1. 30 Jan, 2012 40 commits
    • andersca@apple.com's avatar
      Show repaint counters in individual tiles · f2826416
      andersca@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=77390
      <rdar://problem/10767967>
      
      Reviewed by Darin Adler.
      
      * platform/graphics/ca/mac/TileCache.h:
      * platform/graphics/ca/mac/TileCache.mm:
      (WebCore::TileCache::setNeedsDisplayInRect):
      Make sure to invalidate the repaint counter rect if necessary.
      
      (WebCore::TileCache::drawLayer):
      Draw the repaint counter.
      
      (WebCore::TileCache::showRepaintCounter):
      New function that determines whether we should show repaint counters for the given tile cache.
      
      * platform/graphics/ca/mac/WebTileLayer.h:
      * platform/graphics/ca/mac/WebTileLayer.mm:
      (-[WebTileLayer incrementRepaintCount]):
      Add method for getting the repaint count.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106311 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      f2826416
    • mitz@apple.com's avatar
      <rdar://problem/10778045> REGRESSION (r91935): text-combine fails · 3c1842d0
      mitz@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=77373
      
      Reviewed by Darin Adler.
      
      Source/WebCore: 
      
      Removed tests that were failing because of this bug from the Lion skipped
      list.
      
      * platform/graphics/mac/GlyphPageTreeNodeMac.cpp:
      (WebCore::GlyphPage::fill): Changed to use CTFontGetGlyphsForCharacters,
      rather than wkGetVerticalGlyphsForCharacters, for non-fullwidth fonts.
      
      LayoutTests: 
      
      Removed tests that were failing because of this bug from the skipped list.
      
      * platform/mac-lion/Skipped:
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106310 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      3c1842d0
    • beidson@apple.com's avatar
      Fix typo in test name from my last commit. · 1e14ce0a
      beidson@apple.com authored
      Spotted and rubber-stamped by Darin Adler.
      
      * plugins/crash-restoring-plugin-page-from-page-cache-expected.txt: Copied from plugins/crash-restoring-pluging-page-from-page-cache-expected.txt.
      * plugins/crash-restoring-plugin-page-from-page-cache.html: Copied from plugins/crash-restoring-pluging-page-from-page-cache.html.
      * plugins/crash-restoring-pluging-page-from-page-cache-expected.txt: Removed.
      * plugins/crash-restoring-pluging-page-from-page-cache.html: Removed.
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106309 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      1e14ce0a
    • rniwa@webkit.org's avatar
      Enable memcache on webkit-perf.appspot.com · 94749ac4
      rniwa@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77378
      
      Reviewed by Adam Barth.
      
      Eanble memcache on dashboard, manifest, and runs handlers. Clear appropriate caches when new runs are reported,
      or new models are created. We flush all caches when tests are merged since it's hard to figure out dependencies
      in that case. Luckily, we merge tests only occassionally and manually (or hope so) so this shouldn't be an issue.
      
      * Websites/webkit-perf.appspot.com/create_handler.py:
      (CreateHandler.post):
      * Websites/webkit-perf.appspot.com/dashboard_handler.py:
      (DashboardHandler.get):
      * Websites/webkit-perf.appspot.com/manifest_handler.py:
      (ManifestHandler.get):
      * Websites/webkit-perf.appspot.com/merge_tests_handler.py:
      (MergeTestsHandler.post):
      * Websites/webkit-perf.appspot.com/models.py:
      (createInTransactionWithNumericIdHolder):
      (Test):
      (Test.cacheKey):
      * Websites/webkit-perf.appspot.com/report_handler.py:
      (ReportHandler.post):
      (ReportHandler._addTestIfNeeded):
      * Websites/webkit-perf.appspot.com/runs_handler.py:
      (RunsHandler.get):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106308 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      94749ac4
    • commit-queue@webkit.org's avatar
      BlackBerry - Support Proxy-Authenticate headers when a proxy is configured · 05856f43
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77361
      
      Though we have a proxy configured, we might not have the auth
      credentials it requires. Support Proxy-Authenticate for that case.
      
      Patch by Christopher Hutten-Czapski <chutten@rim.com> on 2012-01-30
      Reviewed by George Staikos.
      
      * platform/network/blackberry/NetworkJob.cpp:
      (WebCore::NetworkJob::handleNotifyHeaderReceived):
      (WebCore::NetworkJob::handleAuthHeader):
      (WebCore::NetworkJob::sendRequestWithCredentials):
      * platform/network/blackberry/NetworkJob.h:
      * platform/network/blackberry/NetworkManager.cpp:
      (WebCore::NetworkManager::startJob):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106307 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      05856f43
    • jamesr@google.com's avatar
      [chromium] Remove unnecessary retry logic in LayerRendererChromium... · fc0bdfbc
      jamesr@google.com authored
      [chromium] Remove unnecessary retry logic in LayerRendererChromium initialization for accelerated painting
      https://bugs.webkit.org/show_bug.cgi?id=77247
      
      Reviewed by Kenneth Russell.
      
      The accelerate painting setting is done differently from other capability-dependent settings for no good reason,
      requiring that we retry initialization with different settings. For all other settings we set the capabilities
      bit to true if the setting is requested and if the required capabilities exist on the underlying context.
      
      * platform/graphics/chromium/LayerRendererChromium.cpp:
      (WebCore::LayerRendererChromium::create):
      (WebCore::LayerRendererChromium::initialize):
      * platform/graphics/chromium/cc/CCLayerTreeHostImpl.cpp:
      (WebCore::CCLayerTreeHostImpl::initializeLayerRenderer):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106306 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      fc0bdfbc
    • beidson@apple.com's avatar
      <rdar://problem/10707072> · 1145c5e7
      beidson@apple.com authored
      Crashes in WebProcess at WebCore::Node::rendererIsNeeded
      
      Reviewed by Darin Adler.
      
      Source/WebCore:
      
      In specific circumstances a plugin element can be without a render style at the point in time where
      the page navigated and enters the page cache.
      
      When this is the cash, the element should not enter into the "custom style for renderer" mode and should
      instead use the default render style machinery.
      
      Test: plugins/crash-restoring-pluging-page-from-page-cache.html
      
      * html/HTMLPlugInImageElement.cpp:
      (WebCore::HTMLPlugInImageElement::documentWillSuspendForPageCache): Only setHasCustomStyleForRenderer and
        forceRecalc if there actually is a custom style to be used.
      (WebCore::HTMLPlugInImageElement::documentDidResumeFromPageCache): Only clearHasCustomStyleForRenderer if there
        actually was a custom style to be cleared.
      (WebCore::HTMLPlugInImageElement::customStyleForRenderer): This should only be called if there actually is a
        custom style to be used. Otherwise the element would have to fallback to the "normal" RenderStyle which might
        not exist.
      
      LayoutTests:
      
      * plugins/crash-restoring-pluging-page-from-page-cache-expected.txt: Added.
      * plugins/crash-restoring-pluging-page-from-page-cache.html: Added.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106305 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      1145c5e7
    • andersca@apple.com's avatar
      Show debug borders for individual tile cache tiles · 7a4286d6
      andersca@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=77388
      
      Reviewed by Sam Weinig.
      
      * platform/graphics/GraphicsLayer.cpp:
      (WebCore::GraphicsLayer::updateDebugIndicators):
      Give tile cache tiles a thin dark blue border.
      
      * platform/graphics/ca/mac/TileCache.h:
      (WebCore::TileCache::tileDebugBorderWidth):
      (WebCore::TileCache::tileDebugBorderColor):
      Add getters and member variables for the tile debug border width and color.
      
      * platform/graphics/ca/mac/TileCache.mm:
      (WebCore::TileCache::TileCache):
      Initialize m_tileDebugBorderWidth.
      
      (WebCore::TileCache::setTileDebugBorderWidth):
      Update the border width of each tile layer.
      
      (WebCore::TileCache::setTileDebugBorderColor):
      Update the border color of each tile layer.
      
      (WebCore::TileCache::createTileLayer):
      Set the border color and border width.
      
      * platform/graphics/ca/mac/WebTileCacheLayer.h:
      * platform/graphics/ca/mac/WebTileCacheLayer.mm:
      (-[WebTileCacheLayer borderColor]):
      (-[WebTileCacheLayer setBorderColor:]):
      (-[WebTileCacheLayer borderWidth]):
      (-[WebTileCacheLayer setBorderWidth:]):
      Call through to the TileCache.
      
      * platform/graphics/mac/WebLayer.mm:
      (drawLayerContents):
      Don't draw the repaint counter for tile cache layers, each tile will maintain its own repaint counter.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106304 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      7a4286d6
    • ojan@chromium.org's avatar
      run-webkit-tests calls out to webkit-build-directory twice · b2a21d44
      ojan@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77248
      
      Reviewed by Dirk Pranke.
      
      This reduces to calls to webkit-build-directory to one call and
      saves 500ms for run-webkit-tests over a single file.
      
      * Scripts/webkit-build-directory:
      If called with no arguments, print out both the top-level directory and the
      configuration directory.
      
      * Scripts/webkit-build-directory:
      * Scripts/webkitpy/layout_tests/port/config.py:
      (Config.build_directory):
      When called with no arguments, get and cache both the top-level and configuration directories.
      
      * Scripts/webkitpy/layout_tests/port/config_standalone.py:
      (main):
      * Scripts/webkitpy/layout_tests/port/config_unittest.py:
      (ConfigTest.assert_configuration):
      (ConfigTest.test_build_directory.mock_webkit_build_directory):
      (ConfigTest.test_build_directory):
      (ConfigTest.test_default_configuration__notfound):
      * Scripts/webkitpy/layout_tests/port/webkit.py:
      (WebKitPort._build_path):
      Cache the root directory in the options object so that other users
      of the options object don't need to call out to webkit-build-directory.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106302 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b2a21d44
    • commit-queue@webkit.org's avatar
      single-file input elements should refuse multi-file drags · 76c0798e
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=25913
      
      Patch by Rakesh KN <rakesh.kn@motorola.com> on 2012-01-30
      Reviewed by Eric Seidel.
      
      Refuse the multiple file drags onto a single file input element.
      
      Source/WebCore:
      
      No new tests: Covered by existing tests, updated the expected results.
      
      * page/DragController.cpp:
      (WebCore::DragController::tryDocumentDrag):
      Setting the dragSession.numberOfItemsToBeAccepted to 0 so that drag operation is none if the
      file input element under mouse is single input type and number of files dragged onto that
      input element are more than 1.
      
      LayoutTests:
      
      * editing/pasteboard/file-input-files-access-expected.txt:
      Expected result changed.
      * editing/pasteboard/script-tests/file-input-files-access.js:
      (runTest):
      Removed the 'FIXME' debug and comment as it is not valid after this change.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106301 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      76c0798e
    • commit-queue@webkit.org's avatar
      [chromium] Use region reported painted opaque for draw culling · 429f925f
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=76015
      
      Patch by Dana Jansens <danakj@chromium.org> on 2012-01-30
      Reviewed by James Robinson.
      
      Source/WebCore:
      
      New unit tests in CCTiledLayerImplTest.cpp, CCQuadCullerTest.cpp, CCLayerTreeHostImplTest.cpp
      
      * platform/graphics/chromium/LayerChromium.cpp:
      (WebCore::LayerChromium::setOpaque):
      * platform/graphics/chromium/TiledLayerChromium.cpp:
      (UpdatableTile):
      (WebCore::TiledLayerChromium::pushPropertiesTo):
      (WebCore::TiledLayerChromium::prepareToUpdateTiles):
      * platform/graphics/chromium/cc/CCDrawQuad.cpp:
      (WebCore::CCDrawQuad::opaqueRect):
      (WebCore):
      * platform/graphics/chromium/cc/CCDrawQuad.h:
      (CCDrawQuad):
      (WebCore::CCDrawQuad::needsBlending):
      * platform/graphics/chromium/cc/CCQuadCuller.cpp:
      (WebCore::CCQuadCuller::cullOccludedQuads):
      * platform/graphics/chromium/cc/CCSolidColorDrawQuad.cpp:
      (WebCore::CCSolidColorDrawQuad::CCSolidColorDrawQuad):
      * platform/graphics/chromium/cc/CCTileDrawQuad.cpp:
      (WebCore::CCTileDrawQuad::create):
      (WebCore::CCTileDrawQuad::CCTileDrawQuad):
      * platform/graphics/chromium/cc/CCTileDrawQuad.h:
      (CCTileDrawQuad):
      * platform/graphics/chromium/cc/CCTiledLayerImpl.cpp:
      (DrawableTile):
      (WebCore::DrawableTile::opaqueRect):
      (WebCore::DrawableTile::setOpaqueRect):
      (WebCore::CCTiledLayerImpl::appendQuads):
      (WebCore::CCTiledLayerImpl::pushTileProperties):
      * platform/graphics/chromium/cc/CCTiledLayerImpl.h:
      (CCTiledLayerImpl):
      
      Source/WebKit/chromium:
      
      * tests/CCLayerTreeHostImplTest.cpp:
      (WebKit::BlendStateCheckLayer::appendQuads):
      (WebKit::BlendStateCheckLayer::setOpaqueColor):
      (BlendStateCheckLayer):
      (WebKit::BlendStateCheckLayer::BlendStateCheckLayer):
      (WebKit::TEST_F):
      * tests/CCQuadCullerTest.cpp:
      (WebCore::MakeTileQuad):
      (WebCore::setQuads):
      (WebCore::TEST):
      (WebCore):
      * tests/CCTiledLayerImplTest.cpp:
      (CCLayerTestCommon::createLayer):
      (CCLayerTestCommon::TEST):
      (CCLayerTestCommon::getQuads):
      (CCLayerTestCommon::coverageVisibleRectOnTileBoundaries):
      (CCLayerTestCommon::coverageVisibleRectIntersectsTiles):
      (CCLayerTestCommon::coverageVisibleRectIntersectsBounds):
      (CCLayerTestCommon):
      
      LayoutTests:
      
      * platform/chromium-linux/compositing/iframes/iframe-in-composited-layer-expected.png:
      * platform/chromium-mac-leopard/fast/repaint/block-selection-gap-in-composited-layer-expected.png: Removed.
      * platform/chromium-mac-snowleopard/fast/repaint/block-selection-gap-in-composited-layer-expected.png: Removed.
      * platform/chromium-win/fast/repaint/block-selection-gap-in-composited-layer-expected.png: Removed.
      * platform/chromium/fast/repaint/block-selection-gap-in-composited-layer-expected.png: Renamed from LayoutTests/platform/chromium-linux/fast/repaint/block-selection-gap-in-composited-layer-expected.png.
      * platform/chromium/test_expectations.txt:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106300 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      429f925f
    • commit-queue@webkit.org's avatar
      [Chromium] Unreviewed test expectation change. · 37bde608
      commit-queue@webkit.org authored
      Remove passing media/media-blocked-by-beforeload.html from test_expectations.txt
      after r103509
      
      Patch by Shadi Khalek <shadi@chromium.org> on 2012-01-30
      
      * platform/chromium/test_expectations.txt:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106299 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      37bde608
    • rniwa@webkit.org's avatar
      Crash in previousLinePosition when moving into a root inline box without leaves · f63044a7
      rniwa@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=76812
      
      Reviewed by Enrica Casucci.
      
      Source/WebCore: 
      
      The crash was caused by us assuming that every root inline box has at least one leaf,
      which isn't true when we create inline boxes for an empty text run with margin, border, etc...
      
      Test: editing/selection/move-into-empty-root-inline-box.html
      
      * editing/visible_units.cpp:
      (WebCore::previousLinePosition):
      (WebCore::nextLinePosition):
      
      LayoutTests: 
      
      Add a regression test for the crash. Unfortunately, we can only test previousLinePosition.
      
      * editing/selection/move-into-empty-root-inline-box-expected.txt: Added.
      * editing/selection/move-into-empty-root-inline-box.html: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106298 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      f63044a7
    • oliver@apple.com's avatar
      Unexpected syntax error · 2ea1fd18
      oliver@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=77340
      
      Reviewed by Gavin Barraclough.
      
      Source/JavaScriptCore:
      
      Function calls and new expressions have the same semantics for
      assignment, so should simply share their lhs handling.
      
      * parser/Parser.cpp:
      (JSC::::parseMemberExpression):
      
      LayoutTests:
      
      Add new tests for correct parsing of new expressions
      
      * fast/js/parser-syntax-check-expected.txt:
      * fast/js/script-tests/parser-syntax-check.js:
      (runTest):
      (invalid):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106297 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      2ea1fd18
    • barraclough@apple.com's avatar
      Unreviewed ARMv7 build fix. · bd8628e7
      barraclough@apple.com authored
      * tools/CodeProfiling.cpp:
      (JSC):
      (JSC::setProfileTimer):
      (JSC::CodeProfiling::begin):
      (JSC::CodeProfiling::end):
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106296 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      bd8628e7
    • levin@chromium.org's avatar
      Using OS(WIN) or OS(MAC) should cause a build error. · 0cf96d96
      levin@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77162
      
      Reviewed by Darin Adler.
      
      * wtf/Platform.h: Expand them into something that will
       cause a compile error.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106295 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      0cf96d96
    • tony@chromium.org's avatar
      css3/flexbox/cross-axis-scrollbar.html ref test fails on Windows Chromium bots · 2f13351f
      tony@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77309
      
      Reviewed by Dirk Pranke.
      
      It looks like the scrollbars on windows overflow by a pixel, so the
      edge of the scrollbar is being covered up by the trailing green box.
      Use z-order to force the scrollbar on top.
      
      This is kind of speculative. I want to see if the test passes on the
      bots before removing the line in test_expectations.txt.
      
      * css3/flexbox/cross-axis-scrollbar-expected.html:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106294 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      2f13351f
    • ojan@chromium.org's avatar
      Parsing test_expecations.txt + Skipped lists takes too long · c1c1bea6
      ojan@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77059
      
      Reviewed by Dirk Pranke.
      
      This saves ~100ms on the Apple Mac port.
      -memoize a bunch of path methods.
      -Avoid doing multiple disk accesses per line.
      -Parse the skipped list directly instead of turning it into a test_expecations.txt
      formatting string and parsing that.
      
      * Scripts/webkitpy/layout_tests/models/test_expectations.py:
      (TestExpectationParser):
      (TestExpectationParser.expectation_for_skipped_test):
      (TestExpectationParser._parse_line):
      (TestExpectationParser._collect_matching_tests):
      (TestExpectations.__init__):
      (TestExpectations._add_skipped_tests):
      * Scripts/webkitpy/layout_tests/models/test_expectations_unittest.py:
      (test_add_skipped_tests):
      (test_add_skipped_tests_duplicate):
      * Scripts/webkitpy/layout_tests/port/base.py:
      (Port):
      (Port.test_isfile):
      (Port.normalize_test_name):
      (Port.layout_tests_dir):
      (Port.abspath_for_test):
      * Scripts/webkitpy/layout_tests/run_webkit_tests.py:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106293 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      c1c1bea6
    • rniwa@webkit.org's avatar
      webkit-perf.appspot.com should have an ability to merge tests · 682120a3
      rniwa@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77250
      
      Reviewed by Adam Barth.
      
      Added admin/merge-tests. This page allows administrators to easily merge test objects.
      
      Also add the forgotten title.png and rename api/create to admin/create since we don't allow
      non-admins to create new models anyway.
      
      * Websites/webkit-perf.appspot.com/app.yaml:
      * Websites/webkit-perf.appspot.com/css: Added.
      * Websites/webkit-perf.appspot.com/css/title.png: Added.
      * Websites/webkit-perf.appspot.com/main.py:
      * Websites/webkit-perf.appspot.com/merge_tests.yaml: Added.
      * Websites/webkit-perf.appspot.com/merge_tests_handler.py: Added.
      (MergeTestHandler):
      (MergeTestHandler.get):
      (MergeTestHandler.post):
      * Websites/webkit-perf.appspot.com/models.py:
      (deleteModelWithNumericIdHolder):
      * Websites/webkit-perf.appspot.com/static/create-models.html:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106292 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      682120a3
    • leviw@chromium.org's avatar
      !m_insideRegionPaint assertion in RenderRegion.cpp is invalid · f4f6ca88
      leviw@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77372
      
      Reviewed by David Hyatt.
      
      Removing the !m_insideRegionPaint assertion that's in three functions in
      RenderRegion. It's triggering in numerous layout tests and isn't valid.
      
      No new tests as this just removes assertions.
      
      * rendering/RenderRegion.cpp:
      (WebCore::RenderRegion::setRenderBoxRegionInfo):
      (WebCore::RenderRegion::takeRenderBoxRegionInfo):
      (WebCore::RenderRegion::removeRenderBoxRegionInfo):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106291 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      f4f6ca88
    • commit-queue@webkit.org's avatar
      [BlackBerry] OS(QNX) also has TM_GMTOFF, TM_ZONE, and TIMEGM · 0676526d
      commit-queue@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77360
      
      Patch by Yong Li <yoli@rim.com> on 2012-01-30
      Reviewed by Rob Buis.
      
      Turn on HAVE(TM_GMTOFF), HAVE(TM_ZONE), and HAVE(TIMEGM)
      for OS(QNX).
      
      * wtf/Platform.h:
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106290 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      0676526d
    • enne@google.com's avatar
      [chromium] Always pre-reserve scrollbar and scroll corner textures · bcd53c54
      enne@google.com authored
      https://bugs.webkit.org/show_bug.cgi?id=77251
      
      Reviewed by James Robinson.
      
      Source/WebCore:
      
      Add a flag to LayerChromium that says that a layer's textures should
      always be reserved. Prior to painting layers, find all layers marked
      as such and reserve their textures. This will prevent texture memory
      limits from being hit before the root layer's scrollbars are reserved
      and painted.
      
      * platform/graphics/chromium/LayerChromium.cpp:
      (WebCore::LayerChromium::LayerChromium):
      * platform/graphics/chromium/LayerChromium.h:
      (LayerChromium):
      (WebCore::LayerChromium::reserveTextures):
      (WebCore::LayerChromium::setAlwaysReserveTextures):
      (WebCore::LayerChromium::alwaysReserveTextures):
      * platform/graphics/chromium/TiledLayerChromium.cpp:
      (WebCore::TiledLayerChromium::reserveTextures):
      * platform/graphics/chromium/TiledLayerChromium.h:
      ():
      * platform/graphics/chromium/cc/CCLayerTreeHost.cpp:
      (WebCore::CCLayerTreeHost::updateLayers):
      (WebCore::CCLayerTreeHost::reserveTextures):
      (WebCore):
      * platform/graphics/chromium/cc/CCLayerTreeHost.h:
      ():
      
      Source/WebKit/chromium:
      
      * src/NonCompositedContentHost.cpp:
      (WebKit::reserveScrollbarLayers):
      (WebKit):
      (WebKit::NonCompositedContentHost::setViewport):
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106289 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      bcd53c54
    • barraclough@apple.com's avatar
      Speculative Windows build fix. · 93be7539
      barraclough@apple.com authored
      * assembler/MacroAssemblerCodeRef.h:
      (FunctionPtr):
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106288 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      93be7539
    • jocelyn.turcotte@nokia.com's avatar
      [Qt] Disconnect the LayerTreeHost from layers when they get removed from their parent. · 349ca233
      jocelyn.turcotte@nokia.com authored
      https://bugs.webkit.org/show_bug.cgi?id=77362
      
      Reviewed by Noam Rosenthal.
      
      This fixes crashes seen in layout tests under certain timing circumstances
      that appeared after r106109 was landed.
      
      * WebProcess/WebCoreSupport/WebGraphicsLayer.cpp:
      (WebCore::WebGraphicsLayer::replaceChild):
      (WebCore::WebGraphicsLayer::removeFromParent):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106287 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      349ca233
    • bdakin@apple.com's avatar
      Speculative 32-bit build-fix. · 675a6f3c
      bdakin@apple.com authored
      * WebCore.exp.in:
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106286 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      675a6f3c
    • leviw@chromium.org's avatar
      Unreviewed gardening. Marking fast/js/dfg-int32array-overflow-values.html as slow · 0930a431
      leviw@chromium.org authored
      for Chromium Debug bots.
      
      * platform/chromium/test_expectations.txt:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106285 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      0930a431
    • rniwa@webkit.org's avatar
      Roll Chromium DEPS from 119623 to 119700. · b8bd0080
      rniwa@webkit.org authored
      * DEPS:
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106284 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      b8bd0080
    • leviw@chromium.org's avatar
      fast/css/fontsize-unit-rems-crash.html should use dumpAsText · 8563cf95
      leviw@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77351
      
      Reviewed by Simon Fraser.
      
      Switching the above test to only use text results as the html element is display:none.
      
      * fast/css/fontsize-unit-rems-crash-expected.txt:
      * fast/css/fontsize-unit-rems-crash.html:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106283 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      8563cf95
    • alexis.menard@openbossa.org's avatar
      Unreviewed. Add myself to CSS, GStreamer, Qt related watchlists. · 14a72abb
      alexis.menard@openbossa.org authored
              
      * Scripts/webkitpy/common/config/watchlist:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106282 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      14a72abb
    • mihnea@adobe.com's avatar
      [CSSRegions]Add support for background-color in region styling · 0ddff378
      mihnea@adobe.com authored
      https://bugs.webkit.org/show_bug.cgi?id=71488
      
      Reviewed by David Hyatt.
      
      Source/WebCore:
      
      Based on work by Alexandru Chiculita (achicu@adobe.com).
      Previous patches for region styling were touching RenderObject::style() method. After several attempts to avoid regressions
      (including caching of RenderObject::style() pointer in most used methods), we decided to attempt a different approach:
      Step1: before each region is repainted, we compute the style for each box that falls into the region
      Step2: before paint, we store the box original style
      Step3: paint the region contents using the style in region
      Step4: after paint is finished, we restore the box original style (and store the box style in region for future region paint)
      
      Tests for region styling are also enabled with this patch.
      
      * WebCore.exp.in:
      * rendering/RenderFlowThread.cpp:
      (WebCore::RenderFlowThread::clearRenderBoxCustomStyle):
      (WebCore::RenderFlowThread::setRegionRangeForBox):
      * rendering/RenderFlowThread.h:
      ():
      * rendering/RenderLayer.cpp:
      (WebCore::RenderLayer::paint):
      (WebCore::RenderLayer::hitTest):
      * rendering/RenderObject.cpp:
      * rendering/RenderObject.h:
      (WebCore::RenderObject::style):
      (RenderObject):
      * rendering/RenderObjectChildList.cpp:
      (WebCore::RenderObjectChildList::removeChildNode):
      * rendering/RenderRegion.cpp:
      (WebCore::RenderRegion::RenderRegion):
      (WebCore::RenderRegion::setRegionBoxesRegionStyle):
      (WebCore):
      (WebCore::RenderRegion::restoreRegionBoxesOriginalStyle):
      (WebCore::RenderRegion::paintReplaced):
      (WebCore::RenderRegion::setRenderBoxRegionInfo):
      (WebCore::RenderRegion::takeRenderBoxRegionInfo):
      (WebCore::RenderRegion::removeRenderBoxRegionInfo):
      (WebCore::RenderRegion::renderBoxRegionStyle):
      (WebCore::RenderRegion::computeStyleInRegion):
      (WebCore::RenderRegion::clearBoxStyleInRegion):
      * rendering/RenderRegion.h:
      (RenderRegion):
      * rendering/RenderView.cpp:
      (WebCore::RenderView::RenderView):
      * rendering/RenderView.h:
      (WebCore):
      (RenderView):
      
      LayoutTests:
      
      Enable the region styling tests again.
      
      * platform/mac-snowleopard/Skipped:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106281 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      0ddff378
    • jberlin@webkit.org's avatar
      WebCore build exceeds address space on 32-bit Windows builders (again). · cfdc9844
      jberlin@webkit.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77357
      
      Reviewed by Adam Roben.
      
      Add an Inspector All-In-One file, but only use it in Release and Production builds.
      This differs from our other All-In-One files, but it is a better approach because it makes
      debugging possible in the Debug configuration (the symbols will be in the correct .obj file
      for the original .cpp files).
      
      * WebCore.vcproj/WebCore.vcproj:
      Also, let VS have its way with the vcproj file.
      * inspector/InspectorAllInOne.cpp: Added.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106280 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      cfdc9844
    • jsbell@chromium.org's avatar
      IndexedDB: Add tests for structured clone data · 129a6d0c
      jsbell@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=74897
      
      Validate that the types listed in the HTML5 "structured clone algorithm"
      round trip properly through Indexed DB storage. Types with incomplete
      support in the Chromium port are excluded (http://crbug.com/108012).
      Landing disabled since cr-linux is failing test for unknown reasons.
      
      Reviewed by Tony Chang.
      
      * platform/chromium/test_expectations.txt:
      * storage/indexeddb/resources/test-data.html: Added.
      * storage/indexeddb/resources/test-data.txt: Added.
      * storage/indexeddb/structured-clone-expected.txt: Added.
      * storage/indexeddb/structured-clone.html: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106279 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      129a6d0c
    • mdelaney@apple.com's avatar
      Limit periodic flushing inside ImageBufferCG to just Lion · 7da9339f
      mdelaney@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=77353
      <rdar://problem/10328309>
      
      Reviewed by Chris Marrin.
      
      * platform/graphics/cg/ImageBufferCG.cpp:
      * platform/graphics/cg/ImageBufferDataCG.h:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106278 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      7da9339f
    • barraclough@apple.com's avatar
      https://bugs.webkit.org/show_bug.cgi?id=77163 · 19463376
      barraclough@apple.com authored
      MacroAssemblerCodeRef.h uses OS(WIN) instead of OS(WINDOWS)
      
      Rubber stamped by Geoff Garen
      
      * assembler/MacroAssemblerCodeRef.h:
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106277 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      19463376
    • barraclough@apple.com's avatar
      Unreviewed build fix for interpreter builds. · a2d7734e
      barraclough@apple.com authored
      * bytecode/CodeBlock.cpp:
      (JSC::CodeBlock::CodeBlock):
      * bytecode/CodeBlock.h:
      (CodeBlock):
      * interpreter/Interpreter.cpp:
      (JSC::Interpreter::privateExecute):
      * tools/CodeProfile.cpp:
      (JSC::CodeProfile::sample):
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106276 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      a2d7734e
    • tommyw@google.com's avatar
      Memory leak caused by PeerConnection add a NULL media stream · 42a4e50d
      tommyw@google.com authored
      https://bugs.webkit.org/show_bug.cgi?id=76150
      
      Source/WebCore:
      
      It was my missunderstanding that the IDL keyword [StrictTypeChecking] also protects against
      null or undefined arguments, it doesn't. Added checks for null pointers.
      
      Reviewed by Adam Barth.
      
      Test: fast/mediastream/peerconnection-addstream.html
      
      * mediastream/PeerConnection.cpp:
      (WebCore::PeerConnection::addStream):
      (WebCore::PeerConnection::removeStream):
      
      LayoutTests:
      
      Added a basic LayoutTest for null/undefined arguments to PeerConnection::addStream().
      
      Reviewed by Adam Barth.
      
      * fast/mediastream/peerconnection-addstream-expected.txt: Added.
      * fast/mediastream/peerconnection-addstream.html: Added.
      * fast/mediastream/script-tests/peerconnection-addstream.js: Added.
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106275 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      42a4e50d
    • aestes@apple.com's avatar
      [Windows] Optionally invert colors when drawing to a WebView's backing store. · f80d5dae
      aestes@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=77168
      
      Reviewed by Sam Weinig.
      
      Source/WebCore:
      
      * css/CSSPrimitiveValueMappings.h: Assert that CompositeDifference is
      not converted to a CSS value. Exposing a new compositing operation to
      CSS is outside the scope of this patch.
      (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
      * platform/graphics/GraphicsTypes.h: Add CompositeDifference as a
      CompositeOperator. Also, remove an outdated comment.
      * platform/graphics/cg/GraphicsContextCG.cpp:
      (WebCore::GraphicsContext::setPlatformCompositeOperation): Map
      CompositeDifference to kCGBlendModeDifference.
      
      Source/WebKit/win:
      
      * WebView.cpp:
      (WebView::WebView): Initialize m_shouldInvertColors to false.
      (WebView::paintIntoBackingStore): If m_shouldInvertColors is true, draw
      an opaque white quad using the CompositeDifference blend mode. This
      blend operation instructs CoreGraphics to take the difference between
      the source pixel (white) and the background pixel, resulting in an
      inverted pixel.
      * WebView.h: Define m_shouldInvertColors.
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106274 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      f80d5dae
    • mdelaney@apple.com's avatar
      Limit the shadow offset CG hack to just SL and Lion · 5d6eeee5
      mdelaney@apple.com authored
      https://bugs.webkit.org/show_bug.cgi?id=77348
      <rdar://problem/10158016>
      
      Reviewed by Chris Marrin.
      
      No new tests, current tests cover this.
      
      * platform/graphics/cg/GraphicsContextCG.cpp:
      (WebCore::GraphicsContext::setPlatformShadow):
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106273 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      5d6eeee5
    • bdakin@apple.com's avatar
      https://bugs.webkit.org/show_bug.cgi?id=77263 · 9770bbf4
      bdakin@apple.com authored
      PlatformScreenMac should not rely on NSWindow for important bits of data
      
      Reviewed by Geoff Garen.
      
      ../WebCore: 
      
      The main problem is that we cannot rely on the NSWindow for information about 
      the deviceScaleFactor because we cannot access an NSWindow from within 
      WebCore for WebKit2 windows. Instead, we can fetch it from 
      WebCore::deviceScaleFactor(), but we need a Frame to call that. So 
      screenAvailableRect and screenRect both now take a FrameView* instead of a 
      Widget*. All existing call sites actually sent a FrameView in anyway, so this 
      is not a big change, but it does require touching a lot of platforms.
      * WebCore.exp.in:
      * platform/PlatformScreen.h:
      (WebCore):
      * platform/blackberry/PlatformScreenBlackBerry.cpp:
      (WebCore::screenAvailableRect):
      (WebCore::screenRect):
      * platform/chromium/PlatformScreenChromium.cpp:
      (WebCore::screenRect):
      (WebCore::screenAvailableRect):
      * platform/chromium/PlatformSupport.h:
      (WebCore):
      (PlatformSupport):
      ():
      * platform/efl/PlatformScreenEfl.cpp:
      (WebCore::screenRect):
      (WebCore::screenAvailableRect):
      * platform/gtk/PlatformScreenGtk.cpp:
      (WebCore::screenRect):
      (WebCore::screenAvailableRect):
      * platform/qt/PlatformScreenQt.cpp:
      (WebCore::screenRect):
      (WebCore::screenAvailableRect):
      * platform/win/PlatformScreenWin.cpp:
      (WebCore::screenRect):
      (WebCore::screenAvailableRect):
      * platform/wx/ScreenWx.cpp:
      (WebCore::screenRect):
      (WebCore::screenAvailableRect):
      
      The Mac-only functions toUserSpace() and toDeviceSpace() were also updated to 
      take a parameter for the deviceScaleFactor.
      * platform/mac/PlatformScreenMac.mm:
      (WebCore::screenRect):
      (WebCore::screenAvailableRect):
      (WebCore::toUserSpace):
      (WebCore::toDeviceSpace):
      
      ../WebKit/chromium: 
      
      These two functions now take a FrameView instead of a Widget.
      * src/PlatformSupport.cpp:
      (WebCore::PlatformSupport::screenRect):
      (WebCore::PlatformSupport::screenAvailableRect):
      
      ../WebKit/mac: 
      
      toUserSpace() and toDeviceSpace() now take a parameter for the 
      deviceScaleFactor.
      * WebCoreSupport/WebChromeClient.mm:
      (windowScaleFactor):
      (WebChromeClient::setWindowRect):
      (WebChromeClient::windowRect):
      
      ../WebKit2: 
      
      toUserSpace() and toDeviceSpace() now take a parameter for the 
      deviceScaleFactor.
      * UIProcess/API/mac/WKView.mm:
      (-[WKView _convertToDeviceSpace:]):
      (-[WKView _convertToUserSpace:]):
      
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106271 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      9770bbf4
    • leviw@chromium.org's avatar
      Enable web intents layout tests for Chromium build. · 036f16ca
      leviw@chromium.org authored
      https://bugs.webkit.org/show_bug.cgi?id=77347
      
      Patch by Greg Billock <gbillock@google.com> on 2012-01-30
      Reviewed by Ryosuke Niwa.
      
      * platform/chromium/test_expectations.txt:
      
      
      git-svn-id: http://svn.webkit.org/repository/webkit/trunk@106266 268f45cc-cd09-0410-ab3c-d52691b4dbfc
      036f16ca