1. 22 Sep, 2010 40 commits
    • jamesr@google.com's avatar
      2010-09-22 James Robinson <jamesr@chromium.org> · 9e0aea63
      jamesr@google.com authored
              [chromium] Re-synchronizes features.gypi with upstream feature_overrides.gypi.
      
              * features.gypi:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68113 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      9e0aea63
    • jamesr@google.com's avatar
      2010-09-22 Ruben <chromium@hybridsource.org> · cab56bd8
      jamesr@google.com authored
              Reviewed by Tony Chang.
      
              [chromium] added ifdefs for FreeBSD support
              https://bugs.webkit.org/show_bug.cgi?id=46316
      
              * bindings/js/JSInspectorFrontendHostCustom.cpp:
              (WebCore::JSInspectorFrontendHost::platform):
              * bindings/v8/custom/V8InspectorFrontendHostCustom.cpp:
              (WebCore::V8InspectorFrontendHost::platformCallback):
              * dom/SelectElement.cpp:
              * loader/CachedFont.cpp:
              * page/EventHandler.cpp:
              * page/chromium/EventHandlerChromium.cpp:
              * platform/Scrollbar.cpp:
              * platform/chromium/ChromiumBridge.h:
              * platform/graphics/chromium/FontPlatformData.h:
              * platform/graphics/skia/FontCustomPlatformData.cpp:
              (WebCore::FontCustomPlatformData::~FontCustomPlatformData):
              (WebCore::FontCustomPlatformData::fontPlatformData):
              (WebCore::createFontCustomPlatformData):
              * platform/graphics/skia/FontCustomPlatformData.h:
      2010-09-22  Ruben  <chromium@hybridsource.org>
      
              Reviewed by Tony Chang.
      
              [chromium] added ifdefs for FreeBSD support
              https://bugs.webkit.org/show_bug.cgi?id=46316
      
              * WebKit.gyp:
              * features.gypi:
              * src/ChromiumBridge.cpp:
              * src/WebFrameImpl.cpp:
              (WebKit::ChromePrintContext::spoolPage):
              * src/WebViewImpl.cpp:
              (WebKit::WebViewImpl::mouseDown):
              (WebKit::WebViewImpl::mouseUp):
              (WebKit::WebViewImpl::keyEvent):
              (WebKit::WebViewImpl::setScrollbarColors):
              (WebKit::WebViewImpl::setSelectionColors):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68112 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      cab56bd8
    • commit-queue@webkit.org's avatar
      2010-09-22 Mario Sanchez Prada <msanchez@igalia.com> · 11d9d820
      commit-queue@webkit.org authored
              Reviewed by Martin Robinson.
      
              [GTK] Replace instances of g_timeout_add() with g_idle_add() in testatk.c
              https://bugs.webkit.org/show_bug.cgi?id=46284
      
              Use g_idle_add() whenever g_timeout_add() was being used.
      
              This is desired because it's better to rely on the main loop to
              decide when it's a good moment to execute the task (when idle)
              than manually setting a timeout, which also sets a minimum amount
              of time needed to get the function executed.
      
              * tests/testatk.c:
              (testWekitAtkTextSelections):
              (testWebkitAtkGetExtents):
              (testWebkitAtkListsOfItems):
              (main):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68111 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      11d9d820
    • commit-queue@webkit.org's avatar
      2010-09-22 Rafael Antognolli <antognolli@profusion.mobi> · 030b422c
      commit-queue@webkit.org authored
              Unreviewed build fix.
      
              [EFL] Fix build due to latest EFL (Edje) changes.
              https://bugs.webkit.org/show_bug.cgi?id=46317
      
              Make the change int -> Edje_Load_Error (enum).
      
              No new functionality, so no new tests.
      
              * platform/efl/RenderThemeEfl.cpp:
              (WebCore::RenderThemeEfl::themePartCacheEntryReset):
              (WebCore::RenderThemeEfl::createEdje):
              (WebCore::RenderThemeEfl::applyPartDescriptions):
              * platform/efl/ScrollbarEfl.cpp:
              (ScrollbarEfl::setParent):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68110 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      030b422c
    • kinuko@chromium.org's avatar
      2010-09-22 Kinuko Yasuda <kinuko@chromium.org> · 09e6da2e
      kinuko@chromium.org authored
              Unreviewed, temporarily disabling the new filesystem test on chromium.
      
              * platform/chromium/test_expectations.txt:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68109 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      09e6da2e
    • cfleizach@apple.com's avatar
      WebCore: AX: aria-hidden change doesn't update the ax tree · da47efdd
      cfleizach@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=45836
      
      Reviewed by Darin Adler.
      
      Test: accessibility/aria-hidden-update.html
      
      * dom/Element.cpp:
      (WebCore::Element::updateAfterAttributeChanged):
      
      LayoutTests: [FileSystem] Fix Flags handling in DirectoryEntry custom binding code
      https://bugs.webkit.org/show_bug.cgi?id=46028
      
      Patch by Kinuko Yasuda <kinuko@chromium.org> on 2010-09-22
      Reviewed by Dumitru Daniliuc.
      
      * fast/filesystem/flags-passing-expected.txt: Added.
      * fast/filesystem/flags-passing.html: Added.
      * fast/filesystem/script-tests/TEMPLATE.html: Added.
      * fast/filesystem/script-tests/flags-passing.js: Added.
      * platform/gtk/Skipped: FileSystem API not supported on gtk.
      * platform/mac/Skipped: FileSystem API not supported on mac.
      * platform/qt/Skipped: FileSystem API not supported on qt.
      * platform/win/Skipped: FileSystem API not supported on win.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68108 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      da47efdd
    • bfulgham@webkit.org's avatar
      [WinCairo] Part 2: Update WebKitTestRunner and DumpRenderTree Build. · aa972b10
      bfulgham@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=46303.
      
      Reviewed by Martin Robinson.
      
      WebKit/win: 
      
      * WebKit.vcproj/WebKit.sln: Update overall Debug_Cairo and
      Release_Cairo configurations to select appropriate build
      targets for WebKitTestRunner and MiniBrowser.
      
      WebKitTools: 
      
      * MiniBrowser/Configurations/MiniBrowserCFLite.vsprops: Added.
      * MiniBrowser/Configurations/MiniBrowserCommon.vsprops: Moved
      CoreFoundation-specific stuff to new MiniBrowserCoreFoundation
      property sheet.
      * MiniBrowser/Configurations/MiniBrowserCoreFoundation.vsprops: Added.
      * MiniBrowser/MiniBrowser.vcproj: Updated configuration to use
      appropriate property sheet for the build types.
      * WebKitTestRunner/win/WebKitTestRunner.vcproj: Updated the
      configuration to use appropriate property sheet for CoreFoundation
      and CFLite-style builds.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68107 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      aa972b10
    • demarchi@webkit.org's avatar
      2010-09-22 Lucas De Marchi <lucas.demarchi@profusion.mobi> · 620dc9cc
      demarchi@webkit.org authored
              Reviewed by Antonio Gomes.
      
              [EFL] Fix build with GCC 4.4.x
              https://bugs.webkit.org/show_bug.cgi?id=40826
      
              Change the -fstrict-aliasing flag to -fno-strict-aliasing as other
              ports like GTK and QT are using in their build systems. It was failing
              to execute when compiled with GCC 4.4.x in Release mode.
      
              Refactor the flags in order to be more readable.
      
              * cmake/WebKitHelpers.cmake:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68106 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      620dc9cc
    • kinuko@chromium.org's avatar
      2010-09-22 Kinuko Yasuda <kinuko@chromium.org> · f0673b22
      kinuko@chromium.org authored
              Reviewed by Dumitru Daniliuc.
      
              [FileSystem] Fix Flags handling in DirectoryEntry custom binding code
              https://bugs.webkit.org/show_bug.cgi?id=46028
      
              Do not use adoptRef to get RefPtr from a raw pointer that is already
              adopted in toNative/toJS.
      
              Test: fast/filesystem/flags-passing.html
      
              * bindings/js/JSDirectoryEntryCustom.cpp:
              (WebCore::JSDirectoryEntry::getFile):
              (WebCore::JSDirectoryEntry::getDirectory):
              * bindings/v8/custom/V8DirectoryEntryCustom.cpp:
              (WebCore::V8DirectoryEntry::getDirectoryCallback):
              (WebCore::V8DirectoryEntry::getFileCallback):
      2010-09-22  Kinuko Yasuda  <kinuko@chromium.org>
      
              Reviewed by Dumitru Daniliuc.
      
              [FileSystem] Fix Flags handling in DirectoryEntry custom binding code
              https://bugs.webkit.org/show_bug.cgi?id=46028
      
              * fast/filesystem/flags-passing-expected.txt: Added.
              * fast/filesystem/flags-passing.html: Added.
              * fast/filesystem/script-tests/TEMPLATE.html: Added.
              * fast/filesystem/script-tests/flags-passing.js: Added.
              * platform/gtk/Skipped: FileSystem API not supported on gtk.
              * platform/mac/Skipped: FileSystem API not supported on mac.
              * platform/qt/Skipped: FileSystem API not supported on qt.
              * platform/win/Skipped: FileSystem API not supported on win.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68105 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      f0673b22
    • tony@chromium.org's avatar
      2010-09-22 Tony Chang <tony@chromium.org> · 762853f4
      tony@chromium.org authored
              Reviewed by David Levin.
      
              [chromium] add baselines for xmlhttprequest tests for chromium
              https://bugs.webkit.org/show_bug.cgi?id=46281
      
              * platform/chromium/http/tests/xmlhttprequest/methods-async-expected.txt: Allow no content-length for non-get/post requests
              * platform/chromium/http/tests/xmlhttprequest/methods-expected.txt: Allow no content-length for non-get/post requests
              * platform/chromium/test_expectations.txt: Enable 3 tests on linux + mac
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68104 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      762853f4
    • commit-queue@webkit.org's avatar
      2010-09-22 Jia Pu <jpu@apple.com> · 151dc7df
      commit-queue@webkit.org authored
              Reviewed by Dan Bernstein.
      
              Need to stop autocorrection panel timer when selection changes.
              https://bugs.webkit.org/show_bug.cgi?id=46201
              <rdar://problem/8441042>
      
              * editing/Editor.cpp:
              (WebCore::Editor::appliedEditing): Stop correction timer after editing.
              (WebCore::Editor::markAllMisspellingsAndBadGrammarInRanges): When text checking is done for
                showing autocorrection panel, we suppress all corrections that are not at the end of
                text checking range. So that we don't replace other words while showing correction panel
                for the last word.
              (WebCore::Editor::stopCorrectionPanelTimer): Stop correction timer.
      
              * editing/Editor.h: Added stopCorrectionPanelTimer() method.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68103 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      151dc7df
    • abarth@webkit.org's avatar
      2010-09-22 Adam Barth <abarth@webkit.org> · 47a19597
      abarth@webkit.org authored
              Reviewed by Eric Seidel.
      
              Optimize commit-queue performance for green trees
              https://bugs.webkit.org/show_bug.cgi?id=46254
      
              This patch redesigns the controller logic for the commit-queue.  In the
              new design, the controller exercises much finer-grained control over
              the landing process.  In particular:
      
              - Patches that fail to apply now get rejected almost immediately.
              - Patches that fail to build get rejects after two builds (instead of
                three builds and one test run).
              - Patches that run into a flaky test now get accepted after one build
                and two test runs instead of three full build-and-test runs.
      
              The main cost of these optimizations is that we don't find out the tree
              has a failing test until the very end of the process, but if the tree
              has a busted test, there's not much we can do anyway.  We might as well
              burn commit-queue resources spinning optimisticly.
      
              * Scripts/webkitpy/tool/bot/commitqueuetask.py: Added.
              * Scripts/webkitpy/tool/bot/commitqueuetask_unittest.py: Added.
              * Scripts/webkitpy/tool/commands/queues.py:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68102 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      47a19597
    • commit-queue@webkit.org's avatar
      2010-09-22 Eric Uhrhane <ericu@chromium.org> · b12498d5
      commit-queue@webkit.org authored
              Reviewed by Dumitru Daniliuc.
      
              Add Chromium implementation for FileWriter
              https://bugs.webkit.org/show_bug.cgi?id=44361
      
              No new tests; the implementation's not yet far enough along for that.
      
              This makes a chain of contruction-and-linking-up for various sorts of
              FileWriter interfaces to call through to Chromium, and for various
              sorts of FileWriterClients to call back, with object ownership always
              pointing towards the Chromium API.  That is, the DOM object owns an
              AsyncFileWriter, which owns its implementation object, and the reverse
              (client) direction is raw pointers.
      
              File metadata is looked up before the FileWriter is returned, so that
              we know the file exists and can cache its length.
      
              The WebCore part of this is simple and generic enough to handle
              Chromium, so it should be easy for me to put in a WebCore-only
              implementation next.
      
              * fileapi/AsyncFileWriter.h:
              * fileapi/DOMFileSystem.cpp:
              (WebCore::DOMFileSystem::createWriter):
              * fileapi/DOMFileSystem.h:
              * fileapi/FileSystemCallbacks.cpp:
              (WebCore::FileSystemCallbacksBase::didCreateFileWriter):
              (WebCore::FileWriterCallbacks::FileWriterCallbacks):
              (WebCore::FileWriterCallbacks::didCreateFileWriter):
              * fileapi/FileSystemCallbacks.h:
              * platform/AsyncFileSystem.h:
              * platform/AsyncFileSystemCallbacks.h:
      2010-09-22  Eric Uhrhane  <ericu@chromium.org>
      
              Reviewed by Dumitru Daniliuc.
      
              Add Chromium implementation for FileWriter
              https://bugs.webkit.org/show_bug.cgi?id=44361
      
              No new tests; the implementation's not yet far enough along for that.
      
              This makes a chain of construction-and-linking-up for various sorts of
              FileWriter interfaces to call through to Chromium, and for various
              sorts of FileWriterClients to call back, with object ownership always
              pointing towards the Chromium API.  That is, the DOM object owns an
              AsyncFileWriter, which owns its implementation object, and the reverse
              (client) direction is raw pointers.
      
              File metadata is looked up before the FileWriter is returned, so that
              we know the file exists and can cache its length.
      
              The WebKit/chromium interface here is modeled after that of the
              FileSystem.
      
              * public/WebFileInfo.h:
              (WebKit::WebFileInfo::WebFileInfo):
              * src/AsyncFileSystemChromium.cpp:
              (WebCore::FileWriterHelperCallbacks::FileWriterHelperCallbacks):
              (WebCore::FileWriterHelperCallbacks::didSucceed):
              (WebCore::FileWriterHelperCallbacks::didReadMetadata):
              (WebCore::FileWriterHelperCallbacks::didReadDirectory):
              (WebCore::FileWriterHelperCallbacks::didOpenFileSystem):
              (WebCore::FileWriterHelperCallbacks::didCreateFileWriter):
              (WebCore::FileWriterHelperCallbacks::didFail):
              (WebCore::AsyncFileSystemChromium::createWriter):
              * src/AsyncFileSystemChromium.h:
              * src/AsyncFileWriterChromium.cpp:
              (WebCore::AsyncFileWriterChromium::setWebFileWriter):
              (WebCore::AsyncFileWriterChromium::write):
              (WebCore::AsyncFileWriterChromium::didFail):
              * src/AsyncFileWriterChromium.h:
              * src/WebFileSystemCallbacksImpl.h:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68101 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b12498d5
    • msaboff@apple.com's avatar
      2010-09-22 Michael Saboff <msaboff@apple.com> · a5b5b181
      msaboff@apple.com authored
              Reviewed by Gavin Barraclough.
      
              Fixed the cross over from alternatives executed once and
              those that loop.  This fixed the problem where the index
              was getting messed up for looping alternatives causing an
              infinite loop.
              https://bugs.webkit.org/show_bug.cgi?id=46189
      
              * yarr/RegexJIT.cpp:
              (JSC::Yarr::RegexGenerator::generateDisjunction):
      2010-09-22  Michael Saboff  <msaboff@apple.com>
      
              Reviewed by Gavin Barraclough.
      
              Updated tests to include patterns similar to what caused the problems
              in https://bugs.webkit.org/show_bug.cgi?id=46189.
      
              * fast/js/regexp-bol-expected.txt:
              * fast/js/script-tests/regexp-bol.js:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68100 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      a5b5b181
    • abecsi@webkit.org's avatar
      2010-09-22 Andras Becsi <abecsi@webkit.org> · 09c050a2
      abecsi@webkit.org authored
              Speculative build fix for the Qt Windows bots.
      
              * Api/DerivedSources.pro:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68099 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      09c050a2
    • bfulgham@webkit.org's avatar
      WebKit/win: [WinCairo] Update WebKitTestRunner and DumpRenderTree Build.rt · 39d134eb
      bfulgham@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=46303.
      
      Reviewed by Martin Robinson.
      
      * WebKit.vcproj/WebKit.sln: Update overall Debug_Cairo and
      Release_Cairo configurations to select appropriate build
      targets for InjectionBundle.
      
      WebKitTools: [WinCairo] Update WebKitTestRunner and DumpRenderTree Build.
      https://bugs.webkit.org/show_bug.cgi?id=46303.
      
      Reviewed by Martin Robinson.
      
      * DumpRenderTree/win/DumpRenderTree.vcproj: Disable local MD5 sources
      for Cairo build. Copy wtf MD5 header to ForwardingHeaders for the
      WinCairo build.
      * WebKitTestRunner/Configurations/InjectedBundleCFLite.vsprops: Added.
      * WebKitTestRunner/win/InjectedBundle.vcproj: Update win new *_Cairo
      build targets that use the new InjectedBundleCFLite.vsprops file.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68098 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      39d134eb
    • abecsi@webkit.org's avatar
      2010-09-22 Andras Becsi <abecsi@webkit.org> · 314324df
      abecsi@webkit.org authored
              Unreviewed build fix after r68079.
      
              [Qt] Also generate WebPage's message-receiving code and message types
              for the Qt port of WebKit2.
      
              * DerivedSources.pro:
              * WebKit2.pro:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68097 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      314324df
    • ap@apple.com's avatar
      Reviewed by Adam Barth. · b4aa58a4
      ap@apple.com authored
              https://bugs.webkit.org/show_bug.cgi?id=46222
              <rdar://problem/8461701> Document.getElementById() malfunctions if ID was changed via Attr node modification.
      
              Test: fast/dom/Attr/change-id-via-attr-node-value.html
      
              * dom/Attr.h:
              * dom/Attr.cpp:
              (WebCore::Attr::setValue): Separated a version callable from WebCore from one avaiable to JS.
              Attr::setValue() can be called from Element::setAttribute(), and we don't want to update
              m_elementsById (or to call attributeChanged()) twice in that case.
              (WebCore::Attr::childrenChanged): If Attr's node children change, id changes.
      
              * dom/Document.cpp: (WebCore::Document::removeElementById): Added an assertion that we are
              not trying to remove something that isn't there. If we are, we probably failed to update
              m_elementsById earlier.
      
              * dom/Element.cpp: (WebCore::Element::setAttribute): If the attribute has an Attr node, its
              children should be updated to match attribute value.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68096 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b4aa58a4
    • sfalken@apple.com's avatar
      Allow jsc.exe to be run against unversioned ICU. · ae3cac30
      sfalken@apple.com authored
      Rubber stamped by Jon Honeycutt.
      
      * JavaScriptCore.vcproj/jsc/jscCommon.vsprops:
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68092 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      ae3cac30
    • tony@chromium.org's avatar
      2010-09-22 Tony Chang <tony@chromium.org> · 21253c2d
      tony@chromium.org authored
              Reviewed by Dimitri Glazkov.
      
              [chromium] Rebaseline 3 tests on win/linux
              https://bugs.webkit.org/show_bug.cgi?id=46315
      
              * platform/chromium-linux/editing/selection/move-by-line-001-expected.checksum: Minor differences in how text and borders are anti-aliased.
              * platform/chromium-linux/editing/selection/move-by-line-001-expected.png:
              * platform/chromium-win/fast/dom/Element/getBoundingClientRect-expected.txt: This test is sensitive to font metrics.  They appear correct visually, just positioned slightly different vertically.
              * platform/chromium-win/fast/dom/Range/getBoundingClientRect-expected.txt: Same as above.
              * platform/chromium/test_expectations.txt:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68091 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      21253c2d
    • tony@chromium.org's avatar
      2010-09-22 Tony Chang <tony@chromium.org> · c9877a0a
      tony@chromium.org authored
              Reviewed by David Levin.
      
              [chromium] remove commented out test failures from test_expectations.txt
              https://bugs.webkit.org/show_bug.cgi?id=46313
      
              * platform/chromium/test_expectations.txt:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68090 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      c9877a0a
    • hyatt@apple.com's avatar
      https://bugs.webkit.org/show_bug.cgi?id=46311, lock form controls and replaced... · 881c2c84
      hyatt@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=46311, lock form controls and replaced elements to vertical block flow.
      
      Reviewed by Dan Bernstein.
      
      Patch form controls that don't use RenderReplaced by hand in html.css.  Patch all RenderReplaced elements
      with an override of setStyle that forces the new style to have a vertical block flow.  We have to do something like
      this to cover the CSS3 content: url(img) method of image replacement.
              
      * css/html.css:
      (input, textarea, keygen, select, button, isindex, meter, progress):
      * rendering/RenderReplaced.cpp:
      (WebCore::RenderReplaced::setStyle):
      * rendering/RenderReplaced.h:
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68089 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      881c2c84
    • andersca@apple.com's avatar
      Stub out a PluginProcessManager class · 901c0e83
      andersca@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=46305
      
      Reviewed by Adam Roben.
      
      * UIProcess/Plugins/PluginProcessManager.cpp: Added.
      (WebKit::PluginProcessManager::shared):
      (WebKit::PluginProcessManager::getPluginProcessConnection):
      (WebKit::PluginProcessManager::removePluginProcessProxy):
      * UIProcess/Plugins/PluginProcessManager.h: Added.
      * WebKit2.xcodeproj/project.pbxproj:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68088 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      901c0e83
    • commit-queue@webkit.org's avatar
      2010-09-22 Kwang Yul Seo <skyul@company100.net> · 271e5035
      commit-queue@webkit.org authored
              Reviewed by Laszlo Gombos.
      
              Use "typedef wchar_t JSChar" when compiled with RVCT
              https://bugs.webkit.org/show_bug.cgi?id=40651
      
              Use wchar_t for JSChar and UChar when compiled with RVCT.
              Linux is the exception for this rule.
      
              * API/JSStringRef.h:
              * wtf/unicode/qt4/UnicodeQt4.h:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68087 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      271e5035
    • atwilson@chromium.org's avatar
      Unreviewed chromium test expectation changes. · 0c7fddfc
      atwilson@chromium.org authored
      * platform/chromium-linux/fast/multicol/span/span-as-immediate-child-property-removal-expected.checksum:
      * platform/chromium-linux/fast/multicol/span/span-as-immediate-child-property-removal-expected.png:
      * platform/chromium-linux/fast/multicol/span/span-as-immediate-child-property-removal-expected.txt:
      * platform/chromium-linux/fast/multicol/span/span-as-immediate-columns-child-removal-expected.checksum:
      * platform/chromium-linux/fast/multicol/span/span-as-immediate-columns-child-removal-expected.png:
      * platform/chromium-linux/fast/multicol/span/span-as-immediate-columns-child-removal-expected.txt:
      * platform/chromium-mac/fast/multicol/span/span-as-immediate-child-property-removal-expected.checksum:
      * platform/chromium-mac/fast/multicol/span/span-as-immediate-child-property-removal-expected.png:
      * platform/chromium-mac/fast/multicol/span/span-as-immediate-child-property-removal-expected.txt: Removed.
      * platform/chromium-mac/fast/multicol/span/span-as-immediate-columns-child-removal-expected.checksum:
      * platform/chromium-mac/fast/multicol/span/span-as-immediate-columns-child-removal-expected.png: Added.
      * platform/chromium-mac/fast/multicol/span/span-as-immediate-columns-child-removal-expected.txt: Removed.
      * platform/chromium-win/fast/multicol/span/span-as-immediate-child-property-removal-expected.checksum:
      * platform/chromium-win/fast/multicol/span/span-as-immediate-child-property-removal-expected.png:
      * platform/chromium-win/fast/multicol/span/span-as-immediate-child-property-removal-expected.txt:
      * platform/chromium-win/fast/multicol/span/span-as-immediate-columns-child-removal-expected.checksum:
      * platform/chromium-win/fast/multicol/span/span-as-immediate-columns-child-removal-expected.png:
      * platform/chromium-win/fast/multicol/span/span-as-immediate-columns-child-removal-expected.txt:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68086 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      0c7fddfc
    • dpranke@chromium.org's avatar
      2010-09-22 Dirk Pranke <dpranke@chromium.org> · 8e26bb6b
      dpranke@chromium.org authored
              Unreviewed, build fix.
      
              Fix bustage of rebaseline-chromium-webkit-tests resulting from
              r67974. Really need better unit tests for this tool :(
      
              * Scripts/webkitpy/layout_tests/rebaseline_chromium_webkit_tests.py:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68085 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      8e26bb6b
    • aroben@apple.com's avatar
      Unbreak test-webkitpy · a26796a0
      aroben@apple.com authored
      * Scripts/webkitpy/test/main.py:
      (Tester.run_tests): Add a line that mysteriously got deleted.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68084 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      a26796a0
    • hyatt@apple.com's avatar
      https://bugs.webkit.org/show_bug.cgi?id=46304 · 399ec5cd
      hyatt@apple.com authored
              
      Reviewed by Dan Bernstein.
      
      display:inline should become display:inline-block when an object's block flow does not match its parent's block flow.
      
      Added fast/blockflow/display-mutation.html
      
      WebCore: 
      
      * css/CSSStyleSelector.cpp:
      (WebCore::CSSStyleSelector::styleForElement):
      (WebCore::CSSStyleSelector::pseudoStyleForElement):
      (WebCore::CSSStyleSelector::adjustRenderStyle):
      * css/CSSStyleSelector.h:
      
      LayoutTests: 
      
      * fast/blockflow/display-mutation-expected.txt: Added.
      * fast/blockflow/display-mutation.html: Added.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68083 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      399ec5cd
    • kenneth@webkit.org's avatar
      Add a ViewportConfiguration class for the Qt WebKit2 API, which · b42082e1
      kenneth@webkit.org authored
      currently calculated a fallback viewport configuration.
      
      Patch by Kenneth Rohde Christiansen <kenneth@webkit.org> on 2010-09-22
      Reviewed by Antonio Gomes.
      
      * UIProcess/API/qt/qwkpage.cpp:
      (QWKPage::ViewportConfiguration::ViewportConfiguration):
      (QWKPage::ViewportConfiguration::~ViewportConfiguration):
      (QWKPage::ViewportConfiguration::operator=):
      (QWKPage::viewportConfigurationForSize):
      * UIProcess/API/qt/qwkpage.h:
      * UIProcess/API/qt/qwkpage_p.h:
      (QtViewportConfigurationPrivate::QtViewportConfigurationPrivate):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68082 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b42082e1
    • aroben@apple.com's avatar
      Windows build fix · f89bd1e2
      aroben@apple.com authored
      * WebProcess/WebKitMain.cpp:
      (WebKitMain): Fix typo
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68081 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      f89bd1e2
    • aroben@apple.com's avatar
      Make test-webkitpy test WebKit2's scripts · ba3c7eea
      aroben@apple.com authored
      These scripts can't be in WebKitTools due to limitations of Apple's
      build process. But that doesn't mean we can't test them!
      
      Fixes <http://webkit.org/b/46297> test-webkitpy should test code in
      WebKit2/Scripts
      
      Reviewed by Adam Barth.
      
      * Scripts/test-webkitpy:
      (_clean_packages_with_test): Renamed from _clean_webkitpy_with_test.
      Now takes an external_package_paths parameter and cleans both webkitpy
      and any external packages.
      (init): Added an external_package_paths parameter which we pass along
      to _clean_packages_with_test.
      (top level): Add WebKit2/Scripts/webkit2 as our only external package
      and pass it along to init and Tester.run_tests.
      
      * Scripts/webkitpy/test/main.py:
      (Tester.run_tests): Added an optional external_package_paths
      parameter. We modify sys.path so that the external packages can be
      imported, and search for unittest files inside all external packages
      in addition to inside webkitpy.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68080 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      ba3c7eea
    • aroben@apple.com's avatar
      Autogenerate WebPage's message-receiving code and message types · 24124b84
      aroben@apple.com authored
      This patch encompasses several changes that allow message types and
      receiving code to be generated by a script, and add some type-safety
      as a bonus. Messages are now represented by structs instead of an ID +
      ArgumentCoder. The struct contains the arguments and the ID together,
      and has a constructor that enforces the use of correct types.
      Correspondingly, a new overload of WebProcessProxy::send that takes a
      message struct (instead of a message ID and separate arguments) has
      been added. Eventually all callers should use this overload and the
      old one can be removed.
      
      This patch only touches WebPage's messages. We should transition other
      message receivers over to this new system eventually.
      
      Fixes <http://webkit.org/b/43636> <rdar://problem/8282462> Add a
      type-safe IPC mechanism to WebKit2
      
      Reviewed by Anders Carlsson.
      
      * DerivedSources.make: Added. Calls generate-message-receiver.py and
      generate-messages-header.py for each message receiver it knows about
      (just WebPage for now).
      
      * Platform/CoreIPC/Arguments.h: Added First/Second/ThirdArgumentType
      typedefs for use in handleMessage.
      
      * Platform/CoreIPC/HandleMessage.h: Added.
      (CoreIPC::handleMessage): This overloaded function template decodes
      arguments and passes them along to the specified function.
      
      * Scripts/generate-message-receiver.py: Added.
      * Scripts/generate-messages-header.py: Added.
      These scripts just wrap functionality in messages.py.
      
      * Scripts/webkit2/__init__.py: Added. This just exists so that Python
      will treat this directory as a package.
      
      * Scripts/webkit2/messages.py: Added. Contains the code to parse
      messages files and generate .cpp/.h files from them.
      (MessageReceiver.__init__): This class represents a single receiver of
      messages.
      (MessageReceiver.iterparameters): Returns a generator that can be used
      to iterate over all the parameters of all the messages of this
      receiver.
      (MessageReceiver.parse): Reads a messages file from a file-like object
      and parses it into a MessageReceiver object.
      (Message.__init__): This class represents a single message.
      (Message.id): Returns the ID name for this message.
      (Parameter.__init__): This class represents a single parameter for a
      message.
      (messages_header_filename): Returns the name of the header that
      defines the messages for a given receiver.
      (surround_in_condition): Surrounds the given string in #if/#endif if
      there is an associated condition.
      (messages_to_kind_enum): Returns a string that defines the Kind enum
      for these messages.
      (function_parameter_type): Returns the type that should be used when
      passing a value of the given type as a parameter to a function.
      (base_class): Returns the base class for a message struct.
      (message_to_struct_declaration): Returns a string that declares the
      struct for this message.
      (forward_declarations_for_namespace): Returns a string that contains
      forward-declarations for a set of types in a given namespace.
      (forward_declarations): Returns a string that contains all the
      forward-declarations needed in order to declare all the message
      structs for this receiver.
      (generate_header_file): Returns a string containing the messages
      header file for this receiver.
      (handler_function): Returns the name of the function that handles a
      given message for a given receiver.
      (case_statement): Returns a string containing a case statement for
      handling the given message.
      (header_for_type): Returns the header needed to define a given type,
      enclosed in quotes or angle brackets as needed.
      (generate_message_handler): Returns a string containing the contents
      of a .cpp file that defines a didReceive*Message function.
      
      * Scripts/webkit2/messages_unittest.py: Added. Contains tests for
      messages.py.
      
      * Shared/CoreIPCSupport/WebPageMessageKinds.h: Removed. This has been
      replaced by a generated WebPageMessages.h header.
      
      * UIProcess/WebEditCommandProxy.cpp:
      * UIProcess/WebPageProxy.cpp:
      Updated to use the new message structs and WebProcessProxy::send
      overload.
      
      * UIProcess/WebProcessProxy.h:
      (WebKit::WebProcessProxy::send): Added this new overload that takes a
      message struct.
      
      * WebKit2.xcodeproj/project.pbxproj: Added a Derived Sources shell
      script target that invokes DerivedSources.make. Added "Derived
      Sources" and "Scripts" groups that contain the various new files.
      
      * WebProcess/WebPage/WebPage.cpp:
      (WebKit::WebPage::didReceivePolicyDecision):
      (WebKit::WebPage::getSourceForFrame):
      Changed these functions to take the raw IPC types and do the necessary
      translation themselves. This keeps the generated code from need to
      know how to perform the translation.
      
      (WebKit::WebPage::didReceiveMessage): Replaced handling of WebPage
      messages with a call to didReceiveWebPageMessage, whose implementation
      is generated by the scripts.
      
      * WebProcess/WebPage/WebPage.h: Added didReceiveWebPageMessage.
      
      * WebProcess/WebPage/WebPage.messages.in: Added. This file declares
      all of the messages that WebPage receives, roughly grouped by
      functionality.
      
      * win/WebKit2.vcproj: Added "Derived Sources" and "Scripts" filters
      that contain the various new files. Let VS resort some other files.
      
      * win/WebKit2Common.vsprops: Added
      $(WebKitOutputDir)\obj\$(ProjectName)\DerivedSources to the include
      path so that the generated messages header can be found.
      
      * win/WebKit2.make: Copy the generated source files to $(DSTROOT).
      
      * win/WebKit2Generated.make: Added a call to build-generated-files.sh.
      
      * win/WebKit2Generated.vcproj: Added build-generated-files.sh.
      
      * win/build-generated-files.sh: Added. Invokes DerivedSources.make.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68079 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      24124b84
    • commit-queue@webkit.org's avatar
      2010-09-22 Mario Sanchez Prada <msanchez@igalia.com> · 2aa8d03c
      commit-queue@webkit.org authored
              Reviewed by Martin Robinson.
      
              [Gtk] object:text-changed events should be emitted for entries and password text
              https://bugs.webkit.org/show_bug.cgi?id=25898
      
              Implement proper 'text-changed' signal emission for the GTK port
      
              Call deleteTextFromNode() when needed while removing text nodes.
              Do it even when removeNode() is going to be called afterwards, in
              order to allow accessibility to get properly notified about the
              text being removed alongside with that node.
      
              * editing/DeleteSelectionCommand.cpp:
              (WebCore::DeleteSelectionCommand::handleGeneralDelete):
      
              New function in AXObjectCache to call when text changes in a node.
              Added one new function to allow notifying something changed in a
              text node through the associated RenderObject, making such a
              function dependant on the platform-specific implementation,
              provided through a protected function (provided a proper
              implementation for the GTK port and a dummy one for the others).
      
              * accessibility/AXObjectCache.cpp:
              (WebCore::AXObjectCache::nodeTextChangeNotification): New
              * accessibility/AXObjectCache.h:
              (WebCore::AXObjectCache::AXTextChange): New enumeration
              (WebCore::AXObjectCache::nodeTextChangeNotification): New
              (WebCore::AXObjectCache::nodeTextChangePlatformNotification): New
              * accessibility/AccessibilityRenderObject.h:
              (WebCore::toAccessibilityRenderObject):
              * accessibility/chromium/AXObjectCacheChromium.cpp:
              (WebCore::AXObjectCache::nodeTextChangePlatformNotification): New
              * accessibility/gtk/AXObjectCacheAtk.cpp:
              (WebCore::emitTextChanged):
              (WebCore::AXObjectCache::nodeTextChangePlatformNotification): New
              * accessibility/mac/AXObjectCacheMac.mm:
              (WebCore::AXObjectCache::nodeTextChangePlatformNotification): New
              * accessibility/win/AXObjectCacheWin.cpp:
              (WebCore::AXObjectCache::nodeTextChangePlatformNotification): New
      
              Notify accessibility when something changes in a text node.
              Call to AXObjectCache::nodeTextChangeNotification() to notify when
              text was inserted/deleted when applying/unapplying a text edition
              command, along with the offset in the original text where the
              change took place and the number of characters that got affected.
      
              * editing/AppendNodeCommand.cpp:
              (WebCore::sendAXTextChangedIgnoringLineBreaks):
              (WebCore::AppendNodeCommand::doApply):
              (WebCore::AppendNodeCommand::doUnapply):
              * editing/DeleteFromTextNodeCommand.cpp:
              (WebCore::DeleteFromTextNodeCommand::doApply):
              (WebCore::DeleteFromTextNodeCommand::doUnapply):
              * editing/InsertIntoTextNodeCommand.cpp:
              (WebCore::InsertIntoTextNodeCommand::doApply):
              (WebCore::InsertIntoTextNodeCommand::doUnapply):
              * editing/InsertNodeBeforeCommand.cpp:
              (WebCore::InsertNodeBeforeCommand::doApply):
              (WebCore::InsertNodeBeforeCommand::doUnapply):
              (WebCore::DeleteSelectionCommand::handleGeneralDelete):
      2010-09-22  Mario Sanchez Prada  <msanchez@igalia.com>
      
              Reviewed by Martin Robinson.
      
              [Gtk] object:text-changed events should be emitted for entries and password text
              https://bugs.webkit.org/show_bug.cgi?id=25898
      
              New unit test to make sure text-changed signals are emitted
      
              * tests/testatk.c:
              (textChangedCb): New. Signal handler for the
              text-changed::insert and text-changed::delete signals.
              (checkTextChangesAndBailOut): New. Source function to check
              the global result of the test and quit from the main loop.
              (testWebkitAtkTextChangedNotifications): New test.
              (main):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68078 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      2aa8d03c
    • hyatt@apple.com's avatar
      https://bugs.webkit.org/show_bug.cgi?id=46298, add logicalWidth/Height... · fbd0a649
      hyatt@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=46298, add logicalWidth/Height getters/setters to RenderBox.
      
      Reviewed by Dan Bernstein.
      
      * rendering/RenderBox.cpp:
      (WebCore::RenderBox::logicalWidth):
      (WebCore::RenderBox::logicalHeight):
      (WebCore::RenderBox::setLogicalWidth):
      (WebCore::RenderBox::setLogicalHeight):
      * rendering/RenderBox.h:
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68077 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      fbd0a649
    • oliver@apple.com's avatar
      2010-09-22 Oliver Hunt <oliver@apple.com> · b1ee29d6
      oliver@apple.com authored
              Reviewed by Gavin Barraclough.
      
              [INTERPRETER] Two tests fail with SputnikError: #1.1: if argArray is neither an array nor an arguments object (see 10.1.8), a TypeError exception is thrown
              https://bugs.webkit.org/show_bug.cgi?id=44245
      
              Remove incorrect code from op_load_varargs in the interpreter.
      
              * interpreter/Interpreter.cpp:
              (JSC::Interpreter::privateExecute):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68076 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b1ee29d6
    • andersca@apple.com's avatar
      Enhance ProcessLauncher to be able to launch plug-in processes · 989d9364
      andersca@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=46295
      
      Reviewed by Darin Adler.
      
      * UIProcess/Launcher/ProcessLauncher.cpp:
      (WebKit::ProcessLauncher::ProcessLauncher):
      Make the ProcessLauncher constructor take a LaunchOptions struct.
      
      (WebKit::ProcessLauncher::processTypeAsString):
      Given a process type, return its string representation.
      
      (WebKit::ProcessLauncher::getProcessTypeFromString):
      And vice versa.
      
      * UIProcess/Launcher/mac/ProcessLauncherMac.mm:
      (WebKit::ProcessLauncher::launchProcess):
      Rename "mode" to "type" and pass the type as a string based on the LaunchOptions struct.
      Also, set the launch architecture from the LaunchOptions struct.
      
      * UIProcess/Launcher/win/ProcessLauncherWin.cpp:
      (WebKit::ProcessLauncher::launchProcess):
      Rename "mode" to "type."
      
      * UIProcess/WebProcessProxy.cpp:
      (WebKit::WebProcessProxy::connect):
      Initialize a LaunchOptions struct.
      
      * WebProcess/WebKitMain.cpp:
      (WebKitMain):
      Factor code that can be shared between the mac and windows WebKitMain code out into a
      WebKitMain overload that takes a CommandLine.
      
      * WebProcess/WebProcessMain.h:
      * WebProcess/mac/WebProcessMainMac.mm:
      (WebKit::WebProcessMain):
      * WebProcess/win/WebProcessMainWin.cpp:
      (WebKit::WebProcessMain):
      Change WebProcessMain to take a const CommandLine reference instead of a pointer.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68075 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      989d9364
    • commit-queue@webkit.org's avatar
      2010-09-22 Luiz Agostini <luiz.agostini@openbossa.org> · b7c66204
      commit-queue@webkit.org authored
              Reviewed by Oliver Hunt.
      
              Script engine agnostic ScriptCallback class
              https://bugs.webkit.org/show_bug.cgi?id=43216
      
              Refactoring class ScriptFunctionCall to create a class (ScriptCallback) that
              receives a script function as a ScriptValue and calls that function with provided parameters.
      
              * bindings/js/ScriptFunctionCall.cpp:
              (WebCore::ScriptCallArgumentHandler::appendArgument):
              (WebCore::ScriptFunctionCall::ScriptFunctionCall):
              (WebCore::ScriptCallback::ScriptCallback):
              (WebCore::ScriptCallback::call):
              * bindings/js/ScriptFunctionCall.h:
              (WebCore::ScriptCallArgumentHandler::ScriptCallArgumentHandler):
              * bindings/v8/ScriptFunctionCall.cpp:
              (WebCore::ScriptCallArgumentHandler::appendArgument):
              (WebCore::ScriptFunctionCall::ScriptFunctionCall):
              (WebCore::ScriptCallback::ScriptCallback):
              (WebCore::ScriptCallback::call):
              * bindings/v8/ScriptFunctionCall.h:
              (WebCore::ScriptCallArgumentHandler::ScriptCallArgumentHandler):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68074 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b7c66204
    • dpranke@chromium.org's avatar
      2010-09-22 Dirk Pranke <dpranke@chromium.org> · 29db70dc
      dpranke@chromium.org authored
              Reviewed by Tony Chang.
      
              new-run-webkit-tests: r68063 broke linux python tests
      
              Add a missing "from __future__ import with_statement" to this
              new file; the perils of testing only on the Mac and by review :(
      
              https://bugs.webkit.org/show_bug.cgi?id=46293
      
              * Scripts/webkitpy/layout_tests/port/chromium_gpu.py:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68073 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      29db70dc
    • bfulgham@webkit.org's avatar
      Unreviewed. · ca8fbe28
      bfulgham@webkit.org authored
      Build fix for invalid assertion after r68059.
      
      * editing/ApplyStyleCommand.cpp:
      (WebCore::getPropertiesNotIn): Correct assertion target after
      name change.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68072 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      ca8fbe28
    • inferno@chromium.org's avatar
      2010-09-22 Abhishek Arya <inferno@chromium.org> · 01a25401
      inferno@chromium.org authored
              Unreviewed.
      
              Rebaseline multicol/span tests for GTK, QT platforms as part of r68062.
      
              * platform/gtk/fast/multicol/span/span-as-immediate-child-property-removal-expected.txt:
              * platform/gtk/fast/multicol/span/span-as-immediate-columns-child-removal-expected.txt:
              * platform/qt/fast/multicol/span/span-as-immediate-child-property-removal-expected.txt:
              * platform/qt/fast/multicol/span/span-as-immediate-columns-child-removal-expected.txt:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@68071 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      01a25401