WebKit-https.git
10 years ago2010-05-11 Joseph Pecoraro <joepeck@webkit.org>
joepeck@webkit.org [Wed, 12 May 2010 05:50:38 +0000 (05:50 +0000)]
2010-05-11  Joseph Pecoraro  <joepeck@webkit.org>

        Reviewed by Timothy Hatcher.

        Web Inspector: Shorthand Properties Getting Deleted When Editing
        https://bugs.webkit.org/show_bug.cgi?id=38958

        * inspector/front-end/DOMAgent.js:
        (WebInspector.CSSStyleDeclaration.prototype.styleTextWithShorthands): was getting regular values instead of shorthand values

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59210 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Chris Jerdonek <cjerdonek@webkit.org>
cjerdonek@webkit.org [Wed, 12 May 2010 05:10:32 +0000 (05:10 +0000)]
2010-05-11  Chris Jerdonek  <cjerdonek@webkit.org>

        Reviewed by Eric Seidel.

        Fixed two FIXME's in svn-apply: eliminated the unnecessary %copiedFiles
        variable and changed gitKnowsOfFile() to use exitStatus().

        https://bugs.webkit.org/show_bug.cgi?id=38862

        * Scripts/svn-apply:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59209 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Eric Seidel <eric@webkit.org>
eric@webkit.org [Wed, 12 May 2010 04:56:54 +0000 (04:56 +0000)]
2010-05-11  Eric Seidel  <eric@webkit.org>

        Unreviewed, just fixing python typo.

        EWS bots should poll more often than every 5 minutes
        https://bugs.webkit.org/show_bug.cgi?id=38968

        Typo in _now(), add a unittest to prove I fixed it.

        * Scripts/webkitpy/tool/bot/queueengine.py:
        * Scripts/webkitpy/tool/bot/queueengine_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59208 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Wed, 12 May 2010 04:44:02 +0000 (04:44 +0000)]
2010-05-11  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Qt build failures cause SheriffBot false positives
        https://bugs.webkit.org/show_bug.cgi?id=38969

        The Qt buildbot randomly fails to compile occasionally because its
        network connection causes SVn to leave zero-byte files around.  These
        compile failures confuse SheriffBot into thinking someone's patch
        caused a build break.

        In this patch, I've temporarily removed Qt from the list of core
        builders.  Ossy is working on a script to clean up the zero byte files.
        Once that goes in, we can add Qt back to the core builders.

        * Scripts/webkitpy/common/net/buildbot.py:
        * Scripts/webkitpy/common/net/buildbot_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59207 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Adam Barth <abarth@webkit.org>
eric@webkit.org [Wed, 12 May 2010 04:36:18 +0000 (04:36 +0000)]
2010-05-11  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        sheriffbot is spammy.
        https://bugs.webkit.org/show_bug.cgi?id=38936

        Reduce sheriffbot spam by not warning about new blameworthy revisions
        that can be explained by previously blamed revisions.  This might cause
        us to not warn about some real failures, but we're getting too much
        spam from slow builders that have large blamelists (and we've already
        poked the responsible folks using data from a fast builder).

        * Scripts/webkitpy/tool/bot/sheriff_unittest.py:
            - Removed unneeded import
        * Scripts/webkitpy/tool/commands/sheriffbot.py:
            - The logic change
        * Scripts/webkitpy/tool/commands/sheriffbot_unittest.py:
            - Tests of the change
        * Scripts/webkitpy/tool/mocktool.py:
            - Merged two declarations of MockBuilder

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59206 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSlight simplification to the calling convention: read RegisterFile::end()
ggaren@apple.com [Wed, 12 May 2010 04:30:35 +0000 (04:30 +0000)]
Slight simplification to the calling convention: read RegisterFile::end()
from an absolute address, instead of from a pointer to the RegisterFile
on the stack.

Reviewed by Gavin Barraclough.

SunSpider reports no change.

This should free us to remove the RegisterFile argument from the stack.

* jit/JIT.cpp:
(JSC::JIT::privateCompile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59205 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Wed, 12 May 2010 04:20:08 +0000 (04:20 +0000)]
2010-05-11  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        sheriffbot is spammy.
        https://bugs.webkit.org/show_bug.cgi?id=38936

        Reduce sheriffbot spam by not warning about new blameworthy revisions
        that can be explained by previously blamed revisions.  This might cause
        us to not warn about some real failures, but we're getting too much
        spam from slow builders that have large blamelists (and we've already
        poked the responsible folks using data from a fast builder).

        * Scripts/webkitpy/tool/bot/sheriff_unittest.py:
            - Removed unneeded import
        * Scripts/webkitpy/tool/commands/sheriffbot.py:
            - The logic change
        * Scripts/webkitpy/tool/commands/sheriffbot_unittest.py:
            - Tests of the change
        * Scripts/webkitpy/tool/mocktool.py:
            - Merged two declarations of MockBuilder

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59204 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Tony Chang <tony@chromium.org>
tony@chromium.org [Wed, 12 May 2010 03:53:35 +0000 (03:53 +0000)]
2010-05-11  Tony Chang  <tony@chromium.org>

        Reviewed by Darin Fisher.

        [chromium] Remove subframe scrollbar routing (r58937)
        https://bugs.webkit.org/show_bug.cgi?id=38894

        * platform/win/Skipped: This test should pass on windows.
2010-05-11  Tony Chang  <tony@chromium.org>

        Reviewed by Darin Fisher.

        [chromium] Remove subframe scrollbar routing (r58937)
        https://bugs.webkit.org/show_bug.cgi?id=38894

        * page/chromium/EventHandlerChromium.cpp:
        (WebCore::EventHandler::passMousePressEventToSubframe):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59203 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoPossible Gtk build fix - use ALWAYS_INLINE on functions that were inlined.
bweinstein@apple.com [Wed, 12 May 2010 03:53:13 +0000 (03:53 +0000)]
Possible Gtk build fix - use ALWAYS_INLINE on functions that were inlined.

* wtf/text/AtomicString.h:
(WebCore::AtomicString::add): Use ALWAYS_INLINE.
* wtf/text/WTFString.h:
(WebCore::String::length): Ditto.
(WebCore::String::isEmpty): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59202 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Tony Chang <tony@chromium.org>
tony@chromium.org [Wed, 12 May 2010 03:51:18 +0000 (03:51 +0000)]
2010-05-11  Tony Chang  <tony@chromium.org>

        Reviewed by David Levin.

        [chromium] detect num processors to pass to make -j on Linux
        https://bugs.webkit.org/show_bug.cgi?id=38833

        * Scripts/webkitdirs.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59201 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Eric Seidel <eric@webkit.org>
eric@webkit.org [Wed, 12 May 2010 03:41:40 +0000 (03:41 +0000)]
2010-05-11  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        commit-queue should commit rollouts without running tests
        https://bugs.webkit.org/show_bug.cgi?id=38940

        Most of this change is improving our test coverage for the commit-queue.
        The only functional change is removing the --test flag when the commit-queue
        is running in rollouts mode.

        I added test coverage for status updates, and updated the commit-queue status
        messages to distinguish rollout vs. normal landing mode in its empty queue
        and land patch messages.

        * Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
         - Got rid of a bunch of copy/paste code using _default_expected_stderr
        * Scripts/webkitpy/tool/commands/queues.py:
         - Moved rollout patch filtering out of _validate_patches_in_commit_queue
           so that we only have to check if the builders are green in one place.
         - Make the "empty queue" message note which queue it is referring to.
         - Don't pass --text to land-attachment when in rollout mode.
        * Scripts/webkitpy/tool/commands/queues_unittest.py:
         - Update results now that we're testing status updates.
         - Test _can_build_and_test since I made a typo in that call while
           writing this change and the unit tests failed to catch it!
        * Scripts/webkitpy/tool/mocktool.py:
         - Log status updates to make them testable in our unit tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59200 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Wed, 12 May 2010 03:25:52 +0000 (03:25 +0000)]
2010-05-11  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Maciej Stachowiak.

        Body not redrawn, and filled with garbage on some composited pages
        https://bugs.webkit.org/show_bug.cgi?id=38951
        <rdar://problem/7891548>

        When the root layer becomes composited (for example, because a negative z-index
        descendant in its stacking context becomes composited), then it has to be made
        large enough to fill the viewport, because the page background draws into it.

        Test: compositing/geometry/composited-html-size.html

        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::calculateCompositedBounds):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59199 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd a compile time assert that sizeof(String) == sizeof(AtomicString).
andersca@apple.com [Wed, 12 May 2010 02:33:32 +0000 (02:33 +0000)]
Add a compile time assert that sizeof(String) == sizeof(AtomicString).

Reviewed by Sam Weinig.

* wtf/text/AtomicString.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59198 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Wed, 12 May 2010 02:28:23 +0000 (02:28 +0000)]
2010-05-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Mark Rowe.

        https://bugs.webkit.org/show_bug.cgi?id=38961
        Move some member variables around to reduce class sizes.

        Make AtomicString no longer inherit from FastAllocBase.

        Since AtomicString's first (and only) member variable, a RefPtr, also inherits from FastAllocBase this
        was causing the size of AtomicString to contain sizeof(void*) bytes of padding.

        * wtf/text/AtomicString.h:
2010-05-11  Anders Carlsson  <andersca@apple.com>

        Reviewed by Mark Rowe.

        https://bugs.webkit.org/show_bug.cgi?id=38961
        Move some member variables around to reduce class sizes.

        If a class inherits (directly or indirectly) from FastAllocBase, we don't want the first member variable
        of that class to also inherit (directly or indirectly) from FastAllocBase as this will add padding to the class
        and thus increasing its size.

        * css/CSSParserValues.h:
        * css/CSSSelector.h:
        (WebCore::CSSSelector::CSSSelector):
        (WebCore::CSSSelector::RareData::RareData):
        * platform/graphics/FontCache.cpp:
        (WebCore::FontPlatformDataCacheKey::FontPlatformDataCacheKey):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59197 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoPatch by Gavin Barraclough.
bweinstein@apple.com [Wed, 12 May 2010 02:25:40 +0000 (02:25 +0000)]
Patch by Gavin Barraclough.

Reviewed by Maciej Stachowiak.

Rest of REGRESSION (r57900-57919): 3% PLT Regression from moving strings into WTF.
https://bugs.webkit.org/show_bug.cgi?id=38930
<rdar://problem/7937188>

Inline string functions that are commonly used in WTFString, CString, and AtomicString.

* JavaScriptCore.exp: Remove exports of inline functions.
* wtf/text/AtomicString.cpp:
(WebCore::AtomicString::addSlowCase): Rename add to addSlowCase, and inline the fast path
    of add.
* wtf/text/AtomicString.h:
(WebCore::AtomicString::add): Inline the fast path of add.
* wtf/text/CString.cpp: Inline implementation of data and length in the header.
* wtf/text/CString.h:
(WTF::CString::data): Inline implementation of data.
(WTF::CString::length): Inline implementation of length.
* wtf/text/WTFString.cpp: Remove implementations of functions to be inlined.
* wtf/text/WTFString.h:
(WebCore::String::String): Inline implementation.
(WebCore::String::length): Ditto.
(WebCore::String::characters): Ditto.
(WebCore::String::operator[]): Ditto.
(WebCore::String::isEmpty): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59196 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdding a function needed by the V8 code generator to turn on/off
dumi@chromium.org [Wed, 12 May 2010 02:20:47 +0000 (02:20 +0000)]
Adding a function needed by the V8 code generator to turn on/off
the binding for WorkerContext::openDatabaseSync().
https://bugs.webkit.org/show_bug.cgi?id=38960

Reviewed by Adam Barth.

* bindings/generic/RuntimeEnabledFeatures.cpp:
(WebCore::RuntimeEnabledFeatures::openDatabaseSyncEnabled):
* bindings/generic/RuntimeEnabledFeatures.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59195 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 MORITA Hajime <morrita@google.com>
morrita@google.com [Wed, 12 May 2010 02:15:00 +0000 (02:15 +0000)]
2010-05-11  MORITA Hajime  <morrita@google.com>

        Reviewed by Dirk Schulze.

        CSS SVG font doesn't recognize URL without element reference
        https://bugs.webkit.org/show_bug.cgi?id=37144

        * platform/mac/svg/custom/svg-fonts-with-no-element-reference-expected.checksum: Added.
        * platform/mac/svg/custom/svg-fonts-with-no-element-reference-expected.png: Added.
        * platform/mac/svg/custom/svg-fonts-with-no-element-reference-expected.txt: Added.
        * svg/custom/resources/graffiti-with-empty-first-font.svg: Added.
        * svg/custom/svg-fonts-with-no-element-reference.html: Added.
2010-05-11  MORITA Hajime  <morrita@google.com>

        Reviewed by Dirk Schulze.

        CSS SVG font doesn't recognize URL without element reference
        https://bugs.webkit.org/show_bug.cgi?id=37144

        CachedFont::getSVGFontById() did assume that element reference (ID) is given.
        So fixed to handle an empty or missing element reference. In which case,
        the first font element is returned if available.

        Test: svg/custom/svg-fonts-with-no-element-reference.html

        * loader/CachedFont.cpp:
        (WebCore::CachedFont::getSVGFontById):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59194 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Kent Tamura <tkent@chromium.org>
tkent@chromium.org [Wed, 12 May 2010 01:53:30 +0000 (01:53 +0000)]
2010-05-11  Kent Tamura  <tkent@chromium.org>

        Reviewed by Dimitri Glazkov.

        [DRT/Chromium] Apply recent changes of test_shell to DRT
        https://bugs.webkit.org/show_bug.cgi?id=38895

        * DEPS: Update Chromium revision to use the recent webkit_support change.
        * WebKit.gyp: Add JavaScriptCore/wtf to 'include' section.
        * src/WebGraphicsContext3D.cpp:
          Remove ENABLE(3D_CANVAS) to provide an empty implementation of
          WebGraphicsContext3D::createDefault().
2010-05-11  Kent Tamura  <tkent@chromium.org>

        Reviewed by Dimitri Glazkov.

        [DRT/Chromium] Apply recent changes of test_shell to DRT
        https://bugs.webkit.org/show_bug.cgi?id=38895

        Port the changes to test_shell during (r40492, r46810] of Chromium.
        Highlights:
        - Introduce NotificationPresenter
        - Fix parameter mismatch of WebViewClient::startDragging()

        This change fixes dozens of unexpected behaviors.

        * DumpRenderTree/DumpRenderTree.gypi:
          Add NotificationPresenter.{cpp,h}
        * DumpRenderTree/chromium/EventSender.cpp:
        (applyKeyModifier):
        (EventSender::EventSender):
        (EventSender::reset):
        (EventSender::webview):
        (EventSender::doDragDrop):
        (EventSender::keyDown):
        (EventSender::addTouchPoint):
        (EventSender::clearTouchPoints):
        (EventSender::releaseTouchPoint):
        (EventSender::setTouchModifier):
        (EventSender::updateTouchPoint):
        (EventSender::cancelTouchPoint):
        (EventSender::sendCurrentTouchEvent):
        (EventSender::touchEnd):
        (EventSender::touchMove):
        (EventSender::touchStart):
        (EventSender::touchCancel):
        * DumpRenderTree/chromium/EventSender.h:
        * DumpRenderTree/chromium/LayoutTestController.cpp:
        (LayoutTestController::LayoutTestController):
        (LayoutTestController::reset):
        (LayoutTestController::showWebInspector):
        (LayoutTestController::setAuthorAndUserStylesEnabled):
        (LayoutTestController::setScrollbarPolicy):
        (LayoutTestController::setWillSendRequestClearHeader):
        (LayoutTestController::callShouldCloseOnWebView):
        (LayoutTestController::grantDesktopNotificationPermission):
        (LayoutTestController::removeOriginAccessWhitelistEntry):
        (LayoutTestController::addUserScript):
        (LayoutTestController::addUserStyleSheet):
        * DumpRenderTree/chromium/LayoutTestController.h:
        * DumpRenderTree/chromium/NotificationPresenter.cpp: Added.
        * DumpRenderTree/chromium/NotificationPresenter.h: Added.
        * DumpRenderTree/chromium/TestShell.cpp:
        (TestShell::TestShell):
        (TestShell::runFileTest):
        (TestShell::resetTestController):
        * DumpRenderTree/chromium/TestShell.h:
        (TestShell::eventSender):
        (TestShell::notificationPresenter):
        (TestShell::showDevTools):
        * DumpRenderTree/chromium/WebViewHost.cpp:
        (WebViewHost::createView):
        (WebViewHost::createPopupMenu):
        (WebViewHost::startDragging):
        (WebViewHost::notificationPresenter):
        (WebViewHost::createApplicationCacheHost):
        (WebViewHost::willSendRequest):
        (WebViewHost::updateAddressBar):
        (WebViewHost::updateURL):
        * DumpRenderTree/chromium/WebViewHost.h:
        (WebViewHost::addClearHeader):
        (WebViewHost::clearHeaders):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59193 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoNo Review.
jer.noble@apple.com [Wed, 12 May 2010 01:35:15 +0000 (01:35 +0000)]
No Review.

Fix the Windows Debug Build: the correct VCProj variable to use when linking against
JavaScriptCore is $(WebKitDLLConfigSuffix), not $(WebKitConfigSuffix), which expands
to an empty string rather than "_debug".

* WebCore.vcproj/QTMovieWin.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59192 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoNo Review.
jer.noble@apple.com [Wed, 12 May 2010 01:16:25 +0000 (01:16 +0000)]
No Review.

Fix build error: The QTMovieWin project is dependent on the JavaScriptCore project.

* WebKit.vcproj/WebKit.sln:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59191 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Jer Noble <jer.noble@apple.com>
jer.noble@apple.com [Wed, 12 May 2010 00:41:48 +0000 (00:41 +0000)]
2010-05-11  Jer Noble  <jer.noble@apple.com>

        Darin Adler.

        19 media tests are crashing on Windows Release
        https://bugs.webkit.org/show_bug.cgi?id=38950
        rdar://problem/7971658

        Link against JavaScriptCore.lib instead of WTF.lib, so the free and malloc used by
        QTMovieWin is always the free and malloc in JavaScriptCore.dll.

        * WebCore.vcproj/QTMovieWin.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59190 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAnother refactoring patch in preparation for a fix to
darin@apple.com [Wed, 12 May 2010 00:35:38 +0000 (00:35 +0000)]
Another refactoring patch in preparation for a fix to
https://bugs.webkit.org/show_bug.cgi?id=38871
REGRESSION: Crash clicking background NPR tab after few minutes of sitting idle

Reviewed by Beth Dakin.

* dom/Document.cpp:
(WebCore::Document::addMediaCanStartListener): Added.
(WebCore::Document::removeMediaCanStartListener): Added.

* dom/Document.h: Added add/removeMediaCanStartListener.

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::~HTMLMediaElement): Call removeMediaCanStartListener
on the document instead of the page.
(WebCore::HTMLMediaElement::willMoveToNewOwnerDocument): Call
removeMediaCanStartListener on the current document.
(WebCore::HTMLMediaElement::didMoveToNewOwnerDocument): Call
addMediaCanStartListener on the new document.
(WebCore::HTMLMediaElement::loadInternal): Call addMediaCanStartListener
on the document instead of the page.

* plugins/PluginView.cpp:
(WebCore::PluginView::startOrAddToUnstartedList): Call addMediaCanStartListener
on the document instead of the page.
(WebCore::PluginView::~PluginView): Call removeMediaCanStartListener
on the document instead of the page, and do it in line instead of calling
a function.

* plugins/PluginView.h: Removed unused removeFromUnstartedListIfNecessary.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59189 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Not reviewed.
ap@apple.com [Wed, 12 May 2010 00:26:03 +0000 (00:26 +0000)]
    Not reviewed.

        https://bugs.webkit.org/show_bug.cgi?id=38956
        [Gtk] http/tests/xmlhttprequest/basic-auth-default.html fails

        * platform/gtk/Skipped: Skipping the new failing test on Gtk.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59188 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoJavaScriptCore: Patch by Gavin Barraclough.
bweinstein@apple.com [Wed, 12 May 2010 00:22:28 +0000 (00:22 +0000)]
JavaScriptCore: Patch by Gavin Barraclough.

Reviewed by Steve Falkenburg and Jon Honeycutt.

REGRESSION (r57900-57919): 3% PLT Regression from moving strings into WTF.
https://bugs.webkit.org/show_bug.cgi?id=38930
<rdar://problem/7937188>

When the String classes were moved from WebCore to WTF, it meant that on Windows, all operations
on Strings in WebCore had to cross a DLL boundary (from WebKit.dll to JavaScript.dll).

We fix this by refactoring some of the WTF string code, so the code in AtomicString, StringImpl, and
WTFString can be built by both WebCore and WTF, and we don't need to talk across a DLL to do operations
on Strings.

* GNUmakefile.am: Add new file to build system.
* JavaScriptCore.gypi: Ditto.
* JavaScriptCore.pro: Ditto.
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Remove string exports, because these are now
    handled in WebCore.
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Remove the post-build step that was added
    here (the post build step is in JavaScriptCoreCommon.vsprops).
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: Copy the three cpp files that need to be compiled
    by WebCore into the WebKitOutputDir directory.
* JavaScriptCore.vcproj/WTF/WTF.vcproj: Add the StringStatics file.
* JavaScriptCore.vcproj/jsc/jsc.vcproj: Add the three WTF string cpp files to this project.
* JavaScriptCore.vcproj/jsc/jscCommon.vsprops: Remove the need to link against WTF.lib (since jsc links against JavaScriptCore).
* JavaScriptCore.xcodeproj/project.pbxproj: Add the StringStatics file.
* wtf/text/AtomicString.cpp: Moved code to StringStatics.
* wtf/text/StringImpl.cpp: Ditto.
* wtf/text/StringStatics.cpp: Added. Move functions in WTF Strings that define static variables to here, so
    the rest of the files can be compiled in WebCore.
(WebCore::StringImpl::empty): Moved from StringImpl.cpp to here.
(WebCore::AtomicString::init): Moved from AtomicString.cpp to here.

WebCore: Patch by Gavin Barraclough.

Reviewed by Steve Falkenburg and Jon Honeycutt.

REGRESSION (r57900-57919): 3% PLT Regression from moving strings into WTF.
https://bugs.webkit.org/show_bug.cgi?id=38930
<rdar://problem/7937188>

Add the WTF strings into the WebCore vcproj, from their copied location in $(WebKitOutputDir).

No new tests because no change in behavior.

* WebCore.vcproj/WebCore.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59187 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSmall re-factoring patch in preparation for fixing https://
bdakin@apple.com [Wed, 12 May 2010 00:04:26 +0000 (00:04 +0000)]
Small re-factoring patch in preparation for fixing https://
bugs.webkit.org/show_bug.cgi?id=38871 REGRESSION: Crash clicking
background NPR tab after few minutes of sitting idle

Reviewed by Darin Adler.

* page/Page.cpp:
(WebCore::Page::takeAnyMediaCanStartListener):
(WebCore::Page::setCanStartMedia):
* page/Page.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59186 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago Reviewed by Brady Eidson.
ap@apple.com [Tue, 11 May 2010 23:41:25 +0000 (23:41 +0000)]
    Reviewed by Brady Eidson.

        https://bugs.webkit.org/show_bug.cgi?id=37989
        Safari (still) doesn't always send basic credentials preemptively

        Test: http/tests/xmlhttprequest/basic-auth-default.html

        After fetching credentials from WebCore storage, put them back right away. This will add
        default credentials for the directory, since this may be the first time we learn about its
        associated protection space.

        * platform/network/cf/ResourceHandleCFNet.cpp:
        (WebCore::ResourceHandle::didReceiveAuthenticationChallenge):
        (WebCore::WebCoreSynchronousLoader::didReceiveChallenge):
        * platform/network/mac/ResourceHandleMac.mm:
        (WebCore::ResourceHandle::didReceiveAuthenticationChallenge):
        (-[WebCoreSynchronousLoader connection:didReceiveAuthenticationChallenge:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59185 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoReviewed by Kevin Ollivier.
kevino@webkit.org [Tue, 11 May 2010 23:41:16 +0000 (23:41 +0000)]
Reviewed by Kevin Ollivier.

Load a blank page on creation to initialize WebCore objects properly.
https://bugs.webkit.org/show_bug.cgi?id=38932

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59184 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoReviewed by Kevin Ollivier.
kevino@webkit.org [Tue, 11 May 2010 23:39:02 +0000 (23:39 +0000)]
Reviewed by Kevin Ollivier.

[wx] Create an NSFont on Mac builds that do not have built-in methods to do so.
https://bugs.webkit.org/show_bug.cgi?id=38931

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59183 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Tue, 11 May 2010 23:22:14 +0000 (23:22 +0000)]
2010-05-10  Dirk Pranke  <dpranke@chromium.org>

        Reviewed by Eric Seidel.

        new-run-webkit-tests: fix a path-handling bug that was breaking the
        dryrun ports on windows, and add a comment about why we don't run
        the chromium dryrun tests by default on every port.

        https://bugs.webkit.org/show_bug.cgi?id=38796

        * Scripts/webkitpy/layout_tests/port/dryrun.py:
        * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59182 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<http://webkit.org/b/38941> build-webkit shouldn't always override ENABLE_FOO Xcode...
mrowe@apple.com [Tue, 11 May 2010 23:12:23 +0000 (23:12 +0000)]
<webkit.org/b/38941> build-webkit shouldn't always override ENABLE_FOO Xcode configuration settings

Reviewed by Darin Adler.

* Scripts/build-webkit: Don't pass the ENABLE setting to xcodebuild if the value matches the default.
This will lead to xcodebuild picking up the settings from FeatureDefines.xcconfig, and will aid in
revealing problems that are due to inconsistent values for settings across projects.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59181 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed. Remove unneeded rebaselined result for enter-clicks-buttons.html
jianli@chromium.org [Tue, 11 May 2010 23:11:02 +0000 (23:11 +0000)]
Unreviewed. Remove unneeded rebaselined result for enter-clicks-buttons.html

* platform/chromium-win/fast/forms/enter-clicks-buttons-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59180 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Avi Drissman <avi@chromium.org>
eric@webkit.org [Tue, 11 May 2010 23:04:34 +0000 (23:04 +0000)]
2010-05-11  Avi Drissman  <avi@chromium.org>

        Reviewed by Darin Fisher.

        [Chromium] Pipe RTL info into WebPopupMenuInfo
        https://bugs.webkit.org/show_bug.cgi?id=38749

        * platform/chromium/PopupMenuChromium.cpp:
        (WebCore::PopupContainer::menuStyle):
        * platform/chromium/PopupMenuChromium.h:
2010-05-11  Avi Drissman  <avi@chromium.org>

        Reviewed by Darin Fisher.

        [Chromium] Pipe RTL info into WebPopupMenuInfo
        https://bugs.webkit.org/show_bug.cgi?id=38749

        * public/WebPopupMenuInfo.h:
        * src/ChromeClientImpl.cpp:
        (WebKit::ChromeClientImpl::getPopupMenuInfo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59179 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix build error when enabling debugging block in WebKit win painting code
alice.liu@apple.com [Tue, 11 May 2010 22:56:19 +0000 (22:56 +0000)]
Fix build error when enabling debugging block in WebKit win painting code

Rubber-stamped by Gavin Barraclough.

JavaScriptCore:

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

WebKit/win:

* WebView.cpp:
(WebView::paintIntoBackingStore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59178 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoBuild fix.
mrowe@apple.com [Tue, 11 May 2010 22:43:23 +0000 (22:43 +0000)]
Build fix.

* DumpRenderTree/mac/LayoutTestControllerMac.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59177 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 David Holloway <dhollowa@chromium.org>
eric@webkit.org [Tue, 11 May 2010 22:39:20 +0000 (22:39 +0000)]
2010-05-11  David Holloway  <dhollowa@chromium.org>

        Reviewed by Darin Fisher.

        [chromium] Add size method to WebInputElement class, providing same data
        as HTMLInputElement::size().  During form AutoFill Chromium needs to
        access the input element's size attribute in aid of form matching and
        filling heuristics.

        https://bugs.webkit.org/show_bug.cgi?id=38825

        * public/WebInputElement.h:
        * src/WebInputElement.cpp:
        (WebKit::WebInputElement::size):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59176 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agohttps://bugs.webkit.org/show_bug.cgi?id=38937
alice.liu@apple.com [Tue, 11 May 2010 22:28:32 +0000 (22:28 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=38937
W7 window preview paints content at the wrong location

Reviewed by Steve Falkenburg.

* WebFrame.cpp:
(WebFrame::paintDocumentRectToContext): Revert r58895
(WebFrame::paintDocumentRectToContextAtPoint): Added

* Interfaces/IWebFramePrivate.idl: Added paintDocumentRectToContextAtPoint
* Interfaces/IWebViewPrivate.idl: ditto
* Interfaces/WebKit.idl: touch to rebuild
* WebFrame.h: ditto
* WebView.cpp: ditto
(WebView::paintDocumentRectToContextAtPoint): ditto
* WebView.h: ditto

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59175 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix the world.
mrowe@apple.com [Tue, 11 May 2010 22:19:52 +0000 (22:19 +0000)]
Fix the world.

In r59162 a change was made to WebCore's FeatureDefines.xcconfig that enabled FILE_READER and FILE_WRITER.
The author and reviewer of that patch ignored the carefully-worded warning at the top of that file asking
that changes to the file be kept in sync across JavaScriptCore, WebCore and WebKit, as well as being kept
in sync with build-webkit.  This led to WebCore and WebKit having different views of Document's vtable
and results in crashes in Safari shortly after launch when virtual function calls resulted in the wrong
function in WebCore being called.

We fix this by bringing the FeatureDefines.xcconfig files in to sync.  Based on the ChangeLog message and
other changes in r59162 it appears that enabling FILE_WRITER was unintentional so that particular change
has been reverted.

JavaScriptCore:

* Configurations/FeatureDefines.xcconfig:

WebCore:

* Configurations/FeatureDefines.xcconfig:

WebKit/mac:

* Configurations/FeatureDefines.xcconfig:

WebKit2:

* Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59174 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Dimitri Glazkov <dglazkov@chromium.org>
dglazkov@chromium.org [Tue, 11 May 2010 22:01:27 +0000 (22:01 +0000)]
2010-05-11  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r58520): Implicit submission on forms with button type="submit" no longer works
        https://bugs.webkit.org/show_bug.cgi?id=38913

        Updated the test to include coverage of button elements.

        * fast/forms/enter-clicks-buttons-expected.txt: Updated expectations to account for implicit
            submission which now correctly happens from the right element (button).
        * fast/forms/implicit-submission-expected.txt: Updated expectations with new test results.
        * fast/forms/implicit-submission.html: Added tests and code to cover button elements.
2010-05-11  Dimitri Glazkov  <dglazkov@chromium.org>

        Reviewed by Darin Adler.

        REGRESSION(r58520): Implicit submission on forms with button type="submit" no longer works
        https://bugs.webkit.org/show_bug.cgi?id=38913

        * Widened the check for a successful submit button to include all HTMLFormControlElements.
        * Clarified implicit submission code by:
            - introducing HTMLFormControlElement::canTriggerImplicitSubmission flag,
            - getting rid of static casting and checking for tag names.

        * html/HTMLFormControlElement.h:
        (WebCore::HTMLFormControlElement::canTriggerImplicitSubmission): Added.
        * html/HTMLFormElement.cpp:
        (WebCore::HTMLFormElement::submitImplicitly): Renamed parameters for clarity, refactored code
            to use canTriggerImplicitSubmission.
        * html/HTMLFormElement.h: Renamed parameter for clarity.
        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::defaultEventHandler):  Used canTriggerImplicitSubmission()
            instead of isTextField() for clarity.
        * html/HTMLInputElement.h:
        (WebCore::HTMLInputElement::canTriggerImplicitSubmission): Added.
        * html/HTMLIsIndexElement.h:
        (WebCore::HTMLIsIndexElement::canTriggerImplicitSubmission): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59173 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Sheriff Bot <webkit.review.bot@gmail.com>
eric@webkit.org [Tue, 11 May 2010 21:35:39 +0000 (21:35 +0000)]
2010-05-11  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r59171.
        http://trac.webkit.org/changeset/59171
        https://bugs.webkit.org/show_bug.cgi?id=38933

        "Broke the world" (Requested by bweinstein on #webkit).

        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:
        * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops:
        * JavaScriptCore.vcproj/WTF/WTF.vcproj:
        * JavaScriptCore.vcproj/jsc/jsc.vcproj:
        * JavaScriptCore.vcproj/jsc/jscCommon.vsprops:
        * JavaScriptCore.xcodeproj/project.pbxproj:
        * wtf/text/AtomicString.cpp:
        (WebCore::AtomicString::init):
        * wtf/text/StringImpl.cpp:
        (WebCore::StringImpl::empty):
        * wtf/text/StringStatics.cpp: Removed.
2010-05-11  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r59171.
        http://trac.webkit.org/changeset/59171
        https://bugs.webkit.org/show_bug.cgi?id=38933

        "Broke the world" (Requested by bweinstein on #webkit).

        * WebCore.vcproj/WebCore.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59172 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoJavaScriptCore: Patch by Gavin Barraclough.
bweinstein@apple.com [Tue, 11 May 2010 20:52:51 +0000 (20:52 +0000)]
JavaScriptCore: Patch by Gavin Barraclough.

Reviewed by Steve Falkenburg.

REGRESSION (r57900-57919): 3% PLT Regression from moving strings into WTF.
https://bugs.webkit.org/show_bug.cgi?id=38930
<rdar://problem/7937188>

When the String classes were moved from WebCore to WTF, it meant that on Windows, all operations
on Strings in WebCore had to cross a DLL boundary (from WebKit.dll to JavaScript.dll).

We fix this by refactoring some of the WTF string code, so the code in AtomicString, StringImpl, and
WTFString can be built by both WebCore and WTF, and we don't need to talk across a DLL to do operations
on Strings.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Remove string exports, because these are now
    handled in WebCore.
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Remove the post-build step that was added
    here (the post build step is in JavaScriptCoreCommon.vsprops).
* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: Copy the three cpp files that need to be compiled
    by WebCore into the WebKitOutputDir directory.
* JavaScriptCore.vcproj/WTF/WTF.vcproj: Add the StringStatics file.
* JavaScriptCore.vcproj/jsc/jsc.vcproj: Add the three WTF string cpp files to this project.
* JavaScriptCore.vcproj/jsc/jscCommon.vsprops: Remove the need to link against WTF.lib (since jsc links against JavaScriptCore).
* JavaScriptCore.xcodeproj/project.pbxproj: Add the StringStatics file.
* wtf/text/AtomicString.cpp: Moved code to StringStatics.
* wtf/text/StringImpl.cpp: Ditto.
* wtf/text/StringStatics.cpp: Added. Move functions in WTF Strings that define static variables to here, so
    the rest of the files can be compiled in WebCore.
(WebCore::StringImpl::empty): Moved from StringImpl.cpp to here.
(WebCore::AtomicString::init): Moved from AtomicString.cpp to here.

WebCore: Patch by Gavin Barraclough.

Reviewed by Steve Falkenburg.

REGRESSION (r57900-57919): 3% PLT Regression from moving strings into WTF.
https://bugs.webkit.org/show_bug.cgi?id=38930
<rdar://problem/7937188>

Add the WTF strings into the WebCore vcproj, from their copied location in $(WebKitOutputDir).

No new tests because no change in behavior.

* WebCore.vcproj/WebCore.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59171 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFixed test failures seen on SnowLeopard buildbot.
ggaren@apple.com [Tue, 11 May 2010 20:38:13 +0000 (20:38 +0000)]
Fixed test failures seen on SnowLeopard buildbot.

* runtime/JSString.cpp:
(JSC::JSString::replaceCharacter): Don't use size_t and wtf::notFound.
Instead, use the individual types and notFound values of the string APIs
we're using, since they're not necessarily the same in 64bit.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59170 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Nate Chapin <japhet@chromium.org>
japhet@chromium.org [Tue, 11 May 2010 20:32:15 +0000 (20:32 +0000)]
2010-05-11  Nate Chapin  <japhet@chromium.org>

        Reviewed by Adam Barth.

        Crash fix in WebPageSerializerImpl::serialize().

        https://bugs.webkit.org/show_bug.cgi?id=38543

        The relevant test is a test_shell_test in src.chromium.org, because neither
        DRT nor the Chromium webkit unit tests can currently cover the serializer
        functionality.

        * src/WebPageSerializerImpl.cpp:
        (WebKit::WebPageSerializerImpl::serialize): Check each frame's url before using it,
            since they are not guaranteed to be valid (e.g., if the frame was treated as a download).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59169 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Ojan Vafai <ojan@chromium.org>
ojan@chromium.org [Tue, 11 May 2010 20:11:27 +0000 (20:11 +0000)]
2010-05-10  Ojan Vafai  <ojan@chromium.org>

        Reviewed by David Levin.

        CheckStyle was eating script errors when there were local-commits and working copy changes
        https://bugs.webkit.org/show_bug.cgi?id=38880

        * Scripts/webkitpy/tool/steps/checkstyle.py:
        * Scripts/webkitpy/tool/steps/checkstyle_unittest.py: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59168 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed. Fix compiling erron on snow leopard.
jianli@chromium.org [Tue, 11 May 2010 19:55:18 +0000 (19:55 +0000)]
Unreviewed. Fix compiling erron on snow leopard.

* html/FileThreadTask.h:
(WebCore::createFileThreadTask):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59166 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] fast/frames/flattening/frameset-flattening-subframesets.html fails intermittentl...
tonikitoo@webkit.org [Tue, 11 May 2010 19:48:39 +0000 (19:48 +0000)]
[Qt] fast/frames/flattening/frameset-flattening-subframesets.html fails intermittently on Qt bot
https://bugs.webkit.org/show_bug.cgi?id=37334

Rubber-stamped by Kenneth Christiansen.
Patch by Antonio Gomes <tonikitoo@webkit.org>

Removing unneeded call to LayoutTestController.display(). It was originally intending
to force a layout, but a re-layout will happen regardless because document.body.offsetTop
is called afterwards

* fast/frames/flattening/frameset-flattening-subframe-resize.html:
* platform/qt/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59165 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed. Skip file-reader.html that is not enabled in chromium yet.
jianli@chromium.org [Tue, 11 May 2010 19:37:49 +0000 (19:37 +0000)]
Unreviewed. Skip file-reader.html that is not enabled in chromium yet.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59164 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[Qt] emit initialLayoutCompleted signal from FrameLoaderClientQt::dispatchDidFirstVis...
tonikitoo@webkit.org [Tue, 11 May 2010 19:20:22 +0000 (19:20 +0000)]
[Qt] emit initialLayoutCompleted signal from FrameLoaderClientQt::dispatchDidFirstVisuallyNonEmptyLayout
https://bugs.webkit.org/show_bug.cgi?id=38921

Reviewed by Kenneth Christiansen.
Patch by Antonio Gomes <tonikitoo@webkit.org>

Emit initialLayoutCompleted signal from FrameLoaderClientQt::dispatchDidFirstVisuallyNonEmptyLayout
instead of FrameLoaderClientQt::dispatchDidFirstLayout , because the former ensures that a
visual content layed out on the frame.

It matches to QWebFrame::initialLayoutCompleted signal documentation at:

"... This is the first time you will see contents displayed on the frame ..."

* WebCoreSupport/FrameLoaderClientQt.cpp:
(WebCore::FrameLoaderClientQt::dispatchDidFirstLayout):
(WebCore::FrameLoaderClientQt::dispatchDidFirstVisuallyNonEmptyLayout):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59163 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoExpose FileReader interface.
jianli@chromium.org [Tue, 11 May 2010 19:04:50 +0000 (19:04 +0000)]
Expose FileReader interface.
https://bugs.webkit.org/show_bug.cgi?id=38609

Reviewed by Dmitry Titov.

WebCore:

Test: fast/files/file-reader.html

* Configurations/FeatureDefines.xcconfig:
* DerivedSources.cpp:
* DerivedSources.make:
* GNUmakefile.am:
* WebCore.gypi:
* WebCore.pri:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSEventTarget.cpp:
(WebCore::toJS):
* bindings/v8/V8DOMWrapper.cpp:
(WebCore::V8DOMWrapper::convertEventTargetToV8Object):
* html/FileError.idl:
* html/FileReader.idl: Added.
* page/DOMWindow.idl:

WebKit/chromium:

* features.gypi: turn on building FileReader for Chromium.

WebKitTools:

* Scripts/build-webkit: turn on building FileReader for Apple's WebKit.

LayoutTests:

Add a layout test to test basic FileReader operations.
Also update layout tests affected by adding the constructors to DOMWindow.

* fast/dom/Window/script-tests/window-property-descriptors.js:
* fast/dom/Window/window-properties.html:
* fast/dom/script-tests/constructed-objects-prototypes.js:
* fast/dom/script-tests/prototype-inheritance-2.js:
* fast/dom/script-tests/prototype-inheritance.js:
* fast/files/file-reader-expected.txt: Added.
* fast/files/file-reader.html: Added.
* fast/files/resources/UTF16BE-BOM.txt: Added.
* fast/files/resources/UTF16BE.txt: Added.
* fast/files/resources/UTF16LE-BOM.txt: Added.
* fast/files/resources/UTF8-BOM.txt: Added.
* fast/files/resources/UTF8.txt: Added.
* fast/files/resources/empty-file: Added.
* fast/js/script-tests/global-constructors.js:
* platform/chromium/test_expectations.txt:
* platform/gtk/Skipped: Skip the new test because eventSender.beginDragWithFiles is not supported in gtk platform.
* platform/qt/Skipped: Skip the new test because eventSender.beginDragWithFiles is not supported in qt platform.
* platform/win/Skipped: Skip the new test because eventSender.beginDragWithFiles is not supported in win platform.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59162 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoStart using ropes in String.prototype.replace.
ggaren@apple.com [Tue, 11 May 2010 18:42:46 +0000 (18:42 +0000)]
Start using ropes in String.prototype.replace.

Reviewed by Oliver Hunt and Darin Adler.

1%-1.5% speedup on SunSpider.

* runtime/JSString.cpp:
(JSC::JSString::resolveRope): Updated for RopeImpl refactoring.

(JSC::JSString::replaceCharacter): Added a replaceCharacter function, which creates
a rope for the resulting replacement.

* runtime/JSString.h: A few changes here:
(JSC::):
(JSC::RopeBuilder::RopeIterator::RopeIterator):
(JSC::RopeBuilder::RopeIterator::operator++):
(JSC::RopeBuilder::RopeIterator::operator*):
(JSC::RopeBuilder::RopeIterator::operator!=):
(JSC::RopeBuilder::RopeIterator::WorkItem::WorkItem):
(JSC::RopeBuilder::RopeIterator::WorkItem::operator!=):
(JSC::RopeBuilder::RopeIterator::skipRopes): Created a RopeIterator abstraction.
We use this to do a substring find without having to resolve the rope.
(We could use this iterator when resolving ropes, too, but resolving
ropes backwards is usually more efficient.)

(JSC::RopeBuilder::JSString): Added constructors for 2 & 3 UStrings.

(JSC::RopeBuilder::appendValueInConstructAndIncrementLength):
(JSC::RopeBuilder::size): Updated for RopeImpl refactoring.

* runtime/Operations.h: Updated for RopeImpl refactoring.
(JSC::jsString): Added jsString functions for 2 & 3 UStrings.

* runtime/RopeImpl.cpp:
(JSC::RopeImpl::derefFibersNonRecursive):
* runtime/RopeImpl.h:
(JSC::RopeImpl::initializeFiber):
(JSC::RopeImpl::size):
(JSC::RopeImpl::fibers):
(JSC::RopeImpl::deref):
(JSC::RopeImpl::RopeImpl): A little refactoring to make this patch easier:
Moved statics to the top of the class; put multi-statement functions on
multiple lines; renamed "fiberCount" to "size" to match other collections;
changed the "fibers" accessor to return the fibers buffer, instead of an
item in the buffer, to make iteration easier.

* runtime/StringPrototype.cpp:
(JSC::stringProtoFuncReplace): Don't resolve a rope unless we need to. Do
use our new replaceCharacter function if possible. Do use a rope to
represent splicing three strings together.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59161 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed.
jianli@chromium.org [Tue, 11 May 2010 18:35:46 +0000 (18:35 +0000)]
Unreviewed.

[chromium] Mark flaky layout tests.

* platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59160 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] Fix build break on Chromium Mac
jam@chromium.org [Tue, 11 May 2010 17:58:28 +0000 (17:58 +0000)]
[chromium] Fix build break on Chromium Mac

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59159 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Tue, 11 May 2010 17:53:49 +0000 (17:53 +0000)]
2010-05-11  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [Gtk] Doesn't support gtk-key-themes
        https://bugs.webkit.org/show_bug.cgi?id=28257

        Add support for GTK+ key themes.

        * WebCoreSupport/EditorClientGtk.cpp:
        (WebKit::backspaceCallback): Added.
        (WebKit::cutClipboardCallback): Added.
        (WebKit::copyClipboardCallback): Added.
        (WebKit::pasteClipboardCallback): Added.
        (WebKit::selectAllCallback): Added.
        (WebKit::deleteFromCursorCallback): Added.
        (WebKit::moveCursorCallback): Added.
        (WebKit::EditorClient::generateEditorCommands): Added.
        (WebKit::EditorClient::executePendingEditorCommands): Added.
        (WebKit::EditorClient::handleKeyboardEvent): Derive editor commands both from internal text widget and a list of fallbacks.
        (WebKit::EditorClient::EditorClient): Add m_nativeWidget initializer.
        * WebCoreSupport/EditorClientGtk.h:
        (WebKit::EditorClient::addPendingEditorCommand): Add m_nativeWidget member.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59158 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Xan Lopez <xlopez@igalia.com>
xan@webkit.org [Tue, 11 May 2010 17:37:47 +0000 (17:37 +0000)]
2010-05-11  Xan Lopez  <xlopez@igalia.com>

        Reviewed by Gustavo Noronha.

        [gtk] critical warning in soup_message_set_first_party() when cancelling a navigation
        https://bugs.webkit.org/show_bug.cgi?id=38653

        Check if willSendRequest cancelled the request before setting the
        first party for the cookies in the message, otherwise we'll get a
        critical warning.

        * platform/network/soup/ResourceHandleSoup.cpp:
        (WebCore::restartedCallback):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59157 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoCheck in gyp changes for r59110
jam@chromium.org [Tue, 11 May 2010 17:11:37 +0000 (17:11 +0000)]
Check in gyp changes for r59110

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59156 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoREGRESSION (r59001): 11 media tests are failing on Windows (38847)
weinig@apple.com [Tue, 11 May 2010 16:47:20 +0000 (16:47 +0000)]
REGRESSION (r59001): 11 media tests are failing on Windows (38847)
rdar://problem/7962997
https://bugs.webkit.org/show_bug.cgi?id=38847

Patch by Jer Noble <jer.noble@apple.com> on 2010-05-11
Reviewed by Adele Peterson.

QTMovieWin was too much refactored; the cacheMovieScale() function should have
been left in QTMovie (instead of being moved into QTMovieGWorld).  This fixes the
video-size-intrinsic-scale test.

QTMovieTask::updateTaskTimer() was ignoring the maxInterval parameter unless
QTGetTimeUntilNextTask() returned an error, which caused the next timer to be
scheduled for extremely far in the future, breaking some of the layout tests.

QTMovieTask::fireTaskClients() now makes a local copy of the task list before
calling clients, in case one of them removes themselves from the task list and
invalidates our iterating pointer.

* platform/graphics/win/QTMovie.cpp:
(QTMoviePrivate::task):
(QTMoviePrivate::createMovieController):
(QTMoviePrivate::cacheMovieScale):
* platform/graphics/win/QTMovieGWorld.cpp:
(QTMovieGWorldPrivate::movieLoadStateChanged):
* platform/graphics/win/QTMovieTask.cpp:
(QTMovieTask::updateTaskTimer):
(QTMovieTask::fireTaskClients):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59155 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Patrick Gansterer <paroga@paroga.com>
eric@webkit.org [Tue, 11 May 2010 16:38:17 +0000 (16:38 +0000)]
2010-05-11  Patrick Gansterer  <paroga@paroga.com>

        Reviewed by Adam Roben.

        Port SharedBuffer to WinCE.
        https://bugs.webkit.org/show_bug.cgi?id=37222

        Use only functions which are available on Win32 and WinCE.

        * platform/win/SharedBufferWin.cpp:
        (WebCore::SharedBuffer::createWithContentsOfFile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59154 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoMerge some pseudo-valid-*.html tests into one.
tkent@chromium.org [Tue, 11 May 2010 16:14:37 +0000 (16:14 +0000)]
Merge some pseudo-valid-*.html tests into one.
https://bugs.webkit.org/show_bug.cgi?id=38846

Reviewed by Shinichiro Hamaji.

No additional test cases and no removed test cases.

* fast/css/pseudo-valid-003-expected.txt: Removed.
* fast/css/pseudo-valid-003.html: Removed.
* fast/css/pseudo-valid-005-expected.txt: Removed.
* fast/css/pseudo-valid-005.html: Removed.
* fast/css/pseudo-valid-006-expected.txt: Removed.
* fast/css/pseudo-valid-006.html: Removed.
* fast/css/pseudo-valid-unapplied-expected.txt: Added.
* fast/css/pseudo-valid-unapplied.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59153 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Jesus Sanchez-Palencia <jesus@webkit.org>
eric@webkit.org [Tue, 11 May 2010 15:45:27 +0000 (15:45 +0000)]
2010-05-11  Jesus Sanchez-Palencia  <jesus@webkit.org>

        Reviewed by Kenneth Rohde Christiansen.

        Add PageClientQGraphicsWidget specific implementation.
        QGraphicsWebViewPrivate doesn't inherit QWebPageClient anymore
        and the needed functions were moved to this new PageClient class,
        including Tiling and Accelerated Composite specific ones.

        [Qt] PageClientQt specific implementation for QGraphicsWidget
        https://bugs.webkit.org/show_bug.cgi?id=37866

        * Api/qgraphicswebview.cpp:
        (QGraphicsWebViewPrivate::QGraphicsWebViewPrivate):
        (QGraphicsWebViewPrivate::~QGraphicsWebViewPrivate):
        (QGraphicsWebViewPrivate::syncLayers):
        (QGraphicsWebViewPrivate::_q_scaleChanged):
        (QGraphicsWebView::event):
        (QGraphicsWebView::setPage):
        (QGraphicsWebView::updateGeometry):
        (QGraphicsWebView::setGeometry):
        (QGraphicsWebView::setResizesToContents):
        * WebCoreSupport/PageClientQt.cpp:
        (WebCore::PageClientQGraphicsWidget::~PageClientQGraphicsWidget):
        (WebCore::PageClientQGraphicsWidget::scroll):
        (WebCore::PageClientQGraphicsWidget::update):
        (WebCore::PageClientQGraphicsWidget::createOrDeleteOverlay):
        (WebCore::PageClientQGraphicsWidget::syncLayers):
        (WebCore::PageClientQGraphicsWidget::setRootGraphicsLayer):
        (WebCore::PageClientQGraphicsWidget::markForSync):
        (WebCore::PageClientQGraphicsWidget::updateCompositingScrollPosition):
        (WebCore::PageClientQGraphicsWidget::updateTiledBackingStoreScale):
        (WebCore::PageClientQGraphicsWidget::setInputMethodEnabled):
        (WebCore::PageClientQGraphicsWidget::inputMethodEnabled):
        (WebCore::PageClientQGraphicsWidget::setInputMethodHint):
        (WebCore::PageClientQGraphicsWidget::cursor):
        (WebCore::PageClientQGraphicsWidget::updateCursor):
        (WebCore::PageClientQGraphicsWidget::palette):
        (WebCore::PageClientQGraphicsWidget::screenNumber):
        (WebCore::PageClientQGraphicsWidget::ownerWidget):
        (WebCore::PageClientQGraphicsWidget::geometryRelativeToOwnerWidget):
        (WebCore::PageClientQGraphicsWidget::pluginParent):
        (WebCore::PageClientQGraphicsWidget::style):
        * WebCoreSupport/PageClientQt.h:
        (WebCore::QGraphicsItemOverlay::page):
        (WebCore::QGraphicsItemOverlay::boundingRect):
        (WebCore::QGraphicsItemOverlay::paint):
        (WebCore::QGraphicsItemOverlay::prepareGraphicsItemGeometryChange):
        (WebCore::PageClientQGraphicsWidget::PageClientQGraphicsWidget):
        (WebCore::PageClientQGraphicsWidget::isQWidgetClient):
        (WebCore::PageClientQGraphicsWidget::allowsAcceleratedCompositing):
        (WebCore::PageClientQGraphicsWidget::):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59152 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Benjamin Poulain <benjamin.poulain@nokia.com>
eric@webkit.org [Tue, 11 May 2010 15:15:54 +0000 (15:15 +0000)]
2010-05-11  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] fast/text/find-hidden-text.html
        https://bugs.webkit.org/show_bug.cgi?id=32922

        Remove fast/text/find-hidden-text.html from the skipped list of Qt.

        * platform/qt/Skipped:
2010-05-11  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] fast/text/find-hidden-text.html
        https://bugs.webkit.org/show_bug.cgi?id=32922

        Use the real page step for populating the QStyleOption otherwhise
        the size can be negative, which can break the QStyle used.

        * platform/qt/ScrollbarThemeQt.cpp:
        (WebCore::styleOptionSlider):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59151 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Pavel Feldman <pfeldman@chromium.org>
pfeldman@chromium.org [Tue, 11 May 2010 14:15:14 +0000 (14:15 +0000)]
2010-05-11  Pavel Feldman  <pfeldman@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: make styles rendering more user-friendly.

        https://bugs.webkit.org/show_bug.cgi?id=36747

        * inspector/front-end/ElementsPanel.js:
        (WebInspector.ElementsPanel):
        * inspector/front-end/StylesSidebarPane.js:
        (WebInspector.StylesSidebarPane.prototype._update):
        (WebInspector.StylesSidebarPane.prototype._rebuildSectionsForStyleRules):
        (WebInspector.ComputedStyleSidebarPane):
        (WebInspector.StylePropertiesSection):
        (WebInspector.StylePropertiesSection.prototype.collapse):
        (WebInspector.StylePropertiesSection.prototype.startEditingSelector):
        * inspector/front-end/inspector.css:
        (#elements-content):
        (#elements-sidebar):
        (.styles-sidebar-separator):
        (.styles-section):
        (.styles-section .header):
        (.styles-section .header .title):
        (.styles-section .header .subtitle):
        (.styles-section .header .subtitle a):
        (.styles-section .properties):
        (.styles-section.no-affect .properties li):
        (.styles-section.no-affect .properties li.editing):
        (.styles-section.expanded .properties):
        (.styles-section .properties li):
        (.styles-section .properties li.parent):
        (.styles-section .properties ol):
        (.styles-section .properties ol.expanded):
        (.styles-section .properties li.parent::before):
        (.styles-section .properties li.parent.expanded::before):
        (.styles-section .properties li .info):
        (.styles-section:hover .properties .enabled-button):
        (.styles-section .properties li.disabled .enabled-button):
        (.styles-section .properties .enabled-button):
        (.styles-section .properties .overloaded, .styles-section .properties .disabled):
        (.styles-section.computed-style .properties .disabled):
        (.styles-section .properties .implicit, .styles-section .properties .inherited):
        (.styles-section .properties .inherited):
        (.styles-section.show-inherited .properties .inherited):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59150 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 John Gregg <johnnyg@google.com>
johnnyg@google.com [Tue, 11 May 2010 13:09:04 +0000 (13:09 +0000)]
2010-05-11  John Gregg  <johnnyg@google.com>

        [chromium] Unreviewed, test expectations.

        * platform/chromium-mac/svg/filters/filteredImage-expected.checksum: Added.
        * platform/chromium-win-xp/svg/batik/text/textFeatures-expected.txt:
        * platform/chromium-win/svg/filters/filteredImage-expected.checksum: Added.
        * platform/chromium-win/svg/filters/filteredImage-expected.png: Added.
        * platform/chromium-win/svg/text/text-hkern-on-vertical-text-expected.checksum: Added.
        * platform/chromium-win/svg/text/text-hkern-on-vertical-text-expected.png: Added.
        * platform/chromium-win/svg/text/text-vkern-on-horizontal-text-expected.checksum: Added.
        * platform/chromium-win/svg/text/text-vkern-on-horizontal-text-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59149 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Nikolas Zimmermann <nzimmermann@rim.com>
zimmermann@webkit.org [Tue, 11 May 2010 12:11:09 +0000 (12:11 +0000)]
2010-05-11  Nikolas Zimmermann  <nzimmermann@rim.com>

        Reviewed by Dirk Schulze.

        Move buildLayoutInformationForTextBox() from SVGRootInlineBox in SVGInlineTextBox where it belongs
        https://bugs.webkit.org/show_bug.cgi?id=38799

        Cleanup SVGRootInlineBox/SVGInlineTextBox. Move buildLayoutInformationForTextBox to SVGInlineTextBox.
        Move lots of helper methods into new SVGTextLayoutUtilities class. Adapted all callsites.

        Fixed bug in kerning handling, hkern was applyable to vertical text as well. Covered by two new tests.

        Tests: svg/text/text-hkern-on-vertical-text.svg
               svg/text/text-vkern-on-horizontal-text.svg

        * Android.mk: Add SVGTextLayoutUtilities.cpp/h to build.
        * GNUmakefile.am: Ditto.
        * WebCore.gypi: Ditto.
        * WebCore.pro: Ditto.
        * WebCore.vcproj/WebCore.vcproj: Ditto.
        * WebCore.xcodeproj/project.pbxproj: Ditto.
        * rendering/SVGInlineTextBox.cpp:
        (WebCore::SVGInlineTextBox::calculateGlyphWidth): Use textRenderer()->characters() instead of textRenderer()->text()->characters() as shortcut.
        (WebCore::SVGInlineTextBox::svgCharacterHitsPosition): Fix style issue, no else if after a return branch.
        (WebCore::SVGInlineTextBox::paintSelection): Use !color.alpha() instead of color.alpha() == 0.
        (WebCore::SVGInlineTextBox::buildLayoutInformation): Moved from SVGRootInlineBox::buildLayoutInformationForTextBox.
        * rendering/SVGInlineTextBox.h: Fix indention.
        (WebCore::SVGInlineTextBox::selectionTop): Inlined.
        (WebCore::SVGInlineTextBox::selectionHeight): Inlined.
        * rendering/SVGRenderTreeAsText.cpp: Include SVGTextLayoutUtilities.h
        * rendering/SVGRootInlineBox.cpp: Move almost all static text layout helper functions in SVGTextLayoutUtilities
        (WebCore::calculateTextLengthCorrectionForTextChunk): Fixed style issue in long if conditions spanning multiple lines.
        (WebCore::SVGRootInlineBox::computePerCharacterLayoutInformation): Use different topLeftPositionOfCharacterRange() taking start/end iterators.
        (WebCore::SVGRootInlineBox::buildLayoutInformation): Use the moved SVGInlineTextBox::buildLayoutInformation() method.
        (WebCore::SVGRootInlineBox::buildTextChunks): Fixed small style issue, remove unneeded parenthesis.
        (WebCore::SVGRootInlineBox::layoutTextChunks): Fix style issue in debugging code.
        * rendering/SVGRootInlineBox.h: Rename 'LastGlyphInfo' to 'SVGLastGlyphInfo' and move to SVGTextLayoutUtilities.
        (WebCore::SVGRootInlineBox::svgTextChunks): Inlined for speed.
        * rendering/SVGTextLayoutUtilities.cpp: Added.
        (WebCore::isVerticalWritingMode):
        (WebCore::dominantBaselineToShift):
        (WebCore::alignmentBaselineToShift):
        (WebCore::glyphOrientationToAngle):
        (WebCore::glyphOrientationIsMultiplyOf180Degrees):
        (WebCore::applyGlyphAdvanceAndShiftRespectingOrientation):
        (WebCore::topLeftPositionOfCharacterRange):
        (WebCore::cummulatedWidthOfInlineBoxCharacterRange):
        (WebCore::cummulatedHeightOfInlineBoxCharacterRange):
        (WebCore::svgTextRunForInlineTextBox):
        (WebCore::calculateCSSKerning):
        (WebCore::applySVGKerning):
        * rendering/SVGTextLayoutUtilities.h: Added.
        (WebCore::):
        (WebCore::SVGTextPaintInfo::SVGTextPaintInfo):
        (WebCore::SVGLastGlyphInfo::SVGLastGlyphInfo):
2010-05-11  Nikolas Zimmermann  <nzimmermann@rim.com>

        Reviewed by Dirk Schulze.

        Move buildLayoutInformationForTextBox() from SVGRootInlineBox in SVGInlineTextBox where it belongs
        https://bugs.webkit.org/show_bug.cgi?id=38799

        * platform/mac/svg/text/text-hkern-on-vertical-text-expected.checksum: Added.
        * platform/mac/svg/text/text-hkern-on-vertical-text-expected.png: Added.
        * platform/mac/svg/text/text-hkern-on-vertical-text-expected.txt: Added.
        * platform/mac/svg/text/text-vkern-on-horizontal-text-expected.checksum: Added.
        * platform/mac/svg/text/text-vkern-on-horizontal-text-expected.png: Added.
        * platform/mac/svg/text/text-vkern-on-horizontal-text-expected.txt: Added.
        * svg/text/text-hkern-on-vertical-text.svg: Added.
        * svg/text/text-vkern-on-horizontal-text.svg: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59148 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore: RenderSVGImage::imageChanged should invalidate the SVG filter cache
zherczeg@webkit.org [Tue, 11 May 2010 09:22:09 +0000 (09:22 +0000)]
WebCore: RenderSVGImage::imageChanged should invalidate the SVG filter cache
https://bugs.webkit.org/show_bug.cgi?id=38838

Reviewed by Dirk Schulze.

Test: svg/filters/filteredImage.svg

* rendering/RenderSVGImage.cpp:
(WebCore::RenderSVGImage::imageChanged):

LayoutTests: Test case for SVG filter cache invalidation, after an image resource arrives.
https://bugs.webkit.org/show_bug.cgi?id=38838

Reviewed by Dirk Schulze.

* platform/gtk/Skipped:
* platform/mac/svg/filters/filteredImage-expected.checksum: Added.
* platform/mac/svg/filters/filteredImage-expected.png: Added.
* platform/mac/svg/filters/filteredImage-expected.txt: Added.
* platform/qt/svg/filters/filteredImage-expected.txt: Added.
* svg/filters/filteredImage.svg: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59147 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 08:08:21 +0000 (08:08 +0000)]
2010-05-11  Simon Fraser  <simon.fraser@apple.com>

        Test result fixes for Leopard.

        Leopard needs custom results for tests that dump the layer tree, because Leopard
        CA does not have geometry flipping, so we do that ourselves by setting a transform.

        * platform/mac-leopard/compositing/iframes/composited-parent-iframe-expected.txt: Added.
        * platform/mac-leopard/compositing/iframes/connect-compositing-iframe-expected.txt: Added.
        * platform/mac-leopard/compositing/iframes/connect-compositing-iframe2-expected.txt: Added.
        * platform/mac-leopard/compositing/iframes/connect-compositing-iframe3-expected.txt: Added.
        * platform/mac-leopard/compositing/iframes/enter-compositing-iframe-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59146 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Tony Chang <tony@chromium.org>
tony@chromium.org [Tue, 11 May 2010 07:24:08 +0000 (07:24 +0000)]
2010-05-10  Tony Chang  <tony@chromium.org>

        Reviewed by Kent Tamura.

        [chromium] don't build DRT in chromium checkouts due to a circular gyp dependency
        https://bugs.webkit.org/show_bug.cgi?id=38889

        * WebKit.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59145 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-11 John Gregg <johnnyg@google.com>
johnnyg@google.com [Tue, 11 May 2010 07:18:53 +0000 (07:18 +0000)]
2010-05-11  John Gregg  <johnnyg@google.com>

        [chromium] Update win-vista version of the baseline recently updated for win-xp.

        * platform/chromium-win-vista/svg/batik/text/textFeatures-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59144 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 06:43:57 +0000 (06:43 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Fix asserting GTK build.

        r59137 changed the behavior of RenderObject::repaintUsingContainer(). I mistakenly
        thought that non-compositing builds would always pass a 0 repaintContainer, but
        actually the RenderView is passed in this case. So use this to repaint if
        ACCELERATED_COMPOSITING is turned off.

        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::repaintUsingContainer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59143 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 06:32:36 +0000 (06:32 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Skip two new flakey tests:

        # Tests are flakey: https://bugs.webkit.org/show_bug.cgi?id=38888
        compositing/iframes/iframe-resize.html
        compositing/iframes/overlapped-iframe.html

        * platform/mac/Skipped:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59142 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 06:22:15 +0000 (06:22 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        No review; attempt to fix failing tests.

        Increase the timeout to give compositing updates a chance to happen.

        * compositing/iframes/iframe-resize.html:
        * compositing/iframes/overlapped-iframe.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59141 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 06:06:18 +0000 (06:06 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Fix warning on Windows about unreachable code.

        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::shouldPropagateCompositingToEnclosingIFrame):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59140 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 06:05:03 +0000 (06:05 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        No review.

        Reduce timeouts from 500 to 50ms to make the tests run more quickly.

        * compositing/iframes/resources/enter-compositing-subframe.html:
        * compositing/iframes/resources/leave-compositing-subframe.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59139 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 05:47:04 +0000 (05:47 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Anders Carlsson.

        Allow compositing layers to be connected across iframe boundaries on Mac
        https://bugs.webkit.org/show_bug.cgi?id=38856

        When painting an iframe into a compositing layer, the "PaintBehaviorFlattenCompositingLayers"
        flag was mistakenly set, because we'd detect that we were drawing into a bitmap. This caused
        content to show up in both compositing layers, and the painted background inside the iframe.

        Fix by taking the flattening state from the parent frame, if there is one. For the root
        frame, we continue to look to see if we're painting into a bitmap.

        * WebView/WebFrame.mm:
        (-[WebFrame _drawRect:contentsOnly:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59138 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 05:47:00 +0000 (05:47 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Anders Carlsson

        Allow compositing layers to be connected across iframe boundaries on Mac
        https://bugs.webkit.org/show_bug.cgi?id=38856

        RenderObject::repaintUsingContainer() incorrectly did a view-based
        repaint if the repaint container was the RenderView. Instead, we need
        to check to see if the RenderView's layer is composited, and, if so,
        whether it's painting into the window or not. This can occur when iframes
        are composited.

        The bug is timing-sensitive, involving compositing in iframes, and I was not able to
        concoct a good testcase.

        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::repaintUsingContainer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59137 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 05:44:06 +0000 (05:44 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Anders Carlsson

        Allow compositing layers to be connected across iframe boundaries on Mac
        https://bugs.webkit.org/show_bug.cgi?id=38856

        Changes to allow compositing layers for iframes to switch between being hosted
        by the iframe's layer-backed NSView, and parented in the GraphicsLayer tree of the
        enclosing document.

        Tests: compositing/iframes/connect-compositing-iframe.html
               compositing/iframes/connect-compositing-iframe2.html
               compositing/iframes/connect-compositing-iframe3.html

        * page/FrameView.h:
        * page/FrameView.cpp:
        (WebCore::FrameView::hasCompositedContent): New convenience method.
        (WebCore::FrameView::setIsOverlapped): If we're composited, poke the owner document in case it
            wants to re-evaluate compositing decisions.
        (WebCore::FrameView::isOverlapped): Just expose the existing flag.

        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::shouldBeNormalFlowOnly):
        (WebCore::RenderLayer::isSelfPaintingLayer):

        * rendering/RenderLayerBacking.cpp:
        (WebCore::RenderLayerBacking::updateGraphicsLayerConfiguration): If this is an iframe, we need
            to ensure that the layers for the iframe content are hooked up.
        (WebCore::RenderLayerBacking::updateDrawsContent): When an iframe toggles between different
            attachments, the 'drawsContent' behavior of its root layer changes, so needs to be updated.

        * rendering/RenderLayerCompositor.h:
        (WebCore::RenderLayerCompositor::updateCompositingLayers): Call destroyRootPlatformLayer()
            instead of detachRootPlatformLayer() and manually zeroing out the OwnPtrs.
        (WebCore::RenderLayerCompositor::updateBacking): If a RenderIFrame changes compositing mode,
            we need to ensure that its content compositor attachment is updated.
        (WebCore::RenderLayerCompositor::repaintOnCompositingChange): The existing code had a bug
            that caused repaints for RenderViews (which have no parent) to bail. We only want to bail
            for non-RenderViews that are not attached.
        (WebCore::RenderLayerCompositor::rebuildCompositingLayerTree): Factored the iframe-connecting
            code into a new method, parentIFrameContentLayers().
        (WebCore::RenderLayerCompositor::parentIFrameContentLayers): New method to share the code that hooks
            up the iframe's compositing layers to the parent.
        (WebCore::RenderLayerCompositor::shouldPropagateCompositingToEnclosingIFrame): Add logic to propagate
            compositing out of iframes on Mac in two situations: 1) when the FrameView is overlapped, and 2)
            if the parent document is already composited.
        (WebCore::RenderLayerCompositor::ensureRootPlatformLayer): Clean up the logic here to better deal
            with dynamic changes of the attachment type.
        (WebCore::RenderLayerCompositor::destroyRootPlatformLayer): Clean up and null out the clipping layer here.
        (WebCore::RenderLayerCompositor::attachRootPlatformLayer): Call rootLayerAttachmentChanged().
        (WebCore::RenderLayerCompositor::detachRootPlatformLayer): Ditto. Also unparent the clipping and platform layers.
        (WebCore::RenderLayerCompositor::updateRootLayerAttachment): Call ensureRootPlatformLayer() to re-evaluate
            the layer attachment.
        (WebCore::RenderLayerCompositor::rootLayerAttachmentChanged): We need to update the drawsContent() status
            of the RenderView's layer's backing, because it changes depending on the attachment.

        * rendering/RenderWidget.cpp:
        (WebCore::RenderWidget::paint): Do overlap testing if the frameView can do fast repaints (as before),
            but also now when the frameView has composited content.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59136 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdding a test that makes sure openDatabaseSync() deals correctly with invalid inputs.
dumi@chromium.org [Tue, 11 May 2010 05:37:42 +0000 (05:37 +0000)]
Adding a test that makes sure openDatabaseSync() deals correctly with invalid inputs.
https://bugs.webkit.org/show_bug.cgi?id=34994

Reviewed by Brady Eidson.

* fast/workers/storage: Added.
* fast/workers/storage/open-database-sync-inputs-expected.txt: Added.
* fast/workers/storage/open-database-sync-inputs.html: Added.
* fast/workers/storage/resources: Added.
* fast/workers/storage/resources/open-database-sync-inputs.js: Added.
(notAString.toString):
* fast/workers/storage/resources/test-inputs-common.js: Added.
(log):
(finishTest):
(runTest.worker.onmessage):
(runTest):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59135 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 05:34:48 +0000 (05:34 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Anders Carlsson

        Allow compositing layers to be connected across iframe boundaries on Mac
        https://bugs.webkit.org/show_bug.cgi?id=38856

        Rename the static shouldPropagateCompositingToIFrameParent() to shouldPropagateCompositingToEnclosingIFrame(),
        to pave the way for runtime switches in the propagation behavior. We have to make sure we call it on
        the correct RenderLayerCompositor (that belonging to the iframe's content document).

        * rendering/RenderLayerBacking.cpp:
        (WebCore::RenderLayerBacking::paintingGoesToWindow): Now that we know the root layer attachment, we
        can simplify this method.

        * rendering/RenderLayerCompositor.h:
        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::enableCompositingMode): No need for the setNeedsStyleRecalc()
        here, because the ensureRootPlatformLayer() or destroyRootPlatformLayer() will have already done it.

        (WebCore::RenderLayerCompositor::rebuildCompositingLayerTree): Use the root layer attachment to determine
        whether to parent the iframe's layers.

        (WebCore::RenderLayerCompositor::didMoveOnscreen): Method name change.
        (WebCore::RenderLayerCompositor::shouldPropagateCompositingToEnclosingIFrame): Name change.
        (WebCore::RenderLayerCompositor::requiresCompositingForIFrame): We need to consult the iframe contents
        document's compositor to ask whether propagation is appropriate.
        (WebCore::RenderLayerCompositor::ensureRootPlatformLayer): Name change.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59134 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 05:33:23 +0000 (05:33 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Anders Carlsson

        Allow compositing layers to be connected across iframe boundaries on Mac
        https://bugs.webkit.org/show_bug.cgi?id=38856

        Use an enum for the type of root layer attachment on a RenderLayerCompositor, so we can
        determine if the attachment is via the ChromeClient, via an enclosing iframe, or unattached.

        * rendering/RenderLayerCompositor.h: New RootLayerAttachment enum.
        (WebCore::RenderLayerCompositor::rootLayerAttachment): getter for the current attachment.

        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::RenderLayerCompositor): Init m_rootLayerAttachment.
        (WebCore::RenderLayerCompositor::~RenderLayerCompositor): Assert that is unattached.
        (WebCore::RenderLayerCompositor::updateCompositingLayers): Call destroyRootPlatformLayer instead of willMoveOffscreen,
        when there are no layers left. Also clear out the clipping layer.
        (WebCore::RenderLayerCompositor::didMoveOnscreen): Call attachRootPlatformLayer.
        (WebCore::RenderLayerCompositor::willMoveOffscreen): Call ensureRootPlatformLayer with the appropriate attachment.
        (WebCore::RenderLayerCompositor::ensureRootPlatformLayer): Only create the m_rootPlatformLayer if we don't have one
           already, but be sure to always set the root layer geometry orientation. Also only create the
           m_clippingLayer if we need to.
        (WebCore::RenderLayerCompositor::destroyRootPlatformLayer): Call detachRootPlatformLayer().
        (WebCore::RenderLayerCompositor::attachRootPlatformLayer): Code moved from didMoveOnscreen, but switching on
            attachment.
        (WebCore::RenderLayerCompositor::detachRootPlatformLayer): Code moved from willMoveOffscreen, but switching on
            attachment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59133 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 05:27:13 +0000 (05:27 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Anders Carlsson

        Allow compositing layers to be connected across iframe boundaries on Mac
        https://bugs.webkit.org/show_bug.cgi?id=38856

        Clean up the geometry logic when propagating compositing out of iframes.

        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::updateRootLayerPosition): Geometry on m_rootPlatformLayer no longer
        needs to be special when there's a clipping layer.

        (WebCore::RenderLayerCompositor::ensureRootPlatformLayer): Don't set the clipping layer to have
        flipped geometry ever, and only set the root layer to have flipped geometry if it is not
        being hosted in an iframe. Also no need to set a custom anchor point on the clipping layer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59132 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 05:22:31 +0000 (05:22 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Fix up test results from last commit, which were generated with a build from before r59129.

        * platform/win/compositing/iframes/composited-parent-iframe-expected.txt:
        * platform/win/compositing/iframes/iframe-resize-expected.txt:
        * platform/win/compositing/iframes/leave-compositing-iframe-expected.txt:
        * platform/win/compositing/iframes/overlapped-iframe-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59131 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 05:12:50 +0000 (05:12 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Anders Carlsson.

        https://bugs.webkit.org/show_bug.cgi?id=38856
        Allow compositing layers to be connected across iframe boundaries on Mac

        Add several more iframe-compositing-related tests. These tests:
            * an iframe dynamically entering compositing mode
            * an iframe dynamically leaving compositing mode
            * an overlapped iframe
            * a compositing iframe being resized

        composited-iframe.html had no value and is removed.

        * compositing/iframes/composited-iframe-expected.txt: Removed.
        * compositing/iframes/composited-iframe.html: Removed.
        * compositing/iframes/enter-compositing-iframe-expected.txt: Copied from LayoutTests/platform/mac/compositing/iframes/composited-iframe-expected.txt.
        * compositing/iframes/enter-compositing-iframe.html: Added.
        * compositing/iframes/iframe-resize-expected.txt: Copied from LayoutTests/platform/mac/compositing/iframes/composited-iframe-expected.txt.
        * compositing/iframes/iframe-resize.html: Added.
        * compositing/iframes/leave-compositing-iframe-expected.txt: Copied from LayoutTests/platform/mac/compositing/iframes/composited-iframe-expected.txt.
        * compositing/iframes/leave-compositing-iframe.html: Added.
        * compositing/iframes/overlapped-iframe-expected.txt: Copied from LayoutTests/platform/mac/compositing/iframes/composited-iframe-expected.txt.
        * compositing/iframes/overlapped-iframe.html: Added.
        * compositing/iframes/resources/enter-compositing-subframe.html: Added.
        * compositing/iframes/resources/leave-compositing-subframe.html: Added.
        * platform/win/compositing/iframes/iframe-resize-expected.txt: Added.
        * platform/win/compositing/iframes/leave-compositing-iframe-expected.txt: Added.
        * platform/win/compositing/iframes/overlapped-iframe-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59130 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Tue, 11 May 2010 05:08:18 +0000 (05:08 +0000)]
2010-05-10  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Anders Carlsson

        Allow compositing layers to be connected across iframe boundaries on Mac
        https://bugs.webkit.org/show_bug.cgi?id=38856

        Clean up some methods related to composited iframes. No behavioral changes.

        * rendering/RenderLayerBacking.h: Move innerRenderLayerCompositor() to be a static method:
            RenderLayerCompositor::iframeContentsCompositor().
        * rendering/RenderLayerBacking.cpp: Remove innerRenderLayerCompositor().
        (WebCore::RenderLayerBacking::updateGraphicsLayerGeometry): Call iframeContentsCompositor().

        * rendering/RenderLayerCompositor.cpp:
        (WebCore::RenderLayerCompositor::enableCompositingMode): Call enclosingIFrameElement()
        (WebCore::RenderLayerCompositor::rebuildCompositingLayerTree): iframeContentsCompositor() is now a static method in this class.
        Also check that the iframe contents are in compositing mode (slightly clearer than relying on rootPlatformLayer() returning
        null).
        (WebCore::RenderLayerCompositor::iframeContentsCompositor): Moved from RenderLayerBacking.
        (WebCore::RenderLayerCompositor::didMoveOnscreen): Use enclosingIFrameElement() method.
        (WebCore::RenderLayerCompositor::willMoveOffscreen): Ditto.
        (WebCore::RenderLayerCompositor::enclosingIFrameElement): New method.
        (WebCore::RenderLayerCompositor::ensureRootPlatformLayer): Call enclosingIFrameElement().
        * rendering/RenderLayerCompositor.h: New method, iframeContentsCompositor().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59129 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 John Gregg <johnnyg@google.com>
johnnyg@google.com [Tue, 11 May 2010 05:06:57 +0000 (05:06 +0000)]
2010-05-10  John Gregg  <johnnyg@google.com>

        [chromium] Final set of baselines and expectations to green up the build.

        * platform/chromium-mac/plugins/geturlnotify-during-document-teardown-expected.txt: Added.
        * platform/chromium-win/plugins/geturlnotify-during-document-teardown-expected.txt: Added.
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59128 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 John Gregg <johnnyg@google.com>
johnnyg@google.com [Tue, 11 May 2010 04:49:38 +0000 (04:49 +0000)]
2010-05-10  John Gregg  <johnnyg@google.com>

        [chromium] rebaseline SVG test for windows after recent changes.

        * platform/chromium-win/svg/W3C-SVG-1.1/filters-blend-01-b-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-blend-01-b-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-blend-01-b-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-color-01-b-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-color-01-b-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-color-01-b-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-composite-02-b-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-composite-02-b-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-composite-02-b-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-comptran-01-b-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-comptran-01-b-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-comptran-01-b-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-diffuse-01-f-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-displace-01-f-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-displace-01-f-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-displace-01-f-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-example-01-b-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-example-01-b-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-example-01-b-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-felem-01-b-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-gauss-01-b-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-gauss-01-b-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-gauss-01-b-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-image-01-b-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-image-01-b-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-image-01-b-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-light-01-f-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-morph-01-f-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-morph-01-f-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-morph-01-f-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-offset-01-b-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-offset-01-b-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-offset-01-b-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-specular-01-f-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-tile-01-b-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-tile-01-b-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-tile-01-b-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/filters-turb-01-f-expected.txt:
        * platform/chromium-win/svg/W3C-SVG-1.1/masking-intro-01-f-expected.txt:
        * platform/chromium-win/svg/batik/filters/feTile-expected.checksum:
        * platform/chromium-win/svg/batik/filters/feTile-expected.png:
        * platform/chromium-win/svg/batik/filters/feTile-expected.txt:
        * platform/chromium-win/svg/batik/filters/filterRegions-expected.checksum:
        * platform/chromium-win/svg/batik/filters/filterRegions-expected.png:
        * platform/chromium-win/svg/batik/filters/filterRegions-expected.txt:
        * platform/chromium-win/svg/batik/masking/maskRegions-expected.txt:
        * platform/chromium-win/svg/batik/text/textEffect-expected.checksum:
        * platform/chromium-win/svg/batik/text/textEffect-expected.png: Added.
        * platform/chromium-win/svg/batik/text/textEffect-expected.txt:
        * platform/chromium-win/svg/batik/text/textEffect3-expected.checksum:
        * platform/chromium-win/svg/batik/text/textEffect3-expected.png: Added.
        * platform/chromium-win/svg/batik/text/textEffect3-expected.txt:
        * platform/chromium-win/svg/batik/text/textFeatures-expected.txt:
        * platform/chromium-win/svg/batik/text/textProperties-expected.txt:
        * platform/chromium-win/svg/custom/empty-mask-expected.txt:
        * platform/chromium-win/svg/custom/empty-merge-expected.checksum:
        * platform/chromium-win/svg/custom/empty-merge-expected.png:
        * platform/chromium-win/svg/custom/feComponentTransfer-Discrete-expected.checksum:
        * platform/chromium-win/svg/custom/feComponentTransfer-Discrete-expected.png:
        * platform/chromium-win/svg/custom/feComponentTransfer-Discrete-expected.txt:
        * platform/chromium-win/svg/custom/feComponentTransfer-Gamma-expected.checksum:
        * platform/chromium-win/svg/custom/feComponentTransfer-Gamma-expected.png:
        * platform/chromium-win/svg/custom/feComponentTransfer-Gamma-expected.txt:
        * platform/chromium-win/svg/custom/feComponentTransfer-Linear-expected.checksum:
        * platform/chromium-win/svg/custom/feComponentTransfer-Linear-expected.png:
        * platform/chromium-win/svg/custom/feComponentTransfer-Linear-expected.txt:
        * platform/chromium-win/svg/custom/feComponentTransfer-Table-expected.checksum:
        * platform/chromium-win/svg/custom/feComponentTransfer-Table-expected.png:
        * platform/chromium-win/svg/custom/feComponentTransfer-Table-expected.txt:
        * platform/chromium-win/svg/custom/feDisplacementMap-01-expected.checksum:
        * platform/chromium-win/svg/custom/feDisplacementMap-01-expected.png:
        * platform/chromium-win/svg/custom/image-with-transform-clip-filter-expected.checksum:
        * platform/chromium-win/svg/custom/image-with-transform-clip-filter-expected.png:
        * platform/chromium-win/svg/custom/image-with-transform-clip-filter-expected.txt:
        * platform/chromium-win/svg/custom/invalid-css-expected.txt:
        * platform/chromium-win/svg/custom/mask-invalidation-expected.txt:
        * platform/chromium-win/svg/custom/non-opaque-filters-expected.txt:
        * platform/chromium-win/svg/custom/resource-invalidate-on-target-update-expected.checksum:
        * platform/chromium-win/svg/custom/resource-invalidate-on-target-update-expected.png:
        * platform/chromium-win/svg/custom/text-filter-expected.checksum:
        * platform/chromium-win/svg/custom/text-filter-expected.png:
        * platform/chromium-win/svg/custom/text-filter-expected.txt:
        * platform/chromium-win/svg/custom/visibility-override-filter-expected.checksum:
        * platform/chromium-win/svg/custom/visibility-override-filter-expected.png:
        * platform/chromium-win/svg/filters/big-sized-filter-2-expected.checksum:
        * platform/chromium-win/svg/filters/big-sized-filter-2-expected.png:
        * platform/chromium-win/svg/filters/big-sized-filter-expected.checksum:
        * platform/chromium-win/svg/filters/big-sized-filter-expected.png:
        * platform/chromium-win/svg/filters/feColorMatrix-values-expected.checksum:
        * platform/chromium-win/svg/filters/feColorMatrix-values-expected.png:
        * platform/chromium-win/svg/filters/feColorMatrix-values-expected.txt:
        * platform/chromium-win/svg/filters/feComposite-expected.checksum:
        * platform/chromium-win/svg/filters/feComposite-expected.png:
        * platform/chromium-win/svg/filters/feComposite-expected.txt:
        * platform/chromium-win/svg/filters/feGaussianBlur-expected.checksum:
        * platform/chromium-win/svg/filters/feGaussianBlur-expected.png:
        * platform/chromium-win/svg/filters/feMerge-expected.checksum:
        * platform/chromium-win/svg/filters/feMerge-expected.png:
        * platform/chromium-win/svg/filters/feOffset-expected.checksum:
        * platform/chromium-win/svg/filters/feOffset-expected.png:
        * platform/chromium-win/svg/filters/feTile-expected.checksum:
        * platform/chromium-win/svg/filters/feTile-expected.png:
        * platform/chromium-win/svg/filters/filter-on-tspan-expected.checksum:
        * platform/chromium-win/svg/filters/filter-on-tspan-expected.png:
        * platform/chromium-win/svg/filters/filter-on-tspan-expected.txt:
        * platform/chromium-win/svg/filters/shadow-on-filter-expected.checksum:
        * platform/chromium-win/svg/filters/shadow-on-filter-expected.png:
        * platform/chromium-win/svg/filters/sourceAlpha-expected.txt:
        * platform/chromium-win/svg/filters/subRegion-in-userSpace-expected.checksum:
        * platform/chromium-win/svg/filters/subRegion-in-userSpace-expected.png:
        * platform/chromium-win/svg/filters/subRegion-one-effect-expected.checksum:
        * platform/chromium-win/svg/filters/subRegion-one-effect-expected.png:
        * platform/chromium-win/svg/filters/subRegion-two-effects-expected.checksum:
        * platform/chromium-win/svg/filters/subRegion-two-effects-expected.png:
        * platform/chromium-win/svg/text/text-vkern-expected.checksum: Added.
        * platform/chromium-win/svg/text/text-vkern-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59127 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed, fixing the expectations for Chromium.
dumi@chromium.org [Tue, 11 May 2010 04:12:22 +0000 (04:12 +0000)]
Unreviewed, fixing the expectations for Chromium.

* platform/chromium/http/tests/security/cross-frame-access-call-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59126 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix a crash when closing a WebView while the Web Inspector is open.
timothy@apple.com [Tue, 11 May 2010 03:51:02 +0000 (03:51 +0000)]
Fix a crash when closing a WebView while the Web Inspector is open.

<rdar://problem/7966830>

Reviewed by Mark Rowe.

* WebCoreSupport/WebInspectorClient.mm:
(-[WebInspectorWindowController destroyInspectorView]): Null check Page since it can be null.
* WebView/WebView.mm:
(-[WebView _close]): Null out _private->page before deleting the page, so code called
during destruction don't access a half deleted Page object.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59125 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Tony Chang <tony@chromium.org>
tony@chromium.org [Tue, 11 May 2010 03:45:56 +0000 (03:45 +0000)]
2010-05-10  Tony Chang  <tony@chromium.org>

        Reviewed by Kent Tamura.

        [chromium] Build DRT when running build-webkit --chromium
        https://bugs.webkit.org/show_bug.cgi?id=38730

        This allows DRT to use the same .o and .lib files when building
        on Windows.

        * WebKit.gyp: Move DRT.gyp targets into this file.
        * gyp_webkit: DRT.gyp is gone
2010-05-10  Tony Chang  <tony@chromium.org>

        Reviewed by Kent Tamura.

        [chromium] Build DRT when running build-webkit --chromium
        https://bugs.webkit.org/show_bug.cgi?id=38730

        * DumpRenderTree/DumpRenderTree.gyp/DumpRenderTree.gyp: Removed.
        * Scripts/build-dumprendertree: This does nothing now.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59124 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Eric Seidel <eric@webkit.org>
eric@webkit.org [Tue, 11 May 2010 03:40:24 +0000 (03:40 +0000)]
2010-05-10  Eric Seidel  <eric@webkit.org>

        Unreviewed, just removing newline from -expected.txt file

        REGRESSION(r59100): "Added test is broken on many platforms." (Requested by eseidel on #webkit).
        https://bugs.webkit.org/show_bug.cgi?id=38881

        Looks like http://trac.webkit.org/changeset/59119
        added an extra newline by mistake.  Removing.

        * plugins/geturlnotify-during-document-teardown-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59123 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed, fixing Qt's expectations.
dumi@chromium.org [Tue, 11 May 2010 02:54:25 +0000 (02:54 +0000)]
Unreviewed, fixing Qt's expectations.

* platform/qt/http/tests/security/cross-frame-access-call-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59122 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRemove conflict markers from ChangeLog
abarth@webkit.org [Tue, 11 May 2010 02:39:33 +0000 (02:39 +0000)]
Remove conflict markers from ChangeLog

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59121 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoREGRESSION(r59100): Added test is broken on many platforms.
jhoneycutt@apple.com [Tue, 11 May 2010 02:35:56 +0000 (02:35 +0000)]
REGRESSION(r59100): Added test is broken on many platforms.
https://bugs.webkit.org/show_bug.cgi?id=38881

Reviewed by Eric Seidel.

* platform/mac-snowleopard/Skipped:
Skipped, because we're not logging NPP_Destroy on Snow Leopard.

* platform/qt/Skipped:
Skipped due to lack of support for setCallCloseOnWebViews on
LayoutTestController.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59120 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoREGRESSION(r59100): Added test is broken on many platforms.
jhoneycutt@apple.com [Tue, 11 May 2010 02:35:35 +0000 (02:35 +0000)]
REGRESSION(r59100): Added test is broken on many platforms.
https://bugs.webkit.org/show_bug.cgi?id=38881

Reviewed by Eric Seidel.

WebKitTools:

* DumpRenderTree/win/TestNetscapePlugin/main.cpp:
(NPP_Destroy):
Use pluginLog, rather than printf, to match other platforms.

LayoutTests:

* plugins/geturlnotify-during-document-teardown-expected.txt:
Update the expected result.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59119 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebCore: Cleaning up and fixing the existing JSC DB bindings.
dumi@chromium.org [Tue, 11 May 2010 02:06:11 +0000 (02:06 +0000)]
WebCore: Cleaning up and fixing the existing JSC DB bindings.
https://bugs.webkit.org/show_bug.cgi?id=34994

Reviewed by Brady Eidson.

* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::JSDOMWindow::openDatabase):
* bindings/js/JSDatabaseCustom.cpp:
(WebCore::JSDatabase::changeVersion):
(WebCore::createTransaction):
(WebCore::JSDatabase::transaction):
(WebCore::JSDatabase::readTransaction):
* bindings/js/JSSQLTransactionCustom.cpp:
(WebCore::JSSQLTransaction::executeSql):

LayoutTests: Fixing the expectations for openDatabase() in cross-frame-access-call.html.
Related to changes in patches #6 and #7 at https://bugs.webkit.org/show_bug.cgi?id=34994.

Reviewed by Brady Eidson.

* http/tests/security/cross-frame-access-call-expected.txt:
* http/tests/security/cross-frame-access-call.html:
* platform/chromium-linux/http/tests/security/cross-frame-access-call-expected.txt: Removed.
* platform/chromium-mac/http/tests/security/cross-frame-access-call-expected.txt: Removed.
* platform/chromium-win/http/tests/security/cross-frame-access-call-expected.txt: Removed.
* platform/chromium/http: Added.
* platform/chromium/http/tests: Added.
* platform/chromium/http/tests/security: Added.
* platform/chromium/http/tests/security/cross-frame-access-call-expected.txt: Copied from LayoutTests/platform/chromium-win/http/tests/security/cross-frame-access-call-expected.txt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59118 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Tue, 11 May 2010 01:52:33 +0000 (01:52 +0000)]
2010-05-10  Dirk Pranke  <dpranke@chromium.org>

        Unreviewed, build fix.

        Fix a build failure caused by assuming the default platform in
        a unit test for new-run-webkit-tests instead of specifying
        --platform test.

        * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59117 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Fumitoshi Ukai <ukai@chromium.org>
ukai@chromium.org [Tue, 11 May 2010 01:50:38 +0000 (01:50 +0000)]
2010-05-10  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Alexey Proskuryakov.

        WebSocket needs to suspend/resume as Active DOM object.
        https://bugs.webkit.org/show_bug.cgi?id=38171

        Implement suspend()/resume() in WebSocket and WebSocketChannel.
        While WebSocketChannel is suspended, it only adds received data in m_buffer
        or record the handle was closed, and report no event to WebSocket.
        When resumed, it will process buffer or handle closing.
        Since suspend/resume would be called while processing JavaScript event handler (e.g. before/after alert()), WebSocketChannel method that would fire an event need to be reentrant.
        So, WebSocketChannel::processBuffer() call WebSocket to fire an event at most once and skips buffer before the calling,
        so that next call of processBuffer() would process the next frame.

        * websockets/ThreadableWebSocketChannel.h:
        * websockets/ThreadableWebSocketChannelClientWrapper.h:
        (WebCore::ThreadableWebSocketChannelClientWrapper::didConnect):
          Mark channel is opened, and process pending events if not suspended.
        (WebCore::ThreadableWebSocketChannelClientWrapper::didReceiveMessage):
          Push message back in pending queue, and process pending events if not suspended.
        (WebCore::ThreadableWebSocketChannelClientWrapper::didClose):
          Mark channel is closed, and process pending events if not suspended.
        (WebCore::ThreadableWebSocketChannelClientWrapper::suspend):
          Mark suspended.
        (WebCore::ThreadableWebSocketChannelClientWrapper::resume):
          Unmark suspended, and process pending events.
        (WebCore::ThreadableWebSocketChannelClientWrapper::ThreadableWebSocketChannelClientWrapper):
        (WebCore::ThreadableWebSocketChannelClientWrapper::processPendingEvents):
        * websockets/WebSocket.cpp:
        (WebCore::WebSocket::canSuspend):
        (WebCore::WebSocket::suspend):
        (WebCore::WebSocket::resume):
        (WebCore::WebSocket::didClose):
        * websockets/WebSocket.h:
        * websockets/WebSocketChannel.cpp:
        (WebCore::WebSocketChannel::WebSocketChannel):
        (WebCore::WebSocketChannel::connect):
        (WebCore::WebSocketChannel::send):
        (WebCore::WebSocketChannel::bufferedAmount):
        (WebCore::WebSocketChannel::close):
        (WebCore::WebSocketChannel::suspend):
        (WebCore::WebSocketChannel::resume):
         When resumed, it will process buffer and
         handle closing if handle was already closed while suspended.
        (WebCore::WebSocketChannel::didClose):
         If suspended, record unhandled bufferedAmount and set m_closed true, so that closing will be processed when resumed.
        (WebCore::WebSocketChannel::didReceiveData):
         Add received data in buffer and process buffer while it is not suspended.
        (WebCore::WebSocketChannel::processBuffer):
         Process handshake header or one frame message.
         Return true if there are more data to be processed.
         Return false otherwise (e.g. incomplete handshake header or incomplete frame).
        * websockets/WebSocketChannel.h:
        * websockets/WorkerThreadableWebSocketChannel.cpp:
        (WebCore::WorkerThreadableWebSocketChannel::suspend):
        (WebCore::WorkerThreadableWebSocketChannel::resume):
        (WebCore::WorkerThreadableWebSocketChannel::Peer::suspend):
        (WebCore::WorkerThreadableWebSocketChannel::Peer::resume):
        (WebCore::WorkerThreadableWebSocketChannel::mainThreadSuspend):
        (WebCore::WorkerThreadableWebSocketChannel::Bridge::suspend):
        (WebCore::WorkerThreadableWebSocketChannel::mainThreadResume):
        (WebCore::WorkerThreadableWebSocketChannel::Bridge::resume):
        * websockets/WorkerThreadableWebSocketChannel.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59116 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 John Gregg <johnnyg@google.com>
johnnyg@google.com [Tue, 11 May 2010 01:45:16 +0000 (01:45 +0000)]
2010-05-10  John Gregg  <johnnyg@google.com>

        [chromium] Fix sort order for expectations after previous change.

        * platform/chromium/fast/dom/prototype-inheritance-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59115 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Dirk Pranke <dpranke@chromium.org>
dpranke@chromium.org [Tue, 11 May 2010 01:30:18 +0000 (01:30 +0000)]
2010-05-10  Dirk Pranke  <dpranke@chromium.org>

        Unreviewed, build-fix

        new-run-webkit-tests: fix test failure caused by me failing to update
        the expected output. Also, run '--platform test' instead of
        'platform dryrun' since the former is guaranteed to work and the
        latter isn't.

        * Scripts/webkitpy/layout_tests/run_webkit_tests_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59114 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoImplement the JSC bindings for the sync DB API.
dumi@chromium.org [Tue, 11 May 2010 01:18:42 +0000 (01:18 +0000)]
Implement the JSC bindings for the sync DB API.
https://bugs.webkit.org/show_bug.cgi?id=34994

Reviewed by Brady Eidson.

* bindings/js/JSDatabaseSyncCustom.cpp:
(WebCore::JSDatabaseSync::changeVersion):
(WebCore::createTransaction):
* bindings/js/JSSQLTransactionSyncCustom.cpp:
(WebCore::JSSQLTransactionSync::executeSql):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59113 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 John Gregg <johnnyg@google.com>
johnnyg@google.com [Tue, 11 May 2010 01:07:32 +0000 (01:07 +0000)]
2010-05-10  John Gregg  <johnnyg@google.com>

        [chromium] Unreviewed, test expectation change for Chromium following r59010.

        * platform/chromium/fast/dom/prototype-inheritance-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59112 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 Dean Jackson <dino@apple.com>
dino@apple.com [Tue, 11 May 2010 00:44:58 +0000 (00:44 +0000)]
2010-05-10  Dean Jackson  <dino@apple.com>

        Reviewed by Simon Fraser.

        https://bugs.webkit.org/show_bug.cgi?id=36566
        The animation-fill-mode property was missing from
        a few places where the animation shorthand is set up.

        Test: animations/animation-shorthand-removed.html

        * css/CSSMutableStyleDeclaration.cpp:
        (WebCore::CSSMutableStyleDeclaration::getPropertyValue): Make sure
        animation-fill-mode is included in the animation shorthand
        * css/CSSPropertyLonghand.cpp:
        (WebCore::initShorthandMap):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59111 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2010-05-10 John Abd-El-Malek <jam@chromium.org>
jam@chromium.org [Tue, 11 May 2010 00:10:07 +0000 (00:10 +0000)]
2010-05-10  John Abd-El-Malek  <jam@chromium.org>

        Reviewed by Darin Fisher.

        [chromium] Add WebKitScrollbar interface to allow Chromium code to reuse the scrollbar code
        https://bugs.webkit.org/show_bug.cgi?id=38854

        * public/WebScrollbar.h: Added.
        (WebKit::WebScrollbar::):
        (WebKit::WebScrollbar::~WebScrollbar):
        * public/WebScrollbarClient.h: Added.
        * public/WebThemeEngine.h:
        * src/AssertMatchingEnums.cpp:
        * src/WebScrollbarImpl.cpp: Added.
        (WebKit::WebScrollbar::create):
        (WebKit::WebScrollbar::defaultThickness):
        (WebKit::WebScrollbarImpl::WebScrollbarImpl):
        (WebKit::WebScrollbarImpl::~WebScrollbarImpl):
        (WebKit::WebScrollbarImpl::setLocation):
        (WebKit::WebScrollbarImpl::value):
        (WebKit::WebScrollbarImpl::setValue):
        (WebKit::WebScrollbarImpl::setDocumentSize):
        (WebKit::WebScrollbarImpl::scroll):
        (WebKit::WebScrollbarImpl::paint):
        (WebKit::WebScrollbarImpl::handleInputEvent):
        (WebKit::WebScrollbarImpl::onMouseDown):
        (WebKit::WebScrollbarImpl::onMouseUp):
        (WebKit::WebScrollbarImpl::onMouseMove):
        (WebKit::WebScrollbarImpl::onMouseLeave):
        (WebKit::WebScrollbarImpl::onMouseWheel):
        (WebKit::WebScrollbarImpl::onKeyDown):
        (WebKit::WebScrollbarImpl::valueChanged):
        (WebKit::WebScrollbarImpl::invalidateScrollbarRect):
        (WebKit::WebScrollbarImpl::isActive):
        (WebKit::WebScrollbarImpl::scrollbarCornerPresent):
        (WebKit::WebScrollbarImpl::getTickmarks):
        * src/WebScrollbarImpl.h: Added.
        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::scrollViewWithKeyboard):
        (WebKit::WebViewImpl::mapKeyCodeForScroll):
        * src/WebViewImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@59110 268f45cc-cd09-0410-ab3c-d52691b4dbfc