WebKit-https.git
8 years agoUpdate perf.webkit.org json example
zoltan@webkit.org [Sat, 8 Mar 2014 22:14:47 +0000 (22:14 +0000)]
Update perf.webkit.org json example
https://bugs.webkit.org/show_bug.cgi?id=129907

Reviewed by Andreas Kling.

The current example is not valid json syntax. I fixed the syntax errors and indented the code properly.

* ReadMe.md:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165338 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoDon't check for shm_open on librt on OpenBSD, lib doesnt exist.
commit-queue@webkit.org [Sat, 8 Mar 2014 22:09:54 +0000 (22:09 +0000)]
Don't check for shm_open on librt on OpenBSD, lib doesnt exist.
https://bugs.webkit.org/show_bug.cgi?id=129973

Patch by Landry Breuil <landry@openbsd.org> on 2014-03-08
Reviewed by Martin Robinson.

* Source/autotools/FindDependencies.m4:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165337 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove unnecessary casting in NetworkProcessMainUnix.cpp
jaepark@webkit.org [Sat, 8 Mar 2014 22:07:49 +0000 (22:07 +0000)]
Remove unnecessary casting in NetworkProcessMainUnix.cpp
https://bugs.webkit.org/show_bug.cgi?id=129968

Reviewed by Andreas Kling.

* NetworkProcess/unix/NetworkProcessMainUnix.cpp:
Since socket is an int, int casting is not necessary.

(WebKit::NetworkProcessMain):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165336 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoLots of cleanups for the GTK+ TestExpectations
mrobinson@webkit.org [Sat, 8 Mar 2014 21:49:27 +0000 (21:49 +0000)]
Lots of cleanups for the GTK+ TestExpectations

Updated TestExpectations, adding bug standins where appropriate, skipping tests that
expected to fail and prone to timeout or crash. Removed some unused expectations. Added
new baselines for some tests.

* platform/gtk-wk1/TestExpectations: Updated expectations.
* platform/gtk-wk2/TestExpectations: Ditto.
* platform/gtk/TestExpectations: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165335 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago32-bit x86 handleUncaughtException returns to wrong location after a stack overflow.
mark.lam@apple.com [Sat, 8 Mar 2014 19:23:01 +0000 (19:23 +0000)]
32-bit x86 handleUncaughtException returns to wrong location after a stack overflow.
<https://webkit.org/b/129969>

Reviewed by Geoffrey Garen.

The 32-bit version of handleUncaughtException was missing the handling of an
edge case for stack overflows where the current frame may already be the
sentinel frame.  This edge case was handled in the 64-bit version.  The fix
is to bring the 32-bit version up to parity.

* jit/JIT.cpp:
(JSC::JIT::privateCompile):
* llint/LowLevelInterpreter32_64.asm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165334 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd and update some test expectations for GTK+
mrobinson@webkit.org [Sat, 8 Mar 2014 19:12:31 +0000 (19:12 +0000)]
Add and update some test expectations for GTK+

* platform/gtk/TestExpectations: These all seem to be failing because of the same bug.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165333 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWeb Inspector: update-InspectorBackendCommands.rb is putting legacy protocol files...
commit-queue@webkit.org [Sat, 8 Mar 2014 13:31:13 +0000 (13:31 +0000)]
Web Inspector: update-InspectorBackendCommands.rb is putting legacy protocol files in the wrong place
https://bugs.webkit.org/show_bug.cgi?id=129959

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2014-03-08
Reviewed by Timothy Hatcher.

The script was putting legacy generated files into UserInterface
instead of UserInterface/Protocol.

* Scripts/update-InspectorBackendCommands.rb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165332 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoLoad source maps and their resources asynchronously.
timothy@apple.com [Sat, 8 Mar 2014 12:56:11 +0000 (12:56 +0000)]
Load source maps and their resources asynchronously.

https://bugs.webkit.org/show_bug.cgi?id=112071

Reviewed by Joseph Pecoraro.

Source/WebCore:

* inspector/InspectorPageAgent.cpp:
(WebCore::InspectorPageAgent::buildObjectForFrameTree):
Honor the hiddenFromInspector flag.

* inspector/InspectorResourceAgent.cpp:
(WebCore::InspectorResourceAgent::willSendRequest):
(WebCore::InspectorResourceAgent::markResourceAsCached):
(WebCore::InspectorResourceAgent::didReceiveResponse):
(WebCore::InspectorResourceAgent::didReceiveData):
(WebCore::InspectorResourceAgent::didFinishLoading):
(WebCore::InspectorResourceAgent::didFailLoading):
(WebCore::InspectorResourceAgent::replayXHR):
(WebCore::InspectorResourceAgent::loadResource):
* inspector/InspectorResourceAgent.h:
Honor the hiddenFromInspector flag by keeping track of
hidden identifiers in a HashSet.

* inspector/protocol/Network.json:
(Network.loadResource): Added.

* platform/network/ResourceRequestBase.h:
(WebCore::ResourceRequestBase::hiddenFromInspector): Added.
(WebCore::ResourceRequestBase::setHiddenFromInspector): Added.
(WebCore::ResourceRequestBase::ResourceRequestBase):
Add a flag to hide the request from the Inspector.

* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::XMLHttpRequest):
(WebCore::XMLHttpRequest::sendForInspector): Added.
(WebCore::XMLHttpRequest::sendForInspectorXHRReplay): Renamed.
(WebCore::XMLHttpRequest::createRequest):
* xml/XMLHttpRequest.h:
Make Inspector loads allow cross-origins and hide their request
from the Inspector itself.

Source/WebInspectorUI:

* UserInterface/Controllers/SourceMapManager.js:
(WebInspector.SourceMapManager.prototype.loadAndParseSourceMap):
(WebInspector.SourceMapManager.prototype.downloadSourceMap):
(WebInspector.SourceMapManager.prototype.sourceMapLoaded):
(WebInspector.SourceMapManager.prototype._loadAndParseSourceMap):
Use NetworkAgent.loadResource.

* UserInterface/Models/SourceMapResource.js:
(WebInspector.SourceMapResource.prototype.requestContentFromBackend.sourceMapResourceLoaded):
(WebInspector.SourceMapResource.prototype.createSourceCodeTextRange):
Use NetworkAgent.loadResource.

* UserInterface/Protocol/InspectorWebBackendCommands.js: Updated.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165331 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMountain Lion WK1 test bots are crashing often
mmaxfield@apple.com [Sat, 8 Mar 2014 09:40:28 +0000 (09:40 +0000)]
Mountain Lion WK1 test bots are crashing often
https://bugs.webkit.org/show_bug.cgi?id=129962

Marking the following tests as flakey:
js/dom/cross-frame-bad-time.html
js/dom/create-lots-of-workers.html

Unreviewed.

Patch by Myles C. Maxfield <mmaxfield@apple.com> on 2014-03-08

* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165330 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoPrevent processes spawned by WebProcess from trying to insert the SecItem shim.
commit-queue@webkit.org [Sat, 8 Mar 2014 08:33:03 +0000 (08:33 +0000)]
Prevent processes spawned by WebProcess from trying to insert the SecItem shim.
https://bugs.webkit.org/show_bug.cgi?id=129864

Patch by Aaron Golden <golden@inkling.com> on 2014-03-08
Reviewed by Anders Carlsson.

* WebProcess/EntryPoint/mac/LegacyProcess/WebContentProcessMain.mm:
(WebKit::WebContentProcessMainDelegate::doPreInitializationWork):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165329 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed build fix for GTKbuild after the switch to CMake
mrobinson@webkit.org [Sat, 8 Mar 2014 08:08:46 +0000 (08:08 +0000)]
Unreviewed build fix for GTKbuild after the switch to CMake

* gtk/common.py:
(is_cmake_build): Like r165284, we need to assume CMake by default, since the
test-only bot doesn't have any build files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165328 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WebKit2][iOS] PDF
timothy_horton@apple.com [Sat, 8 Mar 2014 06:36:02 +0000 (06:36 +0000)]
[WebKit2][iOS] PDF
https://bugs.webkit.org/show_bug.cgi?id=129600
<rdar://problem/15260216>

Reviewed by Anders Carlsson.

Add a single custom content provider which installs a very primitive PDF view
in place of the WKContentView if a PDF is loaded in the main frame.

* Configurations/WebKit2.xcconfig:
Link against CorePDF, for UIPDFPageView.

* UIProcess/Cocoa/WKWebViewContentProviderRegistry.mm:
(-[WKWebViewContentProviderRegistry init]):
Initialize the registry, using WKPDFView for all PDF MIME types.

* WebKit2.xcodeproj/project.pbxproj:
* UIProcess/ios/WKPDFView.h: Added.
* UIProcess/ios/WKPDFView.mm: Added.
Add a very simple PDF view based on UIPDFPageView. Additional functionality
can come in the future; right now, we just create one view for each page, and
depend on CATiledLayer to only paint the pages that are being displayed.

* WebCore.exp.in:
* platform/MIMETypeRegistry.cpp:
(WebCore::initializeSupportedImageMIMETypes):
(WebCore::initializePDFMIMETypes):
(WebCore::initializePostScriptMIMETypes):
(WebCore::initializeMIMETypeRegistry):
(WebCore::MIMETypeRegistry::getPDFMIMETypes):
* platform/MIMETypeRegistry.h:
Split PDF and PostScript MIME types, so WebKit2 can retrieve just the PDF ones
without PostScript included.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165327 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoArgumentEncoder and ArgumentDecoder should be fastmalloced.
akling@apple.com [Sat, 8 Mar 2014 06:34:07 +0000 (06:34 +0000)]
ArgumentEncoder and ArgumentDecoder should be fastmalloced.
<https://webkit.org/b/129960>

Reviewed by Ryosuke Niwa.

* Platform/IPC/ArgumentDecoder.h:
* Platform/IPC/ArgumentEncoder.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165326 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix bugs in 32-bit Structure implementation.
mark.lam@apple.com [Sat, 8 Mar 2014 06:26:16 +0000 (06:26 +0000)]
Fix bugs in 32-bit Structure implementation.
<https://webkit.org/b/129947>

Reviewed by Mark Hahnenberg.

Added the loading of the Structure (from the JSCell) before use that was
missing in a few places.  Also added more test cases to equals-masquerader.js.

* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull):
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* llint/LowLevelInterpreter32_64.asm:
* tests/stress/equals-masquerader.js:
(equalsNull):
(notEqualsNull):
(strictEqualsNull):
(strictNotEqualsNull):
(equalsUndefined):
(notEqualsUndefined):
(strictEqualsUndefined):
(strictNotEqualsUndefined):
(isFalsey):
(test):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165325 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoComment in RenderStyle is no longer accurate
mmaxfield@apple.com [Sat, 8 Mar 2014 05:25:07 +0000 (05:25 +0000)]
Comment in RenderStyle is no longer accurate
https://bugs.webkit.org/show_bug.cgi?id=129956

Reviewed by Simon Fraser.

RenderStyle::createStyleInheritingFromPseudoStyle was moved from RenderImage. Before it
was moved, there was an image-specific comment in the function. However, that comment
is both irrelevant to RenderStyle, and incorrect as the function is necessary for
RenderQuotes as well.

No new tests are necessary because there is no behavior change.

* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::createStyleInheritingFromPseudoStyle):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165324 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMark some more tests as failing on WebKitGTK+ and file bugs for them
mrobinson@webkit.org [Sat, 8 Mar 2014 05:10:30 +0000 (05:10 +0000)]
Mark some more tests as failing on WebKitGTK+ and file bugs for them

* platform/gtk/TestExpectations: Update test expectations.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165323 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoTemporarily disable repeat-out-of-bounds stress tests pending fix for 129953.
commit-queue@webkit.org [Sat, 8 Mar 2014 04:09:06 +0000 (04:09 +0000)]
Temporarily disable repeat-out-of-bounds stress tests pending fix for 129953.
https://bugs.webkit.org/show_bug.cgi?id=129954

Patch by Andrew Trick <atrick@apple.com> on 2014-03-07
Reviewed by Filip Pizlo.

* tests/stress/float32-repeat-out-of-bounds.js:
* tests/stress/int8-repeat-out-of-bounds.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165322 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WebRTC] Throw SYNTAX_ERROR when maxRetransmits and maxRetransmitTime are both set...
thiago.lacerda@openbossa.org [Sat, 8 Mar 2014 03:36:23 +0000 (03:36 +0000)]
[WebRTC] Throw SYNTAX_ERROR when maxRetransmits and maxRetransmitTime are both set in RTCDataChannelInit
https://bugs.webkit.org/show_bug.cgi?id=129894

Reviewed by Eric Carlson.

Source/WebCore:

Existing test was updated.

* Modules/mediastream/RTCDataChannel.cpp:
(WebCore::RTCDataChannel::create):

LayoutTests:

* fast/mediastream/RTCPeerConnection-datachannel-expected.txt:
* fast/mediastream/RTCPeerConnection-datachannel.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165321 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago.cfi directives in LowLevelInterpreter.cpp are providing no benefit
msaboff@apple.com [Sat, 8 Mar 2014 03:33:44 +0000 (03:33 +0000)]
.cfi directives in LowLevelInterpreter.cpp are providing no benefit
https://bugs.webkit.org/show_bug.cgi?id=129945

Reviewed by Mark Lam.

Removed .cfi directive.  Verified that stack traces didn't regress in crash reporter
or in lldb.

* llint/LowLevelInterpreter.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165320 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix the GTK+ build after the recent iOS merge.
mrobinson@webkit.org [Sat, 8 Mar 2014 03:05:34 +0000 (03:05 +0000)]
Fix the GTK+ build after the recent iOS merge.

* UIProcess/API/gtk/PageClientImpl.cpp: Make some methods available to non-Cocoa ports and fix some signatures.
* UIProcess/WebPageProxy.h: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165319 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Win] Unreviewed build fix after r165310.
alex.christensen@flexsim.com [Sat, 8 Mar 2014 02:33:06 +0000 (02:33 +0000)]
[Win] Unreviewed build fix after r165310.

* WebKit.vcxproj/WebKit/WebKitDirectX.props:
Replaced Platform macro, which is either "Win32" or "x64"
with PlatformShortName macro, which is either "x86" or "x64".
This is correct for finding the DirectX libraries.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165315 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Replace GraphicsContext3D with Evas_GL in AcceleratedCompositingContext.
ryuan.choi@samsung.com [Sat, 8 Mar 2014 02:23:50 +0000 (02:23 +0000)]
[EFL] Replace GraphicsContext3D with Evas_GL in AcceleratedCompositingContext.
https://bugs.webkit.org/show_bug.cgi?id=129676

Patch by Hyowon Kim <hw1008.kim@samsung.com> on 2014-03-07
Reviewed by Gyuyoung Kim.

TextureMapperGL has its own GraphicsContext3D created by createForCurrentGLContext().
So AcceleratedCompositingContext doesn't need to create and set it to TextureMapperGL.

We now should do 'makeCurrentContext' with a platform specific context before createForCurrentGLContext().
This patch removes GraphicsContext3D related codes in AcceleratedCompositingContext
and adds EvasGLContext/Surface to it instead.

* WebCoreSupport/AcceleratedCompositingContextEfl.cpp:
(WebCore::AcceleratedCompositingContext::initialize):
(WebCore::AcceleratedCompositingContext::resize):
(WebCore::AcceleratedCompositingContext::renderLayers):
* WebCoreSupport/AcceleratedCompositingContextEfl.h:
* ewk/ewk_view.cpp:
* ewk/ewk_view_private.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165314 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed. Fix build on the EFL port after r165303
ryuan.choi@samsung.com [Sat, 8 Mar 2014 02:20:02 +0000 (02:20 +0000)]
Unreviewed. Fix build on the EFL port after r165303

* UIProcess/CoordinatedGraphics/WebView.cpp:
(WebKit::WebView::didCommitLoadForMainFrame): Added new parameters.
* UIProcess/CoordinatedGraphics/WebView.h: Ditto.
* UIProcess/WebPageProxy.h:
* UIProcess/efl/WebViewEfl.cpp: Added dummy didFinishLoadingDataForCustomContentProvider()
(WebKit::WebViewEfl::exitFullScreen):
(WebKit::WebViewEfl::didFinishLoadingDataForCustomContentProvider):
* UIProcess/efl/WebViewEfl.h: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165313 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[iOS WebKit2] Keyboard accessory comes up when I click a link.
enrica@apple.com [Sat, 8 Mar 2014 02:08:51 +0000 (02:08 +0000)]
[iOS WebKit2] Keyboard accessory comes up when I click a link.
https://bugs.webkit.org/show_bug.cgi?id=129949
<rdar://problem/16265305>

Reviewed by Simon Fraser.

We need to create the accessory view only when needed.

* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView requiresAccessoryView]):
(-[WKContentView inputAccessoryView]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165311 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Win] Unreviewed gardening (for 64-bit)
bfulgham@apple.com [Sat, 8 Mar 2014 02:05:18 +0000 (02:05 +0000)]
[Win] Unreviewed gardening (for 64-bit)

* WebKit.vcxproj/WebKit/WebKitDirectX.props: Use Platform macro, rather than
hard-coded path. This allows it to work on 32-bit or 64-bit builds.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165310 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRebaseline some tests for WebKitGTK+
mrobinson@webkit.org [Sat, 8 Mar 2014 01:59:39 +0000 (01:59 +0000)]
Rebaseline some tests for WebKitGTK+

* platform/gtk: Rebaseline tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165309 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAdd the remaining WebKit2 headers to WEBKIT2_HEADERS
andersca@apple.com [Sat, 8 Mar 2014 01:28:21 +0000 (01:28 +0000)]
Add the remaining WebKit2 headers to WEBKIT2_HEADERS
https://bugs.webkit.org/show_bug.cgi?id=129942
<rdar://problem/15920020>

Reviewed by Tim Horton.

* MigrateHeaders.make:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165307 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoContinue hangs when performing for-of over arguments
oliver@apple.com [Sat, 8 Mar 2014 01:17:57 +0000 (01:17 +0000)]
Continue hangs when performing for-of over arguments
https://bugs.webkit.org/show_bug.cgi?id=129915

Reviewed by Geoffrey Garen.

Source/JavaScriptCore:

Put the continue label in the right place

* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitEnumeration):

LayoutTests:

Add tests

* js/for-of-arguments-continue-hang-expected.txt: Added.
* js/for-of-arguments-continue-hang.html: Added.
* js/script-tests/for-of-arguments-continue-hang.js: Added.
(test):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165306 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[iOS WebKit2] Long-press on a link shows just "Copy".
enrica@apple.com [Sat, 8 Mar 2014 01:17:08 +0000 (01:17 +0000)]
[iOS WebKit2] Long-press on a link shows just "Copy".
https://bugs.webkit.org/show_bug.cgi?id=129938
<rdar://problem/16266054>

Reviewed by Benjamin Poulain.

When adding a FIXME comment, I accidentally removed the ! on
the if (!element) condition.

* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::getPositionInformation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165305 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] built-product-archive does not handle GTK+ CMake build
mrobinson@webkit.org [Sat, 8 Mar 2014 01:13:01 +0000 (01:13 +0000)]
[GTK] built-product-archive does not handle GTK+ CMake build
https://bugs.webkit.org/show_bug.cgi?id=129941

* BuildSlaveSupport/built-product-archive:
(archiveBuiltProduct): Fix the build by adding support to built-product-archive for
the CMake build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165304 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WebKit2][iOS] Main-frame custom content providers
timothy_horton@apple.com [Sat, 8 Mar 2014 01:06:31 +0000 (01:06 +0000)]
[WebKit2][iOS] Main-frame custom content providers
https://bugs.webkit.org/show_bug.cgi?id=129809

Reviewed by Dan Bernstein.

Re-introduce custom content providers to WebKit2 (removed in r152841), but
for iOS this time.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView initWithFrame:configuration:]):
Send the initial MIME types that have custom content providers to the WebProcess.

(-[WKWebView dealloc]):
Unregister our page from the content provider registry.

(-[WKWebView _setHasCustomContentView:loadedMIMEType:]):
Install/uninstall the custom content provider. Create a view, if needed,
based on the class that the registry has associated with the given MIME type.
We unparent the WKContentView while the custom content provider is active,
and re-use the existing WKScrollView to contain the custom content provider.

(-[WKWebView _didFinishLoadingDataForCustomContentProviderWithSuggestedFilename:dataReference:]):
Forward loaded data on to the custom content provider for display.

(-[WKWebView _didCommitLayerTree:WebKit::]):
Assert that we aren't getting layer tree commits while using a custom content provider.

(-[WKWebView viewForZoomingInScrollView:]):
If we're using a custom content provider, its view should be used for zooming
instead of the (unparented) web content view.

(-[WKWebView hasContentView]):
(-[WKWebView scrollViewWillBeginZooming:withView:]):
(-[WKWebView scrollViewWillBeginDragging:]):
(-[WKWebView _didFinishScrolling]):
(-[WKWebView _updateVisibleContentRects]):
If we do not have a parented WKContentView because we are using a custom
content provider, we should not forward scroll view related changes to it.

(-[WKWebView _frameOrBoundsChanged]):
Update the minimum size of the custom content provider if the web view size changes.

* UIProcess/API/Cocoa/WKWebViewConfiguration.mm:
(-[WKWebViewConfiguration copyWithZone:]):
(-[WKWebViewConfiguration _contentProviderRegistry]):
(-[WKWebViewConfiguration _setContentProviderRegistry:]):
* UIProcess/API/Cocoa/WKWebViewConfigurationInternal.h: Added.
Add an internal WKWebViewConfiguration property, the WKWebViewContentProviderRegistry.

* UIProcess/API/Cocoa/WKWebViewInternal.h:

* UIProcess/Cocoa/WKWebViewContentProvider.h: Added.
A minimal protocol for informing custom content providers of changes to
the size of the view, the represented data, and the owning scroll view.

* UIProcess/Cocoa/WKWebViewContentProviderRegistry.h: Added.
* UIProcess/Cocoa/WKWebViewContentProviderRegistry.mm: Added.
(-[WKWebViewContentProviderRegistry addPage:]):
(-[WKWebViewContentProviderRegistry removePage:]):
(-[WKWebViewContentProviderRegistry registerProvider:forMIMEType:]):
(-[WKWebViewContentProviderRegistry providerForMIMEType:]):
(-[WKWebViewContentProviderRegistry mimeTypesWithContentProviders]):
New class, keeps track of a mapping from MIME types to UIViews that
conform to the WKWebViewContentProvider protocol.

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::WebPageProxy):
(WebKit::WebPageProxy::didCommitLoadForFrame):
* UIProcess/WebPageProxy.h:
* UIProcess/WebPageProxy.messages.in:

* UIProcess/PageClient.h:
* UIProcess/ios/PageClientImplIOS.h:
* UIProcess/ios/PageClientImplIOS.mm:
(WebKit::PageClientImpl::PageClientImpl):
(WebKit::PageClientImpl::didCommitLoadForMainFrame):
(WebKit::PageClientImpl::didFinishLoadingDataForCustomContentProvider):
* UIProcess/mac/PageClientImpl.h:
* UIProcess/mac/PageClientImpl.mm:
(WebKit::PageClientImpl::didCommitLoadForMainFrame):
(WebKit::PageClientImpl::didFinishLoadingDataForCustomContentProvider):
Add parameters to didCommitLoadForMainFrame, indicating if we should use
a custom content provider to display the main frame, and what MIME type
was loaded in said frame, and add didFinishLoadingDataForCustomContentProvider.
On iOS, forward these to the WKWebView.
Also, give PageClientImplIOS a reference to the WKWebView.

* UIProcess/ios/WKContentView.mm:
(-[WKContentView initWithFrame:context:WebKit::configuration:WebKit::webView:]):

* WebKit2.xcodeproj/project.pbxproj:
* WebProcess/InjectedBundle/API/Cocoa/WKWebProcessPlugInFrame.mm:
(-[WKWebProcessPlugInFrame _hasCustomContentProvider]):
* WebProcess/InjectedBundle/API/Cocoa/WKWebProcessPlugInFramePrivate.h:
Expose whether or not a given frame has a custom content provider to the plug-in.

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::WebFrameLoaderClient):
(WebKit::WebFrameLoaderClient::hasHTMLView):
(WebKit::WebFrameLoaderClient::dispatchDidCommitLoad):
(WebKit::WebFrameLoaderClient::committedLoad):
(WebKit::WebFrameLoaderClient::finishedLoading):
(WebKit::WebFrameLoaderClient::transitionToCommittedFromCachedFrame):
(WebKit::WebFrameLoaderClient::transitionToCommittedForNewPage):
(WebKit::WebFrameLoaderClient::canCachePage):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:
(WebKit::WebFrameLoaderClient::frameHasCustomContentProvider):
Restore WebFrameLoaderClient custom representation code, which ensures
that data is routed correctly (to the UI process, and not to WebCore) and
that we don't try to use the page cache for frames with content providers.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::mainFrameHasCustomContentProvider):
(WebKit::WebPage::addMIMETypeWithCustomContentProvider):
(WebKit::WebPage::shouldUseCustomContentProviderForResponse):
(WebKit::WebPage::canShowMIMEType):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:
Keep a list of MIME types which can be displayed via custom content providers
in WebPage, which will be pushed down from the registry as new providers
are added. Plug-ins still take precendence over custom content providers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165303 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMigrateHeaders.make should remove C SPI includes from WebKit2 headers
andersca@apple.com [Sat, 8 Mar 2014 01:02:30 +0000 (01:02 +0000)]
MigrateHeaders.make should remove C SPI includes from WebKit2 headers
https://bugs.webkit.org/show_bug.cgi?id=129937
<rdar://problem/15920020>

Reviewed by Tim Horton.

Source/WebKit/mac:

* MigrateHeaders.make:
Add WKPreferences.h to WEBKIT2_HEADERS.
Add a sed rule that deletes all lines that start with a # and contain <WebKit/*Ref.h>.

Source/WebKit2:

* WebKit2.xcodeproj/project.pbxproj:
Make WKPreferences.h and all headers it includes public. Update the post processing of WKBase.h
to account for WKBase.h now being a public header again.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165302 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix the regression introduced by r165288
benjamin@webkit.org [Sat, 8 Mar 2014 00:41:36 +0000 (00:41 +0000)]
Fix the regression introduced by r165288
https://bugs.webkit.org/show_bug.cgi?id=129934

Patch by Benjamin Poulain <bpoulain@apple.com> on 2014-03-07
Reviewed by Ryosuke Niwa.

* page/DOMWindow.cpp:
(WebCore::DOMWindow::scrollX):
(WebCore::DOMWindow::scrollY):
We need to return the new value after layout, not the value from the dirty tree.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165301 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoTraversal failure in a direct adjacent chain with tail backtracking lacks the path...
benjamin@webkit.org [Sat, 8 Mar 2014 00:40:11 +0000 (00:40 +0000)]
Traversal failure in a direct adjacent chain with tail backtracking lacks the path to clear the tail
https://bugs.webkit.org/show_bug.cgi?id=129863

Reviewed by Gavin Barraclough.

Source/WebCore:

Direct adjacent backtracking use the stack to push the backtracking entry point and recover from there.
In case of traversal failure, their is no point in recovering from the indirect adjancent entry point and
we should clear entry point from the stack (which is the purpose of the tail).

The adjancent tail was missing the part for clearing the stack in one case.

The case with adjancent backtracking inside descendant backtracing was doing everything right. This patch
generalize this code and the correct tail is fully generated by generateAdjacentBacktrackingTail().

JumpToClearAdjacentDescendantTail becomes JumpToClearAdjacentTail, and this new backtracking state is added
to the missing traversal action.

Test: fast/selectors/long-adjacent-backtracking.html

* cssjit/SelectorCompiler.cpp:
(WebCore::SelectorCompiler::solveBacktrackingAction):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateSelectorChecker):
(WebCore::SelectorCompiler::SelectorCodeGenerator::linkFailures):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateAdjacentBacktrackingTail):
(WebCore::SelectorCompiler::SelectorCodeGenerator::generateBacktrackingTailsIfNeeded):

LayoutTests:

Test the faulty case.

* fast/selectors/long-adjacent-backtracking-expected.txt: Added.
* fast/selectors/long-adjacent-backtracking.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165300 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Include CMake into install-dependencies script
commit-queue@webkit.org [Sat, 8 Mar 2014 00:33:55 +0000 (00:33 +0000)]
[GTK] Include CMake into install-dependencies script
https://bugs.webkit.org/show_bug.cgi?id=129909

Patch by Lorenzo Tilve <ltilve@igalia.com> on 2014-03-07
Reviewed by Martin Robinson.

As since r165267 buid-webkit uses CMake by default, the dependency
with 'cmake' package should be introduced.

* gtk/install-dependencies: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165299 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSource/WebKit: Make it possible to copy WebKit2 headers to WebKit.framework
andersca@apple.com [Sat, 8 Mar 2014 00:18:21 +0000 (00:18 +0000)]
Source/WebKit: Make it possible to copy WebKit2 headers to WebKit.framework
https://bugs.webkit.org/show_bug.cgi?id=129935

Reviewed by Dan Bernstein.

* WebKit.xcodeproj/project.pbxproj:
Move the migrate headers script out into its own file.

Source/WebKit/mac: Make it possible to copy WebKit2 headers to WebKit.framework
https://bugs.webkit.org/show_bug.cgi?id=129935
<rdar://problem/15920020>

Reviewed by Dan Bernstein.

* MigrateHeaders.make:
Gather a list of public and private WebKit2 headers and filter them against the WEBKIT2_HEADERS list.
The matching headers are copied to the respective Headers or PrivateHeaders directories under WebKit.framework,
and also have their #imports rewritten to reference WebKit instead of WebKit2.

* migrate-headers.sh: Added.
Set up the WEBKIT2_FRAMEWORKS_DIR if needed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165298 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoREGRESSION (r164847): Pressing the command key while hovering an element causes it...
ap@apple.com [Fri, 7 Mar 2014 23:43:57 +0000 (23:43 +0000)]
REGRESSION (r164847): Pressing the command key while hovering an element causes it to un-hover
https://bugs.webkit.org/show_bug.cgi?id=129872

Reviewed by Tim Horton.

* UIProcess/API/Cocoa/WKViewPrivate.h: Changes in performDictionaryLookupAtCurrentMouseLocation
were also not correct, because the current event could be a keyboard one (Cmd+Ctrl+D).
But this SPI is unused, and can be simply removed.

* UIProcess/API/mac/WKView.mm: (-[WKView _postFakeMouseMovedEventForFlagsChangedEvent:]):
locationInWindow is undefined for keyboard events, so we have to use current location.
Thankfully, there is a method that returns the location in appropriate coordinate
space, unlike +[NSEvent mouseLocation] that we used before.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165297 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Win64] Compile error after r165128.
commit-queue@webkit.org [Fri, 7 Mar 2014 23:21:18 +0000 (23:21 +0000)]
[Win64] Compile error after r165128.
https://bugs.webkit.org/show_bug.cgi?id=129807

Patch by peavo@outlook.com <peavo@outlook.com> on 2014-03-07
Reviewed by Mark Lam.

Source/JavaScriptCore:

* JavaScriptCore.vcxproj/LLInt/LLIntAssembly/build-LLIntAssembly.sh:
Check platform environment variable to determine if an assembler file should be generated.

Source/WTF:

* wtf/Platform.h: MSVC does not support computed goto.
Also enabled COMPUTED_GOTO_OPCODES when !ENABLE(LLINT_C_LOOP). This is needed because the ASM LLINT operates like COMPUTED_GOTO_OPCODES,
and relies on the related data structures being defined to support this. On Win32, the platform does not HAVE_COMPUTED_GOTO support,
but does want ENABLE_COMPUTED_GOTO_OPCODES because it uses the ASM LLINT.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165296 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[iOS] Install WebKit2 XPC services to the correct place during iOS Simulator builds
aestes@apple.com [Fri, 7 Mar 2014 22:53:27 +0000 (22:53 +0000)]
[iOS] Install WebKit2 XPC services to the correct place during iOS Simulator builds
https://bugs.webkit.org/show_bug.cgi?id=129914

Reviewed by Dan Bernstein.

We weren't specifying INSTALL_PATH for iOS Simulator builds in
BaseXPCService.xcconfig, and r164576 changed INSTALL_PATH to
INSTALL_PATH_ACTUAL. This caused XPC services to start being installed
to the root of the iOS Simulator SDK. Fix this by setting
INSTALL_PATH_ACTUAL for iphonesimulator.

* Configurations/BaseXPCService.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165295 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix a typo
andersca@apple.com [Fri, 7 Mar 2014 22:52:38 +0000 (22:52 +0000)]
Fix a typo
<rdar://problem/15920020>

Rubber-stamped by Dan Bernstein.

* Configurations/WebKit.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165294 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoClarify how we deal with "special" registers
msaboff@apple.com [Fri, 7 Mar 2014 22:22:46 +0000 (22:22 +0000)]
Clarify how we deal with "special" registers
https://bugs.webkit.org/show_bug.cgi?id=129806

Already reviewed change being relanded.

Relanding change set r165196 as it wasn't responsible for the breakage reported in
https://bugs.webkit.org/show_bug.cgi?id=129822.  That appears to be a build or

Reviewed by Michael Saboff.
configuration issue.

* assembler/ARM64Assembler.h:
(JSC::ARM64Assembler::lastRegister):
* assembler/MacroAssembler.h:
(JSC::MacroAssembler::nextRegister):
* ftl/FTLLocation.cpp:
(JSC::FTL::Location::restoreInto):
* ftl/FTLSaveRestore.cpp:
(JSC::FTL::saveAllRegisters):
(JSC::FTL::restoreAllRegisters):
* ftl/FTLSlowPathCall.cpp:
* jit/RegisterSet.cpp:
(JSC::RegisterSet::reservedHardwareRegisters):
(JSC::RegisterSet::runtimeRegisters):
(JSC::RegisterSet::specialRegisters):
(JSC::RegisterSet::calleeSaveRegisters):
* jit/RegisterSet.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165293 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Mac] Notify system malloc of fake memory pressure.
akling@apple.com [Fri, 7 Mar 2014 22:16:00 +0000 (22:16 +0000)]
[Mac] Notify system malloc of fake memory pressure.
<https://webkit.org/b/129908>

Source/WebCore:

After fixing the leak in r165252, I was surprised to find that it didn't
show up on memory test bots. It turns out that while the memory is now
getting freed, the system malloc implementation doesn't actually release
pages back to the OS until there is memory pressure.

Since we are just faking the memory pressure on bots, we have to let
system malloc in on the fun, so we can get representative numbers.
With this change, we should finally see the effect of not leaking the
URLRequest object tree.

Reviewed by Anders Carlsson.

* platform/mac/MemoryPressureHandlerMac.mm:
(WebCore::MemoryPressureHandler::install):

    Tell system malloc that we are under fake memory pressure.

Source/WebKit2:

Reviewed by Anders Carlsson.

* NetworkProcess/mac/NetworkProcessMac.mm:
(WebKit::NetworkProcess::platformInitializeNetworkProcess):

    Listen for the fake memory pressure notification and respond by
    just forwarding to system malloc. We don't do the full WebCore
    memory pressure handling here, since that will cause stuff that
    the network process doesn't need to get instantiated.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165292 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoTweak the WebKit2 frameworks directories
andersca@apple.com [Fri, 7 Mar 2014 22:06:57 +0000 (22:06 +0000)]
Tweak the WebKit2 frameworks directories
https://bugs.webkit.org/show_bug.cgi?id=129910
<rdar://problem/15920020>

Reviewed by Dan Bernstein.

Use an SDK relative directory for production builds.

* Configurations/WebKit.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165291 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[iOS] WebKit1 scroll position is incorrect
benjamin@webkit.org [Fri, 7 Mar 2014 21:56:18 +0000 (21:56 +0000)]
[iOS] WebKit1 scroll position is incorrect
https://bugs.webkit.org/show_bug.cgi?id=129905

Patch by Benjamin Poulain <bpoulain@apple.com> on 2014-03-07
Reviewed by Simon Fraser.

* page/DOMWindow.cpp:
(WebCore::DOMWindow::scrollX):
(WebCore::DOMWindow::scrollY):
A new optimization was avoiding layout if the position is zero. This was using the scrollX/Y instead
of actualScrollX/Y which caused us to always bail out instead of returning the value to JavaScript.

* platform/ScrollView.h:
(WebCore::ScrollView::actualScrollX):
(WebCore::ScrollView::actualScrollY):
(WebCore::ScrollView::actualScrollPosition):
This was accidentally changed to the iOS compat document visible rect at some point.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165288 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoAttempt to fix the iOS build.
dino@apple.com [Fri, 7 Mar 2014 21:53:19 +0000 (21:53 +0000)]
Attempt to fix the iOS build.

* Configurations/WebKit2.xcconfig: Add -framework OpenGLES to linking.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165287 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove unused StdDeviation from Drop shadow effect
cavalcantii@gmail.com [Fri, 7 Mar 2014 21:33:44 +0000 (21:33 +0000)]
Remove unused StdDeviation from Drop shadow effect
https://bugs.webkit.org/show_bug.cgi?id=129891

Reviewed by Dirk Schulze.

No new tests, no change on behavior.

* platform/graphics/filters/FEDropShadow.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165285 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoREGRESSION(165267): Broke a webkitpy unit tests
mrobinson@webkit.org [Fri, 7 Mar 2014 21:32:51 +0000 (21:32 +0000)]
REGRESSION(165267): Broke a webkitpy unit tests
https://bugs.webkit.org/show_bug.cgi?id=129904

Reviewed by Jon Honeycutt.

* Scripts/webkitpy/port/gtk.py: Flip the default test to assume CMake by default.
(GtkPort._is_cmake_build):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165284 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoLink WebKit.framework against WebKit2.framework and re-export all WebKit2 symbols
andersca@apple.com [Fri, 7 Mar 2014 21:30:52 +0000 (21:30 +0000)]
Link WebKit.framework against WebKit2.framework and re-export all WebKit2 symbols
https://bugs.webkit.org/show_bug.cgi?id=129901
<rdar://problem/15920020>

Reviewed by Andreas Kling.

* Configurations/WebKit.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165283 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMove GCActivityCallback to heap
mhahnenberg@apple.com [Fri, 7 Mar 2014 21:28:43 +0000 (21:28 +0000)]
Move GCActivityCallback to heap
https://bugs.webkit.org/show_bug.cgi?id=129457

Reviewed by Geoffrey Garen.

All the other GC timer related stuff is there already.

* CMakeLists.txt:
* GNUmakefile.list.am:
* JavaScriptCore.vcxproj/JavaScriptCore.vcxproj:
* JavaScriptCore.vcxproj/JavaScriptCore.vcxproj.filters:
* JavaScriptCore.xcodeproj/project.pbxproj:
* heap/GCActivityCallback.cpp: Copied from Source/JavaScriptCore/runtime/GCActivityCallback.cpp.
* heap/GCActivityCallback.h: Copied from Source/JavaScriptCore/runtime/GCActivityCallback.h.
* runtime/GCActivityCallback.cpp: Removed.
* runtime/GCActivityCallback.h: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165282 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoCorrect out-of-band track selection logic.
bfulgham@apple.com [Fri, 7 Mar 2014 21:10:45 +0000 (21:10 +0000)]
Correct out-of-band track selection logic.
https://bugs.webkit.org/show_bug.cgi?id=129900

Reviewed by Jer Noble.

* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::synchronizeTextTrackState): Correct test to
use inequality.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165280 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[iOS] Add an updateID to visibleContentRect updates which is passed back in layer...
simon.fraser@apple.com [Fri, 7 Mar 2014 20:52:09 +0000 (20:52 +0000)]
[iOS] Add an updateID to visibleContentRect updates which is passed back in layer transactions, so we know whether transactions are stale
https://bugs.webkit.org/show_bug.cgi?id=129897

Reviewed by Benjamin Poulain.

In WebKit2 on iOS we need to know when layer updates from the web process
are stale with respect to visible rect updates from the UI process. Do so
by adding an updateID to VisibleContentRectUpdateInfo, and storing it
on each side, returning it in RemoteLayerTreeTransaction.

Did some re-ordering of members and encoding order in RemoteLayerTreeTransaction
to group like data members together.

* Shared/VisibleContentRectUpdateInfo.cpp:
(WebKit::VisibleContentRectUpdateInfo::encode):
(WebKit::VisibleContentRectUpdateInfo::decode):
* Shared/VisibleContentRectUpdateInfo.h:
(WebKit::VisibleContentRectUpdateInfo::VisibleContentRectUpdateInfo):
(WebKit::VisibleContentRectUpdateInfo::updateID):
(WebKit::operator==):
* Shared/mac/RemoteLayerTreeTransaction.h:
(WebKit::RemoteLayerTreeTransaction::setLastVisibleContentRectUpdateID):
(WebKit::RemoteLayerTreeTransaction::lastVisibleContentRectUpdateID):
* Shared/mac/RemoteLayerTreeTransaction.mm:
(WebKit::RemoteLayerTreeTransaction::encode):
(WebKit::RemoteLayerTreeTransaction::decode):
* UIProcess/WebPageProxy.h:
(WebKit::WebPageProxy::nextVisibleContentRectUpdateID):
(WebKit::WebPageProxy::lastVisibleContentRectUpdateID):
* UIProcess/ios/WKContentView.mm:
(-[WKContentView didUpdateVisibleRect:unobscuredRect:scale:inStableState:]):
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::WebPage):
(WebKit::WebPage::willCommitLayerTree):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::updateVisibleContentRects):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165279 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] webkit-patch setup-git-clone fails even after running Tools/gtk/install-depende...
agomez@igalia.com [Fri, 7 Mar 2014 20:39:13 +0000 (20:39 +0000)]
[GTK] webkit-patch setup-git-clone fails even after running Tools/gtk/install-dependencies
https://bugs.webkit.org/show_bug.cgi?id=129893

Reviewed by Martin Robinson.

Added "git-svn" as dependency for deb based distros and,
additionally, "subversion" for rpm based distros.

* gtk/install-dependencies:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165278 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[CSS Shapes] Correctly serialize ellipse positions
betravis@adobe.com [Fri, 7 Mar 2014 20:29:39 +0000 (20:29 +0000)]
[CSS Shapes] Correctly serialize ellipse positions
https://bugs.webkit.org/show_bug.cgi?id=129700

Reviewed by Dirk Schulze.

Source/WebCore:

Updating ellipse serialization to be in line with
the CSS Shapes spec. Positions should serialize as
2 or 4-value positions, converting keywords to
percentages where possible. This uses the same method
as circle position serialization (bug 129404).

Updated existing parsing tests.

* css/CSSBasicShapes.cpp:
(WebCore::CSSBasicShapeEllipse::cssText): Use the
normalization / serialization helpers for shape position.

LayoutTests:

Update tests to use the new position serialization for
ellipse.

* fast/shapes/parsing/parsing-shape-inside-expected.txt:
* fast/shapes/parsing/parsing-shape-outside-expected.txt:
* fast/shapes/parsing/parsing-test-utils.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165277 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoReplace setSystemWebGLLoadPolicy API with methods to query for WebGL blocking policy.
roger_fong@apple.com [Fri, 7 Mar 2014 20:20:17 +0000 (20:20 +0000)]
Replace setSystemWebGLLoadPolicy API with methods to query for WebGL blocking policy.
https://bugs.webkit.org/show_bug.cgi?id=129873.

Reviewed by Dean Jackson.

Remove setSystemWebGLLoadPolicy API.
* UIProcess/API/APILoaderClient.h:
(API::LoaderClient::resolveWebGLLoadPolicy):
* UIProcess/API/C/WKPage.cpp:
(WKPageSetPageLoaderClient):
* UIProcess/API/C/WKPageLoaderClient.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::resolveWebGLPolicyForURL):
* UIProcess/WebPageProxy.h:
* UIProcess/WebPageProxy.messages.in:
* WebProcess/WebPage/mac/WebPageMac.mm:
(WebKit::WebPage::webGLPolicyForURL):
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::createWebViewWithOptions):

Add API to directly query the WebGL blocking policies from WKSI.
* UIProcess/API/C/mac/WKContextPrivateMac.h:
* UIProcess/API/C/mac/WKContextPrivateMac.mm:
(WKContextShouldBlockWebGL):
(WKContextShouldSuggestBlockWebGL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165276 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoCorrect a comment typo from:
commit-queue@webkit.org [Fri, 7 Mar 2014 20:19:28 +0000 (20:19 +0000)]
Correct a comment typo from:
FLT should call fmod directly on platforms where LLVM cannot relocate the libcall
https://bugs.webkit.org/show_bug.cgi?id=129865

Patch by Andrew Trick <atrick@apple.com> on 2014-03-07
Reviewed by Mark Lam.

* ftl/FTLOutput.h:
(JSC::FTL::Output::doubleRem):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165275 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agorun-jsc-stress-tests doesn't eagerly report test failures when using the shell runner
mhahnenberg@apple.com [Fri, 7 Mar 2014 19:40:21 +0000 (19:40 +0000)]
run-jsc-stress-tests doesn't eagerly report test failures when using the shell runner
https://bugs.webkit.org/show_bug.cgi?id=129886

Reviewed by Mark Lam.

* Scripts/jsc-stress-test-helpers/shell-runner.sh: Added a "verbose" mode to the shell runner.
If we're in verbose mode, then we don't redirect any output when running each test. If we're not
in verbose mode then we redirect stdout to /dev/null and allow any stderr output through.
* Scripts/run-jsc-stress-tests: Redirect any sort of error output to stderr so that it will
make it through when in non-verbose mode.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165274 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoInitialize new WKSI functions for WebKit2
eric.carlson@apple.com [Fri, 7 Mar 2014 19:02:41 +0000 (19:02 +0000)]
Initialize new WKSI functions for WebKit2
https://bugs.webkit.org/show_bug.cgi?id=129889

Reviewed by Simon Fraser.

* WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:
(InitWebCoreSystemInterface):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165273 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rebaseline run-binding-tests results after r165242.
bburg@apple.com [Fri, 7 Mar 2014 18:48:55 +0000 (18:48 +0000)]
Unreviewed, rebaseline run-binding-tests results after r165242.

* bindings/scripts/test/JS/JSTestObj.cpp:
(WebCore::jsTestObjPrototypeFunctionMethodWithSequenceArg):
* bindings/scripts/test/JS/JSTestTypedefs.cpp:
(WebCore::jsTestTypedefsPrototypeFunctionCallWithSequenceThatRequiresInclude):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165272 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Win] Convert WebKit.make to an MSBuild file.
bfulgham@apple.com [Fri, 7 Mar 2014 18:43:04 +0000 (18:43 +0000)]
[Win] Convert WebKit.make to an MSBuild file.
https://bugs.webkit.org/show_bug.cgi?id=129888

Reviewed by Dean Jackson.

* WebKit.vcxproj/WebKit.proj: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165270 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove non-working optimization that was attempted on iOS only
darin@apple.com [Fri, 7 Mar 2014 17:51:08 +0000 (17:51 +0000)]
Remove non-working optimization that was attempted on iOS only
https://bugs.webkit.org/show_bug.cgi?id=129595
<rdar://problem/15798825>

Reviewed by Sam Weinig.

Code in GCController tried to optimize cases where the controller
was used, but no JavaScript had been run in the current process.
The code was never effective, and was iOS-only. Another way to fix
the problem would be to change the code so it works, and if we do
that we should do it for all platforms, not just iOS.

* bindings/js/GCController.cpp:
(WebCore::GCController::garbageCollectNow): Remove check of
JSDOMWindow::commonVMExists, since it's called just after a call
to JSDOMWindow::commonVM, which will create it as a side effect.
(WebCore::GCController::releaseExecutableMemory): Ditto.

* bindings/js/JSDOMWindowBase.cpp:
(WebCore::JSDOMWindowBase::commonVM): Removed iOS-specific code
that pulled the commonVM global out into another function so we
can check for its existence without creating it as a side effect.

* bindings/js/JSDOMWindowBase.h: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165269 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed EFL gardening
m.pakula@samsung.com [Fri, 7 Mar 2014 17:33:26 +0000 (17:33 +0000)]
Unreviewed EFL gardening

Clear test expectations for passing tests.

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165268 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] buid-webkit should use CMake by default
mrobinson@webkit.org [Fri, 7 Mar 2014 17:27:42 +0000 (17:27 +0000)]
[GTK] buid-webkit should use CMake by default
https://bugs.webkit.org/show_bug.cgi?id=129855

Reviewed by Carlos Garcia Campos.

Make the CMake build the default GTK+ build when using build-webkit. This will
ensure that all bots are using CMake. We do this by swapping --gtk and --gtkcmake,
so that we have --gtk and --gtkautotools.

* Scripts/webkitdirs.pm: Swap --gtkcmake/--gtk for --gtk/--gtkautotools.
* Scripts/webkitpy/port/gtk.py: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165267 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed, rolling out r165262.
commit-queue@webkit.org [Fri, 7 Mar 2014 17:21:46 +0000 (17:21 +0000)]
Unreviewed, rolling out r165262.
http://trac.webkit.org/changeset/165262
https://bugs.webkit.org/show_bug.cgi?id=129887

Fix wrong commit message (Requested by mpakula on #webkit).

* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165266 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUse OwnPtr in StructureIDTable
mhahnenberg@apple.com [Fri, 7 Mar 2014 17:13:38 +0000 (17:13 +0000)]
Use OwnPtr in StructureIDTable
https://bugs.webkit.org/show_bug.cgi?id=129828

Reviewed by Geoffrey Garen.

This reduces the amount of boilerplate and fixes a memory leak.

* runtime/StructureIDTable.cpp:
(JSC::StructureIDTable::StructureIDTable):
(JSC::StructureIDTable::resize):
(JSC::StructureIDTable::flushOldTables):
(JSC::StructureIDTable::allocateID):
(JSC::StructureIDTable::deallocateID):
* runtime/StructureIDTable.h:
(JSC::StructureIDTable::table):
(JSC::StructureIDTable::get):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165265 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFLT should call fmod directly on platforms where LLVM cannot relocate the libcall
commit-queue@webkit.org [Fri, 7 Mar 2014 17:03:51 +0000 (17:03 +0000)]
FLT should call fmod directly on platforms where LLVM cannot relocate the libcall
https://bugs.webkit.org/show_bug.cgi?id=129865

Patch by Andrew Trick <atrick@apple.com> on 2014-03-07
Reviewed by Filip Pizlo.

* ftl/FTLIntrinsicRepository.h:
* ftl/FTLOutput.h:
(JSC::FTL::Output::doubleRem):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165264 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[CMake] [GTK] Build with ninja when it is available
mrobinson@webkit.org [Fri, 7 Mar 2014 17:00:03 +0000 (17:00 +0000)]
[CMake] [GTK] Build with ninja when it is available
https://bugs.webkit.org/show_bug.cgi?id=129805

Reviewed by Philippe Normand.

* Scripts/webkitdirs.pm:
(canUseNinja): Added this helper which tries to execute Ninja, in an effort to determine
if it's available on the system.
(cmakeGeneratedBuildfile): Return the path to the generated build file. Only Unix support ATM.
(generateBuildSystemFromCMakeProject): For GTK+ compile with Ninja when possible.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165263 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[CSS Shapes] inset does not properly clamp large corner radii
m.pakula@samsung.com [Fri, 7 Mar 2014 16:49:07 +0000 (16:49 +0000)]
[CSS Shapes] inset does not properly clamp large corner radii
https://bugs.webkit.org/show_bug.cgi?id=129726

Patch by Bem Jones-Bey <bjonesbe@adobe.com> on 2014-03-07
Reviewed by Dirk Schulze.

* fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-large-radius-expected.html: Added.
* fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-large-radius.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165262 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[CSS Shapes] inset does not properly clamp large corner radii
bjonesbe@adobe.com [Fri, 7 Mar 2014 16:25:12 +0000 (16:25 +0000)]
[CSS Shapes] inset does not properly clamp large corner radii
https://bugs.webkit.org/show_bug.cgi?id=129726

Reviewed by Dirk Schulze.

Source/WebCore:

Make sure that radii that are larger than 50% of the associated shape
dimension (height or width) are properly reduced the same way as
border-radius radii are.

Test: fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-large-radius.html

* rendering/shapes/Shape.cpp:
(WebCore::ensureRadiiDoNotOverlap): Drive by style fix.
(WebCore::Shape::createShape): Scale radii the same way as border
    radii are scaled.
* platform/graphics/FloatRoundedRect.h:
(WebCore::calcBorderRadiiConstraintScaleFor): Move border radii
    constraint calculation function out so that it can be shared by
    the shape code.
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::getRoundedBorderFor): Call renamed scale function.

LayoutTests:

* fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-large-radius-expected.html: Added.
* fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-large-radius.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165261 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed EFL gardening
m.pakula@samsung.com [Fri, 7 Mar 2014 13:39:00 +0000 (13:39 +0000)]
Unreviewed EFL gardening

Remove wrong test expectations.

* platform/efl-wk2/TestExpectations:
* platform/efl/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165260 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Use final and override in WebPrintOperationGtk
carlosgc@webkit.org [Fri, 7 Mar 2014 13:37:59 +0000 (13:37 +0000)]
[GTK] Use final and override in WebPrintOperationGtk
https://bugs.webkit.org/show_bug.cgi?id=129868

Reviewed by Sergio Villar Senin.

* WebProcess/WebPage/gtk/WebPrintOperationGtk.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165259 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GST] Set name to several timeout sources
carlosgc@webkit.org [Fri, 7 Mar 2014 13:35:57 +0000 (13:35 +0000)]
[GST] Set name to several timeout sources
https://bugs.webkit.org/show_bug.cgi?id=129878

Reviewed by Philippe Normand.

* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
(WebCore::MediaPlayerPrivateGStreamer::videoChanged):
(WebCore::MediaPlayerPrivateGStreamer::videoCapsChanged):
(WebCore::MediaPlayerPrivateGStreamer::audioChanged):
(WebCore::MediaPlayerPrivateGStreamer::textChanged):
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:
(WebCore::MediaPlayerPrivateGStreamerBase::volumeChanged):
(WebCore::MediaPlayerPrivateGStreamerBase::muteChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165258 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[GTK] Timeout sources not correctly removed
carlosgc@webkit.org [Fri, 7 Mar 2014 13:27:55 +0000 (13:27 +0000)]
[GTK] Timeout sources not correctly removed
https://bugs.webkit.org/show_bug.cgi?id=129877

Reviewed by Philippe Normand.

Set source ID variables to 0 after removing the sources from the
context. Also give a name to the sources.

* platform/graphics/gstreamer/TrackPrivateBaseGStreamer.cpp:
(WebCore::TrackPrivateBaseGStreamer::disconnect):
(WebCore::TrackPrivateBaseGStreamer::activeChanged):
(WebCore::TrackPrivateBaseGStreamer::tagsChanged):
(WebCore::TrackPrivateBaseGStreamer::notifyTrackOfActiveChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165257 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove FFTFrameMKL.
commit-queue@webkit.org [Fri, 7 Mar 2014 12:02:59 +0000 (12:02 +0000)]
Remove FFTFrameMKL.
https://bugs.webkit.org/show_bug.cgi?id=129866

Patch by Gergo Balogh <gbalogh.u-szeged@partner.samsung.com> on 2014-03-07
Reviewed by Csaba Osztrogonác.

* platform/audio/FFTFrame.h:
* platform/audio/FFTFrameStub.cpp:
* platform/audio/mkl/FFTFrameMKL.cpp: Removed, it is never used.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165256 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WebRTC] Adding getConfiguration method to RTCPeerConnection
thiago.lacerda@openbossa.org [Fri, 7 Mar 2014 11:46:53 +0000 (11:46 +0000)]
[WebRTC] Adding getConfiguration method to RTCPeerConnection
https://bugs.webkit.org/show_bug.cgi?id=129845

Reviewed by Eric Carlson.

Source/WebCore:

Test: fast/mediastream/RTCPeerConnection-getConfiguration.html

* CMakeLists.txt:
* DerivedSources.make:
* GNUmakefile.list.am:
* Modules/mediastream/RTCConfiguration.idl: Added.
* Modules/mediastream/RTCIceServer.idl: Added.
* Modules/mediastream/RTCPeerConnection.cpp:
(WebCore::RTCPeerConnection::RTCPeerConnection):
(WebCore::RTCPeerConnection::updateIce):
(WebCore::RTCPeerConnection::getConfiguration):
* Modules/mediastream/RTCPeerConnection.h:
* Modules/mediastream/RTCPeerConnection.idl:

LayoutTests:

* fast/mediastream/RTCPeerConnection-getConfiguration-expected.txt: Added.
* fast/mediastream/RTCPeerConnection-getConfiguration.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165255 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoOptimize collectionLast() and collectionTraverseBackward() in LiveNodeList
jinwoo7.song@samsung.com [Fri, 7 Mar 2014 10:42:56 +0000 (10:42 +0000)]
Optimize collectionLast() and collectionTraverseBackward() in LiveNodeList
https://bugs.webkit.org/show_bug.cgi?id=129775

Reviewed by Ryosuke Niwa.

Check the LiveNodeList type if it is HTMLTagNodeListType and ClassNodeListType then
apply inlined matching methods for each type.

* dom/LiveNodeList.cpp:
(WebCore::lastMatchingElement): Added.
(WebCore::previousMatchingElement): Added.
(WebCore::traverseMatchingElementsBackward): Added.
(WebCore::LiveNodeList::collectionLast): Apply lastMatchingElement() with LiveNodeList type.
(WebCore::LiveNodeList::collectionTraverseBackward): Apply traverseMatchingElementsBackward()
with LiveNodeList type.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165254 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMove GTK WebKit1 code to std::unique_ptr
zandobersek@gmail.com [Fri, 7 Mar 2014 10:00:48 +0000 (10:00 +0000)]
Move GTK WebKit1 code to std::unique_ptr
https://bugs.webkit.org/show_bug.cgi?id=129675

Reviewed by Anders Carlsson.

Replace uses of OwnPtr and PassOwnPtr in GTK-specific WebKit1 code with std::unique_ptr.

* WebCoreSupport/AcceleratedCompositingContext.h:
* WebCoreSupport/NavigatorContentUtilsClientGtk.cpp:
* WebCoreSupport/NavigatorContentUtilsClientGtk.h:
* webkit/webkitfavicondatabase.cpp:
(getIconPixbufCancelled):
(webkit_favicon_database_get_favicon_pixbuf):
* webkit/webkitwebview.cpp:
(webkit_web_view_init):
* webkit/webkitwebviewprivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165253 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoWebCoreResourceHandleAsOperationQueueDelegate over-retains NSURLRequest.
akling@apple.com [Fri, 7 Mar 2014 07:20:51 +0000 (07:20 +0000)]
WebCoreResourceHandleAsOperationQueueDelegate over-retains NSURLRequest.
<https://webkit.org/b/129862>

Don't retain the pointer before assigning it to a RetainPtr, since that
will cause the object to leak.

Reviewed by Alexey Proskuryakov.

* platform/network/mac/WebCoreResourceHandleAsOperationQueueDelegate.mm:
(-[WebCoreResourceHandleAsOperationQueueDelegate continueWillSendRequest:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165252 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoDon’t copy WebInputs.json into the framework.
mitz@apple.com [Fri, 7 Mar 2014 06:39:51 +0000 (06:39 +0000)]
Don’t copy WebInputs.json into the framework.

* WebCore.xcodeproj/project.pbxproj: Removed WebInputs.json from the Copy Bundle Resources
build phase of the WebCore target.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165251 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[Mac] Allow com.apple.GSSCred in WebKit sandbox profiles
ap@apple.com [Fri, 7 Mar 2014 06:32:24 +0000 (06:32 +0000)]
[Mac] Allow com.apple.GSSCred in WebKit sandbox profiles
https://bugs.webkit.org/show_bug.cgi?id=129777
<rdar://problem/16050832>

Reviewed by Anders Carlsson.

* NetworkProcess/mac/com.apple.WebKit.NetworkProcess.sb.in:
* Resources/PlugInSandboxProfiles/com.apple.ist.ds.appleconnect.webplugin.sb:
* WebProcess/com.apple.WebProcess.sb.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165250 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoIf the FTL is build-time enabled then it should be run-time enabled.
fpizlo@apple.com [Fri, 7 Mar 2014 06:16:31 +0000 (06:16 +0000)]
If the FTL is build-time enabled then it should be run-time enabled.

Rubber stamped by Geoffrey Garen.

Source/JavaScriptCore:

* runtime/Options.cpp:
(JSC::recomputeDependentOptions):
* runtime/Options.h:

Source/WebKit2:

* Shared/WebProcessCreationParameters.cpp:
(WebKit::WebProcessCreationParameters::WebProcessCreationParameters):
(WebKit::WebProcessCreationParameters::encode):
(WebKit::WebProcessCreationParameters::decode):
* Shared/WebProcessCreationParameters.h:
* UIProcess/mac/WebContextMac.mm:
(WebKit::registerUserDefaultsIfNeeded):
(WebKit::WebContext::platformInitializeWebProcess):
* WebProcess/cocoa/WebProcessCocoa.mm:
(WebKit::WebProcess::platformInitializeWebProcess):

Tools:

* Scripts/run-jsc-stress-tests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165249 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WebRTC] Updating RTCPeerConnection.idl
thiago.lacerda@openbossa.org [Fri, 7 Mar 2014 05:42:57 +0000 (05:42 +0000)]
[WebRTC] Updating RTCPeerConnection.idl
https://bugs.webkit.org/show_bug.cgi?id=129804

Reviewed by Eric Carlson.

Some methods signatures were wrong, marking some arguments as optional when they are mandatory.

Source/WebCore:

Existing tests were updated.

* Modules/mediastream/RTCPeerConnection.cpp:
(WebCore::RTCPeerConnection::getStats):
* Modules/mediastream/RTCPeerConnection.h:
* Modules/mediastream/RTCPeerConnection.idl:
* Modules/mediastream/RTCStatsRequestImpl.cpp:
(WebCore::RTCStatsRequestImpl::create):
(WebCore::RTCStatsRequestImpl::RTCStatsRequestImpl):
(WebCore::RTCStatsRequestImpl::requestFailed):
* Modules/mediastream/RTCStatsRequestImpl.h:
* platform/mediastream/RTCStatsRequest.h:

LayoutTests:

* fast/mediastream/RTCPeerConnection-stats.html:
* fast/mediastream/RTCPeerConnection-statsSelector.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165248 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[WebRTC] Updating RTCIceServer to match spec
thiago.lacerda@openbossa.org [Fri, 7 Mar 2014 05:41:16 +0000 (05:41 +0000)]
[WebRTC] Updating RTCIceServer to match spec
https://bugs.webkit.org/show_bug.cgi?id=129844

Reviewed by Eric Carlson.

Move RTCIceServer from RTCConfiguration to its own file.

* Modules/mediastream/RTCPeerConnection.cpp:
(WebCore::validateIceServerURL):
(WebCore::processIceServer):
* platform/mediastream/RTCConfiguration.h:
(WebCore::RTCConfiguration::iceServers):
* platform/mediastream/RTCIceServer.h: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165247 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[EFL] Move EvasGL classes from WebKit to WebCore namespace.
commit-queue@webkit.org [Fri, 7 Mar 2014 05:06:04 +0000 (05:06 +0000)]
[EFL] Move EvasGL classes from WebKit to WebCore namespace.
https://bugs.webkit.org/show_bug.cgi?id=129797

Patch by Hyowon Kim <hw1008.kim@samsung.com> on 2014-03-06
Reviewed by Gyuyoung Kim.

Source/WebCore:

Though EvasGLContext and EvasGLSurface files were moved from WebKit2/UIProcess/API/efl/
to WebCore/platform/graphics/efl/, they are still in WebKit namespace.
Patch for namespace changes.

* platform/graphics/efl/EvasGLContext.cpp:
* platform/graphics/efl/EvasGLContext.h:
* platform/graphics/efl/EvasGLSurface.cpp:
* platform/graphics/efl/EvasGLSurface.h:

Source/WebKit2:

* UIProcess/API/efl/EwkView.h: namespace changes for EvasGLContext/Surface.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165246 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[OS X] Web Inspector: Allow Apps using JavaScriptCore to access "com.apple.webinspect...
commit-queue@webkit.org [Fri, 7 Mar 2014 04:21:53 +0000 (04:21 +0000)]
[OS X] Web Inspector: Allow Apps using JavaScriptCore to access "com.apple.webinspector" mach port
https://bugs.webkit.org/show_bug.cgi?id=129852

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2014-03-06
Reviewed by Geoffrey Garen.

* framework.sb: Added.
Sandbox extension to allow access to "com.apple.webinspector".

* JavaScriptCore.xcodeproj/project.pbxproj:
Add a Copy Resources build phase and include framework.sb.

* Configurations/JavaScriptCore.xcconfig:
Do not copy framework.sb on iOS.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165245 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoJSGlobalContextRelease incorrectly handles saving/restoring IdentifierTable
mhahnenberg@apple.com [Fri, 7 Mar 2014 03:29:55 +0000 (03:29 +0000)]
JSGlobalContextRelease incorrectly handles saving/restoring IdentifierTable
https://bugs.webkit.org/show_bug.cgi?id=129858

Reviewed by Mark Lam.

It was correct (but really ugly) prior to the combining of APIEntryShim and JSLock,
but now it ends up overwriting the IdentifierTable that JSLock just restored.

* API/JSContextRef.cpp:
(JSGlobalContextRelease):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165244 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoTest that the screen orientation does not change when focusing a <textarea>
aestes@apple.com [Fri, 7 Mar 2014 02:42:02 +0000 (02:42 +0000)]
Test that the screen orientation does not change when focusing a <textarea>
https://bugs.webkit.org/show_bug.cgi?id=129847

Reviewed by Enrica Casucci.

* fast/css/orientation-changes-on-textarea-focus-expected.html: Added.
* fast/css/orientation-changes-on-textarea-focus.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165243 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoCodeGeneratorJS.pm doesn't need to add spaces between consecutive closing template...
bburg@apple.com [Fri, 7 Mar 2014 02:23:54 +0000 (02:23 +0000)]
CodeGeneratorJS.pm doesn't need to add spaces between consecutive closing template brackets
https://bugs.webkit.org/show_bug.cgi?id=129836

Reviewed by Andreas Kling.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateParametersCheck):
(GetNativeVectorInnerType):
(GetSVGPropertyTypes):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165242 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove unused method from BatteryController
jinwoo7.song@samsung.com [Fri, 7 Mar 2014 02:23:02 +0000 (02:23 +0000)]
Remove unused method from BatteryController
https://bugs.webkit.org/show_bug.cgi?id=129850

Reviewed by Gyuyoung Kim.

isActive() method is never called anywhere.

* Modules/battery/BatteryController.cpp:
* Modules/battery/BatteryController.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165241 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[iOS] Crash on launch with website restrictions enabled
psolanki@apple.com [Fri, 7 Mar 2014 02:21:29 +0000 (02:21 +0000)]
[iOS] Crash on launch with website restrictions enabled
https://bugs.webkit.org/show_bug.cgi?id=129854
<rdar://problem/16207016>

Reviewed by Simon Fraser.

* platform/mac/ContentFilterMac.mm:
(WebCore::ContentFilter::ContentFilter): Initialize m_neFilterSourceQueue so that we don't
crash in the dtor due to garbage value in the field.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165240 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agodetectFailures in run-jsc-stress-tests can report everything was ok if too many tests...
mhahnenberg@apple.com [Fri, 7 Mar 2014 02:18:38 +0000 (02:18 +0000)]
detectFailures in run-jsc-stress-tests can report everything was ok if too many tests fail
https://bugs.webkit.org/show_bug.cgi?id=129837

Reviewed by Michael Saboff.

ls test_fail_* can fail if there are too many files that match. We then swallow that error, which makes everything seem ok.
The fix is to use find instead of ls like we do in shell-runner.sh.

* Scripts/run-jsc-stress-tests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165239 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agofilter-build-webkit should not strip .o
commit-queue@webkit.org [Fri, 7 Mar 2014 02:16:20 +0000 (02:16 +0000)]
filter-build-webkit should not strip .o
https://bugs.webkit.org/show_bug.cgi?id=129614

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2014-03-06
Reviewed by Dean Jackson.

This makes it much easier to jump from the failing output at
the end back up to the compile errors earlier in the output.

* Scripts/filter-build-webkit:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165238 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUpdate WebKitSystemInterface.
dino@apple.com [Fri, 7 Mar 2014 02:14:04 +0000 (02:14 +0000)]
Update WebKitSystemInterface.
<rdar://problem/16256214>

* libWebKitSystemInterfaceLion.a:
* libWebKitSystemInterfaceMavericks.a:
* libWebKitSystemInterfaceMountainLion.a:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165237 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove one more setCustomFixedPositionRect() that I missed.
simon.fraser@apple.com [Fri, 7 Mar 2014 02:06:25 +0000 (02:06 +0000)]
Remove one more setCustomFixedPositionRect() that I missed.

* UIProcess/DrawingAreaProxy.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165236 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoSend the fixed position rect to the WebProcess along with the other rects
simon.fraser@apple.com [Fri, 7 Mar 2014 02:01:39 +0000 (02:01 +0000)]
Send the fixed position rect to the WebProcess along with the other rects
https://bugs.webkit.org/show_bug.cgi?id=129856

Reviewed by Benjamin Poulain.

Remove the functions that pass the custom fixed position rect through
the DrawingArea, and replace them by adding this rect to the
VisibleContentRectUpdateInfo, along with the "is stable" flag. We
then set the custom fixed position rect in the web process for
stable updates.

* Shared/VisibleContentRectUpdateInfo.cpp:
(WebKit::VisibleContentRectUpdateInfo::encode):
(WebKit::VisibleContentRectUpdateInfo::decode):
* Shared/VisibleContentRectUpdateInfo.h:
(WebKit::VisibleContentRectUpdateInfo::VisibleContentRectUpdateInfo):
(WebKit::VisibleContentRectUpdateInfo::customFixedPositionRect):
(WebKit::VisibleContentRectUpdateInfo::inStableState):
(WebKit::operator==):
* UIProcess/DrawingAreaProxy.cpp:
* UIProcess/ios/WKContentView.mm:
(-[WKContentView didUpdateVisibleRect:unobscuredRect:scale:inStableState:]):
* WebProcess/WebPage/DrawingArea.h:
* WebProcess/WebPage/DrawingArea.messages.in:
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::updateVisibleContentRects):
* WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.h:
* WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.mm:
* WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165235 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoRemove empty class extension in WKContentViewInteraction
simon.fraser@apple.com [Fri, 7 Mar 2014 01:31:54 +0000 (01:31 +0000)]
Remove empty class extension in WKContentViewInteraction
https://bugs.webkit.org/show_bug.cgi?id=129849

Reviewed by Enrica Casucci.

This empty class extension is cruft left over from refactoring.
It can be removed.

* UIProcess/ios/WKContentViewInteraction.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165234 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoMinor optimization in ScrollingTreeScrollingNodeMac
simon.fraser@apple.com [Fri, 7 Mar 2014 01:31:53 +0000 (01:31 +0000)]
Minor optimization in ScrollingTreeScrollingNodeMac
https://bugs.webkit.org/show_bug.cgi?id=129848

Reviewed by Dean Jackson.

No need to call scrollOffsetForFixedPosition() again if we don't have header
and footer layers.

* page/scrolling/mac/ScrollingTreeScrollingNodeMac.mm:
(WebCore::ScrollingTreeScrollingNodeMac::setScrollLayerPosition):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165233 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoFix FTL build.
oliver@apple.com [Fri, 7 Mar 2014 00:54:34 +0000 (00:54 +0000)]
Fix FTL build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165232 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years ago[iOS] didUpdateVisibleRect: should use fabs, not abs
simon.fraser@apple.com [Fri, 7 Mar 2014 00:47:18 +0000 (00:47 +0000)]
[iOS] didUpdateVisibleRect: should use fabs, not abs
https://bugs.webkit.org/show_bug.cgi?id=129843

Reviewed by Benjamin Poulain.

When zooming we sent messages to the web process with an unchanging
zoom scale because of use of abs() rather than fabs().

* UIProcess/ios/WKContentView.mm:
(-[WKContentView didUpdateVisibleRect:unobscuredRect:scale:inStableState:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165230 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoXMLHttpRequest: status and statusText throw DOM Exception 11 when the state is UNSENT...
mark.lam@apple.com [Fri, 7 Mar 2014 00:26:55 +0000 (00:26 +0000)]
XMLHttpRequest: status and statusText throw DOM Exception 11 when the state is UNSENT or OPENED.
https://bugs.webkit.org/show_bug.cgi?id=45994

Not reviewed.

Re-landing r161051 (originally by Ryosuke Niwa, reviewed by Alexey Proskuryakov)
since https://bugs.webkit.org/show_bug.cgi?id=126219 is no longer an issue.

Source/WebCore:

* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::status):
(WebCore::XMLHttpRequest::statusText):
* xml/XMLHttpRequest.h:
* xml/XMLHttpRequest.idl:

LayoutTests:

* http/tests/xmlhttprequest/exceptions-expected.txt:
* http/tests/xmlhttprequest/exceptions.html:
* http/tests/xmlhttprequest/status-after-abort-expected.txt:
* http/tests/xmlhttprequest/zero-length-response-expected.txt:
* js/dom/dfg-custom-getter-throw-expected.txt:
* js/dom/script-tests/dfg-custom-getter-throw-inlined.js:
(foo):
(bar):
* js/dom/script-tests/dfg-custom-getter-throw.js:
(foo):
(bar):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165229 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8 years agoUnreviewed build fix after r165128.
bfulgham@apple.com [Fri, 7 Mar 2014 00:06:39 +0000 (00:06 +0000)]
Unreviewed build fix after r165128.

* JavaScriptCore.vcxproj/JavaScriptCore.vcxproj: The SEH flag was not getting set when
performing 'Production' and 'DebugSuffix' type builds.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@165228 268f45cc-cd09-0410-ab3c-d52691b4dbfc