Timeline


and

12/16/09:

23:52 Changeset [52241] by abarth@webkit.org

2009-12-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Make UpdateStep quiet
https://bugs.webkit.org/show_bug.cgi?id=32599

  • Scripts/modules/buildsteps.py:
  • Scripts/modules/buildsteps_unittest.py:
23:41 Changeset [52240] by abarth@webkit.org

2009-12-16 Evan Martin <evan@chromium.org>

Reviewed by Adam Barth.

Add Gtk to the early warning system WebKit port list.

https://bugs.webkit.org/show_bug.cgi?id=32629

  • Scripts/modules/commands/early_warning_system.py:
  • Scripts/modules/webkitport.py:
  • Scripts/modules/webkitport_unittest.py:
23:40 Changeset [52239] by abarth@webkit.org

2009-12-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

post-diff is failing with exception under guess_reviewer_from_bug
https://bugs.webkit.org/show_bug.cgi?id=32642

Also refactor output capturing code into
OutputCapture.assert_outputs to share more code between tests.

  • Scripts/modules/buildsteps.py:
    • Add the missing include.
    • Give guess_reviewer_from_bug a private underscore.
  • Scripts/modules/buildsteps_unittest.py: Added.
    • Test to make sure _guess_reviewer_from_bug works as expected.
  • Scripts/modules/commands/commandtest.py:
    • Custom code is now obsoleted by OutputCapture.assert_outputs
  • Scripts/modules/commands/queues_unittest.py:
    • ditto
  • Scripts/modules/credentials_unittest.py:
    • ditto
  • Scripts/modules/mock_bugzillatool.py:
    • fetch_reviewed_patches_from_bug can never return None
  • Scripts/modules/multicommandtool_unittest.py:
    • Custom code is now obsoleted by OutputCapture.assert_outputs
  • Scripts/modules/outputcapture.py:
    • Add assert_outputs to share more code between tests.
  • Scripts/run-webkit-unittests:
    • Add buildsteps_unittest.py
23:37 Changeset [52238] by abarth@webkit.org

2009-12-16 Adam Barth <abarth@webkit.org>

Unreviewed. Added a default argument to _update_status so that callers
don't have to explictly pass None when they don't have a patch object.

  • Scripts/modules/commands/queues.py:
23:17 Changeset [52237] by ukai@chromium.org

Unreviewed, build fix.

[Chromium] Add EditableTextRole to catch up with
http://trac.webkit.org/changeset/52233

  • public/WebAccessibilityRole.h:

(WebKit::): Added WebAccessibilityRoleEditableText

  • src/AssertMatchingEnums.cpp: Added matching compile assert.
23:05 Changeset [52236] by ossy@webkit.org

[Qt] Put test into skiplist because of missing layoutTestController.setSmartInsertDeleteEnabled().

  • platform/qt/Skipped: editing/selection/doubleclick-beside-cr-span.html added.
22:30 Changeset [52235] by hamaji@chromium.org

2009-12-15 MORITA Hajime <morrita@gmail.com>

Reviewed by Eric Seidel.

Bug 28306: double-clicking a word inside <b> beside newline select two words
https://bugs.webkit.org/show_bug.cgi?id=28306


SimplifiedBackwardsTextIterator missed trailing whitespaces just
before folding line-break, which is used to detect word
boundaries. This fix checks strings on RenderText and expand text
range on SimplifiedBackwardsTextIterator to include trailing
whitespaces if availble.

Test: editing/selection/doubleclick-beside-cr-span.html

  • editing/TextIterator.cpp: (WebCore::collapsedSpaceLength): (WebCore::maxOffsetIncludingCollapsedSpaces): (WebCore::SimplifiedBackwardsTextIterator::advance):

2009-12-16 MORITA Hajime <morrita@gmail.com>

Reviewed by Eric Seidel.

Bug 28306: double-clicking a word inside <b> beside newline select two words
https://bugs.webkit.org/show_bug.cgi?id=28306

SimplifiedBackwardsTextIterator missed trailing whitespaces just
before folding line-break, which is used to detect word
boundaries. This fix checks strings on RenderText and expand text
range on SimplifiedBackwardsTextIterator to include trailing
whitespaces if availble.

  • editing/selection/doubleclick-beside-cr-span-expected.txt: Added.
  • editing/selection/doubleclick-beside-cr-span.html: Added.
22:19 Changeset [52234] by ukai@chromium.org

2009-12-16 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Alexey Proskuryakov.

Fix flakiness in WebSocket handshake console message.
https://bugs.webkit.org/show_bug.cgi?id=32598

In chromium, LayoutTests/websocket/tests/handshake-error.html is
flaky, because it outputs the following console message several times:

CONSOLE MESSAGE: line 0: Unexpected response code:101

Update m_mode in WebSocketHandshake correctly: Once it finds
error in handshake message, set m_mode to Failed and don't try
reading handshake message again. If handshake message is not yet
fully received, set m_mode to Incomplete, so next didReceiveData()
will try to check handshake message again.

  • websockets/WebSocketChannel.cpp: (WebCore::WebSocketChannel::didReceiveData):
  • websockets/WebSocketHandshake.cpp: (WebCore::WebSocketHandshake::readServerHandshake):
21:58 Changeset [52233] by jhoneycutt@apple.com

2009-12-16 Jon Honeycutt <jhoneycutt@apple.com>

MSAA: Accessibility role of text nodes is wrong

https://bugs.webkit.org/show_bug.cgi?id=32631
<rdar://problem/7369084>

Reviewed by Alice Liu.

Test: platform/win/accessibility/text-role.html

WebCore/:

  • accessibility/AccessibilityObject.h: (WebCore::): Add a new role to the enum. (WebCore::AccessibilityObject::roleValueForMSAA): Declare a function to return the accessibility role for MSAA; the base implementation returns the old role value.
  • accessibility/AccessibilityRenderObject.cpp: (WebCore::AccessibilityRenderObject::AccessibilityRenderObject): Initialize m_roleForMSAA. (WebCore::AccessibilityRenderObject::roleValueForMSAA): If m_roleForMSAA has been set, return it. If the renderer is a RenderText, set the role to EditableTextRole to match IE and Firefox. Otherwise, set the role for MSAA to the old role value.
  • accessibility/AccessibilityRenderObject.h: Add a member to hold the role for MSAA, and declare an override of roleValueForMSAA().

WebKit/win/:

  • AccessibleBase.cpp: (MSAARole): If the role is WebCore::EditableTextRole, return ROLE_SYSTEM_TEXT. (AccessibleBase::role): Call roleValueForMSAA().

LayoutTests/:

  • platform/win/accessibility/text-role-expected.txt: Added.
  • platform/win/accessibility/text-role.html: Added. Check that the role of the text node that is the first child of the <a> tag has the "editable text" role.
21:30 Changeset [52232] by tkent@chromium.org

2009-12-16 Kent Tamura <tkent@chromium.org>

Reviewed by David Levin.

check-webkit-style supports for TAB check against text files.
https://bugs.webkit.org/show_bug.cgi?id=32538

  • Scripts/check-webkit-style: Move process_patch() to style.py.
  • Scripts/modules/cpp_style.py: Add can_handle().
  • Scripts/modules/cpp_style_unittest.py: Add tests for can_handle().
  • Scripts/modules/style.py: Added. This is a front-end of cpp_style and text_style. It dispatches files to an appropriate linter.
  • Scripts/modules/text_style.py: Added. This is a linter module for generic text files. It supports only for TAB checking at this moment.
  • Scripts/modules/text_style_unittest.py: Added. Tests for text_style.py.
  • Scripts/run-webkit-unittests: Add text_style_unittest.
19:50 Changeset [52231] by ggaren@apple.com

Fixed <rdar://problem/7355025> Interpreter::privateExecute macro generates
bloated code

Reviewed by Oliver Hunt.

This patch cuts Interpreter stack use by about a third.

  • bytecode/Opcode.h: Changed Opcode to const void* to work with the

const static initiliazation we want to do in Interpreter::privateExecute.

  • interpreter/Interpreter.cpp:

(JSC::Interpreter::Interpreter): Moved hashtable initialization here to
avoid polluting Interpreter::privateExecute's stack, and changed it from a
series of add() calls to one add() call in a loop, to cut down on code size.

(JSC::Interpreter::privateExecute): Changed a series of label computations
to a copy of a compile-time constant array to cut down on code size.

16:44 Changeset [52230] by ukai@chromium.org

2009-12-16 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Darin Adler.

Fix LayoutTests/websocket/tests/url-with-query-for-no-query.html
https://bugs.webkit.org/show_bug.cgi?id=32606

It should test with /websocket/tests/no-query

  • websocket/tests/script-tests/url-with-query-for-no-query.js:
  • websocket/tests/url-with-query-for-no-query-expected.txt:
16:41 Changeset [52229] by eric@webkit.org

2009-12-16 Eric Seidel <eric@webkit.org>

No review, rolling out r52206.
http://trac.webkit.org/changeset/52206
https://bugs.webkit.org/show_bug.cgi?id=31187

This broke pixel tests on Mac. Presumably the code in
ImageCG is wrong here.

  • platform/graphics/BitmapImage.h:
  • platform/graphics/GeneratedImage.cpp: (WebCore::GeneratedImage::drawPattern):
  • platform/graphics/GeneratedImage.h:
  • platform/graphics/Image.cpp: (WebCore::Image::drawTiled):
  • platform/graphics/Image.h:
  • platform/graphics/cairo/ImageCairo.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/cg/ImageCG.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/haiku/ImageHaiku.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/qt/ImageQt.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/skia/ImageSkia.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/wince/GraphicsContextWince.cpp: (WebCore::GraphicsContext::drawBitmapPattern):
  • platform/graphics/wince/ImageBufferWince.cpp: (WebCore::): (WebCore::BufferedImage::drawPattern):
  • platform/graphics/wx/ImageWx.cpp: (WebCore::BitmapImage::drawPattern): (WebCore::Image::drawPattern):
16:40 Changeset [52228] by eric@webkit.org

2009-12-16 Eric Seidel <eric@webkit.org>

Reviewed by David Levin.

bugzilla-tool should not require users to install mechanize
https://bugs.webkit.org/show_bug.cgi?id=32635

Centralize our import logic.

  • Scripts/modules/bugzilla.py: use webkit_mechanize
  • Scripts/modules/statusbot.py: use webkit_mechanize
  • Scripts/modules/webkit_mechanize.py: Added.
16:31 Changeset [52227] by ossy@webkit.org

[Qt] editing/selection/empty-cell-right-click.html introduced in r52220 failed.
https://bugs.webkit.org/show_bug.cgi?id=32639

Test was refactored by r52226, and now it passes.

  • platform/qt/Skipped: unskip editing/selection/empty-cell-right-click.html.
16:15 Changeset [52226] by enrica@apple.com

Revised version of the test for:
<rdar://problem/5565465
https://bugs.webkit.org/show_bug.cgi?id=32622

Reviewed by Adele Peterson.

  • editing/selection/empty-cell-right-click-expected.txt:
  • editing/selection/empty-cell-right-click.html:
16:07 Changeset [52225] by ossy@webkit.org

Rubber-stamped by Simon Hausmann.

[Qt] editing/selection/empty-cell-right-click.html introduced in r52220 failed.
https://bugs.webkit.org/show_bug.cgi?id=32639

  • platform/qt/Skipped: editing/selection/empty-cell-right-click.html added until fix.
16:03 Changeset [52224] by dglazkov@chromium.org

2009-12-16 Dimitri Glazkov <dglazkov@chromium.org>

Reviewed by David Levin.

https://bugs.webkit.org/show_bug.cgi?id=32637
[V8] Generate most named property getters.

Covered by existing tests.

  • bindings/scripts/CodeGeneratorV8.pm: Added generation of SetNamedPropertyHandler call.
  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::getTemplate): Removed now-generated SetNamedPropertyHandler callsites.
15:52 Changeset [52223] by eric@webkit.org

2009-12-16 Luiz Agostini <luiz.agostini@openbossa.org>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Implement combobox delegate for Qt
https://bugs.webkit.org/show_bug.cgi?id=32550

Abstract popup menu factory.

A static method from QtAbstractPopupMenu is used to create its instances. If a factory
has been suplied to class QtAbstractPopupMenu the factory will be used to create the
objects. If not an instance of QtFallbackPopupMenu will be created.

The objective is to make it easy to replace the combobox popup at WebCore layer providing
support to the combobox popup delegation API. Future patches will make it possible to
replace the combobox popup at WebKit layer.

No behavior changes.

  • platform/qt/PopupMenuQt.cpp: (WebCore::PopupMenu::PopupMenu):
  • platform/qt/QtAbstractWebPopup.cpp: (WebCore::QtAbstractWebPopup::setFactory): (WebCore::QtAbstractWebPopup::create):
  • platform/qt/QtAbstractWebPopup.h:
  • platform/qt/QtFallbackWebPopup.cpp:
15:39 Changeset [52222] by mrowe@apple.com

Build fix. Disable debug variants of WebKit frameworks.

14:19 Changeset [52221] by eric@webkit.org

2009-12-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Generalize commit-queue recent status page for all queues
https://bugs.webkit.org/show_bug.cgi?id=32633

  • QueueStatusServer/index.html:
    • Generalize to support other queues.
  • QueueStatusServer/queue_status.py:
    • Generalize MainPage to support other queues.
14:02 Changeset [52220] by enrica@apple.com

WebCore: Right clicking in a table cell creates a selection that extends into the next cell.
<rdar://problem/5565465>
https://bugs.webkit.org/show_bug.cgi?id=32622

Reviewed by Eric Seidel.

Avoid extending the selection to the next cell if we start from an empty cell.

Test: editing/selection/empty-cell-right-click.html

  • editing/TypingCommand.cpp:

(WebCore::TypingCommand::deleteKeyPressed): Replaced the code that check for empty table cell
with a call to the new function.

  • editing/VisibleSelection.cpp:

(WebCore::VisibleSelection::setStartAndEndFromBaseAndExtentRespectingGranularity): Extends the
selection only if we are not starting from an empty table cell.

  • editing/htmlediting.cpp:

(WebCore::isEmptyTableCell): Added a function since this check is performed now in two places.

  • editing/htmlediting.h: Added isEmptyTableCell declaration.

LayoutTests: Right clicking in a table cell creates a selection that extends into the next cell.
<rdar://problem/5565465>
https://bugs.webkit.org/show_bug.cgi?id=32622

Reviewed by Eric Seidel.

  • editing/selection/empty-cell-right-click-expected.txt: Added.
  • editing/selection/empty-cell-right-click.html: Added.
13:56 Changeset [52219] by eric@webkit.org

2009-12-16 Dirk Schulze <krit@webkit.org>

Reviewed by Nikolas Zimmermann.

feMerge crahses if feMergeNodes attribute in is empty
https://bugs.webkit.org/show_bug.cgi?id=32489

The previous test only checked if feMerge works on empty node
without a last effect. In this case we get a source image and
everything works fine. This test also covers the case if there
is a last effect.

  • platform/mac/svg/custom/empty-merge-expected.checksum:
  • platform/mac/svg/custom/empty-merge-expected.png:
  • platform/mac/svg/custom/empty-merge-expected.txt:
  • svg/custom/empty-merge.svg:

2009-12-16 Dirk Schulze <krit@webkit.org>

Reviewed by Nikolas Zimmermann.

feMerge crahses if feMergeNodes attribute in is empty
https://bugs.webkit.org/show_bug.cgi?id=32489

SVGFEMergeElement only stored pointers to all effects of it's nodes
in a list. But it should store the RefPtr otherwise the effects get
dereferenced and WebKit crashes.
I also made a short clean-up in SVGFilterBuilder. No change in
functionality.

Test: I updated svg/custom/empty-merge.svg to cover this problem.

  • svg/SVGFEMergeElement.cpp: (WebCore::SVGFEMergeElement::build):
  • svg/graphics/filters/SVGFEMerge.cpp: (WebCore::FEMerge::FEMerge): (WebCore::FEMerge::create): (WebCore::FEMerge::mergeInputs): (WebCore::FEMerge::setMergeInputs):
  • svg/graphics/filters/SVGFEMerge.h:
  • svg/graphics/filters/SVGFilterBuilder.cpp: (WebCore::SVGFilterBuilder::add):
13:24 Changeset [52218] by eric@webkit.org

2009-12-16 Evan Martin <evan@chromium.org>

Reviewed by Adam Barth.

Early warning system server should display output as UTF-8.

https://bugs.webkit.org/show_bug.cgi?id=32625

  • QueueStatusServer/queue_status.py:
13:17 Changeset [52217] by bweinstein@apple.com

Part of <http://webkit.org/b/32568>.
Web Inspector: Context Menus should be used in more places.

Reviewed by Timothy Hatcher.

Add a context menu entry in the console to clear the console. This
is only shown when there is no user selection of text, because if
there is selection of the text, then we want to give the user the Copy +
Spell Check context menu that they are used to.

  • English.lproj/localizedStrings.js: Added "Clear Console".
  • inspector/front-end/ConsoleView.js:

(WebInspector.ConsoleView.prototype._handleContextMenuEvent):

13:17 Changeset [52216] by eric@webkit.org

2009-12-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Add error handling to the early warning system
https://bugs.webkit.org/show_bug.cgi?id=32594

This should be the last step in making the EWS operational. When we
have a build error, we post the log to QueueStatusServer and add a link
to the bug.

  • Scripts/modules/commands/early_warning_system.py:
  • Scripts/modules/commands/queues.py:
  • Scripts/modules/executive.py:
13:15 Changeset [52215] by christian@webkit.org

2009-12-16 Christian Dywan <christian@twotoasts.de>

Reviewed by Gustavo Noronha Silva.

Deprecate the old loading signals load-started, load-committed,
load-finished, load-progress-changed and change unit tests to use
load-status instead.

  • tests/testmimehandling.c: (idle_quit_loop_cb): (test_mime_type): use notify::load-status
  • tests/testwebdatasource.c: (notify_load_status_unreachable_cb): (notify_load_status_cb): (test_webkit_web_data_source): (test_webkit_web_data_source_unreachable_uri): Remove uses of g_main_loop_is_running. It's an error if the loop is wrongly terminated.
  • tests/testwebresource.c: (notify_load_status_cb): (test_web_resource_loading): (notify_load_status_sub_cb): (test_web_resource_sub_resource_loading): use notify::load-status
  • tests/testwebview.c: (idle_quit_loop_cb): (test_webkit_web_view_icon_uri):
  • tests/testwindow.c: (notify_load_status_cb): (test_webkit_window_scrollbar_policy): use notify::load-status
  • webkit/webkitwebframe.cpp: (webkit_web_frame_class_init):
  • webkit/webkitwebview.cpp: (webkit_web_view_class_init): Deprecate old load signals.
12:59 Changeset [52214] by bweinstein@apple.com

Fixes <https://bugs.webkit.org/show_bug.cgi?id=32621>.
Web Inspector: [REGRESSION (52095-52100)] Resources filter disappears when switching
from / to resources panel.

Reviewed by Pavel Feldman.

Changed style in sidebar-resizer-vertical, because if the min-height was 100%, we start
to cover the scope bar.

  • inspector/front-end/inspector.css:
12:14 Changeset [52213] by dglazkov@chromium.org

2009-12-16 Dimitri Glazkov <dglazkov@chromium.org>

Reviewed by Adam Barth.

[V8] Switch all indexers to be generated.
https://bugs.webkit.org/show_bug.cgi?id=32569

Covered by existing tests.

  • WebCore.gypi: Remove V8FileListCustom.cpp and V8ClientRecListCustom.cpp,

as they are now generated.

  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::getTemplate): Remove customization of ClientRectList,

FileList, and DataGridColumnList templates.

  • bindings/v8/custom/V8ClientRectListCustom.cpp: Removed.
  • bindings/v8/custom/V8CustomBinding.h: Remove custom indexer getter decls.
  • bindings/v8/custom/V8DataGridColumnListCustom.cpp: Remove indexer getter definition.
  • bindings/v8/custom/V8FileListCustom.cpp: Removed.
11:34 Changeset [52212] by eric@webkit.org

2009-12-16 anton muhin <antonm@chromium.org>

Reviewed by Darin Adler.

[dom] make dom/Node.cpp compile if DUMP_NODE_STATISTICS is on
https://bugs.webkit.org/show_bug.cgi?id=32617

  • dom/Node.cpp: (WebCore::Node::dumpStatistics):
11:25 Changeset [52211] by dglazkov@chromium.org

2009-12-16 Dimitri Glazkov <dglazkov@chromium.org>

Reviewed by Darin Fisher.

[V8] There is no such thing as HTMLSelectElementCollection.
https://bugs.webkit.org/show_bug.cgi?id=32616

Refactoring, covered by existing tests.

  • WebCore.gypi: Removed V8HTMLSelectElementCollectionCustom.cpp
  • bindings/v8/V8Collection.h: Removed one-off template.
  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::getTemplate): Renamed all references to HTMLSelectElementCollection

to HTMLSelectElement.

  • bindings/v8/custom/V8CustomBinding.h: Renamed custom indexer decls.
  • bindings/v8/custom/V8HTMLSelectElementCollectionCustom.cpp: Removed.
  • bindings/v8/custom/V8HTMLSelectElementCustom.cpp: Moved code as-is from

V8HTMLSelectElementCollectionCustom.cpp

11:20 Changeset [52210] by kov@webkit.org

Fixing bug number in ChangeLog

11:15 Changeset [52209] by sullivan@apple.com

Brought this file up to date, sans review.

  • StringsNotToBeLocalized.txt:
11:14 Changeset [52208] by eric@webkit.org

2009-12-16 Dan Winship <danw@gnome.org>

Reviewed by Gustavo Noronha Silva.

[Gtk] Content-Encoding support

https://bugs.webkit.org/show_bug.cgi?id=522772

  • configure.ac: require libsoup 2.28.2 for SoupContentDecoder

2009-12-16 Dan Winship <danw@gnome.org>

Reviewed by Gustavo Noronha Silva.

[Gtk] Content-Encoding support

https://bugs.webkit.org/show_bug.cgi?id=522772

  • platform/network/soup/ResourceHandleSoup.cpp: (WebCore::startHttp): Don't force "Accept-Encoding: identity" any more

2009-12-16 Dan Winship <danw@gnome.org>

Reviewed by Gustavo Noronha Silva.

Content-Encoding support

https://bugs.webkit.org/show_bug.cgi?id=522772

  • webkit/webkitprivate.cpp: (webkit_init): add a SoupContentDecoder feature to the session
11:03 Changeset [52207] by christian@webkit.org

2009-12-16 Christian Dywan <christian@twotoasts.de>

Reviewed by Gustavo Noronha Silva.

Take into account favicon.ico in the webresource test.

  • tests/testwebresource.c: (test_web_resource_sub_resource_loading):
11:00 Changeset [52206] by eric@webkit.org

2009-12-16 Benjamin Otte <otte@gnome.org>

Reviewed by Darin Adler.

Get rid of the phase argument to Image::drawPattern.
https://bugs.webkit.org/show_bug.cgi?id=31187

The argument can be expressed equally well as part of the
patternTransform. All backends but the Qt one did exactly that
manually anyway.

  • platform/graphics/BitmapImage.h:
  • platform/graphics/GeneratedImage.cpp: (WebCore::GeneratedImage::drawPattern):
  • platform/graphics/GeneratedImage.h:
  • platform/graphics/Image.cpp: (WebCore::Image::drawTiled):
  • platform/graphics/Image.h:
  • platform/graphics/cairo/ImageCairo.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/cg/ImageCG.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/haiku/ImageHaiku.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/qt/ImageQt.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/skia/ImageSkia.cpp: (WebCore::Image::drawPattern):
  • platform/graphics/wince/GraphicsContextWince.cpp: (WebCore::GraphicsContext::drawBitmapPattern):
  • platform/graphics/wince/ImageBufferWince.cpp: (WebCore::): (WebCore::BufferedImage::drawPattern):
  • platform/graphics/wx/ImageWx.cpp: (WebCore::BitmapImage::drawPattern): (WebCore::Image::drawPattern):
10:50 Changeset [52205] by eric@webkit.org

2009-12-16 Kenneth Russell <kbr@google.com>

Reviewed by Darin Fisher.

Performance problems with index validation code for drawElements
https://bugs.webkit.org/show_bug.cgi?id=32466

Added a cache of the maximum index for each element type to
WebGLBuffer, querying it before iterating through the indices in
the client-side copy of the buffer's data. Hoisted checks of the
size of the element array itself into validateElementArraySize to
avoid duplicating code.

The performance improvement has been measured with manual tests.
Existing layout tests cover the index validation functionality
and continue to pass as before.

  • html/canvas/WebGLBuffer.cpp: (WebCore::WebGLBuffer::WebGLBuffer): (WebCore::WebGLBuffer::associateBufferData): (WebCore::WebGLBuffer::associateBufferSubData): (WebCore::WebGLBuffer::getCachedMaxIndex): (WebCore::WebGLBuffer::setCachedMaxIndex): (WebCore::WebGLBuffer::clearCachedMaxIndices):
  • html/canvas/WebGLBuffer.h:
  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::validateElementArraySize): (WebCore::WebGLRenderingContext::validateIndexArrayConservative): (WebCore::WebGLRenderingContext::validateIndexArrayPrecise): (WebCore::WebGLRenderingContext::validateRenderingState): (WebCore::WebGLRenderingContext::drawElements):
  • html/canvas/WebGLRenderingContext.h:
10:43 Changeset [52204] by eric@webkit.org

2009-12-16 Zelidrag Hornung <zelidrag@chromium.org>

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=32261
Added ability to separate the autocomplete suggestion previewing from
the actual accepting of the suggested value in HTMLInputElement.
This element (it's single line text incarnation) can now
be put in suggestion mode where renderer might display the suggested
value without making it automatically exposed to JavaScript.

No new tests. This new methods are not exposed yet. It's use will be
platform specific. PopupMenu and Chromium specific changes will be
separated based on Darin Adler's comments.

  • dom/InputElement.h: (WebCore::InputElementData::suggestedValue): (WebCore::InputElementData::setSuggestedValue):
  • html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::suggestedValue): (WebCore::HTMLInputElement::setSuggestedValue): (WebCore::HTMLInputElement::setValue): (WebCore::HTMLInputElement::setValueFromRenderer):
  • html/HTMLInputElement.h:
  • rendering/RenderTextControlSingleLine.cpp: (WebCore::RenderTextControlSingleLine::updateFromElement):
  • wml/WMLInputElement.cpp: (WebCore::WMLInputElement::suggestedValue):
  • wml/WMLInputElement.h:
10:24 Changeset [52203] by pfeldman@chromium.org

2009-12-16 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Make Elements tree, properties section and resources
view monospace.

https://bugs.webkit.org/show_bug.cgi?id=32609

  • inspector/front-end/ElementsPanel.js:
  • inspector/front-end/PropertiesSection.js: (WebInspector.PropertiesSection):
  • inspector/front-end/ResourceView.js: (WebInspector.ResourceView):
  • inspector/front-end/inspector.css:
10:12 Changeset [52202] by eric@webkit.org

2009-12-16 Dirk Schulze <krit@webkit.org>

Reviewed by Darin Adler.

sourceGraphic wrong if positive x,y is >0 and objectBB is selected on Filter
https://bugs.webkit.org/show_bug.cgi?id=32575

Check if the GraphicsContex get adjusted, if the source rect is clipped
to the filter rect.

  • platform/mac/svg/filters/filter-source-position-expected.checksum: Added.
  • platform/mac/svg/filters/filter-source-position-expected.png: Added.
  • platform/mac/svg/filters/filter-source-position-expected.txt: Added.
  • svg/filters/filter-source-position.svg: Added.

2009-12-16 Dirk Schulze <krit@webkit.org>

Reviewed by Darin Adler.

sourceGraphic wrong if positive x,y is >0 and objectBB is selected on Filter
https://bugs.webkit.org/show_bug.cgi?id=32575

On SVG Filters, the rest of the source graphic gets clipped to the filter
rest. This helps us to minimize the size of intermediate ImageBuffers. If
the position moves to the right or to the bottom, we have to adjust the
GraphicContext of this intermediate ImageBuffer.
This patch fixes the adjustment of the context, according to the clipping.

Test: svg/filters/filter-source-position.svg

  • svg/graphics/SVGResourceFilter.cpp: (WebCore::SVGResourceFilter::prepareFilter):
09:59 Changeset [52201] by enrica@apple.com

CrashTracer: [USER] 1 crash in Safari at com.apple.WebCore: WebCore::endsOfNodeAreVisuallyDistinctPositions + 115
<rdar://problem/7467108>
https://bugs.webkit.org/show_bug.cgi?id=32590

Reviewed by Adele Peterson.

WebCore:

If the element is a button there is no need to look at the children to calculate the caret position.

Test: editing/selection/button-right-click.html

  • dom/Position.cpp:

(WebCore::Position::getInlineBoxAndOffset): Avoid calling recursively this method,
when starting from a button element.

LayoutTests:

  • editing/selection/button-right-click-expected.txt: Added.
  • editing/selection/button-right-click.html: Added.
09:37 Changeset [52200] by kov@webkit.org

Rubber-stamped by Xan Lopez.

Add tab_key_cycles_through_elements to copies of
WebKitWebSettings.

  • webkit/webkitwebsettings.cpp:

(webkit_web_settings_copy):

09:27 Changeset [52199] by eric@webkit.org

2009-12-16 Luiz Agostini <luiz.agostini@openbossa.org>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Implement combobox delegate for Qt
https://bugs.webkit.org/show_bug.cgi?id=32550

Refactoring of class QWebPopup.

Class QWebPopup has been split in QtAbstractWebPopup and QtFallbackWebPopup.
Both new classes are now in corresponding files and files QWebPopup.* have been removed.

Custom combo popup classes will inherit from QtAbstractWebPopup. It is not the public API as it is
in WebCore and will not be visible by users of QtWebKit. It will be used in implementation of
public QtWebKit combobox popup delegation API.

Class QtFallbackWebPopup inherits from QtAbstractWebPopup and implements the currently used combobox popup.
It needs to be improved to provide standard features that are missing like style or indentation.

No behavior changes.

  • WebCore.pro:
  • platform/PopupMenu.h:
  • platform/qt/PopupMenuQt.cpp: (WebCore::PopupMenu::PopupMenu):
  • platform/qt/QWebPopup.cpp: Removed.
  • platform/qt/QWebPopup.h: Removed.
  • platform/qt/QtAbstractWebPopup.cpp: Added. (WebCore::QtAbstractWebPopup::QtAbstractWebPopup): (WebCore::QtAbstractWebPopup::~QtAbstractWebPopup): (WebCore::QtAbstractWebPopup::client):
  • platform/qt/QtAbstractWebPopup.h: Added.
  • platform/qt/QtFallbackWebPopup.cpp: Added. (WebCore::QtFallbackWebPopup::QtFallbackWebPopup): (WebCore::QtFallbackWebPopup::show): (WebCore::QtFallbackWebPopup::populate): (WebCore::QtFallbackWebPopup::showPopup): (WebCore::QtFallbackWebPopup::hidePopup): (WebCore::QtFallbackWebPopup::activeChanged):
  • platform/qt/QtFallbackWebPopup.h: Added. (WebCore::QtFallbackWebPopup::hide):
09:04 Changeset [52198] by eric@webkit.org

2009-12-16 Martin Robinson <martin.james.robinson@gmail.com>

Reviewed by Gustavo Noronha Silva.

[GTK] webkit_web_navigation_action_get_button() documentation
https://bugs.webkit.org/show_bug.cgi?id=25555

Small documentation improvement for webkit_web_navigation_action_get_button.

  • webkit/webkitwebnavigationaction.cpp:
06:21 Changeset [52197] by ossy@webkit.org

Rubber-stamped by Kenneth Rohde Christiansen.

https://bugs.webkit.org/show_bug.cgi?id=31332

[Qt] After http://trac.webkit.org/changeset/52192
there are a failing and a timeouter test.

  • platform/qt/Skipped: http/tests/misc/drag-over-iframe-invalid-source-crash.html added until fix.
  • platform/qt/fast/text/international/khmer-selection-expected.txt: update outdated expected.
06:02 QtWebKit/CodeCamp200912 edited by zecke@selfish.org
Update day2 and day3 (diff)
05:57 QtWebKit/CodeCamp200912 edited by zecke@selfish.org
link to the presentation to freya and mozAfterPaint (diff)
05:54 QtWebKit/CodeCamp200912 edited by zecke@selfish.org
Update performance day 1 summary (diff)
05:46 Performance JS APIs.pdf attached to QtWebKit/CodeCamp200912 by zecke@selfish.org
JavaScript API to analyze performance
02:42 Changeset [52196] by ossy@webkit.org

Rubber-stamped by Simon Hausmann.

[Qt] http://trac.webkit.org/changeset/52194 fixed a side-effect
of DumpRenderTree. The previously wrong expected file updated.

  • platform/qt/css2.1/t1505-c524-font-var-00-b-expected.txt: updated.
02:25 Changeset [52195] by kov@webkit.org

Skipping a new test, and a test that fails only on the release bot,
apparently because of rouding errors.

  • platform/gtk/Skipped:
02:06 Changeset [52194] by hausmann@webkit.org

WebKitTools: Fix crash with tests that use custom font faces.

Reviewed by Holger Freyther.

https://bugs.webkit.org/show_bug.cgi?id=32600

In the DRT we call QApplication::removeAllApplicationFonts(), to make sure
that custom font faces between tests don't influence each other. Calling this
function in Qt however also invalidates all existing handles with QFontDatabase.

In order to make sure that WebCore also drops these handles we call QWebSettings::clearMemoryCaches(),
which implies a call to WebCore::FontCache::fontCache()->invalidate().

  • DumpRenderTree/qt/DumpRenderTreeQt.cpp:

(WebCore::DumpRenderTree::open):

LayoutTests: [Qt] Unskip fast/css/font-face-multiple-families.html.

Reviewed by Holger Freyther.

  • platform/qt/Skipped:
02:05 Changeset [52193] by pfeldman@chromium.org

2009-12-15 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: beautify HTTP info section.

https://bugs.webkit.org/show_bug.cgi?id=32583

  • English.lproj/localizedStrings.js:
  • inspector/front-end/ResourceView.js: (WebInspector.ResourceView): (WebInspector.ResourceView.prototype._refreshURL): (WebInspector.ResourceView.prototype._refreshHTTPInformation):
  • inspector/front-end/inspector.css:
01:54 Changeset [52192] by eric@webkit.org

2009-12-16 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Simon Hausmann.

[Qt] Maintain button state between mouse events.

This allows eventSender to generate drag events.

https://bugs.webkit.org/show_bug.cgi?id=32601

  • DumpRenderTree/qt/EventSenderQt.cpp: (EventSender::mouseDown): (EventSender::mouseUp): (EventSender::mouseMoveTo):
  • DumpRenderTree/qt/EventSenderQt.h:
01:52 Changeset [52191] by kov@webkit.org

Revert r52085, because it causes crashes on media tests in the release
bot.

01:41 QtWebKit edited by hausmann@webkit.org
(diff)
01:09 QtWebKitContrib edited by hausmann@webkit.org
(diff)
01:08 QtWebKit edited by hausmann@webkit.org
(diff)
00:56 Changeset [52190] by ossy@webkit.org

[Qt] fast/css/font-face-unused-source-loaded.html crashes
https://bugs.webkit.org/show_bug.cgi?id=32600

Reviewed by Simon Hausmann.

The crash caused by fast/css/font-face-multiple-families.html, added to skiplist until fix.

  • platform/qt/Skipped: fast/css/font-face-multiple-families.html added.
00:53 QtWebKitTodo edited by hausmann@webkit.org
(diff)

12/15/09:

23:54 Changeset [52189] by tkent@chromium.org

2009-12-15 Chris Jerdonek <chris.jerdonek@gmail.com>

Reviewed by David Levin.

Code clean-up in check-webkit-style as described below.
The only functional changes are minor improvements to the
script's help output.

https://bugs.webkit.org/show_bug.cgi?id=32487

  • Scripts/check-webkit-style:
    • Added the module docstring from cpp_style.py.
    • Improved wording of the --help output and added some of the default values as substitution strings.
    • Called cpp_style.exit_with_usage instead of sys.stderr.write in main function.
    • Made changes caused by changes to cpp_style.py.
  • Scripts/modules/cpp_style.py:
    • Removed module docstring, as well as redundant _USAGE string and main() function obsoleted by check-webkit-style.
    • Moved specification of default script values from within a function definition to global variables.
    • Moved the default webkit filter rules to be near the other global variables.
    • Renamed "FILTER" global variables to "FILTER_RULES" to establish unambiguous terminology.
    • Renamed _ERROR_CATEGORIES to _STYLE_CATEGORIES.
    • Changed _STYLE_CATEGORIES from a string to a list.
    • Added default webkit filter rules to --filter= output.
    • Renamed "print_" methods to "exit_with_" since they exist and added a temporary display_help parameter.
    • Added a temporary display_help parameter to parse_arguments.
  • Scripts/modules/cpp_style_unittest.py:
    • Made changes caused by changes to cpp_style.py.
    • Lower-cased ErrorCollector's ERROR_CATEGORIES instance variables since they are not global.
23:21 Changeset [52188] by abarth@webkit.org

Unreviewed fix for the style-queue.
Typo: _updates_status -> _update_status

We need to improve our testing infrastructure for the queues.

  • Scripts/modules/commands/queues.py:
23:05 Changeset [52187] by abarth@webkit.org

Unreviewed fix for the style-queue. I suspect this is a recent
regression from Eric's change below.

  • Scripts/modules/commands/queues.py:
19:12 Changeset [52186] by mjs@apple.com

2009-12-15 Maciej Stachowiak <mjs@apple.com>

Reviewed by Adele Peterson.

Make SunSpider version more prominent in the title
https://bugs.webkit.org/show_bug.cgi?id=32574

  • hosted/sunspider.html:
  • hosted/versions.html:
  • resources/driver-TEMPLATE.html:
  • resources/results-TEMPLATE.html:
18:26 Changeset [52185] by mjs@apple.com

2009-12-15 Maciej Stachowiak <mjs@apple.com>

Reviewed by Darin Adler.

Make SunSpider refuse to compare results across content versions
https://bugs.webkit.org/show_bug.cgi?id=32573


The results URL now includes a version indicator, and compare mode
will refuse to compare between versions.

  • resources/driver-TEMPLATE.html:
  • resources/results-TEMPLATE.html:
18:02 Changeset [52184] by rolandsteiner@chromium.org

Fix crash when encountering illegal HTML within a continuation inside <ruby>.
(https://bugs.webkit.org/show_bug.cgi?id=32316)

Reviewed by Adam Barth.

Avoid having the anonymous blocks utilized in the ruby renderers being
re-used as continuation blocks.
Avoid having ruby runs collected as "leftover blocks".

Test: fast/ruby/ruby-illegal.html

WebCore:

  • rendering/RenderRuby.h:

(WebCore::RenderRubyAsInline::removeLeftoverAnonymousBlock): mustn't be called
(WebCore::RenderRubyAsInline::isRuby): made private
(WebCore::RenderRubyAsInline::renderName): made private
(WebCore::RenderRubyAsInline::createsAnonymousWrapper): return true
(WebCore::RenderRubyAsBlock::removeLeftoverAnonymousBlock): mustn't be called
(WebCore::RenderRubyAsBlock::isRuby): made private
(WebCore::RenderRubyAsBlock::renderName): made private
(WebCore::RenderRubyAsBlock::createsAnonymousWrapper): return true

  • rendering/RenderRubyRun.h:

(WebCore::RenderRubyRun::removeLeftoverAnonymousBlock): avoid
(WebCore::RenderRubyRun::isRubyRun): made private
(WebCore::RenderRubyRun::renderName): made private
(WebCore::RenderRubyRun::createsAnonymousWrapper): return true

LayoutTests:

  • fast/ruby/ruby-illegal-expected.txt: Added.
  • fast/ruby/ruby-illegal.html: Added.
16:51 Changeset [52183] by eric@webkit.org

2009-12-15 Yaar Schnitman <yaar@chromium.org>

Reviewed by Darin Fisher.

update-webkit-chromium to auto-install gclient

https://bugs.webkit.org/show_bug.cgi?id=32587

  • Scripts/update-webkit-chromium:
16:04 Changeset [52182] by ggaren@apple.com

https://bugs.webkit.org/show_bug.cgi?id=32498
<rdar://problem/7471495>
REGRESSION(r51978-r52039): AJAX "Mark This Forum Read" function no longer
works

Reviewed by Sam "r=me" Weinig.

Fixed a tyop.

  • runtime/Operations.h:

(JSC::jsAdd): Use the '&&' operator, not the ',' operator.

15:53 Changeset [52181] by cmarrin@apple.com

Add delay load of QuartzCore.dll and QuartzCoreInterface.dll to fix Windows build

15:47 Changeset [52180] by eric@webkit.org

2009-12-15 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

bugzilla-tool / commit-queue should add results links to bugs when more results are available on the status bot
https://bugs.webkit.org/show_bug.cgi?id=32546

You can see an example of this working in:
https://bugs.webkit.org/show_bug.cgi?id=32585#c3

  • QueueStatusServer/queue_status.py:
    • Output the id of the newly created status.
  • Scripts/modules/commands/queues.py:
    • Tweak the commit-queue logging to include a full status link.
  • Scripts/modules/statusbot.py:
    • update_status should return the newly created status id.
15:44 Changeset [52179] by ggaren@apple.com

Try to fix the windows build: don't export this inlined function.

15:33 Changeset [52178] by ggaren@apple.com

Inlined JSCell's operator new.

Reviewed by Beth Dakin.

3.7% speedup on bench-allocate-nonretained.js.

(JSC::JSCell::operator new):

15:02 BuildingQtOnWindows edited by dbates@webkit.org
Replaced outdated Trolltech website link to Qt download with Nokia link (diff)
15:02 Changeset [52177] by japhet@chromium.org

2009-12-15 Nate Chapin <japhet@chromium.org>

Reviewed by Darin Adler.

Rename skipCanLoadCheck to skipSecurityCheck in a bunch of places in WebCore/loader.

https://bugs.webkit.org/show_bug.cgi?id=32529

  • loader/Cache.cpp: (WebCore::Cache::requestUserCSSStyleSheet):
  • loader/CachedImage.cpp: (WebCore::CachedImage::load):
  • loader/CachedResource.cpp: (WebCore::CachedResource::load):
  • loader/CachedResource.h: (WebCore::CachedResource::load):
  • loader/DocumentThreadableLoader.cpp: (WebCore::DocumentThreadableLoader::DocumentThreadableLoader): (WebCore::DocumentThreadableLoader::makeSimpleCrossOriginAccessRequest): (WebCore::DocumentThreadableLoader::makeCrossOriginAccessRequestWithPreflight): (WebCore::DocumentThreadableLoader::preflightSuccess): (WebCore::DocumentThreadableLoader::loadRequest):
  • loader/DocumentThreadableLoader.h:
  • loader/FrameLoaderTypes.h: (WebCore::):
  • loader/Request.cpp: (WebCore::Request::Request):
  • loader/Request.h: (WebCore::Request::shouldDoSecurityCheck):
  • loader/SubresourceLoader.cpp: (WebCore::SubresourceLoader::create):
  • loader/SubresourceLoader.h:
  • loader/loader.cpp: (WebCore::Loader::load): (WebCore::Loader::Host::servePendingRequests):
  • loader/loader.h:
14:40 Changeset [52176] by ggaren@apple.com

Removed the number heap, replacing it with a one-item free list for
numbers, taking advantage of the fact that two number cells fit inside
the space for one regular cell, and number cells don't require destruction.

Reviewed by Oliver Hunt.

SunSpider says 1.6% faster in JSVALUE32 mode (the only mode that
heap-allocates numbers).

SunSpider says 1.1% faster in JSVALUE32_64 mode. v8 says 0.8% faster
in JSVALUE32_64 mode. 10% speedup on bench-alloc-nonretained.js. 6%
speedup on bench-alloc-retained.js.

There's a lot of formulaic change in this patch, but not much substance.

(JSC::Debugger::recompileAllJSFunctions):

  • runtime/Collector.cpp:

(JSC::Heap::Heap):
(JSC::Heap::destroy):
(JSC::Heap::allocateBlock):
(JSC::Heap::freeBlock):
(JSC::Heap::freeBlockPtr):
(JSC::Heap::freeBlocks):
(JSC::Heap::recordExtraCost):
(JSC::Heap::allocate):
(JSC::Heap::resizeBlocks):
(JSC::Heap::growBlocks):
(JSC::Heap::shrinkBlocks):
(JSC::Heap::markConservatively):
(JSC::Heap::clearMarkBits):
(JSC::Heap::markedCells):
(JSC::Heap::sweep):
(JSC::Heap::markRoots):
(JSC::Heap::objectCount):
(JSC::Heap::addToStatistics):
(JSC::Heap::statistics):
(JSC::Heap::isBusy):
(JSC::Heap::reset):
(JSC::Heap::collectAllGarbage):
(JSC::Heap::primaryHeapBegin):
(JSC::Heap::primaryHeapEnd):

  • runtime/Collector.h:

(JSC::): Removed all code pertaining to the number heap, and changed all
heap template functions and classes to non-template functions and classes.

(JSC::Heap::allocateNumber): A new optimization to replace the number
heap: allocate half-sized number cells in pairs, returning the first
cell and caching the second cell for the next allocation.

  • runtime/CollectorHeapIterator.h:

(JSC::LiveObjectIterator::LiveObjectIterator):
(JSC::LiveObjectIterator::operator++):
(JSC::DeadObjectIterator::DeadObjectIterator):
(JSC::DeadObjectIterator::operator++):
(JSC::ObjectIterator::ObjectIterator):
(JSC::ObjectIterator::operator++):

  • runtime/JSCell.h:

(JSC::JSCell::isNumber): Removed all code pertaining to the number heap,
and changed all heap template functions and classes to non-template functions
and classes.

14:18 Changeset [52175] by cmarrin@apple.com

updated WebGL demos to latest spec

14:12 Changeset [52174] by mitz@apple.com

<rdar://problem/7470452> Safari/Chromium crashes on complicated @font-face rule
https://bugs.webkit.org/show_bug.cgi?id=32257

Reviewed by Darin Adler.

WebCore:

Test: fast/css/font-face-unused-source-loaded.html

The loading of a font resource caused a CSSFontFace that had the resource in one of its
sources, but wasn’t using it (due to using an earlier source) to invalidate its
CSSSegmentedFontFaces. This caused FontData for the active source (the one that did not just
finish loading) to be deleted, but that went unnoticed by Font::operator==(), since the
corresponding FontFallbackLists were not in the loading state. The fix is for CSSFontFace to
ignore loads from unused sources.

  • css/CSSFontFace.cpp:

(WebCore::CSSFontFace::fontLoaded): Bail out if the loaded source is not the active source.
(WebCore::CSSFontFace::getFontData): Set m_activeSource to the source that supplied the

FontData.

  • css/CSSFontFace.h:

(WebCore::CSSFontFace::CSSFontFace): Initialize m_activeSource.

LayoutTests:

  • fast/css/font-face-unused-source-loaded-expected.txt: Added.
  • fast/css/font-face-unused-source-loaded.html: Added.
14:12 Changeset [52173] by cmarrin@apple.com

Getting rid of out of date WebGL tests

14:07 Changeset [52172] by eric@webkit.org

2009-12-15 Yael Aharon <yael.aharon@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Cannot load flash files from a local file.
https://bugs.webkit.org/show_bug.cgi?id=32572

When loading the main resource and its mime type is application/octet-stream,
use the file extenstion to check if it is a supported plugin.

No new tests were added since any existing swf file can be used as a manual test.

  • loader/MainResourceLoader.cpp: (WebCore::MainResourceLoader::didReceiveResponse):
  • loader/MainResourceLoader.h:
  • plugins/PluginDatabase.h:
13:55 Changeset [52171] by eric@webkit.org

2009-12-15 Zoltan Horvath <zoltan@webkit.org>

Reviewed by Darin Adler.

Allow custom memory allocation control for WeakGCMap class
https://bugs.webkit.org/show_bug.cgi?id=32547

Inherits WeakGCMap from FastAllocBase because it is instantiated by
'new' at: WebCore/dom/Document.cpp:512.

  • runtime/WeakGCMap.h:
13:52 Changeset [52170] by bweinstein@apple.com

Fixes <https://bugs.webkit.org/show_bug.cgi?id=18498>.
REGRESSION: Tips are not shown for resources anymore.

Reviewed by Timothy Hatcher.

As discussed in the bug, Tips are not coming back, and the feature
will be moved to the audits panel. This patch is to remove the Tips
code and the resources that it needed.

  • WebCore.gypi: Removed tips files.
  • inspector/front-end/Images/tipBalloon.png: Removed.
  • inspector/front-end/Images/tipBalloonBottom.png: Removed.
  • inspector/front-end/Images/tipIcon.png: Removed.
  • inspector/front-end/Images/tipIconPressed.png: Removed.
  • inspector/front-end/Resource.js: Removed tips code.

(WebInspector.Resource.prototype.set finished):

  • inspector/front-end/WebKit.qrc: Removed tips files.
  • inspector/front-end/inspector.css: Removes tips style rules.
13:44 Changeset [52169] by pfeldman@chromium.org

2009-12-15 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Display request headers for main resource
having redirects.

https://bugs.webkit.org/show_bug.cgi?id=32577

  • inspector/InspectorController.cpp: (WebCore::InspectorController::identifierForInitialRequest): (WebCore::InspectorController::willSendRequest):
  • inspector/InspectorResource.cpp: (WebCore::InspectorResource::InspectorResource): (WebCore::InspectorResource::appendRedirect): (WebCore::InspectorResource::createCached): (WebCore::InspectorResource::updateRequest):
  • inspector/InspectorResource.h: (WebCore::InspectorResource::create):
13:35 Changeset [52168] by bweinstein@apple.com

Fixes <https://bugs.webkit.org/show_bug.cgi?id=32578>.
Web Inspector: No Need for ?="" now that we have context menus.

Reviewed by Timothy Hatcher.

Remove the toggleNewAttributeButton function and callers to it, as
we don't need this behavior anymore, now that we have context menus.

  • inspector/front-end/ElementsTreeOutline.js:

(WebInspector.ElementsTreeElement.prototype.set hovered):
(WebInspector.ElementsTreeElement.prototype.toggleNewAttributeButton):
(WebInspector.ElementsTreeElement.prototype._startEditing):

12:15 Changeset [52167] by mitz@apple.com

Fixed typos

11:55 Changeset [52166] by dglazkov@chromium.org

2009-12-15 Dimitri Glazkov <dglazkov@chromium.org>

Reviewed by Eric Seidel.

[V8] Generate trivial named property getters.
https://bugs.webkit.org/show_bug.cgi?id=32562

Covered by existing tests.

  • bindings/scripts/CodeGeneratorV8.pm: Added GenerateImplementationNamedPropertyGetter.
  • bindings/v8/V8Collection.h: (WebCore::getNamedPropertyOfCollection): Fixed a booboo by using AtomicString like we're supposed to.
  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::getTemplate): Removed code that's now generated.
11:51 Changeset [52165] by bweinstein@apple.com

Part of <http://webkit.org/b/32568>.
New Web Inspector: The Context Menu should be used in more places.

Reviewed by Timothy Hatcher.

Add support for context menu entries for editing and deleting elements
in the datagrid. This gives support to deleting cookies, and editing and
deleting session storage and local storage.

  • English.lproj/localizedStrings.js: Added "Edit" and "Add New" entries (delete was already there)
  • inspector/front-end/DataGrid.js:

(WebInspector.DataGrid): Added context menu listener.
(WebInspector.DataGrid.prototype._contextMenuInDataTable):

11:08 Changeset [52164] by eric@webkit.org

2009-12-15 Kenneth Russell <kbr@google.com>

Reviewed by Darin Fisher.

Don't allow default framebuffer to be mutated
https://bugs.webkit.org/show_bug.cgi?id=32391

  • fast/canvas/webgl/error-reporting-expected.txt:
  • fast/canvas/webgl/null-object-behaviour-expected.txt:
  • fast/canvas/webgl/script-tests/error-reporting.js:
  • fast/canvas/webgl/script-tests/null-object-behaviour.js:

2009-12-15 Kenneth Russell <kbr@google.com>

Reviewed by Darin Fisher.

Don't allow default framebuffer to be mutated
https://bugs.webkit.org/show_bug.cgi?id=32391

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::framebufferRenderbuffer): (WebCore::WebGLRenderingContext::framebufferTexture2D):
  • platform/graphics/mac/GraphicsContext3DMac.cpp: (WebCore::GraphicsContext3D::bindFramebuffer):

2009-12-15 Kenneth Russell <kbr@google.com>

Reviewed by Darin Fisher.

Don't allow default framebuffer to be mutated
https://bugs.webkit.org/show_bug.cgi?id=32391

  • src/GraphicsContext3D.cpp: (WebCore::GraphicsContext3DInternal::GraphicsContext3DInternal): (WebCore::GraphicsContext3DInternal::reshape): (WebCore::GraphicsContext3DInternal::beginPaint): (WebCore::GraphicsContext3DInternal::bindFramebuffer): (WebCore::GraphicsContext3D::bindFramebuffer):
10:50 Changeset [52163] by eric@webkit.org

2009-12-15 Zoltan Horvath <zoltan@webkit.org>

Reviewed by Darin Adler.

Allow custom memory allocation control for FontPlatformDataCacheKey class
https://bugs.webkit.org/show_bug.cgi?id=32552

Inherits the following struct from FastAllocBase because it is instantiated by 'new':

struct name - instantiated at: WebCore/'location'
struct FontPlatformDataCacheKey - platform/graphics/FontCache.cpp:118

  • platform/graphics/FontCache.cpp:
10:43 Changeset [52162] by eric@webkit.org

2009-12-15 Zoltan Horvath <zoltan@webkit.org>

Reviewed by Darin Adler.

Allow custom memory allocation control for dtoa's P5Node struct
https://bugs.webkit.org/show_bug.cgi?id=32544

Inherits P5Node struct from Noncopyable because it is instantiated by
'new' at wtf/dtoa.cpp:588 and don't need to be copyable.

  • wtf/dtoa.cpp:
10:14 Changeset [52161] by dglazkov@chromium.org

2009-12-15 Dimitri Glazkov <dglazkov@chromium.org>

Unreviewed, build fix.

[Chromium] Add TreeGridRole to catch up with
http://trac.webkit.org/changeset/52159.

  • public/WebAccessibilityRole.h: (WebKit::): Added WebAccessibilityRoleTreeGrid.
  • src/AssertMatchingEnums.cpp: Added matching compile assert.
10:13 Changeset [52160] by jianli@chromium.org

Bug 31090 - [Chromium] Add DownloadURL format to Chromium clipboard.
https://bugs.webkit.org/show_bug.cgi?id=31090

Reviewed by Dmitry Titov.

The proposal to whatwg can be found here:
http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2009-August/022118.html

WebCore:

  • platform/chromium/ChromiumDataObject.cpp:

(WebCore::ChromiumDataObject::clear):
(WebCore::ChromiumDataObject::hasData):
(WebCore::ChromiumDataObject::ChromiumDataObject):

  • platform/chromium/ChromiumDataObject.h:
  • platform/chromium/ClipboardChromium.cpp:

(WebCore::):
(WebCore::clipboardTypeFromMIMEType):
(WebCore::ClipboardChromium::setData):

WebKit/chromium:

  • public/WebDragData.h:
  • src/WebDragData.cpp:

(WebKit::WebDragData::downloadURL):
(WebKit::WebDragData::setDownloadURL):

10:13 BuildBot edited by yaar@chromium.org
(diff)
10:07 Changeset [52159] by cfleizach@apple.com

WAI-ARIA: implement treegrid
https://bugs.webkit.org/show_bug.cgi?id=31299

Reviewed by Darin Adler.

WebCore:

Test: platform/mac/accessibility/aria-treegrid.html

  • accessibility/AXObjectCache.cpp:

(WebCore::AXObjectCache::getOrCreate):

  • accessibility/AccessibilityARIAGrid.h:

(WebCore::AccessibilityARIAGrid::supportsSelectedRows):

  • accessibility/AccessibilityARIAGridRow.cpp:

(WebCore::AccessibilityARIAGridRow::isARIATreeGridRow):
(WebCore::AccessibilityARIAGridRow::disclosedRows):
(WebCore::AccessibilityARIAGridRow::disclosedByRow):

  • accessibility/AccessibilityARIAGridRow.h:
  • accessibility/AccessibilityObject.cpp:

(WebCore::createARIARoleMap):

  • accessibility/AccessibilityObject.h:

(WebCore::):
(WebCore::AccessibilityObject::isARIATreeGridRow):

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::isMultiSelectable):
(WebCore::AccessibilityRenderObject::setIsExpanded):
(WebCore::AccessibilityRenderObject::setSelectedRows):
(WebCore::AccessibilityRenderObject::ariaSelectedRows):
(WebCore::AccessibilityRenderObject::ariaListboxSelectedChildren):
(WebCore::AccessibilityRenderObject::selectedChildren):

  • accessibility/AccessibilityRenderObject.h:
  • accessibility/AccessibilityTable.h:

(WebCore::AccessibilityTable::supportsSelectedRows):

  • accessibility/mac/AccessibilityObjectWrapper.mm:

(-[AccessibilityObjectWrapper additionalAccessibilityAttributeNames]):
(-[AccessibilityObjectWrapper accessibilityAttributeNames]):
(-[AccessibilityObjectWrapper accessibilityAttributeValue:]):
(-[AccessibilityObjectWrapper accessibilitySetValue:forAttribute:]):

LayoutTests:

  • platform/mac-leopard/Skipped:
  • platform/mac/accessibility/aria-treegrid-expected.txt: Added.
  • platform/mac/accessibility/aria-treegrid.html: Added.
09:09 Changeset [52158] by eric@webkit.org

2009-12-15 Zoltan Horvath <zoltan@webkit.org>

Reviewed by Darin Adler.

Allow custom memory allocation control for EmptyEditorClient and EmptyInspectorClient
https://bugs.webkit.org/show_bug.cgi?id=32557

Inherits the following classes from Noncopyable because these are instantiated
by 'new' and no need to be copyable:

class name - instantiated at: WebCore/'location'
class EmptyEditorClient - svg/graphics/SVGImage.cpp:230
class EmptyInspectorClient - svg/graphics/SVGImage.cpp:241

  • loader/EmptyClients.h:
09:02 Changeset [52157] by eric@webkit.org

2009-12-15 Zoltan Horvath <zoltan@webkit.org>

Reviewed by Darin Adler.

Allow custom memory allocation control for WebCore/inspector directory's classes
https://bugs.webkit.org/show_bug.cgi?id=32556

Inherits the following class from Noncopyable because these are instantiated
by 'new' and no need to be copyable:

class name - instantiated at: WebCore/'location'
class InspectorFrontend - inspector/InspectorController.cpp:538
class InspectorTimelineAgent - inspector/InspectorController.cpp:1097

  • inspector/InspectorFrontend.h:
  • inspector/InspectorTimelineAgent.h:
08:54 Changeset [52156] by eric@webkit.org

2009-12-15 Zoltan Horvath <zoltan@webkit.org>

Reviewed by Darin Adler.

Allow custom memory allocation control for WebCore/platform/network directory's classes
https://bugs.webkit.org/show_bug.cgi?id=32553

Inherits the following class from Noncopyable because it is instantiated
by 'new' and no need to be copyable:

class/struct name - instantiated at: WebCore/'location'

struct CrossThreadResourceResponseData - platform/network/ResourceResponseBase.cpp:109

Inherits the following classes from FastAllocBase because these are
instantiated by 'new':

class ResourceResponseBase - (its child class) platform/network/ResourceResponseBase.cpp:90
class ResourceRequestBase - (its child class) loader/DocumentThreadableLoader.cpp:88

  • platform/network/ResourceRequestBase.h:
  • platform/network/ResourceResponseBase.h:
08:44 Changeset [52155] by eric@webkit.org

2009-12-15 Zoltan Horvath <zoltan@webkit.org>

Reviewed by Darin Adler.

Allow custom memory allocation control for AnimationControllerPrivate and FillLayerPropertyWrapperGetter
https://bugs.webkit.org/show_bug.cgi?id=32549

Inherits AnimationControllerPrivate and FillLayerPropertyWrapperGetter
classes from Noncopyable because these are instantiated by 'new' and no need to be copyable:

class/struct name - instantiated at: WebCore/'location'

AnimationControllerPrivate - page/animation/AnimationController.cpp:439
FillLayerPropertyWrapperGetter - (its child class) page/animation/AnimationBase.cpp:464

  • page/animation/AnimationBase.cpp:
  • page/animation/AnimationControllerPrivate.h:
08:36 Changeset [52154] by pfeldman@chromium.org

2009-12-15 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Do not lose HTTP redirect information, display it
in the Resources Panel.

https://bugs.webkit.org/show_bug.cgi?id=32551

  • inspector/InspectorController.cpp: (WebCore::InspectorController::populateScriptObjects): (WebCore::InspectorController::didCommitLoad): (WebCore::InspectorController::didLoadResourceFromMemoryCache): (WebCore::InspectorController::identifierForInitialRequest): (WebCore::InspectorController::willSendRequest): (WebCore::InspectorController::didReceiveResponse): (WebCore::InspectorController::didReceiveContentLength): (WebCore::InspectorController::didFinishLoading): (WebCore::InspectorController::didFailLoading):
  • inspector/InspectorFrontend.cpp:
  • inspector/InspectorFrontend.h:
  • inspector/InspectorResource.cpp: (WebCore::InspectorResource::InspectorResource): (WebCore::InspectorResource::appendRedirect): (WebCore::InspectorResource::updateScriptObject): (WebCore::InspectorResource::releaseScriptObject): (WebCore::InspectorResource::type):
  • inspector/InspectorResource.h: (WebCore::InspectorResource::): (WebCore::InspectorResource::Changes::setAll):
  • inspector/front-end/inspector.js: (WebInspector._addResource): (WebInspector.updateResource):
08:17 Changeset [52153] by zecke@webkit.org

[Qt] Followup for direct signals and slots connections in QNetworkReplyHandler

https://bugs.webkit.org/show_bug.cgi?id=32113

For Qt versions prior to 4.6.1 we will need to use QueuedConnection
for the signal and slot connection to avoid running into a bug in.

Add a comment that we want to remove one signal to forward
queued items when moving our minimum requirements.

  • platform/network/qt/QNetworkReplyHandler.cpp:

(WebCore::QNetworkReplyHandler::start):

08:17 Changeset [52152] by zecke@webkit.org

[Qt] Fix JavaScript prompt behavior for empty/null strings.

https://bugs.webkit.org/show_bug.cgi?id=30914

The patch is based on the work done by Gupta Manish.

In the default implementation of the JavaScript prompt
we are using a QInputDialog to get the text and this has
one quirk with regard to not entering any text.

In other WebKit ports and in Firefox an empty string is
returned but in the Qt case it is a null string.

Change the API documentation in QWebPage to mention we want to
have a non null string but do the fixup in the ChromeClientQt
to support existing code.

WebKit/qt:

  • Api/qwebpage.cpp:

(QWebPage::javaScriptPrompt): Change API documentation

  • WebCoreSupport/ChromeClientQt.cpp:

(WebCore::ChromeClientQt::runJavaScriptPrompt): Fixup null QString

WebCore:

  • manual-tests/qt/java-script-prompt.html: Added.
08:13 Changeset [52151] by eric@webkit.org

2009-12-15 Luiz Agostini <luiz.agostini@openbossa.org>

Reviewed by Kenneth Rohde Christiansen.

Moving list populate methods from PopupMenuQt to QWebPopup.

In preparation to future implementation of a delegation API for the combobox.

  • platform/PopupMenu.h:
  • platform/qt/PopupMenuQt.cpp: (WebCore::PopupMenu::show): (WebCore::PopupMenu::hide):
  • platform/qt/QWebPopup.cpp: (WebCore::QWebPopup::show): (WebCore::QWebPopup::populate):
  • platform/qt/QWebPopup.h: (WebCore::QWebPopup::hide):
06:18 Changeset [52150] by ossy@webkit.org

[Qt] Rename Qt DRT components to match other ports' naming convention and to be more understandable.

Patch by Andras Becsi <abecsi@inf.u-szeged.hu> on 2009-12-15
Reviewed by Tor Arne Vestbø.

  • DumpRenderTree/qt/DumpRenderTree.cpp: Removed.
  • DumpRenderTree/qt/DumpRenderTree.h: Removed.
  • DumpRenderTree/qt/DumpRenderTree.pro: Updated.
  • DumpRenderTree/qt/DumpRenderTreeQt.cpp: Copied from WebKitTools/DumpRenderTree/qt/DumpRenderTree.cpp.
  • DumpRenderTree/qt/DumpRenderTreeQt.h: Copied from WebKitTools/DumpRenderTree/qt/DumpRenderTree.h.
  • DumpRenderTree/qt/GCControllerQt.cpp: Copied from WebKitTools/DumpRenderTree/qt/jsobjects.cpp.

(GCController::GCController):

  • DumpRenderTree/qt/GCControllerQt.h: Copied from WebKitTools/DumpRenderTree/qt/jsobjects.h.
  • DumpRenderTree/qt/LayoutTestControllerQt.cpp:
  • DumpRenderTree/qt/WorkQueueItem.h: Removed.
  • DumpRenderTree/qt/WorkQueueItemQt.cpp: Updated includes.
  • DumpRenderTree/qt/WorkQueueItemQt.h: Copied from WebKitTools/DumpRenderTree/qt/WorkQueueItem.h.
  • DumpRenderTree/qt/jsobjects.cpp: Removed.
  • DumpRenderTree/qt/jsobjects.h: Removed.
  • DumpRenderTree/qt/main.cpp: Updated includes.
05:38 ASF-WebKit-Summit-final.pdf attached to QtWebKit/CodeCamp200912 by kenneth@webkit.org
Considering Tiling for a bette User Experience
04:20 QtWebKit/CodeCamp200912Teams edited by benjamin.poulain@nokia.com
(diff)
04:14 QtWebKit/CodeCamp200912 edited by benjamin.poulain@nokia.com
(diff)
04:08 QtWebKit/CodeCamp200912Teams edited by benjamin.poulain@nokia.com
(diff)
04:05 IMG_0885.jpg attached to QtWebKit/CodeCamp200912Teams by benjamin.poulain@nokia.com
04:00 IMG_0893.jpg attached to QtWebKit/CodeCamp200912Teams by benjamin.poulain@nokia.com
03:58 IMG_0881.jpg attached to QtWebKit/CodeCamp200912Teams by benjamin.poulain@nokia.com
03:58 IMG_0874.jpg attached to QtWebKit/CodeCamp200912Teams by benjamin.poulain@nokia.com
03:51 IMG_0867.jpg attached to QtWebKit/CodeCamp200912Teams by benjamin.poulain@nokia.com
03:29 Changeset [52149] by eric@webkit.org

2009-12-15 Alexander Pavlov <apavlov@chromium.org>

Reviewed by Pavel Feldman.

Extract WebInspector.Section from WebInspector.PropertiesSection.
https://bugs.webkit.org/show_bug.cgi?id=32523

03:12 Changeset [52148] by abarth@webkit.org

2009-12-15 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

We have two mark-fixed commands
https://bugs.webkit.org/show_bug.cgi?id=32073

  • Scripts/mark-bug-fixed: Removed.
  • Scripts/modules/commands/upload.py:
  • Scripts/modules/commands/upload_unittest.py:
02:48 IMG_0858.jpg attached to QtWebKit/CodeCamp200912Teams by benjamin.poulain@nokia.com
All the developers
02:39 Changeset [52147] by eric@webkit.org

2009-12-15 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

bugzilla-tool/commit-queue needs to upload failure logs when patches are rejected
https://bugs.webkit.org/show_bug.cgi?id=28286

  • QueueStatusServer/index.html:
    • Show [results] links if a results log was uploaded.
  • QueueStatusServer/queue_status.py:
    • Empty file uploads appear as u"" and Blob does not handle unicode, so convert to str().
  • Scripts/modules/commands/queues.py:
    • Post to the status bot when a patch fails and include the failure log as a results file.
02:32 QtWebKit/CodeCamp200912Teams created by benjamin.poulain@nokia.com
02:18 Changeset [52146] by eric@webkit.org

2009-12-15 Eric Seidel <eric@webkit.org>

No review, rolling out r52140.
http://trac.webkit.org/changeset/52140
https://bugs.webkit.org/show_bug.cgi?id=31299

The commit broke the Leopard bots and unfortunately
Chris wasn't around to fix them, so I'm rolling the change
out after 3 hours so that the bots can roll green again. :)

  • platform/mac/accessibility/aria-treegrid-expected.txt: Removed.
  • platform/mac/accessibility/aria-treegrid.html: Removed.

2009-12-15 Eric Seidel <eric@webkit.org>

No review, rolling out r52140.
http://trac.webkit.org/changeset/52140
https://bugs.webkit.org/show_bug.cgi?id=31299

The commit broke the Leopard bots and unfortunately
Chris wasn't around to fix them, so I'm rolling the change
out after 3 hours so that the bots can roll green again. :)

  • accessibility/AXObjectCache.cpp: (WebCore::AXObjectCache::getOrCreate):
  • accessibility/AccessibilityARIAGrid.h:
  • accessibility/AccessibilityARIAGridRow.cpp:
  • accessibility/AccessibilityARIAGridRow.h:
  • accessibility/AccessibilityObject.cpp: (WebCore::createARIARoleMap):
  • accessibility/AccessibilityObject.h: (WebCore::):
  • accessibility/AccessibilityRenderObject.cpp: (WebCore::AccessibilityRenderObject::isMultiSelect): (WebCore::AccessibilityRenderObject::setIsExpanded): (WebCore::AccessibilityRenderObject::setSelectedRows): (WebCore::AccessibilityRenderObject::ariaTreeSelectedRows): (WebCore::AccessibilityRenderObject::ariaListboxSelectedChildren): (WebCore::AccessibilityRenderObject::selectedChildren):
  • accessibility/AccessibilityRenderObject.h:
  • accessibility/AccessibilityTable.h:
  • accessibility/mac/AccessibilityObjectWrapper.mm: (-[AccessibilityObjectWrapper additionalAccessibilityAttributeNames]): (-[AccessibilityObjectWrapper accessibilityAttributeNames]): (-[AccessibilityObjectWrapper accessibilityAttributeValue:]): (-[AccessibilityObjectWrapper accessibilitySetValue:forAttribute:]):
02:07 Changeset [52145] by eric@webkit.org

2009-12-15 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

queue sub-commands need --status-host so they can report status
https://bugs.webkit.org/show_bug.cgi?id=32313

Make --status-bot a global option and make
run_bugzilla_tool pass --status-bot to sub-commands.

  • Scripts/bugzilla-tool:
    • Rename _status to status_bot and make it non-lazy.
  • Scripts/modules/commands/queues.py:
    • Move status updates out of WorkQueue and into individual queues.
  • Scripts/modules/commands/queues_unittest.py:
    • Test that --status-host is passed to bugzilla-tool when run as subcommand.
  • Scripts/modules/mock_bugzillatool.py:
    • Add a MockStatusBot
  • Scripts/modules/workqueue.py:
    • Remove status_host and work_work_logs_directory callbacks.
    • Add new work_item_log_path callback so that WorkQueue doesn't need to know about patches!
  • Scripts/modules/workqueue_unittest.py:
    • Update unit tests to reflect new callbacks.
01:29 Changeset [52144] by zecke@webkit.org

[Qt] Do not disable the inspector on show and hide

https://bugs.webkit.org/show_bug.cgi?id=31851

On Qt/X11 with some window managers the window will be
hidden when switching windows. In this case all the results
are gone when coming back to the window.

Attempt to use the CloseEvent to figure out if the window
was closed and withdrawn as this is more friendly to the
user of the inspector client.

  • Api/qwebinspector.cpp:

(QWebInspector::hideEvent):
(QWebInspector::closeEvent):

  • Api/qwebinspector.h:
01:29 Changeset [52143] by zecke@webkit.org

[Qt] Remove dead code from FontPlatformDataQt.cpp

https://bugs.webkit.org/show_bug.cgi?id=32540

In r51758 the code that used familyNames was removed
and we should remove the creation of the familyNames
as well.

  • platform/graphics/qt/FontPlatformDataQt.cpp:
00:57 Changeset [52142] by abarth@webkit.org

2009-12-15 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

scm_unittest.py is too slow
https://bugs.webkit.org/show_bug.cgi?id=31818

Now we don't run the SCM unit tests unless we get the --all flag on the
command line. Eric and I were commenting out this test because it was
too painful to run.

  • Scripts/run-webkit-unittests:
00:17 Changeset [52141] by mitz@apple.com

Reviewed by Adam Roben.

WebCore:

WebCore part of <rdar://problem/7173515> Use LOGFONT support in Core
Graphics when available

When using a version of Core Graphics with LOGFONT support, this fixes
<rdar://problem/5764844> First-time launch is painfully slow due to font
plist creation
and <rdar://problem/6576281> 12px synthetic-italics Tahoma text is less
readable in Safari than in Firefox/IE due to poorly-spaced letters

  • platform/graphics/win/FontCGWin.cpp: (WebCore::Font::drawGlyphs): No longer call drawGDIGlyphs for stroked text.
  • platform/graphics/win/FontCacheWin.cpp: (WebCore::createGDIFont): Added a synthesizeItalic parameter. When set to true, if italics are desired but the best match lacks italics, creates and returns a synthetic italics font. (WebCore::FontCache::createFontPlatformData): Ask to synthesize italics if needed.
  • platform/graphics/win/FontCustomPlatformData.cpp: (WebCore::FontCustomPlatformData::fontPlatformData): Updated an assertion, and changed to create the CGFont from the LOGFONT. (WebCore::createFontCustomPlatformData): Changed cgFont to be a RetainPtr. Do not create a CGFont from the font data if it can be created from the LOGFONT in fontPlatformData().
  • platform/graphics/win/FontDatabase.cpp: (WebCore::populateFontDatabase): Avoid font database work if fonts are not going to be looked up by name.
  • platform/graphics/win/FontPlatformDataCGWin.cpp: (WebCore::FontPlatformData::platformDataInit): Create the CGFont from a LOGFONT based on the HFONT.

WebKitLibraries:

WebKitSystemInterface part of <rdar://problem/7173515> Use LOGFONT support in
Core Graphics when available

  • win/include/WebKitSystemInterface/WebKitSystemInterface.h: Declared

wkCanCreateCGFontWithLOGFONT().

  • win/lib/WebKitSystemInterface.lib:
  • win/lib/WebKitSystemInterface_debug.lib:

12/14/09:

23:31 Changeset [52140] by cfleizach@apple.com

WAI-ARIA: implement treegrid
https://bugs.webkit.org/show_bug.cgi?id=31299

Reviewed by Darin Adler.

WebCore:

Test: platform/mac/accessibility/aria-treegrid.html

  • accessibility/AXObjectCache.cpp:

(WebCore::AXObjectCache::getOrCreate):

  • accessibility/AccessibilityARIAGrid.h:

(WebCore::AccessibilityARIAGrid::supportsSelectedRows):

  • accessibility/AccessibilityARIAGridRow.cpp:

(WebCore::AccessibilityARIAGridRow::isARIATreeGridRow):
(WebCore::AccessibilityARIAGridRow::disclosedRows):
(WebCore::AccessibilityARIAGridRow::disclosedByRow):

  • accessibility/AccessibilityARIAGridRow.h:
  • accessibility/AccessibilityObject.cpp:

(WebCore::createARIARoleMap):

  • accessibility/AccessibilityObject.h:

(WebCore::AccessibilityObject::isARIATreeGridRow):

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::isMultiSelectable):
(WebCore::AccessibilityRenderObject::setIsExpanded):
(WebCore::AccessibilityRenderObject::setSelectedRows):
(WebCore::AccessibilityRenderObject::ariaSelectedRows):
(WebCore::AccessibilityRenderObject::ariaListboxSelectedChildren):
(WebCore::AccessibilityRenderObject::selectedChildren):

  • accessibility/AccessibilityRenderObject.h:
  • accessibility/AccessibilityTable.h:

(WebCore::AccessibilityTable::supportsSelectedRows):

  • accessibility/mac/AccessibilityObjectWrapper.mm:

(-[AccessibilityObjectWrapper additionalAccessibilityAttributeNames]):
(-[AccessibilityObjectWrapper accessibilityAttributeNames]):
(-[AccessibilityObjectWrapper accessibilityAttributeValue:]):
(-[AccessibilityObjectWrapper accessibilitySetValue:forAttribute:]):

LayoutTests:

  • platform/mac/accessibility/aria-treegrid-expected.txt: Added.
  • platform/mac/accessibility/aria-treegrid.html: Added.
22:44 Changeset [52139] by mjs@apple.com

2009-12-14 Maciej Stachowiak <mjs@apple.com>

Reviewed by Darin Adler.

Some Browser-hosted SunSpider files are not valid HTML5
https://bugs.webkit.org/show_bug.cgi?id=32536


Made various fixes to validate as HTML5 with no warnings (mainly removing bogus
close tags and adding a meta charset declaration).

  • hosted/sunspider.html:
  • hosted/versions.html:
  • resources/TEMPLATE.html:
  • resources/driver-TEMPLATE.html:
  • resources/results-TEMPLATE.html:
22:25 Changeset [52138] by abarth@webkit.org

2009-12-14 Adam Barth <abarth@webkit.org>

Unreviewed. Fix failing unittest.

  • Scripts/modules/bugzilla_unittest.py:
22:24 Changeset [52137] by mjs@apple.com

2009-12-14 Maciej Stachowiak <mjs@apple.com>

Reviewed by Eric Seidel.

Make sunspider-0.9.1 the default content set (both command-line and hosted)
https://bugs.webkit.org/show_bug.cgi?id=32537

  • hosted/sunspider.html: Change default suite link.
  • sunspider: Change default suite.
  • sunspider-compare-results: Likewise.
22:16 Changeset [52136] by abarth@webkit.org

2009-12-14 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Kill _create_step_sequence
https://bugs.webkit.org/show_bug.cgi?id=32539

It's cleaner to represent not having these sequences as an empty
sequence instead of None.

  • Scripts/modules/commands/download.py:
  • Scripts/modules/stepsequence.py:
22:00 Changeset [52135] by eric@webkit.org

2009-12-14 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Move Credential handling out into a separate module
https://bugs.webkit.org/show_bug.cgi?id=32531

  • Scripts/modules/bugzilla.py:
  • Scripts/modules/credentials.py: Added.
  • Scripts/modules/credentials_unittest.py: Added.
  • Scripts/run-webkit-unittests:
21:53 Changeset [52134] by abarth@webkit.org

2009-12-14 Adam Barth <abarth@webkit.org>

Unreviewed. Address reviewer comments from an earlier patch. I didn't
do this earlier because I was worried about conflicts in dependent
patches.

  • Scripts/modules/commands/download.py:
21:38 Changeset [52133] by abarth@webkit.org

2009-12-14 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Make download commands declarative
https://bugs.webkit.org/show_bug.cgi?id=32469

This patch "properly" factors most of the download commands. These
commands are now largely declarative, which is the final step of this
grand refactoring.

  • Scripts/modules/buildsteps.py:
  • Scripts/modules/commands/download.py:
21:38 Changeset [52132] by abarth@webkit.org

2009-12-14 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Add AbstractPatchSequencingCommand to remove redundant code
https://bugs.webkit.org/show_bug.cgi?id=32468

Redundant code is bad. This patch moves us towards more declarative
commands.

  • Scripts/modules/commands/download.py:
21:32 Changeset [52131] by abarth@webkit.org

2009-12-14 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Kill WebKitApplyingScripts
https://bugs.webkit.org/show_bug.cgi?id=32467

Ah! I've been wanting to do this for a long time. This patch brings
the applying commands into the patch processing fold.

  • Scripts/bugzilla-tool:
  • Scripts/modules/buildsteps.py:
  • Scripts/modules/commands/download.py:
  • Scripts/modules/commands/download_unittest.py:
  • Scripts/modules/mock_bugzillatool.py:
21:31 Changeset [52130] by abarth@webkit.org

2009-12-14 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Convert rollout to StepSequence
https://bugs.webkit.org/show_bug.cgi?id=32406

  • Scripts/modules/buildsteps.py:
  • Scripts/modules/commands/download.py:
  • Scripts/modules/commands/download_unittest.py:
  • Scripts/modules/mock_bugzillatool.py:
21:21 Changeset [52129] by abarth@webkit.org

2009-12-14 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

[bzt] Kill LandingSequence
https://bugs.webkit.org/show_bug.cgi?id=32464

Removes LandingSequence in favor of StepSequence. This required
changing the Step API slightly to carry a general notion of state
instead of carrying patches specifically.

  • Scripts/modules/buildsteps.py:
  • Scripts/modules/commands/download.py:
  • Scripts/modules/commands/queues.py:
  • Scripts/modules/landingsequence.py: Removed.
  • Scripts/modules/stepsequence.py:
19:27 Changeset [52128] by mjs@apple.com

2009-12-14 Maciej Stachowiak <mjs@apple.com>

Reviewed by Sam Weinig.

in-browser SunSpider suffers excessive penalty under power management
https://bugs.webkit.org/show_bug.cgi?id=32505


I have made a few changes to address this:


1) Change how browser-hosted SunSpider loads tests - preload the content and write it in with
document.write to reduce triggering of progress UI.
2) Reduce time between tests to 10ms from 500ms, so that power management doesn't keep the CPU
stepped all the way down the whole time.
3) Run the test cycle an extra time for warmup before the runs that count.

  • hosted/sunspider-record-result.js: Removed.
  • make-hosted:
  • resources/TEMPLATE.html:
  • resources/driver-TEMPLATE.html:
19:05 Changeset [52127] by mitz@apple.com

<rdar://problem/7460655> REGRESSION (r47255): Content shifted down at software.opensuse.org
https://bugs.webkit.org/show_bug.cgi?id=32382

Reviewed by Darin Adler.

WebCore:

Test: fast/block/float/in-margin.html

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::getClearDelta): Floats that do not intrude into the line (because
they are in the margin) do not need to be cleared.

LayoutTests:

  • fast/block/float/in-margin.html: Added.
  • platform/mac/fast/block/float/in-margin-expected.checksum: Added.
  • platform/mac/fast/block/float/in-margin-expected.png: Added.
  • platform/mac/fast/block/float/in-margin-expected.txt: Added.
17:14 Changeset [52126] by enrica@apple.com

Pressing backspace inside a table cell erases all empty rows below it.
<rdar://problem/5565461>
https://bugs.webkit.org/show_bug.cgi?id=32526

Reviewed by John Sullivan.

WebCore:

No deletion is performed when the caret selection is on an empty table cell.

Test: editing/deleting/delete-empty-table.html

  • editing/TypingCommand.cpp:

(WebCore::TypingCommand::deleteKeyPressed): Added check for empty table cells in
case of caret selection.

LayoutTests:

  • editing/deleting/delete-empty-table-expected.txt: Added.
  • editing/deleting/delete-empty-table.html: Added.
16:19 Changeset [52125] by bweinstein@apple.com

Rubber-stamped by Brady Eidson.

Add two http iFrame dynamic form tests to the Windows Skipped
list.

  • platform/win/Skipped:
14:45 Changeset [52124] by eric@webkit.org

2009-12-14 Evan Stade <estade@chromium.org>

Reviewed by Darin Fisher.

https://bugs.webkit.org/show_bug.cgi?id=32347
Chromium: Context menu launched by menu key has funky behavior

Make the context menu's position depend on the selection only when an
editable field has focus. Also, don't send multiple context menu events
for a single keypress.

  • src/WebViewImpl.cpp: (WebKit::WebViewImpl::keyEvent): (WebKit::WebViewImpl::sendContextMenuEvent):
14:38 Changeset [52123] by eric@webkit.org

2009-12-14 Andrei Popescu <andreip@google.com>

Reviewed by Darin Adler.

[Android] Android needs ResourceHandle implementation
https://bugs.webkit.org/show_bug.cgi?id=32515

Adds implementation of ResourceHandle on Android.

No new tests needed, this is platform-specific code, so existing
tests are sufficient.

  • platform/network/android/ResourceHandleAndroid.cpp: Added. (WebCore::ResourceHandleInternal::~ResourceHandleInternal): (WebCore::ResourceHandle::~ResourceHandle): (WebCore::ResourceHandle::start): (WebCore::ResourceHandle::cancel): (WebCore::ResourceHandle::bufferedData): (WebCore::ResourceHandle::supportsBufferedData): (WebCore::ResourceHandle::setDefersLoading): (WebCore::ResourceHandle::willLoadFromCache): (WebCore::ResourceHandle::loadsBlocked): (WebCore::SyncLoader::SyncLoader): (WebCore::SyncLoader::~SyncLoader): (WebCore::SyncLoader::didReceiveResponse): (WebCore::SyncLoader::didReceiveData): (WebCore::SyncLoader::didFail): (WebCore::ResourceHandle::loadResourceSynchronously):
  • platform/network/android/ResourceLoaderAndroid.h: Added. (WebCore::ResourceLoaderAndroid::~ResourceLoaderAndroid): (WebCore::ResourceLoaderAndroid::ResourceLoaderAndroid):
14:30 Changeset [52122] by eric@webkit.org

2009-12-14 Marco Martin <notmart@gmail.com>

Reviewed by Kenneth Rohde Christiansen.

In the Scrollbar painting of the Qt theme integration,
intersects any previous clip rects with the one needed to paint the scrollbar.
This fixes the painting in QGraphicsview, when the parent of the widget that
paints the QWebPage has the QGraphicsItem::ItemClipsChildrenToShape set and
a piece of the scrollbar should be cipped away
https://bugs.webkit.org/show_bug.cgi?id=30366

No new tests.

  • platform/qt/ScrollbarThemeQt.cpp: (WebCore::ScrollbarThemeQt::paint):
14:23 Changeset [52121] by bfulgham@webkit.org

Provide working printing support for WinCairo port.

Reviewed by Adam Roben.

  • WebFrame.cpp:

(WebFrame::spoolPage): Conditionalize initialization of

PlatformGraphicsContext handling for CG vs. Cairo.

(scaleFactor): Add helper function.
(hdcFromContext): Add helper function.
(WebFrame::drawHeader): Correct Cairo variation.
(WebFrame::drawFooter): Correct Cairo variation.
(WebFrame::spoolPages): Correct Cairo variation.

14:17 Changeset [52120] by dglazkov@chromium.org

2009-12-14 Dimitri Glazkov <dglazkov@chromium.org>

Reviewed by Eric Seidel.

[V8] Generate bindings for trivial indexers.
https://bugs.webkit.org/show_bug.cgi?id=32455

Covered by existing tests.

  • bindings/scripts/CodeGeneratorV8.pm:

Added detecting indexers and generating code for them.

  • bindings/v8/V8Collection.h: Removed code that's now generated.
  • bindings/v8/V8DOMWrapper.cpp: (WebCore::V8DOMWrapper::getTemplate): Ditto.
14:16 Changeset [52119] by eric@webkit.org

2009-12-14 Kinuko Yasuda <kinuko@chromium.com>

Reviewed by Darin Fisher.

Make changing-frame-hierarchy-in-onload.html less flaky and
platform-independent.
https://bugs.webkit.org/show_bug.cgi?id=32319

  • http/tests/navigation/changing-frame-hierarchy-in-onload.html:
14:09 Changeset [52118] by eric@webkit.org

2009-12-14 Robert Hogan <robert@roberthogan.net>

Reviewed by Eric Seidel.

[Qt] Fix https://bugs.webkit.org/show_bug.cgi?id=32437

Amend the behaviour of DRT::notifyDone to dump only when
the page has finished loading.

This fix permits removal of the following tests from the Qt skipped list:
fast/forms/textarea-linewrap-dynamic.html
fast/forms/textarea-setvalue-submit.html
fast/forms/textarea-hard-linewrap-empty.html
fast/forms/submit-to-url-fragment.html
http/tests/misc/percent-sign-in-form-field-name.html
http/tests/security/escape-form-data-field-names.html

However it also requires the following two to be added to the skipped list:

http/tests/xmlhttprequest/access-control-basic-denied-preflight-cache.html
svg/custom/use-instanceRoot-as-event-target.xhtml

It appears the behaviour of notifyDone was masking problems with these two
tests.

  • platform/qt/Skipped:

2009-12-14 Robert Hogan <robert@roberthogan.net>

Reviewed by Eric Seidel.

[Qt] Fix https://bugs.webkit.org/show_bug.cgi?id=32437

Amend the behaviour of DRT::notifyDone to dump only when
the page has finished loading.

This fix permits removal of the following tests from the Qt skipped list:
fast/forms/textarea-linewrap-dynamic.html
fast/forms/textarea-setvalue-submit.html
fast/forms/textarea-hard-linewrap-empty.html
fast/forms/submit-to-url-fragment.html
http/tests/misc/percent-sign-in-form-field-name.html
http/tests/security/escape-form-data-field-names.html

However it also requires the following two to be added to the skipped list:

http/tests/xmlhttprequest/access-control-basic-denied-preflight-cache.html
svg/custom/use-instanceRoot-as-event-target.xhtml

It appears the behaviour of notifyDone was masking problems with these two
tests.

  • DumpRenderTree/qt/DumpRenderTree.cpp: (WebCore::DumpRenderTree::DumpRenderTree):
  • DumpRenderTree/qt/LayoutTestControllerQt.cpp: (LayoutTestController::reset): (LayoutTestController::processWork): (LayoutTestController::maybeDump): (LayoutTestController::notifyDone): (LayoutTestController::provisionalLoad):
  • DumpRenderTree/qt/LayoutTestControllerQt.h: (LayoutTestController::resetLoadFinished):
13:57 Changeset [52117] by eric@webkit.org

2009-12-14 Robert Hogan <robert@roberthogan.net>

Reviewed by Eric Seidel.

Bug 32519 - mouseEvent fires mutiple times in
LayoutTests/svg/custom/use-instanceRoot-as-event-target.xhtml

Add test to Qt skipped list (was passing due to bug in DRT, see
https://bugs.webkit.org/show_bug.cgi?id=32437). Updated detail on bug
in gtk and win skipped lists.

https://bugs.webkit.org/show_bug.cgi?id=32519

  • platform/gtk/Skipped:
  • platform/qt/Skipped:
  • platform/win/Skipped:
13:47 Changeset [52116] by ap@apple.com

Reviewed by Dave Hyatt.

https://bugs.webkit.org/show_bug.cgi?id=31660
REGRESSION (r49394): Image load event fires before the document fragment is attached.

Test: fast/images/image-load-event-in-fragment.html

We cannot fire load events immediately after parsing a fragment - an element inserted via
innerHTML or equivalent should be reachable via getElementById() by the time its load event
fires.

  • dom/Document.cpp: (WebCore::Document::implicitClose): Call dispatchPendingBeforeLoadEvents() and dispatchPendingLoadEvents() individually. I'm not sure why these calls are here, just preserving existing behavior.
  • dom/XMLTokenizer.cpp: (WebCore::XMLTokenizer::write):
  • html/HTMLTokenizer.cpp: (WebCore::HTMLTokenizer::write): Load events for elements that have just been parsed cannot be dispatched yet, we should give the elements a chance to be inserted into a document first. There is no compatibility baggage for beforeload, so it seems OK to dispatch it right away.
  • loader/ImageLoader.cpp: (WebCore::ImageLoader::dispatchPendingBeforeLoadEvents): (WebCore::ImageLoader::dispatchPendingLoadEvents):
  • loader/ImageLoader.h: Separated dispatchPendingEvents() into individual functions for each event kind.
13:27 Changeset [52115] by eric@webkit.org

2009-12-14 Jakob Petsovits <jpetsovits@rim.com>

Reviewed by Simon Fraser.

Add FloatQuad::isRectilinear() to check whether it can be represented as FloatRect.
https://bugs.webkit.org/show_bug.cgi?id=30442

This enables optimizations for distinguishing between
arbitrary FloatQuads and rectilinear ones.

  • platform/graphics/FloatQuad.cpp: (WebCore::FloatQuad::isRectilinear):
  • platform/graphics/FloatQuad.h:
13:08 Changeset [52114] by bweinstein@apple.com

Rubber-stamped by Adam Roben.

Add two XSS tests that intermittently time out to the Windows
Skipped list.

  • platform/win/Skipped:
12:57 Changeset [52113] by eric@webkit.org

2009-12-14 Simon Hausmann <hausmann@webkit.org>

Reviewed by Holger Freyther.

[Qt] Extend basic touch event test with a simple keyboard modifier test.

https://bugs.webkit.org/show_bug.cgi?id=32482

  • fast/events/basic-touch-events-expected.txt:
  • fast/events/script-tests/basic-touch-events.js: (singleTouchSequence):

2009-12-14 Simon Hausmann <hausmann@webkit.org>

Reviewed by Holger Freyther.

[Qt] Add support for keyboard modifiers to TouchEvent

https://bugs.webkit.org/show_bug.cgi?id=32482

Similar to other ui events with keyboard state, get the alt, shift,
meta and ctrl modifiers straight from the platform event.

  • dom/TouchEvent.cpp: (WebCore::TouchEvent::TouchEvent): (WebCore::TouchEvent::initTouchEvent):
  • dom/TouchEvent.h: (WebCore::TouchEvent::create):
  • dom/TouchEvent.idl:
  • page/EventHandler.cpp: (WebCore::EventHandler::handleTouchEvent):
  • platform/PlatformTouchEvent.h: (WebCore::PlatformTouchEvent::PlatformTouchEvent): (WebCore::PlatformTouchEvent::shiftKey): (WebCore::PlatformTouchEvent::ctrlKey): (WebCore::PlatformTouchEvent::altKey): (WebCore::PlatformTouchEvent::metaKey):
  • platform/qt/PlatformTouchEventQt.cpp: (WebCore::PlatformTouchEvent::PlatformTouchEvent):

2009-12-14 Simon Hausmann <hausmann@webkit.org>

Reviewed by Holger Freyther.

[Qt] Add support for keyboard modifiers to Qt DRT's EventSender for touch events

https://bugs.webkit.org/show_bug.cgi?id=32482

  • DumpRenderTree/qt/EventSenderQt.cpp: (EventSender::setTouchModifier): (EventSender::clearTouchPoints): (EventSender::sendTouchEvent):
  • DumpRenderTree/qt/EventSenderQt.h:
12:49 Changeset [52112] by eric@webkit.org

2009-12-14 Benjamin Poulain <benjamin.poulain@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] crash when clicking "Sign out" in GMail
https://bugs.webkit.org/show_bug.cgi?id=32300

When the ResourceLoader cancels the handle we have to make sure not to access
the resource handle afterwards again.
This complete r29515

  • platform/network/qt/QNetworkReplyHandler.cpp: (WebCore::QNetworkReplyHandler::sendResponseIfNeeded):
12:42 Changeset [52111] by eric@webkit.org

2009-12-14 Benjamin Poulain <benjamin.poulain@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Improve the autotests of QtWebkit
https://bugs.webkit.org/show_bug.cgi?id=32216

Refactor tst_qwebelement to remove the qWait()

  • tests/qwebelement/tst_qwebelement.cpp: (tst_QWebElement::style):
12:35 Changeset [52110] by eric@webkit.org

2009-12-14 Andreas Kling <andreas.kling@nokia.com>

Reviewed by Simon Hausmann.

Fix the QWebPage inputMethods() autotest after r51758
to compare the Qt::ImFont property's family against an explicitly
previously configured family.

https://bugs.webkit.org/show_bug.cgi?id=32491

  • tests/qwebpage/tst_qwebpage.cpp: (tst_QWebPage::inputMethods):
12:28 Changeset [52109] by eric@webkit.org

2009-12-14 Robert Hogan <robert@roberthogan.net>

Reviewed by Eric Seidel.

access-control-basic-denied-preflight-cache.html fails manual testing.

LayoutTests/http/tests/xmlhttprequest/access-control-basic-denied-preflight-cache.html
fails manual testing. It passes the automatic layout tests in Qt due to a bug with
notifyDone (see https://bugs.webkit.org/show_bug.cgi?id=32437).

Performing the test manually in QtLauncher shows that an exception is received.

Add access-control-basic-denied-preflight-cache.html to the Qt skipped list.

https://bugs.webkit.org/show_bug.cgi?id=32521

  • platform/qt/Skipped:
12:21 Changeset [52108] by eric@webkit.org

2009-12-14 Alexander Pavlov <apavlov@chromium.org>

Reviewed by Pavel Feldman.

Implement InjectedScript.evaluateOnSelf() that evaluates functions in the InjectedScript context.
https://bugs.webkit.org/show_bug.cgi?id=32517

  • inspector/front-end/InjectedScript.js: (InjectedScript.evaluateOnSelf):
  • inspector/front-end/InjectedScriptAccess.js:
12:13 Changeset [52107] by eric@webkit.org

2009-12-14 Nayan Kumar K <nayankk@gmail.com>

Reviewed by Darin Adler.

Fix for compilation errors in WML enabled builds.

https://bugs.webkit.org/show_bug.cgi?id=32481

  • wml/WMLInputElement.cpp: (WebCore::WMLInputElement::setValue): (WebCore::WMLInputElement::setValueForUser):
  • wml/WMLInputElement.h:
12:06 Changeset [52106] by eric.carlson@apple.com

2009-12-14 Eric Carlson <eric.carlson@apple.com>

Not reviewed, build fix.

Add missing newline at end of test result.

  • platform/mac-tiger/media/video-no-audio-expected.txt:
12:01 Changeset [52105] by benm@google.com

[Android] Android needs ResourceError, ResourceResponse and
ResourceRequest implementation.
https://bugs.webkit.org/show_bug.cgi?id=32522

Patch by Andrei Popescu <andreip@google.com> on 2009-12-14
Reviewed by Darin Adler.

Add implementation of ResourceError, ResourceResponse and ResourceRequest.
No new tests needed, the existing tests should be sufficient.

  • platform/network/android/ResourceError.h: Added.

(WebCore::ResourceError::ResourceError):

  • platform/network/android/ResourceRequest.h: Added.

(WebCore::ResourceRequest::ResourceRequest):
(WebCore::ResourceRequest::doUpdatePlatformRequest):
(WebCore::ResourceRequest::doUpdateResourceRequest):
(WebCore::ResourceRequest::setUserGesture):
(WebCore::ResourceRequest::getUserGesture):

  • platform/network/android/ResourceResponse.h: Added.

(WebCore::ResourceResponse::ResourceResponse):
(WebCore::ResourceResponse::doUpdateResourceResponse):

11:38 Changeset [52104] by weinig@apple.com

Fix the build.

  • page/Geolocation.cpp:
  • page/Geolocation.h:
11:26 Changeset [52103] by weinig@apple.com

Fix for https://bugs.webkit.org/show_bug.cgi?id=32499
Add client based Geolocation provider

Reviewed by Dan Bernstein.

Add first cut of a client based Geolocation provider. This is guarded by
ENABLE(CLIENT_BASED_GEOLOCATION) and is off by default for now. This adds a
GeolocationControllerClient interface that no-one currently implements,
but will in a subsequent patch.

WebCore:

  • GNUmakefile.am:
  • WebCore.base.exp:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • page/Geolocation.cpp:

(WebCore::createGeopositionFromGeolocationPosition):
(WebCore::createPositionErrorFromGeolocationError):
(WebCore::Geolocation::Geolocation):
(WebCore::Geolocation::~Geolocation):
(WebCore::Geolocation::disconnectFrame):
(WebCore::Geolocation::lastPosition):
(WebCore::Geolocation::startRequest):
(WebCore::Geolocation::fatalErrorOccurred):
(WebCore::Geolocation::requestTimedOut):
(WebCore::Geolocation::clearWatch):
(WebCore::Geolocation::handleError):
(WebCore::Geolocation::positionChanged):
(WebCore::Geolocation::makeSuccessCallbacks):
(WebCore::Geolocation::errorOccurred):
(WebCore::Geolocation::geolocationServicePositionChanged):
(WebCore::Geolocation::geolocationServiceErrorOccurred):
(WebCore::Geolocation::startUpdating):
(WebCore::Geolocation::stopUpdating):

  • page/Geolocation.h:
  • page/GeolocationController.cpp: Added.

(WebCore::GeolocationController::GeolocationController):
(WebCore::GeolocationController::~GeolocationController):
(WebCore::GeolocationController::addObserver):
(WebCore::GeolocationController::removeObserver):
(WebCore::GeolocationController::positionChanged):
(WebCore::GeolocationController::errorOccurred):
(WebCore::GeolocationController::lastPosition):

  • page/GeolocationController.h: Added.
  • page/GeolocationControllerClient.h: Added.

(WebCore::GeolocationControllerClient::~GeolocationControllerClient):

  • page/GeolocationError.h: Added.

(WebCore::GeolocationError::):
(WebCore::GeolocationError::create):
(WebCore::GeolocationError::code):
(WebCore::GeolocationError::message):

  • page/GeolocationPosition.h: Added.

(WebCore::GeolocationPosition::create):
(WebCore::GeolocationPosition::timestamp):
(WebCore::GeolocationPosition::latitude):
(WebCore::GeolocationPosition::longitude):
(WebCore::GeolocationPosition::accuracy):
(WebCore::GeolocationPosition::altitude):
(WebCore::GeolocationPosition::altitudeAccuracy):
(WebCore::GeolocationPosition::heading):
(WebCore::GeolocationPosition::speed):
(WebCore::GeolocationPosition::canProvideAltitude):
(WebCore::GeolocationPosition::canProvideAltitudeAccuracy):
(WebCore::GeolocationPosition::canProvideHeading):
(WebCore::GeolocationPosition::canProvideSpeed):
(WebCore::GeolocationPosition::GeolocationPosition):

  • page/Geoposition.h:

(WebCore::Geoposition::create):

  • page/Page.cpp:

(WebCore::Page::Page):

  • page/Page.h:

(WebCore::Page::geolocationController):

  • svg/graphics/SVGImage.cpp:

(WebCore::SVGImage::dataChanged):

WebKit/chromium:

  • src/WebViewImpl.cpp:

(WebKit::WebViewImpl::WebViewImpl):

WebKit/gtk:

  • webkit/webkitwebview.cpp:

(webkit_web_view_init):

WebKit/mac:

  • WebView/WebView.mm:

(-[WebView _commonInitializationWithFrameName:groupName:usesDocumentViews:]):

WebKit/qt:

  • Api/qwebpage.cpp:

(QWebPagePrivate::QWebPagePrivate):

WebKit/win:

  • WebView.cpp:

(WebView::initWithFrame):

11:10 QtWebKit/CodeCamp200912 edited by carol.szabo@nokia.com
(diff)
10:45 Changeset [52102] by agl@chromium.org

2009-12-14 Adam Langley <agl@google.com>

Reviewed by Darin Adler.

Test that an image's size is valid before reading it.

Previously, an image that failed to load (m_failed == true) would
trigger an assertion failure when WebKit tried to get its size.

https://bugs.webkit.org/show_bug.cgi?id=32176
http://code.google.com/p/chromium/issues/detail?id=28566

Test: fast/images/size-failure.html

  • platform/graphics/ImageSource.cpp: (WebCore::ImageSource::createFrameAtIndex):
10:40 Changeset [52101] by ggaren@apple.com

https://bugs.webkit.org/show_bug.cgi?id=32524
REGRESSION(52084): fast/dom/prototypes.html failing two CSS tests

Reviewed by Simon Fraser.

  • wtf/StdLibExtras.h:

(WTF::bitCount): The original patch put the parentheses in the wrong
place, completely changing the calculation and making it almost always
wrong. Moved the parentheses around the '+' operation, like the original
compiler warning suggested.

10:04 Changeset [52100] by mitz@apple.com

<rdar://problem/7458272> REGRESSION (r50470) Reproducible ASSERT in
RenderWidget::setWidgetGeometry()

Reviewed by Darin Adler.

WebCore:

Test: fast/forms/defer-updateFromElement.html

  • html/HTMLFormControlElement.cpp:

(WebCore::updateFromElementCallback): Added this post-attach callback
that invokes updateFromElement().
(WebCore::HTMLFormControlElement::recalcStyle): Instead of calling
updateFromElement(), enqueue a post-attach (in this case, post-style
recalc) callback to do it. updateFromElement() can cause the selection
to change, and in turn trigger synchronous layout, so it must not be
called during style recalc.

LayoutTests:

  • fast/forms/defer-updateFromElement-expected.txt: Added.
  • fast/forms/defer-updateFromElement.html: Added.
10:01 Changeset [52099] by pfeldman@chromium.org

2009-12-14 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: [regression] inspector deletes nodes when editing css.

Now all the treeoutlines are focusable, sections and their titles
are also focusable. Change contains couple of drive-by fixes such as
canceling of the attribute edit nuking attr info and maintaining proper
selection while deleting nodes.

https://bugs.webkit.org/show_bug.cgi?id=32460

  • inspector/front-end/CallStackSidebarPane.js: (WebInspector.CallStackSidebarPane.prototype.handleShortcut):
  • inspector/front-end/ConsoleView.js: (WebInspector.ConsoleView.prototype._promptKeyDown):
  • inspector/front-end/DataGrid.js: (WebInspector.DataGrid.prototype._keyDown):
  • inspector/front-end/DatabaseQueryView.js: (WebInspector.DatabaseQueryView): (WebInspector.DatabaseQueryView.prototype._promptKeyDown):
  • inspector/front-end/ElementsPanel.js: (WebInspector.ElementsPanel.this.treeOutline.focusedNodeChanged): (WebInspector.ElementsPanel.prototype.get defaultFocusedElement): (WebInspector.ElementsPanel.prototype.handleShortcut):
  • inspector/front-end/ElementsTreeOutline.js: (WebInspector.ElementsTreeOutline): (WebInspector.ElementsTreeOutline.prototype._keyDown): (WebInspector.ElementsTreeElement.prototype._textNodeEditingCommitted): (WebInspector.ElementsTreeElement.prototype._editingCancelled):
  • inspector/front-end/Panel.js: (WebInspector.Panel.prototype.show): (WebInspector.Panel.prototype.get defaultFocusedElement):
  • inspector/front-end/PropertiesSection.js: (WebInspector.PropertiesSection):
  • inspector/front-end/ScriptsPanel.js: (WebInspector.ScriptsPanel): (WebInspector.ScriptsPanel.prototype.get defaultFocusedElement): (WebInspector.ScriptsPanel.prototype.handleShortcut):
  • inspector/front-end/SidebarPane.js: (WebInspector.SidebarPane): (WebInspector.SidebarPane.prototype.toggleExpanded): (WebInspector.SidebarPane.prototype._onTitleKeyDown):
  • inspector/front-end/SourceFrame.js: (WebInspector.SourceFrame.prototype._loaded):
  • inspector/front-end/TextPrompt.js: (WebInspector.TextPrompt):
  • inspector/front-end/inspector.css:
  • inspector/front-end/inspector.html:
  • inspector/front-end/inspector.js: (WebInspector.loaded): (WebInspector.documentKeyDown): (WebInspector.documentCanCopy): (WebInspector.documentCopy): (WebInspector._searchFieldManualFocus): (WebInspector._searchKeyDown): (WebInspector.startEditing.cleanUpAfterEditing): (WebInspector.startEditing.keyDownEventListener): (WebInspector.startEditing):
  • inspector/front-end/treeoutline.js: (TreeOutline): (TreeOutline._removeChildAtIndex): (TreeOutline.prototype._treeKeyDown): (TreeElement.prototype.deselect):
  • inspector/front-end/utilities.js: ():
09:46 Changeset [52098] by aroben@apple.com

Change IWebFramePrivate's vtable to be compatible with Safari 4.0.4

Reviewed by Steve Falkenburg.

Fixes <http://webkit.org/b/32433> REGRESSION (r51567): Right click on
a link element crashes WebKit nightly

  • Interfaces/IWebFramePrivate.idl: Moved pauseSVGAnimation to the end

of the interface so the vtable that Safari 4.0.4 sees is unchanged.

09:43 Changeset [52097] by eric.carlson@apple.com

2009-12-14 Eric Carlson <eric.carlson@apple.com>

Not reviewed, build fix.

Update Tiger media test results after r52094.

  • platform/mac-tiger/accessibility: Added.
  • platform/mac-tiger/accessibility/media-element-expected.txt: Added.
  • platform/mac-tiger/media/audio-controls-rendering-expected.txt: Added.
  • platform/mac-tiger/media/controls-after-reload-expected.txt: Added.
  • platform/mac-tiger/media/controls-drag-timebar-expected.txt: Added.
  • platform/mac-tiger/media/controls-strict-expected.txt: Added.
  • platform/mac-tiger/media/controls-styling-expected.txt: Added.
  • platform/mac-tiger/media/video-controls-rendering-expected.txt: Added.
  • platform/mac-tiger/media/video-display-toggle-expected.txt: Added.
  • platform/mac-tiger/media/video-no-audio-expected.txt: Added.
  • platform/mac-tiger/media/video-zoom-controls-expected.txt: Added.
09:20 Changeset [52096] by dglazkov@chromium.org

2009-12-14 Dimitri Glazkov <dglazkov@chromium.org>

Unreviewed, build fix.

[Chromium] Guard all of JSC's wrapper cache-related members on Document.

  • dom/Document.cpp: (WebCore::Document::createWrapperCache): Widened the guards to the whole method.
  • dom/Document.h: Added JSC guards to wrapper cache-related members.
08:25 Changeset [52095] by cfleizach@apple.com

AX: there's an infinite loop going on for some webpages - facebook.com
https://bugs.webkit.org/show_bug.cgi?id=32501

Reviewed by Adele Peterson.

WebCore:

The problem is that parentObjectUnignored() cannot be called
while in accessibilityIsIgnored, because it leads to
recursiveness. Specifically, there were checks to determine
if an object was within a ARIA tree. That check should
1) be moved to all platforms, and 2) not call parentObjectUnignored().

Test: accessibility/hang-in-isignored.html

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::isAllowedChildOfTree):
(WebCore::AccessibilityRenderObject::accessibilityIsIgnored):

  • accessibility/AccessibilityRenderObject.h:
  • accessibility/mac/AccessibilityObjectMac.mm:

(WebCore::AccessibilityObject::accessibilityPlatformIncludesObject):

LayoutTests:

  • accessibility/hang-in-isignored-expected.txt: Added.
  • accessibility/hang-in-isignored.html: Added.
08:16 Changeset [52094] by eric.carlson@apple.com

2009-12-14 Eric Carlson <eric.carlson@apple.com>

Reviewed by Dan Bernstein.

<rdar://problem/6293979> Adopt shared control drawing for <video> controls on Leopard and Windows

Use MediaControllerThemeQuickTime on Leopard.

  • rendering/RenderThemeMac.mm: (WebCore::mediaControllerTheme):

2009-12-14 Eric Carlson <eric.carlson@apple.com>

Reviewed by Dan Bernstein.

<rdar://problem/6293979> Adopt shared control drawing for <video> controls on Leopard and Windows

Regenerate layout and pixel test results for the new UI.

  • platform/mac-leopard/Skipped:
  • platform/mac-leopard/accessibility/media-element-expected.txt:
  • platform/mac-leopard/media/audio-controls-rendering-expected.checksum:
  • platform/mac-leopard/media/audio-controls-rendering-expected.txt:
  • platform/mac-leopard/media/controls-after-reload-expected.checksum: Added.
  • platform/mac-leopard/media/controls-after-reload-expected.png: Added.
  • platform/mac-leopard/media/controls-after-reload-expected.txt:
  • platform/mac-leopard/media/controls-drag-timebar-expected.txt:
  • platform/mac-leopard/media/controls-strict-expected.checksum: Added.
  • platform/mac-leopard/media/controls-strict-expected.png: Added.
  • platform/mac-leopard/media/controls-strict-expected.txt:
  • platform/mac-leopard/media/controls-strict-mode-expected.txt: Added.
  • platform/mac-leopard/media/controls-styling-expected.checksum: Added.
  • platform/mac-leopard/media/controls-styling-expected.png: Added.
  • platform/mac-leopard/media/controls-styling-expected.txt:
  • platform/mac-leopard/media/video-controls-rendering-expected.checksum:
  • platform/mac-leopard/media/video-controls-rendering-expected.txt:
  • platform/mac-leopard/media/video-display-toggle-expected.checksum:
  • platform/mac-leopard/media/video-display-toggle-expected.txt:
  • platform/mac-leopard/media/video-no-audio-expected.checksum: Added.
  • platform/mac-leopard/media/video-no-audio-expected.png: Added.
  • platform/mac-leopard/media/video-no-audio-expected.txt:
  • platform/mac-leopard/media/video-zoom-controls-expected.checksum:
  • platform/mac-leopard/media/video-zoom-controls-expected.txt:
07:44 Changeset [52093] by aroben@apple.com

Build fix

  • WebKit.vcproj/WebKit.vcproj: Added $(DXSDK_DIR)\Lib\x86 to the

libpath for all non-Cairo configurations (it was missing from most of
them).

07:38 Changeset [52092] by aroben@apple.com

Move QuartzCore-related build settings into a vsprops file

Fixes <http://webkit.org/b/32518>

Reviewed by Dan Bernstein.

  • WebCore.vcproj/WebCore.vcproj: Let VS reformat the files list, added

WebCoreQuartzCore.vsprops to the non-Cairo configurations, moved all
other QuartzCore-related build settings from here...

  • WebCore.vcproj/WebCoreQuartzCore.vsprops: ...to here.
07:21 Changeset [52091] by eric@webkit.org

2009-12-14 Yael Aharon <yael.aharon@nokia.com>

Reviewed by Antti Koivisto.

handling scripts can block UI
https://bugs.webkit.org/show_bug.cgi?id=27612

Break execution of external scrips to smaller chunks.
No new tests since no new functionality was introduced.

  • html/HTMLTokenizer.cpp: (WebCore::HTMLTokenizer::HTMLTokenizer): (WebCore::HTMLTokenizer::reset): (WebCore::HTMLTokenizer::notifyFinished): (WebCore::HTMLTokenizer::executeExternalScriptsIfReady): (WebCore::HTMLTokenizer::executeExternalScriptsTimerFired): (WebCore::HTMLTokenizer::continueExecutingExternalScripts):
  • html/HTMLTokenizer.h:
06:13 Changeset [52090] by ossy@webkit.org

Unreviewed trivial buildfix.

Patch by Gabor Loki <loki@inf.u-szeged.hu> on 2009-12-14
Fix crosses initialization of usedPrimaryBlocks for JSValue32

  • runtime/Collector.cpp:

(JSC::Heap::markConservatively):

03:09 Changeset [52089] by zecke@webkit.org

[Qt] Connect QNetworkReply via direct connections.

https://bugs.webkit.org/show_bug.cgi?id=32113

In the past we have switched from auto connections to queued
connections to get some more stability. The problem with the
Queued connections is latency. Even when running simple loading
tests the time it takes to runs highly varies and by using
direct connections these tests produce a stable result.

We will have to look out for crashes coming from within the
QNetworkAccessManager due this patch and address them then.

03:03 Changeset [52088] by philn@webkit.org

2009-12-14 Philippe Normand <pnormand@igalia.com>

Rubber-stamped by Gustavo Noronha Silva.

media/video-played-collapse.html is failing
https://bugs.webkit.org/show_bug.cgi?id=31434

  • platform/gtk/Skipped: Give this test another chance, we believe the bug above has been fixed in a previous commit.
02:51 Changeset [52087] by kov@webkit.org

Reviewed by Xan Lopez.

New setting to allow applications to completely supress the
default context menu that is generated by WebKit. This allows them
to still pass the vent to the default handler without needing to
use ugly hacks such as handling populate-popup, and removing all
the items from the default context menu.

  • webkit/webkitwebsettings.cpp: (webkit_web_settings_class_init): (webkit_web_settings_set_property): (webkit_web_settings_get_property): (webkit_web_settings_copy):
  • webkit/webkitwebview.cpp: (webkit_web_view_forward_context_menu_event):
02:48 Changeset [52086] by zecke@webkit.org

[Qt] QtWebKit should use QImageReader::setQuality(49)

https://bugs.webkit.org/show_bug.cgi?id=30740

Using QImageReader::setQuality(49) will influence the JPEG
decoder of Qt to use the JDCT_IFAST routines for decompressing. This
is showing a 5% improvement in a reduction that decodes all images
used during the page loading test. The libjpeg documentation mentions
that the IFAST routine is not as accurate as other routines but in
our tests we couldn't see a difference.

  • platform/graphics/qt/ImageDecoderQt.cpp:

(WebCore::ImageDecoderQt::setData):

02:01 Changeset [52085] by philn@webkit.org

emit volumeChanged and durationChanged upon notification from GStreamer.

01:56 Changeset [52084] by ossy@webkit.org

GCC 4.3.x warning fixed. Suggested parantheses added.
warning: ../../../JavaScriptCore/wtf/StdLibExtras.h:77: warning: suggest parentheses around + or - in operand of &

Reviewed by Simon Hausmann.

  • wtf/StdLibExtras.h:

(WTF::bitCount):

00:39 Changeset [52083] by adele@apple.com

RS by Maciej Stachowiak.

Adding platform specific results for these test as part of reverting changes from http://trac.webkit.org/changeset/49585 to fix:
<rdar://problem/7382815> REGRESSION (4.0.4-TOT): Home photo slider is too narrow at http://www.ziprealty.com/
https://bugs.webkit.org/show_bug.cgi?id=32078

  • platform/win/fast/css/large-number-round-trip-expected.txt: Added.
  • platform/win/fast/css/opacity-float-expected.txt: Added.
00:13 Changeset [52082] by ggaren@apple.com

JavaScriptCore: Changed GC from mark-sweep to mark-allocate.

Reviewed by Sam Weinig.

Added WeakGCMap to keep WebCore blissfully ignorant about objects that
have become garbage but haven't run their destructors yet.

1% SunSpider speedup.
7.6% v8 speedup (37% splay speedup).
17% speedup on bench-alloc-nonretained.js.
18% speedup on bench-alloc-retained.js.

  • API/JSBase.cpp:

(JSGarbageCollect):

files.

  • debugger/Debugger.cpp:

(JSC::Debugger::recompileAllJSFunctions): Updated to use the Collector
iterator abstraction.

  • jsc.cpp:

(functionGC): Updated for rename.

  • runtime/Collector.cpp: Slightly reduced the number of allocations per

collection, so that small workloads only allocate on collector block,
rather than two.

(JSC::Heap::Heap): Updated to use the new allocateBlock function.

(JSC::Heap::destroy): Updated to use the new freeBlocks function.

(JSC::Heap::allocateBlock): New function to initialize a block when
allocating it.

(JSC::Heap::freeBlock): Consolidated the responsibility for running
destructors into this function.

(JSC::Heap::freeBlocks): Updated to use freeBlock.

(JSC::Heap::recordExtraCost): Sweep the heap in this reporting function,
so that allocation, which is more common, doesn't have to check extraCost.

(JSC::Heap::heapAllocate): Run destructors right before recycling a
garbage cell. This has better cache utilization than a separate sweep phase.

(JSC::Heap::resizeBlocks):
(JSC::Heap::growBlocks):
(JSC::Heap::shrinkBlocks): New set of functions for managing the size of
the heap, now that the heap doesn't maintain any information about its
size.

(JSC::isPointerAligned):
(JSC::isHalfCellAligned):
(JSC::isPossibleCell):
(JSC::isCellAligned):
(JSC::Heap::markConservatively): Cleaned up this code a bit.

(JSC::Heap::clearMarkBits):
(JSC::Heap::markedCells): Some helper functions for examining the the mark
bitmap.

(JSC::Heap::sweep): Simplified this function by using a DeadObjectIterator.

(JSC::Heap::markRoots): Reordered some operations for clarity.

(JSC::Heap::objectCount):
(JSC::Heap::addToStatistics):
(JSC::Heap::statistics): Rewrote these functions to calculate an object
count on demand, since the heap doesn't maintain this information by
itself.

(JSC::Heap::reset): New function for resetting the heap once we've
exhausted heap space.

(JSC::Heap::collectAllGarbage): This function matches the old collect()
behavior, but it's now an uncommon function used only by API.

  • runtime/Collector.h:

(JSC::CollectorBitmap::count):
(JSC::CollectorBitmap::isEmpty): Added some helper functions for managing
the collector mark bitmap.

(JSC::Heap::reportExtraMemoryCost): Changed reporting from cell equivalents
to bytes, so it's easier to understand.

  • runtime/CollectorHeapIterator.h:

(JSC::CollectorHeapIterator::CollectorHeapIterator):
(JSC::CollectorHeapIterator::operator!=):
(JSC::CollectorHeapIterator::operator*):
(JSC::CollectorHeapIterator::advance):
(JSC::::LiveObjectIterator):
(JSC::::operator):
(JSC::::DeadObjectIterator):
(JSC::::ObjectIterator): New iterators for encapsulating details about
heap layout, and what's live and dead on the heap.

  • runtime/JSArray.cpp:

(JSC::JSArray::putSlowCase):
(JSC::JSArray::increaseVectorLength): Delay reporting extra cost until
we're fully constructed, so the heap mark phase won't visit us in an
invalid state.

  • runtime/JSCell.h:

(JSC::JSCell::):
(JSC::JSCell::createDummyStructure):
(JSC::JSCell::JSCell):

  • runtime/JSGlobalData.cpp:

(JSC::JSGlobalData::JSGlobalData):

  • runtime/JSGlobalData.h: Added a dummy cell to simplify allocation logic.
  • runtime/JSString.h:

(JSC::jsSubstring): Don't report extra cost for substrings, since they
share a buffer that's already reported extra cost.

  • runtime/Tracing.d:
  • runtime/Tracing.h: Changed these dtrace hooks not to report object

counts, since they're no longer cheap to compute.

  • runtime/UString.h: Updated for renames.
  • runtime/WeakGCMap.h: Added.

(JSC::WeakGCMap::isEmpty):
(JSC::WeakGCMap::uncheckedGet):
(JSC::WeakGCMap::uncheckedBegin):
(JSC::WeakGCMap::uncheckedEnd):
(JSC::::get):
(JSC::::take):
(JSC::::set):
(JSC::::uncheckedRemove): Mentioned above.

  • wtf/StdLibExtras.h:

(WTF::bitCount): Added a bit population count function, so the heap can
count live objects to fulfill statistics questions.

JavaScriptGlue: Changed GC from mark-sweep to mark-allocate.

Reviewed by Sam Weinig.

  • JavaScriptGlue.cpp:

(JSCollect): Updated for rename. Fixed a bug where JSGlue would not check
to avoid nested GC calls.

WebCore: Changed GC from mark-sweep to mark-allocate.

Reviewed by Sam Weinig.

  • ForwardingHeaders/runtime/WeakGCMap.h: Added.
  • bindings/js/GCController.cpp:

(WebCore::collect):
(WebCore::GCController::gcTimerFired):
(WebCore::GCController::garbageCollectNow): Updated for rename.

  • bindings/js/JSDOMBinding.cpp:

(WebCore::removeWrappers):
(WebCore::hasCachedDOMObjectWrapperUnchecked):
(WebCore::hasCachedDOMObjectWrapper):
(WebCore::hasCachedDOMNodeWrapperUnchecked):
(WebCore::forgetDOMObject):
(WebCore::forgetDOMNode):
(WebCore::isObservableThroughDOM):
(WebCore::markDOMNodesForDocument):
(WebCore::markDOMObjectWrapper):
(WebCore::markDOMNodeWrapper):

  • bindings/js/JSDOMBinding.h: Changed DOM wrapper maps to be WeakGCMaps.

Don't ASSERT that an item must be in the WeakGCMap when its destructor
runs, since it might have been overwritten in the map first.

  • bindings/js/JSDocumentCustom.cpp:

(WebCore::toJS): Changed Document from a DOM object wrapper to a DOM node
wrapper, to simplify some code.

  • bindings/js/JSInspectedObjectWrapper.cpp:

(WebCore::JSInspectedObjectWrapper::JSInspectedObjectWrapper):
(WebCore::JSInspectedObjectWrapper::~JSInspectedObjectWrapper):

  • bindings/js/JSInspectorCallbackWrapper.cpp: Use a WeakGCMap for these

wrappers.

  • bindings/js/JSNodeCustom.cpp:

(WebCore::JSNode::markChildren): Updated for WeakGCMap and Document using
a DOM node wrapper instead of a DOM object wrapper.

  • bindings/js/JSSVGPODTypeWrapper.h:

(WebCore::JSSVGDynamicPODTypeWrapperCache::wrapperMap):
(WebCore::JSSVGDynamicPODTypeWrapperCache::lookupOrCreateWrapper):
(WebCore::JSSVGDynamicPODTypeWrapperCache::forgetWrapper):
(WebCore::::~JSSVGDynamicPODTypeWrapper): Shined a small beam of sanity light
on this code. Use hashtable-based lookup in JSSVGPODTypeWrapper.h instead
of linear lookup through iteration, since that's what hashtables were
invented for. Make JSSVGPODTypeWrapper.h responsible for reomving itself
from the table, instead of its JS wrapper, to decouple these objects from
GC, and because these objects are refCounted, not solely owned by their
JS wrappers.

  • bindings/scripts/CodeGeneratorJS.pm:
  • dom/Document.h: Adopted changes above.

12/13/09:

23:01 Changeset [52081] by adele@apple.com

RS by Maciej Stachowiak.

Readding platform specific results for this test as part of reverting changes from http://trac.webkit.org/changeset/49585 to fix:
<rdar://problem/7382815> REGRESSION (4.0.4-TOT): Home photo slider is too narrow at http://www.ziprealty.com/
https://bugs.webkit.org/show_bug.cgi?id=32078

  • platform/mac/fast/css/opacity-float-expected.txt: Added.
21:01 Changeset [52080] by abarth@webkit.org

2009-12-13 Charles Reis <creis@chromium.org>

Reviewed by Adam Barth.

Refactor some security code out of V8 bindings
https://bugs.webkit.org/show_bug.cgi?id=32326

No new tests. There should be no functionality changes in this patch,
since it is only refactoring code.

  • WebCore.gyp/WebCore.gyp:
  • WebCore.gypi:
  • bindings/BindingSecurity.h: Added. (WebCore::BindingSecurity::BindingSecurity): (WebCore::::canAccessWindow): (WebCore::::canAccessFrame): (WebCore::::checkNodeSecurity):
  • bindings/BindingSecurityBase.cpp: Added. (WebCore::BindingSecurityBase::getDOMWindow): (WebCore::BindingSecurityBase::getFrame): (WebCore::BindingSecurityBase::canAccessWindow):
  • bindings/BindingSecurityBase.h: Added.
  • bindings/GenericBinding.h: Added. (WebCore::):
  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/v8/ScriptController.cpp: (WebCore::ScriptController::isSafeScript):
  • bindings/v8/V8Binding.h:
  • bindings/v8/V8BindingState.cpp: Added. (WebCore::::Only): (WebCore::::getActiveWindow): (WebCore::::immediatelyReportUnsafeAccessTo):
  • bindings/v8/V8BindingState.h: Added. (WebCore::):
  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::reportUnsafeAccessTo): (WebCore::reportUnsafeJavaScriptAccess):
  • bindings/v8/V8Proxy.h: (WebCore::V8Proxy::):
  • bindings/v8/custom/V8CustomBinding.cpp: (WebCore::allowSettingFrameSrcToJavascriptUrl): (WebCore::INDEXED_ACCESS_CHECK): (WebCore::NAMED_ACCESS_CHECK):
  • bindings/v8/custom/V8DOMWindowCustom.cpp: (WebCore::V8Custom::WindowSetTimeoutImpl): (WebCore::ACCESSOR_GETTER): (WebCore::ACCESSOR_SETTER): (WebCore::CALLBACK_FUNC_DECL): (WebCore::V8Custom::ClearTimeoutImpl): (WebCore::NAMED_ACCESS_CHECK): (WebCore::INDEXED_ACCESS_CHECK):
  • bindings/v8/custom/V8LocationCustom.cpp: (WebCore::ACCESSOR_GETTER): (WebCore::CALLBACK_FUNC_DECL): (WebCore::INDEXED_ACCESS_CHECK): (WebCore::NAMED_ACCESS_CHECK):

2009-12-13 Charles Reis <creis@chromium.org>

Reviewed by Adam Barth.

Refactor some security code out of V8 bindings
https://bugs.webkit.org/show_bug.cgi?id=32326

  • src/WebBindings.cpp: (WebKit::getDragDataImpl):
19:10 Changeset [52079] by zecke@webkit.org

[Qt] Implement initializeMaximumHTTPConnectionCountPerHost.

Qt allows to have three connections per host on Symbian, for
every other platform the limit is set to six connections per
host but the default implementation will only schedule four
connections per host.
In a manual test we seem to get the best loading speed by
going with Qt's limit and schedule two more jobs to allow Qt
to prepare them.

  • platform/network/ResourceRequestBase.cpp:
  • platform/network/qt/ResourceRequestQt.cpp:

(WebCore::initializeMaximumHTTPConnectionCountPerHost):

18:38 Changeset [52078] by mjs@apple.com

2009-12-13 Eric Seidel <eric@webkit.org>

Reviewed by Gavin Barraclough.

string-base64 test does not compute a valid base64 string
http://bugs.webkit.org/show_bug.cgi?id=16806

  • tests/string-base64.js: change str[i] to str.charCodeAt(i)

2009-12-13 Maciej Stachowiak <mjs@apple.com>

Reviewed by Gavin Barraclough.

SunSpider/tests/string-base64.js does not compute a valid base64 encoded string
https://bugs.webkit.org/show_bug.cgi?id=16806

Based on a patch by Eric Seidel.


Fix the base64 computation to actually compute correct results. The impact on runtime of
the test is pretty small, but noticeable for some browsers. But at least it's not
doing a wrong and meaningless computation any more.


  • tests/sunspider-0.9.1/string-base64.js: (): (base64ToString):
18:26 Changeset [52077] by mjs@apple.com

Fix minor problem in sunspider-compare-results which left it broken.

Reviewed by Gavin Barraclaugh.

  • Scripts/sunspider-compare-results: Declare $parseonly
17:32 Changeset [52076] by barraclough@apple.com
  • runtime/JSGlobalObjectFunctions.cpp:

Reviewed by NOBODY (speculative Windows build fix).

16:27 Changeset [52075] by barraclough@apple.com

https://bugs.webkit.org/show_bug.cgi?id=32496
Switch remaining cases of string construction to use StringBuilder.
Builds strings using a vector rather than using string append / addition.

Reviewed by Sam Weinig.

(JSC::FunctionExecutable::paramString):

  • runtime/FunctionConstructor.cpp:

(JSC::constructFunction):

  • runtime/JSGlobalObjectFunctions.cpp:

(JSC::encode):
(JSC::decode):
(JSC::globalFuncEscape):
(JSC::globalFuncUnescape):

  • runtime/JSONObject.cpp:

(JSC::Stringifier::stringify):
(JSC::Stringifier::indent):

  • runtime/JSString.h:
  • runtime/LiteralParser.cpp:

(JSC::LiteralParser::Lexer::lexString):

  • runtime/NumberPrototype.cpp:

(JSC::integerPartNoExp):
(JSC::numberProtoFuncToFixed):
(JSC::numberProtoFuncToPrecision):

  • runtime/Operations.h:

(JSC::jsString):

  • runtime/StringPrototype.cpp:

(JSC::substituteBackreferencesSlow):
(JSC::substituteBackreferences):
(JSC::stringProtoFuncConcat):

16:14 Changeset [52074] by mitz@apple.com

Add a style guideline concerning floating point literals
https://bugs.webkit.org/show_bug.cgi?id=32497

Reviewed by Sam Weinig.

  • coding/coding-style.html:
16:12 Changeset [52073] by mitz@apple.com

<rdar://problem/7341364> Crash at HTMLParser::popOneBlockCommon() after
handling misnested residual style tags

Reviewed by Simon Fraser.

WebCore:

Test: fast/parser/residual-style-close-ref-clone.html

  • html/HTMLParser.cpp:

(WebCore::HTMLParser::handleResidualStyleCloseTagAcrossBlocks): Gave the
block stack a strong reference to the cloned residual style element.

LayoutTests:

  • fast/parser/residual-style-close-ref-clone-expected.txt: Added.
  • fast/parser/residual-style-close-ref-clone.html: Added.
15:53 Changeset [52072] by mjs@apple.com

Fixing commit error...

I accidentally committed my last patch in a form that broke Web-hosted SunSpider. Fixing.

  • resources/driver-TEMPLATE.html:
15:00 Changeset [52071] by adele@apple.com

Reverting changes from http://trac.webkit.org/changeset/49585 to fix:
<rdar://problem/7382815> REGRESSION (4.0.4-TOT): Home photo slider is too narrow at http://www.ziprealty.com/
https://bugs.webkit.org/show_bug.cgi?id=32078

Reviewed by Sam Weinig.

WebCore:

  • css/CSSPrimitiveValue.cpp: (WebCore::CSSPrimitiveValue::cssText):

LayoutTests:

  • fast/css/large-number-round-trip-expected.txt:
  • fast/css/large-number-round-trip.html: Updated to run in Firefox.
  • fast/css/opacity-float-expected.txt:
  • fast/css/opacity-float.html:
  • transforms/2d/compound-2d-transforms-expected.txt:
  • transforms/2d/compound-2d-transforms.html:
14:48 Changeset [52070] by kov@webkit.org

Reviewed by Xan Lopez.

Also give the right click event to the frame's event handler, if
there WebCore decides to not generate the default context
menu. This fixes some pages that handle right-button clicks for
things other than context menu creation.

  • webkit/webkitwebview.cpp: (webkit_web_view_forward_context_menu_event):
14:34 Changeset [52069] by eric@webkit.org

2009-12-13 Benjamin Poulain <benjamin.poulain@nokia.com>

Reviewed by Simon Hausmann.

Add a test in Qt for https://bugs.webkit.org/show_bug.cgi?id=29005
https://bugs.webkit.org/show_bug.cgi?id=29008

  • tests/qwebframe/tst_qwebframe.cpp:
12:31 Changeset [52068] by mjs@apple.com

2009-12-13 Maciej Stachowiak <mjs@apple.com>

Reviewed by Sam Weinig.

Use JSON.parse instead of eval for Web-hosted SunSpider results processing
https://bugs.webkit.org/show_bug.cgi?id=32490

  • hosted/json2.js: Added. Incorporated from json.org
  • resources/driver-TEMPLATE.html: Fix an HTML validation bug I noticed.
  • resources/results-TEMPLATE.html: Use JSON.parse instead of eval to process results.
05:33 Changeset [52067] by jeremy@chromium.org

Reviewed by Eric Seidel.

Add code to allow toggling ATSUI/Core Text rendering at runtime in ComplexTextController.
https://bugs.webkit.org/show_bug.cgi?id=31802

The goal here is to allow for a zero runtime hit for ports that decide to select
the API at compile time.
When both USE(ATSUI) and USE(CORE_TEXT) are true, the API is toggled
at runtime. Core Text is used for OS Versions >= 10.6.

No test since this is already covered by existing pixel tests.

  • platform/graphics/SimpleFontData.h: Change #ifdef to define getNSFont() on Chrome/Mac .
  • platform/graphics/mac/ComplexTextController.cpp:

(WebCore::shouldUseATSUIAPI): Cache decision about which API to use.
(WebCore::ComplexTextController::ComplexTextRun::indexAt):
(WebCore::ComplexTextController::collectComplexTextRunsForCharacters): Stub function to call through
to ATSUI/Core text variants.
(WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun): The actual work for object construction
is delegate to createTextRunFromFontDataATSUI/createTextRunFromFontDataCoreText.

  • platform/graphics/mac/ComplexTextController.h:
  • platform/graphics/mac/ComplexTextControllerATSUI.cpp:

(WebCore::ComplexTextController::ComplexTextRun::overrideLayoutOperation): Small style fix.
(WebCore::ComplexTextController::ComplexTextRun::createTextRunFromFontDataATSUI):
(WebCore::ComplexTextController::collectComplexTextRunsForCharactersATSUI):

  • platform/graphics/mac/ComplexTextControllerCoreText.cpp:

(WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun):
(WebCore::ComplexTextController::ComplexTextRun::createTextRunFromFontDataCoreText):
(WebCore::ComplexTextController::collectComplexTextRunsForCharactersCoreText):

04:11 reni_regexp.pdf attached to QtWebKit/CodeCamp200912 by reni@inf.u-szeged.hu
03:32 Changeset [52066] by christian@webkit.org

2009-12-13 Christian Dywan <christian@twotoasts.de>

Reviewed by Holger Freyther.

http://bugs.webkit.org/show_bug.cgi?id=31014
[Gtk] Build fails with gtk+-2.10 (no tested with older releases)

Guard GTK+ 2.12 specific tooltip code with #ifdef.

  • WebCoreSupport/ChromeClientGtk.cpp: (WebKit::ChromeClient::mouseDidMoveOverElement):
01:31 Changeset [52065] by eric@webkit.org

2009-12-13 Simon Hausmann <hausmann@webkit.org>

Reviewed by Holger Freyther.

[Qt] Re-enable QWebView::renderHints property for Qt for Symbian

https://bugs.webkit.org/show_bug.cgi?id=28273

The bug in Qt's moc that triggered a linking error when declaring this
property has been fixed and we can remove the workaround.

  • Api/qwebview.h:
00:44 Changeset [52064] by eric.carlson@apple.com

2009-12-13 Eric Carlson <eric.carlson@apple.com>

No review, rolling out r52059 until I can generate new Tiger
test results.

  • rendering/RenderThemeMac.mm: (WebCore::mediaControllerTheme):

2009-12-13 Eric Carlson <eric.carlson@apple.com>

No review, rolling out r52059 until I can generate new Tiger test
results.

  • platform/mac-leopard/Skipped:
  • platform/mac-leopard/media/controls-after-reload-expected.txt:
  • platform/mac-leopard/media/controls-drag-timebar-expected.txt:
  • platform/mac-leopard/media/controls-strict-expected.txt:
  • platform/mac-leopard/media/controls-styling-expected.txt:
  • platform/mac-leopard/media/video-controls-rendering-expected.checksum:
  • platform/mac-leopard/media/video-controls-rendering-expected.txt:
  • platform/mac-leopard/media/video-display-toggle-expected.checksum:
  • platform/mac-leopard/media/video-display-toggle-expected.txt:
  • platform/mac-leopard/media/video-no-audio-expected.txt:
  • platform/mac-leopard/media/video-zoom-controls-expected.checksum:
  • platform/mac-leopard/media/video-zoom-controls-expected.txt:
  • platform/mac/media/audio-controls-rendering-expected.checksum:
  • platform/mac/media/video-zoom-expected.checksum:

12/12/09:

20:45 Changeset [52063] by weinig@apple.com

Rubber-stamped by Dan Bernstein.

Fix the EXPERIMENTAL_SINGLE_VIEW_MODE build.

  • Configurations/WebCore.xcconfig:
  • page/EventHandler.cpp:

(WebCore::EventHandler::EventHandler):
(WebCore::EventHandler::clearDragState):

  • page/mac/DragControllerMac.mm:

(WebCore::DragController::isCopyKeyDown):

  • page/mac/EventHandlerMac.mm:

(WebCore::EventHandler::eventActivatedView):

  • platform/PlatformKeyboardEvent.h:
  • platform/PlatformMouseEvent.h:
  • platform/PlatformWheelEvent.h:

(WebCore::PlatformWheelEvent::PlatformWheelEvent):

  • platform/ScrollView.cpp:

(WebCore::ScrollView::wheelEvent):

  • platform/mac/KeyEventMac.mm:

(WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent):

  • platform/mac/PlatformMouseEventMac.mm:

(WebCore::PlatformMouseEvent::PlatformMouseEvent):

  • platform/mac/PopupMenuMac.mm:

(WebCore::PopupMenu::show):

  • platform/mac/WidgetMac.mm:
20:26 Changeset [52062] by mjs@apple.com

2009-12-12 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver Hunt.

Enable Web-hosted version of SunSpider to handle multiple versions
https://bugs.webkit.org/show_bug.cgi?id=32478

  • make-hosted: Now generate the hosted version to be able to run both the 0.9 and the 0.9.1 test suites.
  • hosted: Modified property svn:ignore.
  • hosted/sunspider-driver.html: Removed.
  • hosted/sunspider-results.html: Removed.
  • hosted/sunspider.html:
  • hosted/versions.html: Added.
  • resources/TEMPLATE.html:
  • resources/driver-TEMPLATE.html: Copied from hosted/sunspider-driver.html.
  • resources/driver-TEMPLATE.html: Copied from hosted/sunspider-results.html.
19:59 Changeset [52061] by weinig@apple.com

Rubber-stamped by Dan Bernstein.

Fix the !ENABLE(VIDEO) build.

  • rendering/RenderLayerCompositor.cpp:

(WebCore::RenderLayerCompositor::requiresCompositingForVideo):

19:56 Changeset [52060] by weinig@apple.com

Rubber-stamped by Dan Bernstein.

Fix the !ENABLE(DATABASE) build.

  • storage/SQLTransactionClient.cpp:
  • storage/SQLTransactionClient.h:
  • storage/SQLTransactionCoordinator.cpp:
  • storage/SQLTransactionCoordinator.h:
  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::stopLoading):

19:28 Changeset [52059] by eric.carlson@apple.com

2009-12-12 Eric Carlson <eric.carlson@apple.com>

Reviewed by Dan Bernstein.

<rdar://problem/6293979> Adopt shared control drawing for <video> controls on Leopard and Windows

Use MediaControllerThemeQuickTime on Leopard.

  • rendering/RenderThemeMac.mm: (WebCore::mediaControllerTheme):

2009-12-12 Eric Carlson <eric.carlson@apple.com>

Reviewed by Dan Bernstein.

<rdar://problem/6293979> Adopt shared control drawing for <video> controls on Leopard and Windows

Regenerate layout and pixel test results for the new UI.

  • platform/mac-leopard/Skipped:
  • platform/mac-leopard/media/controls-after-reload-expected.txt:
  • platform/mac-leopard/media/controls-drag-timebar-expected.txt:
  • platform/mac-leopard/media/controls-strict-expected.txt:
  • platform/mac-leopard/media/controls-styling-expected.txt:
  • platform/mac-leopard/media/video-controls-rendering-expected.checksum:
  • platform/mac-leopard/media/video-controls-rendering-expected.png:
  • platform/mac-leopard/media/video-controls-rendering-expected.txt:
  • platform/mac-leopard/media/video-display-toggle-expected.checksum:
  • platform/mac-leopard/media/video-display-toggle-expected.png:
  • platform/mac-leopard/media/video-display-toggle-expected.txt:
  • platform/mac-leopard/media/video-no-audio-expected.txt:
  • platform/mac-leopard/media/video-zoom-controls-expected.checksum:
  • platform/mac-leopard/media/video-zoom-controls-expected.png:
  • platform/mac-leopard/media/video-zoom-controls-expected.txt:
  • platform/mac/media/audio-controls-rendering-expected.checksum:
  • platform/mac/media/audio-controls-rendering-expected.png:
  • platform/mac/media/video-zoom-expected.checksum:
  • platform/mac/media/video-zoom-expected.png:
19:02 Changeset [52058] by eric@webkit.org

2009-12-12 Kenneth Russell <kbr@google.com>

Reviewed by Oliver Hunt.

Errors in WebGL[T]Array set() and get() methods
https://bugs.webkit.org/show_bug.cgi?id=32456

Test: fast/canvas/webgl/bug-32456.html

  • fast/canvas/webgl/array-unit-tests.html:
  • fast/canvas/webgl/bug-32456-expected.txt: Added.
  • fast/canvas/webgl/bug-32456.html: Added.

2009-12-12 Kenneth Russell <kbr@google.com>

Reviewed by Oliver Hunt.

Errors in WebGL[T]Array set() and get() methods
https://bugs.webkit.org/show_bug.cgi?id=32456

Test: fast/canvas/webgl/bug-32456.html

  • bindings/js/JSWebGLFloatArrayCustom.cpp: (WebCore::JSWebGLFloatArray::set):
  • bindings/js/JSWebGLIntArrayCustom.cpp: (WebCore::JSWebGLIntArray::set):
  • bindings/js/JSWebGLShortArrayCustom.cpp: (WebCore::JSWebGLShortArray::set):
  • bindings/js/JSWebGLUnsignedByteArrayCustom.cpp: (WebCore::JSWebGLUnsignedByteArray::set):
  • bindings/js/JSWebGLUnsignedIntArrayCustom.cpp: (WebCore::JSWebGLUnsignedIntArray::set):
  • bindings/js/JSWebGLUnsignedShortArrayCustom.cpp: (WebCore::JSWebGLUnsignedShortArray::set):
  • bindings/v8/custom/V8WebGLArrayCustom.h: (WebCore::setWebGLArray):
  • html/canvas/WebGLFloatArray.idl:
  • html/canvas/WebGLUnsignedIntArray.idl:
18:18 Changeset [52057] by mjs@apple.com

2009-12-12 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver Hunt.

Give command-line sunspider the ability to handle multiple suites and versions
https://bugs.webkit.org/show_bug.cgi?id=32477

  • Scripts/run-sunspider: Updated for changes to command-line parameters.
  • Scripts/sunspider-compare-results: ditto

2009-12-12 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver Hunt.

Give command-line sunspider the ability to handle multiple suites and versions
https://bugs.webkit.org/show_bug.cgi?id=32477


Each suite/version's set of tests are now in a separate subdirectory of tests/, with a file named LIST
enumerating the tests. Current suites are:


sunspider-0.9 - riginal SunSpider content
sunspider-0.9.1 - SunSpider suite with a few tests fixed (and a few more to come)
ubench - "ubench" microbenchmark suite
v8-v4 - version 4 of the v8 benchmark
parse-only - some parser tests


Sunspider can now be run with the --suite= parameter to select a
suite. The default is --suite=sunspider-0.9.

  • .: Modified property svn:ignore to ignore new results directories.
  • make-hosted: Generate based on sunspider-0.9 suite.
  • resources/TEMPLATE.html: Fix an HTML compliance problem.
  • resources/sunspider-standalone-driver.js: Become aware of suite names.
  • sunspider: Modified to handle multiple suites as described above.
  • sunspider-compare-results: ditto
  • tests/LIST: Moved to tests/sunspider-0.9/ and tests/sunspider-0.9.1/
  • tests/LIST-PARSE-ONLY: Moved to parse-only/LIST
  • tests/LIST-UBENCH: Removed.
  • tests/LIST-V8: Removed.
  • tests/3d-cube.js: Moved to tests/sunspider-0.9 and tests/sunspider-0.9.1
  • tests/3d-morph.js: ditto
  • tests/3d-raytrace.js: ditto
  • tests/access-binary-trees.js: ditto
  • tests/access-fannkuch.js: ditto
  • tests/access-nbody.js: ditto
  • tests/access-nsieve.js: ditto
  • tests/bitops-3bit-bits-in-byte.js: ditto
  • tests/bitops-bits-in-byte.js: ditto
  • tests/bitops-bitwise-and.js: ditto
  • tests/bitops-nsieve-bits.js: ditto
  • tests/controlflow-recursive.js: ditto
  • tests/crypto-aes.js: ditto
  • tests/crypto-md5.js: ditto
  • tests/crypto-sha1.js: ditto
  • tests/date-format-tofte.js: ditto
  • tests/date-format-xparb.js: ditto
  • tests/math-cordic.js: ditto
  • tests/math-partial-sums.js: ditto
  • tests/math-spectral-norm.js: ditto
  • tests/regexp-dna.js: ditto
  • tests/string-base64.js: ditto
  • tests/string-fasta.js: ditto
  • tests/string-tagcloud.js: ditto
  • tests/string-unpack-code.js: ditto
  • tests/string-validate-input.js: ditto
  • tests/sunspider-0.9/: Added
  • tests/sunspider-0.9/*.js: Moved from one level up
  • tests/sunspider-0.9.1: Added.
  • tests/sunspider-0.9.1/*.js: Moved from one level up.
  • tests/v8-crypto.js: Moved to v8-v4/ directory.
  • tests/v8-deltablue.js: ditto
  • tests/v8-earley-boyer.js: ditto
  • tests/v8-raytrace.js: ditto
  • tests/v8-regexp.js: ditto
  • tests/v8-richards.js: ditto
  • tests/v8-splay.js: ditto
  • tests/v8-v4: Added.
15:12 Changeset [52056] by eric@webkit.org

2009-12-12 Kenneth Russell <kbr@google.com>

Reviewed by Oliver Hunt.

Update WebGL tests to released versions
https://bugs.webkit.org/show_bug.cgi?id=32457

  • fast/canvas/webgl/resources/utils3d.js: (initWebGL.gl.console.window.console.log): (initWebGL): (loadShader): (makeBox): (makeSphere): (processLoadObj): (doLoadObj): (doLoadImageTexture):

2009-12-12 Kenneth Russell <kbr@google.com>

Reviewed by Oliver Hunt.

Update WebGL tests to released versions
https://bugs.webkit.org/show_bug.cgi?id=32457

  • manual-tests/webgl/Earth.html:
  • manual-tests/webgl/ManyPlanetsDeep.html:
  • manual-tests/webgl/SpinningBox.html:
  • manual-tests/webgl/TeapotPerPixel.html:
  • manual-tests/webgl/TeapotPerVertex.html:
  • manual-tests/webgl/resources/CanvasMatrix.js: (CanvasMatrix4.prototype.getAsWebGLFloatArray):
  • manual-tests/webgl/resources/utils3d.js: (initWebGL.gl.console.window.console.log): (initWebGL): (loadShader): (makeBox): (makeSphere): (processLoadObj): (doLoadObj): (doLoadImageTexture):
13:29 Changeset [52055] by eric.carlson@apple.com

2009-12-12 Eric Carlson <eric.carlson@apple.com>

Reviewed by Darin Adler.

<rdar://problem/7453726> Pull shared UI code into WebKit

  • WebKitSystemInterface.h:
  • libWebKitSystemInterfaceLeopard.a:
  • libWebKitSystemInterfaceSnowLeopard.a:
  • libWebKitSystemInterfaceTiger.a:

2009-12-12 Eric Carlson <eric.carlson@apple.com>

Reviewed by Darin Adler.

<rdar://problem/7453726> Pull shared UI code into WebKit

Update media tests for small change in timeline slider thumb position

  • platform/mac/media/audio-controls-rendering-expected.txt:
  • platform/mac/media/controls-after-reload-expected.txt:
  • platform/mac/media/controls-styling-expected.txt:
  • platform/mac/media/video-no-audio-expected.txt:
  • platform/mac/media/video-zoom-controls-expected.txt:
11:10 Changeset [52054] by christian@webkit.org

2009-12-12 Christian Dywan <christian@twotoasts.de>

Reviewed by Dirk Schulze.

Remove the include to SVGFEDisplacementMap.h, it's not needed at all.

  • platform/graphics/filters/FEComponentTransfer.h:
10:32 Changeset [52053] by mjs@apple.com

Not reviewed, fixing previous comment.

Remove accidental change to sunspider-compare-results

  • Scripts/sunspider-compare-results:
10:24 Changeset [52052] by christian@webkit.org

2009-12-12 Christian Dywan <christian@twotoasts.de>

Rubber-stamped by Maciej Stachowiak.

Only include SVGFEDisplacementMap.h if SVG is enabled.

  • platform/graphics/filters/FEComponentTransfer.h:
09:52 Changeset [52051] by mjs@apple.com

2009-12-11 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver Hunt.

Unify codegen for forward and backward variants of branches
https://bugs.webkit.org/show_bug.cgi?id=32463

  • jit/JIT.h: (JSC::JIT::emit_op_loop): Implemented in terms of forward variant. (JSC::JIT::emit_op_loop_if_true): ditto (JSC::JIT::emitSlow_op_loop_if_true): ditto (JSC::JIT::emit_op_loop_if_false): ditto (JSC::JIT::emitSlow_op_loop_if_false): ditto (JSC::JIT::emit_op_loop_if_less): ditto (JSC::JIT::emitSlow_op_loop_if_less): ditto
  • jit/JITOpcodes.cpp:
09:14 Changeset [52050] by christian@webkit.org

2009-12-11 Christian Dywan <christian@twotoasts.de>

Reviewed by Eric Seidel.

[GTK] Build with --disable-svg is broken
http://bugs.webkit.org/show_bug.cgi?id=32361

  • webkit/webkitwebframe.cpp: (webkit_web_frame_pause_svg_animation): Fix ENABLE(SVG) guards.
02:20 Changeset [52049] by eric@webkit.org

2009-12-12 Dirk Schulze <krit@webkit.org>

Reviewed by Eric Seidel.

[Cairo][Qt] Alpha blending in SVG filters is wrong
https://bugs.webkit.org/show_bug.cgi?id=32427

This patch is a follow-up to:
http://trac.webkit.org/changeset/51950
It fixes the alpha blending for Cairo and Qt on SVG Filters.

Covered by: * LayoutTests/svg/filters/feGaussianBlur.svg

  • LayoutTests/svg/filters/feComposite.svg and others.
  • platform/graphics/cairo/ImageBufferCairo.cpp: (WebCore::ImageBuffer::platformTransformColorSpace):
  • platform/graphics/qt/ImageBufferQt.cpp: (WebCore::ImageBuffer::platformTransformColorSpace):
02:13 Changeset [52048] by ossy@webkit.org

[Qt] Missing forward entries when dumping back/forward list
https://bugs.webkit.org/show_bug.cgi?id=32470

Failing tests added to skiplist until bug fixed.

  • platform/qt/Skipped:
  • http/tests/navigation/dynamic-iframe-dynamic-form-back-entry.html added
  • http/tests/navigation/parsed-iframe-dynamic-form-back-entry.html added
Note: See TracTimeline for information about the timeline view.