WebKit.git
10 years ago2011-03-28 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Mon, 28 Mar 2011 09:33:59 +0000 (09:33 +0000)]
2011-03-28  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Dan Bernstein.

        Assigning location.path to something that doesn't start with / mangles the authority
        https://bugs.webkit.org/show_bug.cgi?id=57209
        <rdar://problem/9195132>

        Tests: http/tests/uri/assign-path-with-leading-slash.html
               http/tests/uri/assign-path-without-leading-slash.html

        * platform/KURL.cpp:
        (WebCore::KURL::setPath): If the new path does not have a leading slash, add one.
2011-03-28  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Dan Bernstein.

        Assigning location.path to something that doesn't start with / mangles the authority
        https://bugs.webkit.org/show_bug.cgi?id=57209
        <rdar://problem/9195132>

        * http/tests/uri/assign-path-with-leading-slash-expected.txt: Added.
        * http/tests/uri/assign-path-with-leading-slash.html: Added.
        * http/tests/uri/assign-path-without-leading-slash-expected.txt: Added.
        * http/tests/uri/assign-path-without-leading-slash.html: Added.
        * http/tests/uri/resources/assign-path-landing.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82093 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-28 Yuta Kitamura <yutak@chromium.org>
yutak@chromium.org [Mon, 28 Mar 2011 09:28:44 +0000 (09:28 +0000)]
2011-03-28  Yuta Kitamura  <yutak@chromium.org>

        Unreviewed, add Chromium test expectation for fast/text/justify-ideograph-leading-expansion.html.

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82092 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-28 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Mon, 28 Mar 2011 09:19:50 +0000 (09:19 +0000)]
2011-03-28  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Eric Seidel.

        Extract functions to update logical width from computeInlineDirectionPositionsForLine
        https://bugs.webkit.org/show_bug.cgi?id=57213

        Extracted updateLogicalWidthForLeftAlignedBlock, updateLogicalWidthForRightAlignedBlock,
        and updateLogicalWidthForCenterAlignedBlock from computeInlineDirectionPositionsForLine.

        They are used to update logical widths, logical left, and total logical width for left,
        right, and center aligned blocks.

        * rendering/RenderBlockLineLayout.cpp:
        (WebCore::updateLogicalWidthForLeftAlignedBlock):
        (WebCore::updateLogicalWidthForRightAlignedBlock):
        (WebCore::updateLogicalWidthForCenterAlignedBlock):
        (WebCore::RenderBlock::computeInlineDirectionPositionsForLine):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82091 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-28 Evan Martin <evan@chromium.org>
evan@chromium.org [Mon, 28 Mar 2011 08:57:27 +0000 (08:57 +0000)]
2011-03-28  Evan Martin  <evan@chromium.org>

        Reviewed by Ryosuke Niwa.

        Refactor duplicate code in HTMLTextElement
        https://bugs.webkit.org/show_bug.cgi?id=57215

        * html/HTMLTitleElement.cpp:
        (WebCore::HTMLTitleElement::childrenChanged):
        This method did an equivalent loop to the text() method on
        the same object.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82090 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-28 Yuta Kitamura <yutak@chromium.org>
yutak@chromium.org [Mon, 28 Mar 2011 08:26:16 +0000 (08:26 +0000)]
2011-03-28  Yuta Kitamura  <yutak@chromium.org>

        Unreviewed, update Chromium test results for plugins/plugin-javascript-access.html.

        * platform/chromium-linux/plugins/plugin-javascript-access-expected.txt:
        * platform/chromium-win/plugins/plugin-javascript-access-expected.txt:
        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82089 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Fumitoshi Ukai <ukai@chromium.org>
ukai@chromium.org [Mon, 28 Mar 2011 06:46:23 +0000 (06:46 +0000)]
2011-03-27  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Don't call WebSocket::didClose() more than once.
        https://bugs.webkit.org/show_bug.cgi?id=57081

        If WebSocket close() is called, and connection is established, then
        it will call didClose() that resets m_channel to 0.
        After that, when connection is closed, WebSocketChannel will call
        didClose for the WebSocket instance.

        * http/tests/websocket/tests/close-unref-websocket-expected.txt: Added.
        * http/tests/websocket/tests/close-unref-websocket.html: Added.
        * http/tests/websocket/tests/hanging-handshake_wsh.py: Added.
2011-03-27  Fumitoshi Ukai  <ukai@chromium.org>

        Reviewed by Alexey Proskuryakov.

        Don't call WebSocket::didClose() more than once.
        https://bugs.webkit.org/show_bug.cgi?id=57081

        If WebSocket close() is called, and connection is established, then
        it will call didClose() that resets m_channel to 0.
        After that, when connection is closed, WebSocketChannel will call
        didClose for the WebSocket instance.

        Call WebSocketChannel::disconnect() before m_channel = 0 to make sure
        WebSocketChannel suppress the second didClose().

        Test: http/tests/websocket/tests/close-unref-websocket.html

        * websockets/WebSocket.cpp:
        (WebCore::WebSocket::didClose):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82088 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Andy Estes <aestes@apple.com>
aestes@apple.com [Mon, 28 Mar 2011 06:41:41 +0000 (06:41 +0000)]
2011-03-27  Andy Estes  <aestes@apple.com>

        Reviewed by Maciej Stachowiak.

        Correctly get a plug-in's MIME type when it uses WebPluginMIMETypesFilename
        https://bugs.webkit.org/show_bug.cgi?id=57205

        If the plug-in's Info.plist uses WebPluginMIMETypesFilename to specify
        plug-in MIME types, WebKit has to check for a property list file in
        ~/Library/Preferences for the MIME type dictionary.

        * Shared/Plugins/Netscape/mac/NetscapePluginModuleMac.mm:
        (WebKit::getMIMETypesFromPluginBundle): If the bundle's Info dictionary
        has the key WebPluginMIMETypesFilename, open the property list
        specified by that key's value and return the MIME type dictionary from
        there. Otherwise, return the MIME type dictionary specified by the key
        WebPluginMIMETypes.
        (WebKit::getPluginInfoFromPropertyLists): Call
        getMIMETypesFromPluginBundle().

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82087 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Yuta Kitamura <yutak@chromium.org>
yutak@chromium.org [Mon, 28 Mar 2011 06:09:43 +0000 (06:09 +0000)]
2011-03-27  Yuta Kitamura  <yutak@chromium.org>

        Unreviewed, add Chromium test results for fast/blockflow/fallback-orientation.html.

        * platform/chromium-mac-leopard/fast/blockflow/fallback-orientation-expected.checksum: Added.
        * platform/chromium-mac-leopard/fast/blockflow/fallback-orientation-expected.png: Added.
        * platform/chromium-win-vista/fast/blockflow/fallback-orientation-expected.checksum: Added.
        * platform/chromium-win-vista/fast/blockflow/fallback-orientation-expected.png: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82086 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Mon, 28 Mar 2011 05:15:38 +0000 (05:15 +0000)]
2011-03-27  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Fix script-src redirect handling
        https://bugs.webkit.org/show_bug.cgi?id=57196

        Test both allow => disallow and disallow => allow redirect cases.
        Previously, we had incorrect expectations for one of the redirect
        cases.  Also, I've updated the policy syntax to match the default-src
        syntax.

        * http/tests/security/contentSecurityPolicy/script-src-redirect-expected.txt:
        * http/tests/security/contentSecurityPolicy/script-src-redirect.html:
2011-03-27  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Fix script-src redirect handling
        https://bugs.webkit.org/show_bug.cgi?id=57196

        Resource-loading requirements in CSP apply to each hop in the redirect
        chain.  To make that work properly, we need to move enforcement into
        the loader.  Fortunately, we already have a choke-point in the loader
        for enforcing this kind of policy.

        * dom/ScriptElement.cpp:
        (WebCore::ScriptElement::requestScript):
        * html/parser/HTMLDocumentParser.cpp:
        * html/parser/HTMLDocumentParser.h:
        * html/parser/HTMLScriptRunnerHost.h:
        * loader/cache/CachedResourceLoader.cpp:
        (WebCore::CachedResourceLoader::canRequest):
        * page/ContentSecurityPolicy.cpp:
        (WebCore::ContentSecurityPolicy::allowScriptFromSource):
        * page/ContentSecurityPolicy.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82085 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Jer Noble <jer.noble@apple.com>
jer.noble@apple.com [Mon, 28 Mar 2011 04:27:54 +0000 (04:27 +0000)]
2011-03-27  Jer Noble  <jer.noble@apple.com>

        Reviewed by Maciej Stachowiak.

        Full Screen: disable keyboard access by default
        https://bugs.webkit.org/show_bug.cgi?id=56684

        Pass whether keyboard access was requested up to ChromeClient when asking
        if fullscreen mode is supported.

        * dom/Document.cpp:
        (WebCore::Document::webkitRequestFullScreenForElement):
        * page/ChromeClient.h:
        (WebCore::ChromeClient::supportsFullscreenForElement):
2011-03-27  Jer Noble  <jer.noble@apple.com>

        Reviewed by Maciej Stachowiak.

        Full Screen: disable keyboard access by default
        https://bugs.webkit.org/show_bug.cgi?id=56684

        Take into account whether keyboard access was requested when deciding whether full
        screen mode is supported.

        * WebCoreSupport/WebChromeClient.h:
        * WebCoreSupport/WebChromeClient.mm:
        (WebChromeClient::supportsFullScreenForElement):
        * WebView/WebView.mm:
        (-[WebView _supportsFullScreenForElement:WebCore::withKeyboard:]):
        * WebView/WebViewInternal.h:
2011-03-27  Jer Noble  <jer.noble@apple.com>

        Reviewed by Maciej Stachowiak.

        Full Screen: disable keyboard access by default
        https://bugs.webkit.org/show_bug.cgi?id=56684

        Take into account whether keyboard access was requested when deciding whether full
        screen mode is supported.

        * UIProcess/WebFullScreenManagerProxy.cpp:
        (WebKit::WebFullScreenManagerProxy::supportsFullScreen):
        * UIProcess/WebFullScreenManagerProxy.h:
        * UIProcess/WebFullScreenManagerProxy.messages.in:
        * WebProcess/FullScreen/WebFullScreenManager.cpp:
        (WebKit::WebFullScreenManager::supportsFullScreen):
        * WebProcess/FullScreen/WebFullScreenManager.h:
        * WebProcess/WebCoreSupport/WebChromeClient.cpp:
        (WebKit::WebChromeClient::supportsFullScreenForElement):
        * WebProcess/WebCoreSupport/WebChromeClient.h:
2011-03-27  Jer Noble  <jer.noble@apple.com>

        Reviewed by Maciej Stachowiak.

        Full Screen: disable keyboard access by default
        https://bugs.webkit.org/show_bug.cgi?id=56684

        Accept the withKeyboard parameter to supportsFullScreenForElement.

        * DumpRenderTree/mac/UIDelegate.mm:
        (-[UIDelegate webView:supportsFullScreenForElement:withKeyboard:]):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82084 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Mon, 28 Mar 2011 02:33:12 +0000 (02:33 +0000)]
2011-03-27  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by David Levin.

        check-webkit-style should check ChangeLog for a valid bug number
        https://bugs.webkit.org/show_bug.cgi?id=57184

        * Scripts/webkitpy/style/checker.py:
        * Scripts/webkitpy/style/checker_unittest.py:
        * Scripts/webkitpy/style/checkers/changelog.py: Added.
        * Scripts/webkitpy/style/checkers/changelog_unittest.py: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82083 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Benjamin Poulain <benjamin.poulain@nokia.com>
benjamin.poulain@nokia.com [Sun, 27 Mar 2011 22:54:59 +0000 (22:54 +0000)]
2011-03-27  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Andreas Kling.

        [Qt] QtWebKit will not compile with QT_ASCII_CAST_WARNINGS enabled
        https://bugs.webkit.org/show_bug.cgi?id=57087

        Use explicit conversion for string to avoid depending on the default codec
        installed by the user code.

        * WebKitTestRunner/InjectedBundle/qt/ActivateFontsQt.cpp:
        (WTR::activateFonts):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82082 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sun, 27 Mar 2011 19:56:02 +0000 (19:56 +0000)]
2011-03-27  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Dock doesn't come back after leaving fullscreen on Hulu
        https://bugs.webkit.org/show_bug.cgi?id=57189
        <rdar://problem/9192413>

        Port some code over from the old WebKit1 plug-in host.

        * PluginProcess/mac/PluginProcessShim.mm:
        (WebKit::shimShowMenuBar):
        (WebKit::shimHideMenuBar):
        (WebKit::shimIsMenuBarVisible):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82078 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoKeep a strong reference to the connection to avoid crashes.
andersca@apple.com [Sun, 27 Mar 2011 19:26:47 +0000 (19:26 +0000)]
Keep a strong reference to the connection to avoid crashes.

Reviewed by Sam Weinig.

* Platform/CoreIPC/Connection.cpp:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82077 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Andreas Kling <kling@webkit.org>
andreas.kling@nokia.com [Sun, 27 Mar 2011 19:15:58 +0000 (19:15 +0000)]
2011-03-27  Andreas Kling  <kling@webkit.org>

        Fix build warning about IconDatabaseClient.h (wrong path.)

        * QtWebKit.pro:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82076 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Andreas Kling <kling@webkit.org>
andreas.kling@nokia.com [Sun, 27 Mar 2011 19:12:19 +0000 (19:12 +0000)]
2011-03-27  Andreas Kling  <kling@webkit.org>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Support for CSS color and background-color properties on select element's dropdown list
        https://bugs.webkit.org/show_bug.cgi?id=51627

        Extend the QWebSelectData interface with background and foreground colors
        for the whole menu, as well as per-item. Hook it up to the PopupMenuStyle
        getters in RenderMenuList.

        * Api/qwebkitplatformplugin.h:
        * WebCoreSupport/PopupMenuQt.cpp:
        (SelectData::backgroundColor):
        (SelectData::foregroundColor):
        (SelectData::itemBackgroundColor):
        (SelectData::itemForegroundColor):
        * WebCoreSupport/QtFallbackWebPopup.cpp:
        (WebCore::QtFallbackWebPopup::show):
        (WebCore::QtFallbackWebPopup::populate):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82075 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Yi Shen <yi.4.shen@nokia.com>
yi.4.shen@nokia.com [Sun, 27 Mar 2011 18:26:11 +0000 (18:26 +0000)]
2011-03-27  Yi Shen  <yi.4.shen@nokia.com>

        Reviewed by Andreas Kling.

        [Qt][Symbian] Fix Api test failure -- microFocusCoordinates
        https://bugs.webkit.org/show_bug.cgi?id=57108

        Since the canvas is not self-closing tag, we need to add '</canvas>'.

        * tests/qgraphicswebview/tst_qgraphicswebview.cpp:
        (tst_QGraphicsWebView::microFocusCoordinates):
        * tests/qwebview/tst_qwebview.cpp:
        (tst_QWebView::microFocusCoordinates):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82074 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Sun, 27 Mar 2011 17:52:39 +0000 (17:52 +0000)]
2011-03-27  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Eric Seidel.

        [Qt] Build fix: Define WTF_USE_TEXTURE_MAPPER=1 when CONFIG contains texmap.
        https://bugs.webkit.org/show_bug.cgi?id=57143

        Qt WebKit uses USE(TEXTURE_MAPPER) guard. Check texmap in CONFIG and
        define WTF_USE_TEXTURE_MAPPER=1.

        * QtWebKit.pro:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82073 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Sun, 27 Mar 2011 17:42:08 +0000 (17:42 +0000)]
2011-03-27  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Andreas Kling.

        Fix agentOS() in FrameLoaderClientEfl
        https://bugs.webkit.org/show_bug.cgi?id=57177

        Replace old PLATFORM macro with OS and CPU.

        * WebCoreSupport/FrameLoaderClientEfl.cpp:
        (WebCore::agentOS):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82072 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Sun, 27 Mar 2011 17:38:45 +0000 (17:38 +0000)]
2011-03-27  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Andreas Kling.

        [EFL] Use an appropriate user agent string
        https://bugs.webkit.org/show_bug.cgi?id=57176

        * WebCoreSupport/FrameLoaderClientEfl.cpp:
        (WebCore::composeUserAgent):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82071 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Sun, 27 Mar 2011 17:28:57 +0000 (17:28 +0000)]
2011-03-27  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Andreas Kling.

        [WINCE] Use a appropriate user agent string
        https://bugs.webkit.org/show_bug.cgi?id=57175

        * WebCoreSupport/FrameLoaderClientWinCE.cpp:
        (WebKit::FrameLoaderClientWinCE::userAgent):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82070 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Sun, 27 Mar 2011 17:24:43 +0000 (17:24 +0000)]
2011-03-27  Patrick Gansterer  <paroga@webkit.org>

        Reviewed by Andreas Kling.

        [CMake] Generate WebKitVersion.h
        https://bugs.webkit.org/show_bug.cgi?id=57128

        This file will be used for the user agent string by the CMake based ports.

        * CMakeLists.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82069 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Yuta Kitamura <yutak@chromium.org>
yutak@chromium.org [Sun, 27 Mar 2011 16:36:28 +0000 (16:36 +0000)]
2011-03-27  Yuta Kitamura  <yutak@chromium.org>

        Unreviewed, add new Chromium test result and expectations.

        * platform/chromium-linux/svg/W3C-SVG-1.1/animate-elem-08-t-expected.checksum:
        * platform/chromium-linux/svg/W3C-SVG-1.1/animate-elem-08-t-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/animate-elem-08-t-expected.checksum:
        * platform/chromium-win/svg/W3C-SVG-1.1/animate-elem-08-t-expected.png:
        * platform/chromium-win/svg/W3C-SVG-1.1/animate-elem-08-t-expected.txt:
        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82068 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Ben Taylor <bentaylor.solx86@gmail.com>
commit-queue@webkit.org [Sun, 27 Mar 2011 16:36:13 +0000 (16:36 +0000)]
2011-03-27  Ben Taylor  <bentaylor.solx86@gmail.com>

        Reviewed by Alexey Proskuryakov.

        https://bugs.webkit.org/show_bug.cgi?id=57170  Fix last elements
        in an enum to remove a trailing comma. Sun Studio 12 CC errors out.

        Compile fix only, no actual code change.

        * wtf/MessageQueue.h:
2011-03-27  Ben Taylor  <bentaylor.solx86@gmail.com>

        Reviewed by Alexey Proskuryakov.

        https://bugs.webkit.org/show_bug.cgi?id=57170  Fix last elements
        in an enum to remove a trailing comma. Sun Studio 12 CC errors out.

        Compile fix only, no actual code change.

        * dom/ExceptionCode.h:
        * editing/EditorInsertAction.h:
        * loader/FrameLoaderTypes.h:
        * platform/PlatformKeyboardEvent.h:
        * platform/ScrollTypes.h:
        * platform/graphics/BitmapImage.h:
        * platform/image-decoders/ImageDecoder.h:
        * platform/network/ProtectionSpace.h:
        * platform/network/ResourceHandleClient.h:
        * platform/network/ResourceRequestBase.h:
        * platform/text/TextCodec.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82067 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Rob Buis <rwlbuis@gmail.com>
rwlbuis@webkit.org [Sun, 27 Mar 2011 14:50:17 +0000 (14:50 +0000)]
2011-03-27  Rob Buis  <rwlbuis@gmail.com>

        Unreviewed, update improve test result.

        * platform/mac-leopard/svg/W3C-SVG-1.1/animate-elem-08-t-expected.checksum:
        * platform/mac-leopard/svg/W3C-SVG-1.1/animate-elem-08-t-expected.png:
        * platform/mac/svg/W3C-SVG-1.1/animate-elem-08-t-expected.checksum:
        * platform/mac/svg/W3C-SVG-1.1/animate-elem-08-t-expected.png:
        * platform/mac/svg/W3C-SVG-1.1/animate-elem-08-t-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82066 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Andreas Kling <kling@webkit.org>
andreas.kling@nokia.com [Sun, 27 Mar 2011 14:23:30 +0000 (14:23 +0000)]
2011-03-27  Andreas Kling  <kling@webkit.org>

        Reviewed by Benjamin Poulain.

        REGRESSION(r54712): [Qt] Installed QtWebKit header does not compile.
        https://bugs.webkit.org/show_bug.cgi?id=57183

        The convenience <QtWebKit> header would include \<QtNetwork/QtNetwork\>
        which was due to the outputting code previously being wrapped in eval().

        * Api/DerivedSources.pro:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82065 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Yuta Kitamura <yutak@chromium.org>
yutak@chromium.org [Sun, 27 Mar 2011 13:26:10 +0000 (13:26 +0000)]
2011-03-27  Yuta Kitamura  <yutak@chromium.org>

        Unreviewed, update Chromium test expectations.

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82064 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Yuta Kitamura <yutak@chromium.org>
yutak@chromium.org [Sun, 27 Mar 2011 12:38:49 +0000 (12:38 +0000)]
2011-03-27  Yuta Kitamura  <yutak@chromium.org>

        Unreviewed, add new Chromium test results.

        * platform/chromium-linux/svg/text/text-path-middle-align-expected.checksum: Added.
        * platform/chromium-linux/svg/text/text-path-middle-align-expected.png: Added.
        * platform/chromium-win/fast/blockflow/fallback-orientation-expected.checksum: Added.
        * platform/chromium-win/fast/blockflow/fallback-orientation-expected.png: Added.
        * platform/chromium-win/fast/blockflow/fallback-orientation-expected.txt: Added.
        * platform/chromium-win/svg/text/text-path-middle-align-expected.checksum: Added.
        * platform/chromium-win/svg/text/text-path-middle-align-expected.png: Added.
        * platform/chromium-win/svg/text/text-path-middle-align-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82063 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Yuta Kitamura <yutak@chromium.org>
yutak@chromium.org [Sun, 27 Mar 2011 12:25:40 +0000 (12:25 +0000)]
2011-03-27  Yuta Kitamura  <yutak@chromium.org>

        Unreviewed, fix Chromium text expectation.

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82062 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Rob Buis <rwlbuis@gmail.com>
rwlbuis@webkit.org [Sun, 27 Mar 2011 11:16:44 +0000 (11:16 +0000)]
2011-03-27  Rob Buis  <rwlbuis@gmail.com>

        Reviewed by Nikolas Zimmermann.

        Text on path positioning at zero startOffset
        https://bugs.webkit.org/show_bug.cgi?id=56245

        Since for length 0 no previous point is set, choose epsilon
        length to get normal at starting point of the path.

        Test: svg/text/text-path-middle-align.svg

        * platform/graphics/Path.cpp:
        (WebCore::Path::normalAngleAtLength):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82061 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Ben Taylor <bentaylor.solx86@gmail.com>
commit-queue@webkit.org [Sun, 27 Mar 2011 09:53:52 +0000 (09:53 +0000)]
2011-03-27  Ben Taylor  <bentaylor.solx86@gmail.com>

        Reviewed by Oliver Hunt.

        https://bugs.webkit.org/show_bug.cgi?id=57151, patch derived from set
        created by Thiago Macieria in bug https://bugs.webkit.org/show_bug.cgi?id=24932

        Fix compile issue on Solaris 10/Sun Studio 12 regarding ambiguity on ?: functions

        * bindings/js/JSJavaScriptCallFrameCustom.cpp:
        (WebCore::JSJavaScriptCallFrame::thisObject):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82060 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-27 Benjamin Poulain <benjamin.poulain@nokia.com>
benjamin.poulain@nokia.com [Sun, 27 Mar 2011 07:18:55 +0000 (07:18 +0000)]
2011-03-27  Benjamin Poulain  <benjamin.poulain@nokia.com>

        Reviewed by Andreas Kling.

        [Qt] QtWebKit will not compile with QT_ASCII_CAST_WARNINGS enabled
        https://bugs.webkit.org/show_bug.cgi?id=57087

        Use explicit conversion for string to avoid depending on the default codec
        installed by the user code.

        * Api/qwebkitversion.cpp:
        (qWebKitVersion):
        * Api/qwebpage.cpp:
        (QWebPagePrivate::dynamicPropertyChangeEvent):
        (QWebPage::javaScriptConsoleMessage):
        * WebCoreSupport/DumpRenderTreeSupportQt.cpp:
        (convertToPropertyName):
        (DumpRenderTreeSupportQt::setEditingBehavior):
        (DumpRenderTreeSupportQt::plainText):
        * WebCoreSupport/EditorClientQt.cpp:
        (dumpRange):
        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (drtDescriptionSuitableForTestResult):
        (WebCore::FrameLoaderClientQt::dispatchDecidePolicyForNavigationAction):
        * WebCoreSupport/InspectorClientQt.cpp:
        (WebCore::InspectorClientQt::openInspectorFrontend):
        * WebCoreSupport/InspectorServerQt.cpp:
        (WebCore::parseWebSocketChallengeNumber):
        (WebCore::InspectorServerRequestHandlerQt::tcpReadyRead):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82059 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Jeff Miller <jeffm@apple.com>
jeffm@apple.com [Sun, 27 Mar 2011 05:19:17 +0000 (05:19 +0000)]
2011-03-26  Jeff Miller  <jeffm@apple.com>

        Reviewed by Sam Weinig.

        WKErrorCopyCFError() doesn&apos;t actually return a copy of the CFErrorRef
        https://bugs.webkit.org/show_bug.cgi?id=57172

        * Shared/API/c/cf/WKErrorCF.cpp:
        (WKErrorCopyCFError): Use RetainPtr<> and leakRef() to return a real copy.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82058 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed build fix.
jer.noble@apple.com [Sun, 27 Mar 2011 03:12:06 +0000 (03:12 +0000)]
Unreviewed build fix.

Fix a bug which was causing the "skip back" button to be missing, causing
the media layout tests to fail.

* rendering/RenderThemeMac.mm:
(WebCore::RenderThemeMac::shouldRenderMediaControlPart):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82057 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<rdar://problem/9180716> REGRESSION (r80582): Angle bracket rendered upright in verti...
mitz@apple.com [Sun, 27 Mar 2011 01:38:47 +0000 (01:38 +0000)]
<rdar://problem/9180716> REGRESSION (r80582): Angle bracket rendered upright in vertical mode
https://bugs.webkit.org/show_bug.cgi?id=57169

Reviewed by Maciej Stachowiak.

Source/WebCore:

Test: fast/blockflow/fallback-orientation.html

* platform/graphics/FontFastPath.cpp:
(WebCore::Font::glyphDataForCharacter): If the font has vertical glyphs, use it, rather
than continuing down the fallback list.

LayoutTests:

* fast/blockflow/fallback-orientation.html: Added.
* platform/mac/fast/blockflow/fallback-orientation-expected.checksum: Added.
* platform/mac/fast/blockflow/fallback-orientation-expected.png: Added.
* platform/mac/fast/blockflow/fallback-orientation-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82056 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSource/WebKit2: Revert inadvertantly commited changes.
mjs@apple.com [Sun, 27 Mar 2011 01:10:02 +0000 (01:10 +0000)]
Source/WebKit2: Revert inadvertantly commited changes.

* WebProcess/com.apple.WebProcess.sb:

Tools: Revert inadvertantly committed changes.

* Scripts/old-run-webkit-tests:
* Scripts/webkitpy/layout_tests/run_webkit_tests.py:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82055 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Maciej Stachowiak <mjs@apple.com>
mjs@apple.com [Sun, 27 Mar 2011 00:58:34 +0000 (00:58 +0000)]
2011-03-26  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Darin Adler.

        Follow-up fix for crash with giant inline stylesheets - actually don't crash, and add test
        https://bugs.webkit.org/show_bug.cgi?id=56150

        Test: fast/css/giant-stylesheet-crash.html

        * dom/StyleElement.cpp:
        (WebCore::StyleElement::process):
2011-03-26  Maciej Stachowiak  <mjs@apple.com>

        Reviewed by Darin Adler.

        Follow-up fix for crash with giant inline stylesheets - actually don't crash, and add test
        https://bugs.webkit.org/show_bug.cgi?id=56150

        * fast/css/giant-stylesheet-crash-expected.txt: Added.
        * fast/css/giant-stylesheet-crash.html: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82054 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-22 Jer Noble <jer.noble@apple.com>
jer.noble@apple.com [Sun, 27 Mar 2011 00:15:48 +0000 (00:15 +0000)]
2011-03-22  Jer Noble  <jer.noble@apple.com>

        Reviewed by Eric Carlson.

        Media controls must use full screen style when in new full screen mode.
        https://bugs.webkit.org/show_bug.cgi?id=56851

        Add new full screen styles for the full screen media elements.

        * DerivedSources.make: Add fullScreenQuickTime.css.
        * WebCore.xcodeproj/project.pbxproj: Ditto.
        * css/CSSStyleSelector.cpp: Removed loadFullScreenRulesIfNeeded().
        (WebCore::CSSStyleSelector::CSSStyleSelector): Ditto.
        (WebCore::CSSStyleSelector::styleForElement): Load full screen rules
            if needed.
        * css/fullscreenQuickTime.css: Added.
        * css/mediaControls.css:
        * html/shadow/MediaControls.cpp:
        (WebCore::isFullScreen): Added.
        (WebCore::MediaControls::create): Add new full screen volume controls.
        (WebCore::MediaControls::updateStyle): Ditto.
        (WebCore::MediaControls::update): Ditto.
        (WebCore::MediaControls::updateVolumeSliderContainer): Ditto.
        (WebCore::MediaControls::forwardEvent): Ditto.
        * rendering/MediaControlElements.cpp:
        (WebCore::MediaControlFullscreenVolumeSliderElement::MediaControlFullscreenVolumeSliderElement): Added.
        (WebCore::MediaControlFullscreenVolumeSliderElement::create): Added.
        (WebCore::MediaControlFullscreenVolumeSliderElement::shadowPseudoId): Added.
        (WebCore::MediaControlFullscreenVolumeMinButtonElement::MediaControlFullscreenVolumeMinButtonElement): Addet
        (WebCore::MediaControlFullscreenVolumeMinButtonElement::create): Added.
        (WebCore::MediaControlFullscreenVolumeMinButtonElement::defaultEventHandler): Added.
        (WebCore::MediaControlFullscreenVolumeMinButtonElement::shadowPseudoId): Added.
        (WebCore::MediaControlFullscreenVolumeMaxButtonElement::MediaControlFullscreenVolumeMaxButtonElement): Added.
        (WebCore::MediaControlFullscreenVolumeMaxButtonElement::create): Added.
        (WebCore::MediaControlFullscreenVolumeMaxButtonElement::defaultEventHandler): Added.
        (WebCore::MediaControlFullscreenVolumeMaxButtonElement::shadowPseudoId): Added.
        * rendering/MediaControlElements.h:
            Added m_overridePosition.
        (WebCore::MediaControlVolumeSliderContainerElement::setOverridesPosition): Added.
        * rendering/MediaControlElements.h:
        (WebCore::MediaControlVolumeSliderContainerElement::overridesPosition): Added.
        * rendering/RenderTheme.h:
        (WebCore::RenderTheme::extraFullScreenStyleSheet): Added.
        * rendering/RenderThemeMac.h:
        * rendering/RenderThemeMac.mm:
        (WebCore::RenderThemeMac::extraFullScreenStyleSheet): Added.
        (WebCore::RenderThemeMac::shouldRenderMediaControlPart): Render seek forward and backward
            buttons.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82053 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Jer Noble <jer.noble@apple.com>
jer.noble@apple.com [Sun, 27 Mar 2011 00:03:14 +0000 (00:03 +0000)]
2011-03-26  Jer Noble  <jer.noble@apple.com>

        Reviewed by Eric Carlson.

        Enable the Full Screen API by default in WebKit/mac and WebKit2
        https://bugs.webkit.org/show_bug.cgi?id=56956

        * WebView/WebPreferences.mm:
        (+[WebPreferences initialize]): Default the WebKitFullScreenEnabled preference to true.
2011-03-26  Jer Noble  <jer.noble@apple.com>

        Reviewed by Eric Carlson.

        Enable the Full Screen API by default in WebKit/mac and WebKit2
        https://bugs.webkit.org/show_bug.cgi?id=56956

        * Shared/WebPreferencesStore.h: Default the fullScreenEnabled preference to true.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82052 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Jer Noble <jer.noble@apple.com>
jer.noble@apple.com [Sat, 26 Mar 2011 23:46:08 +0000 (23:46 +0000)]
2011-03-26  Jer Noble  <jer.noble@apple.com>

        Reviewed by Dan Bernstein.

        RenderMedia should obey the view's flattening bit.
        https://bugs.webkit.org/show_bug.cgi?id=57156

        Paint the current frame in software when the associated FrameView
        has its flattening bit set.

        * rendering/RenderVideo.cpp:
        (WebCore::RenderVideo::paintReplaced):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82051 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Sat, 26 Mar 2011 23:05:32 +0000 (23:05 +0000)]
2011-03-26  Patrick Gansterer  <paroga@webkit.org>

        Unreviewed build fix for !ENABLE(DOM_STORAGE).

        * storage/StorageTracker.cpp:
            Added #if ENABLE(DOM_STORAGE) like in the other storage files.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82050 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdd PlatformCertificateInfo::PlatformCertificateInfo(PCCERT_CONTEXT) constructor...
jeffm@apple.com [Sat, 26 Mar 2011 22:53:49 +0000 (22:53 +0000)]
Add PlatformCertificateInfo::PlatformCertificateInfo(PCCERT_CONTEXT) constructor on Windows
https://bugs.webkit.org/show_bug.cgi?id=57152

Reviewed by Steve Falkenburg.

We're going to need to create a PlatformCertificateInfo with a single certificate on Windows to support client certificates.
Also, stop relying on the fact that the Win32 API CertDuplicateCertificateContext() currently returns the same PCCERT_CONTEXT that you pass to it, since that may change in the future.

* Shared/win/PlatformCertificateInfo.cpp:
(WebKit::PlatformCertificateInfo::PlatformCertificateInfo): Added PlatformCertificateInfo(PCCERT_CONTEXT) construtor. Use return value from CertDuplicateCertificateContext().
(WebKit::PlatformCertificateInfo::operator=): Use return value from CertDuplicateCertificateContext().
* Shared/win/PlatformCertificateInfo.h: Added PlatformCertificateInfo(PCCERT_CONTEXT) construtor.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82049 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sat, 26 Mar 2011 21:36:14 +0000 (21:36 +0000)]
2011-03-26  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Graphic corruption appeared with Silverlight contents while resizing window
        https://bugs.webkit.org/show_bug.cgi?id=57167
        <rdar://problem/9094052>

        Pass DispatchMessageEvenWhenWaitingForSyncReply when sending GeometryDidChange messages since that will guarantee
        that those messages will be handled before PaintEntirePlugin messages which prevents the PaintEntirePlugin handler from
        painting into the wrong backing store.

        * WebProcess/Plugins/PluginProxy.cpp:
        (WebKit::PluginProxy::geometryDidChange):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82048 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Sam Weinig <sam@webkit.org>
weinig@apple.com [Sat, 26 Mar 2011 21:18:07 +0000 (21:18 +0000)]
2011-03-26  Sam Weinig  <sam@webkit.org>

        Reviewed by Anders Carlsson.

        Web process considered unresponsive (SPOD shown) when displaying a JavaScript alert() as a result of a click
        <rdar://problem/9067557>
        https://bugs.webkit.org/show_bug.cgi?id=57166

        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::runJavaScriptAlert):
        (WebKit::WebPageProxy::runJavaScriptConfirm):
        (WebKit::WebPageProxy::runJavaScriptPrompt):
        Invalidate the responsiveness timer before calling out to the client.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82047 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Andreas Kling <kling@webkit.org>
commit-queue@webkit.org [Sat, 26 Mar 2011 21:17:57 +0000 (21:17 +0000)]
2011-03-26  Andreas Kling  <kling@webkit.org>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] Show page icons (favicons) in QtTestBrowser location bar.
        https://bugs.webkit.org/show_bug.cgi?id=57162

        * QtTestBrowser/QtTestBrowser.qrc:
        * QtTestBrowser/favicon.png: Added.
        * QtTestBrowser/locationedit.cpp:
        (defaultPageIcon):
        (LocationEdit::LocationEdit):
        (LocationEdit::setPageIcon):
        (LocationEdit::resizeEvent):
        (LocationEdit::updateInternalGeometry):
        * QtTestBrowser/locationedit.h:
        * QtTestBrowser/mainwindow.cpp:
        (MainWindow::buildUI):
        (MainWindow::onIconChanged):
        (MainWindow::onLoadStarted):
        * QtTestBrowser/mainwindow.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82046 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sat, 26 Mar 2011 21:15:07 +0000 (21:15 +0000)]
2011-03-26  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        ASSERTION FAILED: m_operationInProgress == NoOperation loading nytimes.com
        https://bugs.webkit.org/show_bug.cgi?id=57165
        <rdar://problem/9024311>

        The assertion fired because during GC, the web process sends a synchronous NPObjectMessageReceiver::Deallocate
        message to the plug-in process. Since this is a synchronous message, the web process needs to process incoming synchronous
        messages. While waiting, we get an incoming PluginProxy::Evaluate message from the plug-in. This causes JavaScript to run
        during GC which is very bad.

        The fix for this is to add a flag on the connection that will cause synchronous messages sent by the connection (in this case the
        plug-in process) to not be processed while the other side (the web process) is waiting for a synchronous reply _unless_ the connection
        is actually processing a synchronous message. (The last part is to avoid deadlocks).

        Since the call to NPN_Evaluate by the plug-in (that ends up sending the PluginProxy::Evaluate message) comes from a run loop timer firing,
        it's OK to wait for it to be processed by the web process when it returns to the run loop.

        * Platform/CoreIPC/Connection.cpp:
        (CoreIPC::Connection::Connection):
        Initialize m_onlySendMessagesAsDispatchWhenWaitingForSyncReplyWhenProcessingSuchAMessage and m_inDispatchMessageMarkedDispatchWhenWaitingForSyncReplyCount.

        (CoreIPC::Connection::setOnlySendMessagesAsDispatchWhenWaitingForSyncReplyWhenProcessingSuchAMessage):
        Set m_onlySendMessagesAsDispatchWhenWaitingForSyncReplyWhenProcessingSuchAMessage.

        (CoreIPC::Connection::sendMessage):
        Don't add the MessageID::DispatchMessageWhenWaitingForSyncReply flag when the right flags has been set on the connection, and it's not processing a synchronous message.

        (CoreIPC::Connection::dispatchMessage):
        Increment and decrement m_inDispatchMessageMarkedDispatchWhenWaitingForSyncReplyCount accordingly.

        * PluginProcess/WebProcessConnection.cpp:
        (WebKit::WebProcessConnection::WebProcessConnection):
        Call setOnlySendMessagesAsDispatchWhenWaitingForSyncReplyWhenProcessingSuchAMessage on the connection.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82045 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Sat, 26 Mar 2011 20:50:47 +0000 (20:50 +0000)]
2011-03-26  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Benjamin Poulain.

        [Qt] Remove GraphicsLayerTextureMapper::nativeLayer
        https://bugs.webkit.org/show_bug.cgi?id=57141

        GraphicsLayer::nativeLayer was removed in r73380.

        * platform/graphics/texmap/GraphicsLayerTextureMapper.cpp:
        * platform/graphics/texmap/GraphicsLayerTextureMapper.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82044 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRollout r82042 (If a user doesn't have a Database/LocalStorage directory, it can...
weinig@apple.com [Sat, 26 Mar 2011 20:48:58 +0000 (20:48 +0000)]
Rollout r82042 (If a user doesn't have a Database/LocalStorage directory, it can't be created (sandbox violations))

-[NSString stringByResolvingSymlinksInPath] does not do what we needed for paths within /private.
This caused all SSL sites to stop working.

* WebProcess/mac/WebProcessMac.mm:
(WebKit::appendSandboxParameterPath):
(WebKit::initializeSandbox):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82043 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Sam Weinig <sam@webkit.org>
weinig@apple.com [Sat, 26 Mar 2011 20:32:34 +0000 (20:32 +0000)]
2011-03-26  Sam Weinig  <sam@webkit.org>

        Reviewed by Anders Carlsson.

        If a user doesn't have a Database/LocalStorage directory, it can't be created (sandbox violations)
        <rdar://problem/9127411>
        https://bugs.webkit.org/show_bug.cgi?id=57164

        * WebProcess/mac/WebProcessMac.mm:
        (WebKit::appendSandboxParameterPath):
        (WebKit::initializeSandbox):
        Use -[NSString stringByResolvingSymlinksInPath] instead of realpath to ensure we can resolve
        symlinks even if the file/directory doesn't exist yet.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82042 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sat, 26 Mar 2011 20:20:46 +0000 (20:20 +0000)]
2011-03-26  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Factor processing of incoming sync replies out into processIncomingSyncReply
        https://bugs.webkit.org/show_bug.cgi?id=57161

        * Platform/CoreIPC/Connection.cpp:
        (CoreIPC::Connection::sendSyncMessage):
        Pass DispatchMessageEvenWhenWaitingForSyncReply to sendMessage.

        (CoreIPC::Connection::processIncomingSyncReply):
        Move code from processIncomingMessage to here.

        (CoreIPC::Connection::processIncomingMessage):
        Call processIncomingSyncReply.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82041 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Sat, 26 Mar 2011 20:14:04 +0000 (20:14 +0000)]
2011-03-26  Martin Robinson  <mrobinson@igalia.com>

        [GTK] layoutTestController.overridePreference("WebKitDefaultFontSize"...) does not take into account screen DPI
        https://bugs.webkit.org/show_bug.cgi?id=57160

        Skip failing test. This bug was uncovered by r82033.

        * platform/gtk/Skipped:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82040 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Sat, 26 Mar 2011 20:07:08 +0000 (20:07 +0000)]
2011-03-26  Martin Robinson  <mrobinson@igalia.com>

        Remove a test from the GTK+ skipped list that no longer exists.

        * platform/gtk/Skipped: Update the skipped list.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82039 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Sat, 26 Mar 2011 20:04:51 +0000 (20:04 +0000)]
2011-03-26  Martin Robinson  <mrobinson@igalia.com>

        Rebaseline a few more GTK+ tests after r82033.

        * platform/gtk/fast/forms/input-appearance-spinbutton-disabled-readonly-expected.checksum:
        * platform/gtk/fast/forms/input-appearance-spinbutton-disabled-readonly-expected.png:
        * platform/gtk/fast/forms/input-appearance-spinbutton-disabled-readonly-expected.txt:
        * platform/gtk/fast/forms/input-appearance-spinbutton-layer-expected.checksum:
        * platform/gtk/fast/forms/input-appearance-spinbutton-layer-expected.png:
        * platform/gtk/fast/forms/input-appearance-spinbutton-layer-expected.txt:
        * platform/gtk/fast/forms/input-appearance-spinbutton-visibility-expected.checksum:
        * platform/gtk/fast/forms/input-appearance-spinbutton-visibility-expected.png:
        * platform/gtk/fast/forms/input-appearance-spinbutton-visibility-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82038 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 26 Mar 2011 19:55:02 +0000 (19:55 +0000)]
2011-03-26  Adam Barth  <abarth@webkit.org>

        Remove the JavaScriptGlue GYP build.  We're not going to use GYP for
        JavaScriptGlue because JavaScriptGlue is used only by one port.

        * gyp: Removed.
        * gyp/JavaScriptGlue.gyp: Removed.
        * gyp/JavaScriptGlue.gypi: Removed.
        * gyp/remove-headers-if-needed.sh: Removed.
        * gyp/run-if-exists.sh: Removed.
        * gyp/update-info-plist.sh: Removed.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82037 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Sam Weinig <sam@webkit.org>
weinig@apple.com [Sat, 26 Mar 2011 19:37:10 +0000 (19:37 +0000)]
2011-03-26  Sam Weinig  <sam@webkit.org>

        Reviewed by Anders Carlsson.

        WebKit2: Assert in CoreIPC::typeFromCFTypeRef() when loading a site with an invalid certificate
        <rdar://problem/9188041>
        https://bugs.webkit.org/show_bug.cgi?id=57159

        * Shared/cf/ArgumentCodersCF.cpp:
        (CoreIPC::typeFromCFTypeRef):
        (CoreIPC::encode):
        (CoreIPC::decode):
        * Shared/cf/ArgumentCodersCF.h:
        Add encoding/decoding for SecCertificateRef type.

        * Shared/mac/PlatformCertificateInfo.mm:
        (WebKit::PlatformCertificateInfo::encode):
        (WebKit::PlatformCertificateInfo::decode):
        Use ArgumentCodersCF now that it knows how to encode/decode SecCertificateRefs.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82036 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sat, 26 Mar 2011 19:34:45 +0000 (19:34 +0000)]
2011-03-26  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Handle synchronous replies coming in out of order
        https://bugs.webkit.org/show_bug.cgi?id=57158

        When processing an incoming reply, don't assume that it belongs to the last sent synchronous request.
        Instead, iterate over the m_pendingSyncReplies vector backwards looking for the corresponding request.

        * Platform/CoreIPC/Connection.cpp:
        (CoreIPC::Connection::processIncomingMessage):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82035 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Andreas Kling <kling@webkit.org>
andreas.kling@nokia.com [Sat, 26 Mar 2011 19:28:19 +0000 (19:28 +0000)]
2011-03-26  Andreas Kling  <kling@webkit.org>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] QWebFrame::iconChanged() not emitted when icon is cached but not yet loaded
        https://bugs.webkit.org/show_bug.cgi?id=57157

        Add an IconDatabaseClient for the Qt port to ensure that QWebFrame::iconChanged()
        is always emitted when appropriate.

        * QtWebKit.pro: Add new files.

        * WebCoreSupport/IconDatabaseClientQt.h: Added.
        * WebCoreSupport/IconDatabaseClientQt.cpp: Added.
        (WebCore::IconDatabaseClientQt::instance):
        (WebCore::IconDatabaseClientQt::IconDatabaseClientQt):
        (WebCore::IconDatabaseClientQt::~IconDatabaseClientQt):
        (WebCore::IconDatabaseClientQt::performImport):
        (WebCore::IconDatabaseClientQt::didRemoveAllIcons):
        (WebCore::IconDatabaseClientQt::didImportIconURLForPageURL):
        (WebCore::IconDatabaseClientQt::didImportIconDataForPageURL):
        (WebCore::IconDatabaseClientQt::didChangeIconForPageURL):
        (WebCore::IconDatabaseClientQt::didFinishURLImport):

        * WebCoreSupport/FrameLoaderClientQt.h:
        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (WebCore::FrameLoaderClientQt::registerForIconNotification):
        (WebCore::FrameLoaderClientQt::onIconLoadedForPageURL): New slot connected
        to the IconDatabaseClientQt::iconLoadedForPageURL() signal. This emits the
        QWebFrame::iconChanged() signal when the IconDatabases finishes loading
        a cached favicon for the frame's URL.

        * Api/qwebsettings.cpp:
        (QWebSettings::setIconDatabasePath): Make sure that IconDatabaseClientQt is
        instantiated. An IconDatabaseClient has to be registered before the IconDatabase
        spawns its reader thread.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82034 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Sat, 26 Mar 2011 18:52:09 +0000 (18:52 +0000)]
2011-03-26  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Force DumpRenderTree to use 96 DPI
        https://bugs.webkit.org/show_bug.cgi?id=56866

        Update GTK+ test results which now have differ.

        * platform/gtk/editing/pasteboard/4641033-expected.checksum:
        * platform/gtk/editing/pasteboard/4641033-expected.png:
        * platform/gtk/editing/pasteboard/4641033-expected.txt:
        * platform/gtk/editing/pasteboard/4944770-1-expected.checksum:
        * platform/gtk/editing/pasteboard/4944770-1-expected.png:
        * platform/gtk/editing/pasteboard/4944770-1-expected.txt:
        * platform/gtk/editing/pasteboard/4944770-2-expected.checksum:
        * platform/gtk/editing/pasteboard/4944770-2-expected.png:
        * platform/gtk/editing/pasteboard/4944770-2-expected.txt:
        * platform/gtk/editing/selection/caret-before-select-expected.checksum:
        * platform/gtk/editing/selection/caret-before-select-expected.png:
        * platform/gtk/editing/selection/caret-before-select-expected.txt:
        * platform/gtk/editing/selection/replaced-boundaries-3-expected.checksum:
        * platform/gtk/editing/selection/replaced-boundaries-3-expected.png:
        * platform/gtk/editing/selection/replaced-boundaries-3-expected.txt:
        * platform/gtk/editing/selection/select-box-expected.checksum:
        * platform/gtk/editing/selection/select-box-expected.png:
        * platform/gtk/editing/selection/select-box-expected.txt:
        * platform/gtk/editing/selection/select-element-paragraph-boundary-expected.checksum:
        * platform/gtk/editing/selection/select-element-paragraph-boundary-expected.png:
        * platform/gtk/editing/selection/select-element-paragraph-boundary-expected.txt:
        * platform/gtk/fast/block/float/float-avoidance-expected.checksum:
        * platform/gtk/fast/block/float/float-avoidance-expected.png:
        * platform/gtk/fast/block/margin-collapse/103-expected.checksum:
        * platform/gtk/fast/block/margin-collapse/103-expected.png:
        * platform/gtk/fast/forms/001-expected.checksum:
        * platform/gtk/fast/forms/001-expected.png:
        * platform/gtk/fast/forms/001-expected.txt:
        * platform/gtk/fast/forms/003-expected.checksum:
        * platform/gtk/fast/forms/003-expected.png:
        * platform/gtk/fast/forms/003-expected.txt:
        * platform/gtk/fast/forms/004-expected.checksum:
        * platform/gtk/fast/forms/004-expected.png:
        * platform/gtk/fast/forms/004-expected.txt:
        * platform/gtk/fast/forms/HTMLOptionElement_label01-expected.checksum:
        * platform/gtk/fast/forms/HTMLOptionElement_label01-expected.png:
        * platform/gtk/fast/forms/HTMLOptionElement_label01-expected.txt:
        * platform/gtk/fast/forms/HTMLOptionElement_label02-expected.checksum:
        * platform/gtk/fast/forms/HTMLOptionElement_label02-expected.png:
        * platform/gtk/fast/forms/HTMLOptionElement_label02-expected.txt:
        * platform/gtk/fast/forms/HTMLOptionElement_label03-expected.checksum:
        * platform/gtk/fast/forms/HTMLOptionElement_label03-expected.png:
        * platform/gtk/fast/forms/HTMLOptionElement_label03-expected.txt:
        * platform/gtk/fast/forms/HTMLOptionElement_label04-expected.checksum:
        * platform/gtk/fast/forms/HTMLOptionElement_label04-expected.png:
        * platform/gtk/fast/forms/HTMLOptionElement_label04-expected.txt:
        * platform/gtk/fast/forms/HTMLOptionElement_label06-expected.checksum:
        * platform/gtk/fast/forms/HTMLOptionElement_label06-expected.png:
        * platform/gtk/fast/forms/HTMLOptionElement_label06-expected.txt:
        * platform/gtk/fast/forms/HTMLOptionElement_label07-expected.checksum:
        * platform/gtk/fast/forms/HTMLOptionElement_label07-expected.png:
        * platform/gtk/fast/forms/HTMLOptionElement_label07-expected.txt:
        * platform/gtk/fast/forms/basic-selects-expected.checksum:
        * platform/gtk/fast/forms/basic-selects-expected.png:
        * platform/gtk/fast/forms/basic-selects-expected.txt:
        * platform/gtk/fast/forms/control-clip-overflow-expected.checksum:
        * platform/gtk/fast/forms/control-clip-overflow-expected.png:
        * platform/gtk/fast/forms/control-restrict-line-height-expected.checksum:
        * platform/gtk/fast/forms/control-restrict-line-height-expected.png:
        * platform/gtk/fast/forms/control-restrict-line-height-expected.txt:
        * platform/gtk/fast/forms/disabled-select-change-index-expected.checksum:
        * platform/gtk/fast/forms/disabled-select-change-index-expected.png:
        * platform/gtk/fast/forms/disabled-select-change-index-expected.txt:
        * platform/gtk/fast/forms/form-element-geometry-expected.checksum:
        * platform/gtk/fast/forms/form-element-geometry-expected.png:
        * platform/gtk/fast/forms/form-element-geometry-expected.txt:
        * platform/gtk/fast/forms/menulist-deselect-update-expected.checksum:
        * platform/gtk/fast/forms/menulist-deselect-update-expected.png:
        * platform/gtk/fast/forms/menulist-deselect-update-expected.txt:
        * platform/gtk/fast/forms/menulist-narrow-width-expected.checksum:
        * platform/gtk/fast/forms/menulist-narrow-width-expected.png:
        * platform/gtk/fast/forms/menulist-narrow-width-expected.txt:
        * platform/gtk/fast/forms/menulist-no-overflow-expected.checksum:
        * platform/gtk/fast/forms/menulist-no-overflow-expected.png:
        * platform/gtk/fast/forms/menulist-no-overflow-expected.txt:
        * platform/gtk/fast/forms/menulist-restrict-line-height-expected.checksum:
        * platform/gtk/fast/forms/menulist-restrict-line-height-expected.png:
        * platform/gtk/fast/forms/menulist-restrict-line-height-expected.txt:
        * platform/gtk/fast/forms/menulist-separator-painting-expected.checksum:
        * platform/gtk/fast/forms/menulist-separator-painting-expected.png:
        * platform/gtk/fast/forms/menulist-separator-painting-expected.txt:
        * platform/gtk/fast/forms/menulist-style-color-expected.checksum:
        * platform/gtk/fast/forms/menulist-style-color-expected.png:
        * platform/gtk/fast/forms/menulist-style-color-expected.txt:
        * platform/gtk/fast/forms/menulist-width-change-expected.checksum:
        * platform/gtk/fast/forms/menulist-width-change-expected.png:
        * platform/gtk/fast/forms/menulist-width-change-expected.txt:
        * platform/gtk/fast/forms/option-script-expected.checksum:
        * platform/gtk/fast/forms/option-script-expected.png:
        * platform/gtk/fast/forms/option-script-expected.txt:
        * platform/gtk/fast/forms/option-strip-whitespace-expected.checksum:
        * platform/gtk/fast/forms/option-strip-whitespace-expected.png:
        * platform/gtk/fast/forms/option-strip-whitespace-expected.txt:
        * platform/gtk/fast/forms/option-text-clip-expected.checksum:
        * platform/gtk/fast/forms/option-text-clip-expected.png:
        * platform/gtk/fast/forms/select-align-expected.checksum:
        * platform/gtk/fast/forms/select-align-expected.png:
        * platform/gtk/fast/forms/select-baseline-expected.checksum:
        * platform/gtk/fast/forms/select-baseline-expected.png:
        * platform/gtk/fast/forms/select-baseline-expected.txt:
        * platform/gtk/fast/forms/select-change-listbox-to-popup-expected.checksum:
        * platform/gtk/fast/forms/select-change-listbox-to-popup-expected.png:
        * platform/gtk/fast/forms/select-change-listbox-to-popup-expected.txt:
        * platform/gtk/fast/forms/select-dirty-parent-pref-widths-expected.checksum:
        * platform/gtk/fast/forms/select-dirty-parent-pref-widths-expected.png:
        * platform/gtk/fast/forms/select-dirty-parent-pref-widths-expected.txt:
        * platform/gtk/fast/forms/select-disabled-appearance-expected.checksum:
        * platform/gtk/fast/forms/select-disabled-appearance-expected.png:
        * platform/gtk/fast/forms/select-disabled-appearance-expected.txt:
        * platform/gtk/fast/forms/select-initial-position-expected.checksum:
        * platform/gtk/fast/forms/select-initial-position-expected.png:
        * platform/gtk/fast/forms/select-initial-position-expected.txt:
        * platform/gtk/fast/forms/select-selected-expected.checksum:
        * platform/gtk/fast/forms/select-selected-expected.png:
        * platform/gtk/fast/forms/select-selected-expected.txt:
        * platform/gtk/fast/forms/select-size-expected.checksum:
        * platform/gtk/fast/forms/select-size-expected.png:
        * platform/gtk/fast/forms/select-size-expected.txt:
        * platform/gtk/fast/forms/select-style-expected.checksum:
        * platform/gtk/fast/forms/select-style-expected.png:
        * platform/gtk/fast/forms/select-style-expected.txt:
        * platform/gtk/fast/forms/select-visual-hebrew-expected.checksum:
        * platform/gtk/fast/forms/select-visual-hebrew-expected.png:
        * platform/gtk/fast/forms/select-visual-hebrew-expected.txt:
        * platform/gtk/fast/forms/select-writing-direction-natural-expected.checksum:
        * platform/gtk/fast/forms/select-writing-direction-natural-expected.png:
        * platform/gtk/fast/forms/select-writing-direction-natural-expected.txt:
        * platform/gtk/fast/forms/stuff-on-my-optgroup-expected.checksum:
        * platform/gtk/fast/forms/stuff-on-my-optgroup-expected.png:
        * platform/gtk/fast/forms/stuff-on-my-optgroup-expected.txt:
        * platform/gtk/fast/html/keygen-expected.checksum:
        * platform/gtk/fast/html/keygen-expected.png:
        * platform/gtk/fast/html/keygen-expected.txt:
        * platform/gtk/fast/invalid/014-expected.checksum:
        * platform/gtk/fast/invalid/014-expected.png:
        * platform/gtk/fast/invalid/014-expected.txt:
        * platform/gtk/fast/invalid/residual-style-expected.txt:
        * platform/gtk/fast/parser/document-write-option-expected.checksum:
        * platform/gtk/fast/parser/document-write-option-expected.png:
        * platform/gtk/fast/parser/document-write-option-expected.txt:
        * platform/gtk/fast/replaced/replaced-breaking-expected.checksum:
        * platform/gtk/fast/replaced/replaced-breaking-expected.png:
        * platform/gtk/fast/replaced/replaced-breaking-expected.txt:
        * platform/gtk/fast/replaced/replaced-breaking-mixture-expected.checksum:
        * platform/gtk/fast/replaced/replaced-breaking-mixture-expected.png:
        * platform/gtk/fast/replaced/replaced-breaking-mixture-expected.txt:
        * platform/gtk/fast/replaced/three-selects-break-expected.checksum:
        * platform/gtk/fast/replaced/three-selects-break-expected.png:
        * platform/gtk/fast/replaced/three-selects-break-expected.txt:
        * platform/gtk/fast/replaced/width100percent-menulist-expected.checksum:
        * platform/gtk/fast/replaced/width100percent-menulist-expected.png:
        * platform/gtk/fast/replaced/width100percent-menulist-expected.txt:
        * platform/gtk/fast/text/international/bidi-menulist-expected.checksum:
        * platform/gtk/fast/text/international/bidi-menulist-expected.png:
        * platform/gtk/fast/text/international/bidi-menulist-expected.txt:
        * platform/gtk/fast/text/international/pop-up-button-text-alignment-and-direction-expected.checksum:
        * platform/gtk/fast/text/international/pop-up-button-text-alignment-and-direction-expected.png:
        * platform/gtk/fast/text/international/pop-up-button-text-alignment-and-direction-expected.txt:
        * platform/gtk/http/tests/navigation/javascriptlink-frames-expected.txt:
        * platform/gtk/tables/mozilla/bugs/bug1188-expected.checksum:
        * platform/gtk/tables/mozilla/bugs/bug1188-expected.png:
        * platform/gtk/tables/mozilla/bugs/bug1188-expected.txt:
        * platform/gtk/tables/mozilla/bugs/bug18359-expected.checksum:
        * platform/gtk/tables/mozilla/bugs/bug18359-expected.png:
        * platform/gtk/tables/mozilla/bugs/bug18359-expected.txt:
        * platform/gtk/tables/mozilla/bugs/bug2479-3-expected.checksum:
        * platform/gtk/tables/mozilla/bugs/bug2479-3-expected.png:
        * platform/gtk/tables/mozilla/bugs/bug2479-3-expected.txt:
        * platform/gtk/tables/mozilla/bugs/bug2479-4-expected.txt:
        * platform/gtk/tables/mozilla/bugs/bug29326-expected.checksum:
        * platform/gtk/tables/mozilla/bugs/bug29326-expected.png:
        * platform/gtk/tables/mozilla/bugs/bug29326-expected.txt:
        * platform/gtk/tables/mozilla/bugs/bug33855-expected.checksum:
        * platform/gtk/tables/mozilla/bugs/bug33855-expected.png:
        * platform/gtk/tables/mozilla/bugs/bug33855-expected.txt:
        * platform/gtk/tables/mozilla/bugs/bug4382-expected.checksum:
        * platform/gtk/tables/mozilla/bugs/bug4382-expected.png:
        * platform/gtk/tables/mozilla/bugs/bug4382-expected.txt:
        * platform/gtk/tables/mozilla/bugs/bug96334-expected.checksum:
        * platform/gtk/tables/mozilla/bugs/bug96334-expected.png:
        * platform/gtk/tables/mozilla/bugs/bug96334-expected.txt:
        * platform/gtk/tables/mozilla/core/margins-expected.checksum:
        * platform/gtk/tables/mozilla/core/margins-expected.png:
        * platform/gtk/tables/mozilla/core/margins-expected.txt:
        * platform/gtk/tables/mozilla/dom/tableDom-expected.checksum:
        * platform/gtk/tables/mozilla/dom/tableDom-expected.png:
        * platform/gtk/tables/mozilla/dom/tableDom-expected.txt:
        * platform/gtk/tables/mozilla_expected_failures/bugs/bug2479-5-expected.checksum:
        * platform/gtk/tables/mozilla_expected_failures/bugs/bug2479-5-expected.png:
        * platform/gtk/tables/mozilla_expected_failures/bugs/bug2479-5-expected.txt:
        * platform/gtk/transforms/2d/zoom-menulist-expected.checksum:
        * platform/gtk/transforms/2d/zoom-menulist-expected.png:
        * platform/gtk/transforms/2d/zoom-menulist-expected.txt:
2011-03-23  Martin Robinson  <mrobinson@igalia.com>

        Reviewed by Xan Lopez.

        [GTK] Force DumpRenderTree to use 96 DPI
        https://bugs.webkit.org/show_bug.cgi?id=56866

        Make sure that GTK+ is set to use 96 DPI when running tests. This ensures
        broader compatibility with WebKit2 and makes setting font sizes more straightforward.

        * DumpRenderTree/gtk/DumpRenderTree.cpp:
        (initializeGtkFontSettings): Set the screen resolution and GTK+ xft property to 96 DPI.
        (resetDefaultsToConsistentValues):Remove the call which repeatedly set the GDK screen resolution.
        (setDefaultsToConsistentStateValuesForTesting): Update the font sizes to reflect the new DPI.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82033 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sat, 26 Mar 2011 18:23:53 +0000 (18:23 +0000)]
2011-03-26  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        Fix a possible deadlock when two synchronous messages are sent at the same time
        https://bugs.webkit.org/show_bug.cgi?id=57155

        Simplify code and fix a possible (although highly improbable) dead lock.

        * Platform/CoreIPC/Connection.cpp:
        Make SyncMessageState atomically ref counted since it can be ref()'ed from the connection queue.
        Get rid of m_waitForSyncReplyCount and add m_didScheduleDispatchMessagesWork.

        (CoreIPC::Connection::SyncMessageState::SyncMessageState):
        Initialize m_didScheduleDispatchMessagesWork to false.

        (CoreIPC::Connection::SyncMessageState::processIncomingMessage):
        if m_didScheduleDispatchMessagesWork is false, schedule a call to dispatchMessageAndResetDidScheduleDispatchMessagesWork
        on the client run loop.

        (CoreIPC::Connection::SyncMessageState::dispatchMessageAndResetDidScheduleDispatchMessagesWork):
        Dispatch messages and set m_didScheduleDispatchMessagesWork back to false.

        (CoreIPC::Connection::sendSyncMessage):
        Remove calls to beginWaitForSyncReply and endWaitForSyncReply.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82032 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Sat, 26 Mar 2011 18:15:04 +0000 (18:15 +0000)]
2011-03-26  Martin Robinson  <mrobinson@igalia.com>

        Rebaseline many GTK+ tests after r81994.

        * platform/gtk/fast/block/positioning/056-expected.txt:
        * platform/gtk/fast/block/positioning/auto/005-expected.checksum:
        * platform/gtk/fast/block/positioning/auto/005-expected.png:
        * platform/gtk/fast/block/positioning/auto/005-expected.txt:
        * platform/gtk/fast/block/positioning/auto/006-expected.checksum:
        * platform/gtk/fast/block/positioning/auto/006-expected.png:
        * platform/gtk/fast/block/positioning/auto/006-expected.txt:
        * platform/gtk/fast/block/positioning/auto/vertical-lr/005-expected.checksum:
        * platform/gtk/fast/block/positioning/auto/vertical-lr/005-expected.png:
        * platform/gtk/fast/block/positioning/auto/vertical-lr/005-expected.txt:
        * platform/gtk/fast/block/positioning/auto/vertical-lr/006-expected.checksum:
        * platform/gtk/fast/block/positioning/auto/vertical-lr/006-expected.png:
        * platform/gtk/fast/block/positioning/auto/vertical-lr/006-expected.txt:
        * platform/gtk/fast/block/positioning/auto/vertical-rl/005-expected.checksum:
        * platform/gtk/fast/block/positioning/auto/vertical-rl/005-expected.png:
        * platform/gtk/fast/block/positioning/auto/vertical-rl/005-expected.txt:
        * platform/gtk/fast/block/positioning/auto/vertical-rl/006-expected.checksum:
        * platform/gtk/fast/block/positioning/auto/vertical-rl/006-expected.png:
        * platform/gtk/fast/block/positioning/auto/vertical-rl/006-expected.txt:
        * platform/gtk/fast/block/positioning/fixed-positioning-scrollbar-bug-expected.txt:
        * platform/gtk/fast/block/positioning/inline-block-relposition-expected.checksum:
        * platform/gtk/fast/block/positioning/inline-block-relposition-expected.png:
        * platform/gtk/fast/block/positioning/inline-block-relposition-expected.txt:
        * platform/gtk/fast/block/positioning/relative-overflow-block-expected.txt:
        * platform/gtk/fast/clip/001-expected.txt:
        * platform/gtk/fast/clip/004-expected.txt:
        * platform/gtk/fast/clip/006-expected.txt:
        * platform/gtk/fast/clip/007-expected.txt:
        * platform/gtk/fast/clip/008-expected.txt:
        * platform/gtk/fast/clip/009-expected.txt:
        * platform/gtk/fast/clip/010-expected.txt:
        * platform/gtk/fast/clip/011-expected.txt:
        * platform/gtk/fast/clip/012-expected.txt:
        * platform/gtk/fast/clip/013-expected.txt:
        * platform/gtk/fast/clip/014-expected.txt:
        * platform/gtk/fast/clip/nestedTransparencyClip-expected.txt:
        * platform/gtk/fast/clip/outline-overflowClip-expected.txt:
        * platform/gtk/fast/forms/input-appearance-preventDefault-expected.txt:
        * platform/gtk/fast/invalid/014-expected.txt:
        * platform/gtk/fast/layers/layer-visibility-expected.txt:
        * platform/gtk/fast/repaint/layout-state-scrolloffset-expected.txt:
        * platform/gtk/fast/repaint/layout-state-scrolloffset2-expected.txt:
        * platform/gtk/fast/repaint/layout-state-scrolloffset3-expected.txt:
        * platform/gtk/tables/mozilla/bugs/bug51140-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82031 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 26 Mar 2011 14:38:43 +0000 (14:38 +0000)]
2011-03-26  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        We shouldn't bother running GYP if the generated files are newer than the GYP files
        https://bugs.webkit.org/show_bug.cgi?id=57146

        In the common case, this check will avoid any overhead from processing
        the GYP files.  Another approach to doing this is to add the feature to
        GYP directly, but GYP's approach to this problem is to compute the
        output in its entirety and compare it byte-for-byte against the output
        file.  In the future, it might make sense to add this approach as an
        alternative approach for GYP itself.

        I also removed JavaScriptGlue from the script because we're not really
        going to change JavaScriptGlue over to GYP.  We were using
        JavaScriptGlue as a learning experience.

        * Source/gyp/configure:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82030 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUpdate Qt specific expected results after r81992.
ossy@webkit.org [Sat, 26 Mar 2011 14:25:02 +0000 (14:25 +0000)]
Update Qt specific expected results after r81992.

* platform/qt/fast/repaint/layout-state-scrolloffset-expected.checksum:
* platform/qt/fast/repaint/layout-state-scrolloffset-expected.png:
* platform/qt/fast/repaint/layout-state-scrolloffset-expected.txt:
* platform/qt/fast/repaint/layout-state-scrolloffset2-expected.checksum:
* platform/qt/fast/repaint/layout-state-scrolloffset2-expected.png:
* platform/qt/fast/repaint/layout-state-scrolloffset2-expected.txt:
* platform/qt/fast/repaint/layout-state-scrolloffset3-expected.checksum:
* platform/qt/fast/repaint/layout-state-scrolloffset3-expected.png:
* platform/qt/fast/repaint/layout-state-scrolloffset3-expected.txt:
* platform/qt/tables/mozilla/bugs/bug51140-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82029 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Sat, 26 Mar 2011 11:59:34 +0000 (11:59 +0000)]
2011-03-26  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Teach Content Security Policy how to parse source-list
        https://bugs.webkit.org/show_bug.cgi?id=54799

        Test a variety of source-list parsing cases.  There's a bunch more
        cases we could be testing.  We'll add more over time.

        * http/tests/security/contentSecurityPolicy/source-list-parsing-expected.txt: Added.
        * http/tests/security/contentSecurityPolicy/source-list-parsing.html: Added.
2011-03-26  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        Teach Content Security Policy how to parse source-list
        https://bugs.webkit.org/show_bug.cgi?id=54799

        This patch is larger than I would like, but I wasn't sure how to make
        it any smaller while still being reasonably testable.  I've left out
        some features (such as host wildcarding and 'self') so I can add them
        in later patches with tests.

        Test: http/tests/security/contentSecurityPolicy/source-list-parsing.html

        * bindings/ScriptControllerBase.cpp:
        * dom/ScriptElement.cpp:
        * html/parser/HTMLDocumentParser.cpp:
        * loader/FrameLoader.cpp:
            - Add include explicitly now that we're not spamming the include
              everywhere.
        * dom/Document.cpp:
        (WebCore::Document::initSecurityContext):
            - We need to pass the SecurityOrigin object to
              ContentSecurityPolicy so that it can resolve implicit parts of
              source patterns, such as the scheme.
        * dom/Document.h:
            - Forward declare ContentSecurityPolicy rather than including the
              header.  Technically this could be a separate change, but I was
              getting annoyed at the world re-builds.
        * page/ContentSecurityPolicy.cpp:
        (WebCore::skipExactly):
        (WebCore::skipUtil):
        (WebCore::skipWhile):
            - Clean up these parser helper functions.  We might consider moving
              them to a more general location.  They're very helpful for
              writing secure HTTP header parsers.
        (WebCore::CSPSource::CSPSource):
            - New class to represent one source in a source-list.
        (WebCore::CSPSource::matches):
        (WebCore::CSPSource::schemeMatches):
        (WebCore::CSPSource::hostMatches):
        (WebCore::CSPSource::portMatches):
        (WebCore::CSPSource::isSchemeOnly):
            - Currently we represent scheme-only sources using with an empty
              m_host.  Another approach I considered was using another bool,
              but that seemed slighly messier.
        (WebCore::CSPSourceList::CSPSourceList):
            - CSPSourceList doesn't need to ref SecurityOrigin because
              CSPSourceList is owned by ContentSecurityPolicy, which holds a
              ref.
        (WebCore::CSPSourceList::parse):
        (WebCore::CSPSourceList::matches):
        (WebCore::CSPSourceList::parseSource):
        (WebCore::CSPSourceList::parseScheme):
        (WebCore::CSPSourceList::parseHost):
        (WebCore::CSPSourceList::parsePort):
            - A basic "segment and recurse" parser.  This parser causes us to
              take more branches than we need, but I don't think we need to
              squeeze every last ouch of performance out of this parser.  This
              approach is more simple than some of the other approaches I
              tried.
        (WebCore::CSPSourceList::addSourceSelf):
        (WebCore::CSPDirective::CSPDirective):
        (WebCore::CSPDirective::allows):
        (WebCore::ContentSecurityPolicy::ContentSecurityPolicy):
        (WebCore::ContentSecurityPolicy::parse):
        (WebCore::ContentSecurityPolicy::parseDirective):
        (WebCore::ContentSecurityPolicy::addDirective):
            - I couldn't resist re-writing this parser to use the helper
              functions and to match the style of the source-list parser.
        * page/ContentSecurityPolicy.h:
        (WebCore::ContentSecurityPolicy::create):
            - Accept a SecurityOrigin context object.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82028 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Sat, 26 Mar 2011 10:03:22 +0000 (10:03 +0000)]
2011-03-26  Patrick Gansterer  <paroga@webkit.org>

        Unreviewed EFL and WinCE build fix for r81977.

        * platform/graphics/FontPlatformData.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82027 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoAdded platform-specific expected results for a new test.
mitz@apple.com [Sat, 26 Mar 2011 07:49:28 +0000 (07:49 +0000)]
Added platform-specific expected results for a new test.

* platform/mac-snowleopard/fast/text/justify-ideograph-leading-expansion-expected.txt: Added.
* platform/win/fast/text/justify-ideograph-leading-expansion-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82026 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Sat, 26 Mar 2011 07:22:44 +0000 (07:22 +0000)]
2011-03-26  Patrick Gansterer  <paroga@webkit.org>

        Unreviewed WinCE build fix for r82000.

        * platform/wince/FileSystemWinCE.cpp:
        (WebCore::openFile): Added missing function.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82025 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-26 Patrick Gansterer <paroga@webkit.org>
paroga@webkit.org [Sat, 26 Mar 2011 07:07:50 +0000 (07:07 +0000)]
2011-03-26  Patrick Gansterer  <paroga@webkit.org>

        Unreviewed WinCE build fix.

        * CMakeListsWinCE.txt: Added missing include directory.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82024 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Vincent Scheib <scheib@chromium.org>
scheib@chromium.org [Sat, 26 Mar 2011 03:06:55 +0000 (03:06 +0000)]
2011-03-25  Vincent Scheib  <scheib@chromium.org>

        Chromium test_expectations: bug added for tests impacted by r81992

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82021 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoplugins/embed-prefers-plugins-for-images.html only passing on SnowLeopard.
aestes@apple.com [Sat, 26 Mar 2011 01:42:17 +0000 (01:42 +0000)]
plugins/embed-prefers-plugins-for-images.html only passing on SnowLeopard.
https://bugs.webkit.org/show_bug.cgi?id=57137

Check in failing results for this test (except for on SnowLeopard) until
the failure can be diagnosed.

* platform/mac-snowleopard/plugins: Added.
* platform/mac-snowleopard/plugins/embed-prefers-plugins-for-images-expected.txt: Added.
* plugins/embed-prefers-plugins-for-images-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82020 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Sam Weinig <sam@webkit.org>
weinig@apple.com [Sat, 26 Mar 2011 01:22:39 +0000 (01:22 +0000)]
2011-03-25  Sam Weinig  <sam@webkit.org>

        Reviewed by Adele Peterson.

        WKPageLoadAlternateHTMLString doesn't re-spawn a dead WebProcess, but should
        <rdar://problem/9191493>
        https://bugs.webkit.org/show_bug.cgi?id=57134

        Make all load functions respawn a dead WebProcess for consistency.

        * UIProcess/WebPageProxy.cpp:
        (WebKit::WebPageProxy::loadHTMLString):
        Respawn a dead WebProcess.

        (WebKit::WebPageProxy::loadAlternateHTMLString):
        Respawn a dead WebProcess.

        (WebKit::WebPageProxy::loadPlainTextString):
        Respawn a dead WebProcess.

        (WebKit::WebPageProxy::didStartProvisionalLoadForFrame):
        * UIProcess/WebPageProxy.h:
        * UIProcess/WebPageProxy.messages.in:
        * WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
        (WebKit::WebFrameLoaderClient::dispatchDidStartProvisionalLoad):
        * WebProcess/WebPage/WebPage.cpp:
        (WebKit::WebPage::loadAlternateHTMLString):
        Since we may no longer always have a main frame when calling WKPageLoadAlternateHTMLString,
        set the unreachable URL, if there is one, on provisional load.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82019 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[wx] Build fix, don't use the new FPD implementation yet, until we can merge ours...
kevino@webkit.org [Sat, 26 Mar 2011 01:21:30 +0000 (01:21 +0000)]
[wx] Build fix, don't use the new FPD implementation yet, until we can merge ours with it.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82018 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoLayoutTestController.setPluginsEnabled() isn't implemented in WebKit2,
aestes@apple.com [Sat, 26 Mar 2011 01:17:31 +0000 (01:17 +0000)]
LayoutTestController.setPluginsEnabled() isn't implemented in WebKit2,
and that is now required for fast/images/embed-image.html.

* platform/mac-wk2/Skipped: Skip fast/images/embed-image.html

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82017 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSkip fast/images/embed-image.html on Windows XP
aestes@apple.com [Sat, 26 Mar 2011 01:13:37 +0000 (01:13 +0000)]
Skip fast/images/embed-image.html on Windows XP
https://bugs.webkit.org/show_bug.cgi?id=57135

It would appear that the Windows XP bot is loading a plug-in despite
the test calling layoutTestController.setPluginsEnabled(false).

* platform/win-xp/Skipped: Skip fast/images/embed-image.html

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82016 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoRe-work a test to not be flaky.
aestes@apple.com [Sat, 26 Mar 2011 00:52:29 +0000 (00:52 +0000)]
Re-work a test to not be flaky.

* plugins/embed-prefers-plugins-for-images.html:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82015 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Jer Noble <jer.noble@apple.com>
jer.noble@apple.com [Sat, 26 Mar 2011 00:51:00 +0000 (00:51 +0000)]
2011-03-25  Jer Noble  <jer.noble@apple.com>

        Reviewed by Eric Carlson.

        MediaPlayerPrivateQuickTimeVisualContext should use the Application Cache during load.
        https://bugs.webkit.org/show_bug.cgi?id=57047

        No new tests.

        When loading a URL, checkk osee if the Appplication Cache has a version of that URL
        stored; if so, use the local path to that cached media instead of the remote URL.

        * platform/graphics/win/MediaPlayerPrivateQuickTimeVisualContext.cpp:
        (WebCore::MediaPlayerPrivateQuickTimeVisualContext::loadInternal):
        * platform/graphics/win/QTMovie.cpp:
        (QTMovie::loadPath):
        * platform/graphics/win/QTMovie.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82014 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<rdar://problem/9134330> Missing expansion before ideograph at the beginning or end...
mitz@apple.com [Sat, 26 Mar 2011 00:46:26 +0000 (00:46 +0000)]
<rdar://problem/9134330> Missing expansion before ideograph at the beginning or end of a text run
https://bugs.webkit.org/show_bug.cgi?id=57106

Reviewed by Darin Adler.

Source/WebCore:

Test: fast/text/justify-ideograph-leading-expansion.html

* platform/graphics/TextRun.h:
Replaced TrailingExpansionBehavior enum with ExpansionBehavior flags.
(WebCore::TextRun::TextRun): Constructors now take an expansionBehavior parameter.
(WebCore::TextRun::allowsLeadingExpansion): Added this accessor.
(WebCore::TextRun::allowsTrailingExpansion): Changed to use the m_expansionBehavior member.
* platform/graphics/WidthIterator.cpp:
(WebCore::WidthIterator::WidthIterator): Initialize m_isAfterExpansion from the TextRun, allowing
leading expansion when appropriate.
(WebCore::WidthIterator::advance): Moved the last-glyph-in-run check to only apply to expansion
after the glyph, not expansion before the glyph, since that is not trailing expansion. Added code
to handle expansion before the first glyph.
* platform/graphics/mac/ComplexTextController.cpp:
(WebCore::ComplexTextController::ComplexTextController): Initialize m_afterExpansion from the
TextRun, allowing leading expansion when appropriate. Set m_runWidthSoFar to the leading expansion.
(WebCore::ComplexTextController::offsetForPosition): Account for leading expansion.
(WebCore::ComplexTextController::adjustGlyphsAndAdvances): Similar to WidthIterator::advance()
* platform/graphics/mac/ComplexTextController.h: Added m_leadingExpansion member variable.
* rendering/InlineBox.h:
(WebCore::InlineBox::InlineBox): Renamed m_hasSelectedChildren to m_hasSelectedChildrenOrCanHaveLeadingExpansion
to reflect the use of this bit by InlineTextBox.
* rendering/InlineTextBox.cpp:
(WebCore::InlineTextBox::selectionRect): Replaced calls to trailingExpansionBehavior() with expansionBehavior().
(WebCore::InlineTextBox::paint): Ditto.
(WebCore::InlineTextBox::paintSelection): Ditto.
(WebCore::InlineTextBox::paintCompositionBackground): Ditto.
(WebCore::InlineTextBox::paintSpellingOrGrammarMarker): Ditto.
(WebCore::InlineTextBox::paintTextMatchMarker): Ditto.
(WebCore::InlineTextBox::computeRectForReplacementMarker): Ditto.
(WebCore::InlineTextBox::offsetForPosition): Ditto.
(WebCore::InlineTextBox::positionForOffset): Ditto.
* rendering/InlineTextBox.h:
(WebCore::InlineTextBox::canHaveLeadingExpansion): Added this accessor.
(WebCore::InlineTextBox::setCanHaveLeadingExpansion): Ditto.
(WebCore::InlineTextBox::expansionBehavior): Replaced trailingExpansionBehavior() with this function,
which also considers canHaveLeadingExpansion().
* rendering/RenderBlockLineLayout.cpp:
(WebCore::RenderBlock::computeInlineDirectionPositionsForLine): Call setCanHaveLeadingExpansion() on
text boxes that can have leading expansion. Avoid negative expansion.
* rendering/RootInlineBox.cpp:
* rendering/RootInlineBox.h:
(WebCore::RootInlineBox::hasSelectedChildren): Updated for renaming of the flag.
(WebCore::RootInlineBox::setHasSelectedChildren): Ditto.

LayoutTests:

* fast/text/justify-ideograph-leading-expansion.html: Added.
* platform/mac/fast/text/justify-ideograph-leading-expansion-expected.checksum: Added.
* platform/mac/fast/text/justify-ideograph-leading-expansion-expected.png: Added.
* platform/mac/fast/text/justify-ideograph-leading-expansion-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82013 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Anders Carlsson <andersca@apple.com>
andersca@apple.com [Sat, 26 Mar 2011 00:45:02 +0000 (00:45 +0000)]
2011-03-25  Anders Carlsson  <andersca@apple.com>

        Reviewed by Sam Weinig.

        PluginView's call to invalidateContentsAndWindow() in invalidateRect() is wrong for plug-ins which paint into compositing layers
        https://bugs.webkit.org/show_bug.cgi?id=57133
        <rdar://problem/9029442>

        * WebProcess/Plugins/Netscape/mac/NetscapePluginMac.mm:
        (WebKit::NetscapePlugin::platformInvalidate):
        Return true under the Core Animation drawing model, since that prevents the plug-in from calling PluginController::invalidate.

        * WebProcess/Plugins/PluginView.cpp:
        (WebKit::PluginView::invalidateRect):
        Don't call invalidateContentsAndWindow() if the plug-in has a Core Animation layer.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82012 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-23 Jer Noble <jer.noble@apple.com>
jer.noble@apple.com [Sat, 26 Mar 2011 00:41:12 +0000 (00:41 +0000)]
2011-03-23  Jer Noble  <jer.noble@apple.com>

        Reviewed by Maciej Stachowiak.

        MediaPlayerPrivateAVFoundation should use the Application Cache during load.
        https://bugs.webkit.org/show_bug.cgi?id=56997

        No new tests.

        When loading a URL, check to see if the Application Cache has a version of that URL
        stored; if so, use the local path to that cached media instead of the remote URL.

        * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.cpp:
        (WebCore::MediaPlayerPrivateAVFoundation::prepareToPlay): Check to see if the media should be loaded
            from the application cache.
        * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundation.h:
        * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundationObjC.h:
        * platform/graphics/avfoundation/MediaPlayerPrivateAVFoundationObjC.mm:
        (WebCore::MediaPlayerPrivateAVFoundationObjC::createAVPlayerForURL):
        (WebCore::MediaPlayerPrivateAVFoundationObjC::createAVPlayerForCacheResource): Added.
        (WebCore::MediaPlayerPrivateAVFoundationObjC::createAVPlayer): Split out from createAVPlayerForURL.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82011 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Ben Taylor <bentaylor.solx86@gmail.com>
commit-queue@webkit.org [Sat, 26 Mar 2011 00:30:40 +0000 (00:30 +0000)]
2011-03-25  Ben Taylor  <bentaylor.solx86@gmail.com>

        Reviewed by Darin Adler.

        https://bugs.webkit.org/show_bug.cgi?id=57122. Solaris 10/SunStudio 12 expect
        both sides of a ?: operation to have the same types.  Extracted from original
        https://bugs.webkit.org/show_bug.cgi?id=24932, patch 13 of 17, and originally
        created by Thiago Macieira.

        fixes a compile issue on Solaris 10/SunStudio 12

        * loader/DocumentLoader.h:
        (WebCore::DocumentLoader::serverRedirectSourceForHistory):
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::load):
        (WebCore::FrameLoader::loadWithNavigationAction):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82010 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-02-03 Jer Noble <jer.noble@apple.com>
jer.noble@apple.com [Sat, 26 Mar 2011 00:29:03 +0000 (00:29 +0000)]
2011-02-03  Jer Noble  <jer.noble@apple.com>

        Reviewed by Maciej Stachowiak.

        MediaPlayerPrivateQTKit should use the Application Cache during load.
        https://bugs.webkit.org/show_bug.cgi?id=53818

        No new tests.

        When loading a URL, check to see if the Application Cache has a version of that URL
        stored; if so, use that data instead of the remote URL.

        * platform/graphics/mac/MediaPlayerPrivateQTKit.h:
        * platform/graphics/mac/MediaPlayerPrivateQTKit.mm:
        (WebCore::MediaPlayerPrivateQTKit::commonMovieAttributes):
        (WebCore::MediaPlayerPrivateQTKit::createQTMovie):
        (WebCore::MediaPlayerPrivateQTKit::loadInternal):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82009 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago[chromium] add missing include from previous checkin
jam@chromium.org [Sat, 26 Mar 2011 00:14:02 +0000 (00:14 +0000)]
[chromium] add missing include from previous checkin

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82008 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoplugins/embed-prefers-plugins-for-images.html fails on Windows.
aestes@apple.com [Sat, 26 Mar 2011 00:12:16 +0000 (00:12 +0000)]
plugins/embed-prefers-plugins-for-images.html fails on Windows.
https://bugs.webkit.org/show_bug.cgi?id=57132

* platform/win/Skipped: Skip plugins/embed-prefers-plugins-for-images.html

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82007 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 James Robinson <jamesr@chromium.org>
jamesr@google.com [Sat, 26 Mar 2011 00:10:07 +0000 (00:10 +0000)]
2011-03-25  James Robinson  <jamesr@chromium.org>

        Reviewed by Kenneth Russell.

        [chromium] Move draw implementations to CCLayerImpl for everything except content layers
        https://bugs.webkit.org/show_bug.cgi?id=56793

        Moves the implementation of the layer's draw() function from LayerChromium subclasses to CCLayerImpl subclasses for
        all layer types except content layers.  This gets us closer to decoupling the composite step itself from updating the layers.

        The biggest change in this patch is adding a set of CCLayerImpl subclasses to implement the different drawing routines
        and moving the code from each XXXLayerChromium to CCXXXLayerImpl.  In order to render from the CCLayerImpl side all state
        needed at draw time also has to be synchronized, which is performed in pushPropertiesTo().

        On the LayerRendererChromium side there are a few changes.  I've split the updateContents calls on LayerChromiums into two operations
        tentatively named 'paintContentsIfNeeded' and 'updateCompositorResources'.  paintContents() is used for any code that calls into WebKit in order to generate
        new pixel data.  It's expected that this call may take a long period of time and may involve "odd" side effects.  updateCompositorResources() is used for
        code that needs to update the compositor's texture data or other state.  It is not expected that this callback will take long (since it's just inserting
        commands into the GL command stream, ideally) and this call is expected to have access to the compositor's context, unlike paintContents().

        The updateAndDrawLayers cycle now looks like this:
        1.) update the root content layer
        2.) update the root layer scrollbars
        3.) for each LayerChromium in tree order:
          a.) ensure a CCLayerImpl of the correct type exists for this layer
          b.) push all draw time properties from the LayerChromium to the CCLayerImpl
          c.) construct the appropriate draw transforms, render surfaces and render surface z-order sublayer lists
        4.) for each LayerChromium in tree order, paint the layer's contents
        5.) for each LayerChromium in tree order, update the associate compositor resources (textures, etc)
        6.) draw the root layer and its scrollbars
        7.) for each CCLayerImpl in render surface order, draw it

        Step 3 should really happen after step 5, but right now painting a content layer requires knowledge of the render surface properties and draw transforms
        in order to paint.  We also currently push layer properties from LayerChromium->CCLayerImpls twice now - once at 3.b and once after 5 so we can pick
        up any texture updates.  This will also get fixed when the paint dependency on render surfaces is resolved.

        I haven't modified the root layer or content layers in order to minimize conflicts with the other pending work in that area.

        Tests: compositing/

        * WebCore.gypi:
        * platform/graphics/chromium/Canvas2DLayerChromium.cpp:
        (WebCore::Canvas2DLayerChromium::updateCompositorResources):
        * platform/graphics/chromium/Canvas2DLayerChromium.h:
        * platform/graphics/chromium/CanvasLayerChromium.cpp:
        (WebCore::CanvasLayerChromium::createCCLayerImpl):
        (WebCore::CanvasLayerChromium::pushPropertiesTo):
        * platform/graphics/chromium/CanvasLayerChromium.h:
        * platform/graphics/chromium/ContentLayerChromium.cpp:
        (WebCore::ContentLayerChromium::requiresClippedUpdateRect):
        (WebCore::ContentLayerChromium::paintContentsIfDirty):
        (WebCore::ContentLayerChromium::updateCompositorResources):
        (WebCore::ContentLayerChromium::bindContentsTexture):
        * platform/graphics/chromium/ContentLayerChromium.h:
        * platform/graphics/chromium/ImageLayerChromium.cpp:
        (WebCore::ImageLayerChromium::paintContentsIfDirty):
        * platform/graphics/chromium/ImageLayerChromium.h:
        * platform/graphics/chromium/LayerChromium.cpp:
        (WebCore::LayerChromium::LayerChromium):
        (WebCore::LayerChromium::cleanupResources):
        (WebCore::LayerChromium::setLayerRenderer):
        (WebCore::LayerChromium::setName):
        (WebCore::LayerChromium::pushPropertiesTo):
        (WebCore::LayerChromium::dumpLayer):
        (WebCore::LayerChromium::createCCLayerImpl):
        (WebCore::LayerChromium::createCCLayerImplIfNeeded):
        (WebCore::LayerChromium::ccLayerImpl):
        (WebCore::LayerChromium::layerRenderer):
        * platform/graphics/chromium/LayerChromium.h:
        (WebCore::LayerChromium::paintContentsIfDirty):
        (WebCore::LayerChromium::updateCompositorResources):
        * platform/graphics/chromium/LayerRendererChromium.cpp:
        (WebCore::LayerRendererChromium::updateLayers):
        (WebCore::LayerRendererChromium::updatePropertiesAndRenderSurfaces):
        (WebCore::LayerRendererChromium::paintContentsRecursive):
        (WebCore::LayerRendererChromium::updateCompositorResourcesRecursive):
        (WebCore::LayerRendererChromium::drawLayer):
        (WebCore::LayerRendererChromium::initializeSharedObjects):
        * platform/graphics/chromium/LayerRendererChromium.h:
        (WebCore::LayerRendererChromium::canvasLayerProgram):
        (WebCore::LayerRendererChromium::pluginLayerProgram):
        (WebCore::LayerRendererChromium::videoLayerRGBAProgram):
        (WebCore::LayerRendererChromium::videoLayerYUVProgram):
        * platform/graphics/chromium/PluginLayerChromium.cpp:
        (WebCore::PluginLayerChromium::createCCLayerImpl):
        (WebCore::PluginLayerChromium::setTextureId):
        (WebCore::PluginLayerChromium::pushPropertiesTo):
        * platform/graphics/chromium/PluginLayerChromium.h:
        (WebCore::PluginLayerChromium::textureId):
        * platform/graphics/chromium/VideoLayerChromium.cpp:
        (WebCore::VideoLayerChromium::createCCLayerImpl):
        (WebCore::VideoLayerChromium::updateCompositorResources):
        (WebCore::VideoLayerChromium::pushPropertiesTo):
        * platform/graphics/chromium/VideoLayerChromium.h:
        * platform/graphics/chromium/WebGLLayerChromium.cpp:
        (WebCore::WebGLLayerChromium::updateCompositorResources):
        * platform/graphics/chromium/WebGLLayerChromium.h:
        * platform/graphics/chromium/cc/CCCanvasLayerImpl.cpp: Copied from Source/WebCore/platform/graphics/chromium/PluginLayerChromium.cpp.
        (WebCore::CCCanvasLayerImpl::CCCanvasLayerImpl):
        (WebCore::CCCanvasLayerImpl::~CCCanvasLayerImpl):
        (WebCore::CCCanvasLayerImpl::draw):
        (WebCore::CCCanvasLayerImpl::dumpLayerProperties):
        * platform/graphics/chromium/cc/CCCanvasLayerImpl.h: Copied from Source/WebCore/platform/graphics/chromium/PluginLayerChromium.h.
        (WebCore::CCCanvasLayerImpl::create):
        (WebCore::CCCanvasLayerImpl::setTextureId):
        (WebCore::CCCanvasLayerImpl::setPremultipliedAlpha):
        * platform/graphics/chromium/cc/CCLayerImpl.cpp:
        (WebCore::CCLayerImpl::descendantsDrawsContent):
        (WebCore::CCLayerImpl::updateCompositorResources):
        (WebCore::CCLayerImpl::writeIndent):
        * platform/graphics/chromium/cc/CCLayerImpl.h:
        (WebCore::CCLayerImpl::setAnchorPoint):
        (WebCore::CCLayerImpl::setAnchorPointZ):
        (WebCore::CCLayerImpl::setMasksToBounds):
        (WebCore::CCLayerImpl::setOpacity):
        (WebCore::CCLayerImpl::setPosition):
        (WebCore::CCLayerImpl::setPreserves3D):
        (WebCore::CCLayerImpl::setSublayerTransform):
        (WebCore::CCLayerImpl::setTransform):
        * platform/graphics/chromium/cc/CCPluginLayerImpl.cpp: Copied from Source/WebCore/platform/graphics/chromium/PluginLayerChromium.cpp.
        (WebCore::CCPluginLayerImpl::CCPluginLayerImpl):
        (WebCore::CCPluginLayerImpl::~CCPluginLayerImpl):
        (WebCore::CCPluginLayerImpl::draw):
        (WebCore::CCPluginLayerImpl::dumpLayerProperties):
        * platform/graphics/chromium/cc/CCPluginLayerImpl.h: Copied from Source/WebCore/platform/graphics/chromium/PluginLayerChromium.h.
        (WebCore::CCPluginLayerImpl::create):
        (WebCore::CCPluginLayerImpl::setTextureId):
        * platform/graphics/chromium/cc/CCVideoLayerImpl.cpp: Added.
        (WebCore::CCVideoLayerImpl::CCVideoLayerImpl):
        (WebCore::CCVideoLayerImpl::~CCVideoLayerImpl):
        (WebCore::CCVideoLayerImpl::setTexture):
        (WebCore::CCVideoLayerImpl::draw):
        (WebCore::CCVideoLayerImpl::drawYUV):
        (WebCore::CCVideoLayerImpl::drawRGBA):
        (WebCore::CCVideoLayerImpl::dumpLayerProperties):
        * platform/graphics/chromium/cc/CCVideoLayerImpl.h: Copied from Source/WebCore/platform/graphics/chromium/PluginLayerChromium.h.
        (WebCore::CCVideoLayerImpl::create):
        (WebCore::CCVideoLayerImpl::setSkipsDraw):
        (WebCore::CCVideoLayerImpl::setFrameFormat):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82006 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago<rdar://problem/8648311> and https://bugs.webkit.org/show_bug.cgi?id=56425
beidson@apple.com [Sat, 26 Mar 2011 00:09:07 +0000 (00:09 +0000)]
<rdar://problem/8648311> and https://bugs.webkit.org/show_bug.cgi?id=56425
WebKit2 IconDatabase and API

Reviewed by Sam Weinig.

This gets a basic, usable API in place.
Enhancements can come in other bugs as they're needed.

* UIProcess/API/C/WKIconDatabase.cpp:
(WKIconDatabaseRetainIconForURL):
(WKIconDatabaseReleaseIconForURL):
(WKIconDatabaseEnableDatabaseCleanup):
* UIProcess/API/C/WKIconDatabase.h:

The first actual accessor for an "icon for a URL" is for a CGImageRef:
* UIProcess/API/C/cg: Added.
* UIProcess/API/C/cg/WKIconDatabaseCG.cpp: Added.
(WKIconDatabaseGetCGImageForURL):
* UIProcess/API/C/cg/WKIconDatabaseCG.h: Added.

Expose relevant methods to support the above API:
* UIProcess/WebIconDatabase.cpp:
(WebKit::WebIconDatabase::WebIconDatabase): New IconDatabases will have cleanup disabled from the start,
  and require a single call to "enableDatabaseCleanup" to allow cleanup.
(WebKit::WebIconDatabase::setDatabasePath):
(WebKit::WebIconDatabase::enableDatabaseCleanup):
(WebKit::WebIconDatabase::imageForPageURL):
* UIProcess/WebIconDatabase.h:

Project file stuff:
* WebKit2.xcodeproj/project.pbxproj:
* win/WebKit2.vcproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82005 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoWebScrollBar not updating when calling WebScrollBar::setValue.
weinig@apple.com [Sat, 26 Mar 2011 00:04:41 +0000 (00:04 +0000)]
WebScrollBar not updating when calling WebScrollBar::setValue.
<rdar://problem/9143871>

Reviewed by Jon Honeycutt.

* WebScrollBar.cpp:
(WebScrollBar::setValue):
Call scrollToOffsetWithoutAnimation to actually update the scroll position.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82004 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Oliver Hunt <oliver@apple.com>
oliver@apple.com [Fri, 25 Mar 2011 23:20:39 +0000 (23:20 +0000)]
2011-03-25  Oliver Hunt  <oliver@apple.com>

        Reviewed by Darin Adler.

        Allow defineOwnProperty to work on DOMObjects
        https://bugs.webkit.org/show_bug.cgi?id=57129

        Now we disallow the cross origin accesses, rather than just disallowing
        the defineOwnProperty at all.

        * http/tests/security/xss-DENIED-defineProperty-expected.txt:
2011-03-25  Oliver Hunt  <oliver@apple.com>

        Reviewed by Darin Adler.

        Allow defineOwnProperty to work on DOMObjects
        https://bugs.webkit.org/show_bug.cgi?id=57129

        Fix a couple of places where we uses getter()/setter() rather
        than [gs]etterPresent().

        * runtime/JSObject.cpp:
        (JSC::JSObject::defineOwnProperty):
2011-03-25  Oliver Hunt  <oliver@apple.com>

        Reviewed by Darin Adler.

        Allow defineOwnProperty to work on DOMObjects
        https://bugs.webkit.org/show_bug.cgi?id=57129

        As other engines are allowing defineOwnProperty to be applied
        to host objects there's no reason for us to retain this
        restriction.

        * bindings/js/JSDOMWrapper.cpp:
        * bindings/js/JSDOMWrapper.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82003 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Jeff Miller <jeffm@apple.com>
jeffm@apple.com [Fri, 25 Mar 2011 23:15:07 +0000 (23:15 +0000)]
2011-03-25  Jeff Miller  <jeffm@apple.com>

        Reviewed by Darin Adler.

        Add CoreIPC::encodeResourceError() and CoreIPC::decodeResourceError() for all platforms
        https://bugs.webkit.org/show_bug.cgi?id=57125

        Currently, these are only implemented on the Mac, but we're going to need them for Windows as well, so we might as well make them always platform-specific.
        In this patch, just replicate the existing common non-Mac implementation in WebCoreArgumentCoders.h to platform-specific files.

        * Shared/WebCoreArgumentCoders.h: Move non-Mac implementations of encodeResourceError() and decodeResourceError() to common files.
        * Shared/gtk/WebCoreArgumentCodersGtk.cpp:
        (CoreIPC::encodeResourceError): Added.
        (CoreIPC::decodeResourceError): Added.
        * Shared/qt/WebCoreArgumentCodersQt.cpp:
        (CoreIPC::encodeResourceError): Added.
        (CoreIPC::decodeResourceError): Added.
        * Shared/win/WebCoreArgumentCodersWin.cpp:
        (CoreIPC::encodeResourceError): Added.
        (CoreIPC::decodeResourceError): Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82002 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Andy Estes <aestes@apple.com>
aestes@apple.com [Fri, 25 Mar 2011 23:07:43 +0000 (23:07 +0000)]
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        TestNetscapePlugIn needs to register support for an image MIME type so
        we can test that the embed tag prefers plug-ins to render image MIME
        types.

        * DumpRenderTree/TestNetscapePlugIn/mac/Info.plist: Register image/png.
        * DumpRenderTree/TestNetscapePlugIn/main.cpp:
        (NP_GetMIMEDescription): Ditto.
        * DumpRenderTree/TestNetscapePlugIn/win/TestNetscapePlugin.rc: Ditto.
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        * fast/images/embed-image.html: Disable plug-ins to continue testing
        native image rendering with embed.
        * plugins/embed-prefers-plugins-for-images-expected.txt: Added.
        * plugins/embed-prefers-plugins-for-images.html: Added.
        * plugins/plugin-javascript-access-expected.txt: Updated to include the
        new MIME type registered by TestNetscapePlugIn.
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        Update objectContentType() implementation to handle the
        shouldPreferPlugInsForImages flag.

        * src/FrameLoaderClientImpl.cpp:
        (WebKit::FrameLoaderClientImpl::objectContentType):
        * src/FrameLoaderClientImpl.h:
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        Update objectContentType() implementation to handle the
        shouldPreferPlugInsForImages flag.

        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (WebCore::FrameLoaderClientQt::objectContentType):
        * WebCoreSupport/FrameLoaderClientQt.h:
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        Update objectContentType() implementation to handle the
        shouldPreferPlugInsForImages flag.

        * WebCoreSupport/FrameLoaderClientGtk.cpp:
        (WebKit::FrameLoaderClient::objectContentType):
        * WebCoreSupport/FrameLoaderClientGtk.h:
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        Update objectContentType() implementation to handle the
        shouldPreferPlugInsForImages flag.

        * WebFrame.cpp:
        (WebFrame::objectContentType):
        * WebFrame.h:
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        Update objectContentType() implementation to handle the
        shouldPreferPlugInsForImages flag.

        * WebKitSupport/FrameLoaderClientWx.cpp:
        (WebCore::FrameLoaderClientWx::objectContentType):
        * WebKitSupport/FrameLoaderClientWx.h:
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        Update objectContentType() implementation to handle the
        shouldPreferPlugInsForImages flag.

        * WebCoreSupport/WebFrameLoaderClient.h:
        * WebCoreSupport/WebFrameLoaderClient.mm:
        (WebFrameLoaderClient::objectContentType):
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        Update objectContentType() implementation to handle the
        shouldPreferPlugInsForImages flag.

        * WebCoreSupport/FrameLoaderClientHaiku.cpp:
        (WebCore::FrameLoaderClientHaiku::objectContentType):
        * WebCoreSupport/FrameLoaderClientHaiku.h:
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        Update objectContentType() implementation to handle the
        shouldPreferPlugInsForImages flag.

        * WebCoreSupport/FrameLoaderClientWinCE.cpp:
        (WebKit::FrameLoaderClientWinCE::objectContentType):
        * WebCoreSupport/FrameLoaderClientWinCE.h:
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        Update objectContentType() implementation to handle the
        shouldPreferPlugInsForImages flag.

        * WebCoreSupport/FrameLoaderClientEfl.cpp:
        (WebCore::FrameLoaderClientEfl::objectContentType):
        * WebCoreSupport/FrameLoaderClientEfl.h:
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        AC_QuickTime.js uses the common <object>/<embed> paradigm to embed the
        QuickTime plug-in in web pages. r70748 removed our mapping of classids
        to MIME types, which causes WebKit to fall back from the object to the
        embed tag when QuickTime is embedded by this script. The script emits
        the following embed tag to embed a QuickTime movie with a poster frame:

        <embed src="poster-frame.png" target="quicktimeplayer" href="movie.mov">

        The expectation is that a QuickTime plug-in is instantiated to display the
        poster frame, since QuickTime registers for many common image MIME
        types. This is how Gecko behaves for embed. However, WebKit prefers to
        use its native image rendering for image embeds, in which case no movie
        is played when the poster frame is clicked.

        Fix this by changing embed to check for a plug-in that can handle the
        image type before rendering the image natively. This matches Gecko.

        Test: plugins/embed-prefers-plugins-for-images.html

        * html/HTMLEmbedElement.cpp:
        (WebCore::HTMLEmbedElement::HTMLEmbedElement):
        * html/HTMLObjectElement.cpp:
        (WebCore::HTMLObjectElement::HTMLObjectElement):
        (WebCore::HTMLObjectElement::parametersForPlugin):
        * html/HTMLPlugInImageElement.cpp:
        (WebCore::HTMLPlugInImageElement::HTMLPlugInImageElement):
        (WebCore::HTMLPlugInImageElement::isImageType):
        (WebCore::HTMLPlugInImageElement::wouldLoadAsNetscapePlugin):
        * html/HTMLPlugInImageElement.h:
        (WebCore::HTMLPlugInImageElement::shouldPreferPlugInsForImages):
        * loader/EmptyClients.h:
        (WebCore::EmptyFrameLoaderClient::objectContentType):
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::defaultObjectContentType):
        * loader/FrameLoader.h:
        * loader/FrameLoaderClient.h:
        * loader/SubframeLoader.cpp:
        (WebCore::SubframeLoader::resourceWillUsePlugin):
        (WebCore::SubframeLoader::requestPlugin):
        (WebCore::SubframeLoader::requestObject):
        (WebCore::SubframeLoader::shouldUsePlugin):
        * loader/SubframeLoader.h:
2011-03-25  Andy Estes  <aestes@apple.com>

        Reviewed by Adele Peterson.

        REGRESSION (r70748): latest nightly builds kills AC_QuickTime.js
        https://bugs.webkit.org/show_bug.cgi?id=49016

        Update objectContentType() implementation to handle the
        shouldPreferPlugInsForImages flag.

        * WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
        (WebKit::WebFrameLoaderClient::objectContentType):
        * WebProcess/WebCoreSupport/WebFrameLoaderClient.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82001 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-23 Jer Noble <jer.noble@apple.com>
jer.noble@apple.com [Fri, 25 Mar 2011 23:04:54 +0000 (23:04 +0000)]
2011-03-23  Jer Noble  <jer.noble@apple.com>

        Reviewed by Eric Carlson.

        Application Cache should save audio/ and video/ mime types as flat files
        https://bugs.webkit.org/show_bug.cgi?id=53784
        <rdar://problem/8932473>

        No new tests, as this behavior is not possible to test without changes to the MediaPlayer engines.

        Store certain mime types as flat files alongside the Application Cache database.
        This requires plumbing the saved file path from ApplicationCacheStorage through
        to ApplicationCacheResource.

        (WebCore::ApplicationCacheStorage::openDatabase): Modify the CacheResourceData schema and
            add a new DeletedCacheResources table, add a new CacheResourceDataDeleted trigger.
        (WebCore::ApplicationCacheStorage::store): Add the new path data when adding new rows in
            CacheResourceData, and store media resources as flat files.
        (WebCore::ApplicationCacheStorage::loadCache): Pull the "path" column from CacheResourceData
            when loading cache items.
        (WebCore::ApplicationCacheStorage::remove): Call checkForDeletedResources.
        (WebCore::ApplicationCacheStorage::empty): Ditto.
        (WebCore::ApplicationCacheStorage::storeCopyOfCache): Ditto.
        (WebCore::ApplicationCacheStorage::deleteCacheGroup): Ditto.
        (WebCore::ApplicationCacheStorage::checkForDeletedResources): Walk through DeletedCacheResourceData
            looking for entries with non-empty "path" columns; if found, delete.
        (WebCore::ApplicationCacheStorage::flatFileAreaSize): Walk through CacheResourceData rows
            and sum the file size of those rows with flat file storage.
        (WebCore::ApplicationCacheStorage::verifySchemaVersion): Call deleteTables() instead of
            clearAllTables() directly.
        (WebCore::ApplicationCacheStorage::deleteTables): Call empty() before deleting the tables,
            so that flat files get deleted.
        (WebCore::ApplicationCacheStorage::shouldStoreResourceAsFlatFile): Added.
        (WebCore::ApplicationCacheStorage::writeDataToUniqueFileInDirectory): Added.
        * loader/appcache/ApplicationCacheStorage.h:
        * platform/win/FileSystemWin.cpp:
        (WebCore::openFile): Implement openFile on Windows.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@82000 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Kinuko Yasuda <kinuko@chromium.org>
kinuko@chromium.org [Fri, 25 Mar 2011 23:01:00 +0000 (23:01 +0000)]
2011-03-25  Kinuko Yasuda  <kinuko@chromium.org>

        Reviewed by David Levin.

        Data race between ~WorkerFileSystemCallbacksBridge and runTasksOnWorkerThread
        https://bugs.webkit.org/show_bug.cgi?id=56138
        Stopped passing the WorkerFileSystemCallbacksBridge's reference to
        MainThreadFileSystemCallbacks so that in most cases (in normal cases)
        its reference is only maintained by WorkerThread.

        * src/WorkerFileSystemCallbacksBridge.cpp:
        (WebKit::MainThreadFileSystemCallbacks::createLeakedPtr):
        (WebKit::MainThreadFileSystemCallbacks::MainThreadFileSystemCallbacks):
        (WebKit::WorkerFileSystemCallbacksBridge::mayPostTaskToWorker):
        * src/WorkerFileSystemCallbacksBridge.h:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81999 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Vincent Scheib <scheib@chromium.org>
scheib@chromium.org [Fri, 25 Mar 2011 22:56:52 +0000 (22:56 +0000)]
2011-03-25  Vincent Scheib  <scheib@chromium.org>

        fast/forms/input-file-directory-upload.html timing out.
        http://crbug.com/77468

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81998 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed build correction.
jberlin@webkit.org [Fri, 25 Mar 2011 22:33:54 +0000 (22:33 +0000)]
Unreviewed build correction.

Patch by Brent Fulgham <bfulgham@webkit.org> on 2011-03-25
Add stubs for WK2 cookie handling with cURL.

* WebProcess/Cookies/curl: Added.
* WebProcess/Cookies/curl/WebCookieManagerCurl.cpp: Added.
(WebKit::WebCookieManager::platformSetHTTPCookieAcceptPolicy):
(WebKit::WebCookieManager::platformGetHTTPCookieAcceptPolicy):
* win/WebKit2.vcproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81997 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Vincent Scheib <scheib@chromium.org>
scheib@chromium.org [Fri, 25 Mar 2011 22:15:59 +0000 (22:15 +0000)]
2011-03-25  Vincent Scheib  <scheib@chromium.org>

        Attempting SLOW to resolve fast/forms/input-file-directory-upload.html

        * platform/chromium/test_expectations.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81996 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Fri, 25 Mar 2011 22:04:14 +0000 (22:04 +0000)]
2011-03-25  Martin Robinson  <mrobinson@igalia.com>

        Add a GTK+ specific baseline for a text test that seems to have whitespace difference,
        probably due to font sizes resulting in newlines.

        * platform/gtk/fast/events/onchange-text-form-field-expected.txt: Added.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81995 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-24 David Hyatt <hyatt@apple.com>
ggaren@apple.com [Fri, 25 Mar 2011 22:01:18 +0000 (22:01 +0000)]
2011-03-24  David Hyatt  <hyatt@apple.com>

        Reviewed by Dan Bernstein.

        Add a simplified normal flow layout path optimization for overflow recomputation
        and for positioned objects inside relative positioned containers.

        Currently there is an optimized code path for positioned objects, but as soon as
        we encounter a normal flow object in the containing block chain, we lose the
        optimization.

        This patch adds a new type of style difference called SimplifiedLayout that is
        returned when only overflow needs to be recomputed. Whenever a transform changes,
        this is the hint returned now instead of a full layout.

        In addition, when positioned objects need layout and start marking up the
        containing block chain, we now propagate the fact that the layout is simplified
        all the way up to the root, even when we encounter normal flow containing
        blocks.

        The layoutOnlyPositionedObjects function has been renamed to simplifiedLayout()
        and is now used for all of these cases (in addition to what it handled before).

        The simplified layout optimization (even in ToT) did not work correctly when
        static distances needed to be recomputed. In order to make static distance
        computations work with simplified layout, positioned objects now always compute
        their static offsets, even if they explicitly specify left/top.  That way normal
        flow layout never has to re-run when the positioned object moves.  This makes
        movement of a positioned object along a single non-auto axis much faster when the
        other axis is auto. Because this code kicked in more often for absolutely positioned
        objects whose original display was inline, I went ahead and fixed the trailing space
        issue with those objects.  This causes a bunch of layout tests to progress.

        Added fast/block/positioning/static-inline-position-dynamic.html and trailing-space-test.html.

        * animations/dynamic-stylesheet-loading-expected.txt:
        * animations/keyframe-timing-functions-expected.txt:
        * fast/block/positioning/static-inline-position-dynamic.html: Added.
        * fast/block/positioning/trailing-space-test.html: Added.
        * fast/dynamic/ancestor-to-absolute-expected.txt:
        * fast/forms/input-appearance-elementFromPoint-expected.txt:
        * fast/forms/input-hit-test-border-expected.txt:
        * platform/mac/fast/block/positioning/002-expected.png:
        * platform/mac/fast/block/positioning/003-expected.png:
        * platform/mac/fast/block/positioning/004-expected.png:
        * platform/mac/fast/block/positioning/005-expected.png:
        * platform/mac/fast/block/positioning/006-expected.png:
        * platform/mac/fast/block/positioning/007-expected.png:
        * platform/mac/fast/block/positioning/008-expected.png:
        * platform/mac/fast/block/positioning/009-expected.png:
        * platform/mac/fast/block/positioning/010-expected.png:
        * platform/mac/fast/block/positioning/011-expected.png:
        * platform/mac/fast/block/positioning/012-expected.png:
        * platform/mac/fast/block/positioning/013-expected.png:
        * platform/mac/fast/block/positioning/014-expected.png:
        * platform/mac/fast/block/positioning/015-expected.png:
        * platform/mac/fast/block/positioning/016-expected.png:
        * platform/mac/fast/block/positioning/017-expected.png:
        * platform/mac/fast/block/positioning/018-expected.png:
        * platform/mac/fast/block/positioning/019-expected.png:
        * platform/mac/fast/block/positioning/020-expected.png:
        * platform/mac/fast/block/positioning/021-expected.png:
        * platform/mac/fast/block/positioning/022-expected.png:
        * platform/mac/fast/block/positioning/023-expected.png:
        * platform/mac/fast/block/positioning/024-expected.png:
        * platform/mac/fast/block/positioning/025-expected.png:
        * platform/mac/fast/block/positioning/026-expected.png:
        * platform/mac/fast/block/positioning/027-expected.png:
        * platform/mac/fast/block/positioning/028-expected.png:
        * platform/mac/fast/block/positioning/029-expected.png:
        * platform/mac/fast/block/positioning/030-expected.png:
        * platform/mac/fast/block/positioning/031-expected.png:
        * platform/mac/fast/block/positioning/032-expected.png:
        * platform/mac/fast/block/positioning/033-expected.png:
        * platform/mac/fast/block/positioning/034-expected.png:
        * platform/mac/fast/block/positioning/035-expected.png:
        * platform/mac/fast/block/positioning/036-expected.png:
        * platform/mac/fast/block/positioning/037-expected.png:
        * platform/mac/fast/block/positioning/038-expected.png:
        * platform/mac/fast/block/positioning/039-expected.png:
        * platform/mac/fast/block/positioning/040-expected.png:
        * platform/mac/fast/block/positioning/041-expected.png:
        * platform/mac/fast/block/positioning/042-expected.png:
        * platform/mac/fast/block/positioning/043-expected.png:
        * platform/mac/fast/block/positioning/044-expected.png:
        * platform/mac/fast/block/positioning/045-expected.png:
        * platform/mac/fast/block/positioning/046-expected.png:
        * platform/mac/fast/block/positioning/047-expected.checksum:
        * platform/mac/fast/block/positioning/047-expected.png:
        * platform/mac/fast/block/positioning/048-expected.png:
        * platform/mac/fast/block/positioning/049-expected.png:
        * platform/mac/fast/block/positioning/050-expected.png:
        * platform/mac/fast/block/positioning/056-expected.txt:
        * platform/mac/fast/block/positioning/061-expected.png:
        * platform/mac/fast/block/positioning/062-expected.png:
        * platform/mac/fast/block/positioning/auto/005-expected.txt:
        * platform/mac/fast/block/positioning/auto/006-expected.txt:
        * platform/mac/fast/block/positioning/auto/007-expected.png:
        * platform/mac/fast/block/positioning/auto/vertical-lr/005-expected.txt:
        * platform/mac/fast/block/positioning/auto/vertical-lr/006-expected.txt:
        * platform/mac/fast/block/positioning/auto/vertical-lr/007-expected.checksum:
        * platform/mac/fast/block/positioning/auto/vertical-lr/007-expected.png:
        * platform/mac/fast/block/positioning/auto/vertical-rl/005-expected.txt:
        * platform/mac/fast/block/positioning/auto/vertical-rl/006-expected.txt:
        * platform/mac/fast/block/positioning/auto/vertical-rl/007-expected.checksum:
        * platform/mac/fast/block/positioning/auto/vertical-rl/007-expected.png:
        * platform/mac/fast/block/positioning/fixed-positioning-scrollbar-bug-expected.txt:
        * platform/mac/fast/block/positioning/inline-block-relposition-expected.checksum:
        * platform/mac/fast/block/positioning/inline-block-relposition-expected.png:
        * platform/mac/fast/block/positioning/inline-block-relposition-expected.txt:
        * platform/mac/fast/block/positioning/move-with-auto-width-expected.png:
        * platform/mac/fast/block/positioning/negative-rel-position-expected.png:
        * platform/mac/fast/block/positioning/relative-overconstrained-expected.png:
        * platform/mac/fast/block/positioning/relative-overflow-block-expected.txt:
        * platform/mac/fast/block/positioning/static-inline-position-dynamic-expected.checksum: Added.
        * platform/mac/fast/block/positioning/static-inline-position-dynamic-expected.png: Added.
        * platform/mac/fast/block/positioning/static-inline-position-dynamic-expected.txt: Added.
        * platform/mac/fast/block/positioning/trailing-space-test-expected.checksum: Added.
        * platform/mac/fast/block/positioning/trailing-space-test-expected.png: Added.
        * platform/mac/fast/block/positioning/trailing-space-test-expected.txt: Added.
        * platform/mac/fast/clip/001-expected.txt:
        * platform/mac/fast/clip/004-expected.txt:
        * platform/mac/fast/clip/006-expected.txt:
        * platform/mac/fast/clip/007-expected.txt:
        * platform/mac/fast/clip/008-expected.txt:
        * platform/mac/fast/clip/009-expected.txt:
        * platform/mac/fast/clip/010-expected.txt:
        * platform/mac/fast/clip/011-expected.txt:
        * platform/mac/fast/clip/012-expected.txt:
        * platform/mac/fast/clip/013-expected.txt:
        * platform/mac/fast/clip/014-expected.txt:
        * platform/mac/fast/clip/nestedTransparencyClip-expected.txt:
        * platform/mac/fast/clip/outline-overflowClip-expected.txt:
        * platform/mac/fast/forms/input-appearance-preventDefault-expected.txt:
        * platform/mac/fast/invalid/014-expected.txt:
        * platform/mac/fast/layers/layer-visibility-expected.txt:
        * platform/mac/fast/repaint/layout-state-scrolloffset-expected.txt:
        * platform/mac/fast/repaint/layout-state-scrolloffset2-expected.txt:
        * platform/mac/fast/repaint/layout-state-scrolloffset3-expected.txt:
        * platform/mac/tables/mozilla/bugs/bug51140-expected.txt:
2011-03-25  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Crash when paused at a breakpoint caused by inaccurate Activation records.
        https://bugs.webkit.org/show_bug.cgi?id=57120

        * runtime/JSActivation.cpp:
        (JSC::JSActivation::symbolTableGet):
        (JSC::JSActivation::symbolTablePut):
        (JSC::JSActivation::getOwnPropertyNames):
        (JSC::JSActivation::symbolTablePutWithAttributes):

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81994 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoUnreviewed build correction.
bfulgham@webkit.org [Fri, 25 Mar 2011 21:59:10 +0000 (21:59 +0000)]
Unreviewed build correction.

Add stubs for WK2 cookie handling with cURL.

* WebProcess/Cookies/curl: Added.
* WebProcess/Cookies/curl/WebCookieManagerCurl.cpp: Added.
(WebKit::WebCookieManager::platformSetHTTPCookieAcceptPolicy):
(WebKit::WebCookieManager::platformGetHTTPCookieAcceptPolicy):
* win/WebKit2.vcproj:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81993 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoSource/WebCore: https://bugs.webkit.org/show_bug.cgi?id=56909
hyatt@apple.com [Fri, 25 Mar 2011 21:57:07 +0000 (21:57 +0000)]
Source/WebCore: https://bugs.webkit.org/show_bug.cgi?id=56909

Reviewed by Dan Bernstein.

Add a simplified normal flow layout path optimization for overflow recomputation
and for positioned objects inside relative positioned containers.

Currently there is an optimized code path for positioned objects, but as soon as
we encounter a normal flow object in the containing block chain, we lose the
optimization.

This patch adds a new type of style difference called SimplifiedLayout that is
returned when only overflow needs to be recomputed. Whenever a transform changes,
this is the hint returned now instead of a full layout.

In addition, when positioned objects need layout and start marking up the
containing block chain, we now propagate the fact that the layout is simplified
all the way up to the root, even when we encounter normal flow containing
blocks.

The layoutOnlyPositionedObjects function has been renamed to simplifiedLayout()
and is now used for all of these cases (in addition to what it handled before).

The simplified layout optimization (even in ToT) did not work correctly when
static distances needed to be recomputed. In order to make static distance
computations work with simplified layout, positioned objects now always compute
their static offsets, even if they explicitly specify left/top.  That way normal
flow layout never has to re-run when the positioned object moves.  This makes
movement of a positioned object along a single non-auto axis much faster when the
other axis is auto. Because this code kicked in more often for absolutely positioned
objects whose original display was inline, I went ahead and fixed the trailing space
issue with those objects.  This causes a bunch of layout tests to progress.

Added fast/block/positioning/static-inline-position-dynamic.html and trailing-space-test.html.

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::layoutBlock):
(WebCore::RenderBlock::adjustPositionedBlock):
(WebCore::RenderBlock::simplifiedNormalFlowLayout):
(WebCore::RenderBlock::simplifiedLayout):
* rendering/RenderBlock.h:
* rendering/RenderBlockLineLayout.cpp:
(WebCore::setStaticPositions):
(WebCore::RenderBlock::findNextLineBreak):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::styleDidChange):
(WebCore::RenderBox::positionLineBox):
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::styleWillChange):
* rendering/RenderFlexibleBox.cpp:
(WebCore::RenderFlexibleBox::layoutBlock):
(WebCore::RenderFlexibleBox::layoutHorizontalBox):
(WebCore::RenderFlexibleBox::layoutVerticalBox):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::RenderObject):
(WebCore::RenderObject::adjustStyleDifference):
(WebCore::RenderObject::setStyle):
(WebCore::RenderObject::styleDidChange):
* rendering/RenderObject.h:
(WebCore::RenderObject::needsLayout):
(WebCore::RenderObject::needsPositionedMovementLayoutOnly):
(WebCore::RenderObject::needsSimplifiedNormalFlowLayout):
(WebCore::RenderObject::setNeedsLayout):
(WebCore::RenderObject::setChildNeedsLayout):
(WebCore::RenderObject::setNeedsSimplifiedNormalFlowLayout):
(WebCore::RenderObject::markContainingBlocksForLayout):
* rendering/RenderTable.cpp:
(WebCore::RenderTable::layout):
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::diff):
* rendering/style/RenderStyleConstants.h:
* rendering/svg/RenderSVGText.cpp:
(WebCore::RenderSVGText::layout):

LayoutTests: Add a simplified normal flow layout path optimization for overflow recomputation
and for positioned objects inside relative positioned containers.

Reviewed by Dan Bernstein.

Currently there is an optimized code path for positioned objects, but as soon as
we encounter a normal flow object in the containing block chain, we lose the
optimization.

This patch adds a new type of style difference called SimplifiedLayout that is
returned when only overflow needs to be recomputed. Whenever a transform changes,
this is the hint returned now instead of a full layout.

In addition, when positioned objects need layout and start marking up the
containing block chain, we now propagate the fact that the layout is simplified
all the way up to the root, even when we encounter normal flow containing
blocks.

The layoutOnlyPositionedObjects function has been renamed to simplifiedLayout()
and is now used for all of these cases (in addition to what it handled before).

The simplified layout optimization (even in ToT) did not work correctly when
static distances needed to be recomputed. In order to make static distance
computations work with simplified layout, positioned objects now always compute
their static offsets, even if they explicitly specify left/top.  That way normal
flow layout never has to re-run when the positioned object moves.  This makes
movement of a positioned object along a single non-auto axis much faster when the
other axis is auto. Because this code kicked in more often for absolutely positioned
objects whose original display was inline, I went ahead and fixed the trailing space
issue with those objects.  This causes a bunch of layout tests to progress.

Added fast/block/positioning/static-inline-position-dynamic.html and trailing-space-test.html.

* animations/dynamic-stylesheet-loading-expected.txt:
* animations/keyframe-timing-functions-expected.txt:
* fast/block/positioning/static-inline-position-dynamic.html: Added.
* fast/block/positioning/trailing-space-test.html: Added.
* fast/dynamic/ancestor-to-absolute-expected.txt:
* fast/forms/input-appearance-elementFromPoint-expected.txt:
* fast/forms/input-hit-test-border-expected.txt:
* platform/mac/fast/block/positioning/002-expected.png:
* platform/mac/fast/block/positioning/003-expected.png:
* platform/mac/fast/block/positioning/004-expected.png:
* platform/mac/fast/block/positioning/005-expected.png:
* platform/mac/fast/block/positioning/006-expected.png:
* platform/mac/fast/block/positioning/007-expected.png:
* platform/mac/fast/block/positioning/008-expected.png:
* platform/mac/fast/block/positioning/009-expected.png:
* platform/mac/fast/block/positioning/010-expected.png:
* platform/mac/fast/block/positioning/011-expected.png:
* platform/mac/fast/block/positioning/012-expected.png:
* platform/mac/fast/block/positioning/013-expected.png:
* platform/mac/fast/block/positioning/014-expected.png:
* platform/mac/fast/block/positioning/015-expected.png:
* platform/mac/fast/block/positioning/016-expected.png:
* platform/mac/fast/block/positioning/017-expected.png:
* platform/mac/fast/block/positioning/018-expected.png:
* platform/mac/fast/block/positioning/019-expected.png:
* platform/mac/fast/block/positioning/020-expected.png:
* platform/mac/fast/block/positioning/021-expected.png:
* platform/mac/fast/block/positioning/022-expected.png:
* platform/mac/fast/block/positioning/023-expected.png:
* platform/mac/fast/block/positioning/024-expected.png:
* platform/mac/fast/block/positioning/025-expected.png:
* platform/mac/fast/block/positioning/026-expected.png:
* platform/mac/fast/block/positioning/027-expected.png:
* platform/mac/fast/block/positioning/028-expected.png:
* platform/mac/fast/block/positioning/029-expected.png:
* platform/mac/fast/block/positioning/030-expected.png:
* platform/mac/fast/block/positioning/031-expected.png:
* platform/mac/fast/block/positioning/032-expected.png:
* platform/mac/fast/block/positioning/033-expected.png:
* platform/mac/fast/block/positioning/034-expected.png:
* platform/mac/fast/block/positioning/035-expected.png:
* platform/mac/fast/block/positioning/036-expected.png:
* platform/mac/fast/block/positioning/037-expected.png:
* platform/mac/fast/block/positioning/038-expected.png:
* platform/mac/fast/block/positioning/039-expected.png:
* platform/mac/fast/block/positioning/040-expected.png:
* platform/mac/fast/block/positioning/041-expected.png:
* platform/mac/fast/block/positioning/042-expected.png:
* platform/mac/fast/block/positioning/043-expected.png:
* platform/mac/fast/block/positioning/044-expected.png:
* platform/mac/fast/block/positioning/045-expected.png:
* platform/mac/fast/block/positioning/046-expected.png:
* platform/mac/fast/block/positioning/047-expected.checksum:
* platform/mac/fast/block/positioning/047-expected.png:
* platform/mac/fast/block/positioning/048-expected.png:
* platform/mac/fast/block/positioning/049-expected.png:
* platform/mac/fast/block/positioning/050-expected.png:
* platform/mac/fast/block/positioning/056-expected.txt:
* platform/mac/fast/block/positioning/061-expected.png:
* platform/mac/fast/block/positioning/062-expected.png:
* platform/mac/fast/block/positioning/auto/005-expected.txt:
* platform/mac/fast/block/positioning/auto/006-expected.txt:
* platform/mac/fast/block/positioning/auto/007-expected.png:
* platform/mac/fast/block/positioning/auto/vertical-lr/005-expected.txt:
* platform/mac/fast/block/positioning/auto/vertical-lr/006-expected.txt:
* platform/mac/fast/block/positioning/auto/vertical-lr/007-expected.checksum:
* platform/mac/fast/block/positioning/auto/vertical-lr/007-expected.png:
* platform/mac/fast/block/positioning/auto/vertical-rl/005-expected.txt:
* platform/mac/fast/block/positioning/auto/vertical-rl/006-expected.txt:
* platform/mac/fast/block/positioning/auto/vertical-rl/007-expected.checksum:
* platform/mac/fast/block/positioning/auto/vertical-rl/007-expected.png:
* platform/mac/fast/block/positioning/fixed-positioning-scrollbar-bug-expected.txt:
* platform/mac/fast/block/positioning/inline-block-relposition-expected.checksum:
* platform/mac/fast/block/positioning/inline-block-relposition-expected.png:
* platform/mac/fast/block/positioning/inline-block-relposition-expected.txt:
* platform/mac/fast/block/positioning/move-with-auto-width-expected.png:
* platform/mac/fast/block/positioning/negative-rel-position-expected.png:
* platform/mac/fast/block/positioning/relative-overconstrained-expected.png:
* platform/mac/fast/block/positioning/relative-overflow-block-expected.txt:
* platform/mac/fast/block/positioning/static-inline-position-dynamic-expected.checksum: Added.
* platform/mac/fast/block/positioning/static-inline-position-dynamic-expected.png: Added.
* platform/mac/fast/block/positioning/static-inline-position-dynamic-expected.txt: Added.
* platform/mac/fast/block/positioning/trailing-space-test-expected.checksum: Added.
* platform/mac/fast/block/positioning/trailing-space-test-expected.png: Added.
* platform/mac/fast/block/positioning/trailing-space-test-expected.txt: Added.
* platform/mac/fast/clip/001-expected.txt:
* platform/mac/fast/clip/004-expected.txt:
* platform/mac/fast/clip/006-expected.txt:
* platform/mac/fast/clip/007-expected.txt:
* platform/mac/fast/clip/008-expected.txt:
* platform/mac/fast/clip/009-expected.txt:
* platform/mac/fast/clip/010-expected.txt:
* platform/mac/fast/clip/011-expected.txt:
* platform/mac/fast/clip/012-expected.txt:
* platform/mac/fast/clip/013-expected.txt:
* platform/mac/fast/clip/014-expected.txt:
* platform/mac/fast/clip/nestedTransparencyClip-expected.txt:
* platform/mac/fast/clip/outline-overflowClip-expected.txt:
* platform/mac/fast/forms/input-appearance-preventDefault-expected.txt:
* platform/mac/fast/invalid/014-expected.txt:
* platform/mac/fast/layers/layer-visibility-expected.txt:
* platform/mac/fast/repaint/layout-state-scrolloffset-expected.txt:
* platform/mac/fast/repaint/layout-state-scrolloffset2-expected.txt:
* platform/mac/fast/repaint/layout-state-scrolloffset3-expected.txt:
* platform/mac/tables/mozilla/bugs/bug51140-expected.txt:

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81992 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years agoFix the GTK+ build until we can implement the new cross-platform
slewis@apple.com [Fri, 25 Mar 2011 21:21:17 +0000 (21:21 +0000)]
Fix the GTK+ build until we can implement the new cross-platform
FontPlatformData.h bits.

Patch by Martin Robinson <mrobinson@igalia.com> on 2011-03-25
* platform/graphics/FontPlatformData.h: Include the proper old font headers.
(WebCore::FontPlatformData::FontPlatformData): Eliminate GTK+ specific bits of this file
which reference our obsolete font implementation.
(WebCore::FontPlatformData::hash): Ditto.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81991 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Martin Robinson <mrobinson@igalia.com>
mrobinson@webkit.org [Fri, 25 Mar 2011 21:13:54 +0000 (21:13 +0000)]
2011-03-25  Martin Robinson  <mrobinson@igalia.com>

        Fix the GTK+ build until we can implement the new cross-platform
        FontPlatformData.h bits.

        * platform/graphics/FontPlatformData.h: Include the proper old font headers.
        (WebCore::FontPlatformData::FontPlatformData): Eliminate GTK+ specific bits of this file
        which reference our obsolete font implementation.
        (WebCore::FontPlatformData::hash): Ditto.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81990 268f45cc-cd09-0410-ab3c-d52691b4dbfc

10 years ago2011-03-25 Steve Falkenburg <sfalken@apple.com>
sfalken@apple.com [Fri, 25 Mar 2011 21:04:54 +0000 (21:04 +0000)]
2011-03-25  Steve Falkenburg  <sfalken@apple.com>

        Reviewed by Brian Weinstein.

        Microsoft Windows Presentation Foundation (WPF) plug-in complains about missing xpcom.dll
        https://bugs.webkit.org/show_bug.cgi?id=57119
        <rdar://problem/9054148>

        This plug-in from Microsoft links against xpcom.dll, which is a Firefox-specific DLL
        not available in WebKit. The plug-in is fairly widespread, since it was included in
        a Windows Update push at one point.

        * plugins/win/PluginPackageWin.cpp:
        (WebCore::PluginPackage::isPluginBlacklisted): Blacklist npwpf.dll.
2011-03-25  Steve Falkenburg  <sfalken@apple.com>

        Reviewed by Brian Weinstein.

        Microsoft Windows Presentation Foundation (WPF) plug-in complains about missing xpcom.dll
        https://bugs.webkit.org/show_bug.cgi?id=57119
        <rdar://problem/9054148>

        This plug-in from Microsoft links against xpcom.dll, which is a Firefox-specific DLL
        not available in WebKit. The plug-in is fairly widespread, since it was included in
        a Windows Update push at one point.

        * UIProcess/Plugins/win/PluginInfoStoreWin.cpp:
        (WebKit::PluginInfoStore::shouldUsePlugin): Blacklist npwpf.dll.

git-svn-id: http://svn.webkit.org/repository/webkit/trunk@81989 268f45cc-cd09-0410-ab3c-d52691b4dbfc