WebKit-https.git
11 years ago2010-09-14 Chris Rogers <crogers@google.com>
crogers@google.com [Wed, 15 Sep 2010 03:19:02 +0000 (03:19 +0000)]
2010-09-14  Chris Rogers  <crogers@google.com>

        Reviewed by Kenneth Russell.

        audio engine: add AudioFileReader files (Mac implementation)
        https://bugs.webkit.org/show_bug.cgi?id=36475

        No new tests since audio API is not yet implemented.

        * platform/audio/AudioFileReader.h: Added.
        * platform/audio/mac/AudioFileReaderMac.cpp: Added.
        (WebCore::createAudioBufferList):
        (WebCore::destroyAudioBufferList):
        (WebCore::AudioFileReader::AudioFileReader):
        (WebCore::AudioFileReader::~AudioFileReader):
        (WebCore::AudioFileReader::readProc):
        (WebCore::AudioFileReader::getSizeProc):
        (WebCore::AudioFileReader::createBus):
        (WebCore::createBusFromAudioFile):
        (WebCore::createBusFromInMemoryAudioFile):
        * platform/audio/mac/AudioFileReaderMac.h: Added.
        (WebCore::AudioFileReader::data):
        (WebCore::AudioFileReader::dataSize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67530 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Hyung Song <beergun@company100.net>
commit-queue@webkit.org [Wed, 15 Sep 2010 02:31:30 +0000 (02:31 +0000)]
2010-09-14  Hyung Song  <beergun@company100.net>

        Reviewed by Kent Tamura.

        [BREWMP] Add IMemGroup and IMemSpace to OwnPtr type.
        https://bugs.webkit.org/show_bug.cgi?id=44764

        * wtf/OwnPtrCommon.h:
        * wtf/brew/OwnPtrBrew.cpp:
        (WTF::deleteOwnedPtr):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67529 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Wed, 15 Sep 2010 02:04:55 +0000 (02:04 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        bugs.webkit.org should autocomplete WebKit Committers and Reviewers
        https://bugs.webkit.org/show_bug.cgi?id=45799

        This patch implements an autocomplete dropdown that helps users type
        the names of WebKit committers and reviewers.  The script grabs the
        data from SVN now that wms set up CORS for http://svn.webkit.org.
        Security technology for the win.

        This script is based on (well, essentially copied from) the popular
        autocomplete extension:

        https://chrome.google.com/extensions/detail/olaabhcgdogcbcoiolomlcodkngnemfb

        I asked Ojan, and said he was fine licensing the code under the BSD
        license.  I'm not sure how the patch interacts with the extension, but
        we can retire the extension if they interact poorly.

        * committers-autocomplete.js: Added.
        * template/en/default/global/footer.html.tmpl:
            - Add the script to every page on bugs.webkit.org

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67528 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Mihai Parparita <mihaip@chromium.org>
abarth@webkit.org [Wed, 15 Sep 2010 01:59:19 +0000 (01:59 +0000)]
2010-09-14  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by Adam Barth.

        Send webkit accessibility notifications to Chromium
        https://bugs.webkit.org/show_bug.cgi?id=45156

        Move printf inside shouldDumpAccessibilityNotifications check.

        * DumpRenderTree/chromium/WebViewHost.cpp:
        (WebViewHost::postAccessibilityNotification):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67527 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Simon Fraser <simon.fraser@apple.com>
simon.fraser@apple.com [Wed, 15 Sep 2010 01:55:11 +0000 (01:55 +0000)]
2010-09-14  Simon Fraser  <simon.fraser@apple.com>

        Reviewed by Oliver Hunt.

        Drawing an image outside the bounds of a canvas still causes repaints
        https://bugs.webkit.org/show_bug.cgi?id=45792

        Some canvas examples, such as http://www.phoboslab.org/biolab/, draw
        images outide the bounds of the canvas. If the canvas already had a
        non-empty m_dirtyRect, such draws would cause us to do a repaint of
        m_dirtyRect again (which goes through the platform-specific invalidation
        machinery, which can be slow).

        Optimize this by avoiding extra invalidation when the dirty rect
        does not intersect the canvas bounds.

        Not testable because this avoids additional repaints of m_dirtyRect, but
        m_dirtyRect has to be non-empty for this to benefit.

        * html/HTMLCanvasElement.cpp:
        (WebCore::HTMLCanvasElement::didDraw):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67526 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Yury Semikhatsky <yurys@chromium.org>
yurys@chromium.org [Wed, 15 Sep 2010 01:42:54 +0000 (01:42 +0000)]
2010-09-14  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: Links from "Profiles" don't lead to "Scripts"
        https://bugs.webkit.org/show_bug.cgi?id=45756

        * inspector/front-end/ProfileDataGridTree.js:
        (WebInspector.ProfileDataGridNode.prototype.createCell): use standard routine to create resource link
        * inspector/front-end/inspector.js: code cleanup: removed no-op method
        (WebInspector.addMainEventListeners):
2010-09-14  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: Links from "Profiles" don't lead to "Scripts"
        https://bugs.webkit.org/show_bug.cgi?id=45756

        * src/js/DevTools.js: remove obsolete code

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67525 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Tony Chang <tony@chromium.org>
tony@chromium.org [Wed, 15 Sep 2010 01:25:07 +0000 (01:25 +0000)]
2010-09-14  Tony Chang  <tony@chromium.org>

        Reviewed by Kent Tamura.

        [chromium] fix http/tests/security/local-user-CSS-from-remote.html
        https://bugs.webkit.org/show_bug.cgi?id=45788

        * platform/chromium/test_expectations.txt:
2010-09-14  Tony Chang  <tony@chromium.org>

        Reviewed by Kent Tamura.

        [chromium] fix http/tests/security/local-user-CSS-from-remote.html
        https://bugs.webkit.org/show_bug.cgi?id=45788

        * DEPS: Include webkit_support::LocalFileToDataURL
2010-09-14  Tony Chang  <tony@chromium.org>

        Reviewed by Kent Tamura.

        [chromium] fix http/tests/security/local-user-CSS-from-remote.html
        https://bugs.webkit.org/show_bug.cgi?id=45788

        * DumpRenderTree/chromium/LayoutTestController.cpp:
        (LayoutTestController::setUserStyleSheetLocation): use a base64 url like in Chrome

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67524 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Patrick Gansterer <paroga@paroga.com>
commit-queue@webkit.org [Wed, 15 Sep 2010 01:23:17 +0000 (01:23 +0000)]
2010-09-14  Patrick Gansterer  <paroga@paroga.com>

        Reviewed by Adam Roben.

        [WINCE] Use codePage instead of TextEncoding
        https://bugs.webkit.org/show_bug.cgi?id=45466

        Use codePage as member variable to avoid repeated lookup.

        * platform/text/wince/TextCodecWinCE.cpp:
        (WebCore::newTextCodecWinCE):
        (WebCore::TextCodecWinCE::TextCodecWinCE):
        (WebCore::decode):
        (WebCore::TextCodecWinCE::decode):
        (WebCore::TextCodecWinCE::encode):
        * platform/text/wince/TextCodecWinCE.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67523 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Wed, 15 Sep 2010 00:48:36 +0000 (00:48 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Add a license block to the reviewtool.  I should have added this from
        the start.  According to the SVN log, I'm the only one to have ever
        touched this file.

        * code-review.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67522 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Tony Chang <tony@chromium.org>
tony@chromium.org [Wed, 15 Sep 2010 00:45:24 +0000 (00:45 +0000)]
2010-09-14  Tony Chang  <tony@chromium.org>

        Reviewed by Ojan Vafai.

        [chromium] Remove WebKit::areLayoutTestImagesOpaque since it's no longer needed
        https://bugs.webkit.org/show_bug.cgi?id=45768

        * public/WebKit.h:
        * src/WebKit.cpp:
2010-09-14  Tony Chang  <tony@chromium.org>

        Reviewed by Ojan Vafai.

        [chromium] Remove WebKit::areLayoutTestImagesOpaque since it's no longer needed
        https://bugs.webkit.org/show_bug.cgi?id=45768

        * DumpRenderTree/chromium/TestShell.cpp:
        (TestShell::dumpImage): new baselines are checked in and Linux will match Windows going forward.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67521 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Wed, 15 Sep 2010 00:36:39 +0000 (00:36 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Attempted Chromium build fix.  The compiler can't tell that
        ASSERT_NOT_REACHED is not reached and so it demands an initialization
        of this variable.

        * html/HTMLFontElement.cpp:
        (WebCore::HTMLFontElement::parseMappedAttribute):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67520 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Ryuan Choi <ryuan.choi@samsung.com>
commit-queue@webkit.org [Wed, 15 Sep 2010 00:34:36 +0000 (00:34 +0000)]
2010-09-14  Ryuan Choi  <ryuan.choi@samsung.com>

        Reviewed by Kenneth Rohde Christiansen.

        [EFL] ewk_frame_view_create_for_view set null as theme.
        https://bugs.webkit.org/show_bug.cgi?id=45741

        Add theme in ewk_view to use instead of frame's theme which is null in
        ewk_frame_view_create_for_view.

        * ewk/ewk_frame.cpp:
        (ewk_frame_view_create_for_view):
        * ewk/ewk_view.cpp:
        (ewk_view_theme_set):
        (ewk_view_theme_get):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67519 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoPart of <rdar://problem/8420003>. Make it possible to override the version number...
mrowe@apple.com [Wed, 15 Sep 2010 00:19:27 +0000 (00:19 +0000)]
Part of <rdar://problem/8420003>.  Make it possible to override the version number used in linked-in-or-after checks.

Reviewed by John Sullivan.

* Misc/WebKitVersionChecks.h: Remove two unused functions from the header and add a method to set the overriden version.
* Misc/WebKitVersionChecks.m:
(WebKitLinkedOnOrAfter):
(setWebKitLinkTimeVersion): Set the overridden version.
(WebKitLinkTimeVersion): Return the overridden version if set, otherwise do the old thing.
* WebView/WebPreferences.mm:
(+[WebPreferences setWebKitLinkTimeVersion:]): Call through to set the overridden version.
* WebView/WebPreferencesPrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67518 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Mihai Parparita <mihaip@chromium.org>
tony@chromium.org [Wed, 15 Sep 2010 00:19:23 +0000 (00:19 +0000)]
2010-09-14  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by Tony Chang.

        blob: URL scheme does not work with Chromium DRT
        https://bugs.webkit.org/show_bug.cgi?id=45772

        Remove blob-related tests that now pass.

        * platform/chromium/drt_expectations.txt:
2010-09-14  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by Tony Chang.

        blob: URL scheme does not work with Chromium DRT
        https://bugs.webkit.org/show_bug.cgi?id=45772

        Roll Chromium forward to r59406, which registers the blob
        URL scheme for the Chromium DRT too.

        * DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67517 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Andreas Kling <andreas.kling@nokia.com>
andreas.kling@nokia.com [Tue, 14 Sep 2010 23:53:54 +0000 (23:53 +0000)]
2010-09-14  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Ariya Hidayat.

        [Qt] Wheeling over windowed plugins sometimes cause them to disappear

        * plugins/qt/PluginContainerQt.cpp:
        (PluginClientWrapper::x11Event): Pass on captured mouse wheel events
        to the effective native window rather than the grandparent widget.
        We were calling QWidget::winId() which caused a new X11 window to be
        created for the grandparent widget and generally mess things up.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67516 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[wx] Build fixes after recent changes.
kevino@webkit.org [Tue, 14 Sep 2010 23:46:58 +0000 (23:46 +0000)]
[wx] Build fixes after recent changes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67515 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 14 Sep 2010 23:32:26 +0000 (23:32 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Handle <font size=0> as <font size=1> like any other browser
        https://bugs.webkit.org/show_bug.cgi?id=39148

        Update font size parser to match HTML5 spec.  The main difference here
        is the handling of 0, which now maps to 1 instead of 3.

        * html/HTMLFontElement.cpp:
        (WebCore::parseFontSize):
        (WebCore::HTMLFontElement::cssValueFromFontSizeNumber):
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Handle <font size=0> as <font size=1> like any other browser
        https://bugs.webkit.org/show_bug.cgi?id=39148

        Update tests and results to match new behavior.

        * editing/execCommand/query-font-size-expected.txt:
        * fast/dom/HTMLFontElement/script-tests/size-attribute.js:
        * fast/dom/HTMLFontElement/size-attribute-expected.txt:
        * platform/mac/tables/mozilla/bugs/bug16012-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 14 Sep 2010 23:29:38 +0000 (23:29 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        document.lastModified gives no output if the response doesn't have a Last-Modified header
        https://bugs.webkit.org/show_bug.cgi?id=8475

        Implement document.lastModified as per HTML5.

        Tests: http/tests/misc/last-modified-parsing.html
               http/tests/misc/no-last-modified.html

        * dom/Document.cpp:
        (WebCore::Document::lastModified):
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        document.lastModified gives no output if the response doesn't have a Last-Modified header
        https://bugs.webkit.org/show_bug.cgi?id=8475

        * http/tests/misc/last-modified-parsing-expected.txt: Added.
        * http/tests/misc/last-modified-parsing.html: Added.
            - Tests some details of our date parsing and serialization.  The
              serialization format is kind of goofy, but that's what the spec
              seems to want.
        * http/tests/misc/no-last-modified-expected.txt: Added.
        * http/tests/misc/no-last-modified.html: Added.
            - Test was happens if the server doesn't send a Last-Modified
              header.  We're supposed to use the current date and time.
        * http/tests/resources/last-modified.php: Added.
            - PHP helper script.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67513 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 14 Sep 2010 23:27:19 +0000 (23:27 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        `document.compatMode` is `undefined` when using `application/xhtml+xml`
        https://bugs.webkit.org/show_bug.cgi?id=44422

        Moving the compatMode IDL attribute from HTMLDocument.idl to
        Document.idl fixes this bug and improve compliance with HTML5.

        Test: fast/parser/compatMode-in-xhtml.xhtml

        * dom/Document.cpp:
        (WebCore::Document::compatMode):
        * dom/Document.h:
        * dom/Document.idl:
        * html/HTMLDocument.cpp:
        * html/HTMLDocument.h:
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        `document.compatMode` is `undefined` when using `application/xhtml+xml`
        https://bugs.webkit.org/show_bug.cgi?id=44422

        Test the value of compatMode in XHTML.  I'm terrible at XML, but this
        test seems to work.

        * fast/parser/compatMode-in-xhtml-expected.txt: Added.
        * fast/parser/compatMode-in-xhtml.xhtml: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67512 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 James Kozianski <soulplant@gmail.com>
commit-queue@webkit.org [Tue, 14 Sep 2010 23:15:07 +0000 (23:15 +0000)]
2010-09-14  James Kozianski  <soulplant@gmail.com>

        Reviewed by Ojan Vafai.

        Setting of CSSStyleDeclaration.prototype.foo on outer window affects prototype of inner window
        https://bugs.webkit.org/show_bug.cgi?id=18447

        Add a test for CSSStyleDeclaration.prototype.foo leaking into inner
        windows.

        * fast/frames/css-style-declaration-isolation-expected.txt: Added.
        * fast/frames/css-style-declaration-isolation.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67511 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix windows build.
weinig@apple.com [Tue, 14 Sep 2010 23:13:45 +0000 (23:13 +0000)]
Fix windows build.

* win/WebKit2Generated.make:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67510 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 14 Sep 2010 23:12:49 +0000 (23:12 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Nate Chapin.

        V8 security checks don't account for shadowing named frames
        https://bugs.webkit.org/show_bug.cgi?id=45700

        Test: http/tests/security/xss-DENIED-frame-name.html

        * bindings/v8/custom/V8DOMWindowCustom.cpp:
        (WebCore::V8DOMWindow::namedSecurityCheck):
            - If the property name exists on the object, it will shadow the
              named property lookup on the window object.  That means we need
              to block access if there's shadowing going on.
        (WebCore::V8DOMWindow::indexedSecurityCheck):
            - I made the corresponding change to this function too, but I don't
              think this one can actually be triggered because JavaScript
              variable names need to start with a non-digit.
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Nate Chapin.

        V8 security checks don't account for shadowing named frames
        https://bugs.webkit.org/show_bug.cgi?id=45700

        Test whether cross-origin observers can see global variables shadowing
        named frames.

        * http/tests/security/resources/frame-for-parent-name.html: Added.
        * http/tests/security/xss-DENIED-frame-name-expected.txt: Added.
        * http/tests/security/xss-DENIED-frame-name.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoRemove WKBundleNodeRef, replacing uses with WKBundleNodeHandleRef.
weinig@apple.com [Tue, 14 Sep 2010 22:53:31 +0000 (22:53 +0000)]
Remove WKBundleNodeRef, replacing uses with WKBundleNodeHandleRef.
https://bugs.webkit.org/show_bug.cgi?id=45785

Reviewed by Anders Carlsson.

WebKit2:

* WebKit2.xcodeproj/project.pbxproj:
* win/WebKit2.vcproj:
Remove files.

* WebProcess/InjectedBundle/API/c/WKBundleAPICast.h:
* WebProcess/InjectedBundle/API/c/WKBundleBase.h:
* WebProcess/InjectedBundle/API/c/WKBundleNode.cpp: Removed.
* WebProcess/InjectedBundle/API/c/WKBundleNode.h: Removed.
* WebProcess/InjectedBundle/API/c/WKBundlePage.h:
* WebProcess/InjectedBundle/API/c/WKBundleRange.cpp:
(WKBundleRangeCopyStartContainer):
(WKBundleRangeCopyEndContainer):
* WebProcess/InjectedBundle/API/c/WKBundleRange.h:
* WebProcess/InjectedBundle/InjectedBundlePageEditorClient.cpp:
(WebKit::InjectedBundlePageEditorClient::shouldInsertNode):
Change to be in terms of WKBundleNodeHandleRef.

WebKitTools:

* WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
(WTR::propertyValue):
(WTR::numericWindowPropertyValue):
(WTR::dumpPath):
(WTR::toStr):
(WTR::operator<<):
(WTR::InjectedBundlePage::InjectedBundlePage):
(WTR::InjectedBundlePage::shouldInsertNode):
(WTR::InjectedBundlePage::shouldBeginEditing):
(WTR::InjectedBundlePage::shouldEndEditing):
(WTR::InjectedBundlePage::shouldInsertText):
(WTR::InjectedBundlePage::shouldDeleteRange):
(WTR::InjectedBundlePage::shouldChangeSelectedRange):
(WTR::InjectedBundlePage::shouldApplyStyle):
* WebKitTestRunner/InjectedBundle/InjectedBundlePage.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 14 Sep 2010 22:53:02 +0000 (22:53 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        incorrect tabindex parsing
        https://bugs.webkit.org/show_bug.cgi?id=21076

        Updated our integer parsing for tabindex to use the algorithm from the
        HTML5 spec.

        Test: fast/parser/tabindex-parsing-2.html

        * html/HTMLElement.cpp:
        (WebCore::HTMLElement::parseMappedAttribute):
        * html/parser/HTMLParserIdioms.cpp:
        (WebCore::parseHTMLInteger):
        * html/parser/HTMLParserIdioms.h:
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        incorrect tabindex parsing
        https://bugs.webkit.org/show_bug.cgi?id=21076

        * fast/parser/tabindex-parsing-2-expected.txt: Added.
        * fast/parser/tabindex-parsing-2.html: Added.
            - Test a bunch of corner cases in tabindex parsing.
        * fast/parser/tabindex-parsing-expected.txt:
        * fast/parser/tabindex-parsing.html:
            - Update expected result.  This case is actually the one that the
              original reporter was complaining about.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBuild correction, no review.
bfulgham@webkit.org [Tue, 14 Sep 2010 22:48:46 +0000 (22:48 +0000)]
Build correction, no review.

* WebCore.vcproj/WebCore.vcproj: Exclude OwnPtrCairo.cpp from
non-Cairo based builds.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67505 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 14 Sep 2010 22:38:13 +0000 (22:38 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Fix two typos in commit-queue.  The first is harmless.  The second
        causes the queue to reject patches when the build is broken.  :(

        * Scripts/webkitpy/tool/commands/queues.py:
        * Scripts/webkitpy/tool/commands/queues_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67504 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agofast/files/apply-blob-url-to-img.html timeout on Leopard
jianli@chromium.org [Tue, 14 Sep 2010 22:26:27 +0000 (22:26 +0000)]
fast/files/apply-blob-url-to-img.html timeout on Leopard
https://bugs.webkit.org/show_bug.cgi?id=45576

Reviewed by David Levin.

WebCore:

The failure on Leopard is caused by the static method
ResourceHandle::supportsBufferedData that returns true to indicate that
the buffered data support is enabled for blob resource. However,
BlobResourceHandle does not support this and we need to override this.

The fix is to change supportsBufferedData method from static to virtual
so that we can override the default behavior in BlobResourceHandle.

* fileapi/ThreadableBlobRegistry.cpp: Copy BlobData to make it cross-thread.
(WebCore::BlobRegistryContext::BlobRegistryContext):
* loader/ResourceLoader.cpp:
(WebCore::ResourceLoader::resourceData):
(WebCore::ResourceLoader::addData):
* platform/network/BlobResourceHandle.cpp:
(WebCore::BlobResourceHandle::supportsBufferedData): Override the default
behavior.
* platform/network/BlobResourceHandle.h:
* platform/network/ResourceHandle.h: Made supportsBufferedData virtual.
* platform/network/mac/ResourceHandleMac.mm:
(WebCore::ResourceHandle::bufferedData):

WebKit/chromium:

* src/ResourceHandle.cpp: Remove the no longer applicable static comment.

LayoutTests:

Update the test to compute the right coordinates to drag the file to.
This fixes the problem that onInputFileChange() is not called on
Leopard.

* fast/files/apply-blob-url-to-img.html:
* platform/mac/Skipped: Remove the passing test from Skipped list.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67503 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Victor Wang <victorw@chromium.org>
victorw@chromium.org [Tue, 14 Sep 2010 22:19:26 +0000 (22:19 +0000)]
2010-09-14  Victor Wang  <victorw@chromium.org>

        Reviewed by Dimitri Glazkov.

        [Chromium] Remove the wrong chromium-win-vista expectations.

        Layout test results generated from Chromium vista bots match
        the expectations in chromium-win. These baselines are either
        wrong or out of date. Need to be cleaned up so chromium
        vista bots can pass on these tests.

        https://bugs.webkit.org/show_bug.cgi?id=45766

        * platform/chromium-win-vista/css2.1/t0805-c5519-brdr-r-01-e-expected.checksum: Removed.
        * platform/chromium-win-vista/css2.1/t0805-c5519-brdr-r-01-e-expected.png: Removed.
        * platform/chromium-win-vista/css2.1/t0805-c5519-brdr-r-01-e-expected.txt: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c414-flt-fit-01-d-g-expected.checksum: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c414-flt-fit-01-d-g-expected.png: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c414-flt-fit-01-d-g-expected.txt: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c5525-fltblck-00-d-ag-expected.checksum: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c5525-fltblck-00-d-ag-expected.png: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c5525-fltblck-00-d-ag-expected.txt: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c5525-flthw-00-c-g-expected.checksum: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c5525-flthw-00-c-g-expected.png: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c5525-flthw-00-c-g-expected.txt: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c5526-flthw-00-c-g-expected.checksum: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c5526-flthw-00-c-g-expected.png: Removed.
        * platform/chromium-win-vista/css2.1/t0905-c5526-flthw-00-c-g-expected.txt: Removed.
        * platform/chromium-win-vista/css2.1/t1202-counter-09-b-expected.checksum: Removed.
        * platform/chromium-win-vista/css2.1/t1202-counter-09-b-expected.png: Removed.
        * platform/chromium-win-vista/css2.1/t1202-counters-09-b-expected.checksum: Removed.
        * platform/chromium-win-vista/css2.1/t1202-counters-09-b-expected.png: Removed.
        * platform/chromium-win-vista/fast/gradients/radial-centered-expected.checksum: Removed.
        * platform/chromium-win-vista/fast/gradients/radial-centered-expected.png: Removed.
        * platform/chromium-win-vista/fast/gradients/radial-centered-expected.txt: Removed.
        * platform/chromium-win-vista/svg/custom/text-linking-expected.checksum: Removed.
        * platform/chromium-win-vista/svg/custom/text-linking-expected.png: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Jessie Berlin <jberlin@apple.com>
jberlin@webkit.org [Tue, 14 Sep 2010 22:00:54 +0000 (22:00 +0000)]
2010-09-14  Jessie Berlin  <jberlin@apple.com>

        Reviewed by Adam Roben.

        Exceptions are getting ignored on 64-bit Windows in the Web Process
        https://bugs.webkit.org/show_bug.cgi?id=45779

        WebKit2:

        * WebProcess/WebKitMain.cpp:
        (disableProcessCallbackFilterForExceptions):
        Clear the PROCESS_CALLBACK_FILTER_ENABLED flag so that exceptions thrown in callback routines
        are not ignored, as described in http://support.microsoft.com/kb/976038 and
        http://blog.paulbetts.org/index.php/2010/07/20/the-case-of-the-disappearing-onload-exception-user-mode-callback-exceptions-in-x64/.
        (WebKitMain):

        WebKitSite:

        * building/tools.html:
        Add instructions about how to optionally get hotfix 976038.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoNot reviewed.
ggaren@apple.com [Tue, 14 Sep 2010 21:55:23 +0000 (21:55 +0000)]
Not reviewed.

Fixed layout test failure seen on buildbot by correcting expected results.

It looks like Darin accidentally checked in the wrong expected results.

* fast/css/visited-link-hang-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67500 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBuild correction, no review.
bfulgham@webkit.org [Tue, 14 Sep 2010 21:43:54 +0000 (21:43 +0000)]
Build correction, no review.

WebCore:

* WebCore.vcproj/WebCore.vcproj: Add missing OwnPtrCairo.cpp
  implementation file.
* platform/graphics/cairo/OwnPtrCairo.cpp: Add missing include
  for <cairo.h>

WebKit2:

* win/WebKit2.vcproj: Disable build for Cairo targets of
the WebURLRequestWin.cpp and WebURLResponseWin.cpp files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67499 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Chris Guillory <chris.guillory@google.com>
abarth@webkit.org [Tue, 14 Sep 2010 21:29:35 +0000 (21:29 +0000)]
2010-09-14  Chris Guillory   <chris.guillory@google.com>

        Reviewed by Chris Fleizach.

        Send all accessibility notifications to Chromium.
        https://bugs.webkit.org/show_bug.cgi?id=45156

        Use postAccessibilityNotification to pass accessibility notifications
        to chromium.

        * WebKit.gyp:
        * public/WebAccessibilityNotification.h: Added.
        * public/WebViewClient.h:
        (WebKit::WebViewClient::postAccessibilityNotification):
        * src/AssertMatchingEnums.h: Added.
        * src/ChromeClientImpl.cpp:
        (WebKit::toWebAccessibilityNotification):
        (WebKit::ChromeClientImpl::didChangeAccessibilityObjectChildren):
        (WebKit::ChromeClientImpl::postAccessibilityNotification):
        * src/ChromeClientImpl.h:
2010-09-14  Chris Guillory   <chris.guillory@google.com>

        Reviewed by Chris Fleizach.

        Send all accessibility notifications to Chromium.
        https://bugs.webkit.org/show_bug.cgi?id=45156

        Use postAccessibilityNotification to pass accessibility notifications
        to chromium.

        Tests: platform/chromium/accessibility/post-notification-ActiveDescendantChanged.html
               platform/chromium/accessibility/post-notification-CheckedStateChanged.html
               platform/chromium/accessibility/post-notification-ChildrenChanged.html
               platform/chromium/accessibility/post-notification-FocusedUIElementChanged.html
               platform/chromium/accessibility/post-notification-LayoutComplete.html
               platform/chromium/accessibility/post-notification-LiveRegionChanged.html
               platform/chromium/accessibility/post-notification-LoadComplete.html
               platform/chromium/accessibility/post-notification-MenuListValueChanged.html
               platform/chromium/accessibility/post-notification-RowCollapsed.html
               platform/chromium/accessibility/post-notification-RowCountChanged.html
               platform/chromium/accessibility/post-notification-RowExpanded.html
               platform/chromium/accessibility/post-notification-ScrolledToAnchor.html
               platform/chromium/accessibility/post-notification-SelectedChildrenChanged.html
               platform/chromium/accessibility/post-notification-SelectedTextChanged.html
               platform/chromium/accessibility/post-notification-ValueChanged.html

        * accessibility/chromium/AXObjectCacheChromium.cpp:
        (WebCore::AXObjectCache::postPlatformNotification):
        * page/chromium/ChromeClientChromium.h:
2010-09-14  Chris Guillory   <chris.guillory@google.com>

        Reviewed by Chris Fleizach.

        Send all accessibility notifications to Chromium.
        https://bugs.webkit.org/show_bug.cgi?id=45156

        * platform/chromium/accessibility/children-changed-notification-expected.txt: Removed.
        * platform/chromium/accessibility/children-changed-notification.html: Removed.
        * platform/chromium/accessibility/post-notification-ActiveDescendantChanged-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-ActiveDescendantChanged.html: Added.
        * platform/chromium/accessibility/post-notification-CheckedStateChanged-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-CheckedStateChanged.html: Added.
        * platform/chromium/accessibility/post-notification-ChildrenChanged-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-ChildrenChanged.html: Added.
        * platform/chromium/accessibility/post-notification-FocusedUIElementChanged-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-FocusedUIElementChanged.html: Added.
        * platform/chromium/accessibility/post-notification-LayoutComplete-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-LayoutComplete.html: Added.
        * platform/chromium/accessibility/post-notification-LiveRegionChanged-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-LiveRegionChanged.html: Added.
        * platform/chromium/accessibility/post-notification-LoadComplete-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-LoadComplete.html: Added.
        * platform/chromium/accessibility/post-notification-MenuListValueChanged-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-MenuListValueChanged.html: Added.
        * platform/chromium/accessibility/post-notification-RowCollapsed-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-RowCollapsed.html: Added.
        * platform/chromium/accessibility/post-notification-RowCountChanged-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-RowCountChanged.html: Added.
        * platform/chromium/accessibility/post-notification-RowExpanded-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-RowExpanded.html: Added.
        * platform/chromium/accessibility/post-notification-ScrolledToAnchor-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-ScrolledToAnchor.html: Added.
        * platform/chromium/accessibility/post-notification-SelectedChildrenChanged-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-SelectedChildrenChanged.html: Added.
        * platform/chromium/accessibility/post-notification-SelectedTextChanged-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-SelectedTextChanged.html: Added.
        * platform/chromium/accessibility/post-notification-ValueChanged-expected.txt: Added.
        * platform/chromium/accessibility/post-notification-ValueChanged.html: Added.
        * platform/chromium/accessibility/post-notification.js: Added.
        (enumAccessibilityObjects):
        (testDone):
        (test):
        * platform/chromium/drt_expectations.txt:
        * platform/chromium/test_expectations.txt:
2010-09-14  Chris Guillory   <chris.guillory@google.com>

        Reviewed by Chris Fleizach.

        Send all accessibility notifications to Chromium.
        https://bugs.webkit.org/show_bug.cgi?id=45156

        * DumpRenderTree/chromium/WebViewHost.cpp:
        (WebViewHost::postAccessibilityNotification):
        * DumpRenderTree/chromium/WebViewHost.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67498 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago Reviewed by Adam Roben.
adachan@apple.com [Tue, 14 Sep 2010 21:22:59 +0000 (21:22 +0000)]
    Reviewed by Adam Roben.

        Add an IWebFramePrivate API to load string as plain text into the WebFrame.
        https://bugs.webkit.org/show_bug.cgi?id=45782

        * Interfaces/IWebFramePrivate.idl:
        * Interfaces/WebKit.idl: Touch the file.
        * WebFrame.cpp:
        (WebFrame::loadPlainTextString):
        * WebFrame.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67497 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Darin Adler <darin@apple.com>
darin@apple.com [Tue, 14 Sep 2010 21:10:18 +0000 (21:10 +0000)]
2010-09-14  Darin Adler  <darin@apple.com>

        Reviewed by Geoffrey Garen.

        Hang processing href attribute containing a million slashes
        https://bugs.webkit.org/show_bug.cgi?id=45767

        Test: fast/css/visited-link-hang.html

        * platform/LinkHash.cpp:
        (WebCore::findSlashDotDotSlash): Added a start position argument and
        changed types to use size_t consistently instead of a mix.
        (WebCore::findSlashSlash): Ditto.
        (WebCore::findSlashDotSlash): Ditto.
        (WebCore::squeezeOutNullCharacters): Added.
        (WebCore::cleanSlashDotDotSlashes): Added. Factored out part
        of cleanPath (see below).
        (WebCore::mergeDoubleSlashes): Ditto.
        (WebCore::cleanSlashDotSlashes): Ditto.
        (WebCore::cleanPath): Changed algorithm to not remove as we go to
        avoid N^2 behavior; instead replace with null characters and then
        do a squeeze operation after the fact. Also moved the body of the
        function out of line since we normally don't have to do any cleaning.
        This whole thing should go at some point -- it's not the right
        algorithm -- but this should eliminate the performance problems
        without changing behavior.
2010-09-14  Darin Adler  <darin@apple.com>

        Reviewed by Geoffrey Garen.

        Hang processing href attribute containing a million slashes
        https://bugs.webkit.org/show_bug.cgi?id=45767

        * fast/css/visited-link-hang-expected.txt: Added.
        * fast/css/visited-link-hang.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67496 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 14 Sep 2010 21:07:14 +0000 (21:07 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        commit-queue is slow during the day
        https://bugs.webkit.org/show_bug.cgi?id=45780

        Thanks to the new logging, we've noticed that checkout-is-out-of-date
        errors in the first pass of landing don't retry the land.  Instead,
        they're treated as failures and cause the commit-queue to do two more
        builds before really trying to land the patch.  Worse, in the second
        build, we can get bitten by a flaky test.

        This patch takes a slightly different approach to the commit-queue's
        main control logic.  We now use a separate subprocess for building and
        testing and for landing.  This means we should very rarely see the
        checkout-is-out-of-date error, and when we do see it, we should retry
        more quickly.  If my understanding is correct, this should be a big
        speed win for the commit-queue.

        * Scripts/webkitpy/tool/commands/download.py:
        * Scripts/webkitpy/tool/commands/queues.py:
        * Scripts/webkitpy/tool/commands/queues_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67495 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Darin Adler <darin@apple.com>
darin@apple.com [Tue, 14 Sep 2010 21:06:01 +0000 (21:06 +0000)]
2010-09-14  Darin Adler  <darin@apple.com>

        Reviewed by Geoffrey Garen.

        Sort with non-numeric custom sort function fails on array with length but no values
        https://bugs.webkit.org/show_bug.cgi?id=45781

        * runtime/JSArray.cpp:
        (JSC::JSArray::sort): Replaced early exit for an array of length zero to instead
        exit for any array without values, even if it has a non-0 length.
2010-09-14  Darin Adler  <darin@apple.com>

        Reviewed by Geoffrey Garen.

        Sort with non-numeric custom sort function fails on array with length but no values
        https://bugs.webkit.org/show_bug.cgi?id=45781

        * fast/js/script-tests/sort-large-array.js: Added test cases.
        * fast/js/sort-large-array-expected.txt: Updated.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67494 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Nico Weber <thakis@chromium.org>
thakis@chromium.org [Tue, 14 Sep 2010 21:02:06 +0000 (21:02 +0000)]
2010-09-14  Nico Weber  <thakis@chromium.org>

        Reviewed by Kenneth Russell.

        [chromium]: On Mac, let image layer data row order match skia.
        https://bugs.webkit.org/show_bug.cgi?id=45400

        Make CoreGraphics texture row order match skia's row order, remove all
        code that worked around this difference. This also fixes a problem where
        image layers would be drawn upside down.

        Covered by existing layout tests, but we don't run these with
        compositing yet.

        * platform/graphics/chromium/ContentLayerChromium.cpp:
        (WebCore::ContentLayerChromium::SharedValues::SharedValues):
        (WebCore::ContentLayerChromium::updateContents):
        (WebCore::ContentLayerChromium::updateTextureRect):
        * platform/graphics/chromium/LayerRendererChromium.cpp:
        (WebCore::LayerRendererChromium::setRootLayerCanvasSize):
        (WebCore::LayerRendererChromium::drawLayers):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 14 Sep 2010 20:29:00 +0000 (20:29 +0000)]
2010-09-14  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Tony Chang.

        applyInlineStyleToNodeRange does not extend run properly
        https://bugs.webkit.org/show_bug.cgi?id=45616

        The bug was caused by applyInlineStyleToNodeRange's skipping a node whenever it has children.
        Fixed the bug by only skipping a node that contains the past end node, contains non-editable region,
        or has a non-editable parent node. Also fixed a bug that applyInlineStyleToNodeRange does not remove
        styled elements in a run that are identical to the one added. addInlineStyleIfNeeded now removes
        all styled elements in a run using removeStyleFromRunBeforeApplyingStyle before calling addInlineStyleIfNeeded.

        Test: editing/style/inline-style-extend-run.html

        * editing/ApplyStyleCommand.cpp:
        (WebCore::containsNonEditableRegion): Added.
        (WebCore::ApplyStyleCommand::applyInlineStyleToNodeRange): See above.
        (WebCore::ApplyStyleCommand::removeStyleFromRunBeforeApplyingStyle): Added, see above.
        (WebCore::ApplyStyleCommand::removeInlineStyleFromElement): Added support for RemoveAlways.
        (WebCore::ApplyStyleCommand::removeImplicitlyStyledElement): Replaced RemoveAttributesAndElements by RemoveIfNeeded.
        (WebCore::ApplyStyleCommand::extractInlineStyleToPushDown): Replaced RemoveAttributesAndElements by RemoveIfNeeded.
        * editing/ApplyStyleCommand.h:
2010-09-14  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Tony Chang.

        applyInlineStyleToNodeRange does not extend a run properly
        https://bugs.webkit.org/show_bug.cgi?id=45616

        Added a test to ensure WebKit adds just one element when applying inline style and removes redundant styled elements.
        Rebaseline of other tests are required because WebKit now generates less elements.

        * editing/execCommand/script-tests/toggle-style-3.js: Merged two b elements.
        * editing/execCommand/toggle-style-3-expected.txt: Ditto.
        * editing/execCommand/unlink-expected.txt: Changed the order in which span and anchor appear.
        * editing/style/inline-style-extend-run-expected.txt: Added.
        * editing/style/inline-style-extend-run.html: Added.
        * editing/style/script-tests/inline-style-extend-run.js: Added.
        (testSingleToggle):
        * editing/style/script-tests/make-text-writing-direction-inline.js: Merged two spans.
        * fast/events/event-input-contentEditable-expected.txt: Merged two anchor elements.
        * fast/events/event-input-contentEditable.html: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67490 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Andreas Kling <andreas.kling@nokia.com>
andreas.kling@nokia.com [Tue, 14 Sep 2010 20:08:11 +0000 (20:08 +0000)]
2010-09-14  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] V8 port: Get inspector up and running
        https://bugs.webkit.org/show_bug.cgi?id=45771

        * WebCoreSupport/InspectorClientQt.cpp:
        (WebCore::ensureDebuggerScriptLoaded): Added, loads DebuggerScript.js into ScriptDebugServer.
        (WebCore::InspectorClientQt::openInspectorFrontend): Ensure that DebuggerScript.js is loaded
        before opening an inspector.
2010-09-14  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Kenneth Rohde Christiansen.

        [Qt] V8 port: Get inspector up and running
        https://bugs.webkit.org/show_bug.cgi?id=45771

        * inspector/front-end/WebKit.qrc: Add DebuggerScript.js

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67488 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Tony Chang <tony@chromium.org>
tony@chromium.org [Tue, 14 Sep 2010 19:56:59 +0000 (19:56 +0000)]
2010-09-14  Tony Chang  <tony@chromium.org>

        Unreviewed, fix chromium win compile.

        * public/WebView.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67487 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoPPC Build fix.
oliver@apple.com [Tue, 14 Sep 2010 19:48:05 +0000 (19:48 +0000)]
PPC Build fix.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67486 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Tue, 14 Sep 2010 19:44:13 +0000 (19:44 +0000)]
2010-09-14  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Andreas Kling.

        [chromium] Remove unused ICU headers
        https://bugs.webkit.org/show_bug.cgi?id=45698

        Remove unused headers.

        * platform/graphics/chromium/FontCacheLinux.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67485 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 14 Sep 2010 19:23:25 +0000 (19:23 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Eric Seidel.

        HTML5 spec violation: </><!doctype html> or </x><!doctype html> SHOULD NOT imply standards mode
        https://bugs.webkit.org/show_bug.cgi?id=17722

        I believe our behavior matches the HTML5 spec.  When we see a </x>, we
        correctly kick out of standards mode.  In the </> case, however, the
        tokenizer just eats the bytes, without emitting a token, which means we
        don't get kicked out of standards mode.  Our behavior matches Minefield.

        * fast/parser/strange-doctype-compatmode-1-expected.txt: Added.
        * fast/parser/strange-doctype-compatmode-1.html: Added.
        * fast/parser/strange-doctype-compatmode-2-expected.txt: Added.
        * fast/parser/strange-doctype-compatmode-2.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67484 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Andreas Kling <andreas.kling@nokia.com>
andreas.kling@nokia.com [Tue, 14 Sep 2010 19:02:57 +0000 (19:02 +0000)]
2010-09-14  Andreas Kling  <andreas.kling@nokia.com>

        Reviewed by Antonio Gomes.

        [Qt] Numpad Enter key returns keyCode 0 instead of 13
        https://bugs.webkit.org/show_bug.cgi?id=45014

        Original patch by Jarred Nicholls <jarred.nicholls@gmail.com>

        There is a missing case statement in the PlatformKeyboardEventQt.cpp
        file to handle the Enter button on the number pad.  Because it is not
        handled, the keyCode returned in a JavaScript event is 0 rather than 13.

        Test: manual-tests/qt/numpad-enter-key.html

        * manual-tests/qt/numpad-enter-key.html: Added.
        * platform/qt/PlatformKeyboardEventQt.cpp:
        (WebCore::windowsKeyCodeForKeyEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67483 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoWindows production build fix.
sfalken@apple.com [Tue, 14 Sep 2010 18:49:25 +0000 (18:49 +0000)]
Windows production build fix.
Roll out r65143.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.make:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67482 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Evan Martin <evan@chromium.org>
evan@chromium.org [Tue, 14 Sep 2010 18:15:52 +0000 (18:15 +0000)]
2010-09-14  Evan Martin  <evan@chromium.org>

        Reviewed by Darin Fisher.

        [chromium] building without debug syms broke after recent refactoring
        https://bugs.webkit.org/show_bug.cgi?id=45720

        Now that we build WebCore as multiple libraries, we need to adjust the
        flags to turn off debug symbols for all of these libraries.

        No new tests; should only affect an uncommon build flag.

        * WebCore.gyp/WebCore.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67481 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Tony Chang <tony@chromium.org>
tony@chromium.org [Tue, 14 Sep 2010 17:59:49 +0000 (17:59 +0000)]
2010-09-14  Tony Chang  <tony@chromium.org>

        Unreviewed, remove tests that are now passing after Noel Gordon's
        fix to test_shell delete key handling.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67480 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Tony Chang <tony@chromium.org>
tony@chromium.org [Tue, 14 Sep 2010 17:38:17 +0000 (17:38 +0000)]
2010-09-14  Tony Chang  <tony@chromium.org>

        Reviewed by Dimitri Glazkov.

        [chromium] implement layoutTestController.setDomainRelaxationForbiddenForURLScheme
        https://bugs.webkit.org/show_bug.cgi?id=45762

        * platform/chromium/test_expectations.txt:
2010-09-14  Tony Chang  <tony@chromium.org>

        Reviewed by Dimitri Glazkov.

        [chromium] implement layoutTestController.setDomainRelaxationForbiddenForURLScheme
        https://bugs.webkit.org/show_bug.cgi?id=45762

        * public/WebView.h: add WebView::setDomainRelaxationForbidden
        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::setDomainRelaxationForbidden):
        * src/WebViewImpl.h:
2010-09-14  Tony Chang  <tony@chromium.org>

        Reviewed by Dimitri Glazkov.

        [chromium] implement layoutTestController.setDomainRelaxationForbiddenForURLScheme
        https://bugs.webkit.org/show_bug.cgi?id=45762

        * DumpRenderTree/chromium/LayoutTestController.cpp:
        (LayoutTestController::LayoutTestController):
        (LayoutTestController::setDomainRelaxationForbiddenForURLScheme):
        * DumpRenderTree/chromium/LayoutTestController.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67479 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Tony Chang <tony@chromium.org>
tony@chromium.org [Tue, 14 Sep 2010 17:27:51 +0000 (17:27 +0000)]
2010-09-14  Tony Chang  <tony@chromium.org>

        Unreviewed, fix chromium compile.

        * platform/graphics/chromium/ImageLayerChromium.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67478 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Nico Weber <thakis@chromium.org>
thakis@chromium.org [Tue, 14 Sep 2010 17:16:32 +0000 (17:16 +0000)]
2010-09-14  Nico Weber  <thakis@chromium.org>

        Reviewed by Darin Fisher.

        [chromium]: crash less in compositor mode on mac
        https://bugs.webkit.org/show_bug.cgi?id=45714

        The contents image can go away when the dom is updated. Since the
        layer tree is updated asynchronously, it needs to ref the native image,
        else it will hold on to a stale object.

        This still needs to be done for skia, too.

        * platform/graphics/chromium/ImageLayerChromium.h:
        * platform/graphics/chromium/ImageLayerChromium.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67477 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Leandro Pereira <leandro@profusion.mobi>
leandro@webkit.org [Tue, 14 Sep 2010 17:04:20 +0000 (17:04 +0000)]
2010-09-14  Leandro Pereira  <leandro@profusion.mobi>

        [EFL] Unreviewed build fix: remove "virtual" keyword from
        implementation file.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67476 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Tue, 14 Sep 2010 16:49:55 +0000 (16:49 +0000)]
2010-09-14  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Darin Adler.

        Share UnicodeMacrosFromICU.h
        https://bugs.webkit.org/show_bug.cgi?id=45710

        glib, qt4 and wince use the same macros from ICU.
        Remove the code duplication and use the same header file.

        * wtf/unicode/UnicodeMacrosFromICU.h: Copied from JavaScriptCore/wtf/unicode/glib/UnicodeMacrosFromICU.h.
        * wtf/unicode/glib/UnicodeMacrosFromICU.h: Removed.
        * wtf/unicode/qt4/UnicodeQt4.h:
        * wtf/unicode/wince/UnicodeWince.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67475 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Andrey Kosyakov <caseq@chromium.org>
caseq@chromium.org [Tue, 14 Sep 2010 16:30:45 +0000 (16:30 +0000)]
2010-09-14  Andrey Kosyakov  <caseq@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: [Resources panel] [HAR] Need a way to save timing data.
        Added support for resource timings to HAREntry.
        https://bugs.webkit.org/show_bug.cgi?id=45663

        * inspector/front-end/HAREntry.js:
        (WebInspector.HAREntry.prototype._buildTimings):
        (WebInspector.HAREntry.prototype._toMilliseconds):
        (WebInspector.HAREntry.prototype._interval):

2010-09-14  Andrey Kosyakov  <caseq@chromium.org>

        Reviewed by Yury Semikhatsky.

        Web Inspector: [Resources panel] [HAR] Need a way to save timing data.
        https://bugs.webkit.org/show_bug.cgi?id=45663

        * http/tests/inspector/resource-har-conversion-expected.txt:
        * http/tests/inspector/resource-parameters-expected.txt:
        * http/tests/inspector/resource-test2.js: Added entire timings object to list of non-deterministic fields.
        * http/tests/inspector/resource-tests.js: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Tue, 14 Sep 2010 16:19:32 +0000 (16:19 +0000)]
2010-09-14  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Kent Tamura.

        [BREWMP] Always set LeftButton in PlatformMouseEvent
        https://bugs.webkit.org/show_bug.cgi?id=45705

        AEE_POINTER_GET_MOUSE_MODIFIERS always returns 0, so it is impossible to know
        which button is pressed or released. The current code erroneously sets NoButton.
        Set LeftButton unconditionally because Brew MP usually runs on touch device.

        * platform/brew/PlatformMouseEventBrew.cpp:
        (WebCore::PlatformMouseEvent::PlatformMouseEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67473 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Yury Semikhatsky <yurys@chromium.org>
yurys@chromium.org [Tue, 14 Sep 2010 15:40:50 +0000 (15:40 +0000)]
2010-09-14  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Andreas Kling.

        Move DebuggerScript.js from WebKit/chromium/src/js to WebCore/bindings/v8
        https://bugs.webkit.org/show_bug.cgi?id=45739

        * bindings/v8/DebuggerScript.js: Added.
2010-09-14  Yury Semikhatsky  <yurys@chromium.org>

        Reviewed by Andreas Kling.

        Move DebuggerScript.js from WebKit/chromium/src/js to WebCore/bindings/v8
        https://bugs.webkit.org/show_bug.cgi?id=45739

        * WebKit.grd:
        * WebKit.gypi:
        * src/js/DebuggerScript.js: Removed.
        * src/js/DevToolsHostStub.js: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67472 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Hans Wennborg <hans@chromium.org>
hans@chromium.org [Tue, 14 Sep 2010 15:31:06 +0000 (15:31 +0000)]
2010-09-14  Hans Wennborg  <hans@chromium.org>

        Reviewed by Jeremy Orlow.

        Rebaseline DeviceOrientation related layout tests for Chromium.
        https://bugs.webkit.org/show_bug.cgi?id=45542

        Some DeviceOrientation layout tests need Chromium-specific
        baselines because of differences between JSC and V8.

        Also enable prototype-inheritance.html again and unfork its
        expectations.

        * fast/dom/script-tests/prototype-inheritance.js:
        * platform/chromium-linux/fast/dom/DeviceOrientation/window-property-expected.txt: Added.
        * platform/chromium-linux/fast/dom/Window/window-properties-device-orientation-expected.txt: Added.
        * platform/chromium-mac/fast/dom/DeviceOrientation/window-property-expected.txt: Added.
        * platform/chromium-mac/fast/dom/Window/window-properties-device-orientation-expected.txt: Added.
        * platform/chromium-mac/fast/dom/prototype-inheritance-expected.txt: Removed.
        * platform/chromium-win/fast/dom/DeviceOrientation/window-property-expected.txt: Added.
        * platform/chromium-win/fast/dom/Window/window-properties-device-orientation-expected.txt: Added.
        * platform/chromium-win/fast/dom/prototype-inheritance-expected.txt: Removed.
        * platform/chromium/fast/dom/prototype-inheritance-expected.txt:
        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67471 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Pavel Podivilov <podivilov@chromium.org>
yurys@chromium.org [Tue, 14 Sep 2010 15:18:53 +0000 (15:18 +0000)]
2010-09-14  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Yury Semikhatsky.

        [Chromium] add new tests from r67463 to expectations
        https://bugs.webkit.org/show_bug.cgi?id=45753

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67470 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-10 Alexander Pavlov <apavlov@chromium.org>
apavlov@chromium.org [Tue, 14 Sep 2010 14:59:20 +0000 (14:59 +0000)]
2010-09-10  Alexander Pavlov  <apavlov@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: Implement on-demand reporting of empty CSS rules matched for a node by WebCore
        https://bugs.webkit.org/show_bug.cgi?id=45530

        WebCore:
        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::matchRules):
        (WebCore::CSSStyleSelector::matchRulesForList):
        (WebCore::CSSStyleSelector::matchUARules):
        (WebCore::CSSStyleSelector::styleForElement):
        (WebCore::CSSStyleSelector::pseudoStyleForElement):
        (WebCore::CSSStyleSelector::styleRulesForElement):
        (WebCore::CSSStyleSelector::pseudoStyleRulesForElement):
        * css/CSSStyleSelector.h:
        * inspector/InspectorDOMAgent.cpp:
        (WebCore::InspectorDOMAgent::getStyles):
        (WebCore::InspectorDOMAgent::buildArrayForPseudoElements):

        LayoutTests:
        * inspector/elements-panel-styles-expected.txt:
        * inspector/resources/elements-panel-styles.css:
        (.foo):
        (.foo::before):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67469 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Hyung Song <beergun@company100.net>
commit-queue@webkit.org [Tue, 14 Sep 2010 14:42:41 +0000 (14:42 +0000)]
2010-09-14  Hyung Song  <beergun@company100.net>

        Reviewed by Darin Adler.

        [BREWMP] Set m_size in SharedBuffer::createWithContentsOfFile();
        https://bugs.webkit.org/show_bug.cgi?id=45701

        SharedBuffer::createWithContentsOfFile() reads file content and saves to
        result.m_buffer directly, leaving result.m_size unchanged.

        * platform/brew/SharedBufferBrew.cpp:
        (WebCore::SharedBuffer::createWithContentsOfFile):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Adam Barth <abarth@webkit.org>
abarth@webkit.org [Tue, 14 Sep 2010 14:25:38 +0000 (14:25 +0000)]
2010-09-14  Adam Barth  <abarth@webkit.org>

        Reviewed by Darin Adler.

        Move adjustLexerState to the HTMLTokenizer
        https://bugs.webkit.org/show_bug.cgi?id=45649

        This function used let us share code between the tree builder and the
        two other users of the HTMLTokenizer.  However, now that the tree
        builder is all grown up, it can't use this code anymore.  The other
        users would much rather this code lived on HTMLTokenizer and understood
        <script>.  This patch attempts to make them happy.

        * html/parser/HTMLPreloadScanner.cpp:
        (WebCore::HTMLPreloadScanner::processToken):
        * html/parser/HTMLTokenizer.cpp:
        (WebCore::HTMLTokenizer::updateStateFor):
        * html/parser/HTMLTokenizer.h:
        * html/parser/HTMLTreeBuilder.cpp:
        (WebCore::HTMLTreeBuilder::scriptEnabled):
        * html/parser/HTMLTreeBuilder.h:
        * html/parser/HTMLViewSourceParser.cpp:
        (WebCore::HTMLViewSourceParser::updateTokenizerState):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Mihai Parparita <mihaip@chromium.org>
commit-queue@webkit.org [Tue, 14 Sep 2010 14:06:00 +0000 (14:06 +0000)]
2010-09-14  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by Dimitri Glazkov.

        post-basic.html and post-frames.html in http/tests/navigation should not be pixel tests
        https://bugs.webkit.org/show_bug.cgi?id=45610

        post-basic.html and post-frames.html would be a lot simpler if
        they weren't pixel tests (post-frames.html doesn't even have the right
        expectations for Chromium/Mac). Since all they test is that form
        POSTs are handled correctly, they can just use dumpAsText.

        * http/tests/navigation/post-basic-expected.txt: Added.
        * http/tests/navigation/post-basic.html:
        * http/tests/navigation/post-frames-expected.txt: Added.
        * http/tests/navigation/post-frames.html:
        * http/tests/navigation/resources/form-target.pl: Added
        * http/tests/navigation/resources/page-that-posts.html: Added.
        * platform/chromium-linux/http/tests/navigation/post-basic-expected.checksum: Removed.
        * platform/chromium-linux/http/tests/navigation/post-basic-expected.png: Removed.
        * platform/chromium-linux/http/tests/navigation/post-frames-expected.checksum: Removed.
        * platform/chromium-linux/http/tests/navigation/post-frames-expected.png: Removed.
        * platform/chromium-win/http/tests/navigation/post-basic-expected.checksum: Removed.
        * platform/chromium-win/http/tests/navigation/post-basic-expected.png: Removed.
        * platform/chromium-win/http/tests/navigation/post-basic-expected.txt: Removed.
        * platform/chromium-win/http/tests/navigation/post-frames-expected.checksum: Removed.
        * platform/chromium-win/http/tests/navigation/post-frames-expected.png: Removed.
        * platform/chromium-win/http/tests/navigation/post-frames-expected.txt: Removed.
        * platform/chromium/test_expectations.txt:
        * platform/mac/http/tests/navigation/post-basic-expected.checksum: Removed.
        * platform/mac/http/tests/navigation/post-basic-expected.png: Removed.
        * platform/mac/http/tests/navigation/post-basic-expected.txt: Removed.
        * platform/mac/http/tests/navigation/post-frames-expected.checksum: Removed.
        * platform/mac/http/tests/navigation/post-frames-expected.png: Removed.
        * platform/mac/http/tests/navigation/post-frames-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Andras Becsi <abecsi@webkit.org>
abecsi@webkit.org [Tue, 14 Sep 2010 14:04:56 +0000 (14:04 +0000)]
2010-09-14  Andras Becsi  <abecsi@webkit.org>

        Reviewed by Csaba Osztrogon√°c.

        [Qt] qmake doesn't generate correct dependencies for rcc resource files
        https://bugs.webkit.org/show_bug.cgi?id=45747

        * WebKit.pri: add OUT_PWD to DEPENDPATH

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Anton Muhin <antonm@chromium.org>
antonm@chromium.org [Tue, 14 Sep 2010 13:33:43 +0000 (13:33 +0000)]
2010-09-14  Anton Muhin  <antonm@chromium.org>

        Unreviewed cleanup of test expectations for Chromium.

        ietestcenter/Javascript/15.4.4.15-3-14.html, ietestcenter/Javascript/15.4.4.15-3-29.html and
        ietestcenter/Javascript/15.4.4.15-3-8.html are passing now.

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67464 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Pierre-Antoine LaFayette <plafayet@codeaurora.org>
commit-queue@webkit.org [Tue, 14 Sep 2010 13:25:06 +0000 (13:25 +0000)]
2010-09-14  Pierre-Antoine LaFayette  <plafayet@codeaurora.org>

        Reviewed by Darin Adler.

        REGRESSION (r52839): Incomplete repaint of IMG with text-align:center
        https://bugs.webkit.org/show_bug.cgi?id=40142

        Add pixel tests to verify proper repaint of RenderReplaced as well as non-inline children.

        * fast/repaint/block-layout-inline-children-float-positioned.html: Added.
        * fast/repaint/block-layout-inline-children-replaced.html: Added.
        * platform/mac/fast/repaint/block-layout-inline-children-float-positioned-expected.checksum: Added.
        * platform/mac/fast/repaint/block-layout-inline-children-float-positioned-expected.png: Added.
        * platform/mac/fast/repaint/block-layout-inline-children-float-positioned-expected.txt: Added.
        * platform/mac/fast/repaint/block-layout-inline-children-replaced-expected.checksum: Added.
        * platform/mac/fast/repaint/block-layout-inline-children-replaced-expected.png: Added.
        * platform/mac/fast/repaint/block-layout-inline-children-replaced-expected.txt: Added.
2010-09-14  Pierre-Antoine LaFayette  <plafayet@codeaurora.org>

        Reviewed by Darin Adler.

        REGRESSION (r52839): Incomplete repaint of IMG with text-align:center
        https://bugs.webkit.org/show_bug.cgi?id=40142

        http://trac.webkit.org/changeset/52839 didn't really deal with the inline replaced element case.  This just generalizes to all inline content.

        Patch by: Dave Hyatt

        Tests: fast/repaint/block-layout-inline-children-float-positioned.html
               fast/repaint/block-layout-inline-children-replaced.html

        * rendering/RenderBlockLineLayout.cpp:
        (WebCore::RenderBlock::layoutInlineChildren):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67463 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Hans Wennborg <hans@chromium.org>
hans@chromium.org [Tue, 14 Sep 2010 13:20:18 +0000 (13:20 +0000)]
2010-09-14  Hans Wennborg  <hans@chromium.org>

        Reviewed by Jeremy Orlow.

        Make WebDeviceOrientationClientMock constructor private.
        https://bugs.webkit.org/show_bug.cgi?id=45743

        Clients should use the create member function instead. DumpRenderTree
        and the Chromium test_shell have previously been updated to call create.

        * public/WebDeviceOrientationClientMock.h:
        (WebKit::WebDeviceOrientationClientMock::WebDeviceOrientationClientMock):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67462 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Balazs Kelemen <kb@inf.u-szeged.hu>
abecsi@webkit.org [Tue, 14 Sep 2010 13:16:42 +0000 (13:16 +0000)]
2010-09-14  Balazs Kelemen  <kb@inf.u-szeged.hu>

        Reviewed by Andreas Kling.

        [Qt] Cookie operations should use NetworkingContext
        https://bugs.webkit.org/show_bug.cgi?id=45745
        No new functionality so no new tests.

        * platform/qt/CookieJarQt.cpp:
        (WebCore::cookieJar): Use the NetworkingContext for obtaining the QNetworkAccessManager
        instead of directly accessing the FrameLoaderClient.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Pavel Podivilov <podivilov@chromium.org>
caseq@chromium.org [Tue, 14 Sep 2010 12:38:19 +0000 (12:38 +0000)]
2010-09-14  Pavel Podivilov  <podivilov@chromium.org>

        Reviewed by Pavel Feldman.

        [Chromium] mark fast/frames/frame-limit.html as DEBUG SLOW CRASH
        https://bugs.webkit.org/show_bug.cgi?id=45738

        * platform/chromium/test_expectations.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67460 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Sheriff Bot <webkit.review.bot@gmail.com>
commit-queue@webkit.org [Tue, 14 Sep 2010 12:26:36 +0000 (12:26 +0000)]
2010-09-14  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r67449.
        http://trac.webkit.org/changeset/67449
        https://bugs.webkit.org/show_bug.cgi?id=45746

        This change makes tree red. (Requested by pfeldman on
        #webkit).

        * editing/execCommand/script-tests/toggle-style-3.js:
        * editing/execCommand/toggle-style-3-expected.txt:
        * editing/execCommand/unlink-expected.txt:
        * editing/style/inline-style-extend-run-expected.txt: Removed.
        * editing/style/inline-style-extend-run.html: Removed.
        * editing/style/script-tests/inline-style-extend-run.js: Removed.
        * editing/style/script-tests/make-text-writing-direction-inline.js:
2010-09-14  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r67449.
        http://trac.webkit.org/changeset/67449
        https://bugs.webkit.org/show_bug.cgi?id=45746

        This change makes tree red. (Requested by pfeldman on
        #webkit).

        * editing/ApplyStyleCommand.cpp:
        (WebCore::ApplyStyleCommand::applyInlineStyleToNodeRange):
        (WebCore::ApplyStyleCommand::removeInlineStyleFromElement):
        (WebCore::ApplyStyleCommand::removeImplicitlyStyledElement):
        (WebCore::ApplyStyleCommand::extractInlineStyleToPushDown):
        * editing/ApplyStyleCommand.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Eric Seidel <eric@webkit.org>
eric@webkit.org [Tue, 14 Sep 2010 11:07:30 +0000 (11:07 +0000)]
2010-09-14  Eric Seidel  <eric@webkit.org>

        Unreviewed, reverting changes r67451 and r67451.
        Broke lots of builders.

        Only intercept ESC key press when autocorrection UI is visible.
        https://bugs.webkit.org/show_bug.cgi?id=45071

        * editing/Editor.cpp:
        * editing/Editor.h:
        * editing/EditorCommand.cpp:
        (WebCore::createCommandMap):
        * loader/EmptyClients.h:
        * page/EditorClient.h:
2010-09-14  Eric Seidel  <eric@webkit.org>

        Unreviewed, reverting changes r67451 and r67451.
        Broke lots of builders.

        Only intercept ESC key press when autocorrection UI is visible.
        https://bugs.webkit.org/show_bug.cgi?id=45071

        * WebCoreSupport/WebEditorClient.h:
        * WebCoreSupport/WebEditorClient.mm:
        (WebEditorClient::WebEditorClient):
        (WebEditorClient::dismissCorrectionPanel):
2010-09-14  Eric Seidel  <eric@webkit.org>

        Unreviewed, reverting changes r67451 and r67451.
        Broke lots of builders.

        Only intercept ESC key press when autocorrection UI is visible.
        https://bugs.webkit.org/show_bug.cgi?id=45071

        * WebProcess/WebCoreSupport/WebEditorClient.cpp:
        (WebKit::WebEditorClient::dismissCorrectionPanel):
        * WebProcess/WebCoreSupport/WebEditorClient.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67458 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoAdding myself as a reviewer.
ossy@webkit.org [Tue, 14 Sep 2010 09:52:40 +0000 (09:52 +0000)]
Adding myself as a reviewer.

* Scripts/webkitpy/common/config/committers.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago[Qt] Build-webkit shows a warning when WebKitBuild directory is not exist
zoltan@webkit.org [Tue, 14 Sep 2010 09:46:50 +0000 (09:46 +0000)]
[Qt] Build-webkit shows a warning when WebKitBuild directory is not exist
https://bugs.webkit.org/show_bug.cgi?id=45736

Reviewed by Kenneth Rohde Christiansen.

Build-webkit has showed a warning when WebKitBuild didn't exist, add
an additional condition to avoid it.

* Scripts/build-webkit:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Kinuko Yasuda <kinuko@chromium.org>
kinuko@chromium.org [Tue, 14 Sep 2010 09:23:52 +0000 (09:23 +0000)]
2010-09-14  Kinuko Yasuda  <kinuko@chromium.org>

        Unreviewed; an attempt to fix Chromium Windows build.

        * fileapi/FileEntry.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67455 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Andrey Kosyakov <caseq@chromium.org>
caseq@chromium.org [Tue, 14 Sep 2010 09:10:20 +0000 (09:10 +0000)]
2010-09-13  Andrey Kosyakov  <caseq@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: provide support for opening external navigation for links within certain elements
        Use default handling for clicks on links that have target="_blank"
        Add option for external navigation to URL audit formatter.
        https://bugs.webkit.org/show_bug.cgi?id=45671

        * inspector/front-end/AuditFormatters.js:
        (WebInspector.AuditFormatters.url):
        * inspector/front-end/inspector.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Enrico Ros <eros@codeaurora.org>
andreas.kling@nokia.com [Tue, 14 Sep 2010 08:26:15 +0000 (08:26 +0000)]
2010-09-14  Enrico Ros  <eros@codeaurora.org>

        Reviewed by Andreas Kling.

        [Qt] Respect title attribute on option elements
        https://bugs.webkit.org/show_bug.cgi?id=45084

        Set the tooltip in the combo box model, so it's reflected in the view.

        * WebCoreSupport/QtFallbackWebPopup.cpp:
        (WebCore::QtFallbackWebPopup::populate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67453 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoBuild fix.
mitz@apple.com [Tue, 14 Sep 2010 08:03:21 +0000 (08:03 +0000)]
Build fix.

* editing/EditorCommand.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67452 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoOnly intercept ESC key press when autocorrection UI is visible.
mitz@apple.com [Tue, 14 Sep 2010 07:50:32 +0000 (07:50 +0000)]
Only intercept ESC key press when autocorrection UI is visible.
https://bugs.webkit.org/show_bug.cgi?id=45071

Patch by Jia Pu <jpu@apple.com> on 2010-09-14
Reviewed by Dan Bernstein.

WebCore:

Only intercept ESC key when autocorrection panel is shown. Otherwise pressing
ESC will cancel current loading like usual.

* editing/Editor.cpp:
(WebCore::Editor::isShowingCorrectionPanel): Query editor client whether autocorrection
  panel is shown.

* editing/Editor.h: Added declaration of isShowingCorrectionPanel().

* editing/EditorCommand.cpp:
(WebCore::supportedDismissCorrectionPanel): Only support executeCancelOperation()
  when autocorrection panel is shown.
(WebCore::createCommandMap): Replaced isSupported function pointer for
  executeCancelOperation() command with supportedDismissCorrectionPanel().

* loader/EmptyClients.h:
(WebCore::EmptyEditorClient::isShowingCorrectionPanel): Dummy implementation of
  new member method declared in base class.

* page/EditorClient.h: Declared new member method isShowingCorrectionPanel().

WebKit/mac:

* WebCoreSupport/WebEditorClient.h: Added declaration of isShowingCorrectionPanel(),
  which provides an inteface to query whether autocorrection panel is shown.

* WebCoreSupport/WebEditorClient.mm:
(WebEditorClient::WebEditorClient): Defined a constant, InvalidCorrectionPanelTag,
  for invalid correction panel tag. Replaced -1 with this constant.
(WebEditorClient::dismissCorrectionPanel): Ditto
(WebEditorClient::isShowingCorrectionPanel): Query whether autocorrection panel is shown.

WebKit2:

* WebProcess/WebCoreSupport/WebEditorClient.cpp:
(WebKit::WebEditorClient::isShowingCorrectionPanel): Dummy implementation.

* WebProcess/WebCoreSupport/WebEditorClient.h: Adopt new method delcared in base class.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67451 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Tue, 14 Sep 2010 07:40:08 +0000 (07:40 +0000)]
2010-09-14  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Kent Tamura.

        [BREWMP] Remove TextBreakIteratorInternalICUBrew.cpp
        https://bugs.webkit.org/show_bug.cgi?id=45691

        Brew MP no longer depends on ICU for internationalization.
        Remove TextBreakIteratorInternalICUBrew.cpp.

        * platform/text/brew/TextBreakIteratorInternalICUBrew.cpp: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-14 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 14 Sep 2010 07:12:00 +0000 (07:12 +0000)]
2010-09-14  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Tony Chang.

        applyInlineStyleToNodeRange does not extend run properly
        https://bugs.webkit.org/show_bug.cgi?id=45616

        The bug was caused by applyInlineStyleToNodeRange's skipping a node whenever it has children.
        Fixed the bug by only skipping a node that contains the past end node, contains non-editable region,
        or has a non-editable parent node. Also fixed a bug that applyInlineStyleToNodeRange does not remove
        styled elements in a run that are identical to the one added. addInlineStyleIfNeeded now removes
        all styled elements in a run using removeStyleFromRunBeforeApplyingStyle before calling addInlineStyleIfNeeded.

        Test: editing/style/inline-style-extend-run.html

        * editing/ApplyStyleCommand.cpp:
        (WebCore::containsNonEditableRegion): Added.
        (WebCore::ApplyStyleCommand::applyInlineStyleToNodeRange): See above.
        (WebCore::ApplyStyleCommand::removeStyleFromRunBeforeApplyingStyle): Added, see above.
        (WebCore::ApplyStyleCommand::removeInlineStyleFromElement): Added support for RemoveAlways.
        (WebCore::ApplyStyleCommand::removeImplicitlyStyledElement): Replaced RemoveAttributesAndElements by RemoveIfNeeded.
        (WebCore::ApplyStyleCommand::extractInlineStyleToPushDown): Replaced RemoveAttributesAndElements by RemoveIfNeeded.
        * editing/ApplyStyleCommand.h:
2010-09-14  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Tony Chang.

        applyInlineStyleToNodeRange does not extend a run properly
        https://bugs.webkit.org/show_bug.cgi?id=45616

        Added a test to ensure WebKit adds just one element when applying inline style and removes redundant styled elements.
        Rebaseline of other tests are required because WebKit now generates less elements.

        * editing/execCommand/script-tests/toggle-style-3.js: Merged two b elements.
        * editing/execCommand/toggle-style-3-expected.txt: Ditto.
        * editing/execCommand/unlink-expected.txt: Changed the order in which span and anchor appear.
        * editing/style/inline-style-extend-run-expected.txt: Added.
        * editing/style/inline-style-extend-run.html: Added.
        * editing/style/script-tests/inline-style-extend-run.js: Added.
        (testSingleToggle):
        * editing/style/script-tests/make-text-writing-direction-inline.js: Merged two spans.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67449 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Yuta Kitamura <yutak@chromium.org>
tkent@chromium.org [Tue, 14 Sep 2010 06:51:34 +0000 (06:51 +0000)]
2010-09-13  Yuta Kitamura  <yutak@chromium.org>

        Reviewed by Kent Tamura.

        Add one #if guard block to fix Qt builds.

        Web Inspector: WebSocket in Resources tab
        https://bugs.webkit.org/show_bug.cgi?id=40768

        * inspector/InspectorResource.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67448 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Yuta Kitamura <yutak@chromium.org>
commit-queue@webkit.org [Tue, 14 Sep 2010 06:27:38 +0000 (06:27 +0000)]
2010-09-13  Yuta Kitamura  <yutak@chromium.org>

        Reviewed by Pavel Feldman.

        Web Inspector: WebSocket in Resources tab
        https://bugs.webkit.org/show_bug.cgi?id=40768

        * inspector/extensions-api-expected.txt: Added a new resource type.
2010-09-13  Yuta Kitamura  <yutak@chromium.org>

        Reviewed by Pavel Feldman.

        Add WebSocket resource type to Web Inspector.

        When a new WebSocket connection is established, a line for that connection
        will appear in Web Inspector's Resources tab. If the resource name is
        clicked, the details of handshake request and response will be shown.

        Web Inspector: WebSocket in Resources tab
        https://bugs.webkit.org/show_bug.cgi?id=40768

        * inspector/InspectorController.cpp:
        (WebCore::InspectorController::addResource): WebSocket resource does not
        have an associated loader, thus frame might be null. Need to check it.
        (WebCore::InspectorController::removeResource): Ditto.
        (WebCore::InspectorController::didCreateWebSocket):
        (WebCore::InspectorController::willSendWebSocketHandshakeRequest):
        (WebCore::InspectorController::didReceiveWebSocketHandshakeResponse):
        (WebCore::InspectorController::didCloseWebSocket):
        * inspector/InspectorController.h:
        * inspector/InspectorResource.cpp: Add null checks of m_loader and m_frame,
        because WebSocket does not have a loader and we need to allow null for
        these variables.
        (WebCore::createReadableStringFromBinary):
        (WebCore::InspectorResource::InspectorResource):
        (WebCore::InspectorResource::create): Factory function of
        regular (non-WebSocket) resources.
        (WebCore::InspectorResource::createWebSocket): Factory function of
        WebSocket resources.
        (WebCore::InspectorResource::updateWebSocketRequest):
        (WebCore::InspectorResource::updateWebSocketResponse):
        (WebCore::InspectorResource::updateScriptObject): m_frame->document() becomes
        available when Frame::setDocument() is called. We cannot obtain documentURL
        during the constructor or updateRequest() function, because m_frame->document()
        is not available yet at that point and documentURL will contain a wrong URL.
        As far as I know, updateScriptObject() is the only place where we can safely
        obtain documentURL.
        (WebCore::InspectorResource::cachedResource):
        (WebCore::InspectorResource::type):
        (WebCore::InspectorResource::resourceData):
        * inspector/InspectorResource.h:
        (WebCore::InspectorResource::):
        (WebCore::InspectorResource::markWebSocket):
        * inspector/front-end/Resource.js:
        (WebInspector.Resource.Type.toString):
        (WebInspector.Resource.prototype.set type):
        (WebInspector.Resource.prototype._mimeTypeIsConsistentWithType):
        * inspector/front-end/ResourceView.js:
        (WebInspector.ResourceView.prototype._refreshRequestHeaders):
        (WebInspector.ResourceView.prototype._refreshResponseHeaders):
        (WebInspector.ResourceView.prototype._refreshHeaders):
        * inspector/front-end/inspector.css:
        (.resources-category-websockets, .resources-category-other):
        (.resources-category-websockets .resources-graph-bar):
        (.resources-category-websockets.resource-cached .resources-graph-bar):
        * inspector/front-end/inspector.js:
        (WebInspector.loaded):
        (WebInspector.updateResource):
        * websockets/WebSocketChannel.cpp:
        (WebCore::WebSocketChannel::WebSocketChannel):
        (WebCore::WebSocketChannel::disconnect):
        (WebCore::WebSocketChannel::didOpen):
        (WebCore::WebSocketChannel::didClose):
        (WebCore::WebSocketChannel::processBuffer):
        (WebCore::WebSocketChannel::identifier):
        * websockets/WebSocketChannel.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67447 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Eric Uhrhane <ericu@chromium.org>
commit-queue@webkit.org [Tue, 14 Sep 2010 06:06:59 +0000 (06:06 +0000)]
2010-09-13  Eric Uhrhane  <ericu@chromium.org>

        Reviewed by Darin Fisher.

        Add Chromium API for FileWriter
        https://bugs.webkit.org/show_bug.cgi?id=44360

        Build file changes.
        * WebKit.gyp:

        Added a way to get a WebFileWriter.
        * public/WebFileSystem.h:
        (WebKit::WebFileSystem::createFileWriter):

        WebFileWriter writes and truncates files.
        * public/WebFileWriter.h: Added.

        WebFileWriterClient reports success/failure and progress events.
        * public/WebFileWriterClient.h: Added.

        AsyncFileWriterChromium is a connector that links FileWriter and WebFileWriter, FileWriterClient and WebFileWriterClient.
        * src/AsyncFileWriterChromium.cpp: Added.
        * src/AsyncFileWriterChromium.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67446 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 W. James MacLean <wjmaclean@google.com>
commit-queue@webkit.org [Tue, 14 Sep 2010 05:26:24 +0000 (05:26 +0000)]
2010-09-13  W. James MacLean  <wjmaclean@google.com>

        Reviewed by Darin Fisher.

        [chromium] Thumbnails not generated for GPU Rendered Pages
        https://bugs.webkit.org/show_bug.cgi?id=44127

        Replicates existing functionality, use existing tests.
        Adds pixel-readback for GPU composited pages to allow for thumbnailing,
        printing and other services to work with GPU rendered pages.

        * platform/graphics/chromium/LayerRendererChromium.cpp:
        (WebCore::LayerRendererChromium::getFramebufferPixels):
        * platform/graphics/chromium/LayerRendererChromium.h:
        (WebCore::LayerRendererChromium::rootLayerTextureSize):
2010-09-13  W. James MacLean  <wjmaclean@google.com>

        Reviewed by Darin Fisher.

        [chromium] Thumbnails not generated for GPU Rendered Pages
        https://bugs.webkit.org/show_bug.cgi?id=44127

        Modified WebViewImpl::paint() to detect non-null canvas pointers when
        accelerated compositing is active, and instead fills the pixel buffer
        from the GPU framebuffer. Includes re-scaling support when provided
        canvas does not match size of current render layer. Limits pixel
        readback to rect passed to paint(), clipped by size of rootLayerTexture.

        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::doPixelReadbackToCanvas):
        (WebKit::WebViewImpl::paint):
        * src/WebViewImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67445 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Kinuko Yasuda <kinuko@chromium.org>
kinuko@chromium.org [Tue, 14 Sep 2010 05:05:28 +0000 (05:05 +0000)]
2010-09-13  Kinuko Yasuda  <kinuko@chromium.org>

        Unreviewed, build fix for Gtk.

        * bindings/js/JSEntryCustom.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67444 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Kinuko Yasuda <kinuko@chromium.org>
kinuko@chromium.org [Tue, 14 Sep 2010 04:59:23 +0000 (04:59 +0000)]
2010-09-13  Kinuko Yasuda  <kinuko@chromium.org>

        Reviewed by Adam Barth.

        Make FileEntry and DirectoryEntry methods accessible via Entry instance
        https://bugs.webkit.org/show_bug.cgi?id=45553

        FileEntry- and DirectoryEntry- specific methods must be available on Entry object based on its internal type.

        Added custom toJS/toV8 implementation to Entry interface.

        No new tests; test will be added later when we have complete implementation for the API.

        * CMakeLists.txt:
        * GNUmakefile.am:
        * WebCore.gypi:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * bindings/js/JSEntryCustom.cpp: Added.
        (WebCore::toJS):
        * bindings/v8/custom/V8EntryCustom.cpp: Added.
        (WebCore::toV8):
        * fileapi/DirectoryEntry.h: Nits fix.
        * fileapi/DirectoryEntry.idl: Added ConvertUndefinedOrNullToString to path parameter.
        * fileapi/Entry.idl: Added CustomToJS attribute.
        * fileapi/FileEntry.cpp:
        (WebCore::FileEntry::createWriter):
        (WebCore::FileEntry::file):
        * fileapi/FileEntry.h:
        * fileapi/FileEntry.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67443 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Daniel Bates <dbates@rim.com>
dbates@webkit.org [Tue, 14 Sep 2010 04:29:58 +0000 (04:29 +0000)]
2010-09-13  Daniel Bates  <dbates@rim.com>

        https://bugs.webkit.org/show_bug.cgi?id=45732

        Fix the Qt build. I missed this when reviewing the patch
        for Bug #45732.

        * WebCoreSupport/DumpRenderTreeSupportQt.cpp:
        (DumpRenderTreeSupportQt::suspendActiveDOMObjects): Change enum value ActiveDOMObject::JavaScriptPaused to
        ActiveDOMObject::JavaScriptDebuggerPaused

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67442 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Ryosuke Niwa <rniwa@webkit.org>
rniwa@webkit.org [Tue, 14 Sep 2010 04:18:26 +0000 (04:18 +0000)]
2010-09-13  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Darin Adler.

        REGRESSION: In Gmail, a crash occurs at getDoubleValue() when applying a text color to a new line
        https://bugs.webkit.org/show_bug.cgi?id=45632

        Test: editing/execCommand/query-font-size-with-typing-style.html

        The crash was caused by selectionStartCSSPropertyValue's deleting nodeToRemove before
        retrieving the font-size property. Fixed the bug by moving the removal code to the end of the function.

        * editing/Editor.cpp:
        (WebCore::Editor::selectionStartCSSPropertyValue):
2010-09-13  Ryosuke Niwa  <rniwa@webkit.org>

        Reviewed by Darin Adler.

        REGRESSION: In Gmail, a crash occurs at getDoubleValue() when applying a text color to a new line
        https://bugs.webkit.org/show_bug.cgi?id=45632

        Added a test to ensure WebKit does not crash when querying font size even if there is a typing style.

        * editing/execCommand/query-font-size-with-typing-style-expected.txt: Added.
        * editing/execCommand/query-font-size-with-typing-style.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67441 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Tue, 14 Sep 2010 04:17:01 +0000 (04:17 +0000)]
2010-09-13  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Daniel Bates.

        [Qt] DumpRenderTreeSupportQt::suspendActiveDOMObjects needs a new parameter
        https://bugs.webkit.org/show_bug.cgi?id=45732

        * WebCoreSupport/DumpRenderTreeSupportQt.cpp:
        (DumpRenderTreeSupportQt::suspendActiveDOMObjects): Pass ActiveDOMObject::JavaScriptPaused
        to suspendActiveDOMObjects.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67440 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 James Robinson <jamesr@chromium.org>
jamesr@google.com [Tue, 14 Sep 2010 04:08:51 +0000 (04:08 +0000)]
2010-09-13  James Robinson  <jamesr@chromium.org>

        Reviewed by Dimitri Glazkov.

        [chromium] Return NULL from getSharedGraphicsContext3D if we couldn't initialize
        https://bugs.webkit.org/show_bug.cgi?id=45731

        If we can't create a GraphicsContext3D, return NULL from getSharedGraphicsContext3D rather
        than an object with NULL internal pointers.  Callers know how to handle a null return value
        but not one that has inconsistent internal state.

        * src/WebViewImpl.cpp:
        (WebKit::WebViewImpl::getSharedGraphicsContext3D):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67439 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Kwang Yul Seo <skyul@company100.net>
commit-queue@webkit.org [Tue, 14 Sep 2010 04:00:11 +0000 (04:00 +0000)]
2010-09-13  Kwang Yul Seo  <skyul@company100.net>

        Reviewed by Darin Adler.

        [BREWMP] Remove WebCore namespace in SSLKeyGeneratorBrew.cpp
        https://bugs.webkit.org/show_bug.cgi?id=45690

        WebCore namespace is not required because getSupportedKeySizes and signedPublicKeyAndChallengeString are already
        in namespace WebCore block.

        * platform/brew/SSLKeyGeneratorBrew.cpp:
        (WebCore::getSupportedKeySizes):
        (WebCore::signedPublicKeyAndChallengeString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67438 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoURL in address bar doesn't update when navigating to http://webkit.org/new-bug when...
weinig@apple.com [Tue, 14 Sep 2010 03:40:59 +0000 (03:40 +0000)]
URL in address bar doesn't update when navigating to webkit.org/new-bug when using WebKit2
<rdar://problem/8272775>
https://bugs.webkit.org/show_bug.cgi?id=45729

Reviewed by Jon Honeycutt.

Pass an updated provisional url while notifying the UIProcess of
didReceiveServerRedirectForProvisionalLoadForFrame.

* UIProcess/WebFrameProxy.cpp:
(WebKit::WebFrameProxy::didReceiveServerRedirectForProvisionalLoad):
* UIProcess/WebFrameProxy.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::didReceiveMessage):
(WebKit::WebPageProxy::didReceiveServerRedirectForProvisionalLoadForFrame):
* UIProcess/WebPageProxy.h:
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDidReceiveServerRedirectForProvisionalLoad):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67437 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix windows build.
weinig@apple.com [Tue, 14 Sep 2010 03:23:10 +0000 (03:23 +0000)]
Fix windows build.

* UIProcess/Launcher/win/ProcessLauncherWin.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67436 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Mario Sanchez Prada <msanchez@igalia.com>
commit-queue@webkit.org [Tue, 14 Sep 2010 03:15:39 +0000 (03:15 +0000)]
2010-09-13  Mario Sanchez Prada  <msanchez@igalia.com>

        Reviewed by Martin Robinson.

        [GTK] Code simplification needed in Atk Wrapper
        https://bugs.webkit.org/show_bug.cgi?id=45687

        Use TextIterator::rangeLength() to calculate endOffset

        * accessibility/gtk/AccessibilityObjectWrapperAtk.cpp:
        (getSelectionOffsetsForObject): Don't manually iterate over a
        artificial new range but just call TextIterator::rangeLength on
        the range created between the start and end positions.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67435 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoClean up xcode project.
weinig@apple.com [Tue, 14 Sep 2010 02:54:44 +0000 (02:54 +0000)]
Clean up xcode project.

Reviewed by Alice Liu.

* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67434 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoThreaded mode should work in WebKit2 MiniBrowser
weinig@apple.com [Tue, 14 Sep 2010 02:48:48 +0000 (02:48 +0000)]
Threaded mode should work in WebKit2 MiniBrowser
https://bugs.webkit.org/show_bug.cgi?id=45727

Reviewed by Jon Honeycutt.

- Make starting the web thread more like starting a new
  process (encapsulate it in a thread launcher class).
- Make the call to didFinishLaunching asynchronous so that
  WebProcess is fully constructed before it is called.

* Platform/WorkItem.h:
(MemberFunctionWorkItem1::MemberFunctionWorkItem1):
(MemberFunctionWorkItem1::~MemberFunctionWorkItem1):
(MemberFunctionWorkItem1::execute):
(WorkItem::create):
* UIProcess/Launcher/ProcessLauncher.h:
* UIProcess/Launcher/ThreadLauncher.cpp: Added.
(WebKit::ThreadLauncher::ThreadLauncher):
(WebKit::ThreadLauncher::launchThread):
(WebKit::ThreadLauncher::didFinishLaunchingThread):
(WebKit::ThreadLauncher::invalidate):
* UIProcess/Launcher/ThreadLauncher.h: Added.
(WebKit::ThreadLauncher::Client::~Client):
(WebKit::ThreadLauncher::create):
(WebKit::ThreadLauncher::isLaunching):
* UIProcess/Launcher/mac/ProcessLauncherMac.mm:
(WebKit::ProcessLauncher::terminateProcess):
* UIProcess/Launcher/mac/ThreadLauncherMac.mm: Added.
(WebKit::webThreadBody):
(WebKit::ThreadLauncher::createWebThread):
* UIProcess/Launcher/qt/ProcessLauncherQt.cpp:
* UIProcess/Launcher/qt/ThreadLauncherQt.cpp: Added.
(WebKit::webThreadBody):
(WebKit::ProcessLauncher::createWebThread):
* UIProcess/Launcher/win/ProcessLauncherWin.cpp:
* UIProcess/Launcher/win/ThreadLauncherWin.cpp: Added.
(WebKit::webThreadBody):
(WebKit::ProcessLauncher::createWebThread):
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::~WebProcessProxy):
(WebKit::WebProcessProxy::connect):
(WebKit::WebProcessProxy::isLaunching):
(WebKit::WebProcessProxy::didFinishLaunching):
* UIProcess/WebProcessProxy.h:
* WebKit2.pro:
* WebKit2.xcodeproj/project.pbxproj:
* win/WebKit2.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67433 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Eric Carlson <eric.carlson@apple.com>
eric.carlson@apple.com [Tue, 14 Sep 2010 02:41:56 +0000 (02:41 +0000)]
2010-09-13  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Geoff Garen.

        JavaScript dialog should not deactivate media element
        https://bugs.webkit.org/show_bug.cgi?id=45688

        Test: http/tests/media/reload-after-dialog.html

        * bindings/generic/ActiveDOMCallback.cpp:
        (WebCore::ActiveDOMObjectCallbackImpl::suspend): Add ReasonForSuspension parameter.

        * bindings/js/ScriptDebugServer.cpp:
        (WebCore::ScriptDebugServer::setJavaScriptPaused): Pass new parameter to suspendActiveDOMObjects.

        * dom/ActiveDOMObject.cpp:
        (WebCore::ActiveDOMObject::suspend): Add ReasonForSuspension parameter.
        * dom/ActiveDOMObject.h:

        * dom/ScriptExecutionContext.cpp:
        (WebCore::ScriptExecutionContext::suspendActiveDOMObjects): Ditto.
        * dom/ScriptExecutionContext.h:

        * history/CachedFrame.cpp:
        (WebCore::CachedFrame::CachedFrame): Pass new parameter to suspendActiveDOMObjects.

        * html/HTMLMarqueeElement.cpp:
        (WebCore::HTMLMarqueeElement::suspend): Add ReasonForSuspension parameter.
        * html/HTMLMarqueeElement.h:

        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::stop): Deal with element being suspended.
        (WebCore::HTMLMediaElement::suspend): Do nothing unless reason for suspension is document
        becoming inactive.
        * html/HTMLMediaElement.h:

        * page/PageGroupLoadDeferrer.cpp:
        (WebCore::PageGroupLoadDeferrer::PageGroupLoadDeferrer): Pass WillShowDialog to suspendActiveDOMObjects.

        * page/SuspendableTimer.cpp:
        (WebCore::SuspendableTimer::suspend): Add ReasonForSuspension parameter.
        * page/SuspendableTimer.h:

        * websockets/WebSocket.cpp:
        (WebCore::WebSocket::suspend): Ditto.
        * websockets/WebSocket.h:

        * xml/XMLHttpRequest.cpp:
        (WebCore::XMLHttpRequest::suspend): Ditto.
        * xml/XMLHttpRequest.h:

2010-09-13  Eric Carlson  <eric.carlson@apple.com>

        Reviewed by Geoff Garen.

        JavaScript dialog should not deactivate media element
        https://bugs.webkit.org/show_bug.cgi?id=45688

        * http/tests/media/reload-after-dialog-expected.txt: Added.
        * http/tests/media/reload-after-dialog.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67432 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoFix the mac build.
weinig@apple.com [Tue, 14 Sep 2010 02:15:12 +0000 (02:15 +0000)]
Fix the mac build.

* page/mac/EventHandlerMac.mm:
(WebCore::EventHandler::passMousePressEventToSubframe):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67431 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years agoCan't scroll on Bugzilla patch review pages on Mac with WebKit2
weinig@apple.com [Tue, 14 Sep 2010 02:03:28 +0000 (02:03 +0000)]
Can't scroll on Bugzilla patch review pages on Mac with WebKit2
https://bugs.webkit.org/show_bug.cgi?id=43555

Reviewed by Anders Carlsson.

<rdar://problem/8421386> WebKit2: gmail.com is completely unresponsive to clicks
<rdar://problem/8276298> Can't scroll on Bugzilla patch review pages on Mac with WebKit2 (43555)

* page/mac/EventHandlerMac.mm:
(WebCore::EventHandler::passWheelEventToWidget):
(WebCore::frameHasPlatformWidget):
(WebCore::EventHandler::passMousePressEventToSubframe):
(WebCore::EventHandler::passMouseMoveEventToSubframe):
(WebCore::EventHandler::passMouseReleaseEventToSubframe):
Add WebKit2 mode for these functions based on the presence of a platform widget for a
FrameView.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67430 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Eric Seidel <eric@webkit.org>
eric@webkit.org [Tue, 14 Sep 2010 01:38:50 +0000 (01:38 +0000)]
2010-09-13  Eric Seidel  <eric@webkit.org>

        Reviewed by Adam Barth.

        commit-queue should log when it's retrying due to checkout out of date error
        https://bugs.webkit.org/show_bug.cgi?id=45725

        This will allow us to better track how often we see these errors
        and how long it takes the queue to recover from them.

        In order to pass the tests, I had to correct an error
        from a previous change.  The UnitTestPort should not have
        ever been using the base port method as that will read
        a file from disk.  If the unit tests haven't passed the contents
        it should raise an IOError as though the file doesn't exist.

        * Scripts/webkitpy/layout_tests/port/base_unittest.py:
        * Scripts/webkitpy/tool/commands/queues.py:
        * Scripts/webkitpy/tool/commands/queues_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67429 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Mihai Parparita <mihaip@chromium.org>
commit-queue@webkit.org [Tue, 14 Sep 2010 01:37:22 +0000 (01:37 +0000)]
2010-09-13  Mihai Parparita  <mihaip@chromium.org>

        Reviewed by Adam Barth.

        Upload results when cq patch landing fails
        https://bugs.webkit.org/show_bug.cgi?id=45683

        Include script errors when CommitQueue._land fails.

        * Scripts/webkitpy/tool/commands/queues.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67428 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11 years ago2010-09-13 Patrick Gansterer <paroga@paroga.com>
commit-queue@webkit.org [Tue, 14 Sep 2010 01:10:21 +0000 (01:10 +0000)]
2010-09-13  Patrick Gansterer  <paroga@paroga.com>

        Reviewed by Kenneth Rohde Christiansen.

        [WINCE] Buildfix for ViewportArguments.cpp
        https://bugs.webkit.org/show_bug.cgi?id=45678

        Use fabs() instead of abs().
        This patch also fixes the return value, when
        the value is between -1 and 1.

        * dom/ViewportArguments.cpp:
        (WebCore::findUserScalableValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@67427 268f45cc-cd09-0410-ab3c-d52691b4dbfc