flag and functions used to enable workers in runtime.

  • bindings/v8/WorkerContextExecutionProxy.cpp:
  • bindings/v8/WorkerContextExecutionProxy.h:
13:00 Changeset [45345] by simon.fraser@apple.com

2009-06-29 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dave Hyatt.

<rdar://problem/6976712> Text antialiasing problems when rendering into compositing layers.

Improve the appearance of text in compositing layers by making use of
CA's layer geometry flipping, rather that doing it ourselves with a flip
transform. This allows CG to use font autohinting in the layer text.

  • platform/graphics/GraphicsLayer.h: (WebCore::GraphicsLayer::setGeometryOrientation): (WebCore::GraphicsLayer::geometryOrientation): New methods to set whether this layer uses flipped geometry.
  • platform/graphics/mac/GraphicsLayerCA.h:
  • platform/graphics/mac/GraphicsLayerCA.mm: (WebCore::GraphicsLayerCA::setGeometryOrientation): (WebCore::GraphicsLayerCA::geometryOrientation): Subclass in order to call into CA

(WebCore::GraphicsLayerCA::setContentsLayer):
Now that CA is doing the geometry flipping, we no longer need to flip the content
layers for image and video manually

  • rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::ensureRootPlatformLayer): Turn on flipping on the root layer.
12:52 Changeset [45344] by hyatt@apple.com

2009-06-29 David Hyatt <hyatt@apple.com>

Reviewed by Mark Rowe.

Update layout tests for addition of HTMLDataGridCellElement constructor.

  • fast/dom/Window/window-properties-expected.txt:
  • fast/js/global-constructors-expected.txt:
12:36 Changeset [45343] by zimmermann@webkit.org

Reviewed by Adam Treat.
Rename go-task-animation.* to ontimer-inline-event.* to reflect what it really tests.

12:11 Changeset [45342] by zimmermann@webkit.org

Reviewed by Adam Treat.

Fixes: https://bugs.webkit.org/show_bug.cgi?id=26246

Implement most WML specific <select> element features.
Add 'iname' / 'ivalue' support and support variable references.

Add two tests covering most select/variable related functionality.
Needs more tests when adding full 'iname' / 'ivalue' support (used in conjuction with onpick).

Tests: http/tests/wml/post-data-to-server.html

wml/select-element-variables.html

11:11 Changeset [45341] by weinig@apple.com

2009-06-29 Sam Weinig <sam@webkit.org>

Reviewed by Geoffrey Garen.

Add inline code dealing with doubles for op_jfalse and op_jtrue.

  • assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::): (JSC::MacroAssemblerX86Common::zeroDouble):
  • jit/JITOpcodes.cpp: (JSC::JIT::emit_op_jfalse): (JSC::JIT::emit_op_jtrue):
09:34 Changeset [45340] by jberlin@apple.com

2009-06-29 Jessie Berlin <jberlin@apple.com>

Fix the windows build.

Reviewed by Adam Roben.

  • WebCoreSupport/WebContextMenuClient.cpp: (WebContextMenuClient::isSpeaking):
  • WebCoreSupport/WebContextMenuClient.h:
09:03 Changeset [45339] by hyatt@apple.com

2009-06-29 David Hyatt <hyatt@apple.com>

Reviewed by Adam Roben.

Make sure to patch all configurations to have datagrid enabled.

  • WebKit.vcproj/WebKit.vcproj:
08:52 Changeset [45338] by hyatt@apple.com

WebCore:

2009-06-29 David Hyatt <hyatt@apple.com>

Reviewed by Adam Roben.

Put <datagrid> behind an ifdef.

  • Configurations/FeatureDefines.xcconfig:
  • bindings/js/JSDataGridColumnListCustom.cpp:
  • bindings/js/JSDataGridDataSource.cpp:
  • bindings/js/JSDataGridDataSource.h:
  • bindings/js/JSHTMLDataGridElementCustom.cpp:
  • html/DataGridColumn.cpp:
  • html/DataGridColumn.h:
  • html/DataGridColumn.idl:
  • html/DataGridColumnList.cpp:
  • html/DataGridColumnList.h:
  • html/DataGridColumnList.idl:
  • html/DataGridDataSource.h:
  • html/HTMLDataGridCellElement.cpp:
  • html/HTMLDataGridCellElement.h:
  • html/HTMLDataGridCellElement.idl:
  • html/HTMLDataGridColElement.cpp:
  • html/HTMLDataGridColElement.h:
  • html/HTMLDataGridColElement.idl:
  • html/HTMLDataGridElement.cpp:
  • html/HTMLDataGridElement.h:
  • html/HTMLDataGridElement.idl:
  • html/HTMLDataGridRowElement.cpp:
  • html/HTMLDataGridRowElement.h:
  • html/HTMLDataGridRowElement.idl:
  • html/HTMLTagNames.in:
  • page/DOMWindow.idl:
  • rendering/RenderDataGrid.cpp:
  • rendering/RenderDataGrid.h:

WebKit/mac:

2009-06-29 David Hyatt <hyatt@apple.com>

Reviewed by Adam Roben.

Put datagrid behind an #ifdef.

  • Configurations/FeatureDefines.xcconfig:

WebKit/win:

2009-06-29 David Hyatt <hyatt@apple.com>

Reviewed by Adam Roben.

Put <datagrid> behind an ifdef.

  • WebKit.vcproj/WebKit.vcproj:
06:57 Changeset [45337] by hausmann@webkit.org

2009-06-29 Simon Hausmann <simon.hausmann@nokia.com>

Reviewed by Ariya Hidayat and Adam Roben.

Fix compilation with MINGW. Ported MSVC inline assembly to
GNU inline assembly. Also fixed casting errors where gcc
refused to cast a pointer-to-function to a pointer-to-object,
without an intermediate cast to a non-pointer type.

02:17 Changeset [45336] by hausmann@webkit.org

2009-06-29 Simon Hausmann <simon.hausmann@nokia.com>

Fix the Qt build, add missing isSpeaking() implementation to
ContextMenuClient.

06/28/09:

22:24 Changeset [45335] by xan@webkit.org

2009-06-28 Xan Lopez <xlopez@igalia.com>

Fix the GTK+ build.

  • WebCoreSupport/ContextMenuClientGtk.cpp: (WebKit::ContextMenuClient::isSpeaking):
  • WebCoreSupport/ContextMenuClientGtk.h:
21:28 Changeset [45334] by ggaren@apple.com

2009-06-28 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.

Standardized op_get_by_id to put { tag, payload } in { regT1, regT0 }.


SunSpider and v8 report maybe 0.2%-0.4% regressions, but the optimization
this enables will win much more than that back.

  • jit/JIT.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines):
  • jit/JIT.h:
  • jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_method_check): (JSC::JIT::emit_op_get_by_id): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::compileGetByIdSlowCase): (JSC::JIT::patchGetByIdSelf): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain):
20:06 Changeset [45333] by sullivan@apple.com

WebCore:

2009-06-26 John Sullivan <sullivan@apple.com>

Added Speech submenu to context menu on Mac when there's a non-editable selection
(it was already present when there's an editable selection).
Also added support for disabling "Stop Speaking" when there is no speaking to stop.

Reviewed by Tim Hatcher.

  • loader/EmptyClients.h: (WebCore::EmptyContextMenuClient::isSpeaking): implemented this new virtual function to return false


  • page/ContextMenuClient.h: declared this new pure virtual function


  • platform/ContextMenu.cpp: (WebCore::ContextMenu::populate): insert Speech item (after a separator) on Mac when there's selected non-editable text (WebCore::ContextMenu::checkOrEnableIfNeeded): enable Stop Speaking item only if the context menu client returns true for isSpeaking()

WebKit/mac:

2009-06-26 John Sullivan <sullivan@apple.com>

Added support for disabling Stop Speaking when there is no speaking to stop.
Also fixed Stop Speaking so that it works (formerly it would throw an exception, and
selecting it from the menu would perform a web search instead (!)).

Reviewed by Tim Hatcher.

  • WebCoreSupport/WebContextMenuClient.h: declare isSpeaking()


  • WebCoreSupport/WebContextMenuClient.mm: (WebContextMenuClient::isSpeaking): implement by calling through to NSApplication (WebContextMenuClient::stopSpeaking): fixed implementation by correcting the signature of the NSApplication method
19:54 Changeset [45332] by mitz@apple.com

WebCore:

Reviewed by Simon Fraser.

Test: added a case to fast/dom/Element/scrollWidth.html

Ensure that scroll{Width, Height} is greater or equal to
client{Width, Height}.

  • rendering/RenderBox.cpp: (WebCore::RenderBox::scrollWidth): (WebCore::RenderBox::scrollHeight):

LayoutTests:

Reviewed by Simon Fraser.

  • fast/dom/Element/scrollWidth-expected.txt:
  • fast/dom/Element/scrollWidth.html:
15:06 Changeset [45331] by mrowe@apple.com

Versioning.

10:25 Changeset [45330] by simon.fraser@apple.com

2009-06-27 Simon Fraser <simon.fraser@apple.com>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=26780

Do not make compositing layers for non-self-painting RenderLayers,
since these exist only to push a clip onto the clipping stack. If such
a layer gets compositied for some other reason, it should not paint.

Also ensure that we update composited layer positions correctly
inside overflow:scroll layers. We can't assume that the contents
are child layers, so we have to go up to the compositing ancestor,
and tell it to update all its child layer positions as we do
after layout.

Tests: compositing/layers-inside-overflow-scroll.html

compositing/self-painting-layers.html

  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::scrollToOffset):
  • rendering/RenderLayerBacking.cpp: (WebCore::RenderLayerBacking::paintIntoLayer):
  • rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::calculateCompositedBounds): (WebCore::RenderLayerCompositor::needsToBeComposited):
05:40 Changeset [45329] by jmalonzo@webkit.org

2009-06-28 Sriram Yadavalli <sriram.yadavalli@nokia.com>

Reviewed by Eric Seidel.

[Qt] Fix build break for Qt
https://bugs.webkit.org/show_bug.cgi?id=26779

  • Api/qwebpage.cpp: (QWebPage::swallowContextMenuEvent):
04:58 Changeset [45328] by jmalonzo@webkit.org

2009-06-28 Jan Michael Alonzo <jmalonzo@webkit.org>

Gtk build fix - include stdarg.h for va_start/va_end.

  • DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.cpp:
04:15 Changeset [45327] by eric@webkit.org

2009-06-28 Luke Kenneth Casson Leighton <lkcl@lkcl.net>

Reviewed by Eric Seidel.

PurgeableBuffer #defines leave out functions on gtk MacOSX 10.4 build
https://bugs.webkit.org/show_bug.cgi?id=23057

Define these functions for Gtk as well.

  • platform/PurgeableBuffer.h:
04:14 Changeset [45326] by eric@webkit.org

2009-06-28 Nate Chapin <japhet@chromium.org>

Reviewed by Dimitri Glazkov.

Update CodeGeneratorV8.pm to match the new api for V8Proxy.

https://bugs.webkit.org/show_bug.cgi?id=26765

  • bindings/scripts/CodeGeneratorV8.pm: Match the current version of V8Proxy.
04:13 Changeset [45325] by eric@webkit.org

2009-06-28 Nicolas Sylvain <nsylvain@chromium.org>

Reviewed by Dimitri Glazkov.

If loading a font fails because of the sandbox, we ask the browser process to
try to load it by calling ensureFontLoaded. If it still fails after
ensureFontLoaded, we hit a ASSERT_NOT_REACHED.

This case happens once in a while during browser shutdown. The browser will
queue a message to the renderer to shutdown, and will then stop answering sync
messages from the renderer. If the renderer is still loading a page during this
time, it might try to call the browser process to ask to load a font. The
browser process will ignore the request, and the font will fail to load, even
after the second try.

This is unfortunate, but there is no real risk here, since the renderer will be
going away as soon as it processes another message.

This can't be layout tested as it depends on the sandbox.

https://bugs.webkit.org/show_bug.cgi?id=26743

  • platform/graphics/chromium/FontChromiumWin.cpp:
  • platform/graphics/chromium/FontPlatformDataChromiumWin.cpp:
  • platform/graphics/chromium/GlyphPageTreeNodeChromiumWin.cpp:
  • platform/graphics/chromium/SimpleFontDataChromiumWin.cpp:
03:31 Changeset [45324] by eric@webkit.org

2009-06-28 John Abd-El-Malek <jam@chromium.org>

https://bugs.webkit.org/show_bug.cgi?id=15457

Layout tests to ensure that "src" paramater is added when it's missing
but "data" is present.

  • plugins/netscape-plugin-map-data-to-src-expected.txt: Added.
  • plugins/netscape-plugin-map-data-to-src.html: Added.

2009-06-28 John Abd-El-Malek <jam@chromium.org>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=15457

Test: plugins/netscape-plugin-map-data-to-src.html

Fix problems with Real or WMP plugins not displaying because "data" was set
on the OBJECT tag instead of "src". This is based on what Firefox does, see
http://mxr.mozilla.org/mozilla-central/source/layout/generic/nsObjectFrame.cpp#3045

  • rendering/RenderPartObject.cpp: (WebCore::mapDataParamToSrc): (WebCore::RenderPartObject::updateWidget):

2009-06-28 John Abd-El-Malek <jam@chromium.org>

https://bugs.webkit.org/show_bug.cgi?id=15457

Modify NPAPI test plugin to verify that "src" paramater is added when
it's missing but "data" is present.

  • DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.cpp: (pluginLog):
  • DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.h:
  • DumpRenderTree/TestNetscapePlugIn.subproj/main.cpp: (NPP_New): (NPP_Destroy): (NPP_SetWindow): (handleEventCarbon): (handleEventCocoa):
  • DumpRenderTree/win/TestNetscapePlugin/main.cpp: (NPP_New):
03:19 Changeset [45323] by eric@webkit.org

2009-06-28 Eric Seidel <eric@webkit.org>

No review. Just updating test results Kevin seems to have missed.

Test failure started with:
http://trac.webkit.org/changeset/45287
The new results definitely look better.

  • fast/profiler/profile-with-no-title-expected.txt:
02:36 Changeset [45322] by jmalonzo@webkit.org

2009-06-28 Miloš Popović <gpopac@gmail.com>

Rubber-stamped by Jan Alonzo.

New Serbian translation.

06/27/09:

23:05 Changeset [45321] by simon.fraser@apple.com

2009-06-27 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dan Bernstein.

https://bugs.webkit.org/show_bug.cgi?id=26780

Fix up previous change. When computeCompositingRequirements() determines
that the current layer is composited, it needs to inform its parent
by setting compositingState.m_subtreeIsCompositing() to true. That didn't
always happen after the previous patch. Clarified the logic here.

  • rendering/RenderLayerCompositor.cpp: (WebCore::CompositingState::CompositingState): (WebCore::RenderLayerCompositor::computeCompositingRequirements):
22:54 Changeset [45320] by jmalonzo@webkit.org

2009-06-27 Laszlo Gombos <laszlo.1.gombos@nokia.com>

Reviewed by Jan Alonzo.

[Qt] Build fix after r45290
https://bugs.webkit.org/show_bug.cgi?id=26769

  • WebCore.pro:
22:54 Changeset [45319] by jmalonzo@webkit.org

2009-06-27 Emilio Pozuelo Monfort <pochu27@gmail.com>

Reviewed by Jan Alonzo.

[GTK] Don't use deprecated GTK+ symbols.
https://bugs.webkit.org/show_bug.cgi?id=26583

  • plugins/gtk/gtk2xtbin.c: (gtk_xtbin_class_init): (gtk_xtbin_new): (gtk_xtbin_destroy):
  • GtkLauncher/main.c: (create_toolbar):
21:38 Changeset [45318] by simon.fraser@apple.com

2009-06-27 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dan Bernstein.

https://bugs.webkit.org/show_bug.cgi?id=26780

First part: fix the RenderLayer::hasCompositingDescendant() flag to be set
correctly.

  • rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::computeCompositingRequirements): Do not unconditionally set compositingState.m_subtreeIsCompositing, because that can clobber the value from an earlier sibling. Add some more comments.

Remove a final use of Vector iterators.

(WebCore::RenderLayerCompositor::recursiveRepaintLayerRect):
Move the normalFlowList() processing outside the test for isStackingContext().

21:19 Changeset [45317] by mitz@apple.com

Reviewed by Simon Fraser and Antti Koivisto.

  • make paintFillLayerExtended() non-virtual and remove its clipY and clipH parameters

These parameters were computed and passed along to
paintFillLayerExtended in order to vertically constrain the rect fill to
the damage rect, because Qt cannot paint tall rectangles (see
<http://websvn.kde.org/?view=rev&revision=42721>). Since the damage rect
is passed along in the PaintInfo, the extra parameters are redundant,
and the intersection can just take place in paintFillLayerExtended().

  • rendering/InlineFlowBox.cpp: (WebCore::InlineFlowBox::paintFillLayers): (WebCore::InlineFlowBox::paintFillLayer): (WebCore::InlineFlowBox::paintBoxDecorations): (WebCore::InlineFlowBox::paintMask):
  • rendering/InlineFlowBox.h:
  • rendering/RenderBox.cpp: (WebCore::RenderBox::paintRootBoxDecorations): (WebCore::RenderBox::paintBoxDecorations): (WebCore::RenderBox::paintMask): (WebCore::RenderBox::paintMaskImages): (WebCore::RenderBox::paintFillLayers): (WebCore::RenderBox::paintFillLayer):
  • rendering/RenderBox.h:
  • rendering/RenderBoxModelObject.cpp: (WebCore::RenderBoxModelObject::paintFillLayerExtended):
  • rendering/RenderBoxModelObject.h:
  • rendering/RenderFieldset.cpp: (WebCore::RenderFieldset::paintBoxDecorations): (WebCore::RenderFieldset::paintMask):
  • rendering/RenderTable.cpp: (WebCore::RenderTable::paintBoxDecorations): (WebCore::RenderTable::paintMask):
  • rendering/RenderTableCell.cpp: (WebCore::RenderTableCell::paintBackgroundsBehindCell): (WebCore::RenderTableCell::paintMask):
20:13 Changeset [45316] by bfulgham@webkit.org

WebCore:

2009-06-27 Ryosuke Niwa <rniwa@google.com>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=26762

Clean up for IndentOutdentCommand::indentRegion, and solved most of problems related to the bug 21712.
https://bugs.webkit.org/show_bug.cgi?id=21712

Added few utility functions to htmlediting.h/cpp

isVisibilyAdjacent checks whether the first position is visibly next to the second position.
i.e. there is no visible node between the first and second positions

canMergeLists checks whether two lists can be merged.
It checks the type of list, the editing boundary, and adjacency of the lists.

Tests: editing/execCommand/indent-nested-lists-1.html

editing/execCommand/indent-nested-lists-2.html
editing/execCommand/indent-nested-lists-3.html
editing/execCommand/indent-nested-lists-4.html
editing/execCommand/indent-nested-lists-5.html
editing/execCommand/indent-nested-lists-6.html
editing/execCommand/indent-nested-lists-7.html
editing/execCommand/outdent-nested-lists-1.html
editing/execCommand/outdent-nested-lists-2.html
editing/execCommand/outdent-nested-lists-3.html
editing/execCommand/outdent-nested-lists-4.html

  • editing/IndentOutdentCommand.cpp: (WebCore::IndentOutdentCommand::prepareBlockquoteLevelForInsertion): (WebCore::IndentOutdentCommand::tryIndentingAsListItem): (WebCore::IndentOutdentCommand::indentIntoBlockquote): (WebCore::IndentOutdentCommand::indentRegion):
  • editing/IndentOutdentCommand.h:
  • editing/htmlediting.cpp: (WebCore::enclosingListChild): (WebCore::canMergeLists): (WebCore::isVisibilyAdjacent):
  • editing/htmlediting.h:

LayoutTests:

2009-06-27 Ryosuke Niwa <set EMAIL_ADDRESS environment variable>

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=26762

Clean up for IndentOutdentCommand::indentRegion, and solved most of problems related to the bug 21712.
https://bugs.webkit.org/show_bug.cgi?id=21712


The following test cases are intended for 21712, but included here
since this patch already solved a lot of problems.

  • editing/execCommand/indent-nested-lists-1-expected.txt: Added.
  • editing/execCommand/indent-nested-lists-1.html: Added.
  • editing/execCommand/indent-nested-lists-2-expected.txt: Added.
  • editing/execCommand/indent-nested-lists-2.html: Added.
  • editing/execCommand/indent-nested-lists-3-expected.txt: Added.
  • editing/execCommand/indent-nested-lists-3.html: Added.
  • editing/execCommand/indent-nested-lists-4-expected.txt: Added.
  • editing/execCommand/indent-nested-lists-4.html: Added.
  • editing/execCommand/indent-nested-lists-5-expected.txt: Added.
  • editing/execCommand/indent-nested-lists-5.html: Added.
  • editing/execCommand/indent-nested-lists-6-expected.txt: Added.
  • editing/execCommand/indent-nested-lists-6.html: Added.
  • editing/execCommand/indent-nested-lists-7-expected.txt: Added.
  • editing/execCommand/indent-nested-lists-7.html: Added.
  • editing/execCommand/outdent-nested-lists-1-expected.txt: Added.
  • editing/execCommand/outdent-nested-lists-1.html: Added.
  • editing/execCommand/outdent-nested-lists-2-expected.txt: Added.
  • editing/execCommand/outdent-nested-lists-2.html: Added.
  • editing/execCommand/outdent-nested-lists-3-expected.txt: Added.
  • editing/execCommand/outdent-nested-lists-3.html: Added.
  • editing/execCommand/outdent-nested-lists-4-expected.txt: Added.
  • editing/execCommand/outdent-nested-lists-4.html: Added.
19:31 Changeset [45315] by bfulgham@webkit.org

2009-06-27 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

WebInspector: Fix completion when iterating options using Tab.

https://bugs.webkit.org/show_bug.cgi?id=26722

  • inspector/front-end/TextPrompt.js: (WebInspector.TextPrompt.prototype._completionsReady):
17:32 Changeset [45314] by abarth@webkit.org

2009-06-27 Daniel Bates <dbates@intudata.com>

Reviewed by Adam Barth.


https://bugs.webkit.org/show_bug.cgi?id=26776


More tests for the XSSAuditor.

  • http/tests/security/xssAuditor/anchor-url-dom-write-location-expected.txt: Added.
  • http/tests/security/xssAuditor/anchor-url-dom-write-location-inline-event-expected.txt: Added.
  • http/tests/security/xssAuditor/anchor-url-dom-write-location-inline-event.html: Added.
  • http/tests/security/xssAuditor/anchor-url-dom-write-location-javascript-URL-expected.txt: Added.
  • http/tests/security/xssAuditor/anchor-url-dom-write-location-javascript-URL.html: Added.
  • http/tests/security/xssAuditor/anchor-url-dom-write-location.html: Added.
  • http/tests/security/xssAuditor/dom-write-URL-expected.txt: Added.
  • http/tests/security/xssAuditor/dom-write-URL.html: Added.
  • http/tests/security/xssAuditor/dom-write-innerHTML-expected.txt: Added.
  • http/tests/security/xssAuditor/dom-write-innerHTML.html: Added.
  • http/tests/security/xssAuditor/dom-write-location-expected.txt: Added.
  • http/tests/security/xssAuditor/dom-write-location-inline-event-expected.txt: Added.
  • http/tests/security/xssAuditor/dom-write-location-inline-event.html: Added.
  • http/tests/security/xssAuditor/dom-write-location-javascript-URL-expected.txt: Added.
  • http/tests/security/xssAuditor/dom-write-location-javascript-URL.html: Added.
  • http/tests/security/xssAuditor/dom-write-location.html: Added.
  • http/tests/security/xssAuditor/faux-script1-expected.txt: Added.
  • http/tests/security/xssAuditor/faux-script1.html: Added.
  • http/tests/security/xssAuditor/faux-script2-expected.txt: Added.
  • http/tests/security/xssAuditor/faux-script2.html: Added.
  • http/tests/security/xssAuditor/faux-script3-expected.txt: Added.
  • http/tests/security/xssAuditor/faux-script3.html: Added.
  • http/tests/security/xssAuditor/get-from-iframe-expected.txt: Added.
  • http/tests/security/xssAuditor/get-from-iframe.html: Added.
  • http/tests/security/xssAuditor/malformed-HTML-expected.txt: Added.
  • http/tests/security/xssAuditor/malformed-HTML.html: Added.
  • http/tests/security/xssAuditor/post-from-iframe-expected.txt: Added.
  • http/tests/security/xssAuditor/post-from-iframe.html: Added.
  • http/tests/security/xssAuditor/resources/anchor-url-dom-write-location-click.html: Added.
  • http/tests/security/xssAuditor/resources/echo-dom-write-URL.html: Added.
  • http/tests/security/xssAuditor/resources/echo-dom-write-innerHTML.html: Added.
  • http/tests/security/xssAuditor/resources/echo-dom-write-location.html: Added.
  • http/tests/security/xssAuditor/resources/utilities.js: Added. (sendRequestFromIFrame):
14:38 Changeset [45313] by kov@webkit.org

2009-06-27 Gustavo Noronha Silva <gns@gnome.org>

Reviewed by Holger Freyther.

https://bugs.webkit.org/show_bug.cgi?id=25889
[GTK] scrollbar policy for main frame is not implementable

Override visibleContentRect to handle GTK+'s case, in which
scrollbars or equivalent decoration are painted by the parent
widget.

  • platform/ScrollView.cpp:
  • platform/gtk/ScrollViewGtk.cpp: (WebCore::ScrollView::visibleContentRect):
12:36 Changeset [45312] by abarth@webkit.org

WebCore:

2009-06-27 Daniel Bates <dbates@intudata.com>

Reviewed by Adam Barth.

https://bugs.webkit.org/show_bug.cgi?id=26708


Fix addresses false negatives with respect to scheme relative paths, iFrame JavaScript URLs,
and UTF-7 encoded payloads.

Tests: http/tests/security/xssAuditor/http-equiv-utf-7-encoded.html

http/tests/security/xssAuditor/iframe-javascript-url.html
http/tests/security/xssAuditor/script-tag-utf-7-encoded.html
http/tests/security/xssAuditor/script-tag-with-source-relative-scheme.html

  • html/HTMLTokenizer.cpp: (WebCore::HTMLTokenizer::scriptHandler): Moved XSSAuditor check to HTMLTokenizer::parseTag. (WebCore::HTMLTokenizer::parseTag):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::loadSubframe): Modified to inform XSSAuditor of parent frame so as to compare against iFrame javascript URL.
  • page/XSSAuditor.cpp: Removed method XSSAuditor::isControlCharacter. Instead, exposed method isControlCharacter in ResourceResponseBase.cpp. (WebCore::XSSAuditor::XSSAuditor): (WebCore::XSSAuditor::decodeURL): Modified to decode string using specified encoder. (WebCore::XSSAuditor::findInRequest): Generalized to arbitrary frame so as to prevent execution of iFrame javascript URL.
  • page/XSSAuditor.h: Added field m_parentFrame.
  • platform/network/ResourceResponseBase.cpp: (WebCore::isControlCharacter):
  • platform/network/ResourceResponseBase.h:

LayoutTests:

2009-06-27 Daniel Bates <dbates@intudata.com>

Reviewed by Adam Barth.

Tests for https://bugs.webkit.org/show_bug.cgi?id=26708

Tests prevention of injected scripts with scheme relative paths, iFrame JavaScript URLs,
and UTF-7 encoding.

  • http/tests/security/xssAuditor/http-equiv-utf-7-encoded-expected.txt: Added.
  • http/tests/security/xssAuditor/http-equiv-utf-7-encoded.html: Added.
  • http/tests/security/xssAuditor/iframe-javascript-url-expected.txt: Added.
  • http/tests/security/xssAuditor/iframe-javascript-url.html: Added.
  • http/tests/security/xssAuditor/resources/echo-intertag-default-encode.pl: Added.
  • http/tests/security/xssAuditor/script-tag-utf-7-encoded-expected.txt: Added.
  • http/tests/security/xssAuditor/script-tag-utf-7-encoded.html: Added.
  • http/tests/security/xssAuditor/script-tag-with-source-relative-scheme-expected.txt: Added.
  • http/tests/security/xssAuditor/script-tag-with-source-relative-scheme.html: Added.
01:34 Changeset [45311] by oliver@apple.com

Bug 26771: Canvas is incorrectly tainted when drawing from a video element that uses <source> elements

Reviewed by Maciej Stachowiak.

The drawImage(<video>) logic naively assumes that it just needs
to check the src attribute of the video element when in fact it
needs to look at the url that is being played instead. Failure
to do this means that video provided through source elements
taints the canvas.

Test: media/video-canvas-source.html

00:34 Changeset [45310] by hausmann@webkit.org

2009-06-27 Simon Hausmann <simon.hausmann@nokia.com>

Build fix for Qt under Windows.

  • Api/qwebhistory.h: Use consistent export linkage for the datastream operators.
Note: See TracTimeline for information about the timeline view.