⚠ Archived content — this site is no longer maintained.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

Timeline



Sep 4, 2011:

11:48 PM Changeset in webkit [94514] by commit-queue@webkit.org
  • 4 edits in trunk/Source/JavaScriptCore

REGRESSION (r94452): 20 http/tests tests failing on Qt Linux Release
​https://bugs.webkit.org/show_bug.cgi?id=67562

Patch by Mark Hahnenberg <​mhahnenberg@apple.com> on 2011-09-04
Reviewed by Darin Adler.

Fixing the build (again which was broken by the dtoa patch. Needed
to make sure WTF::double_conversion::initialize() is called for Qt
as well as adding a check for WinCE in dtoa/utils.h

  • runtime/InitializeThreading.cpp:

(JSC::initializeThreadingOnce):

  • wtf/dtoa/cached-powers.cc:
  • wtf/dtoa/utils.h:
11:43 PM Changeset in webkit [94513] by koz@chromium.org
  • 6 edits
    2 deletes in trunk

Unreviewed, rolling out r94510.
​http://trac.webkit.org/changeset/94510
​https://bugs.webkit.org/show_bug.cgi?id=66531

Causes layout test crashes.

Source/WebCore:

  • dom/Document.cpp:

(WebCore::Document::webkitWillEnterFullScreenForElement):
(WebCore::Document::webkitDidExitFullScreenForElement):

  • dom/NodeRenderingContext.cpp:

(WebCore::wrapWithRenderFullScreen):
(WebCore::NodeRendererFactory::createRendererIfNeeded):

  • rendering/RenderFullScreen.cpp:

(RenderFullScreen::createFullScreenStyle):

  • rendering/RenderFullScreen.h:

LayoutTests:

  • plugins/fullscreen-plugins-dont-reload-expected.txt: Removed.
  • plugins/fullscreen-plugins-dont-reload.html: Removed.
10:59 PM Changeset in webkit [94512] by inferno@chromium.org
  • 2 edits in trunk/Source/WebCore

Unreviewed. Compile fix for r94511.

  • dom/Range.cpp:

(WebCore::Range::processContents):
(WebCore::Range::processAncestorsAndTheirSiblings):

10:22 PM Changeset in webkit [94511] by inferno@chromium.org
  • 3 edits
    2 adds in trunk

Crash in Range::processAncestorsAndTheirSiblings.
​https://bugs.webkit.org/show_bug.cgi?id=67556

Reviewed by Ryosuke Niwa.

Source/WebCore:

Create a temporary RefPtr Node vector to keep all the ancestor's
childs so that we don't access removed child nodes.

Test: fast/dom/Range/range-delete-contents-event-fire-crash.html

  • dom/Range.cpp:

(WebCore::Range::processContents):
(WebCore::Range::processAncestorsAndTheirSiblings):

LayoutTests:

Tests that we do not crash when removing contents of
a range from the document.

  • fast/dom/Range/range-delete-contents-event-fire-crash-expected.txt: Added.
  • fast/dom/Range/range-delete-contents-event-fire-crash.html: Added.
9:46 PM Changeset in webkit [94510] by commit-queue@webkit.org
  • 6 edits
    2 adds in trunk

Don't detach elements from the render tree when entering fullscreen mode
​https://bugs.webkit.org/show_bug.cgi?id=66531

Source/WebCore:

This prevents plugin instances from being destroyed and reinstantiated
when entering fullscreen mode.

Patch by Jeremy Apthorp <​jeremya@google.com> on 2011-09-04
Reviewed by Darin Fisher.

Test: plugins/fullscreen-plugins-dont-reload.html

  • dom/Document.cpp:

(WebCore::Document::webkitWillEnterFullScreenForElement):
(WebCore::Document::webkitDidExitFullScreenForElement):

  • dom/NodeRenderingContext.cpp:

(WebCore::NodeRendererFactory::createRendererIfNeeded):

  • rendering/RenderFullScreen.cpp:

(createFullScreenStyle):
(RenderFullScreen::wrapRenderer):
(RenderFullScreen::unwrapRenderer):

  • rendering/RenderFullScreen.h:

LayoutTests:

Test that the plugin is not reloaded when entering fullscreen.

Patch by Jeremy Apthorp <​jeremya@google.com> on 2011-09-04
Reviewed by Darin Fisher.

  • plugins/fullscreen-plugins-dont-reload-expected.txt: Added.
  • plugins/fullscreen-plugins-dont-reload.html: Added.
9:18 PM Changeset in webkit [94509] by ryuan.choi@samsung.com
  • 2 edits
    2 moves in trunk/Source/WebKit/efl

[EFL] Rename ewk_view_(single|tiled).c to ewk_view_(single|tiled).cpp
​https://bugs.webkit.org/show_bug.cgi?id=66542

Use a C++ compiler for these files, so that they can include the C++-using
ewk_private.h.

Reviewed by Daniel Bates.

  • CMakeListsEfl.txt:
  • ewk/ewk_view_single.cpp: Renamed from Source/WebKit/efl/ewk/ewk_view_single.c.

(_ewk_view_single_on_del):
(_ewk_view_single_smart_add):
(_ewk_view_single_smart_backing_store_add):
(_ewk_view_single_smart_resize):
(_ewk_view_4b_move_region_up):
(_ewk_view_4b_move_region_down):
(_ewk_view_4b_move_line_left):
(_ewk_view_4b_move_line_right):
(_ewk_view_4b_move_region_left):
(_ewk_view_4b_move_region_right):
(_ewk_view_4b_move_region):
(_ewk_view_single_scroll_process_single):
(_ewk_view_single_smart_scrolls_process):
(_ewk_view_single_smart_repaints_process):
(_ewk_view_single_smart_zoom_weak_set):
(_ewk_view_single_smart_zoom_weak_smooth_scale_set):
(_ewk_view_single_smart_bg_color_set):
(ewk_view_single_smart_set):
(_ewk_view_single_smart_class_new):
(ewk_view_single_add):

  • ewk/ewk_view_tiled.cpp: Renamed from Source/WebKit/efl/ewk/ewk_view_tiled.c.

(_ewk_view_tiled_render_cb):
(_ewk_view_tiled_updates_process_pre):
(_ewk_view_tiled_smart_backing_store_add):
(_ewk_view_tiled_contents_size_changed_cb):
(_ewk_view_tiled_smart_add):
(_ewk_view_tiled_smart_scrolls_process):
(_ewk_view_tiled_smart_repaints_process):
(_ewk_view_tiled_smart_contents_resize):
(_ewk_view_tiled_smart_zoom_set):
(_ewk_view_tiled_smart_zoom_weak_set):
(_ewk_view_tiled_smart_zoom_weak_smooth_scale_set):
(_ewk_view_tiled_smart_flush):
(_ewk_view_tiled_smart_pre_render_region):
(_ewk_view_tiled_smart_pre_render_relative_radius):
(_ewk_view_tiled_smart_pre_render_cancel):
(_ewk_view_tiled_smart_disable_render):
(_ewk_view_tiled_smart_enable_render):
(ewk_view_tiled_smart_set):
(_ewk_view_tiled_smart_class_new):
(ewk_view_tiled_add):
(ewk_view_tiled_unused_cache_get):
(ewk_view_tiled_unused_cache_set):

5:01 PM Changeset in webkit [94508] by mitz@apple.com
  • 10 edits
    2 adds in trunk

<rdar://problem/10071256> Retain retired custom fonts until the next style recalc

Reviewed by Darin Adler.

Source/WebCore:

Test: fast/css/font-face-used-after-retired.html

During style recalc, existing renderers may reference their old style, including font data.
Allow them to do so safely by keeping retired custom font data around until after style recalc.

  • css/CSSFontFace.cpp:

(WebCore::CSSFontFace::retireCustomFont): Added. Calls through to CSSFontSelector, if the font
face is still part of any segmented font face. Otherwise, deletes the custom font data.

  • css/CSSFontFace.h:
  • css/CSSFontFaceSource.cpp:

(WebCore::CSSFontFaceSource::pruneTable): Changed to call retireCustomFont() instead of deleting
retired font data.

  • css/CSSFontSelector.cpp:

(WebCore::CSSFontSelector::retireCustomFont): Added. Calls through to the Document, if this is
still the active font selector for a document. Otherwise, deletes the custom font data.

  • css/CSSFontSelector.h:
  • css/CSSSegmentedFontFace.cpp:

(WebCore::CSSSegmentedFontFace::pruneTable): Changed to call retireCustomFont() instead of
deleting retired font data.

  • dom/Document.cpp:

(WebCore::Document::~Document): Added a call to deleteRetiredCustomFonts(), in case the Document
is destroyed before getting a chance to recalc style after custom fonts have been retired.
(WebCore::Document::recalcStyle): Added a call to deleteRetiredCustomFonts() after style recalc.
(WebCore::Document::deleteRetiredCustomFonts): Added. Deletes all previously-retired custom font
data.

  • dom/Document.h:

(WebCore::Document::retireCustomFont): Added.

LayoutTests:

  • fast/css/font-face-used-after-retired-expected.txt: Added.
  • fast/css/font-face-used-after-retired.html: Added.
2:24 PM Changeset in webkit [94507] by abarth@webkit.org
  • 3 edits in trunk/Source/WebKit/chromium

[Chromium] Add mutliple to WebPopupMenuInfo for Android
​https://bugs.webkit.org/show_bug.cgi?id=67576

Reviewed by Darin Fisher.

This flag requests that the embedder allow multiple selection for the
popup menu. (See also ​https://bugs.webkit.org/show_bug.cgi?id=67574.)

  • public/WebPopupMenuInfo.h:
  • src/ExternalPopupMenu.cpp:

(WebKit::ExternalPopupMenu::getPopupMenuInfo):

2:19 PM Changeset in webkit [94506] by weinig@apple.com
  • 2 edits in trunk/LayoutTests

Forgot to check in new results for fast/events/event-creation.html

  • fast/events/event-creation-expected.txt:
2:04 PM Changeset in webkit [94505] by weinig@apple.com
  • 12 edits
    3 deletes in trunk

Document.createEvent should support all the interfaces of Event we got
​https://bugs.webkit.org/show_bug.cgi?id=67568

Reviewed by Anders Carlsson.

Source/WebCore:

Updated fast/events/event-creation.html

  • dom/BeforeLoadEvent.h:

(WebCore::BeforeLoadEvent::create):
(WebCore::BeforeLoadEvent::BeforeLoadEvent):
Added empty create.

  • dom/Document.cpp:

(WebCore::Document::createEvent):
Add missing interfaces.

  • dom/HashChangeEvent.h:

(WebCore::HashChangeEvent::create):
(WebCore::HashChangeEvent::HashChangeEvent):
Added empty create.

  • page/SpeechInputEvent.cpp:

(WebCore::SpeechInputEvent::create):
(WebCore::SpeechInputEvent::SpeechInputEvent):
(WebCore::SpeechInputEvent::~SpeechInputEvent):

  • page/SpeechInputEvent.h:

Added empty create.

LayoutTests:

  • fast/events/event-creation.html:

Converted to shouldBe style test, changed prototype check to us instanceof and constructor check,
added enumeration based catchall.

  • fast/events/event-instanceof-expected.txt: Removed.
  • fast/events/event-instanceof.html: Removed.
  • fast/events/script-tests/event-instanceof.js: Removed.

Moved contents to fast/events/event-creation.html.

11:10 AM Changeset in webkit [94504] by abarth@webkit.org
  • 7 edits in trunk/Source

[Chromium] Add memory threshold values to WebKitPlatformSupport.h
​https://bugs.webkit.org/show_bug.cgi?id=67575

Reviewed by Darin Fisher.

Source/WebCore:

Grab these memory thresholds from PlatformSupport rather than hard-coding them.

  • bindings/v8/V8GCController.cpp:

(WebCore::V8GCController::checkMemoryUsage):

  • platform/chromium/PlatformSupport.h:
  • platform/qt/PlatformSupport.h:

(WebCore::PlatformSupport::lowMemoryUsageMB):
(WebCore::PlatformSupport::highMemoryUsageMB):
(WebCore::PlatformSupport::highUsageDeltaMB):

Source/WebKit/chromium:

Instead of hard-coding these values, we now grab them from
WebKitPlatformSupport because Android wishes to configure them on a
per-device basis.

One could view grabing these values from PlatformSupport as a bit of a
layering violation (because they feel like policy), but another view is
that the alternative implementation is to query the OS for various
memory parameters. In either case, getting these values via
PlatformSupport is much easier and consistent with the other memory
related queries in PlatformSupport already.

  • public/WebKitPlatformSupport.h:

(WebKit::WebKitPlatformSupport::lowMemoryUsageMB):
(WebKit::WebKitPlatformSupport::highMemoryUsageMB):
(WebKit::WebKitPlatformSupport::highUsageDeltaMB):

  • src/PlatformSupport.cpp:

(WebCore::PlatformSupport::lowMemoryUsageMB):
(WebCore::PlatformSupport::highMemoryUsageMB):
(WebCore::PlatformSupport::highUsageDeltaMB):

11:04 AM Changeset in webkit [94503] by loislo@chromium.org
  • 2 edits in trunk/LayoutTests

2011-09-04 Ilya Tikhonovsky <​loislo@chromium.org>

Unreviewed fix for chromium expectations.
The expectations for http/tests/security/xssAuditor/dom-write-innerHTML.html
were in conflict. The first was for LINUX the second was for DEBUG. I just added RELEASE flag for LINUX one.

  • platform/chromium/test_expectations.txt:
10:51 AM Changeset in webkit [94502] by kevino@webkit.org
  • 1 edit
    5 adds in trunk/Source/WebCore

[wx] Unreviewed build fix. Add new / moved files missing from last commit.

9:49 AM Changeset in webkit [94501] by kevino@webkit.org
  • 19 edits in trunk

[wx] Enable wxWebKit to run using the wxGC Cairo backend on platforms other than GTK.
​https://bugs.webkit.org/show_bug.cgi?id=67577

Reviewed by Kevin Ollivier.

Sep 3, 2011:

10:43 PM Changeset in webkit [94500] by fpizlo@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

ThunkGenerators does not convert positive double zero into integer zero
​https://bugs.webkit.org/show_bug.cgi?id=67553

Reviewed by Gavin Barraclough.

This is an 0.5% speed-up on V8 and neutral elsewhere.

  • jit/SpecializedThunkJIT.h:

(JSC::SpecializedThunkJIT::returnDouble):

7:58 PM Changeset in webkit [94499] by Laszlo Gombos
  • 8 edits in trunk/Source

REGRESSION (r86268): Fix for qt_networkAccessAllowed()
​https://bugs.webkit.org/show_bug.cgi?id=67570

Reviewed by Noam Rosenthal.

Source/WebCore:

No new tests as this change only removes dead code.
Support for Qt 4.6 has been removed a while back.

  • WebCore.pri:
  • WebCore.pro:
  • features.pri:
  • platform/network/NetworkStateNotifier.h:
  • platform/network/qt/NetworkStateNotifierQt.cpp:

Source/WebKit/qt:

Replace the ENABLE(QT_BEARER) guard with
!defined(QT_NO_BEARERMANAGEMENT) which is true by
default in Qt 4.7 or later versions.

  • Api/qwebsettings.cpp:

(qt_networkAccessAllowed):

4:31 PM Changeset in webkit [94498] by rniwa@webkit.org
  • 1 edit in trunk/LayoutTests/platform/mac/Skipped

Remove the corresponding entry from the skipped list.

4:30 PM Changeset in webkit [94497] by rniwa@webkit.org
  • 3 edits
    2 copies in trunk

REGRESSION(r94274): selection-change-closes-typing.html fails
​https://bugs.webkit.org/show_bug.cgi?id=67377

Reviewed by Kent Tamura.

Source/WebCore:

The problem was that when the shadow DOM is updated by setInnerTextValue, WebKit layer detects the selection
change and calls confirmCompositionWithoutDisturbingSelection, which in turn modifies the shadow DOM by
inserting text.

Fixed the bug by not inserting text in confirmCompositionWithoutDisturbingSelection. It turned out that this
function is only used to cancel composition but never to confirming composition and restoring selection.

Test: platform/mac/editing/input/selection-change-closes-typing-2.html

  • editing/Editor.cpp:

(WebCore::Editor::confirmCompositionWithoutDisturbingSelection):
(WebCore::Editor::confirmComposition):

LayoutTests:

Add a regression test to ensure the same bug doesn't exist in textarea element.

  • platform/mac/editing/input/selection-change-closes-typing-2-expected.txt: Copied from

LayoutTests/platform/mac/editing/input/selection-change-closes-typing-expected.txt.

  • platform/mac/editing/input/selection-change-closes-typing-2.html: Copied from

LayoutTests/platform/mac/editing/input/selection-change-closes-typing.html.

3:43 PM Changeset in webkit [94496] by commit-queue@webkit.org
  • 1 edit
    2 adds in trunk/LayoutTests

Test that document.all.tags() matches IE behavior with too few arguments
​https://bugs.webkit.org/show_bug.cgi?id=66535

Patch by Mark Pilgrim <​pilgrim@chromium.org> on 2011-09-03
Reviewed by Darin Adler.

  • fast/dom/document-all-tags-expected.txt: Added.
  • fast/dom/document-all-tags.html: Added.
2:00 PM Changeset in webkit [94495] by robert@webkit.org
  • 4 edits
    4 adds in trunk/LayoutTests

Add platform-specific results for r94492

Unreviewed.

  • platform/chromium-win/fast/repaint/block-layout-inline-children-float-positioned-expected.png:
  • platform/gtk/fast/css/absolute-child-with-percent-height-inside-relative-parent-expected.txt: Added.
  • platform/gtk/fast/css/bug4860-absolute-inline-child-inherits-alignment-expected.txt: Added.
  • platform/gtk/fast/inline/absolute-positioned-inline-in-centred-block-expected.txt: Added.
  • platform/gtk/fast/repaint/block-layout-inline-children-float-positioned-expected.txt:
  • platform/mac/fast/inline/absolute-positioned-inline-in-centred-block-expected.txt: Added.
  • platform/mac/fast/repaint/block-layout-inline-children-float-positioned-expected.txt:
1:08 PM Changeset in webkit [94494] by kevino@webkit.org
  • 4 edits in trunk

[wx] Unreviewed build fix. Add wtf/dtoa directory to build.

12:26 PM Changeset in webkit [94493] by weinig@apple.com
  • 8 edits in trunk

Add missing Event constructors to DOMWindow.idl
​https://bugs.webkit.org/show_bug.cgi?id=67449

Reviewed by Anders Carlsson.

Source/WebCore:

Covered by existing tests.

  • page/DOMWindow.idl:

LayoutTests:

  • platform/mac/fast/dom/Window/window-properties-expected.txt:
  • platform/mac/fast/dom/Window/window-property-descriptors-expected.txt:
  • platform/mac/fast/dom/prototype-inheritance-expected.txt:
  • platform/mac/fast/js/global-constructors-expected.txt:

Update results for new constructors.

11:28 AM Changeset in webkit [94492] by robert@webkit.org
  • 8 edits
    12 adds in trunk

div align="center" rendering problem
​https://bugs.webkit.org/show_bug.cgi?id=4860

Reviewed by David Hyatt.

Source/WebCore:

When an inline element with absolute position was the sole or first child of a render block with
centred alignment, it wasn't obeying its parent's alignment. However it would obey the
alignment if it was preceded by some text. The problem was that the element's render object
was getting skipped as leading white space, so it was not included in a normal line block in
a bidi run. Instead, its position was getting set by RenderBlockLineLayout::setStaticPositions()
which does not pay attention to alignment. Preceding the element with some text allowed the object
to get included in a Bidi run and so get a linebox which would get properly aligned.

The fix is to get RenderBlockLineLayout::setStaticPositions() to obey the alignment specified by
the object's container. This allows WebKit to get the same result on the test as Firefox and IE.
Opera has the same bug as unpatched WebKit.

Tests: fast/css/bug4860-absolute-block-child-does-not-inherit-alignment.html

  • Ensure positioned block elements inherit alignment. fast/css/bug4860-absolute-inline-child-inherits-alignment.html
  • Ensure positioned inline elements inherit alignment. fast/inline/absolute-positioned-inline-in-centred-block.html
  • Ensure positioned inline element that's the sole or first child of a rendered block obeys parents alignment. fast/inline/absolute-positioned-block-in-centred-block.html
  • As above, but a positioned block should not inherit alignment.
  • rendering/RenderBlock.h:
  • rendering/RenderBlockLineLayout.cpp:

(WebCore::RenderBlock::updateLogicalWidthForAlignment):
(WebCore::RenderBlock::computeInlineDirectionPositionsForLine): Move the alignment check to updateLogicalWidthForAlignment.
(WebCore::setStaticPositions): use startAlignedOffsetForLine and use startAlignedOffsetForBlock
(WebCore::RenderBlock::startAlignedOffsetForLine): New function, find the aligned offset using updateLogicalWidthForAlignment

LayoutTests:

  • fast/css/bug4860-absolute-block-child-does-not-inherit-alignment-expected.png: Added.
  • fast/css/bug4860-absolute-block-child-does-not-inherit-alignment-expected.txt: Added.
  • fast/css/bug4860-absolute-block-child-does-not-inherit-alignment.html: Added.
  • fast/css/bug4860-absolute-inline-child-inherits-alignment-expected.png: Added.
  • fast/css/bug4860-absolute-inline-child-inherits-alignment-expected.txt: Added.
  • fast/css/bug4860-absolute-inline-child-inherits-alignment.html: Added.
  • fast/inline/absolute-positioned-block-in-centred-block-expected.png: Added.
  • fast/inline/absolute-positioned-block-in-centred-block-expected.txt: Added.
  • fast/inline/absolute-positioned-block-in-centred-block.html: Added.
  • fast/inline/absolute-positioned-inline-in-centred-block-expected.png: Added.
  • fast/inline/absolute-positioned-inline-in-centred-block-expected.txt: Added.
  • fast/inline/absolute-positioned-inline-in-centred-block.html: Added.
  • platform/chromium-linux/fast/repaint/block-layout-inline-children-float-positioned-expected.png:
  • platform/chromium-linux/fast/repaint/block-layout-inline-children-float-positioned-expected.txt: Both of these two tests were based on the incorrect behaviour of ignoring the alignment specified by the container of an element absolute position when there was no text preceding the element. The updated results agree with Firefox and IE.
  • fast/css/absolute-child-with-percent-height-inside-relative-parent-expected.txt:
  • platform/chromium-win/fast/css/absolute-child-with-percent-height-inside-relative-parent-expected.png: This test expected the wrong alignment of the red block - it should be centred, not aligned to the left.
11:18 AM Changeset in webkit [94491] by andreas.kling@nokia.com
  • 3 edits in trunk/Source/WebCore

Remove two unused functions from Element.
​https://bugs.webkit.org/show_bug.cgi?id=67492

Reviewed by Benjamin Poulain.

Removed openTagStartToString() and setCStringAttribute() as they are
not called from anywhere.

  • dom/Element.cpp:
  • dom/Element.h:
10:13 AM Changeset in webkit [94490] by mitz@apple.com
  • 2 edits in trunk/Tools

32-bit build fix.

  • TestWebKitAPI/mac/PlatformUtilitiesMac.mm:

(TestWebKitAPI::Util::toSTD):

9:13 AM Changeset in webkit [94489] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

[Qt] Enable support for WebGL OES_standard_derivatives for Qt
​https://bugs.webkit.org/show_bug.cgi?id=67430

Patch by Andrew Wason <​rectalogic@rectalogic.com> on 2011-09-03
Reviewed by Noam Rosenthal.

Tested using ​https://cvs.khronos.org/svn/repos/registry/trunk/public/webgl/sdk/tests/conformance/extensions/oes-standard-derivatives.html

Enable existing support for OES_standard_derivatives for Qt.

  • platform/graphics/opengl/Extensions3DOpenGL.cpp:

(WebCore::Extensions3DOpenGL::ensureEnabled):
(WebCore::Extensions3DOpenGL::isEnabled):

8:57 AM Changeset in webkit [94488] by mitz@apple.com
  • 5 edits
    1 add in trunk/Tools

Move testStringByEvaluatingJavaScriptFromString() from DumpRenderTree to TestWebKitAPI
​https://bugs.webkit.org/show_bug.cgi?id=67559

Reviewed by Darin Adler.

  • DumpRenderTree/mac/DumpRenderTree.mm:

(dumpRenderTree): Moved testStringByEvaluatingJavaScriptFromString() from here and removed
the call to it.

  • TestWebKitAPI/PlatformUtilities.h: Declared Util::toSTD(NSString *)
  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: Added StringByEvaluatingJavaScriptFromString.mm.
  • TestWebKitAPI/Tests/mac/StringByEvaluatingJavaScriptFromString.mm: Added.

(TestWebKitAPI::TEST): Added. Moved testStringByEvaluatingJavaScriptFromString() to here.

  • TestWebKitAPI/mac/PlatformUtilitiesMac.mm:

(TestWebKitAPI::Util::toSTD): Added.

1:29 AM Changeset in webkit [94487] by abarth@webkit.org
  • 6 edits in trunk/Tools

garden-o-matic results view should have a better results selector
​https://bugs.webkit.org/show_bug.cgi?id=67542

Reviewed by Dimitri Glazkov.

This patch adds a basic grid-based results selector to the results
view. Currently, this grid just shows how the tests fail on the
various bots. A future patch will let you select which failure to view
and to mark the failures for rebaselining.

  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
    • Add some more basic functions for manipulating dictionaries.
  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/controllers.js:
  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results.js:
  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/ui/results_unittests.js:
1:27 AM Changeset in webkit [94486] by abarth@webkit.org
  • 5 edits in trunk/Tools

garden-o-matic results view should show Actual and Expected results for text failures
​https://bugs.webkit.org/show_bug.cgi?id=67536

Reviewed by Eric Seidel.

In the past, we didn't show actual and expected text results (we just
showed the diff) because we didn't have the awesome six-by-two results
grid. Now that we do, we can move show these results.

  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base_unittests.js:
    • Fixed some typos that were preventing these unit tests from running.
  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:

(.):

  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results_unittests.js:
    • Report -actual.txt and -expected.txt as possible results URLs.
1:25 AM Changeset in webkit [94485] by abarth@webkit.org
  • 4 edits in trunk/Tools

garden-o-matic shows "ghost" failures that are already fixed
​https://bugs.webkit.org/show_bug.cgi?id=67535

Reviewed by Eric Seidel.

About 10% of writes fail to stick on test-results.appspot.com. This
patch causes us to fetch the full_results.json blob directly from
build.chromium.org rather than relying upon AppEngine.

Unforuntately, we still need to get historical information from
AppEngine. That will take several patches to fix becuase I need to
change the bots to write the full_results.json file to the
build-specific directory in addition to the zip file.

  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/base.js:
  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/model.js:
  • BuildSlaveSupport/build.webkit.org-config/public_html/TestFailures/scripts/results.js:
12:49 AM Changeset in webkit [94484] by fpizlo@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

DFG variable predictions only work for local variables, not temporaries
​https://bugs.webkit.org/show_bug.cgi?id=67554

Reviewed by Gavin Barraclough.

This appears to be a slight speed-up in Kraken (0.3% but significant)
and neutral elsewhere.

  • dfg/DFGGraph.h:

(JSC::DFG::Graph::predict):

12:45 AM Changeset in webkit [94483] by rniwa@webkit.org
  • 3 edits in trunk/Tools

Unreviewed, rolling out r94458.
​http://trac.webkit.org/changeset/94458
​https://bugs.webkit.org/show_bug.cgi?id=67558

Caused appcache test to fail on various bots (Requested by
rniwa on #webkit).

Patch by Sheriff Bot <​webkit.review.bot@gmail.com> on 2011-09-03

  • Scripts/webkitpy/layout_tests/port/webkit.py:
  • Scripts/webkitpy/layout_tests/port/webkit_unittest.py:
12:45 AM Changeset in webkit [94482] by yutak@chromium.org
  • 18 edits
    1 copy
    2 moves
    4 adds in trunk

WebSocket: Send ArrayBuffer as WebSocket binary message
​https://bugs.webkit.org/show_bug.cgi?id=67477

Reviewed by Kent Tamura.

Source/WebCore:

Tests: http/tests/websocket/tests/hybi/send-arraybuffer.html

http/tests/websocket/tests/hybi/workers/send-arraybuffer.html
http/tests/websocket/tests/hybi/bufferedAmount-after-close.html (updated)

  • bindings/js/JSWebSocketCustom.cpp:

(WebCore::JSWebSocket::send):

  • bindings/v8/custom/V8WebSocketCustom.cpp:

(WebCore::V8WebSocket::sendCallback):

  • websockets/ThreadableWebSocketChannel.h:
  • websockets/ThreadableWebSocketChannelClientWrapper.cpp:

(WebCore::ThreadableWebSocketChannelClientWrapper::ThreadableWebSocketChannelClientWrapper):
(WebCore::ThreadableWebSocketChannelClientWrapper::sendRequestResult):
(WebCore::ThreadableWebSocketChannelClientWrapper::setSendRequestResult):

  • websockets/ThreadableWebSocketChannelClientWrapper.h:

Rename "sent" to "sendRequestResult" to clarify the meaning. Messages from the script may not
be sent immediately, thus the return value of WebSocketChannel::send() indicates whether the
message has been queued successfully, rather than whether the message has been sent or not.

  • websockets/WebSocket.cpp:

(WebCore::WebSocket::send):
Case of sending "[object ArrayBuffer]" is covered by an existing test
http/tests/websocket/tests/{hybi,hixie76}/send-object.html.

  • websockets/WebSocket.h:
  • websockets/WebSocket.idl:
  • websockets/WebSocketChannel.cpp:

(WebCore::WebSocketChannel::send):

  • websockets/WebSocketChannel.h:
  • websockets/WorkerThreadableWebSocketChannel.cpp:

(WebCore::WorkerThreadableWebSocketChannel::send):
(WebCore::workerContextDidSend):
(WebCore::WorkerThreadableWebSocketChannel::Peer::send):
(WebCore::WorkerThreadableWebSocketChannel::mainThreadSendArrayBuffer):
Construct an ArrayBuffer from the data on Vector<char>.
(WebCore::WorkerThreadableWebSocketChannel::Bridge::send):
Copy the content into temporary buffer of Vector<char>, and send it to the main thread.

  • websockets/WorkerThreadableWebSocketChannel.h:

LayoutTests:

  • http/tests/websocket/tests/hybi/bufferedAmount-after-close-expected.txt:
  • http/tests/websocket/tests/hybi/bufferedAmount-after-close.html:

Add tests to send ArrayBuffers.

  • http/tests/websocket/tests/hybi/check-binary-messages_wsh.py:

Renamed from LayoutTests/http/tests/websocket/tests/hybi/send-blob_wsh.py, because this handler
is used from send-blob.html and send-arraybuffer.html.

  • http/tests/websocket/tests/hybi/send-arraybuffer-expected.txt: Added.
  • http/tests/websocket/tests/hybi/send-arraybuffer.html:

Added. Send three small binary messages as ArrayBuffers.

  • http/tests/websocket/tests/hybi/send-blob.html:
  • http/tests/websocket/tests/hybi/workers/resources/check-binary-messages_wsh.py:

Renamed from LayoutTests/http/tests/websocket/tests/hybi/workers/resources/send-blob_wsh.py.

  • http/tests/websocket/tests/hybi/workers/resources/send-arraybuffer.js:

Added. Same as send-arraybuffer.html, except that this test is run on the worker.

  • http/tests/websocket/tests/hybi/workers/resources/send-blob.js:

Removed the definition of unnecessary function startsWith().

  • http/tests/websocket/tests/hybi/workers/send-arraybuffer-expected.txt: Added.
  • http/tests/websocket/tests/hybi/workers/send-arraybuffer.html: Added
12:29 AM Changeset in webkit [94481] by rniwa@webkit.org
  • 2 edits in trunk/Tools

Unreviewed, rolling out r94453.
​http://trac.webkit.org/changeset/94453
​https://bugs.webkit.org/show_bug.cgi?id=67557

Caused appcache test to fail on various bots (Requested by
rniwa on #webkit).

Patch by Sheriff Bot <​webkit.review.bot@gmail.com> on 2011-09-03

  • Scripts/webkitpy/layout_tests/port/webkit.py:
Note: See TracTimeline for information about the timeline view.