WebKit-https.git
7 years ago[Qt][WK2] Fix build after removal of MessageID.h
rafael.lobo@openbossa.org [Thu, 31 Jan 2013 23:26:01 +0000 (23:26 +0000)]
[Qt][WK2] Fix build after removal of MessageID.h
https://bugs.webkit.org/show_bug.cgi?id=108534

Reviewed by Anders Carlsson.

* Platform/CoreIPC/unix/ConnectionUnix.cpp:
(CoreIPC::MessageInfo::MessageInfo):
(CoreIPC::MessageInfo::setMessageBodyIsOutOfLine):
(CoreIPC::MessageInfo::isMessageBodyIsOutOfLine):
(MessageInfo):
(CoreIPC::Connection::processMessage):
(CoreIPC::Connection::sendOutgoingMessage):
* Target.pri:
* UIProcess/CoordinatedGraphics/LayerTreeRenderer.cpp:
* UIProcess/DrawingAreaProxy.cpp:
* WebProcess/WebPage/CoordinatedGraphics/CoordinatedLayerTreeHost.cpp:
* WebProcess/soup/WebSoupRequestManager.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141495 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoBegin to make XSSAuditor thread aware
tonyg@chromium.org [Thu, 31 Jan 2013 23:18:35 +0000 (23:18 +0000)]
Begin to make XSSAuditor thread aware
https://bugs.webkit.org/show_bug.cgi?id=108394

Reviewed by Adam Barth.

This patch moves the parts of filterToken() that depend on Frame, Document, etc. to a delegate class.
The new didBlockScript delegate method is invoked with the resulting DidBlockScriptRequest which will
be owned by the CompactHTMLToken in the threaded case.

This is just the first step as there is more to do to make XSSAuditor thread safe.

No new tests because covered by existing tests.

* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.vcxproj/WebCore.vcxproj:
* WebCore.vcxproj/WebCore.vcxproj.filters:
* WebCore.xcodeproj/project.pbxproj:
* html/parser/BackgroundHTMLParser.cpp:
(WebCore::BackgroundHTMLParser::pumpTokenizer):
* html/parser/HTMLDocumentParser.cpp:
(WebCore::HTMLDocumentParser::HTMLDocumentParser):
(WebCore::HTMLDocumentParser::processTokensFromBackgroundParser):
(WebCore::HTMLDocumentParser::pumpTokenizer):
* html/parser/HTMLDocumentParser.h:
(HTMLDocumentParser):
* html/parser/XSSAuditor.cpp:
(WebCore::XSSAuditor::filterToken):
* html/parser/XSSAuditor.h:
(WebCore):
(XSSAuditor):
* html/parser/XSSAuditorDelegate.cpp: Added.
(WebCore):
(WebCore::XSSAuditorDelegate::XSSAuditorDelegate):
(WebCore::XSSAuditorDelegate::didBlockScript):
* html/parser/XSSAuditorDelegate.h: Added.
(WebCore):
(DidBlockScriptRequest):
(WebCore::DidBlockScriptRequest::create):
(WebCore::DidBlockScriptRequest::DidBlockScriptRequest):
(XSSAuditorDelegate):

* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.vcxproj/WebCore.vcxproj:
* WebCore.vcxproj/WebCore.vcxproj.filters:
* WebCore.xcodeproj/project.pbxproj:
* html/parser/BackgroundHTMLParser.cpp:
(WebCore::BackgroundHTMLParser::pumpTokenizer):
* html/parser/HTMLDocumentParser.cpp:
(WebCore::HTMLDocumentParser::HTMLDocumentParser):
(WebCore::HTMLDocumentParser::processParsedChunkFromBackgroundParser):
(WebCore::HTMLDocumentParser::pumpTokenizer):
* html/parser/HTMLDocumentParser.h:
(HTMLDocumentParser):
* html/parser/XSSAuditor.cpp:
(WebCore::XSSAuditor::XSSAuditor):
(WebCore::XSSAuditor::filterToken):
* html/parser/XSSAuditor.h:
(WebCore):
(XSSAuditor):
* html/parser/XSSAuditorDelegate.cpp: Added.
(WebCore):
(WebCore::XSSAuditorDelegate::XSSAuditorDelegate):
(WebCore::XSSAuditorDelegate::didBlockScript):
* html/parser/XSSAuditorDelegate.h: Added.
(WebCore):
(DidBlockScriptRequest):
(WebCore::DidBlockScriptRequest::create):
(WebCore::DidBlockScriptRequest::DidBlockScriptRequest):
(XSSAuditorDelegate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141494 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAssertion failure in WebResourceLoadScheduler::remove when loading .webarchives
beidson@apple.com [Thu, 31 Jan 2013 23:13:16 +0000 (23:13 +0000)]
Assertion failure in WebResourceLoadScheduler::remove when loading .webarchives
<rdar://problem/12888145> and https://bugs.webkit.org/show_bug.cgi?id=108520

Reviewed by Alexey Proskuryakov.

* WebProcess/Network/WebResourceLoadScheduler.cpp:
(WebKit::WebResourceLoadScheduler::scheduleLoad): Even if it isn't to be scheduled with the
  NetworkProcess, still add this ResourceLoader to the scheduler's records.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCSS3's vh attribute is not adjusting while browser resizes
commit-queue@webkit.org [Thu, 31 Jan 2013 23:12:05 +0000 (23:12 +0000)]
CSS3's vh attribute is not adjusting while browser resizes
https://bugs.webkit.org/show_bug.cgi?id=86418

Patch by Uday Kiran <udaykiran@motorola.com> on 2013-01-31
Reviewed by Antti Koivisto.

Source/WebCore:

Elements with viewport percentage units needs layout whenever
viewport size changes. Check for viewport percentage unit
was missing for height.

Test: css3/viewport-percentage-lengths/vh-resize.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::updateBlockChildDirtyBitsBeforeLayout):
* rendering/RenderBox.cpp:
(WebCore::RenderBox::hasViewportPercentageLogicalHeight): Check if style
is specified in viewport percentage units.
(WebCore):
* rendering/RenderBox.h:
(RenderBox):
* rendering/RenderView.cpp:
(WebCore::RenderView::layout): Elements with viewport percentage units
needs relayout when viewport size changes.

LayoutTests:

Added test to check element with vh units gets resized when
viewport height is increased or decreased.

* css3/viewport-percentage-lengths/vh-resize-expected.html: Added.
* css3/viewport-percentage-lengths/vh-resize.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141492 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoDumpRenderTree should put NSSound calls inside USE(APPKIT)
psolanki@apple.com [Thu, 31 Jan 2013 22:56:47 +0000 (22:56 +0000)]
DumpRenderTree should put NSSound calls inside USE(APPKIT)
https://bugs.webkit.org/show_bug.cgi?id=108499

Reviewed by Andy Estes.

* DumpRenderTree/mac/DumpRenderTree.mm:
(dumpRenderTree):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141491 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoObjective-C API: JSContext exception property causes reference cycle
mhahnenberg@apple.com [Thu, 31 Jan 2013 22:33:55 +0000 (22:33 +0000)]
Objective-C API: JSContext exception property causes reference cycle
https://bugs.webkit.org/show_bug.cgi?id=107778

Reviewed by Darin Adler.

JSContext has a (retain) JSValue * exception property which, when non-null, creates a
reference cycle (since the JSValue * holds a strong reference back to the JSContext *).

* API/JSContext.mm: Instead of JSValue *, we now use a plain JSValueRef, which eliminates the reference cycle.
(-[JSContext initWithVirtualMachine:]):
(-[JSContext setException:]):
(-[JSContext exception]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141490 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoTextAutosizing: refactor the cluster related method parameters into a single struct.
commit-queue@webkit.org [Thu, 31 Jan 2013 22:31:45 +0000 (22:31 +0000)]
TextAutosizing: refactor the cluster related method parameters into a single struct.
https://bugs.webkit.org/show_bug.cgi?id=108384

This anticipates adding more cluster related information in the following patches.

Patch by Anton Vayvod <avayvod@chromium.org> on 2013-01-31
Reviewed by Kenneth Rohde Christiansen.

Just a refactoring so no new tests.

* rendering/TextAutosizer.cpp: define the new struct and pass it to every method instead of
a cluster and its descendant block containing all text.
* rendering/TextAutosizer.h: change method prototypes to receive the new struct
as a parameter defining an autosizing cluster.

* rendering/TextAutosizer.cpp:
(WebCore::TextAutosizingClusterInfo::TextAutosizingClusterInfo):
(TextAutosizingClusterInfo):
(WebCore):
(WebCore::TextAutosizer::processSubtree):
(WebCore::TextAutosizer::processCluster):
(WebCore::TextAutosizer::processContainer):
(WebCore::TextAutosizer::isAutosizingCluster):
(WebCore::TextAutosizer::clusterShouldBeAutosized):
(WebCore::TextAutosizer::measureDescendantTextWidth):
* rendering/TextAutosizer.h:
(WebCore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141489 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed trivial Unix build fix.
hausmann@webkit.org [Thu, 31 Jan 2013 22:28:41 +0000 (22:28 +0000)]
Unreviewed trivial Unix build fix.

Use Vector<> instead of Deque<> when iterating
over m_attachments in the USE(UNIX_DOMAIN_SOCKETS)
case.

* Platform/CoreIPC/ArgumentDecoder.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141488 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRebaseline after r141459.
jberlin@webkit.org [Thu, 31 Jan 2013 22:23:43 +0000 (22:23 +0000)]
Rebaseline after r141459.

* platform/mac/fast/multicol/shrink-to-column-height-for-pagination-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141487 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAdd a call to the page UI client to determine if a plug-in should load
bweinstein@apple.com [Thu, 31 Jan 2013 22:17:52 +0000 (22:17 +0000)]
Add a call to the page UI client to determine if a plug-in should load
https://bugs.webkit.org/show_bug.cgi?id=108407
<rdar://problem/13066332>

Source/WebKit2:

Reviewed by Anders Carlsson.

This patch adds a client call to the WKPageUIClient to be called to determine
whether or not a plug-in should load.

* UIProcess/API/C/WKPage.h: Add shouldLoadPlugin.
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::getPluginPath): Moved from WebProcessProxy, and added a call to
    m_uiClient.shouldInstantiatePlugin.
* UIProcess/WebPageProxy.h:
* UIProcss/WebPageProxy.messages.in: Moved GetPluginPath from WebProcessProxy to WebPageProxy.
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::getPluginPath): Moved to WebPageProxy.
* UIProcess/WebProcessProxy.h:
* UIProcess/WebUIClient.cpp:
(WebKit::WebUIClient::shouldInstantiatePlugin): Return that we should load the plug-in if
    the client function isn't defined, and call the function if it is.
* UIProcess/WebUIClient.h:
* UIProcess/mac/WebInspectorProxyMac.mm:
(WebKit::WebInspectorProxy::platformCreateInspectorPage): Add an entry for the new
    client function.
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::createPlugin): Send the message to the WebPageProxy, not the WebProcessProxy.
(WebKit::WebPage::canPluginHandleResponse): Made a member function, so it can call sendSync, and
    send the message to the WebPageProxy, not the WebProcessProxy.
* WebProcess/WebPage/WebPage.h:

Tools:

Add entries for the new function in the necessary structs.

Reviewed by Anders Carlsson.

* MiniBrowser/mac/WK2BrowserWindowController.m:
(-[WK2BrowserWindowController awakeFromNib]):
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::createOtherPage):
(WTR::TestController::createWebViewWithOptions):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141486 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed build fix. Win7 port.
roger_fong@apple.com [Thu, 31 Jan 2013 22:13:44 +0000 (22:13 +0000)]
Unreviewed build fix. Win7 port.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreExports.def:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141485 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoTweak ChangeLog.
andersca@apple.com [Thu, 31 Jan 2013 22:12:33 +0000 (22:12 +0000)]
Tweak ChangeLog.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141484 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Rework page scale factor limits initialization
aelias@chromium.org [Thu, 31 Jan 2013 22:02:23 +0000 (22:02 +0000)]
[chromium] Rework page scale factor limits initialization
https://bugs.webkit.org/show_bug.cgi?id=108446

Reviewed by James Robinson.

When loading a page with viewportEnabled, both the limits
specified by the viewport tag and the content width need to be
considered before we initialize the minimum page scale (and
pageScaleFactor itself usually to the same value). The timing has
proven tricky to get correct.

This patch simplifies the flow by computing the
limits only at the end of layouts and at no other time. In combination
with https://bugs.webkit.org/show_bug.cgi?id=107922 which sets
needsLayout() appropriately, this results in a more robust and easy
to understand sequence.

Fixes FixedLayoutInitializeAtMinimumPageScale test.

* src/ChromeClientImpl.cpp:
(WebKit::ChromeClientImpl::dispatchViewportPropertiesDidChange):
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::WebViewImpl):
(WebKit::WebViewImpl::resize):
(WebKit::WebViewImpl::setPageScaleFactorLimits):
(WebKit::WebViewImpl::computePageScaleFactorLimits):
(WebKit::WebViewImpl::layoutUpdated):
(WebKit::WebViewImpl::didChangeContentsSize):
* src/WebViewImpl.h:
(WebKit::WebViewImpl::setInitialPageScaleFactor):
(WebViewImpl):
* tests/WebFrameTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141483 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUse a Vector for IPC attachments
andersca@apple.com [Thu, 31 Jan 2013 22:00:22 +0000 (22:00 +0000)]
Use a Vector for IPC attachments
https://bugs.webkit.org/show_bug.cgi?id=108517

Reviewed by Sam Weinig.

* Platform/CoreIPC/ArgumentDecoder.cpp:
(CoreIPC::ArgumentDecoder::create):
(CoreIPC::ArgumentDecoder::ArgumentDecoder):
(CoreIPC::ArgumentDecoder::removeAttachment):
* Platform/CoreIPC/ArgumentDecoder.h:
(ArgumentDecoder):
* Platform/CoreIPC/Connection.h:
* Platform/CoreIPC/MessageDecoder.cpp:
(CoreIPC::MessageDecoder::create):
(CoreIPC::MessageDecoder::MessageDecoder):
* Platform/CoreIPC/MessageDecoder.h:
(MessageDecoder):
* Platform/CoreIPC/mac/ConnectionMac.cpp:
(CoreIPC::createMessageDecoder):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141482 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] Bing Images viewport causes layout "fun"
zhajiang@rim.com [Thu, 31 Jan 2013 21:57:00 +0000 (21:57 +0000)]
[BlackBerry] Bing Images viewport causes layout "fun"
https://bugs.webkit.org/show_bug.cgi?id=108393

Patch by Jacky Jiang  <zhajiang@rim.com>.
Reviewed by Yong Li.
Internally reviewed by Arvid Nilsson and partially reviewed by Konrad Piascik.

Source/WebCore:

PR: 277855
On bing.com image search page, JS generated styles for the divs which
contained images based on the window.screen and window.innerWidth.
Unfortunately the styles for the divs weren't good enough on our device
which resulted in three images being overlapped in the same viewport.
On our port, we calculate the screen rect base on the physical screen
pixels. However, iOS and lastest Android Chrome port already changed it
to density-independent (logical) pixels. And also to my knowledge that
some websites would generate bad styles for us as well through
@media device-width/device-height.
To be consistent and fix such kinds of issues, we should use
density-independent pixels instead of physical screen pixels. Mainly
ported from Chrome port change r139356. The patch will affect
screen.width/height, window.outerWidth/outerHeight and
@media device-width/device-height.

* platform/blackberry/PlatformScreenBlackBerry.cpp:
(WebCore):
(WebCore::toUserSpace):
(WebCore::screenAvailableRect):
(WebCore::screenRect):

Source/WebKit/blackberry:

PR: 277855
Set applyPageScaleFactorInCompositor settting to false by default.
Scale down chrome window rect to density-independent pixels.

* Api/WebPage.cpp:
(BlackBerry::WebKit::WebPagePrivate::didChangeSettings):
* Api/WebSettings.cpp:
(WebKit):
(BlackBerry::WebKit::WebSettings::standardSettings):
(BlackBerry::WebKit::WebSettings::applyDeviceScaleFactorInCompositor):
(BlackBerry::WebKit::WebSettings::setApplyPageScaleFactorInCompositor):
* Api/WebSettings.h:
* WebCoreSupport/ChromeClientBlackBerry.cpp:
(WebCore::ChromeClientBlackBerry::windowRect):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141481 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[CSS Shaders] Parse custom filter function with the at-rule reference syntax
mvujovic@adobe.com [Thu, 31 Jan 2013 21:52:04 +0000 (21:52 +0000)]
[CSS Shaders] Parse custom filter function with the at-rule reference syntax
https://bugs.webkit.org/show_bug.cgi?id=108351

Reviewed by Dean Jackson.

Source/WebCore:

This patch implements the parsing for the new custom filter function "at-rule reference"
syntax. It does not implement the style resolution yet.

The custom function syntax has changed in the spec. Instead of including all of the custom
filter information inline, the custom function can now reference an @filter rule by name.
The custom function can still accept a list of parameters.

The syntax is defined as the following, where IDENT is the name of the @filter rule:
filter: custom(IDENT [, <param>]*)

In practice, it can look like this:
filter: custom(page-curl, direction 90, amount 0.5);

Spec: https://dvcs.w3.org/hg/FXTF/raw-file/tip/filters/index.html#customident-ltparamgt

Tests: css3/filters/custom-with-at-rule-syntax/parsing-custom-function-invalid.html
       css3/filters/custom-with-at-rule-syntax/parsing-custom-function-valid.html

* css/CSSParser.cpp:
(WebCore::CSSParser::parseCustomFilterParameters):
    Factor out a new function that just parses custom filter parameters. This is used by the
    previous inline syntax and the new at-rule reference syntax.
(WebCore):
(WebCore::CSSParser::parseCustomFilterFunctionWithAtRuleReferenceSyntax):
    This method parses the new at-rule reference syntax. When we remove the inline syntax,
    we can rename this method to "parseCustomFilterFunction" and call it directly.
(WebCore::CSSParser::parseCustomFilterFunctionWithInlineSyntax):
    Move the previous inline syntax parsing code into a new function.
(WebCore::CSSParser::parseCustomFilterFunction):
    This method looks ahead in the CSS parser values and decides whether to parse the custom
    function with the previous inline syntax or with the new at-rule reference syntax.
(WebCore::CSSParser::parseFilter):
    I renamed the "parseCustomFilter" method to "parseCustomFilterFunction" to be more
    explicit. Here, we update the name in the call site.
* css/CSSParser.h:
* css/StyleResolver.cpp:
(WebCore::StyleResolver::createCustomFilterOperationWithAtRuleReferenceSyntax):
    This is a stub method. In the future, style resolution for the at-rule reference
    syntax will happen here. When we remove the inline syntax, we can rename this method
(WebCore::StyleResolver::createCustomFilterOperationWithInlineSyntax):
    This method contains the previous inline syntax style resolution code.
(WebCore):
(WebCore::StyleResolver::createCustomFilterOperation):
    This method decides which syntax we should use to resolve the style.
* css/StyleResolver.h:
(StyleResolver):

LayoutTests:

Add positive and negative parsing tests for the new custom function syntax.

Add a new folder "css3/filters/custom-with-at-rule-syntax". This will contain all the tests
using the new custom filters at-rule syntax. We will gradually copy tests in
"css3/filters/custom" over to "css3/filters/custom-with-at-rule-syntax" and modify them to
use the new at-rule syntax. When all of the tests have been replicated using the new syntax,
we will remove the previous syntax and the tests in "css3/filters/custom".

* css3/filters/custom-with-at-rule-syntax/parsing-custom-function-invalid-expected.txt: Added.
* css3/filters/custom-with-at-rule-syntax/parsing-custom-function-invalid.html: Added.
* css3/filters/custom-with-at-rule-syntax/parsing-custom-function-valid-expected.txt: Added.
* css3/filters/custom-with-at-rule-syntax/parsing-custom-function-valid.html: Added.
* css3/filters/custom-with-at-rule-syntax/script-tests/parsing-custom-function-invalid.js: Added.
(testInvalidFilterRule):
* css3/filters/custom-with-at-rule-syntax/script-tests/parsing-custom-function-valid.js: Added.
(testFilterProperty):
* css3/filters/script-tests/custom-filter-parsing-common.js:
(heading):
    New function to print a heading to make groups of related parsing tests easier to see.
(shouldHaveConstructor):
    New function to check the JS type of an object on JSC as well as V8. This is intended to
    eventually replace shouldBeType, which works differently on V8 vs. JSC and requires us
    to create Chromium-specific expectations for the custom filters parsing tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141480 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoEditor::m_compositionNode not updated on HTMLInputElement::setValue()
commit-queue@webkit.org [Thu, 31 Jan 2013 21:50:17 +0000 (21:50 +0000)]
Editor::m_compositionNode not updated on HTMLInputElement::setValue()
https://bugs.webkit.org/show_bug.cgi?id=107737

Patch by Aurimas Liutikas <aurimas@chromium.org> on 2013-01-31
Reviewed by Ryosuke Niwa.

Source/WebCore:

Chromium has a bug where the IME composition did not get cancelled on JavaScript changes
to the focused editing field. Most of other WebKit ports were already doing this check
in their EditorClient::respondToChangedSelection. I took that logic and moved it to the
Editor so every port and use the same code.

An existing test editing/input/setting-input-value-cancel-ime-composition.html covers this change.
This test used to have an expectation to fail on Chromium and after this patch it will start passing.

* editing/Editor.cpp:
(WebCore::Editor::cancelCompositionIfSelectionIsInvalid):
    Adding a call that can be used by any the port to cancel the composition if it's no longer valid.
(WebCore):
* editing/Editor.h:
(Editor):

Source/WebKit/chromium:

* public/WebViewClient.h:
(WebKit::WebViewClient::didCancelCompositionOnSelectionChange):
    Adding a callback to let the WebViewClient know that the composition has been cancelled.
* src/EditorClientImpl.cpp:
(WebKit::EditorClientImpl::respondToChangedSelection):
    Adding a call composition if it is no longer valid.

Source/WebKit/efl:

* WebCoreSupport/EditorClientEfl.cpp:
(WebCore::EditorClientEfl::respondToChangedSelection):
    Adding a call to the newly refactored method.

Source/WebKit/gtk:

* WebCoreSupport/EditorClientGtk.cpp:
(WebKit::EditorClient::respondToChangedSelection):
    Adding a call to the newly refactored Editor method.

Source/WebKit/mac:

* WebView/WebHTMLView.mm:
(-[WebHTMLView _updateSelectionForInputManager]):

Source/WebKit/win:

* WebView.cpp:
(WebView::updateSelectionForIME):
    Adding a call to the newly refactored method.

LayoutTests:

* platform/chromium/TestExpectations: Removed fail expectation for the editing/input/setting-input-value-cancel-ime-composition.html since this patch fixes the bug https://bugs.webkit.org/show_bug.cgi?id=55560

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141479 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Unreviewed gardening.
fmalita@chromium.org [Thu, 31 Jan 2013 21:47:25 +0000 (21:47 +0000)]
[Chromium] Unreviewed gardening.

Win build fix after http://trac.webkit.org/changeset/141471.

* DumpRenderTree/chromium/TestRunner/src/MockSpellCheck.cpp:
(MockSpellCheck::spellCheckWord):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141478 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoDisable ENABLE_FULLSCREEN_API on iOS
joepeck@webkit.org [Thu, 31 Jan 2013 21:42:46 +0000 (21:42 +0000)]
Disable ENABLE_FULLSCREEN_API on iOS
https://bugs.webkit.org/show_bug.cgi?id=108250

Reviewed by Benjamin Poulain.

Source/JavaScriptCore:

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit/mac:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit2:

* Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141477 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Move LocalizedStrings to WebCore
pilgrim@chromium.org [Thu, 31 Jan 2013 21:18:16 +0000 (21:18 +0000)]
[Chromium] Move LocalizedStrings to WebCore
https://bugs.webkit.org/show_bug.cgi?id=108488

Reviewed by Adam Barth.

Part of a larger refactoring series; see tracking bug 106829.

Source/WebCore:

* WebCore.gypi:
* platform/chromium/LocalizedStringsChromium.cpp: Copied from Source/WebKit/chromium/src/LocalizedStrings.cpp.
(WebCore::imageTitle):

Source/WebKit/chromium:

* WebKit.gyp:
* src/LocalizedStrings.cpp: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWebKit2: provide new bundle APIs to allow bundle clients to be notified of pasteboard...
enrica@apple.com [Thu, 31 Jan 2013 21:14:58 +0000 (21:14 +0000)]
WebKit2: provide new bundle APIs to allow bundle clients to be notified of pasteboard access.
https://bugs.webkit.org/show_bug.cgi?id=108396.
<rdar://problem/12920461>

Source/WebCore:

Reviewed by Alexey Proskuryakov.

Adds support in WebCore to new EditorClient methods to support the modified
client bundle API in WebKit2.

* WebCore.exp.in:
* editing/Editor.cpp:
(WebCore::Editor::cut): Added call to willWriteSelectionToPasteboard.
(WebCore::Editor::copy): Ditto.
(WebCore::Editor::willWriteSelectionToPasteboard): Added.
* editing/Editor.h:
* loader/EmptyClients.h: Added empty client implementation
for the new methods.
(WebCore::EmptyEditorClient::willWriteSelectionToPasteboard):
(WebCore::EmptyEditorClient::getClientPasteboardDataForRange):
* page/EditorClient.h:
* platform/mac/PasteboardMac.mm:
(WebCore::Pasteboard::writeSelectionForTypes): Added call to getClientPasteboardDataForRange.

Source/WebKit/blackberry:

Reviewed by Alexey Proskuryakov.

Adds stub implementation for WebKit of the new EditorClient methods.

* WebCoreSupport/EditorClientBlackBerry.cpp:
(WebCore::EditorClientBlackBerry::willWriteSelectionToPasteboard):
(WebCore::EditorClientBlackBerry::getClientPasteboardDataForRange):
* WebCoreSupport/EditorClientBlackBerry.h:

Source/WebKit/chromium:

Reviewed by Alexey Proskuryakov.

Adds stub implementation for WebKit of the new EditorClient methods.

* src/EditorClientImpl.cpp:
(WebKit::EditorClientImpl::willWriteSelectionToPasteboard):
(WebKit::EditorClientImpl::getClientPasteboardDataForRange):
* src/EditorClientImpl.h:

Source/WebKit/efl:

Reviewed by Alexey Proskuryakov.

Adds stub implementation for WebKit of the new EditorClient methods.

* WebCoreSupport/EditorClientEfl.cpp:
(WebCore::EditorClientEfl::willWriteSelectionToPasteboard):
(WebCore::EditorClientEfl::getClientPasteboardDataForRange):
* WebCoreSupport/EditorClientEfl.h:

Source/WebKit/gtk:

Reviewed by Alexey Proskuryakov.

Adds stub implementation for WebKit of the new EditorClient methods.

* WebCoreSupport/EditorClientGtk.cpp:
(WebKit::EditorClient::willWriteSelectionToPasteboard):
(WebKit::EditorClient::getClientPasteboardDataForRange):
* WebCoreSupport/EditorClientGtk.h:

Source/WebKit/mac:

Reviewed by Alexey Proskuryakov.

Adds stub implementation for WebKit of the new EditorClient methods.

* WebCoreSupport/WebEditorClient.h:
* WebCoreSupport/WebEditorClient.mm:
(WebEditorClient::willWriteSelectionToPasteboard):
(WebEditorClient::getClientPasteboardDataForRange):

Source/WebKit/qt:

Reviewed by Alexey Proskuryakov.

Adds stub implementation for WebKit of the new EditorClient methods.

* WebCoreSupport/EditorClientQt.cpp:
(WebCore::EditorClientQt::willWriteSelectionToPasteboard):
(WebCore::EditorClientQt::getClientPasteboardDataForRange):
* WebCoreSupport/EditorClientQt.h:

Source/WebKit/win:

Reviewed by Alexey Proskuryakov.

Adds stub implementation for WebKit of the new EditorClient methods.

* WebCoreSupport/WebEditorClient.cpp:
(WebEditorClient::willWriteSelectionToPasteboard):
(WebEditorClient::getClientPasteboardDataForRange):
* WebCoreSupport/WebEditorClient.h:

Source/WebKit/wince:

Reviewed by Alexey Proskuryakov.

Adds stub implementation for WebKit of the new EditorClient methods.

* WebCoreSupport/EditorClientWinCE.cpp:
(WebKit::EditorClientWinCE::willWriteSelectionToPasteboard):
(WebKit::EditorClientWinCE::getClientPasteboardDataForRange):
* WebCoreSupport/EditorClientWinCE.h:

Source/WebKit/wx:

Reviewed by Alexey Proskuryakov.

Adds stub implementation for WebKit of the new EditorClient methods.

* WebKitSupport/EditorClientWx.cpp:
(WebCore::EditorClientWx::willWriteSelectionToPasteboard):
(WebCore::EditorClientWx::getClientPasteboardDataForRange):
* WebKitSupport/EditorClientWx.h:

Source/WebKit2:

Reviewed by Alexey Proskuryakov.

This patch adds new bundle client API to receive notifications
relative the pasteboard activity. There are 2 new API added to
InjectedBundleEditorClient, to receive notification before and
after the pasteboard content is added and one API to provide
additional content to add to the pasteboard.
In order to create content to add to the pasteboard, WKWebArchiveRef
and WKWebArchiveResourcesRef have been added to the set of API level
object.
This work is a joint effort with Sam Weinig who contributed the
support for WKWebArchiveRef, WKWebArchiveResourcesRef and related
files. Sam is the author of the first chunk of changes listed below.

* Shared/API/c/WKBase.h:
* Shared/API/c/WKSharedAPICast.h:
* Shared/API/c/mac/WKWebArchive.cpp: Added.
(WKWebArchiveGetTypeID):
(WKWebArchiveCreate):
(WKWebArchiveCreateWithData):
(WKWebArchiveCreateFromRange):
(WKWebArchiveCopyMainResource):
(WKWebArchiveCopySubresources):
(WKWebArchiveCopySubframeArchives):
(WKWebArchiveCopyData):
* Shared/API/c/mac/WKWebArchive.h: Added.
* Shared/API/c/mac/WKWebArchiveResource.cpp: Added.
(WKWebArchiveResourceGetTypeID):
(WKWebArchiveResourceCreate):
(WKWebArchiveResourceCopyData):
(WKWebArchiveResourceCopyURL):
(WKWebArchiveResourceCopyMIMEType):
(WKWebArchiveResourceCopyTextEncoding):
* Shared/API/c/mac/WKWebArchiveResource.h: Added.
* Shared/APIObject.h:
* Shared/WebArchive.cpp: Added.
(WebKit::WebArchive::create):
(WebKit::WebArchive::WebArchive):
(WebKit::WebArchive::~WebArchive):
(WebKit::WebArchive::mainResource):
(WebKit::WebArchive::subresources):
(WebKit::WebArchive::subframeArchives):
(WebKit::releaseCFData):
(WebKit::WebArchive::data):
(WebKit::WebArchive::coreLegacyWebArchive):
* Shared/WebArchive.h: Added.
(WebKit::WebArchive::type):
* Shared/WebArchiveResource.cpp: Added.
(WebKit::WebArchiveResource::create):
(WebKit::WebArchiveResource::WebArchiveResource):
(WebKit::WebArchiveResource::~WebArchiveResource):
(WebKit::releaseCFData):
(WebKit::WebArchiveResource::data):
(WebKit::WebArchiveResource::URL):
(WebKit::WebArchiveResource::MIMEType):
(WebKit::WebArchiveResource::textEncoding):
(WebKit::WebArchiveResource::coreArchiveResource):
* Shared/WebArchiveResource.h: Added.
(WebKit::WebArchiveResource::type):
* WebKit2.xcodeproj/project.pbxproj:

* Shared/APIClientTraits.cpp: Added versioning to InjectedBundlePageEditorClient.
* Shared/APIClientTraits.h:
* WebProcess/InjectedBundle/API/c/WKBundlePage.h:
* WebProcess/InjectedBundle/InjectedBundlePageEditorClient.cpp:
(WebKit::InjectedBundlePageEditorClient::willWriteToPasteboard): Added.
(WebKit::InjectedBundlePageEditorClient::getPasteboardDataForRange): Added.
(WebKit::InjectedBundlePageEditorClient::didWriteToPasteboard): Added.
* WebProcess/InjectedBundle/InjectedBundlePageEditorClient.h:
* WebProcess/WebCoreSupport/WebEditorClient.cpp:
(WebKit::WebEditorClient::didWriteSelectionToPasteboard):
(WebKit::WebEditorClient::willWriteSelectionToPasteboard):
(WebKit::WebEditorClient::getClientPasteboardDataForRange):
* WebProcess/WebCoreSupport/WebEditorClient.h:

Tools:

Reviewed by Alexey Proskuryakov.

Adding new WebKit2 test with relevant bundle, to test the new APIs.

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKit2/PasteboardNotifications.mm: Added.
(TestWebKitAPI::didReceiveMessageFromInjectedBundle):
(TestWebKitAPI::setInjectedBundleClient):
* TestWebKitAPI/Tests/WebKit2/PasteboardNotifications_Bundle.cpp: Added.
(PasteboardNotificationsTest):
(TestWebKitAPI::willWriteToPasteboard):
(TestWebKitAPI::getPasteboardDataForRange):
(TestWebKitAPI::didWriteToPasteboard):
(TestWebKitAPI::PasteboardNotificationsTest::PasteboardNotificationsTest):
(TestWebKitAPI::PasteboardNotificationsTest::didCreatePage):
* TestWebKitAPI/Tests/WebKit2/execCopy.html: Added.
* WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp: Updated to reflect
changes in InjectedBundleEditorClient.
(WTR::InjectedBundlePage::InjectedBundlePage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141473 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRemove MessageID.h
andersca@apple.com [Thu, 31 Jan 2013 21:03:22 +0000 (21:03 +0000)]
Remove MessageID.h
https://bugs.webkit.org/show_bug.cgi?id=108516

Reviewed by Sam Weinig.

* Platform/CoreIPC/Connection.cpp:
(CoreIPC::Connection::processIncomingMessage):
(CoreIPC::Connection::sendOutgoingMessages):
(CoreIPC::Connection::dispatchSyncMessage):
(CoreIPC::Connection::dispatchMessage):
* Platform/CoreIPC/Connection.h:
(CoreIPC):
(Connection):
* Platform/CoreIPC/MessageID.h: Removed.
* Platform/CoreIPC/mac/ConnectionMac.cpp:
(CoreIPC::Connection::sendOutgoingMessage):
(CoreIPC::Connection::receiveSourceEventHandler):
* Shared/CoreIPCSupport/WebConnectionMessageKinds.h: Removed.
* Shared/CoreIPCSupport/WebContextMessageKinds.h:
* UIProcess/DrawingAreaProxy.h:
(CoreIPC):
* UIProcess/Network/CustomProtocols/CustomProtocolManagerProxy.h:
* UIProcess/WebApplicationCacheManagerProxy.h:
* UIProcess/WebContext.cpp:
(WebKit::WebContext::WebContext):
(WebKit::WebContext::didReceiveMessage):
(WebKit::WebContext::didReceiveSyncMessage):
* UIProcess/WebCookieManagerProxy.h:
* UIProcess/WebFrameProxy.h:
(CoreIPC):
* UIProcess/WebFullScreenManagerProxy.h:
(CoreIPC):
* UIProcess/WebIconDatabase.h:
(CoreIPC):
* UIProcess/WebPageProxy.cpp:
* UIProcess/WebPageProxy.h:
(CoreIPC):
* WebKit2.xcodeproj/project.pbxproj:
* WebProcess/FullScreen/WebFullScreenManager.cpp:
* WebProcess/FullScreen/WebFullScreenManager.h:
(CoreIPC):
* WebProcess/InjectedBundle/InjectedBundle.cpp:
(WebKit::InjectedBundle::postMessage):
(WebKit::InjectedBundle::postSynchronousMessage):
* WebProcess/InjectedBundle/InjectedBundle.h:
(CoreIPC):
* WebProcess/WebPage/DrawingArea.h:
(CoreIPC):
* WebProcess/WebPage/LayerTreeHost.h:
(CoreIPC):
* WebProcess/WebPage/WebPage.cpp:
* WebProcess/WebPage/WebPage.h:
(CoreIPC):
* WebProcess/WebPage/WebPageGroupProxy.h:
(CoreIPC):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141472 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoTools: [Chromium] Add two-word misspelling to mock spellchecker
commit-queue@webkit.org [Thu, 31 Jan 2013 20:58:51 +0000 (20:58 +0000)]
Tools: [Chromium] Add two-word misspelling to mock spellchecker
https://bugs.webkit.org/show_bug.cgi?id=108498

Patch by Rouslan Solomakhin <rouslan@chromium.org> on 2013-01-31
Reviewed by Tony Chang.

Some layout tests verify spellcheck behavior when multiple words are
marked as a single misspelling. This change adds a two-word misspelling
'upper case' to the mock spellchecker used by layout tests.

* DumpRenderTree/chromium/TestRunner/src/MockSpellCheck.cpp:
(MockSpellCheck::spellCheckWord): Modified to handle spaces in misspellings.
(MockSpellCheck::initializeIfNeeded):  Modified to use a vector instead of a hash table.
* DumpRenderTree/chromium/TestRunner/src/MockSpellCheck.h:
(MockSpellCheck): Changed misspellings container from hash table to vector.

LayoutTests: [Chromium] Expect spellcheck to work for exactly-selected multi-word misspellings
https://bugs.webkit.org/show_bug.cgi?id=108498

Patch by Rouslan Solomakhin <rouslan@chromium.org> on 2013-01-31
Reviewed by Tony Chang.

* platform/chromium/TestExpectations: Update spellcheck tests expectations.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141471 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Make selection handling work in applyPageScaleInCompositor mode
commit-queue@webkit.org [Thu, 31 Jan 2013 20:55:33 +0000 (20:55 +0000)]
[chromium] Make selection handling work in applyPageScaleInCompositor mode
https://bugs.webkit.org/show_bug.cgi?id=107831

Patch by Chris Hopman <cjhopman@chromium.org> on 2013-01-31
Reviewed by Ryosuke Niwa.

These functions expect a window point. When in
applyPageScaleFactorInCompositor mode, the points need to be unscaled
by the page scale factor.

* src/WebFrameImpl.cpp:
(WebKit::WebFrameImpl::selectRange):
(WebKit::WebFrameImpl::moveCaretSelectionTowardsWindowPoint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141470 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION(r141136): Apple's internal PLT test suite doesn't finish
jberlin@webkit.org [Thu, 31 Jan 2013 20:55:15 +0000 (20:55 +0000)]
REGRESSION(r141136): Apple's internal PLT test suite doesn't finish
https://bugs.webkit.org/show_bug.cgi?id=108380

Mark some tests whose results depend on main resource caching being enabled as failing until
main resource caching can be re-enabled on mac.

* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141469 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUse OS() and PLATFORM() macros in WebCorePrefix.h for readability
ddkilzer@apple.com [Thu, 31 Jan 2013 20:52:29 +0000 (20:52 +0000)]
Use OS() and PLATFORM() macros in WebCorePrefix.h for readability
<http://webkit.org/b/108374>

Reviewed by Laszlo Gombos.

* WebCorePrefix.h: Replaced all possible checks with OS() or
PLATFORM() macros other than __APPLE__.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Prospective build fix.
hausmann@webkit.org [Thu, 31 Jan 2013 20:51:09 +0000 (20:51 +0000)]
[Qt] Prospective build fix.

Reviewed by Jocelyn Turcotte.

Don't create .prl files for intermediate libs as their contents get
injected within -Wl,-whole-archive -lFoo -wl,-no-whole-archive.

* qmake/mkspecs/features/default_post.prf:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCleanup: Use ScriptExecutionContext::topOrigin when relevant.
mkwst@chromium.org [Thu, 31 Jan 2013 20:48:49 +0000 (20:48 +0000)]
Cleanup: Use ScriptExecutionContext::topOrigin when relevant.
https://bugs.webkit.org/show_bug.cgi?id=108476

Source/WebCore:

Reviewed by Jochen Eisinger.

Rather than diving through 'frame()->top()' or 'topDocument()' to get
the SecurityOrigin of the top-level document in a frame, we can now
directly grab the 'topOrigin()' off of a ScriptExecutionContext. This
patch adjusts a few callsites to use the new hotness rather than the
old brokenness.

This is a pure refactoring: No web-visible changes should result.

* Modules/webdatabase/DOMWindowWebDatabase.cpp:
(WebCore::DOMWindowWebDatabase::openDatabase):
* page/DOMWindow.cpp:
(WebCore::DOMWindow::sessionStorage):
(WebCore::DOMWindow::localStorage):
* workers/SharedWorker.cpp:
(WebCore::SharedWorker::create):
* workers/WorkerMessagingProxy.cpp:
(WebCore::WorkerMessagingProxy::startWorkerContext):

Source/WebKit2:

Reviewed by Anders Carlsson.

* WebProcess/Plugins/PluginView.cpp:
(WebKit::PluginView::storageBlockingStateChanged):
(WebKit::PluginView::isPrivateBrowsingEnabled):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoGet rid of IncomingMessage
andersca@apple.com [Thu, 31 Jan 2013 20:47:12 +0000 (20:47 +0000)]
Get rid of IncomingMessage
https://bugs.webkit.org/show_bug.cgi?id=108514

Reviewed by Sam Weinig.

* Platform/CoreIPC/Connection.cpp:
(Connection::SyncMessageState):
(ConnectionAndIncomingMessage):
(CoreIPC::Connection::SyncMessageState::~SyncMessageState):
(CoreIPC::Connection::SyncMessageState::processIncomingMessage):
(CoreIPC::Connection::SyncMessageState::dispatchMessages):
(CoreIPC::Connection::waitForMessage):
(CoreIPC::Connection::processIncomingMessage):
(CoreIPC::Connection::enqueueIncomingMessage):
(CoreIPC::Connection::dispatchMessage):
(CoreIPC::Connection::dispatchOneMessage):
* Platform/CoreIPC/Connection.h:
(Connection):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Curl] There is no way for a WebKit client to set the Curl cookie jar path
commit-queue@webkit.org [Thu, 31 Jan 2013 20:44:11 +0000 (20:44 +0000)]
[Curl] There is no way for a WebKit client to set the Curl cookie jar path
https://bugs.webkit.org/show_bug.cgi?id=107692

Patch by peavo@outlook.com <peavo@outlook.com> on 2013-01-31
Reviewed by Brent Fulgham.

Set default cookie jar path.
Get the cookie jar path for Curl from the environment variable CURL_COOKIE_JAR_PATH.

* platform/network/curl/ResourceHandleManager.cpp:
(WebCore::cookieJarPath): Added helper function to set cookie jar path.
(WebCore::ResourceHandleManager::ResourceHandleManager): Call helper function.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141464 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRemove PLATFORM(WIN_CAIRO) from NetscapePluginX11.cpp
paroga@webkit.org [Thu, 31 Jan 2013 20:41:21 +0000 (20:41 +0000)]
Remove PLATFORM(WIN_CAIRO) from NetscapePluginX11.cpp
https://bugs.webkit.org/show_bug.cgi?id=108439

Reviewed by Brent Fulgham.

PLATFORM(WIN_CAIRO) is Windows only, where no X11 exists.

* WebProcess/Plugins/Netscape/x11/NetscapePluginX11.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141463 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION (r138962): Fails to show "confirm form resubmission", hangs browser
mjs@apple.com [Thu, 31 Jan 2013 20:38:19 +0000 (20:38 +0000)]
REGRESSION (r138962): Fails to show "confirm form resubmission", hangs browser
https://bugs.webkit.org/show_bug.cgi?id=108214

Reviewed by Adam Barth.

Attempt a workaround suggested by Nate Chapin by putting some code
in #ifdef !PLATFORM(CHROMIUM).

* loader/MainResourceLoader.cpp:
(WebCore::MainResourceLoader::notifyFinished):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141462 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[v8] Enable binding integrity on mac
tsepez@chromium.org [Thu, 31 Jan 2013 20:24:04 +0000 (20:24 +0000)]
[v8] Enable binding integrity on mac
https://bugs.webkit.org/show_bug.cgi?id=108500

Reviewed by Adam Barth.

Patch is correct if existing tests pass without new crashes.

* features.gypi:
Set ENABLE_BINDING_INTEGRITY=1 when OS=="mac"

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION(r128517): Percentage heights in quirks mode collapse when printing
ojan@chromium.org [Thu, 31 Jan 2013 20:00:12 +0000 (20:00 +0000)]
REGRESSION(r128517): Percentage heights in quirks mode collapse when printing
https://bugs.webkit.org/show_bug.cgi?id=108382

Reviewed by David Hyatt.

Source/WebCore:

r128517 clean up our containing block finding logic, but broke percentage
heights in quirks mode during printing since the RenderView would have 0 height.
Turns out we already had a long-standing bug where we'd incorrectly
treat collapse percentage heights on the body when printing as well.

Fix both bugs by changing the way we grab the logical height on the RenderView.
RenderView::computeLogicalHeight returns 0 when printing. For the purposes of
stretching and percentage heights, we instead need to return the pageLogicalHeight.

Tests: printing/quirks-percentage-height-body.html
       printing/quirks-percentage-height.html
       printing/standards-percentage-heights.html

* rendering/RenderBox.cpp:
(WebCore::RenderBox::computeLogicalHeight):
This FIXME is outdated and already fixed. Also, call viewLogicalHeightForPercentages
which does the same logic except also correctly handles column RenderViews.
(WebCore::RenderBox::viewLogicalHeightForPercentages):
(WebCore::RenderBox::computePercentageLogicalHeight):
* rendering/RenderBox.h:

* rendering/RenderBox.cpp:
(WebCore::RenderBox::computeLogicalHeight):
(WebCore::RenderBox::viewLogicalHeightForPercentages):
(WebCore):
(WebCore::RenderBox::computePercentageLogicalHeight):
* rendering/RenderBox.h:
(RenderBox):

LayoutTests:

* platform/chromium/fast/multicol/shrink-to-column-height-for-pagination-expected.txt:
We're just clipping more content that you can't scroll to anyways. This looks
like an improvement to me.
* printing/css2.1/page-break-inside-000-expected.txt:
We pass this test now!
* printing/quirks-percentage-height-body-expected.html: Added.
* printing/quirks-percentage-height-body.html: Added.
* printing/quirks-percentage-height-expected.html: Added.
* printing/quirks-percentage-height.html: Added.
* printing/standards-percentage-heights-expected.html: Added.
* printing/standards-percentage-heights.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoinspector/editor/text-editor-ctrl-movements.html was renamed to
jberlin@webkit.org [Thu, 31 Jan 2013 19:56:23 +0000 (19:56 +0000)]
inspector/editor/text-editor-ctrl-movements.html was renamed to
inspector/editor/text-editor-word-jumps.html in r141399. Update the TestExpectations file.

* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141458 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoASSERTION FAILED: m_clipRectsCache->m_respectingOverflowClip[clipRectsType] == (clipR...
jberlin@webkit.org [Thu, 31 Jan 2013 19:56:20 +0000 (19:56 +0000)]
ASSERTION FAILED: m_clipRectsCache->m_respectingOverflowClip[clipRectsType] == (clipRectsContext.respectOverflowClip == RespectOverflowClip) in RenderLayer.
https://bugs.webkit.org/show_bug.cgi?id=108257

* platform/mac/TestExpectations:
Mark the tests as "crashing" in debug.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[canvas] Implement currentPath to get and set the current path of the context
krit@webkit.org [Thu, 31 Jan 2013 19:46:50 +0000 (19:46 +0000)]
[canvas] Implement currentPath to get and set the current path of the context
https://bugs.webkit.org/show_bug.cgi?id=108246

Patch by Dirk Schulze <krit@webkit.org> on 2013-01-31
Reviewed by Dean Jackson.

Source/WebCore:

Add currentPath attribute to CanvasRenderingContext2d interface. This allows
setting the current context path by an existing Path object as well as getting
the current context path as new Path object. The returned and the set Paths
are not live.

This feature is behind the CANVAS_PATH compiler flag which is disabled by
default for now.

Test: fast/canvas/canvas-currentPath.html

* html/canvas/CanvasPathMethods.cpp: Rename transformIsInvertible to
    isTransformInvertible for harmonizing naming schema.
(WebCore::CanvasPathMethods::moveTo): Ditto.
(WebCore::CanvasPathMethods::lineTo): Ditto.
(WebCore::CanvasPathMethods::quadraticCurveTo): Ditto.
(WebCore::CanvasPathMethods::bezierCurveTo): Ditto.
(WebCore::CanvasPathMethods::arcTo): Ditto.
(WebCore::CanvasPathMethods::arc): Ditto.
(WebCore::CanvasPathMethods::rect): Ditto.
* html/canvas/CanvasPathMethods.h: Ditto.
(CanvasPathMethods):
(WebCore::CanvasPathMethods::isTransformInvertible):
* html/canvas/CanvasRenderingContext2D.cpp:
(WebCore):
(WebCore::CanvasRenderingContext2D::currentPath): Getter for currentPath.
(WebCore::CanvasRenderingContext2D::setCurrentPath): Setter for currentPath.
* html/canvas/CanvasRenderingContext2D.h:
(WebCore):
(CanvasRenderingContext2D):
(WebCore::CanvasRenderingContext2D::isTransformInvertible):
* html/canvas/CanvasRenderingContext2D.idl: Add currentPath attribute.
* html/canvas/DOMPath.h: Add new constructor and getter for Path object.
(WebCore::DOMPath::create): New static function for new ctor.
(DOMPath):
(WebCore::DOMPath::path): Getter for Path object.
(WebCore::DOMPath::DOMPath):

LayoutTests:

Added tests for canvas.currentPath in various combinations of the context state.

* fast/canvas/canvas-currentPath-expected.txt: Added.
* fast/canvas/canvas-currentPath.html: Added.
* fast/canvas/script-tests/canvas-currentPath.js: Added.
(testPointCollection):
* platform/chromium/TestExpectations: Skip test until enabling CANVAS_PATH.
* platform/efl/TestExpectations: Ditto.
* platform/gtk/TestExpectations: Ditto.
* platform/mac/TestExpectations: Ditto.
* platform/qt/TestExpectations: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed build fix: Remove Web Intents files from
hausmann@webkit.org [Thu, 31 Jan 2013 19:42:40 +0000 (19:42 +0000)]
Unreviewed build fix: Remove Web Intents files from
the Qt build system.

* Target.pri:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141455 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWebFrameTest.DisambiguationPopup is failing
commit-queue@webkit.org [Thu, 31 Jan 2013 19:40:14 +0000 (19:40 +0000)]
WebFrameTest.DisambiguationPopup is failing
https://bugs.webkit.org/show_bug.cgi?id=108160

Patch by Dan Alcantara <dfalcantara@chromium.org> on 2013-01-31
Reviewed by Adam Barth.

Fix the unit test so that it is using the right HTML file.

* tests/WebFrameTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] webkit_unit_tests should depend on base/allocator for ASAN
tony@chromium.org [Thu, 31 Jan 2013 19:37:53 +0000 (19:37 +0000)]
[Chromium] webkit_unit_tests should depend on base/allocator for ASAN
https://bugs.webkit.org/show_bug.cgi?id=108497

Reviewed by James Robinson.

* WebKitUnitTests.gyp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141453 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK build fix.
zandobersek@gmail.com [Thu, 31 Jan 2013 19:35:21 +0000 (19:35 +0000)]
Unreviewed GTK build fix.
Removing build targets for Web Intents files that were removed in r141448.

* GNUmakefile.list.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141452 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoEWS graphs have unusual data for patch waiting times
commit-queue@webkit.org [Thu, 31 Jan 2013 19:34:38 +0000 (19:34 +0000)]
EWS graphs have unusual data for patch waiting times
https://bugs.webkit.org/show_bug.cgi?id=108427

Patch by Alan Cutter <alancutter@chromium.org> on 2013-01-31
Reviewed by Eric Seidel.

The max_patches_waiting QueueLog property may have missed recording patches waiting in idle times when no events are fired to update it. I added a cron job to take care of this issue.
The patch wait duration in PatchLogs was being incorrectly updated after a patch expired and was picked up again from the queue. That period should really count towards the processing time instead.
Added a WarningLog to record if any erroneous situations occur due to invalid contents in the datastore. These basically highlight if any "impossible" execution paths get executed.

* QueueStatusServer/app.yaml:
* QueueStatusServer/cron.yaml:
* QueueStatusServer/handlers/syncqueuelogs.py: Copied from Tools/QueueStatusServer/model/patchlog.py.
(SyncQueueLogs):
(SyncQueueLogs.get):
* QueueStatusServer/loggers/recordpatchevent.py:
(RecordPatchEvent.added):
(RecordPatchEvent.retrying):
(RecordPatchEvent.started):
(RecordPatchEvent.stopped):
(RecordPatchEvent.updated):
* QueueStatusServer/main.py:
* QueueStatusServer/model/patchlog.py:
(PatchLog.lookup):
(PatchLog):
(PatchLog.lookup_if_exists):
(PatchLog.calculate_process_duration):
(PatchLog._generate_key):
* QueueStatusServer/model/queuelog.py:
(QueueLog.update_max_patches_waiting):
(QueueLog._get_or_create_txn):
(QueueLog):
(QueueLog._get_patches_waiting):
* QueueStatusServer/model/warninglog.py: Copied from Tools/QueueStatusServer/model/patchlog.py.
(WarningLog):
(WarningLog.record):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141451 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCall FrameView::contentsResized() when setting fixed layout size
aelias@chromium.org [Thu, 31 Jan 2013 19:25:41 +0000 (19:25 +0000)]
Call FrameView::contentsResized() when setting fixed layout size
https://bugs.webkit.org/show_bug.cgi?id=107922

Reviewed by James Robinson.

In fixed layout mode, we should be calling contentsResized() when the
fixed layout size is changed, but not laying out when the visible
content rect changes.

Previously landed as r140869 but was reverted due to a bug in bundled
Chromium-specific code. This patch includes just the minimum needed in
WebCore.

New WebFrameTest: FrameViewNeedsLayoutOnFixedLayoutResize.  Some
flaky and obsolete tests for the old page scale mode are also deleted.

Source/WebCore:

* page/FrameView.cpp:
(WebCore::FrameView::visibleContentsResized):
* platform/ScrollView.cpp:
(WebCore::ScrollView::setFixedLayoutSize):
(WebCore::ScrollView::setUseFixedLayout):

Source/WebKit/chromium:

* tests/WebFrameTest.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, update expectation after the order of elements was codified.
tony@chromium.org [Thu, 31 Jan 2013 19:16:07 +0000 (19:16 +0000)]
Unreviewed, update expectation after the order of elements was codified.
https://bugs.webkit.org/show_bug.cgi?id=98686

Also delete the Qt result since this test is skipped on Qt.

* editing/pasteboard/data-transfer-items-expected.txt:
* platform/qt/editing/pasteboard/data-transfer-items-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141449 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRemove Web Intents code from WebKit2
andersca@apple.com [Thu, 31 Jan 2013 19:11:25 +0000 (19:11 +0000)]
Remove Web Intents code from WebKit2
https://bugs.webkit.org/show_bug.cgi?id=108506

Reviewed by Simon Fraser.

Source/WebKit2:

Since nobody builds with Web Intents enabled anymore, and since the code is going to
be removed from WebCore, remove it from WebKit2.

* Shared/API/c/WKBase.h:
* Shared/APIClientTraits.cpp:
(WebKit):
* Shared/APIObject.h:
* Shared/IntentData.cpp: Removed.
* Shared/IntentData.h: Removed.
* Shared/IntentServiceInfo.cpp: Removed.
* Shared/IntentServiceInfo.h: Removed.
* Shared/WebIntentServiceInfo.cpp: Removed.
* Shared/WebIntentServiceInfo.h: Removed.
* UIProcess/API/C/WKAPICast.h:
(WebKit):
* UIProcess/API/C/WKIntentData.cpp: Removed.
* UIProcess/API/C/WKIntentData.h: Removed.
* UIProcess/API/C/WKIntentServiceInfo.cpp: Removed.
* UIProcess/API/C/WKIntentServiceInfo.h: Removed.
* UIProcess/API/C/WKPage.cpp:
* UIProcess/API/C/WKPage.h:
* UIProcess/WebIntentData.cpp: Removed.
* UIProcess/WebIntentData.h: Removed.
* UIProcess/WebLoaderClient.cpp:
* UIProcess/WebLoaderClient.h:
(WebKit):
(WebLoaderClient):
* UIProcess/WebPageProxy.cpp:
* UIProcess/WebPageProxy.h:
(WebKit):
(WebPageProxy):
* UIProcess/WebPageProxy.messages.in:
* UIProcess/WebProcessProxy.cpp:
* UIProcess/WebProcessProxy.h:
(WebProcessProxy):
* WebProcess/InjectedBundle/API/c/WKBundleAPICast.h:
(WebKit):
* WebProcess/InjectedBundle/API/c/WKBundleIntent.cpp: Removed.
* WebProcess/InjectedBundle/API/c/WKBundleIntent.h: Removed.
* WebProcess/InjectedBundle/API/c/WKBundleIntentRequest.cpp: Removed.
* WebProcess/InjectedBundle/API/c/WKBundleIntentRequest.h: Removed.
* WebProcess/InjectedBundle/API/c/WKBundlePage.cpp:
* WebProcess/InjectedBundle/API/c/WKBundlePage.h:
* WebProcess/InjectedBundle/InjectedBundleIntent.cpp: Removed.
* WebProcess/InjectedBundle/InjectedBundleIntent.h: Removed.
* WebProcess/InjectedBundle/InjectedBundleIntentRequest.cpp: Removed.
* WebProcess/InjectedBundle/InjectedBundleIntentRequest.h: Removed.
* WebProcess/InjectedBundle/InjectedBundlePageLoaderClient.cpp:
* WebProcess/InjectedBundle/InjectedBundlePageLoaderClient.h:
(WebKit):
(InjectedBundlePageLoaderClient):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:
(WebFrameLoaderClient):
* WebProcess/WebPage/WebFrame.cpp:
* WebProcess/WebPage/WebFrame.h:
(WebCore):
(WebKit):
(WebFrame):
* WebProcess/WebPage/WebPage.cpp:
* WebProcess/WebPage/WebPage.h:
(WebCore):
(WebKit):
(WebPage):
* WebProcess/WebPage/WebPage.messages.in:
* WebProcess/WebProcess.cpp:
* WebProcess/WebProcess.h:
(WebCore):
(WebProcess):
* WebProcess/WebProcess.messages.in:

Tools:

* WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
* WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:
(WTR::InjectedBundlePage::InjectedBundlePage):
* WebKitTestRunner/InjectedBundle/InjectedBundlePage.h:
(InjectedBundlePage):
* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
* WebKitTestRunner/InjectedBundle/TestRunner.h:
(TestRunner):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141448 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoLack of a log level string should not obliterate compiled in logging channel state.
beidson@apple.com [Thu, 31 Jan 2013 19:06:51 +0000 (19:06 +0000)]
Lack of a log level string should not obliterate compiled in logging channel state.
https://bugs.webkit.org/show_bug.cgi?id=108502

Reviewed by Alexey Proskuryakov and Sam Weinig.

Source/WebKit/mac:

* Misc/WebKitLogging.m: If there's no log level string, leave the channel state alone.

Source/WebKit2:

* Platform/mac/Logging.mac.mm:
(WebKit::initializeLogChannel): If there's no log level string, leave the channel state alone.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141447 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[BlackBerry] CookieParser should ignore invalid/unsupported attributes instead of...
commit-queue@webkit.org [Thu, 31 Jan 2013 19:06:28 +0000 (19:06 +0000)]
[BlackBerry] CookieParser should ignore invalid/unsupported attributes instead of ignoring the cookie
https://bugs.webkit.org/show_bug.cgi?id=108494

PR 287832
Patch by Otto Derek Cheung <otcheung@rim.com> on 2013-01-31
Reviewed by Rob Buis.
Internally Reviewed by Liam Quinn.

A short-term fix for CookieParser to not blow up when a Set-Cookie header has an invalid
or unsupported cookie attribute.

We will ignore the attribute only, not the entire cookie.

Also, making sure the parser won't replace values of valid attributes when evaluating an unsupported
cookie with the same length and same first character as a supported attribute.

Tested using opera cookie and browser.swlab.rim.net cookies test suite. Tested using regular sites that
has a login mechanism.

* platform/blackberry/CookieParser.cpp:
(WebCore::CookieParser::parseOneCookie):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141446 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago WebProcess sandbox profile overhaul.
ap@apple.com [Thu, 31 Jan 2013 19:05:30 +0000 (19:05 +0000)]
    WebProcess sandbox profile overhaul.

        Reviewed by Sam Weinig.

        Moves some rules together by susbystem for easier maintenance.

        Addresses <rdar://problem/9276393>, <rdar://problem/10844321>, <rdar://problem/12408537>,
        <rdar://problem/12558524>.

        * WebProcess/com.apple.WebProcess.sb.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141445 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] MediaStream API: Rename WebMediaStreamDescriptor and WebMediaStreamCompone...
tommyw@google.com [Thu, 31 Jan 2013 19:01:31 +0000 (19:01 +0000)]
[chromium] MediaStream API: Rename WebMediaStreamDescriptor and WebMediaStreamComponent to WebMediaStream and WebMediaStreamTrack
https://bugs.webkit.org/show_bug.cgi?id=108458

Source/Platform:

Only renames, no other code changes.

Reviewed by Adam Barth.

* Platform.gypi:
* chromium/public/WebMediaStream.h: Copied from Source/Platform/chromium/public/WebMediaStreamDescriptor.h.
(WebCore):
(WebKit):
(WebMediaStream):
(ExtraData):
(WebKit::WebMediaStream::ExtraData::~ExtraData):
(WebKit::WebMediaStream::WebMediaStream):
(WebKit::WebMediaStream::~WebMediaStream):
(WebKit::WebMediaStream::operator=):
(WebKit::WebMediaStream::isNull):
* chromium/public/WebMediaStreamCenter.h:
(WebKit):
(WebMediaStreamCenter):
(WebKit::WebMediaStreamCenter::didAddMediaStreamTrack):
(WebKit::WebMediaStreamCenter::didRemoveMediaStreamTrack):
* chromium/public/WebMediaStreamCenterClient.h:
(WebKit):
(WebMediaStreamCenterClient):
* chromium/public/WebMediaStreamComponent.h:
* chromium/public/WebMediaStreamDescriptor.h:
* chromium/public/WebMediaStreamTrack.h: Added.
(WebCore):
(WebKit):
(WebMediaStreamTrack):
(WebKit::WebMediaStreamTrack::WebMediaStreamTrack):
(WebKit::WebMediaStreamTrack::~WebMediaStreamTrack):
(WebKit::WebMediaStreamTrack::operator=):
(WebKit::WebMediaStreamTrack::isNull):
* chromium/public/WebRTCPeerConnectionHandler.h:
(WebKit):
(WebRTCPeerConnectionHandler):
* chromium/public/WebRTCPeerConnectionHandlerClient.h:
(WebKit):
(WebRTCPeerConnectionHandlerClient):
* chromium/public/WebRTCStatsRequest.h:
(WebKit):
(WebRTCStatsRequest):

Source/WebCore:

Reviewed by Adam Barth.

Only renames, no other code changes.

* WebCore.gypi:
* platform/chromium/support/WebMediaStream.cpp: Added.
(WebKit::WebMediaStream::WebMediaStream):
(WebKit):
(WebKit::WebMediaStream::reset):
(WebKit::WebMediaStream::label):
(WebKit::WebMediaStream::id):
(WebKit::WebMediaStream::extraData):
(WebKit::WebMediaStream::setExtraData):
(WebKit::WebMediaStream::audioSources):
(WebKit::WebMediaStream::videoSources):
(WebKit::WebMediaStream::operator=):
(WebKit::WebMediaStream::operator PassRefPtr<WebCore::MediaStreamDescriptor>):
(WebKit::WebMediaStream::operator WebCore::MediaStreamDescriptor*):
(WebKit::WebMediaStream::initialize):
(WebKit::WebMediaStream::assign):
* platform/chromium/support/WebMediaStreamComponent.cpp: Removed.
* platform/chromium/support/WebMediaStreamDescriptor.cpp: Removed.
* platform/chromium/support/WebMediaStreamTrack.cpp: Added.
(WebKit):
(WebKit::WebMediaStreamTrack::WebMediaStreamTrack):
(WebKit::WebMediaStreamTrack::operator=):
(WebKit::WebMediaStreamTrack::initialize):
(WebKit::WebMediaStreamTrack::reset):
(WebKit::WebMediaStreamTrack::operator PassRefPtr<MediaStreamComponent>):
(WebKit::WebMediaStreamTrack::operator MediaStreamComponent*):
(WebKit::WebMediaStreamTrack::isEnabled):
(WebKit::WebMediaStreamTrack::id):
(WebKit::WebMediaStreamTrack::stream):
(WebKit::WebMediaStreamTrack::source):
(WebKit::WebMediaStreamTrack::assign):
* platform/chromium/support/WebRTCStatsRequest.cpp:
(WebKit::WebRTCStatsRequest::stream):
(WebKit::WebRTCStatsRequest::component):
* platform/mediastream/chromium/MediaStreamCenterChromium.cpp:
(WebCore::MediaStreamCenterChromium::didCreateMediaStream):
(WebCore::MediaStreamCenterChromium::stopLocalMediaStream):
(WebCore::MediaStreamCenterChromium::addMediaStreamTrack):
(WebCore::MediaStreamCenterChromium::removeMediaStreamTrack):
* platform/mediastream/chromium/MediaStreamCenterChromium.h:
(WebKit):
(MediaStreamCenterChromium):
* platform/mediastream/chromium/RTCPeerConnectionHandlerChromium.cpp:
(WebCore::RTCPeerConnectionHandlerChromium::didAddRemoteStream):
(WebCore::RTCPeerConnectionHandlerChromium::didRemoveRemoteStream):
* platform/mediastream/chromium/RTCPeerConnectionHandlerChromium.h:
(WebKit):
(RTCPeerConnectionHandlerChromium):

Source/WebKit/chromium:

Reviewed by Adam Barth.

Only renames, no other code changes.

* public/WebMediaStreamRegistry.h:
(WebKit):
(WebMediaStreamRegistry):
* public/WebUserMediaRequest.h:
(WebKit):
(WebUserMediaRequest):
* src/WebMediaStreamRegistry.cpp:
(WebKit::WebMediaStreamRegistry::lookupMediaStreamDescriptor):
* src/WebUserMediaRequest.cpp:
(WebKit::WebUserMediaRequest::requestSucceeded):

Tools:

Reviewed by Adam Barth.

Only renames, no other code changes.

* DumpRenderTree/chromium/MockWebMediaStreamCenter.cpp:
(MockWebMediaStreamCenter::didEnableMediaStreamTrack):
(MockWebMediaStreamCenter::didDisableMediaStreamTrack):
(MockWebMediaStreamCenter::didAddMediaStreamTrack):
(MockWebMediaStreamCenter::didRemoveMediaStreamTrack):
(MockWebMediaStreamCenter::didStopLocalMediaStream):
(MockWebMediaStreamCenter::didCreateMediaStream):
* DumpRenderTree/chromium/MockWebMediaStreamCenter.h:
(MockWebMediaStreamCenter):
* DumpRenderTree/chromium/MockWebRTCPeerConnectionHandler.cpp:
(MockWebRTCPeerConnectionHandler::addStream):
(MockWebRTCPeerConnectionHandler::removeStream):
(MockWebRTCPeerConnectionHandler::getStats):
* DumpRenderTree/chromium/MockWebRTCPeerConnectionHandler.h:
(MockWebRTCPeerConnectionHandler):
* DumpRenderTree/chromium/WebUserMediaClientMock.cpp:
(UserMediaRequestTask::UserMediaRequestTask):
(WebUserMediaClientMock::requestUserMedia):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141444 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoObjective-C API: Fix insertion of values greater than the max index allowed by the...
mhahnenberg@apple.com [Thu, 31 Jan 2013 18:54:56 +0000 (18:54 +0000)]
Objective-C API: Fix insertion of values greater than the max index allowed by the spec
https://bugs.webkit.org/show_bug.cgi?id=108264

Reviewed by Oliver Hunt.

Fixed a bug, added a test to the API tests, cleaned up some code.

* API/JSValue.h: Changed some of the documentation on setValue:atIndex: to indicate that
setting values at indices greater than UINT_MAX - 1 wont' affect the length of JS arrays.
* API/JSValue.mm:
(-[JSValue valueAtIndex:]): We weren't returning when we should have been.
(-[JSValue setValue:atIndex:]): Added a comment about why we do the early check for being larger than UINT_MAX.
(objectToValueWithoutCopy): Removed two redundant cases that were already checked previously.
* API/tests/testapi.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141443 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed trivial build fix: Pre C++11 the use of
hausmann@webkit.org [Thu, 31 Jan 2013 18:54:33 +0000 (18:54 +0000)]
Unreviewed trivial build fix: Pre C++11 the use of
>> in nested templates is ambiguous in the grammar and
requires the insertion of a space here. Since these files are
not Mac specific we don't require C++11 yet and a space
fixes the build.

* Platform/CoreIPC/Connection.h:
(Connection):
* Shared/ChildProcessProxy.h:
(ChildProcessProxy):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141442 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoREGRESSION (r141402): incorrectly set injected bundle path when running GTK unit...
zandobersek@gmail.com [Thu, 31 Jan 2013 18:50:06 +0000 (18:50 +0000)]
REGRESSION (r141402): incorrectly set injected bundle path when running GTK unit tests
https://bugs.webkit.org/show_bug.cgi?id=108496

Reviewed by Philippe Normand.

* Scripts/run-gtk-tests:
(TestRunner.__init__): Add the build type as a member on the TestRunner interface.
(TestRunner._setup_testing_environment): Use the build type member to determine correct path
to the injected bundle.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141441 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Unreviewed gardening.
fmalita@chromium.org [Thu, 31 Jan 2013 18:46:22 +0000 (18:46 +0000)]
[Chromium] Unreviewed gardening.

* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141440 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[EFL] Disable Web Intents
commit-queue@webkit.org [Thu, 31 Jan 2013 18:43:41 +0000 (18:43 +0000)]
[EFL] Disable Web Intents
https://bugs.webkit.org/show_bug.cgi?id=108457

Patch by Christophe Dumez <dchris@gmail.com> on 2013-01-31
Reviewed by Alexey Proskuryakov.

.:

Turn off WEB_INTENTS flag in EFL CMake project.

* Source/cmake/OptionsEfl.cmake:

Source/WebKit:

Remove intents files from EFL CMake project.

* PlatformEfl.cmake:

Source/WebKit/efl:

Remove code related to Web Intents from EFL
WebKit.

* WebCoreSupport/DumpRenderTreeSupportEfl.cpp:
* WebCoreSupport/DumpRenderTreeSupportEfl.h:
* WebCoreSupport/FrameLoaderClientEfl.cpp:
* WebCoreSupport/FrameLoaderClientEfl.h:
(FrameLoaderClientEfl):
* ewk/EWebKit.h:
* ewk/ewk_frame.cpp:
* ewk/ewk_frame.h:
* ewk/ewk_frame_private.h:
* ewk/ewk_intent.cpp: Removed.
* ewk/ewk_intent.h: Removed.
* ewk/ewk_intent_private.h: Removed.
* ewk/ewk_intent_request.cpp: Removed.
* ewk/ewk_intent_request.h: Removed.

Source/WebKit2:

Remove code related to Web Intents from EFL
WebKit2.

* CMakeLists.txt:
* PlatformEfl.cmake:
* UIProcess/API/efl/EWebKit2.h:
* UIProcess/API/efl/EwkViewCallbacks.h:
* UIProcess/API/efl/ewk_intent.cpp: Removed.
* UIProcess/API/efl/ewk_intent.h: Removed.
* UIProcess/API/efl/ewk_intent_private.h: Removed.
* UIProcess/API/efl/ewk_intent_service.cpp: Removed.
* UIProcess/API/efl/ewk_intent_service.h: Removed.
* UIProcess/API/efl/ewk_intent_service_private.h: Removed.
* UIProcess/API/efl/ewk_view.cpp:
* UIProcess/API/efl/ewk_view.h:
* UIProcess/API/efl/tests/resources/intent-request.html: Removed.
* UIProcess/API/efl/tests/resources/intent-service.html: Removed.
* UIProcess/API/efl/tests/test_ewk2_intents.cpp: Removed.
* UIProcess/efl/PageLoadClientEfl.cpp:
(WebKit::PageLoadClientEfl::PageLoadClientEfl):
* UIProcess/efl/PageLoadClientEfl.h:
(PageLoadClientEfl):

Tools:

Remove EFL DRT code related to Web intents.

* DumpRenderTree/efl/DumpRenderTreeChrome.cpp:
(DumpRenderTreeChrome::createView):
(DumpRenderTreeChrome::resetDefaultsToConsistentValues):
(DumpRenderTreeChrome::onFrameCreated):
* DumpRenderTree/efl/DumpRenderTreeChrome.h:
(DumpRenderTreeChrome):
* DumpRenderTree/efl/TestRunnerEfl.cpp:
(TestRunner::sendWebIntentResponse):
(TestRunner::deliverWebIntent):
* Scripts/webkitperl/FeatureList.pm: Turn off WEB_INTENTS flag
for EFL port.

LayoutTests:

Skip webintents/ test cases for EFL port now that
the feature is disabled.

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141439 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoIndexedDB: Remove WebKit API for old onSuccess/onUpgradeNeeded
alecflett@chromium.org [Thu, 31 Jan 2013 18:39:24 +0000 (18:39 +0000)]
IndexedDB: Remove WebKit API for old onSuccess/onUpgradeNeeded
https://bugs.webkit.org/show_bug.cgi?id=108399

Reviewed by Dimitri Glazkov.

Cleanup now that chromium proxies the new signatures through.

* src/IDBCallbacksProxy.cpp:
(WebKit::IDBCallbacksProxy::onSuccess):
(WebKit::IDBCallbacksProxy::onUpgradeNeeded):
* src/IDBCallbacksProxy.h:
(IDBCallbacksProxy):
* src/WebIDBCallbacksImpl.cpp:
* src/WebIDBCallbacksImpl.h:
(WebIDBCallbacksImpl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141438 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Unreviewed, land baselines with textarea resizer enabled.
tony@chromium.org [Thu, 31 Jan 2013 18:38:46 +0000 (18:38 +0000)]
[chromium] Unreviewed, land baselines with textarea resizer enabled.
https://bugs.webkit.org/show_bug.cgi?id=108383

* platform/chromium-linux/fast/block/float/overhanging-tall-block-expected.png: Removed.
* platform/chromium-mac-lion/editing/inserting/4960120-1-expected.png:
* platform/chromium-mac-lion/editing/pasteboard/pasting-tabs-expected.png:
* platform/chromium-mac-lion/fast/dom/HTMLTextAreaElement/reset-textarea-expected.png:
* platform/chromium-mac-lion/fast/forms/basic-textareas-expected.png:
* platform/chromium-mac-lion/fast/forms/basic-textareas-quirks-expected.png:
* platform/chromium-mac-lion/fast/forms/form-element-geometry-expected.png:
* platform/chromium-mac-lion/fast/forms/linebox-overflow-in-textarea-padding-expected.png:
* platform/chromium-mac-lion/fast/forms/negativeLineHeight-expected.png:
* platform/chromium-mac-lion/fast/forms/placeholder-position-expected.png:
* platform/chromium-mac-lion/fast/forms/textAreaLineHeight-expected.png:
* platform/chromium-mac-lion/fast/forms/textarea-align-expected.png:
* platform/chromium-mac-lion/fast/forms/textarea-placeholder-pseudo-style-expected.png:
* platform/chromium-mac-lion/fast/forms/textarea-placeholder-visibility-1-expected.png:
* platform/chromium-mac-lion/fast/forms/textarea-placeholder-visibility-2-expected.png:
* platform/chromium-mac-lion/fast/forms/textarea-scroll-height-expected.png:
* platform/chromium-mac-lion/fast/forms/textarea-scrollbar-expected.png:
* platform/chromium-mac-lion/fast/forms/textarea-scrolled-type-expected.png:
* platform/chromium-mac-lion/fast/forms/textarea-setinnerhtml-expected.png:
* platform/chromium-mac-lion/fast/forms/textarea-width-expected.png:
* platform/chromium-mac-lion/fast/overflow/overflow-x-y-expected.png:
* platform/chromium-mac-lion/fast/parser/entity-comment-in-textarea-expected.png:
* platform/chromium-mac-lion/fast/parser/open-comment-in-textarea-expected.png:
* platform/chromium-mac-lion/fast/replaced/width100percent-textarea-expected.png:
* platform/chromium-mac-lion/fast/table/003-expected.png:
* platform/chromium-mac-lion/fast/text/international/rtl-white-space-pre-wrap-expected.png:
* platform/chromium-mac-lion/fast/text/international/unicode-bidi-plaintext-in-textarea-expected.png:
* platform/chromium-mac-lion/tables/mozilla/bugs/bug194024-expected.png:
* platform/chromium-mac-lion/tables/mozilla/bugs/bug30559-expected.png:
* platform/chromium-mac-lion/tables/mozilla/bugs/bug30692-expected.png:
* platform/chromium-mac-snowleopard/editing/inserting/4960120-1-expected.png:
* platform/chromium-mac-snowleopard/editing/pasteboard/pasting-tabs-expected.png:
* platform/chromium-mac-snowleopard/fast/dom/HTMLTextAreaElement/reset-textarea-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/basic-textareas-quirks-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/form-element-geometry-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/linebox-overflow-in-textarea-padding-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/negativeLineHeight-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/placeholder-position-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/textAreaLineHeight-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/textarea-align-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/textarea-placeholder-pseudo-style-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/textarea-placeholder-visibility-1-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/textarea-placeholder-visibility-2-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/textarea-scroll-height-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/textarea-scrollbar-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/textarea-scrolled-type-expected.png:
* platform/chromium-mac-snowleopard/fast/forms/textarea-width-expected.png:
* platform/chromium-mac-snowleopard/fast/overflow/overflow-x-y-expected.png:
* platform/chromium-mac-snowleopard/fast/parser/entity-comment-in-textarea-expected.png:
* platform/chromium-mac-snowleopard/fast/parser/open-comment-in-textarea-expected.png:
* platform/chromium-mac-snowleopard/fast/replaced/width100percent-textarea-expected.png:
* platform/chromium-mac-snowleopard/fast/table/003-expected.png:
* platform/chromium-mac-snowleopard/fast/text/international/rtl-white-space-pre-wrap-expected.png:
* platform/chromium-mac-snowleopard/fast/text/international/unicode-bidi-plaintext-in-textarea-expected.png:
* platform/chromium-mac-snowleopard/tables/mozilla/bugs/bug194024-expected.png:
* platform/chromium-mac-snowleopard/tables/mozilla/bugs/bug30559-expected.png:
* platform/chromium-mac-snowleopard/tables/mozilla/bugs/bug30692-expected.png:
* platform/chromium-mac/editing/inserting/4960120-1-expected.png:
* platform/chromium-mac/editing/pasteboard/pasting-tabs-expected.png:
* platform/chromium-mac/fast/dom/HTMLTextAreaElement/reset-textarea-expected.png:
* platform/chromium-mac/fast/forms/basic-textareas-expected.png:
* platform/chromium-mac/fast/forms/basic-textareas-quirks-expected.png:
* platform/chromium-mac/fast/forms/form-element-geometry-expected.png:
* platform/chromium-mac/fast/forms/linebox-overflow-in-textarea-padding-expected.png:
* platform/chromium-mac/fast/forms/negativeLineHeight-expected.png:
* platform/chromium-mac/fast/forms/placeholder-position-expected.png:
* platform/chromium-mac/fast/forms/textAreaLineHeight-expected.png:
* platform/chromium-mac/fast/forms/textarea-align-expected.png:
* platform/chromium-mac/fast/forms/textarea-placeholder-pseudo-style-expected.png:
* platform/chromium-mac/fast/forms/textarea-placeholder-visibility-1-expected.png:
* platform/chromium-mac/fast/forms/textarea-placeholder-visibility-2-expected.png:
* platform/chromium-mac/fast/forms/textarea-scroll-height-expected.png:
* platform/chromium-mac/fast/forms/textarea-scrollbar-expected.png:
* platform/chromium-mac/fast/forms/textarea-scrolled-type-expected.png:
* platform/chromium-mac/fast/forms/textarea-setinnerhtml-expected.png:
* platform/chromium-mac/fast/forms/textarea-width-expected.png:
* platform/chromium-mac/fast/overflow/overflow-x-y-expected.png:
* platform/chromium-mac/fast/parser/entity-comment-in-textarea-expected.png:
* platform/chromium-mac/fast/parser/open-comment-in-textarea-expected.png:
* platform/chromium-mac/fast/repaint/textarea-set-disabled-expected.png:
* platform/chromium-mac/fast/replaced/width100percent-textarea-expected.png:
* platform/chromium-mac/fast/table/003-expected.png:
* platform/chromium-mac/fast/text/international/rtl-white-space-pre-wrap-expected.png:
* platform/chromium-mac/fast/text/international/unicode-bidi-plaintext-in-textarea-expected.png:
* platform/chromium-mac/tables/mozilla/bugs/bug194024-expected.png:
* platform/chromium-mac/tables/mozilla/bugs/bug30559-expected.png:
* platform/chromium-mac/tables/mozilla/bugs/bug30692-expected.png:
* platform/chromium-win/editing/inserting/4960120-1-expected.png:
* platform/chromium-win/editing/pasteboard/pasting-tabs-expected.png:
* platform/chromium-win/fast/block/float/overhanging-tall-block-expected.png:
* platform/chromium-win/fast/dom/HTMLTextAreaElement/reset-textarea-expected.png:
* platform/chromium-win/fast/forms/basic-textareas-expected.png:
* platform/chromium-win/fast/forms/basic-textareas-quirks-expected.png:
* platform/chromium-win/fast/forms/form-element-geometry-expected.png:
* platform/chromium-win/fast/forms/linebox-overflow-in-textarea-padding-expected.png:
* platform/chromium-win/fast/forms/negativeLineHeight-expected.png:
* platform/chromium-win/fast/forms/placeholder-position-expected.png:
* platform/chromium-win/fast/forms/textAreaLineHeight-expected.png:
* platform/chromium-win/fast/forms/textarea-align-expected.png:
* platform/chromium-win/fast/forms/textarea-placeholder-pseudo-style-expected.png:
* platform/chromium-win/fast/forms/textarea-placeholder-visibility-1-expected.png:
* platform/chromium-win/fast/forms/textarea-placeholder-visibility-2-expected.png:
* platform/chromium-win/fast/forms/textarea-scroll-height-expected.png:
* platform/chromium-win/fast/forms/textarea-scrollbar-expected.png:
* platform/chromium-win/fast/forms/textarea-scrolled-type-expected.png:
* platform/chromium-win/fast/forms/textarea-setinnerhtml-expected.png:
* platform/chromium-win/fast/forms/textarea-width-expected.png:
* platform/chromium-win/fast/overflow/overflow-x-y-expected.png:
* platform/chromium-win/fast/parser/entity-comment-in-textarea-expected.png:
* platform/chromium-win/fast/parser/open-comment-in-textarea-expected.png:
* platform/chromium-win/fast/replaced/width100percent-textarea-expected.png:
* platform/chromium-win/fast/table/003-expected.png:
* platform/chromium-win/fast/text/international/rtl-white-space-pre-wrap-expected.png:
* platform/chromium-win/fast/text/international/unicode-bidi-plaintext-in-textarea-expected.png:
* platform/chromium-win/tables/mozilla/bugs/bug194024-expected.png:
* platform/chromium-win/tables/mozilla/bugs/bug30559-expected.png:
* platform/chromium-win/tables/mozilla/bugs/bug30692-expected.png:
* platform/chromium/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141437 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: searching for <u> in elements panel finds all tags containing "u"
commit-queue@webkit.org [Thu, 31 Jan 2013 18:34:11 +0000 (18:34 +0000)]
Web Inspector: searching for <u> in elements panel finds all tags containing "u"
https://bugs.webkit.org/show_bug.cgi?id=108176

Patch by Dmitry Zvorygin <zvorygin@chromium.org> on 2013-01-31
Reviewed by Pavel Feldman.

When searching for tag name check that tag name must either start from
search query, or must end with it.

Source/WebCore:

* inspector/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::performSearch):

LayoutTests:

* inspector/elements/elements-panel-search-expected.txt:
* inspector/elements/elements-panel-search.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141436 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Disable force_static_libs_as_shared in developer builds if wk2 is enabled
hausmann@webkit.org [Thu, 31 Jan 2013 18:32:12 +0000 (18:32 +0000)]
[Qt] Disable force_static_libs_as_shared in developer builds if wk2 is enabled

Reviewed by Jocelyn Turcotte.

This is a temporary workaround during the development of #108471, where we move files around
but some link time dependencies might remain.

* qmake/mkspecs/features/default_post.prf:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141435 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed GTK gardening.
zandobersek@gmail.com [Thu, 31 Jan 2013 18:24:04 +0000 (18:24 +0000)]
Unreviewed GTK gardening.

Adding a flaky crasher expectation for fast/css-generated-content/block-and-box-hit-testing.html.
Removing failure expectations for SVGViewSpec tests.

* platform/gtk-wk1/TestExpectations:
* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141434 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago <rdar://problem/12695827> PPT: Make loading file URLs work with a sandboxed...
ap@apple.com [Thu, 31 Jan 2013 18:22:23 +0000 (18:22 +0000)]
    <rdar://problem/12695827> PPT: Make loading file URLs work with a sandboxed NetworkProcess

        Address review comments.

        * NetworkProcess/mac/com.apple.WebKit.NetworkProcess.sb.in:
        * WebProcess/com.apple.WebProcess.sb.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141433 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoStop using OutgoingMessage
andersca@apple.com [Thu, 31 Jan 2013 18:18:55 +0000 (18:18 +0000)]
Stop using OutgoingMessage
https://bugs.webkit.org/show_bug.cgi?id=108495

Reviewed by Sam Weinig.

* Platform/CoreIPC/Connection.cpp:
(CoreIPC::Connection::sendMessage):
(CoreIPC::Connection::sendOutgoingMessages):
* Platform/CoreIPC/Connection.h:
(Connection):
* Shared/ChildProcessProxy.cpp:
(WebKit::ChildProcessProxy::~ChildProcessProxy):
(WebKit::ChildProcessProxy::sendMessage):
(WebKit::ChildProcessProxy::didFinishLaunching):
* Shared/ChildProcessProxy.h:
(ChildProcessProxy):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141432 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoDatabaseContext needs to outlive the ScriptExecutionContext.
mark.lam@apple.com [Thu, 31 Jan 2013 18:12:23 +0000 (18:12 +0000)]
DatabaseContext needs to outlive the ScriptExecutionContext.
https://bugs.webkit.org/show_bug.cgi?id=108355.

Reviewed by Geoffrey Garen.

Added a RefPtr<DatabaseContext> in ScriptExecutionContext to keep the
DatabaseContext alive until after ScriptExecutionContext destructs.

No new tests.

* Modules/webdatabase/DatabaseContext.cpp:
(WebCore::DatabaseContext::DatabaseContext):
* dom/ScriptExecutionContext.cpp:
(WebCore::ScriptExecutionContext::setDatabaseContext):
* dom/ScriptExecutionContext.h:
(ScriptExecutionContext):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141431 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Make it possible to compile WebKit2 Qt related files without access to internal...
hausmann@webkit.org [Thu, 31 Jan 2013 18:04:16 +0000 (18:04 +0000)]
[Qt] Make it possible to compile WebKit2 Qt related files without access to internal WK2 C++ API
https://bugs.webkit.org/show_bug.cgi?id=108472

Reviewed by Andreas Kling.

.:

When linking WebKit2, also link the WebKit2QML module.

* Source/api.pri:

Source/WebKit2:

Add a new module to the qmake build system that represents the part of the WebKit2 Qt integration
that doesn't depend on WebKit2 internals.

Changed qwebnavigationhistory.cpp to not use any internal headers and compile it as part of the
internals-free module.

* Target.pri:
* UIProcess/API/qt/qwebnavigationhistory.cpp:
* WebKit2.pro:
* WebKit2QML.pri: Added.

Tools:

Add a new module to the qmake build system that represents the part of the WebKit2 Qt integration
that doesn't depend on WebKit2 internals.

* qmake/mkspecs/features/webkit_modules.prf:
* qmake/mkspecs/modules/webkit2qml.prf: Copied from Source/WebKit2/WebKit2.pro.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141430 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoOn Linux, can't get spelling suggestions without selection when unified text checker...
g.czajkowski@samsung.com [Thu, 31 Jan 2013 17:59:22 +0000 (17:59 +0000)]
On Linux, can't get spelling suggestions without selection when unified text checker is enabled
https://bugs.webkit.org/show_bug.cgi?id=107684

Reviewed by Ryosuke Niwa.

No new tests, covered by context-menu-suggestions.html.

* editing/Editor.cpp:
(WebCore::Editor::guessesForMisspelledOrUngrammatical):
For Linux, extend selection range to the closest word to get
the spelling suggestions when unified text checker is enabled.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141429 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] Remove dead transitional code from WebViewImpl
commit-queue@webkit.org [Thu, 31 Jan 2013 17:44:08 +0000 (17:44 +0000)]
[chromium] Remove dead transitional code from WebViewImpl
https://bugs.webkit.org/show_bug.cgi?id=107889

Patch by James Robinson <jamesr@chromium.org> on 2013-01-31
Reviewed by Adam Barth.

Source/WebKit/chromium:

The chromium side of this landed at r178256 and seems stable.

* public/WebWidget.h:
(WebKit::WebWidget::setCompositorSurfaceReady):
* src/WebViewImpl.cpp:
(WebKit::WebViewImpl::WebViewImpl):
(WebKit::WebViewImpl::~WebViewImpl):
(WebKit):
(WebKit::WebViewImpl::setIsTransparent):
(WebKit::WebViewImpl::setIsAcceleratedCompositingActive):
* src/WebViewImpl.h:
* tests/ScrollingCoordinatorChromiumTest.cpp:
(WebKit::FakeWebViewClient::initializeLayerTreeView):
(FakeWebViewClient):
(WebKit::FakeWebViewClient::layerTreeView):
(WebKit::ScrollingCoordinatorChromiumTest::ScrollingCoordinatorChromiumTest):
(ScrollingCoordinatorChromiumTest):

Tools:

ScrollingCoordinatorChromiumTests need to initialize compositing, so its implementation of
WebWidgetClient has to support the new compositor initialization path.

* DumpRenderTree/chromium/WebViewHost.cpp:
(WebViewHost::initializeLayerTreeView):
(WebViewHost::setWebWidget):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141428 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Skia] Update SkiaImageFilterBuilder's brightness matrix to match Chromium's implemen...
fmalita@chromium.org [Thu, 31 Jan 2013 17:36:46 +0000 (17:36 +0000)]
[Skia] Update SkiaImageFilterBuilder's brightness matrix to match Chromium's implementation
https://bugs.webkit.org/show_bug.cgi?id=107841

Reviewed by Stephen White.

Update Skia's brightness filter to match the spec after http://trac.webkit.org/changeset/139770.

No new tests: covered by existing tests (currently disabled pending rebaseline).

* platform/graphics/filters/skia/SkiaImageFilterBuilder.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141427 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: show cursor location in DTE
commit-queue@webkit.org [Thu, 31 Jan 2013 17:17:30 +0000 (17:17 +0000)]
Web Inspector: show cursor location in DTE
https://bugs.webkit.org/show_bug.cgi?id=108478

Patch by Andrey Lushnikov <lushnikov@chromium.org> on 2013-01-31
Reviewed by Pavel Feldman.

Add a new status bar element to SourceFrame to display current
DefaultTextEditor cursor position.

No new tests.

* English.lproj/localizedStrings.js:
* inspector/front-end/DefaultTextEditor.js:
(WebInspector.DefaultTextEditor.prototype.copyRange):
* inspector/front-end/SourceFrame.js:
(WebInspector.SourceFrame):
(WebInspector.SourceFrame.prototype.statusBarItems):
(WebInspector.SourceFrame.prototype.selectionChanged):
(WebInspector.SourceFrame.prototype._updateSourcePosition):
* inspector/front-end/TextEditor.js:
(WebInspector.TextEditor.prototype.copyRange):
* inspector/front-end/inspector.css:
(.source-frame-position):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141426 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWebKitTestRunner needs an implementation of eventSender.beginDragWithFiles
jberlin@webkit.org [Thu, 31 Jan 2013 16:45:44 +0000 (16:45 +0000)]
WebKitTestRunner needs an implementation of eventSender.beginDragWithFiles
https://bugs.webkit.org/show_bug.cgi?id=64285

* platform/wk2/TestExpectations:
Skip another test that uses eventSender.beginDragWithFiles.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141425 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: overlay highlight in DTE gets messed up when zoom factor changes.
commit-queue@webkit.org [Thu, 31 Jan 2013 16:33:06 +0000 (16:33 +0000)]
Web Inspector: overlay highlight in DTE gets messed up when zoom factor changes.
https://bugs.webkit.org/show_bug.cgi?id=108486

Patch by Andrey Lushnikov <lushnikov@chromium.org> on 2013-01-31
Reviewed by Pavel Feldman.

Repaint overlay highlight when zoom factor changes.

No new tests.

* inspector/front-end/DefaultTextEditor.js:
(WebInspector.TextEditorMainPanel.prototype.highlightRegex):
(WebInspector.TextEditorMainPanel.prototype.removeHighlight):
(WebInspector.TextEditorMainPanel.prototype.highlightRange):
(WebInspector.TextEditorMainPanel.prototype._repaintLineRowsAffectedByHighlightDescriptors):
(WebInspector.TextEditorMainPanel.prototype.resize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141424 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Unreviewed gardening. Skip a failing test after r141269.
kadam@inf.u-szeged.hu [Thu, 31 Jan 2013 16:13:35 +0000 (16:13 +0000)]
[Qt] Unreviewed gardening. Skip a failing test after r141269.
https://bugs.webkit.org/show_bug.cgi?id=108481.
Renamed inspector/editor/text-editor-ctrl-movements.html after r141399.

* platform/qt/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141423 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Fix minor memory leak in slot execution
hausmann@webkit.org [Thu, 31 Jan 2013 15:07:34 +0000 (15:07 +0000)]
[Qt] Fix minor memory leak in slot execution

Reviewed by Allan Sandfeld Jensen.

Don't leak the "length" string when executing a slot in JavaScript.

* bridge/qt/qt_runtime.cpp:
(JSC::Bindings::QtConnectionObject::execute):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141420 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed trivial build fix: Add missing virtual destructor to
hausmann@webkit.org [Thu, 31 Jan 2013 14:59:29 +0000 (14:59 +0000)]
Unreviewed trivial build fix: Add missing virtual destructor to
LayerTreeRendererClient. Otherwise the build with -Werror breaks, which
complains (rightly so) that we're deleting a sub-class where the super class
doesn't have a virtual destructor.

* UIProcess/CoordinatedGraphics/LayerTreeRenderer.h:
(WebKit::LayerTreeRendererClient::~LayerTreeRendererClient):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141419 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoAllow blocking of IndexedDB in third-party contexts
mkwst@chromium.org [Thu, 31 Jan 2013 14:57:55 +0000 (14:57 +0000)]
Allow blocking of IndexedDB in third-party contexts
https://bugs.webkit.org/show_bug.cgi?id=94171

Reviewed by Jochen Eisinger.

Source/WebCore:

This patch ensures that the origin of the top window is passed into
SecurityOrigin::canAccessDatabase when working with IndexedDB. Giving
SecurityOrigin access to this data means that it can properly check
whether the database is being opened in a third-party context, and
therefore properly enforce the third-party access checks that were
added in http://trac.webkit.org/changeset/125736.

Third-party checks are added to IDBFactory::open,
IDBFactory::deleteDatabase, and IDBFactory::getDatabaseNames; each will
now throw a SECURITY_ERR exception when access in a third-party context
if third-party access checks are enabled.

To make this process slightly more clear, and avoid some ugly casting
logic, this patch adds a 'topOrigin'  method to ScriptExecutionContext,
and implements it on both WorkerContext and Document.

Tests: http/tests/security/cross-origin-indexeddb-allowed.html
       http/tests/security/cross-origin-indexeddb.html
       http/tests/security/cross-origin-worker-indexeddb-allowed.html
       http/tests/security/cross-origin-worker-indexeddb.html

* Modules/indexeddb/IDBFactory.cpp:
(WebCore::IDBFactory::getDatabaseNames):
(WebCore::IDBFactory::openInternal):
(WebCore::IDBFactory::deleteDatabase):
    Grab the SecurityOrigin of the current context's top-level origin,
    and pass it to SecurityOrigin::canAccessDatabase to ensure that
    access checks are properly applied to these three methods.
* dom/Document.cpp:
(WebCore::Document::topOrigin):
(WebCore):
* dom/Document.h:
(Document):
* dom/ScriptExecutionContext.h:
(ScriptExecutionContext):
    Add a topOrigin() method to ScriptExecutionContext, and implement it
    on Document in order to give callers access to the top document's
    SecurityOrigin without casting ScriptExecutionContext.
* workers/WorkerContext.h:
    Change the existing topOrigin() method to override the new method
    on ScriptExecutionContext.

LayoutTests:

Add tests to ensure that IndexedDB can be blocked in a third-party
context in both normal documents and in workers. These tests are
modeled after the existing cross-origin-websql* tests; it might be
possible to reuse some code in the future.

* http/tests/security/cross-origin-indexeddb-allowed-expected.txt: Added.
* http/tests/security/cross-origin-indexeddb-allowed.html: Added.
* http/tests/security/cross-origin-indexeddb-expected.txt: Added.
* http/tests/security/cross-origin-indexeddb.html: Added.
* http/tests/security/cross-origin-worker-indexeddb-allowed-expected.txt: Added.
* http/tests/security/cross-origin-worker-indexeddb-allowed.html: Added.
* http/tests/security/cross-origin-worker-indexeddb-expected.txt: Added.
* http/tests/security/cross-origin-worker-indexeddb.html: Added.
* http/tests/security/resources/cross-origin-iframe-for-indexeddb.html: Added.
* http/tests/security/resources/cross-origin-iframe-for-worker-indexeddb.html: Added.
* http/tests/security/resources/document-for-cross-origin-worker-indexeddb.html: Added.
* http/tests/security/resources/worker-for-indexeddb.js: Added.
(self.onmessage):
    Add exciting new tests, with more boilerplate than I expected!
* platform/efl/TestExpectations:
* platform/mac-snowleopard/TestExpectations:
* platform/mac/TestExpectations:
* platform/qt/TestExpectations:
* platform/win/TestExpectations:
* platform/wincairo/TestExpectations:
    Skip these IndexedDB tests on platforms where the feature isn't
    enabled.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141418 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: [Network] Add cookie column to show presence of request/response cookies.
commit-queue@webkit.org [Thu, 31 Jan 2013 14:52:58 +0000 (14:52 +0000)]
Web Inspector: [Network] Add cookie column to show presence of request/response cookies.
https://bugs.webkit.org/show_bug.cgi?id=107441

Patch by Eugene Klyuchnikov <eustas@chromium.org> on 2013-01-31
Reviewed by Pavel Feldman.

In some use cases only responses that contain "set-cookie" headers
are interesting.

* English.lproj/localizedStrings.js: Added "Set-Cookies" string.
* inspector/front-end/NetworkPanel.js:
(WebInspector.NetworkLogView.prototype._createTable): Added columns.
(WebInspector.NetworkLogView.prototype._createSortingFunctions): Ditto.
(WebInspector.NetworkDataGridNode.prototype.createCells): Ditto.
(WebInspector.NetworkDataGridNode.prototype.refreshRequest): Ditto.
(WebInspector.NetworkDataGridNode.prototype._refreshCookiesCell): Added.
(WebInspector.NetworkDataGridNode.prototype._refreshSetCookiesCell): Added.
(WebInspector.NetworkDataGridNode.RequestCookiesCountComparator): Added.
(WebInspector.NetworkDataGridNode.ResponseCookiesCountComparator): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141417 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Unreviewed gardening. Skip two compositing tests, because They hit assertion...
kadam@inf.u-szeged.hu [Thu, 31 Jan 2013 14:30:49 +0000 (14:30 +0000)]
[Qt] Unreviewed gardening. Skip two compositing tests, because They hit assertion fail after 140999.
https://bugs.webkit.org/show_bug.cgi?id=108257.

* platform/qt/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141416 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: Exception in HeapSnapshotView.js
aandrey@chromium.org [Thu, 31 Jan 2013 14:15:28 +0000 (14:15 +0000)]
Web Inspector: Exception in HeapSnapshotView.js
https://bugs.webkit.org/show_bug.cgi?id=108456

Reviewed by Yury Semikhatsky.

Listen "profile added" events in the HeapSnapshotView targeted only to the Heap Snapshot type.

* inspector/front-end/HeapSnapshotView.js:
(WebInspector.HeapSnapshotView.prototype._onProfileHeaderAdded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141415 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: do not set any textContent in overlay highlight spans
commit-queue@webkit.org [Thu, 31 Jan 2013 14:10:36 +0000 (14:10 +0000)]
Web Inspector: do not set any textContent in overlay highlight spans
https://bugs.webkit.org/show_bug.cgi?id=108460

Patch by Andrey Lushnikov <lushnikov@chromium.org> on 2013-01-31
Reviewed by Pavel Feldman.

Source/WebCore:

Do not set any text content for overlay highlight spans to implicitly
set its height. Instead, measure its height in the same manner as its
done for "width" and "left" and set it explicitly.

No new tests: no change in behaviour.

* inspector/front-end/DefaultTextEditor.js:
(WebInspector.TextEditorMainPanel.prototype._appendOverlayHighlight):
(WebInspector.TextEditorMainPanel.ElementMetrics):
* inspector/front-end/textEditor.css:
(.text-editor-token-highlight): Update highlight to correspond to
slightly different elements height.

LayoutTests:

Replace "height" pixel value with "<number>" to avoid difference in
test expectations on varioius platforms and correct test expectations.

* inspector/editor/editor-test.js:
(initialize_EditorTests.InspectorTest.dumpEditorHTML):
* inspector/editor/text-editor-highlight-api-expected.txt:
* inspector/editor/text-editor-highlight-token-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141414 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Qt] Box shadows on a transparency layer is very slow
allan.jensen@digia.com [Thu, 31 Jan 2013 14:07:58 +0000 (14:07 +0000)]
[Qt] Box shadows on a transparency layer is very slow
https://bugs.webkit.org/show_bug.cgi?id=107547

Reviewed by Noam Rosenthal.

Include the window boundaries in the clip returned by GraphicsContext,
since QPainter may remember clips larger than the destination, but
ShadowBlur uses the clipBounds to determine the size of the shadow layer.

* platform/graphics/qt/GraphicsContextQt.cpp:
(WebCore::GraphicsContext::clipBounds):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141413 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoCoordinated Graphics: view the debug border/repaint count of the non composited layer.
commit-queue@webkit.org [Thu, 31 Jan 2013 14:07:22 +0000 (14:07 +0000)]
Coordinated Graphics: view the debug border/repaint count of the non composited layer.
https://bugs.webkit.org/show_bug.cgi?id=108401

Patch by Seulgi Kim <seulgikim@company100.net> on 2013-01-31
Reviewed by Noam Rosenthal.

Make non-compositing layer draw debug border and show repaint counter
accroding to settings.

* WebProcess/WebPage/CoordinatedGraphics/CoordinatedLayerTreeHost.cpp:
(WebKit::CoordinatedLayerTreeHost::CoordinatedLayerTreeHost):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141412 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: highlight backend languages as scripts
pfeldman@chromium.org [Thu, 31 Jan 2013 14:05:06 +0000 (14:05 +0000)]
Web Inspector: highlight backend languages as scripts
https://bugs.webkit.org/show_bug.cgi?id=108336

Reviewed by Vsevolod Vlasov.

Most languages have structure similar to js, so enabling default highlighter for them won't hurt.

* inspector/front-end/FileSystemWorkspaceProvider.js:
(WebInspector.FileSystemWorkspaceProvider.prototype._contentTypeForPath):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141411 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnjumble ChangeLog authors after r141407.
akling@apple.com [Thu, 31 Jan 2013 13:48:12 +0000 (13:48 +0000)]
Unjumble ChangeLog authors after r141407.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141410 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: [Profiles] show launcher view upon deleting last profile type's header
aandrey@chromium.org [Thu, 31 Jan 2013 13:42:42 +0000 (13:42 +0000)]
Web Inspector: [Profiles] show launcher view upon deleting last profile type's header
https://bugs.webkit.org/show_bug.cgi?id=108468

Reviewed by Pavel Feldman.

We should show launcher view and hide profile type sidebar section upon deleting last profile type's header from the sidebar.
Right now we only show launcher view upon deleting the last profile header (of any profile type).

* inspector/front-end/ProfilesPanel.js:
(WebInspector.ProfilesPanel.prototype._removeProfileHeader):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141409 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Chromium] Unreviewed gardening.
fmalita@chromium.org [Thu, 31 Jan 2013 13:39:59 +0000 (13:39 +0000)]
[Chromium] Unreviewed gardening.

* platform/chromium-mac-snowleopard/fast/forms/basic-textareas-expected.png:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141408 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoVector should consult allocator about ideal size when choosing capacity.
akling@apple.com [Thu, 31 Jan 2013 13:34:32 +0000 (13:34 +0000)]
Vector should consult allocator about ideal size when choosing capacity.
<http://webkit.org/b/108410>
<rdar://problem/13124002>

Source/JavaScriptCore:

Patch by Filip Pizlo <fpizlo@apple.com> on 2013-01-30
Reviewed by Benjamin Poulain.

Remove assertion about Vector capacity that won't hold anymore since capacity()
may not be what you passed to reserveCapacity().

* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::CodeBlock):

Source/WTF:

Reviewed by Benjamin Poulain.

Added WTF::fastMallocGoodSize(), a workalike/wrapper for OS X's malloc_good_size().
It returns the actual size of the block that will get allocated for a given byte size.

Vector's internal buffer now checks with the allocator if the resulting allocation
could actually house more objects and updates its capacity to make use of the space.

* wtf/Deque.h:
(WTF::::expandCapacity):
* wtf/FastMalloc.cpp:
(WTF::fastMallocGoodSize):
* wtf/FastMalloc.h:
* wtf/Vector.h:
(WTF::VectorBufferBase::allocateBuffer):
(WTF::VectorBufferBase::tryAllocateBuffer):
(WTF::VectorBufferBase::reallocateBuffer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141407 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: [Network] Columns are misaligned as first row goes out of sight.
commit-queue@webkit.org [Thu, 31 Jan 2013 13:25:41 +0000 (13:25 +0000)]
Web Inspector: [Network] Columns are misaligned as first row goes out of sight.
https://bugs.webkit.org/show_bug.cgi?id=107933

Patch by Eugene Klyuchnikov <eustas@chromium.org> on 2013-01-31
Reviewed by Pavel Feldman.

For some reason if "td" elements in first table row have "display: none"
style, then column width calculation breaks.

* inspector/front-end/networkLogView.css:
(.network-log-grid.data-grid tr.offscreen > td > div):
Hide content instead of "td" element itself.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141406 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoformMethod to have empty string as default value and 'get' as invalid.
rgf748@motorola.com [Thu, 31 Jan 2013 13:01:58 +0000 (13:01 +0000)]
formMethod to have empty string as default value and 'get' as invalid.
https://bugs.webkit.org/show_bug.cgi?id=108263

Reviewed by Kent Tamura.

The spec says formmethod should only have an invalid value default, not a missing value default.
Spec: http://www.whatwg.org/specs/web-apps/current-work/#form-submission-0
      http://www.w3.org/html/wg/drafts/html/master/forms.html#attr-fs-formmethod

Source/WebCore:

Test: fast/forms/formmethod-attribute-test.html

* html/HTMLFormControlElement.cpp: For the missing formMethod attr return empty string.
(WebCore::HTMLFormControlElement::formMethod):

LayoutTests:

* fast/forms/formmethod-attribute-test-expected.txt: Added.
* fast/forms/formmethod-attribute-test.html: Added.
* fast/forms/submit-form-attributes-expected.txt:
* fast/forms/submit-form-attributes.html: Modified test to behave as expected.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141405 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[Efl][WebGL] Add better support to track and free XResources.
commit-queue@webkit.org [Thu, 31 Jan 2013 12:47:22 +0000 (12:47 +0000)]
[Efl][WebGL] Add better support to track and free XResources.
https://bugs.webkit.org/show_bug.cgi?id=107397

Patch by Kondapally Kalyan <kalyan.kondapally@intel.com> on 2013-01-20
Reviewed by Noam Rosenthal.

We leak Memory during config selection as we dont free temporary
allocated visuals, config etc. This patch ensures that we dont leak any
memory.

Covered by existing WebGL tests.

* platform/graphics/surfaces/egl/EGLSurface.cpp:
(WebCore::EGLWindowTransportSurface::EGLWindowTransportSurface):
* platform/graphics/surfaces/glx/GLXConfigSelector.h:
(WebCore::GLXConfigSelector::GLXConfigSelector):
(WebCore::GLXConfigSelector::visualInfo):
(WebCore::GLXConfigSelector::pBufferContextConfig):
(WebCore::GLXConfigSelector::surfaceContextConfig):
(WebCore::GLXConfigSelector::createSurfaceConfig):
(GLXConfigSelector):
Removed XVisualInfo as member variable. Fixed memory leaks.

* platform/graphics/surfaces/glx/GLXSurface.cpp:
(WebCore::GLXTransportSurface::GLXTransportSurface):
* platform/graphics/surfaces/glx/OwnPtrX11.h: Added.
(OwnPtrX11):
(WebCore::OwnPtrX11::OwnPtrX11):
(WebCore::OwnPtrX11::~OwnPtrX11):
(WebCore::OwnPtrX11::operator=):
(WebCore::OwnPtrX11::operator T*):
(WebCore::OwnPtrX11::operator->):
(WebCore::OwnPtrX11::get):
Calls XFree on pointer to memory allocated by X when the class goes out of scope.
OwnPtr cannot be used here as GLXFBConfig is a pointer to an array of pointers.

* platform/graphics/surfaces/glx/X11WindowResources.cpp:
(WebCore::X11OffScreenWindow::X11OffScreenWindow):
(WebCore::X11OffScreenWindow::~X11OffScreenWindow):
(WebCore):
(WebCore::X11OffScreenWindow::createOffScreenWindow):
(WebCore::X11OffScreenWindow::destroyWindow):
(WebCore::X11OffScreenWindow::nativeSharedDisplay):
* platform/graphics/surfaces/glx/X11WindowResources.h:
(X11OffScreenWindow):
Removed unused code. Fixed memory leaks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141404 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: [Canvas] remove invalid canvas profile trace logs upon frame navigation
aandrey@chromium.org [Thu, 31 Jan 2013 12:45:12 +0000 (12:45 +0000)]
Web Inspector: [Canvas] remove invalid canvas profile trace logs upon frame navigation
https://bugs.webkit.org/show_bug.cgi?id=108454

Reviewed by Pavel Feldman.

Source/WebCore:

When canvas profile trace logs stored in the backend get destroyed (for example, on frame navigation), send an event to frontend and update the UI.
Drive-by: replace String types to corresponding TypeBuilder::Canvas::* typedefs.
* inspector/InjectedScriptCanvasModule.cpp:
(WebCore::InjectedScriptCanvasModule::captureFrame):
(WebCore::InjectedScriptCanvasModule::startCapturing):
(WebCore::InjectedScriptCanvasModule::callStartCapturingFunction):
(WebCore::InjectedScriptCanvasModule::stopCapturing):
(WebCore::InjectedScriptCanvasModule::dropTraceLog):
(WebCore::InjectedScriptCanvasModule::callVoidFunctionWithTraceLogIdArgument):
(WebCore::InjectedScriptCanvasModule::traceLog):
(WebCore::InjectedScriptCanvasModule::replayTraceLog):
(WebCore::InjectedScriptCanvasModule::resourceInfo):
(WebCore::InjectedScriptCanvasModule::resourceState):
* inspector/InjectedScriptCanvasModule.h:
(InjectedScriptCanvasModule):
* inspector/Inspector.json:
* inspector/InspectorCanvasAgent.cpp:
(WebCore::InspectorCanvasAgent::dropTraceLog):
(WebCore::InspectorCanvasAgent::stopCapturing):
(WebCore::InspectorCanvasAgent::getTraceLog):
(WebCore::InspectorCanvasAgent::replayTraceLog):
(WebCore::InspectorCanvasAgent::getResourceInfo):
(WebCore::InspectorCanvasAgent::getResourceState):
(WebCore::InspectorCanvasAgent::frameNavigated):
* inspector/InspectorCanvasAgent.h:
(InspectorCanvasAgent):
* inspector/InspectorPageAgent.cpp:
(WebCore::InspectorPageAgent::hasIdForFrame):
(WebCore):
* inspector/InspectorPageAgent.h:
(InspectorPageAgent):
* inspector/front-end/CanvasProfileView.js:
(WebInspector.CanvasProfileType.prototype._runSingleFrameCapturing):
(WebInspector.CanvasProfileType.prototype._startFrameCapturing):
(WebInspector.CanvasProfileType.prototype._didStartCapturingFrame):
(WebInspector.CanvasProfileType.prototype._traceLogsRemoved):
(WebInspector.CanvasDispatcher.prototype.contextCreated):
(WebInspector.CanvasDispatcher.prototype.traceLogsRemoved):
(WebInspector.CanvasProfileHeader):
(WebInspector.CanvasProfileHeader.prototype.frameId):

LayoutTests:

* inspector/profiler/canvas-profiler-test.js:
(initialize_CanvasWebGLProfilerTest.InspectorTest.enableCanvasAgent.InspectorBackend.registerCanvasDispatcher):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141403 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[GTK] run-gtk-tests does not respect the -d argument
kov@webkit.org [Thu, 31 Jan 2013 12:43:15 +0000 (12:43 +0000)]
[GTK] run-gtk-tests does not respect the -d argument
https://bugs.webkit.org/show_bug.cgi?id=107822

Reviewed by Philippe Normand.

* Scripts/run-gtk-tests:
(TestRunner.__init__): use the value for the debug option to decide what
build_type to request a path for.
* gtk/common.py:
(get_build_path): now accepts a build_type argument instead of trying both
Release and Debug in order, defaults to release.
(build_path): takes and passes a build_type argument, no default value.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141402 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: DTE doesn't highlight words if the selected one is the last in the...
commit-queue@webkit.org [Thu, 31 Jan 2013 12:04:09 +0000 (12:04 +0000)]
Web Inspector: DTE doesn't highlight words if the selected one is the last in the line
https://bugs.webkit.org/show_bug.cgi?id=108344

Patch by Andrey Lushnikov <lushnikov@chromium.org> on 2013-01-31
Reviewed by Pavel Feldman.

Source/WebCore:

Fix error in _isWord function which made an erroneous line-end check.

Improved test: inspector/editor/text-editor-highlight-token.html

* inspector/front-end/DefaultTextEditor.js:
(WebInspector.TextEditorMainPanel.TokenHighlighter.prototype._isWord):

LayoutTests:

Add a test to cover specific case which caused the bug.

* inspector/editor/text-editor-highlight-token-expected.txt:
* inspector/editor/text-editor-highlight-token.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141401 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoInjectedBundle is being built even with --disable-webkit2
commit-queue@webkit.org [Thu, 31 Jan 2013 12:01:16 +0000 (12:01 +0000)]
InjectedBundle is being built even with --disable-webkit2
https://bugs.webkit.org/show_bug.cgi?id=108364

Patch by PaweĊ‚ Forysiuk <tuxator@o2.pl> on 2013-01-31
Reviewed by Gustavo Noronha Silva.

* GNUmakefile.am: Wrap Injected bundle with ENABLE_WEBKIT2 condition

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141400 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: text-editor-ctrl-movements.html timeouts on mac
commit-queue@webkit.org [Thu, 31 Jan 2013 12:01:01 +0000 (12:01 +0000)]
Web Inspector: text-editor-ctrl-movements.html timeouts on mac
https://bugs.webkit.org/show_bug.cgi?id=108440

Patch by Andrey Lushnikov <lushnikov@chromium.org> on 2013-01-31
Reviewed by Pavel Feldman.

Use "alt-arrows" key shortcuts to jump over words instead of
"ctrl-arrows" on Mac platform.

* inspector/editor/text-editor-word-jumps-expected.txt: Renamed from LayoutTests/inspector/editor/text-editor-ctrl-movements-expected.txt.
* inspector/editor/text-editor-word-jumps.html: Renamed from LayoutTests/inspector/editor/text-editor-ctrl-movements.html.
* platform/chromium/TestExpectations: Do not skip this test anymore.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141399 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed, Qt build fix after r141265.
philn@webkit.org [Thu, 31 Jan 2013 11:44:58 +0000 (11:44 +0000)]
Unreviewed, Qt build fix after r141265.

* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
Avoid uninitialized variable error when NATIVE_FULLSCREEN_VIDEO is disabled.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141398 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUpdate Christophe Dumez's email address
mikhail.pozdnyakov@intel.com [Thu, 31 Jan 2013 11:44:49 +0000 (11:44 +0000)]
Update Christophe Dumez's email address
https://bugs.webkit.org/show_bug.cgi?id=108453

Patch by Christophe Dumez <dchris@gmail.com> on 2013-01-31
Reviewed by Kenneth Rohde Christiansen.

Update my email address in committers.py.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141397 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: test that references from DOM nodes to event listeners are presented...
yurys@chromium.org [Thu, 31 Jan 2013 11:25:30 +0000 (11:25 +0000)]
Web Inspector: test that references from DOM nodes to event listeners are presented in heap snapshots
https://bugs.webkit.org/show_bug.cgi?id=108322

Reviewed by Vsevolod Vlasov.

Test that links from DOM node wrappers to event listener functions are presented
in heap snapshots.

* inspector-protocol/heap-profiler/heap-snapshot-with-event-listener-expected.txt: Added.
* inspector-protocol/heap-profiler/heap-snapshot-with-event-listener.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141396 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoClick on a label element won't select input[type=date]
tkent@chromium.org [Thu, 31 Jan 2013 11:17:40 +0000 (11:17 +0000)]
Click on a label element won't select input[type=date]
https://bugs.webkit.org/show_bug.cgi?id=108428

Reviewed by Kentaro Hara.

Source/WebCore:

date/time input types with INPUT_MULTIPLE_FIELDS_UI are not
mouse-focusable. <input> as a shadow host never gets focus and
sub-fields in its shadow tree are focusable. Howevever, the click
handling of <label> checked isMouseFocusable. We introduce new function
isFocusableByClickOnLabel to HTMLElement. It's default implementation is
same as isMouseFocusable, and we add special handling for
BaseMultipleFieldsDateAndTimeInputType::isFocusableByClickOnLabel.

Also, date/time input types without INPUT_MULTIPLE_FIELDS_UI were not
mouse-focusable by a mistake. It should be mouse-focusable.

Test: fast/forms/date/date-click-on-label.html

* html/HTMLElement.cpp:
(WebCore::HTMLElement::isFocusableByClickOnLabel):
Added. Just calls isMouseFocusable.
* html/HTMLElement.h:
(HTMLElement): Declare isFocusableByClickOnLabel.
* html/HTMLLabelElement.cpp:
(WebCore::HTMLLabelElement::defaultEventHandler):
Calls isFocusableByClickOnLabel instead of isMouseFocusable.

* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::isFocusableByClickOnLabel):
Delegate to InputType::isFocusableByClickOnLabel.
* html/HTMLInputElement.h:
(HTMLInputElement): Declare isFocusableByClickOnLabel.
* html/InputType.cpp:
(WebCore::InputType::isFocusableByClickOnLabel):
Added. Calls isMouseFocusable.
* html/InputType.h:
(InputType): Declare isFocusableByClickOnLabel.

* html/BaseMultipleFieldsDateAndTimeInputType.cpp:
(WebCore::BaseMultipleFieldsDateAndTimeInputType::isFocusableByClickOnLabel):
Added. Use the same logic with textfields.
* html/BaseMultipleFieldsDateAndTimeInputType.h:
(BaseMultipleFieldsDateAndTimeInputType): Declare isFocusableByClickOnLabel.
* html/BaseChooserOnlyDateAndTimeInputType.cpp:
(WebCore::BaseChooserOnlyDateAndTimeInputType::isMouseFocusable):
Added. Use the same logic with textfields.
* html/BaseChooserOnlyDateAndTimeInputType.h:
(BaseChooserOnlyDateAndTimeInputType): Declare isMouseFocusable.

LayoutTests:

* fast/forms/date/date-click-on-label-expected.txt: Added.
* fast/forms/date/date-click-on-label.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141395 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoRename WheelEvent::Granularity to WheelEvent::DeltaMode
haraken@chromium.org [Thu, 31 Jan 2013 11:10:44 +0000 (11:10 +0000)]
Rename WheelEvent::Granularity to WheelEvent::DeltaMode
https://bugs.webkit.org/show_bug.cgi?id=108434

Reviewed by Ryosuke Niwa.

Per the spec, WheelEvent::Granularity should be renamed to WheelEvent::DeltaMode.

Spec: http://www.w3.org/TR/DOM-Level-3-Events/#events-WheelEvent
https://dvcs.w3.org/hg/d4e/raw-file/tip/source_respec.htm#constructor-wheelevent

No tests. No change in behavior.

Source/WebCore:

* dom/WheelEvent.cpp:
(WebCore::WheelEvent::WheelEvent):
(WebCore::WheelEvent::initWheelEvent):
(WebCore::deltaMode):
(WebCore::WheelEventDispatchMediator::WheelEventDispatchMediator):
* dom/WheelEvent.h:
(WebCore::WheelEvent::create):
(WebCore::WheelEvent::deltaMode):
(WheelEvent):
* page/EventHandler.cpp:
(WebCore::wheelGranularityToScrollGranularity):
(WebCore::EventHandler::defaultWheelEventHandler):

Source/WebKit/chromium:

* src/WebInputEventConversion.cpp:
(WebKit::PlatformWheelEventBuilder::PlatformWheelEventBuilder):
(WebKit::WebMouseWheelEventBuilder::WebMouseWheelEventBuilder):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141394 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoWeb Inspector: test that nodes from the same detached DOM tree will get into one...
yurys@chromium.org [Thu, 31 Jan 2013 10:52:41 +0000 (10:52 +0000)]
Web Inspector: test that nodes from the same detached DOM tree will get into one group in heap snapshot
https://bugs.webkit.org/show_bug.cgi?id=108202

Reviewed by Vsevolod Vlasov.

Source/WebCore:

Test: inspector-protocol/heap-profiler/heap-snapshot-with-detached-dom-tree.html

* inspector/front-end/JSHeapSnapshot.js:
(WebInspector.JSHeapSnapshotNode.prototype.className): removed unnecessary i18n.

LayoutTests:

Test that JS wrappers for all DOM nodes from the same detached DOM tree will get into
single "Detached DOM Tree" entry in the JS heap snapshot.

* http/tests/inspector-protocol/resources/InspectorTest.js:
(InspectorTest.importScript):
* inspector-protocol/heap-profiler/heap-snapshot-with-detached-dom-tree-expected.txt: Added.
* inspector-protocol/heap-profiler/heap-snapshot-with-detached-dom-tree.html: Added. I started
with writing simplified version of WebInspector.JSHeapSnapshot just for tests
but it soon it became clear that we would need to reimplement too much functionality
of WebInspector.JSHeapSnapshot so I decided not to reinvent the wheel and just import
original heap snapshot model.
* inspector-protocol/heap-profiler/resources/heap-snapshot-common.js: Added.
(InspectorTest.takeHeapSnapshot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141393 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years ago[chromium] move runModalBeforeUnloadDialog to TestRunner library
jochen@chromium.org [Thu, 31 Jan 2013 10:02:04 +0000 (10:02 +0000)]
[chromium] move runModalBeforeUnloadDialog to TestRunner library
https://bugs.webkit.org/show_bug.cgi?id=108442

Reviewed by Adam Barth.

* DumpRenderTree/chromium/TestRunner/public/WebTestProxy.h:
(WebTestRunner::WebTestProxy::runModalBeforeUnloadDialog):
* DumpRenderTree/chromium/TestRunner/src/WebTestProxy.cpp:
(WebTestRunner::WebTestProxyBase::runModalBeforeUnloadDialog):
* DumpRenderTree/chromium/WebViewHost.cpp:
* DumpRenderTree/chromium/WebViewHost.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141392 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7 years agoUnreviewed. Bump Chromium dependency to 179332
yurys@chromium.org [Thu, 31 Jan 2013 09:46:02 +0000 (09:46 +0000)]
Unreviewed. Bump Chromium dependency to 179332

* DEPS:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@141391 268f45cc-cd09-0410-ab3c-d52691b4dbfc