[webkit-changes] [205470] trunk/Source/WebKit2

[zandobersek]

Title: [205470] trunk/Source/WebKit2








Revision 205470
Author zandober...@gmail.com
Date 2016-09-05 22:57:16 -0700 (Mon, 05 Sep 2016)


Log Message
Fix ENABLE(GAMEPAD) build errors on non-Cocoa platforms
https://bugs.webkit.org/show_bug.cgi?id=161596

Reviewed by Darin Adler.

This patch fixes build errors popping up when enabling Gamepad
feature on non-Cocoa platforms.

* CMakeLists.txt: Add the Source/WebCore/platform/gamepad/ directory to
WebKit2_INCLUDE_DIRECTORIES. Add GamepadData.cpp, UIGamepad.cpp files to the build.
* UIProcess/Gamepad/UIGamepadProvider.cpp: Don't include Cocoa-specific
HIDGamepadProvider.h file. Update methods for the UIGamepadProvider class.
(WebKit::UIGamepadProvider::platformWebPageProxyForGamepadInput):
(WebKit::UIGamepadProvider::platformStartMonitoringGamepads): Deleted.
(WebKit::UIGamepadProvider::platformStopMonitoringGamepads): Deleted.
(WebKit::UIGamepadProvider::platformGamepads): Deleted.
(WebKit::UIGamepadProvider::platformWebProcessProxyForGamepadInput): Deleted.

Modified Paths

trunk/Source/WebKit2/CMakeLists.txt
trunk/Source/WebKit2/ChangeLog
trunk/Source/WebKit2/UIProcess/Gamepad/UIGamepadProvider.cpp




Diff

Modified: trunk/Source/WebKit2/CMakeLists.txt (205469 => 205470)

--- trunk/Source/WebKit2/CMakeLists.txt	2016-09-06 04:21:23 UTC (rev 205469)
+++ trunk/Source/WebKit2/CMakeLists.txt	2016-09-06 05:57:16 UTC (rev 205470)
@@ -116,6 +116,7 @@
 "${WEBCORE_DIR}/platform"
 "${WEBCORE_DIR}/platform/animation"
 "${WEBCORE_DIR}/platform/audio"
+"${WEBCORE_DIR}/platform/gamepad"
 "${WEBCORE_DIR}/platform/graphics"
 "${WEBCORE_DIR}/platform/graphics/filters"
 "${WEBCORE_DIR}/platform/graphics/harfbuzz"
@@ -331,6 +332,8 @@
 Shared/Databases/IndexedDB/IDBUtilities.cpp
 Shared/Databases/IndexedDB/WebIDBResult.cpp
 
+Shared/Gamepad/GamepadData.cpp
+
 Shared/Plugins/NPIdentifierData.cpp
 Shared/Plugins/NPObjectMessageReceiver.cpp
 Shared/Plugins/NPObjectProxy.cpp
@@ -484,6 +487,7 @@
 UIProcess/Downloads/DownloadProxy.cpp
 UIProcess/Downloads/DownloadProxyMap.cpp
 
+UIProcess/Gamepad/UIGamepad.cpp
 UIProcess/Gamepad/UIGamepadProvider.cpp
 
 UIProcess/InspectorServer/HTTPRequest.cpp


Modified: trunk/Source/WebKit2/ChangeLog (205469 => 205470)

--- trunk/Source/WebKit2/ChangeLog	2016-09-06 04:21:23 UTC (rev 205469)
+++ trunk/Source/WebKit2/ChangeLog	2016-09-06 05:57:16 UTC (rev 205470)
@@ -1,3 +1,23 @@
+2016-09-05  Zan Dobersek  
+
+Fix ENABLE(GAMEPAD) build errors on non-Cocoa platforms
+https://bugs.webkit.org/show_bug.cgi?id=161596
+
+Reviewed by Darin Adler.
+
+This patch fixes build errors popping up when enabling Gamepad
+feature on non-Cocoa platforms.
+
+* CMakeLists.txt: Add the Source/WebCore/platform/gamepad/ directory to
+WebKit2_INCLUDE_DIRECTORIES. Add GamepadData.cpp, UIGamepad.cpp files to the build.
+* UIProcess/Gamepad/UIGamepadProvider.cpp: Don't include Cocoa-specific
+HIDGamepadProvider.h file. Update methods for the UIGamepadProvider class.
+(WebKit::UIGamepadProvider::platformWebPageProxyForGamepadInput):
+(WebKit::UIGamepadProvider::platformStartMonitoringGamepads): Deleted.
+(WebKit::UIGamepadProvider::platformStopMonitoringGamepads): Deleted.
+(WebKit::UIGamepadProvider::platformGamepads): Deleted.
+(WebKit::UIGamepadProvider::platformWebProcessProxyForGamepadInput): Deleted.
+
 2016-09-05  Tim Horton  
 
 Remove some dead code in ViewGestureControllerMac


Modified: trunk/Source/WebKit2/UIProcess/Gamepad/UIGamepadProvider.cpp (205469 => 205470)

--- trunk/Source/WebKit2/UIProcess/Gamepad/UIGamepadProvider.cpp	2016-09-06 04:21:23 UTC (rev 205469)
+++ trunk/Source/WebKit2/UIProcess/Gamepad/UIGamepadProvider.cpp	2016-09-06 05:57:16 UTC (rev 205470)
@@ -31,8 +31,8 @@
 #include "GamepadData.h"
 #include "UIGamepad.h"
 #include "WebProcessPool.h"
-#include 
 #include 
+#include 
 #include 
 
 using namespace WebCore;
@@ -230,29 +230,12 @@
 // FIXME: Implement for other platforms
 }
 
-void UIGamepadProvider::platformStartMonitoringGamepads()
+WebPageProxy* UIGamepadProvider::platformWebPageProxyForGamepadInput()
 {
 // FIXME: Implement for other platforms
+return nullptr;
 }
 
-void UIGamepadProvider::platformStopMonitoringGamepads()
-{
-// FIXME: Implement for other platforms
-}
-
-const Vector& UIGamepadProvider::platformGamepads()
-{
-static NeverDestroyed> emptyGamepads;
-return emptyGamepads;
-
-// FIXME: Implement for other platforms
-}
-
-WebProcessProxy* UIGamepadProvider::platformWebProcessProxyForGamepadInput()
-{
-// FIXME: Implement for other platforms
-}
-
 void UIGamepadProvider::platformStopMonitoringInput()
 {
 }






___
webkit-changes mailing list

[webkit-changes] [205469] trunk/Source/WebKit2

[timothy_horton]

Title: [205469] trunk/Source/WebKit2








Revision 205469
Author timothy_hor...@apple.com
Date 2016-09-05 21:21:23 -0700 (Mon, 05 Sep 2016)


Log Message
Remove some dead code in ViewGestureControllerMac
https://bugs.webkit.org/show_bug.cgi?id=161616

Reviewed by Darin Adler.

* UIProcess/mac/ViewGestureControllerMac.mm:
(WebKit::ViewGestureController::endSwipeGesture):
All other LEGACY_SWIPE_SHADOW_STYLE code has been removed, except for this.

Modified Paths

trunk/Source/WebKit2/ChangeLog
trunk/Source/WebKit2/UIProcess/mac/ViewGestureControllerMac.mm




Diff

Modified: trunk/Source/WebKit2/ChangeLog (205468 => 205469)

--- trunk/Source/WebKit2/ChangeLog	2016-09-06 03:33:50 UTC (rev 205468)
+++ trunk/Source/WebKit2/ChangeLog	2016-09-06 04:21:23 UTC (rev 205469)
@@ -1,3 +1,14 @@
+2016-09-05  Tim Horton  
+
+Remove some dead code in ViewGestureControllerMac
+https://bugs.webkit.org/show_bug.cgi?id=161616
+
+Reviewed by Darin Adler.
+
+* UIProcess/mac/ViewGestureControllerMac.mm:
+(WebKit::ViewGestureController::endSwipeGesture):
+All other LEGACY_SWIPE_SHADOW_STYLE code has been removed, except for this.
+
 2016-09-05  Darin Adler  
 
 More bindings improvements, particularly things not needed for _javascript_ bindings


Modified: trunk/Source/WebKit2/UIProcess/mac/ViewGestureControllerMac.mm (205468 => 205469)

--- trunk/Source/WebKit2/UIProcess/mac/ViewGestureControllerMac.mm	2016-09-06 03:33:50 UTC (rev 205468)
+++ trunk/Source/WebKit2/UIProcess/mac/ViewGestureControllerMac.mm	2016-09-06 04:21:23 UTC (rev 205469)
@@ -752,12 +752,6 @@
 
 m_swipeCancellationTracker = nullptr;
 
-#if ENABLE(LEGACY_SWIPE_SHADOW_STYLE)
-CALayer *rootLayer = m_webPageProxy.acceleratedCompositingRootLayer();
-rootLayer.shadowOpacity = 0;
-rootLayer.shadowRadius = 0;
-#endif
-
 if (cancelled) {
 removeSwipeSnapshot();
 m_webPageProxy.navigationGestureDidEnd(false, *targetItem);






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205468] trunk

[darin]

Title: [205468] trunk








Revision 205468
Author da...@apple.com
Date 2016-09-05 20:33:50 -0700 (Mon, 05 Sep 2016)


Log Message
More bindings improvements, particularly things not needed for _javascript_ bindings
https://bugs.webkit.org/show_bug.cgi?id=161572

Reviewed by Sam Weinig.

Source/WebCore:

* WebCore.xcodeproj/project.pbxproj: Export a couple of files needed for legacy bindings.

* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::addCrossOriginWindowPropertyNames): Mark array const.
* bindings/js/JSLocationCustom.cpp:
(WebCore::addCrossOriginLocationPropertyNames): Ditto.

* dom/DOMImplementation.cpp:
(WebCore::addString): Deleted.
(WebCore::isSupportedSVG10Feature): Deleted.
(WebCore::isSupportedSVG11Feature): Deleted.
(WebCore::DOMImplementation::hasFeature): Deleted.

* dom/DOMImplementation.h: Changed hasFeature to take no arguments and to always
return true, as specified in the DOM spec.

* dom/DOMImplementation.idl: Removed the arguments to hasFeature.

* dom/Document.cpp:
(WebCore::Document::defaultCharsetForLegacyBindings): Renamed from
defaultCharsetForBindings, because this function is used only by the
non-_javascript_ bindings.
* dom/Document.h: Updated for the above.

* dom/DocumentType.h: Removed entitiesForBindings,
notationsForBindings, and internalSubsetForBindings, all of which were
empty functions. The empty implementations are now in the legacy bindings.

* dom/Element.cpp:
(WebCore::Element::dispatchSimulatedClickForBindings): Deleted.
This code is now at the single call site, in HTMLElement.
* dom/Element.h: Updated for the above change.

* dom/Node.cpp:
(WebCore::Node::isSupportedForBindings): Deleted. This is only used in the
legacy bindings, and since it only returns a value other than "true" for
SVG features, it is now in the SVGTests class.
* dom/Node.h: Updated for the above change.

* html/HTMLElement.cpp:
(WebCore::HTMLElement::click): Call simulateClick with the appropriate
arguments rather than calling dispatchSimulatedClickForBindings.

* svg/SVGTests.cpp:
(WebCore::supportedSVGFeatures): Added. Moved here from DOMImplementation,
but also refactored.
(WebCore::SVGTests::isValid): Changed to use the function above, rather than
calling DOMImplementation::hasFeature. This change fixes a bug where feature names
that do not look like valid SVG feature names, names such as "xxx", were returning
true claiming that the feature was supported. This behavior was helpful in the
general DOMImplementation function, but harmful here. This bug was causing test
failures in some of our SVG tests.
(WebCore::SVGTests::hasFeatureForLegacyBindings): Added. Uses the set above to
implement the legacy behavior of hasFeature. It's better to have this in parallel
with the SVGTests::isValid function rather than having either depend on the other.

* svg/SVGTests.h: Added exported SVGTests::isValid function.

Source/WebKit/mac:

* DOM/DOMDOMImplementation.mm:
(-[DOMImplementation hasFeature:version:]): Call SVGTests::hasFeatureForLegacyBindings
instead of DOMImplementation::hasFeature. Also removed JSMainThreadNullState
since this does not do anything that can invoke _javascript_.

* DOM/DOMDocument.mm:
(-[DOMDocument defaultCharset]): Call defaultCharsetForLegacyBindings. Also removed
JSMainThreadNullState.

* DOM/DOMDocumentType.mm:
(-[DOMDocumentType entities]): Changed to simply return nil; same behavior as before,
but with no dependency on WebCore. Also removed JSMainThreadNullState.
(-[DOMDocumentType notations]): Ditto.
(-[DOMDocumentType internalSubset]): Ditto, except empty string rather than nil.

* DOM/DOMNode.mm:
(-[DOMNode isSupported:version:]): Call SVGTests::hasFeatureForLegacyBindings
instead of Node::isSupportedForBindings. Also removed JSMainThreadNullState.

Source/WebKit2:

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDOMImplementation.cpp:
(webkit_dom_dom_implementation_has_feature): Call SVGTests::hasFeatureForLegacyBindings
instead of DOMImplementation::hasFeature. Also removed JSMainThreadNullState
since this does not do anything that can invoke _javascript_.

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDeprecated.cpp:
(webkit_dom_document_get_default_charset): Call defaultCharsetForLegacyBindings.
Also removed JSMainThreadNullState.

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDocumentType.cpp:
(webkit_dom_document_type_get_entities): Changed to simply return nullptr; same behavior
as before, but with no dependency on WebCore. Also removed JSMainThreadNullState.
(webkit_dom_document_type_get_notations): Ditto.
(webkit_dom_document_type_get_internal_subset): Ditto.

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMNode.cpp:
(webkit_dom_node_is_supported): Call SVGTests::hasFeatureForLegacyBindings
instead of Node::isSupportedForBindings. Also removed JSMainThreadNullState.

LayoutTests:

* TestExpectations: Updated to expect success instead of failure for the
imported/mozilla/svg tests named conditions-01.svg, conditions-06.svg,

[webkit-changes] [205467] trunk/Source/WebKit2

[mcatanzaro]

Title: [205467] trunk/Source/WebKit2








Revision 205467
Author mcatanz...@igalia.com
Date 2016-09-05 20:16:22 -0700 (Mon, 05 Sep 2016)


Log Message
Unreviewed, update stale variable name

* UIProcess/API/gtk/WebKitWebViewBase.cpp:
(webkitWebViewBaseCreateWebPage):

Modified Paths

trunk/Source/WebKit2/ChangeLog
trunk/Source/WebKit2/UIProcess/API/gtk/WebKitWebViewBase.cpp




Diff

Modified: trunk/Source/WebKit2/ChangeLog (205466 => 205467)

--- trunk/Source/WebKit2/ChangeLog	2016-09-06 02:39:11 UTC (rev 205466)
+++ trunk/Source/WebKit2/ChangeLog	2016-09-06 03:16:22 UTC (rev 205467)
@@ -1,3 +1,10 @@
+2016-09-05  Michael Catanzaro  
+
+Unreviewed, update stale variable name
+
+* UIProcess/API/gtk/WebKitWebViewBase.cpp:
+(webkitWebViewBaseCreateWebPage):
+
 2016-08-31  Filip Pizlo  
 
 Butterflies should be allocated in Auxiliary MarkedSpace instead of CopiedSpace and we should rewrite as much of the GC as needed to make this not a regression


Modified: trunk/Source/WebKit2/UIProcess/API/gtk/WebKitWebViewBase.cpp (205466 => 205467)

--- trunk/Source/WebKit2/UIProcess/API/gtk/WebKitWebViewBase.cpp	2016-09-06 02:39:11 UTC (rev 205466)
+++ trunk/Source/WebKit2/UIProcess/API/gtk/WebKitWebViewBase.cpp	2016-09-06 03:16:22 UTC (rev 205467)
@@ -1176,8 +1176,8 @@
 void webkitWebViewBaseCreateWebPage(WebKitWebViewBase* webkitWebViewBase, Ref&& configuration)
 {
 WebKitWebViewBasePrivate* priv = webkitWebViewBase->priv;
-WebProcessPool* context = configuration->processPool();
-priv->pageProxy = context->createWebPage(*priv->pageClient, WTFMove(configuration));
+WebProcessPool* processPool = configuration->processPool();
+priv->pageProxy = processPool->createWebPage(*priv->pageClient, WTFMove(configuration));
 priv->pageProxy->initializeWebPage();
 
 priv->acceleratedBackingStore = AcceleratedBackingStore::create(*priv->pageProxy);






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205466] trunk/Source/JavaScriptCore

[fpizlo]

Title: [205466] trunk/Source/_javascript_Core








Revision 205466
Author fpi...@apple.com
Date 2016-09-05 19:39:11 -0700 (Mon, 05 Sep 2016)


Log Message
Heap::isMarked() should use concurrent lazy flipping
https://bugs.webkit.org/show_bug.cgi?id=161613

Reviewed by Michael Catanzaro.

I found out about this race condition via
https://bugs.webkit.org/show_bug.cgi?id=160125#c233.

The problem is that we use isMarked, and maybe even isLive, inside the concurrent mark
phase. So, they need to lazy-flip in a non-racy way.

* heap/HeapInlines.h:
(JSC::Heap::isLive):
(JSC::Heap::isMarked):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/heap/HeapInlines.h




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (205465 => 205466)

--- trunk/Source/_javascript_Core/ChangeLog	2016-09-06 02:16:03 UTC (rev 205465)
+++ trunk/Source/_javascript_Core/ChangeLog	2016-09-06 02:39:11 UTC (rev 205466)
@@ -1,5 +1,22 @@
 2016-09-05  Filip Pizlo  
 
+Heap::isMarked() should use concurrent lazy flipping
+https://bugs.webkit.org/show_bug.cgi?id=161613
+
+Reviewed by Michael Catanzaro.
+
+I found out about this race condition via
+https://bugs.webkit.org/show_bug.cgi?id=160125#c233.
+
+The problem is that we use isMarked, and maybe even isLive, inside the concurrent mark
+phase. So, they need to lazy-flip in a non-racy way.
+
+* heap/HeapInlines.h:
+(JSC::Heap::isLive):
+(JSC::Heap::isMarked):
+
+2016-09-05  Filip Pizlo  
+
 Unreviewed, reset generator test results after the butterflies.
 
 * Scripts/tests/builtins/expected/_javascript_Core-Builtin.Promise-Combined.js-result:


Modified: trunk/Source/_javascript_Core/heap/HeapInlines.h (205465 => 205466)

--- trunk/Source/_javascript_Core/heap/HeapInlines.h	2016-09-06 02:16:03 UTC (rev 205465)
+++ trunk/Source/_javascript_Core/heap/HeapInlines.h	2016-09-06 02:39:11 UTC (rev 205466)
@@ -80,7 +80,7 @@
 if (cell->isLargeAllocation())
 return cell->largeAllocation().isLive();
 MarkedBlock& block = cell->markedBlock();
-block.flipIfNecessary(block.vm()->heap.objectSpace().version());
+block.flipIfNecessaryConcurrently(block.vm()->heap.objectSpace().version());
 return block.handle().isLiveCell(cell);
 }
 
@@ -90,7 +90,7 @@
 if (cell->isLargeAllocation())
 return cell->largeAllocation().isMarked();
 MarkedBlock& block = cell->markedBlock();
-block.flipIfNecessary(block.vm()->heap.objectSpace().version());
+block.flipIfNecessaryConcurrently(block.vm()->heap.objectSpace().version());
 return block.isMarked(cell);
 }
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205465] trunk/Source/JavaScriptCore

[fpizlo]

Title: [205465] trunk/Source/_javascript_Core








Revision 205465
Author fpi...@apple.com
Date 2016-09-05 19:16:03 -0700 (Mon, 05 Sep 2016)


Log Message
Unreviewed, reset generator test results after the butterflies.

* Scripts/tests/builtins/expected/_javascript_Core-Builtin.Promise-Combined.js-result:
* Scripts/tests/builtins/expected/_javascript_Core-Builtin.Promise-Separate.js-result:
* Scripts/tests/builtins/expected/_javascript_Core-Builtin.prototype-Combined.js-result:
* Scripts/tests/builtins/expected/_javascript_Core-Builtin.prototype-Separate.js-result:
* Scripts/tests/builtins/expected/_javascript_Core-BuiltinConstructor-Combined.js-result:
* Scripts/tests/builtins/expected/_javascript_Core-BuiltinConstructor-Separate.js-result:
* Scripts/tests/builtins/expected/_javascript_Core-InternalClashingNames-Combined.js-result:
* Scripts/tests/builtins/expected/WebCore-AnotherGuardedInternalBuiltin-Separate.js-result:
* Scripts/tests/builtins/expected/WebCore-ArbitraryConditionalGuard-Separate.js-result:
* Scripts/tests/builtins/expected/WebCore-GuardedBuiltin-Separate.js-result:
* Scripts/tests/builtins/expected/WebCore-GuardedInternalBuiltin-Separate.js-result:
* Scripts/tests/builtins/expected/WebCore-UnguardedBuiltin-Separate.js-result:
* Scripts/tests/builtins/expected/WebCore-xmlCasingTest-Separate.js-result:
* Scripts/tests/builtins/expected/WebCoreJSBuiltins.h-result:

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/_javascript_Core-Builtin.Promise-Combined.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/_javascript_Core-Builtin.Promise-Separate.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/_javascript_Core-Builtin.prototype-Combined.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/_javascript_Core-Builtin.prototype-Separate.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/_javascript_Core-BuiltinConstructor-Combined.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/_javascript_Core-BuiltinConstructor-Separate.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/_javascript_Core-InternalClashingNames-Combined.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/WebCore-AnotherGuardedInternalBuiltin-Separate.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/WebCore-ArbitraryConditionalGuard-Separate.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/WebCore-GuardedBuiltin-Separate.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/WebCore-GuardedInternalBuiltin-Separate.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/WebCore-UnguardedBuiltin-Separate.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/WebCore-xmlCasingTest-Separate.js-result
trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/WebCoreJSBuiltins.h-result




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (205464 => 205465)

--- trunk/Source/_javascript_Core/ChangeLog	2016-09-06 02:08:52 UTC (rev 205464)
+++ trunk/Source/_javascript_Core/ChangeLog	2016-09-06 02:16:03 UTC (rev 205465)
@@ -1,5 +1,24 @@
 2016-09-05  Filip Pizlo  
 
+Unreviewed, reset generator test results after the butterflies.
+
+* Scripts/tests/builtins/expected/_javascript_Core-Builtin.Promise-Combined.js-result:
+* Scripts/tests/builtins/expected/_javascript_Core-Builtin.Promise-Separate.js-result:
+* Scripts/tests/builtins/expected/_javascript_Core-Builtin.prototype-Combined.js-result:
+* Scripts/tests/builtins/expected/_javascript_Core-Builtin.prototype-Separate.js-result:
+* Scripts/tests/builtins/expected/_javascript_Core-BuiltinConstructor-Combined.js-result:
+* Scripts/tests/builtins/expected/_javascript_Core-BuiltinConstructor-Separate.js-result:
+* Scripts/tests/builtins/expected/_javascript_Core-InternalClashingNames-Combined.js-result:
+* Scripts/tests/builtins/expected/WebCore-AnotherGuardedInternalBuiltin-Separate.js-result:
+* Scripts/tests/builtins/expected/WebCore-ArbitraryConditionalGuard-Separate.js-result:
+* Scripts/tests/builtins/expected/WebCore-GuardedBuiltin-Separate.js-result:
+* Scripts/tests/builtins/expected/WebCore-GuardedInternalBuiltin-Separate.js-result:
+* Scripts/tests/builtins/expected/WebCore-UnguardedBuiltin-Separate.js-result:
+* Scripts/tests/builtins/expected/WebCore-xmlCasingTest-Separate.js-result:
+* Scripts/tests/builtins/expected/WebCoreJSBuiltins.h-result:
+
+2016-09-05  Filip Pizlo  
+
 Unreviewed, fix cloop build.
 
 * bytecode/SuperSampler.cpp:


Modified: trunk/Source/_javascript_Core/Scripts/tests/builtins/expected/_javascript_Core-Builtin.Promise-Combined.js-result (205464 => 205465)

--- 

[webkit-changes] [205463] trunk/Source/JavaScriptCore

[fpizlo]

Title: [205463] trunk/Source/_javascript_Core








Revision 205463
Author fpi...@apple.com
Date 2016-09-05 18:11:19 -0700 (Mon, 05 Sep 2016)


Log Message
Unreviewed, fix cloop build.

* bytecode/SuperSampler.cpp:

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/bytecode/SuperSampler.cpp




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (205462 => 205463)

--- trunk/Source/_javascript_Core/ChangeLog	2016-09-06 01:02:22 UTC (rev 205462)
+++ trunk/Source/_javascript_Core/ChangeLog	2016-09-06 01:11:19 UTC (rev 205463)
@@ -1,3 +1,9 @@
+2016-09-05  Filip Pizlo  
+
+Unreviewed, fix cloop build.
+
+* bytecode/SuperSampler.cpp:
+
 2016-08-31  Filip Pizlo  
 
 Butterflies should be allocated in Auxiliary MarkedSpace instead of CopiedSpace and we should rewrite as much of the GC as needed to make this not a regression


Modified: trunk/Source/_javascript_Core/bytecode/SuperSampler.cpp (205462 => 205463)

--- trunk/Source/_javascript_Core/bytecode/SuperSampler.cpp	2016-09-06 01:02:22 UTC (rev 205462)
+++ trunk/Source/_javascript_Core/bytecode/SuperSampler.cpp	2016-09-06 01:11:19 UTC (rev 205463)
@@ -30,6 +30,7 @@
 #include "Options.h"
 #include 
 #include 
+#include 
 #include 
 
 namespace JSC {






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205461] trunk/Tools

[mcatanzaro]

Title: [205461] trunk/Tools








Revision 205461
Author mcatanz...@igalia.com
Date 2016-09-05 17:03:08 -0700 (Mon, 05 Sep 2016)


Log Message
"Redundant argument in sprintf" warning spam from prepare-ChangeLog
https://bugs.webkit.org/show_bug.cgi?id=161606

Reviewed by Darin Adler.

Avoid extra arguments in call to sprintf.

* Scripts/prepare-ChangeLog:
(statusDescription):

Modified Paths

trunk/Tools/ChangeLog
trunk/Tools/Scripts/prepare-ChangeLog




Diff

Modified: trunk/Tools/ChangeLog (205460 => 205461)

--- trunk/Tools/ChangeLog	2016-09-05 23:57:17 UTC (rev 205460)
+++ trunk/Tools/ChangeLog	2016-09-06 00:03:08 UTC (rev 205461)
@@ -1,5 +1,17 @@
 2016-09-05  Michael Catanzaro  
 
+"Redundant argument in sprintf" warning spam from prepare-ChangeLog
+https://bugs.webkit.org/show_bug.cgi?id=161606
+
+Reviewed by Darin Adler.
+
+Avoid extra arguments in call to sprintf.
+
+* Scripts/prepare-ChangeLog:
+(statusDescription):
+
+2016-09-05  Michael Catanzaro  
+
 [GTK] Skip several failing API tests
 
 Unfortunately we don't have any form of test expectations for API tests, so we'll just skip


Modified: trunk/Tools/Scripts/prepare-ChangeLog (205460 => 205461)

--- trunk/Tools/Scripts/prepare-ChangeLog	2016-09-05 23:57:17 UTC (rev 205460)
+++ trunk/Tools/Scripts/prepare-ChangeLog	2016-09-06 00:03:08 UTC (rev 205461)
@@ -2325,21 +2325,21 @@
 my $propertyDescription = defined $propertyChanges ? propertyChangeDescription($propertyChanges) : "";
 
 my %svn = (
-"A" => defined $original ? " Copied from \%s." : " Added.",
+"A" => defined $original ? sprintf(" Copied from \%s.", $original) : " Added.",
 "D" => " Removed.",
 "M" => "",
-"R" => defined $original ? " Replaced with \%s." : " Replaced.",
+"R" => defined $original ? sprintf(" Replaced with \%s.", $original) : " Replaced.",
 " " => "",
 );
 
 my %git = %svn;
 $git{"A"} = " Added.";
-$git{"C"} = " Copied from \%s.";
-$git{"R"} = " Renamed from \%s.";
+$git{"C"} = sprintf(" Copied from \%s.", $original);
+$git{"R"} = sprintf(" Renamed from \%s.", $original);
 
 my $description;
-$description = sprintf($svn{$status}, $original) if isSVN() && exists $svn{$status};
-$description = sprintf($git{$status}, $original) if isGit() && exists $git{$status};
+$description = $svn{$status} if isSVN() && exists $svn{$status};
+$description = $git{$status} if isGit() && exists $git{$status};
 return unless defined $description;
 
 $description .= $propertyDescription unless isAddedStatus($status);






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205460] trunk/Source/WebKit2

[mcatanzaro]

Title: [205460] trunk/Source/WebKit2








Revision 205460
Author mcatanz...@igalia.com
Date 2016-09-05 16:57:17 -0700 (Mon, 05 Sep 2016)


Log Message
Unreviewed error message typo fix

* WebProcess/UserContent/WebUserContentController.cpp:
(WebKit::WebUserContentController::removeUserContentWorlds):

Modified Paths

trunk/Source/WebKit2/ChangeLog
trunk/Source/WebKit2/WebProcess/UserContent/WebUserContentController.cpp




Diff

Modified: trunk/Source/WebKit2/ChangeLog (205459 => 205460)

--- trunk/Source/WebKit2/ChangeLog	2016-09-05 20:54:06 UTC (rev 205459)
+++ trunk/Source/WebKit2/ChangeLog	2016-09-05 23:57:17 UTC (rev 205460)
@@ -1,3 +1,10 @@
+2016-09-05  Michael Catanzaro  
+
+Unreviewed error message typo fix
+
+* WebProcess/UserContent/WebUserContentController.cpp:
+(WebKit::WebUserContentController::removeUserContentWorlds):
+
 2016-09-05  Dan Bernstein  
 
 [Cocoa] Drop ".Development" suffix from XPC service bundle IDs, leave it only on the executable


Modified: trunk/Source/WebKit2/WebProcess/UserContent/WebUserContentController.cpp (205459 => 205460)

--- trunk/Source/WebKit2/WebProcess/UserContent/WebUserContentController.cpp	2016-09-05 20:54:06 UTC (rev 205459)
+++ trunk/Source/WebKit2/WebProcess/UserContent/WebUserContentController.cpp	2016-09-05 23:57:17 UTC (rev 205460)
@@ -111,7 +111,7 @@
 
 auto it = worldMap().find(worldIdentifier);
 if (it == worldMap().end()) {
-WTFLogAlways("Trying to remove a UserContentWorld (id=%" PRIu64 ") that is does not exist.", worldIdentifier);
+WTFLogAlways("Trying to remove a UserContentWorld (id=%" PRIu64 ") that does not exist.", worldIdentifier);
 return;
 }
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205459] trunk/Source/WebKit2

[mitz]

Title: [205459] trunk/Source/WebKit2








Revision 205459
Author m...@apple.com
Date 2016-09-05 13:54:06 -0700 (Mon, 05 Sep 2016)


Log Message
[Cocoa] Drop ".Development" suffix from XPC service bundle IDs, leave it only on the executable
https://bugs.webkit.org/show_bug.cgi?id=161575

Reviewed by Sam Weinig.

Append the ".Development" suffix to the executable names, but not to the bundle names and
bundle IDs.

* Configurations/BaseTarget.xcconfig: No need for a preprocessor definition anymore.
* Configurations/BaseXPCService.xcconfig: Set EXECUTABLE_SUFFIX.
* Configurations/DatabaseService.xcconfig: No need to suffix the service name and bundle ID.
* Configurations/NetworkService.xcconfig: Ditto.
* Configurations/PluginService.32.xcconfig: Ditto.
* Configurations/PluginService.64.xcconfig: Ditto.
* Configurations/WebContentService.xcconfig: Ditto.
* UIProcess/Launcher/mac/ProcessLauncherMac.mm:
(WebKit::serviceName): Service names are no longer suffixed.
* WebKit2.xcodeproj/project.pbxproj: Ditto.

Modified Paths

trunk/Source/WebKit2/ChangeLog
trunk/Source/WebKit2/Configurations/BaseTarget.xcconfig
trunk/Source/WebKit2/Configurations/BaseXPCService.xcconfig
trunk/Source/WebKit2/Configurations/DatabaseService.xcconfig
trunk/Source/WebKit2/Configurations/NetworkService.xcconfig
trunk/Source/WebKit2/Configurations/PluginService.32.xcconfig
trunk/Source/WebKit2/Configurations/PluginService.64.xcconfig
trunk/Source/WebKit2/Configurations/WebContentService.xcconfig
trunk/Source/WebKit2/UIProcess/Launcher/mac/ProcessLauncherMac.mm
trunk/Source/WebKit2/WebKit2.xcodeproj/project.pbxproj




Diff

Modified: trunk/Source/WebKit2/ChangeLog (205458 => 205459)

--- trunk/Source/WebKit2/ChangeLog	2016-09-05 17:25:52 UTC (rev 205458)
+++ trunk/Source/WebKit2/ChangeLog	2016-09-05 20:54:06 UTC (rev 205459)
@@ -1,3 +1,24 @@
+2016-09-05  Dan Bernstein  
+
+[Cocoa] Drop ".Development" suffix from XPC service bundle IDs, leave it only on the executable
+https://bugs.webkit.org/show_bug.cgi?id=161575
+
+Reviewed by Sam Weinig.
+
+Append the ".Development" suffix to the executable names, but not to the bundle names and
+bundle IDs.
+
+* Configurations/BaseTarget.xcconfig: No need for a preprocessor definition anymore.
+* Configurations/BaseXPCService.xcconfig: Set EXECUTABLE_SUFFIX.
+* Configurations/DatabaseService.xcconfig: No need to suffix the service name and bundle ID.
+* Configurations/NetworkService.xcconfig: Ditto.
+* Configurations/PluginService.32.xcconfig: Ditto.
+* Configurations/PluginService.64.xcconfig: Ditto.
+* Configurations/WebContentService.xcconfig: Ditto.
+* UIProcess/Launcher/mac/ProcessLauncherMac.mm:
+(WebKit::serviceName): Service names are no longer suffixed.
+* WebKit2.xcodeproj/project.pbxproj: Ditto.
+
 2016-09-05  Gustavo Noronha Silva  
 
 [GTK] GL_PACK_ROW_LENGTH is not available in GLES2


Modified: trunk/Source/WebKit2/Configurations/BaseTarget.xcconfig (205458 => 205459)

--- trunk/Source/WebKit2/Configurations/BaseTarget.xcconfig	2016-09-05 17:25:52 UTC (rev 205458)
+++ trunk/Source/WebKit2/Configurations/BaseTarget.xcconfig	2016-09-05 20:54:06 UTC (rev 205459)
@@ -37,7 +37,7 @@
 FRAMEWORK_SEARCH_PATHS = $(FRAMEWORK_SEARCH_PATHS_base);
 
 GCC_PREFIX_HEADER = WebKit2Prefix.h;
-GCC_PREPROCESSOR_DEFINITIONS = $(DEBUG_DEFINES) $(FEATURE_DEFINES) $(WK_MANUAL_SANDBOXING_DEFINES) WK_XPC_SERVICE_SUFFIX=\"$(WK_XPC_SERVICE_SUFFIX)\" FRAMEWORK_NAME=WebKit;
+GCC_PREPROCESSOR_DEFINITIONS = $(DEBUG_DEFINES) $(FEATURE_DEFINES) $(WK_MANUAL_SANDBOXING_DEFINES) FRAMEWORK_NAME=WebKit;
 WEBKITADDITIONS_HEADER_SEARCH_PATHS = $(BUILT_PRODUCTS_DIR)/usr/local/include/WebKitAdditions $(SDKROOT)/usr/local/include/WebKitAdditions;
 HEADER_SEARCH_PATHS = $(BUILT_PRODUCTS_DIR)/usr/local/include "$(WEBCORE_PRIVATE_HEADERS_DIR)/ForwardingHeaders" "$(WEBCORE_PRIVATE_HEADERS_DIR)/icu" $(BUILT_PRODUCTS_DIR)/DerivedSources/WebKit2 $(WEBKITADDITIONS_HEADER_SEARCH_PATHS) $(HEADER_SEARCH_PATHS);
 


Modified: trunk/Source/WebKit2/Configurations/BaseXPCService.xcconfig (205458 => 205459)

--- trunk/Source/WebKit2/Configurations/BaseXPCService.xcconfig	2016-09-05 17:25:52 UTC (rev 205458)
+++ trunk/Source/WebKit2/Configurations/BaseXPCService.xcconfig	2016-09-05 20:54:06 UTC (rev 205459)
@@ -27,6 +27,8 @@
 VALID_ARCHS_macosx_YES = $(VALID_ARCHS);
 VALID_ARCHS_macosx_NO = $(ARCHS_STANDARD);
 
+EXECUTABLE_SUFFIX = $(WK_XPC_SERVICE_SUFFIX);
+
 INSTALL_PATH[sdk=iphone*] = $(WEBKIT_FRAMEWORKS_DIR)/WebKit.framework/XPCServices;
 INSTALL_PATH[sdk=macosx*] = $(WEBKIT_FRAMEWORKS_DIR)/WebKit.framework/Versions/A/XPCServices;
 


Modified: trunk/Source/WebKit2/Configurations/DatabaseService.xcconfig (205458 => 205459)

--- trunk/Source/WebKit2/Configurations/DatabaseService.xcconfig	2016-09-05 17:25:52 UTC (rev 205458)
+++ 

[webkit-changes] [205458] trunk/Source/WebCore

[commit-queue]

Title: [205458] trunk/Source/WebCore








Revision 205458
Author commit-qu...@webkit.org
Date 2016-09-05 10:25:52 -0700 (Mon, 05 Sep 2016)


Log Message
run-bindings-tests fails since r205422
https://bugs.webkit.org/show_bug.cgi?id=161595

Patch by Fujii Hironori  on 2016-09-05
Reviewed by Darin Adler.

Rebaseline binding tests after r205422.

* bindings/scripts/test/JS/JSInterfaceName.cpp:
* bindings/scripts/test/JS/JSInterfaceName.h:
* bindings/scripts/test/JS/JSTestActiveDOMObject.cpp:
* bindings/scripts/test/JS/JSTestActiveDOMObject.h:
* bindings/scripts/test/JS/JSTestClassWithJSBuiltinConstructor.cpp:
* bindings/scripts/test/JS/JSTestClassWithJSBuiltinConstructor.h:
* bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.cpp:
* bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.h:
* bindings/scripts/test/JS/JSTestCustomNamedGetter.cpp:
* bindings/scripts/test/JS/JSTestCustomNamedGetter.h:
* bindings/scripts/test/JS/JSTestEventConstructor.cpp:
* bindings/scripts/test/JS/JSTestEventConstructor.h:
* bindings/scripts/test/JS/JSTestEventTarget.cpp:
* bindings/scripts/test/JS/JSTestEventTarget.h:
* bindings/scripts/test/JS/JSTestException.cpp:
* bindings/scripts/test/JS/JSTestException.h:
* bindings/scripts/test/JS/JSTestGenerateIsReachable.cpp:
* bindings/scripts/test/JS/JSTestGenerateIsReachable.h:
* bindings/scripts/test/JS/JSTestGlobalObject.cpp:
* bindings/scripts/test/JS/JSTestGlobalObject.h:
* bindings/scripts/test/JS/JSTestInterface.cpp:
* bindings/scripts/test/JS/JSTestInterface.h:
* bindings/scripts/test/JS/JSTestIterable.cpp:
* bindings/scripts/test/JS/JSTestIterable.h:
* bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:
* bindings/scripts/test/JS/JSTestMediaQueryListListener.h:
* bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
* bindings/scripts/test/JS/JSTestNamedConstructor.h:
* bindings/scripts/test/JS/JSTestNode.cpp:
* bindings/scripts/test/JS/JSTestNode.h:
* bindings/scripts/test/JS/JSTestNondeterministic.cpp:
* bindings/scripts/test/JS/JSTestNondeterministic.h:
* bindings/scripts/test/JS/JSTestObj.cpp:
* bindings/scripts/test/JS/JSTestObj.h:
* bindings/scripts/test/JS/JSTestOverloadedConstructors.cpp:
* bindings/scripts/test/JS/JSTestOverloadedConstructors.h:
* bindings/scripts/test/JS/JSTestOverloadedConstructorsWithSequence.cpp:
* bindings/scripts/test/JS/JSTestOverloadedConstructorsWithSequence.h:
* bindings/scripts/test/JS/JSTestOverrideBuiltins.cpp:
* bindings/scripts/test/JS/JSTestOverrideBuiltins.h:
* bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:
* bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.h:
* bindings/scripts/test/JS/JSTestTypedefs.cpp:
* bindings/scripts/test/JS/JSTestTypedefs.h:
* bindings/scripts/test/JS/JSattribute.cpp:
* bindings/scripts/test/JS/JSattribute.h:
* bindings/scripts/test/JS/JSreadonly.cpp:
* bindings/scripts/test/JS/JSreadonly.h:

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/bindings/scripts/test/JS/JSInterfaceName.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSInterfaceName.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestActiveDOMObject.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestActiveDOMObject.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestClassWithJSBuiltinConstructor.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestClassWithJSBuiltinConstructor.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestCustomNamedGetter.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestCustomNamedGetter.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestEventConstructor.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestEventConstructor.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestEventTarget.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestEventTarget.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestException.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestException.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestGenerateIsReachable.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestGenerateIsReachable.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestGlobalObject.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestGlobalObject.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestInterface.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestInterface.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestIterable.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestIterable.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestMediaQueryListListener.h
trunk/Source/WebCore/bindings/scripts/test/JS/JSTestNamedConstructor.cpp

[webkit-changes] [205457] trunk/Source/WebCore

[commit-queue]

Title: [205457] trunk/Source/WebCore








Revision 205457
Author commit-qu...@webkit.org
Date 2016-09-05 10:21:33 -0700 (Mon, 05 Sep 2016)


Log Message
[CMake] Duplicated IDL files in WebCore_IDL_FILES
https://bugs.webkit.org/show_bug.cgi?id=161592

Patch by Fujii Hironori  on 2016-09-05
Reviewed by Darin Adler.

* CMakeLists.txt: Removed duplicated IDL files.
* DerivedSources.cpp: Added #if ENABLE(USER_MESSAGE_HANDLERS)

Modified Paths

trunk/Source/WebCore/CMakeLists.txt
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/DerivedSources.cpp




Diff

Modified: trunk/Source/WebCore/CMakeLists.txt (205456 => 205457)

--- trunk/Source/WebCore/CMakeLists.txt	2016-09-05 17:19:44 UTC (rev 205456)
+++ trunk/Source/WebCore/CMakeLists.txt	2016-09-05 17:21:33 UTC (rev 205457)
@@ -542,7 +542,6 @@
 html/ValidityState.idl
 html/VoidCallback.idl
 
-html/canvas/ANGLEInstancedArrays.idl
 html/canvas/CanvasGradient.idl
 html/canvas/CanvasPath.idl
 html/canvas/CanvasPattern.idl
@@ -549,61 +548,7 @@
 html/canvas/CanvasProxy.idl
 html/canvas/CanvasRenderingContext2D.idl
 html/canvas/DOMPath.idl
-html/canvas/EXTBlendMinMax.idl
-html/canvas/EXTFragDepth.idl
-html/canvas/EXTShaderTextureLOD.idl
-html/canvas/EXTTextureFilterAnisotropic.idl
-html/canvas/EXTsRGB.idl
-html/canvas/OESElementIndexUint.idl
-html/canvas/OESStandardDerivatives.idl
-html/canvas/OESTextureFloat.idl
-html/canvas/OESTextureFloatLinear.idl
-html/canvas/OESTextureHalfFloat.idl
-html/canvas/OESTextureHalfFloatLinear.idl
-html/canvas/OESVertexArrayObject.idl
-html/canvas/WebGL2RenderingContext.idl
-html/canvas/WebGLActiveInfo.idl
-html/canvas/WebGLBuffer.idl
-html/canvas/WebGLCompressedTextureATC.idl
-html/canvas/WebGLCompressedTexturePVRTC.idl
-html/canvas/WebGLCompressedTextureS3TC.idl
-html/canvas/WebGLContextAttributes.idl
-html/canvas/WebGLContextEvent.idl
-html/canvas/WebGLDebugRendererInfo.idl
-html/canvas/WebGLDebugShaders.idl
-html/canvas/WebGLDepthTexture.idl
-html/canvas/WebGLDrawBuffers.idl
-html/canvas/WebGLFramebuffer.idl
-html/canvas/WebGLLoseContext.idl
-html/canvas/WebGLProgram.idl
-html/canvas/WebGLQuery.idl
-html/canvas/WebGLRenderbuffer.idl
-html/canvas/WebGLRenderingContext.idl
-html/canvas/WebGLRenderingContextBase.idl
-html/canvas/WebGLSampler.idl
-html/canvas/WebGLShader.idl
-html/canvas/WebGLShaderPrecisionFormat.idl
-html/canvas/WebGLSync.idl
-html/canvas/WebGLTexture.idl
-html/canvas/WebGLTransformFeedback.idl
-html/canvas/WebGLUniformLocation.idl
-html/canvas/WebGLVertexArrayObject.idl
-html/canvas/WebGLVertexArrayObjectOES.idl
 
-html/track/AudioTrack.idl
-html/track/AudioTrackList.idl
-html/track/DataCue.idl
-html/track/TextTrack.idl
-html/track/TextTrackCue.idl
-html/track/TextTrackCueList.idl
-html/track/TextTrackList.idl
-html/track/TrackEvent.idl
-html/track/VTTCue.idl
-html/track/VTTRegion.idl
-html/track/VTTRegionList.idl
-html/track/VideoTrack.idl
-html/track/VideoTrackList.idl
-
 inspector/CommandLineAPIHost.idl
 inspector/InspectorFrontendHost.idl
 
@@ -630,9 +575,6 @@
 page/PerformanceResourceTiming.idl
 page/PerformanceTiming.idl
 page/Screen.idl
-page/UserMessageHandler.idl
-page/UserMessageHandlersNamespace.idl
-page/WebKitNamespace.idl
 page/WebKitPoint.idl
 page/WindowEventHandlers.idl
 page/WindowOrWorkerGlobalScope.idl
@@ -3165,6 +3107,8 @@
 html/track/TextTrackList.idl
 html/track/TrackEvent.idl
 html/track/VTTCue.idl
+html/track/VTTRegion.idl
+html/track/VTTRegionList.idl
 html/track/VideoTrack.idl
 html/track/VideoTrackList.idl
 )


Modified: trunk/Source/WebCore/ChangeLog (205456 => 205457)

--- trunk/Source/WebCore/ChangeLog	2016-09-05 17:19:44 UTC (rev 205456)
+++ trunk/Source/WebCore/ChangeLog	2016-09-05 17:21:33 UTC (rev 205457)
@@ -1,3 +1,13 @@
+2016-09-05  Fujii Hironori  
+
+[CMake] Duplicated IDL files in WebCore_IDL_FILES
+https://bugs.webkit.org/show_bug.cgi?id=161592
+
+Reviewed by Darin Adler.
+
+* CMakeLists.txt: Removed duplicated IDL files.
+* DerivedSources.cpp: Added #if ENABLE(USER_MESSAGE_HANDLERS)
+
 2016-09-05  Joseph Pecoraro  
 
 [Mac] Static Analyzer warnings about unused values in CDMSessionAVStreamSession.mm


Modified: trunk/Source/WebCore/DerivedSources.cpp (205456 => 205457)

--- trunk/Source/WebCore/DerivedSources.cpp	2016-09-05 17:19:44 UTC (rev 205456)
+++ trunk/Source/WebCore/DerivedSources.cpp	2016-09-05 17:21:33 UTC (rev 205457)
@@ -630,8 +630,11 @@
 #include "JSUIEvent.cpp"
 #include "JSUIRequestEvent.cpp"
 #include "JSURLUtils.cpp"
+#if ENABLE(USER_MESSAGE_HANDLERS)
 #include "JSUserMessageHandler.cpp"
 

[webkit-changes] [205456] trunk/Source/WebCore

[commit-queue]

Title: [205456] trunk/Source/WebCore








Revision 205456
Author commit-qu...@webkit.org
Date 2016-09-05 10:19:44 -0700 (Mon, 05 Sep 2016)


Log Message
[Mac] Static Analyzer warnings about unused values in CDMSessionAVStreamSession.mm
https://bugs.webkit.org/show_bug.cgi?id=161598

Patch by Joseph Pecoraro  on 2016-09-05
Reviewed by Darin Adler.

* platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h:
* platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.mm:
(WebCore::CDMSessionAVStreamSession::generateKeyReleaseMessage):
Make this parameter a reference again, accidentally lost in r195410.
Now assignments to the parameter make sense.

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h
trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.mm




Diff

Modified: trunk/Source/WebCore/ChangeLog (205455 => 205456)

--- trunk/Source/WebCore/ChangeLog	2016-09-05 16:58:48 UTC (rev 205455)
+++ trunk/Source/WebCore/ChangeLog	2016-09-05 17:19:44 UTC (rev 205456)
@@ -1,3 +1,16 @@
+2016-09-05  Joseph Pecoraro  
+
+[Mac] Static Analyzer warnings about unused values in CDMSessionAVStreamSession.mm
+https://bugs.webkit.org/show_bug.cgi?id=161598
+
+Reviewed by Darin Adler.
+
+* platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h:
+* platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.mm:
+(WebCore::CDMSessionAVStreamSession::generateKeyReleaseMessage):
+Make this parameter a reference again, accidentally lost in r195410.
+Now assignments to the parameter make sense.
+
 2016-09-05  Youenn Fablet  
 
 Clean CSS stylesheets should be accessible from _javascript_


Modified: trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h (205455 => 205456)

--- trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h	2016-09-05 16:58:48 UTC (rev 205455)
+++ trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h	2016-09-05 17:19:44 UTC (rev 205456)
@@ -58,7 +58,7 @@
 void setStreamSession(AVStreamSession*);
 
 protected:
-PassRefPtr generateKeyReleaseMessage(unsigned short& errorCode, uint32_t systemCode);
+PassRefPtr generateKeyReleaseMessage(unsigned short& errorCode, uint32_t& systemCode);
 
 WeakPtrFactory m_weakPtrFactory;
 RetainPtr m_streamSession;


Modified: trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.mm (205455 => 205456)

--- trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.mm	2016-09-05 16:58:48 UTC (rev 205455)
+++ trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.mm	2016-09-05 17:19:44 UTC (rev 205456)
@@ -303,7 +303,7 @@
 [m_streamSession removeStreamDataParser:parser];
 }
 
-PassRefPtr CDMSessionAVStreamSession::generateKeyReleaseMessage(unsigned short& errorCode, uint32_t systemCode)
+PassRefPtr CDMSessionAVStreamSession::generateKeyReleaseMessage(unsigned short& errorCode, uint32_t& systemCode)
 {
 ASSERT(m_mode == KeyRelease);
 m_certificate = m_initData;






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205455] trunk

[commit-queue]

Title: [205455] trunk








Revision 205455
Author commit-qu...@webkit.org
Date 2016-09-05 09:58:48 -0700 (Mon, 05 Sep 2016)


Log Message
Clean CSS stylesheets should be accessible from _javascript_
https://bugs.webkit.org/show_bug.cgi?id=158728

Patch by Youenn Fablet  on 2016-09-05
Reviewed by Darin Adler.

Source/WebCore:

Covered by updated tests.

Making use of the resource response type to evaluate whether to make the CSS stylesheets visible
for stylesheet link elements and for stylesheet processing instructions.
Ideally, the origin flag should be a boolean.

To keep behavior consistent in cors-check-unaware cases, the flag might remain unset.
In that case, the behavior remains the same (checking of the stylesheet URL).
The origin flag is set to true or false only in case of fetch mode being set to cors using the crossorigin attribute.

Updated CSSStyleSheet::create to take a Node reference. Updated callers accordingly.

* contentextensions/ContentExtensionStyleSheet.cpp:
(WebCore::ContentExtensions::ContentExtensionStyleSheet::ContentExtensionStyleSheet): Updated to pass a Node reference and not pointer.
* css/CSSStyleSheet.cpp:
(WebCore::CSSStyleSheet::create): Makes origin flag undefined if boolean parameter is not passed.
Otherwise origin flag is set according given boolean value.
(WebCore::CSSStyleSheet::createInline): Makes origin flag set to true.
(WebCore::CSSStyleSheet::CSSStyleSheet): Adding origin flag initialization.
(WebCore::CSSStyleSheet::canAccessRules): Returning according the origin flag if set.
Returning as before in case the flag is not set.
* css/CSSStyleSheet.h: Adding origin clean flag.
* dom/ExtensionStyleSheets.cpp:
(WebCore::ExtensionStyleSheets::addUserStyleSheet):
(WebCore::ExtensionStyleSheets::addAuthorStyleSheetForTesting):
(WebCore::ExtensionStyleSheets::maybeAddContentExtensionSheet):
* html/HTMLLinkElement.cpp:
(WebCore::HTMLLinkElement::initializeStyleSheet): Helper routine used in setCSSStyleSheet.
Sets origin clean flag if the resource is clean.
(WebCore::HTMLLinkElement::setCSSStyleSheet): Making use of initializeStyleSheet.
* html/HTMLLinkElement.h:

LayoutTests:

* http/tests/security/cannot-read-cssrules-expected.txt:
* http/tests/security/cannot-read-cssrules-redirect-expected.txt:
* http/tests/security/cannot-read-cssrules-redirect.html:
* http/tests/security/cannot-read-cssrules.html:
* http/tests/security/cross-origin-css-9.html:
* http/tests/security/resources/xorigincss1-allow-star.php: Added.

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/http/tests/security/cannot-read-cssrules-expected.txt
trunk/LayoutTests/http/tests/security/cannot-read-cssrules-redirect-expected.txt
trunk/LayoutTests/http/tests/security/cannot-read-cssrules-redirect.html
trunk/LayoutTests/http/tests/security/cannot-read-cssrules.html
trunk/LayoutTests/http/tests/security/cross-origin-css-9.html
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/contentextensions/ContentExtensionStyleSheet.cpp
trunk/Source/WebCore/css/CSSStyleSheet.cpp
trunk/Source/WebCore/css/CSSStyleSheet.h
trunk/Source/WebCore/dom/ExtensionStyleSheets.cpp
trunk/Source/WebCore/dom/ProcessingInstruction.cpp
trunk/Source/WebCore/html/HTMLLinkElement.cpp
trunk/Source/WebCore/html/HTMLLinkElement.h


Added Paths

trunk/LayoutTests/http/tests/security/resources/xorigincss1-allow-star.php




Diff

Modified: trunk/LayoutTests/ChangeLog (205454 => 205455)

--- trunk/LayoutTests/ChangeLog	2016-09-05 16:42:23 UTC (rev 205454)
+++ trunk/LayoutTests/ChangeLog	2016-09-05 16:58:48 UTC (rev 205455)
@@ -1,3 +1,17 @@
+2016-09-05  Youenn Fablet  
+
+Clean CSS stylesheets should be accessible from _javascript_
+https://bugs.webkit.org/show_bug.cgi?id=158728
+
+Reviewed by Darin Adler.
+
+* http/tests/security/cannot-read-cssrules-expected.txt:
+* http/tests/security/cannot-read-cssrules-redirect-expected.txt:
+* http/tests/security/cannot-read-cssrules-redirect.html:
+* http/tests/security/cannot-read-cssrules.html:
+* http/tests/security/cross-origin-css-9.html:
+* http/tests/security/resources/xorigincss1-allow-star.php: Added.
+
 2016-09-05  Philippe Normand  
 
 WebRTC: Several media stream tests crashing


Modified: trunk/LayoutTests/http/tests/security/cannot-read-cssrules-expected.txt (205454 => 205455)

--- trunk/LayoutTests/http/tests/security/cannot-read-cssrules-expected.txt	2016-09-05 16:42:23 UTC (rev 205454)
+++ trunk/LayoutTests/http/tests/security/cannot-read-cssrules-expected.txt	2016-09-05 16:58:48 UTC (rev 205455)
@@ -1,10 +1,21 @@
+CONSOLE MESSAGE: Origin http://127.0.0.1:8000 is not allowed by Access-Control-Allow-Origin.
 This test whether a script can read the rules from a cross-origin style sheet. For more information on why we block this, please see https://bugs.webkit.org/show_bug.cgi?id=20527.
 
 Test begins.
-== Cross-Origin ==
+== Cross-Origin 

[webkit-changes] [205454] trunk/Tools

[mcatanzaro]

Title: [205454] trunk/Tools








Revision 205454
Author mcatanz...@igalia.com
Date 2016-09-05 09:42:23 -0700 (Mon, 05 Sep 2016)


Log Message
[GTK] Skip several failing API tests

Unfortunately we don't have any form of test expectations for API tests, so we'll just skip
them after bugs are filed.

* Scripts/run-gtk-tests:
(TestRunner):

Modified Paths

trunk/Tools/ChangeLog
trunk/Tools/Scripts/run-gtk-tests




Diff

Modified: trunk/Tools/ChangeLog (205453 => 205454)

--- trunk/Tools/ChangeLog	2016-09-05 14:34:09 UTC (rev 205453)
+++ trunk/Tools/ChangeLog	2016-09-05 16:42:23 UTC (rev 205454)
@@ -1,3 +1,13 @@
+2016-09-05  Michael Catanzaro  
+
+[GTK] Skip several failing API tests
+
+Unfortunately we don't have any form of test expectations for API tests, so we'll just skip
+them after bugs are filed.
+
+* Scripts/run-gtk-tests:
+(TestRunner):
+
 2016-09-05  Youenn Fablet  
 
 Fix W3C test importer unit tests after r205447


Modified: trunk/Tools/Scripts/run-gtk-tests (205453 => 205454)

--- trunk/Tools/Scripts/run-gtk-tests	2016-09-05 14:34:09 UTC (rev 205453)
+++ trunk/Tools/Scripts/run-gtk-tests	2016-09-05 16:42:23 UTC (rev 205454)
@@ -72,7 +72,13 @@
 SkippedTest("WebKit2Gtk/TestUIClient", "/webkit2/WebKitWebView/audio-usermedia-permission-request", "Test times out", 158257),
 SkippedTest("WebKit2Gtk/TestCookieManager", "/webkit2/WebKitCookieManager/persistent-storage", "Test is flaky", 134580),
 SkippedTest("WebKit2Gtk/TestWebViewEditor", "/webkit2/WebKitWebView/editable/editable", "Test hits an assertion in Debug builds", 151654, "Debug"),
+SkippedTest("WebKit2Gtk/TestWebViewEditor", "/webkit2/WebKitWebView/editable/editable", "Same test, hits release assertion too", 161608),
+SkippedTest("WebKit2Gtk/TestWebViewEditor", "/webkit2/WebKitWebView/cut-copy-paste/non-editable", "Test hits assertion", 161608),
+SkippedTest("WebKit2Gtk/TestWebViewEditor", "/webkit2/WebKitWebView/cut-copy-paste/editable", "Test hits assertion", 161608),
+SkippedTest("WebKit2Gtk/TestWebViewEditor", "/webkit2/WebKitWebView/select-all/non-editable", "Test hits assertion", 161608),
+SkippedTest("WebKit2Gtk/TestWebViewEditor", "/webkit2/WebKitWebView/select-all/editable", "Test hits assertion", 161608),
 SkippedTest("WebKit2Gtk/TestWebExtensions", "/webkit2/WebKitWebView/install-missing-plugins-permission-request", "Test times out", 147822),
+SkippedTest("WebKit2Gtk/TestWebKitWebContext", "/webkit2/WebKitSecurityManager/file-xhr", "Test hits assertion", 161607),
 SkippedTest("WebKit2/TestWebKit2", "WebKit2.MouseMoveAfterCrash", "Test is flaky", 85066),
 SkippedTest("WebKit2/TestWebKit2", "WebKit2.NewFirstVisuallyNonEmptyLayoutForImages", "Test is flaky", 85066),
 SkippedTest("WebKit2/TestWebKit2", "WebKit2.NewFirstVisuallyNonEmptyLayoutFrames", "Test fails", 85037),






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205453] trunk/Tools

[commit-queue]

Title: [205453] trunk/Tools








Revision 205453
Author commit-qu...@webkit.org
Date 2016-09-05 07:34:09 -0700 (Mon, 05 Sep 2016)


Log Message
Fix W3C test importer unit tests after r205447
https://bugs.webkit.org/show_bug.cgi?id=161604

Unreviewed.

Patch by Youenn Fablet  on 2016-09-05

* Scripts/webkitpy/w3c/test_importer.py:
(TestImporter.__init__): Skip cleaning test_resource_files files if resource-files.json cannot be read.

Modified Paths

trunk/Tools/ChangeLog
trunk/Tools/Scripts/webkitpy/w3c/test_importer.py




Diff

Modified: trunk/Tools/ChangeLog (205452 => 205453)

--- trunk/Tools/ChangeLog	2016-09-05 14:22:07 UTC (rev 205452)
+++ trunk/Tools/ChangeLog	2016-09-05 14:34:09 UTC (rev 205453)
@@ -1,3 +1,13 @@
+2016-09-05  Youenn Fablet  
+
+Fix W3C test importer unit tests after r205447
+https://bugs.webkit.org/show_bug.cgi?id=161604
+
+Unreviewed.
+
+* Scripts/webkitpy/w3c/test_importer.py:
+(TestImporter.__init__): Skip cleaning test_resource_files files if resource-files.json cannot be read.
+
 2016-09-05  Philippe Normand  
 
 WebRTC: Several media stream tests crashing


Modified: trunk/Tools/Scripts/webkitpy/w3c/test_importer.py (205452 => 205453)

--- trunk/Tools/Scripts/webkitpy/w3c/test_importer.py	2016-09-05 14:22:07 UTC (rev 205452)
+++ trunk/Tools/Scripts/webkitpy/w3c/test_importer.py	2016-09-05 14:34:09 UTC (rev 205453)
@@ -180,7 +180,7 @@
 
 self._test_resource_files_json_path = self.filesystem.join(self.layout_tests_w3c_path, "resources", "resource-files.json")
 self._test_resource_files = json.loads(self.filesystem.read_text_file(self._test_resource_files_json_path)) if self.filesystem.exists(self._test_resource_files_json_path) else None
-if self.options.clean_destination_directory:
+if self.options.clean_destination_directory and self._test_resource_files:
 self._test_resource_files["files"] = []
 
 def do_import(self):






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205452] trunk/Source/WebKit2

[gns]

Title: [205452] trunk/Source/WebKit2








Revision 205452
Author g...@gnome.org
Date 2016-09-05 07:22:07 -0700 (Mon, 05 Sep 2016)


Log Message
[GTK] GL_PACK_ROW_LENGTH is not available in GLES2
https://bugs.webkit.org/show_bug.cgi?id=161484

Reviewed by Carlos Garcia Campos.

* UIProcess/gtk/AcceleratedBackingStoreWayland.cpp:
(WebKit::AcceleratedBackingStoreWayland::paint): when under GLES2 we cannot rely on
GL_PACK_ROW_LENGTH; use glReadPixel directly when stride matches width, read line
by line manually otherwise. Colour conversion is also required to get the data out
correctly.

Modified Paths

trunk/Source/WebKit2/ChangeLog
trunk/Source/WebKit2/UIProcess/gtk/AcceleratedBackingStoreWayland.cpp




Diff

Modified: trunk/Source/WebKit2/ChangeLog (205451 => 205452)

--- trunk/Source/WebKit2/ChangeLog	2016-09-05 13:54:25 UTC (rev 205451)
+++ trunk/Source/WebKit2/ChangeLog	2016-09-05 14:22:07 UTC (rev 205452)
@@ -1,3 +1,16 @@
+2016-09-05  Gustavo Noronha Silva  
+
+[GTK] GL_PACK_ROW_LENGTH is not available in GLES2
+https://bugs.webkit.org/show_bug.cgi?id=161484
+
+Reviewed by Carlos Garcia Campos.
+
+* UIProcess/gtk/AcceleratedBackingStoreWayland.cpp:
+(WebKit::AcceleratedBackingStoreWayland::paint): when under GLES2 we cannot rely on
+GL_PACK_ROW_LENGTH; use glReadPixel directly when stride matches width, read line
+by line manually otherwise. Colour conversion is also required to get the data out
+correctly.
+
 2016-09-05  Carlos Garcia Campos  
 
 [GTK] SetNativeSurfaceHandleForCompositing should not be compiled in for wayland build


Modified: trunk/Source/WebKit2/UIProcess/gtk/AcceleratedBackingStoreWayland.cpp (205451 => 205452)

--- trunk/Source/WebKit2/UIProcess/gtk/AcceleratedBackingStoreWayland.cpp	2016-09-05 13:54:25 UTC (rev 205451)
+++ trunk/Source/WebKit2/UIProcess/gtk/AcceleratedBackingStoreWayland.cpp	2016-09-05 14:22:07 UTC (rev 205452)
@@ -89,13 +89,28 @@
 glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_2D, texture, 0);
 
 glPixelStorei(GL_PACK_ALIGNMENT, 4);
-glPixelStorei(GL_PACK_ROW_LENGTH, cairo_image_surface_get_stride(m_surface.get()) / 4);
+
 #if USE(OPENGL_ES_2)
-glReadPixels(0, 0, textureSize.width(), textureSize.height(), GL_RGBA, GL_UNSIGNED_BYTE, cairo_image_surface_get_data(m_surface.get()));
+unsigned char* data = ""
+if (cairo_image_surface_get_stride(m_surface.get()) == textureSize.width() * 4)
+glReadPixels(0, 0, textureSize.width(), textureSize.height(), GL_RGBA, GL_UNSIGNED_BYTE, data);
+else {
+int strideBytes = cairo_image_surface_get_stride(m_surface.get());
+for (int i = 0; i < textureSize.height(); i++) {
+unsigned char* dataOffset = data + i * strideBytes;
+glReadPixels(0, i, textureSize.width(), 1, GL_RGBA, GL_UNSIGNED_BYTE, dataOffset);
+}
+}
+
+// Convert to BGRA.
+int totalBytes = size.width() * size.height() * 4;
+for (int i = 0; i < totalBytes; i += 4)
+std::swap(data[i], data[i + 2]);
 #else
+glPixelStorei(GL_PACK_ROW_LENGTH, cairo_image_surface_get_stride(m_surface.get()) / 4);
 glReadPixels(0, 0, textureSize.width(), textureSize.height(), GL_BGRA, GL_UNSIGNED_INT_8_8_8_8_REV, cairo_image_surface_get_data(m_surface.get()));
+glPixelStorei(GL_PACK_ROW_LENGTH, 0);
 #endif
-glPixelStorei(GL_PACK_ROW_LENGTH, 0);
 
 glBindFramebuffer(GL_FRAMEBUFFER, 0);
 glDeleteFramebuffers(1, );






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205451] trunk

[philn]

Title: [205451] trunk








Revision 205451
Author ph...@webkit.org
Date 2016-09-05 06:54:25 -0700 (Mon, 05 Sep 2016)


Log Message
WebRTC: Several media stream tests crashing
https://bugs.webkit.org/show_bug.cgi?id=161582

Reviewed by Alejandro G. Castro.

Tools:

* gtk/jhbuild.modules: Bump to latest OpenWebRTC master fixing the crash on missing
glvideoflip GStreamer element.

LayoutTests:

* platform/gtk/TestExpectations: Unskip now-passing tests.

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/gtk/TestExpectations
trunk/Tools/ChangeLog
trunk/Tools/gtk/jhbuild.modules




Diff

Modified: trunk/LayoutTests/ChangeLog (205450 => 205451)

--- trunk/LayoutTests/ChangeLog	2016-09-05 11:02:37 UTC (rev 205450)
+++ trunk/LayoutTests/ChangeLog	2016-09-05 13:54:25 UTC (rev 205451)
@@ -1,3 +1,12 @@
+2016-09-05  Philippe Normand  
+
+WebRTC: Several media stream tests crashing
+https://bugs.webkit.org/show_bug.cgi?id=161582
+
+Reviewed by Alejandro G. Castro.
+
+* platform/gtk/TestExpectations: Unskip now-passing tests.
+
 2016-09-05  Youenn Fablet  
 
 CachedResourceLoader is not taking into account fetch options to use or not cached resources


Modified: trunk/LayoutTests/platform/gtk/TestExpectations (205450 => 205451)

--- trunk/LayoutTests/platform/gtk/TestExpectations	2016-09-05 11:02:37 UTC (rev 205450)
+++ trunk/LayoutTests/platform/gtk/TestExpectations	2016-09-05 13:54:25 UTC (rev 205451)
@@ -823,11 +823,6 @@
 
 webkit.org/b/153937 http/tests/misc/detach-during-notifyDone.html [ Crash Pass ]
 
-webkit.org/b/161582 fast/mediastream/apply-constraints-video.html [ Crash ]
-webkit.org/b/161582 fast/mediastream/MediaStream-video-element-displays-buffer.html [ Crash ]
-webkit.org/b/161582 fast/mediastream/MediaStream-video-element-video-tracks-disabled-then-enabled.html [ Crash ]
-webkit.org/b/161582 fast/mediastream/MediaStream-video-element-video-tracks-disabled.html [ Crash ]
-
 #
 # End of Crashing tests
 #


Modified: trunk/Tools/ChangeLog (205450 => 205451)

--- trunk/Tools/ChangeLog	2016-09-05 11:02:37 UTC (rev 205450)
+++ trunk/Tools/ChangeLog	2016-09-05 13:54:25 UTC (rev 205451)
@@ -1,3 +1,13 @@
+2016-09-05  Philippe Normand  
+
+WebRTC: Several media stream tests crashing
+https://bugs.webkit.org/show_bug.cgi?id=161582
+
+Reviewed by Alejandro G. Castro.
+
+* gtk/jhbuild.modules: Bump to latest OpenWebRTC master fixing the crash on missing
+glvideoflip GStreamer element.
+
 2016-09-05  Youenn Fablet  
 
 W3C test importer should generate the list of resource files


Modified: trunk/Tools/gtk/jhbuild.modules (205450 => 205451)

--- trunk/Tools/gtk/jhbuild.modules	2016-09-05 11:02:37 UTC (rev 205450)
+++ trunk/Tools/gtk/jhbuild.modules	2016-09-05 13:54:25 UTC (rev 205451)
@@ -514,7 +514,7 @@
   
   
  
-
+
   
 
   





___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205449] releases/WebKitGTK/webkit-2.12.5/

[carlosgc]

Title: [205449] releases/WebKitGTK/webkit-2.12.5/








Revision 205449
Author carlo...@webkit.org
Date 2016-09-05 03:31:44 -0700 (Mon, 05 Sep 2016)


Log Message
WebKitGTK+ 2.12.5

Added Paths

releases/WebKitGTK/webkit-2.12.5/




Diff




___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205448] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205448] releases/WebKitGTK/webkit-2.12








Revision 205448
Author carlo...@webkit.org
Date 2016-09-05 03:31:14 -0700 (Mon, 05 Sep 2016)


Log Message
Unreviewed. Update OptionsGTK.cmake and NEWS for 2.12.5 release.

.:

* Source/cmake/OptionsGTK.cmake: Bump version numbers.

Source/WebKit2:

* gtk/NEWS: Add release notes for 2.12.5.

Modified Paths

releases/WebKitGTK/webkit-2.12/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebKit2/gtk/NEWS
releases/WebKitGTK/webkit-2.12/Source/cmake/OptionsGTK.cmake




Diff

Modified: releases/WebKitGTK/webkit-2.12/ChangeLog (205447 => 205448)

--- releases/WebKitGTK/webkit-2.12/ChangeLog	2016-09-05 10:30:07 UTC (rev 205447)
+++ releases/WebKitGTK/webkit-2.12/ChangeLog	2016-09-05 10:31:14 UTC (rev 205448)
@@ -1,3 +1,9 @@
+2016-09-05  Carlos Garcia Campos  
+
+Unreviewed. Update OptionsGTK.cmake and NEWS for 2.12.5 release.
+
+* Source/cmake/OptionsGTK.cmake: Bump version numbers.
+
 2016-08-24  Carlos Garcia Campos  
 
 Unreviewed. Update OptionsGTK.cmake and NEWS for 2.12.4 release.


Modified: releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog (205447 => 205448)

--- releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog	2016-09-05 10:30:07 UTC (rev 205447)
+++ releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog	2016-09-05 10:31:14 UTC (rev 205448)
@@ -1,3 +1,9 @@
+2016-09-05  Carlos Garcia Campos  
+
+Unreviewed. Update OptionsGTK.cmake and NEWS for 2.12.5 release.
+
+* gtk/NEWS: Add release notes for 2.12.5.
+
 2016-08-30  Carlos Garcia Campos  
 
 REGRESSION(r194846): [GTK] UI process crash visiting sites protected with HTTP auth when using GTK+ < 3.14


Modified: releases/WebKitGTK/webkit-2.12/Source/WebKit2/gtk/NEWS (205447 => 205448)

--- releases/WebKitGTK/webkit-2.12/Source/WebKit2/gtk/NEWS	2016-09-05 10:30:07 UTC (rev 205447)
+++ releases/WebKitGTK/webkit-2.12/Source/WebKit2/gtk/NEWS	2016-09-05 10:31:14 UTC (rev 205448)
@@ -1,4 +1,13 @@
 ==
+WebKitGTK+  2.12.5
+==
+
+What's new in WebKitGTK+ 2.12.5?
+
+  - Fix a regression introduced in 2.12.4 that caused a hang in the network process after a load failure.
+  - Fix several crashes and rendering issues.
+
+==
 WebKitGTK+  2.12.4
 ==
 


Modified: releases/WebKitGTK/webkit-2.12/Source/cmake/OptionsGTK.cmake (205447 => 205448)

--- releases/WebKitGTK/webkit-2.12/Source/cmake/OptionsGTK.cmake	2016-09-05 10:30:07 UTC (rev 205447)
+++ releases/WebKitGTK/webkit-2.12/Source/cmake/OptionsGTK.cmake	2016-09-05 10:31:14 UTC (rev 205448)
@@ -2,7 +2,7 @@
 
 set(PROJECT_VERSION_MAJOR 2)
 set(PROJECT_VERSION_MINOR 12)
-set(PROJECT_VERSION_MICRO 4)
+set(PROJECT_VERSION_MICRO 5)
 set(PROJECT_VERSION ${PROJECT_VERSION_MAJOR}.${PROJECT_VERSION_MINOR}.${PROJECT_VERSION_MICRO})
 set(WEBKITGTK_API_VERSION 4.0)
 
@@ -15,8 +15,8 @@
 
 # Libtool library version, not to be confused with API version.
 # See http://www.gnu.org/software/libtool/manual/html_node/Libtool-versioning.html
-CALCULATE_LIBRARY_VERSIONS_FROM_LIBTOOL_TRIPLE(WEBKIT2 50 10 13)
-CALCULATE_LIBRARY_VERSIONS_FROM_LIBTOOL_TRIPLE(_javascript_CORE 21 12 3)
+CALCULATE_LIBRARY_VERSIONS_FROM_LIBTOOL_TRIPLE(WEBKIT2 50 11 13)
+CALCULATE_LIBRARY_VERSIONS_FROM_LIBTOOL_TRIPLE(_javascript_CORE 21 13 3)
 
 # These are shared variables, but we special case their definition so that we can use the
 # CMAKE_INSTALL_* variables that are populated by the GNUInstallDirs macro.






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205447] trunk

[commit-queue]

Title: [205447] trunk








Revision 205447
Author commit-qu...@webkit.org
Date 2016-09-05 03:30:07 -0700 (Mon, 05 Sep 2016)


Log Message
W3C test importer should generate the list of resource files
https://bugs.webkit.org/show_bug.cgi?id=161578

Patch by Youenn Fablet  on 2016-09-05
Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

* resources/resource-files.json: Reformatting the file.

Tools:

Generating LayoutTests/imported/w3c/resources/resource-files.json when importing tests.

* Scripts/webkitpy/w3c/test_importer.py:
(TestImporter.__init__): Reading test resource files configuration file.
(TestImporter.find_importable_tests):
(TestImporter.import_tests): Update the test resource files based on test parsed files.
(TestImporter):
(TestImporter._already_identified_as_resource_file): Helper routine to know whether a file is already set as a resource file.
(TestImporter._is_in_resources_directory: Helper routine to know whether a file is in a resources directory.

Modified Paths

trunk/LayoutTests/imported/w3c/ChangeLog
trunk/LayoutTests/imported/w3c/resources/resource-files.json
trunk/Tools/ChangeLog
trunk/Tools/Scripts/webkitpy/w3c/test_importer.py




Diff

Modified: trunk/LayoutTests/imported/w3c/ChangeLog (205446 => 205447)

--- trunk/LayoutTests/imported/w3c/ChangeLog	2016-09-05 09:53:17 UTC (rev 205446)
+++ trunk/LayoutTests/imported/w3c/ChangeLog	2016-09-05 10:30:07 UTC (rev 205447)
@@ -1,3 +1,12 @@
+2016-09-05  Youenn Fablet  
+
+W3C test importer should generate the list of resource files
+https://bugs.webkit.org/show_bug.cgi?id=161578
+
+Reviewed by Ryosuke Niwa.
+
+* resources/resource-files.json: Reformatting the file.
+
 2016-09-03  Chris Dumez  
 
 Align meta element http-equiv="refresh" parsing with the HTML specification


Modified: trunk/LayoutTests/imported/w3c/resources/resource-files.json (205446 => 205447)

--- trunk/LayoutTests/imported/w3c/resources/resource-files.json	2016-09-05 09:53:17 UTC (rev 205446)
+++ trunk/LayoutTests/imported/w3c/resources/resource-files.json	2016-09-05 10:30:07 UTC (rev 205447)
@@ -1,219 +1,222 @@
-{"directories": [
-"web-platform-tests/common",
-"web-platform-tests/dom/nodes/Document-createElement-namespace-tests",
-"web-platform-tests/fonts",
-"web-platform-tests/html/browsers/browsing-the-web/navigating-across-documents/source/support",
-"web-platform-tests/html/browsers/browsing-the-web/unloading-documents/support",
-"web-platform-tests/html/browsers/history/the-history-interface/non-automated",
-"web-platform-tests/html/browsers/history/the-location-interface/non-automated",
-"web-platform-tests/images",
-"web-platform-tests/service-workers",
-"web-platform-tests/tools"
-], "files": [
-"web-platform-tests/XMLHttpRequest/xmlhttprequest-sync-block-defer-scripts-subframe.html",
-"web-platform-tests/XMLHttpRequest/xmlhttprequest-sync-not-hang-scriptloader-subframe.html",
-"web-platform-tests/dom/events/EventListener-incumbent-global-subframe-1.sub.html",
-"web-platform-tests/dom/events/EventListener-incumbent-global-subframe-2.sub.html",
-"web-platform-tests/dom/events/EventListener-incumbent-global-subsubframe.sub.html",
-"web-platform-tests/dom/nodes/Element-getElementsByTagName-change-document-HTMLNess-iframe.xml",
-"web-platform-tests/dom/nodes/Node-isEqualNode-iframe1.xml",
-"web-platform-tests/dom/nodes/Node-isEqualNode-iframe2.xml",
-"web-platform-tests/dom/nodes/Node-parentNode-iframe.html",
-"web-platform-tests/dom/nodes/ParentNode-querySelector-All-content.html",
-"web-platform-tests/dom/nodes/ParentNode-querySelector-All-content.xht",
-"web-platform-tests/dom/nodes/getElementsByClassNameFrame.htm",
-"web-platform-tests/dom/ranges/Range-test-iframe.html",
-"web-platform-tests/html/browsers/browsing-the-web/history-traversal/001-1.html",
-"web-platform-tests/html/browsers/browsing-the-web/history-traversal/001-2.html",
-"web-platform-tests/html/browsers/browsing-the-web/history-traversal/browsing_context_name-1.html",
-"web-platform-tests/html/browsers/browsing-the-web/history-traversal/browsing_context_name-2.html",
-"web-platform-tests/html/browsers/browsing-the-web/history-traversal/browsing_context_name-3.html",
-"web-platform-tests/html/browsers/browsing-the-web/history-traversal/browsing_context_name-4.html",
-"web-platform-tests/html/browsers/browsing-the-web/history-traversal/unset_context_name.html",
-"web-platform-tests/html/browsers/browsing-the-web/navigating-across-documents/001-1.html",
-"web-platform-tests/html/browsers/browsing-the-web/navigating-across-documents/001-2.html",
-"web-platform-tests/html/browsers/browsing-the-web/navigating-across-documents/001-3.html",
-"web-platform-tests/html/browsers/browsing-the-web/navigating-across-documents/002-1.html",
-"web-platform-tests/html/browsers/browsing-the-web/navigating-across-documents/002-2.html",

[webkit-changes] [205446] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205446] releases/WebKitGTK/webkit-2.12








Revision 205446
Author carlo...@webkit.org
Date 2016-09-05 02:53:17 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r204699 - [DFG] Should not fixup AnyIntUse in 32_64
https://bugs.webkit.org/show_bug.cgi?id=161029

Reviewed by Saam Barati.

JSTests:

* typeProfiler/int52-dfg.js: Added.
(test):

Source/_javascript_Core:

DFG fixup phase uses AnyIntUse even in 32bit DFG. This patch removes this incorrect filtering.
If the 32bit DFG see the TypeAnyInt, it should fallback to the NumberUse case.

* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):

Modified Paths

releases/WebKitGTK/webkit-2.12/Source/_javascript_Core/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/_javascript_Core/dfg/DFGFixupPhase.cpp


Added Paths

releases/WebKitGTK/webkit-2.12/JSTests/
releases/WebKitGTK/webkit-2.12/JSTests/typeProfiler/
releases/WebKitGTK/webkit-2.12/JSTests/typeProfiler/int52-dfg.js




Diff

Added: releases/WebKitGTK/webkit-2.12/JSTests/typeProfiler/int52-dfg.js (0 => 205446)

--- releases/WebKitGTK/webkit-2.12/JSTests/typeProfiler/int52-dfg.js	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/JSTests/typeProfiler/int52-dfg.js	2016-09-05 09:53:17 UTC (rev 205446)
@@ -0,0 +1,17 @@
+load("./driver/driver.js");
+
+function test()
+{
+var ok = 0;
+for (var i = 0; i < 1e4; ++i) {
+// Int52. ProfileType should not use AnyIntUse edge in 32bit environment.
+// If 32bit uses AnyIntUse, it leads crashing.
+ok += 0xf;
+}
+return ok;
+}
+test();
+
+var types = findTypeForExpression(test, "ok += 0x");
+assert(types.instructionTypeSet.primitiveTypeNames.length === 1, "Primitive type names should one candidate.");
+assert(types.instructionTypeSet.primitiveTypeNames.indexOf(T.Integer) !== -1, "Primitive type names should contain 'Integer'");


Modified: releases/WebKitGTK/webkit-2.12/Source/_javascript_Core/ChangeLog (205445 => 205446)

--- releases/WebKitGTK/webkit-2.12/Source/_javascript_Core/ChangeLog	2016-09-05 09:35:33 UTC (rev 205445)
+++ releases/WebKitGTK/webkit-2.12/Source/_javascript_Core/ChangeLog	2016-09-05 09:53:17 UTC (rev 205446)
@@ -1,3 +1,16 @@
+2016-08-21  Yusuke Suzuki  
+
+[DFG] Should not fixup AnyIntUse in 32_64
+https://bugs.webkit.org/show_bug.cgi?id=161029
+
+Reviewed by Saam Barati.
+
+DFG fixup phase uses AnyIntUse even in 32bit DFG. This patch removes this incorrect filtering.
+If the 32bit DFG see the TypeAnyInt, it should fallback to the NumberUse case.
+
+* dfg/DFGFixupPhase.cpp:
+(JSC::DFG::FixupPhase::fixupNode):
+
 2016-05-03  Michael Saboff  
 
 Crash: Array.prototype.slice() and .splice() can call fastSlice() after an array is truncated


Modified: releases/WebKitGTK/webkit-2.12/Source/_javascript_Core/dfg/DFGFixupPhase.cpp (205445 => 205446)

--- releases/WebKitGTK/webkit-2.12/Source/_javascript_Core/dfg/DFGFixupPhase.cpp	2016-09-05 09:35:33 UTC (rev 205445)
+++ releases/WebKitGTK/webkit-2.12/Source/_javascript_Core/dfg/DFGFixupPhase.cpp	2016-09-05 09:53:17 UTC (rev 205446)
@@ -1369,12 +1369,22 @@
 RefPtr typeSet = node->typeLocation()->m_instructionTypeSet;
 RuntimeTypeMask seenTypes = typeSet->seenTypes();
 if (typeSet->doesTypeConformTo(TypeMachineInt)) {
-if (node->child1()->shouldSpeculateInt32())
+if (node->child1()->shouldSpeculateInt32()) {
 fixEdge(node->child1());
-else
+node->remove();
+break;
+}
+
+if (enableInt52()) {
 fixEdge(node->child1());
-node->remove();
-} else if (typeSet->doesTypeConformTo(TypeNumber | TypeMachineInt)) {
+node->remove();
+break;
+}
+
+// Must not perform fixEdge here since the type set only includes TypeMachineInt. Double values should be logged.
+}
+
+if (typeSet->doesTypeConformTo(TypeNumber | TypeMachineInt)) {
 fixEdge(node->child1());
 node->remove();
 } else if (typeSet->doesTypeConformTo(TypeString)) {






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205445] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205445] releases/WebKitGTK/webkit-2.12








Revision 205445
Author carlo...@webkit.org
Date 2016-09-05 02:35:33 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r205392 - Should never be reached failure in WebCore::floatValueForLength
https://bugs.webkit.org/show_bug.cgi?id=139397


Reviewed by Simon Fraser.

Source/WebCore:

floatValueForLength can't resolve unspecified Length types. Filter them out and return 0 as if they were auto.

Test: svg/css/assert-on-non-resolvable-dimension.html

* svg/SVGLengthContext.cpp:
(WebCore::SVGLengthContext::valueForLength):

LayoutTests:

* svg/css/assert-on-non-resolvable-dimension-expected.txt: Added.
* svg/css/assert-on-non-resolvable-dimension.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/svg/SVGLengthContext.cpp


Added Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/svg/css/assert-on-non-resolvable-dimension-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/svg/css/assert-on-non-resolvable-dimension.html




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205444 => 205445)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:34:19 UTC (rev 205444)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:35:33 UTC (rev 205445)
@@ -1,5 +1,16 @@
 2016-09-02  Zalan Bujtas  
 
+Should never be reached failure in WebCore::floatValueForLength
+https://bugs.webkit.org/show_bug.cgi?id=139397
+
+
+Reviewed by Simon Fraser.
+
+* svg/css/assert-on-non-resolvable-dimension-expected.txt: Added.
+* svg/css/assert-on-non-resolvable-dimension.html: Added.
+
+2016-09-02  Zalan Bujtas  
+
 ASSERT_NOT_REACHED() is touched in WebCore::valueForLength
 https://bugs.webkit.org/show_bug.cgi?id=123337
 


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/svg/css/assert-on-non-resolvable-dimension-expected.txt (0 => 205445)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/svg/css/assert-on-non-resolvable-dimension-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/svg/css/assert-on-non-resolvable-dimension-expected.txt	2016-09-05 09:35:33 UTC (rev 205445)
@@ -0,0 +1 @@
+PASS if no assert in debug. 


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/svg/css/assert-on-non-resolvable-dimension.html (0 => 205445)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/svg/css/assert-on-non-resolvable-dimension.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/svg/css/assert-on-non-resolvable-dimension.html	2016-09-05 09:35:33 UTC (rev 205445)
@@ -0,0 +1,23 @@
+
+
+
+This tests that we don't assert on non-resolvable values.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+
+
+PASS if no assert in debug.
+
+
+foobar
+foobar
+foobar
+foobar
+foobar
+foobar
+foobar
+
+
+


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog (205444 => 205445)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:34:19 UTC (rev 205444)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:35:33 UTC (rev 205445)
@@ -1,5 +1,20 @@
 2016-09-02  Zalan Bujtas  
 
+Should never be reached failure in WebCore::floatValueForLength
+https://bugs.webkit.org/show_bug.cgi?id=139397
+
+
+Reviewed by Simon Fraser.
+
+floatValueForLength can't resolve unspecified Length types. Filter them out and return 0 as if they were auto.
+
+Test: svg/css/assert-on-non-resolvable-dimension.html
+
+* svg/SVGLengthContext.cpp:
+(WebCore::SVGLengthContext::valueForLength):
+
+2016-09-02  Zalan Bujtas  
+
 ASSERT_NOT_REACHED() is touched in WebCore::valueForLength
 https://bugs.webkit.org/show_bug.cgi?id=123337
 


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/svg/SVGLengthContext.cpp (205444 => 205445)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/svg/SVGLengthContext.cpp	2016-09-05 09:34:19 UTC (rev 205444)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/svg/SVGLengthContext.cpp	2016-09-05 09:35:33 UTC (rev 205445)
@@ -91,7 +91,7 @@
 {
 if (length.isPercent())
 return convertValueFromPercentageToUserUnits(length.value() / 100, mode, IGNORE_EXCEPTION);
-if (length.isAuto())
+if (length.isAuto() || !length.isSpecified())
 return 0;
 
 FloatSize viewportSize;






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205444] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205444] releases/WebKitGTK/webkit-2.12








Revision 205444
Author carlo...@webkit.org
Date 2016-09-05 02:34:19 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r205384 - ASSERT_NOT_REACHED() is touched in WebCore::valueForLength
https://bugs.webkit.org/show_bug.cgi?id=123337


Reviewed by Simon Fraser.

Source/WebCore:

Do not try to use unspecified height value while resolving logical height for table row.

Test: fast/table/assert-on-non-resolvable-row-dimension.html

* rendering/RenderTableCell.h:
(WebCore::RenderTableCell::logicalHeightForRowSizing):

LayoutTests:

* fast/table/assert-on-non-resolvable-row-dimension-expected.txt: Added.
* fast/table/assert-on-non-resolvable-row-dimension.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableCell.h


Added Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-non-resolvable-row-dimension-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-non-resolvable-row-dimension.html




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205443 => 205444)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:33:03 UTC (rev 205443)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:34:19 UTC (rev 205444)
@@ -1,5 +1,16 @@
 2016-09-02  Zalan Bujtas  
 
+ASSERT_NOT_REACHED() is touched in WebCore::valueForLength
+https://bugs.webkit.org/show_bug.cgi?id=123337
+
+
+Reviewed by Simon Fraser.
+
+* fast/table/assert-on-non-resolvable-row-dimension-expected.txt: Added.
+* fast/table/assert-on-non-resolvable-row-dimension.html: Added.
+
+2016-09-02  Zalan Bujtas  
+
 ASSERTION FAILED: !m_committedWidth in WebCore::LineWidth::fitBelowFloats
 https://bugs.webkit.org/show_bug.cgi?id=149462
 


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-non-resolvable-row-dimension-expected.txt (0 => 205444)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-non-resolvable-row-dimension-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-non-resolvable-row-dimension-expected.txt	2016-09-05 09:34:19 UTC (rev 205444)
@@ -0,0 +1,2 @@
+PASS if no assert in debug.
+foobar	foobar	foobar	foobar	foobar	foobar	foobar


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-non-resolvable-row-dimension.html (0 => 205444)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-non-resolvable-row-dimension.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-non-resolvable-row-dimension.html	2016-09-05 09:34:19 UTC (rev 205444)
@@ -0,0 +1,24 @@
+
+
+
+This tests that we don't assert on non-resolvable values.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+
+
+PASS if no assert in debug.
+ 
+  
+foobar
+foobar
+foobar
+foobar
+foobar
+foobar
+foobar
+  
+
+
+


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog (205443 => 205444)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:33:03 UTC (rev 205443)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:34:19 UTC (rev 205444)
@@ -1,5 +1,20 @@
 2016-09-02  Zalan Bujtas  
 
+ASSERT_NOT_REACHED() is touched in WebCore::valueForLength
+https://bugs.webkit.org/show_bug.cgi?id=123337
+
+
+Reviewed by Simon Fraser.
+
+Do not try to use unspecified height value while resolving logical height for table row.
+
+Test: fast/table/assert-on-non-resolvable-row-dimension.html
+
+* rendering/RenderTableCell.h:
+(WebCore::RenderTableCell::logicalHeightForRowSizing):
+
+2016-09-02  Zalan Bujtas  
+
 ASSERTION FAILED: !m_committedWidth in WebCore::LineWidth::fitBelowFloats
 https://bugs.webkit.org/show_bug.cgi?id=149462
 


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableCell.h (205443 => 205444)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableCell.h	2016-09-05 09:33:03 UTC (rev 205443)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableCell.h	2016-09-05 09:34:19 UTC (rev 205444)
@@ -284,6 +284,8 @@
 {
 // FIXME: This function does too much work, and is very hot during table layout!
 LayoutUnit adjustedLogicalHeight = logicalHeight() - (intrinsicPaddingBefore() + intrinsicPaddingAfter());
+if (!style().logicalHeight().isSpecified())
+return adjustedLogicalHeight;
 LayoutUnit styleLogicalHeight = valueForLength(style().logicalHeight(), 0);
 // In strict mode, box-sizing: content-box do the right thing and actually add 

[webkit-changes] [205443] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205443] releases/WebKitGTK/webkit-2.12








Revision 205443
Author carlo...@webkit.org
Date 2016-09-05 02:33:03 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r205374 - ASSERTION FAILED: !m_committedWidth in WebCore::LineWidth::fitBelowFloats
https://bugs.webkit.org/show_bug.cgi?id=149462


Reviewed by David Hyatt.

Source/WebCore:

In certain cases (multiple spans on the same line with negativ marings), the LineWidth::m_committedWidth > 0
check is not sufficient to decide if some content has already been committed to the current line.
This patch adds a flag to indicate if we ever committed to the current line.

Test: fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit.html

* rendering/line/BreakingContext.h:
(WebCore::BreakingContext::handleText):
* rendering/line/LineWidth.h:
(WebCore::LineWidth::hasCommitted):

LayoutTests:

* fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit-expected.txt: Added.
* fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/line/BreakingContext.h
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/line/LineWidth.cpp
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/line/LineWidth.h


Added Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit.html




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205442 => 205443)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:27:31 UTC (rev 205442)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:33:03 UTC (rev 205443)
@@ -1,3 +1,14 @@
+2016-09-02  Zalan Bujtas  
+
+ASSERTION FAILED: !m_committedWidth in WebCore::LineWidth::fitBelowFloats
+https://bugs.webkit.org/show_bug.cgi?id=149462
+
+
+Reviewed by David Hyatt.
+
+* fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit-expected.txt: Added.
+* fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit.html: Added.
+
 2016-08-31  Zalan Bujtas  
 
 ASSERTION FAILED: !flow->layer() && !flow->isInlineElementContinuation() in WebCore::RenderBlock::addContinuationWithOutline


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit-expected.txt (0 => 205443)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit-expected.txt	2016-09-05 09:33:03 UTC (rev 205443)
@@ -0,0 +1,2 @@
+PASS if no assert in debug.
+foobarfo b


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit.html (0 => 205443)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit.html	2016-09-05 09:33:03 UTC (rev 205443)
@@ -0,0 +1,23 @@
+
+
+
+This tests that inline does not assert when its composite state changes.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+
+div {
+max-width: 1px;
+}
+
+.title {
+margin-right: -60px;
+}
+
+
+
+PASS if no assert in debug.
+foobarfo b
+
+


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog (205442 => 205443)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:27:31 UTC (rev 205442)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:33:03 UTC (rev 205443)
@@ -1,3 +1,22 @@
+2016-09-02  Zalan Bujtas  
+
+ASSERTION FAILED: !m_committedWidth in WebCore::LineWidth::fitBelowFloats
+https://bugs.webkit.org/show_bug.cgi?id=149462
+
+
+Reviewed by David Hyatt.
+
+In certain cases (multiple spans on the same line with negativ marings), the LineWidth::m_committedWidth > 0
+check is not sufficient to decide if some content has already been committed to the current line.
+This patch adds a flag to indicate if we ever committed to the current line. 
+
+Test: fast/text/assert-when-text-with-negative-margin-sibling-does-not-fit.html
+
+* rendering/line/BreakingContext.h:
+(WebCore::BreakingContext::handleText):
+* rendering/line/LineWidth.h:
+(WebCore::LineWidth::hasCommitted):
+
 2016-08-31  Zalan Bujtas  
 
   

[webkit-changes] [205442] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205442] releases/WebKitGTK/webkit-2.12








Revision 205442
Author carlo...@webkit.org
Date 2016-09-05 02:27:31 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r205277 - ASSERTION FAILED: !flow->layer() && !flow->isInlineElementContinuation() in WebCore::RenderBlock::addContinuationWithOutline
https://bugs.webkit.org/show_bug.cgi?id=151022


Reviewed by Simon Fraser.

Source/WebCore:

Certain CSS properties like opacity, mask etc (graphical group members) require the layer to be a non-normal flow,
self painting layer so that we can manage painting properly. However in addition to those CSS properties, some of the will-change
values (again, graphical group members) also require the layer to be non-normal flow, self painting.
This patch ensures that when deciding whether a layer should be a normal flow layer, we take the will-change values into account as well.

Tests: compositing/layer-creation/assert-on-inline-element-with-will-change.html
   compositing/layer-creation/will-change-on-normal-flow-content.html

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::calculateClipRects):
* rendering/style/WillChangeData.cpp:
(WebCore::propertyCreatesGraphicalGroup):
(WebCore::WillChangeData::addFeature):
* rendering/style/WillChangeData.h:
(WebCore::WillChangeData::canCreateGraphicalGroup):
(WebCore::WillChangeData::canTriggerCompositingOnInline): Deleted.

LayoutTests:

* compositing/layer-creation/assert-on-inline-element-with-will-change-expected.txt: Added.
* compositing/layer-creation/assert-on-inline-element-with-will-change.html: Added.
* compositing/layer-creation/will-change-on-normal-flow-content-expected.txt: Added.
* compositing/layer-creation/will-change-on-normal-flow-content.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderLayer.cpp
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/style/WillChangeData.cpp
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/style/WillChangeData.h


Added Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/compositing/layer-creation/assert-on-inline-element-with-will-change-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/compositing/layer-creation/assert-on-inline-element-with-will-change.html
releases/WebKitGTK/webkit-2.12/LayoutTests/compositing/layer-creation/will-change-on-normal-flow-content-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/compositing/layer-creation/will-change-on-normal-flow-content.html




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205441 => 205442)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:19:48 UTC (rev 205441)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:27:31 UTC (rev 205442)
@@ -1,3 +1,16 @@
+2016-08-31  Zalan Bujtas  
+
+ASSERTION FAILED: !flow->layer() && !flow->isInlineElementContinuation() in WebCore::RenderBlock::addContinuationWithOutline
+https://bugs.webkit.org/show_bug.cgi?id=151022
+
+
+Reviewed by Simon Fraser.
+
+* compositing/layer-creation/assert-on-inline-element-with-will-change-expected.txt: Added.
+* compositing/layer-creation/assert-on-inline-element-with-will-change.html: Added.
+* compositing/layer-creation/will-change-on-normal-flow-content-expected.txt: Added.
+* compositing/layer-creation/will-change-on-normal-flow-content.html: Added.
+
 2016-08-26  Zalan Bujtas  
 
 ASSERT_NOT_REACHED() is touched in WebCore::minimumValueForLength


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/compositing/layer-creation/assert-on-inline-element-with-will-change-expected.txt (0 => 205442)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/compositing/layer-creation/assert-on-inline-element-with-will-change-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/compositing/layer-creation/assert-on-inline-element-with-will-change-expected.txt	2016-09-05 09:27:31 UTC (rev 205442)
@@ -0,0 +1,2 @@
+PASS if no assert in debug.
+


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/compositing/layer-creation/assert-on-inline-element-with-will-change.html (0 => 205442)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/compositing/layer-creation/assert-on-inline-element-with-will-change.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/compositing/layer-creation/assert-on-inline-element-with-will-change.html	2016-09-05 09:27:31 UTC (rev 205442)
@@ -0,0 +1,21 @@
+
+
+
+This tests that we make a self painting layer for the inline when will-change requires one.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+
+span {
+   outline: 5px solid green;
+   -webkit-columns: 100px 1;
+   will-change: opacity;
+}
+
+
+
+PASS if no assert in debug.
+
+
+


Added: 

[webkit-changes] [205441] trunk/Source/WebCore

[fred . wang]

Title: [205441] trunk/Source/WebCore








Revision 205441
Author fred.w...@free.fr
Date 2016-09-05 02:19:48 -0700 (Mon, 05 Sep 2016)


Log Message
More refactoring of RenderMathMLScripts
https://bugs.webkit.org/show_bug.cgi?id=161371

Patch by Frederic Wang  on 2016-09-05
Reviewed by Darin Adler.

This is a follow-up of bug 161084. The function getScriptMetricsAndLayoutIfNeeded was quite
complicated and it was not obvious that we have to call it twice with the same reference
to a struture holding vertical metrics. We extract the part retrieving layout parameters
into verticalParameters and move its layoutIfNeeded calls into layoutBlock. Then it can
be reduced to a simple function that retrieve the vertical metrics in one call.
We also improve getBaseAndScripts to make clear that it is performing validation. It returns
a ReferenceChildren structure encapsulating pointers to important children so that we no
longer pass these pointers as function parameters. We continue to need them to browse the
list of prescripts & postscripts but we refactor a bit the loop to avoid explicit mention
of RenderBox*.

No new tests, already covered by existing tests.

* rendering/mathml/RenderMathMLScripts.cpp:
(WebCore::RenderMathMLScripts::validateAndGetReferenceChildren): We now store the pointers to
the base, firstPostScript and firstPreScript children in the ReferenceChildren structure. We
also add a pointer to the prescriptDelimiter for convenience.
(WebCore::RenderMathMLScripts::italicCorrection): Use the ReferenceChildren structure so that
we are sure the base has been validated before calling this function.
(WebCore::RenderMathMLScripts::computePreferredLogicalWidths): Retrieve the reference
children with validateAndGetReferenceChildren instead of calling getBaseAndScripts and use
ReferenceChildren to handle these children and to call italicCorrection. The loops for
SubSup, UnderOver, Multiscripts are also rewritten a bit to avoid declaring a null RenderBox*
outside of them and hence allow to use auto.
(WebCore::RenderMathMLScripts::verticalParameters): This part to extract the layout
parameters is extracted from getScriptMetricsAndLayoutIfNeeded. The parameters are returned
as a VerticalParameters struct.
(WebCore::RenderMathMLScripts::verticalMetrics): This is the remaining part of
getScriptMetricsAndLayoutIfNeeded It used to call layoutIfNeeded on children and to
calculate maximum vertical metrics. For Multiscripts it was called twice: We did a first
call to handle the prescripts and then pass the result again in the second call to handle
the postscripts. We modify a bit the loop so that all the scripts are handled in one call and
hence we can directly return a VerticalMetrics. Again, the reference children are now handled
using the ReferenceChildren structure passed as a parameter.
(WebCore::RenderMathMLScripts::layoutBlock): We retrieve the reference children with
validateAndGetReferenceChildren instead of calling getBaseAndScripts and use
ReferenceChildren to handle these children and to call italicCorrection. We layout all the
children if needed in one loop at the beginning instead of doing that when their vertical
metrics are needed. We can now also retrieve vertical metrics with a single call.
(WebCore::RenderMathMLScripts::getBaseAndScripts): Renamed validateAndGetReferenceChildren.
(WebCore::RenderMathMLScripts::getScriptMetricsAndLayoutIfNeeded): Deleted. Split into
verticalParameters and verticalMetrics.
* rendering/mathml/RenderMathMLScripts.h: New structure to handle the pointers to reference
children. Update the signature of getBaseAndScripts to use this struture and give a clearer
name. Update the signature of italicCorrection to use this structure too. Add a new structure
VerticalParameters and declare the helper function to retrieve them. Rename ScriptMetrics
to VerticalMetrics and update the signature of the function needed to retrieve it.

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/rendering/mathml/RenderMathMLScripts.cpp
trunk/Source/WebCore/rendering/mathml/RenderMathMLScripts.h




Diff

Modified: trunk/Source/WebCore/ChangeLog (205440 => 205441)

--- trunk/Source/WebCore/ChangeLog	2016-09-05 09:18:47 UTC (rev 205440)
+++ trunk/Source/WebCore/ChangeLog	2016-09-05 09:19:48 UTC (rev 205441)
@@ -1,3 +1,58 @@
+2016-09-05  Frederic Wang  
+
+More refactoring of RenderMathMLScripts
+https://bugs.webkit.org/show_bug.cgi?id=161371
+
+Reviewed by Darin Adler.
+
+This is a follow-up of bug 161084. The function getScriptMetricsAndLayoutIfNeeded was quite
+complicated and it was not obvious that we have to call it twice with the same reference
+to a struture holding vertical metrics. We extract the part retrieving layout parameters
+into verticalParameters and move its layoutIfNeeded calls into layoutBlock. Then it can
+be reduced to a simple function that retrieve the vertical 

[webkit-changes] [205440] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205440] releases/WebKitGTK/webkit-2.12








Revision 205440
Author carlo...@webkit.org
Date 2016-09-05 02:18:47 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r205056 - ASSERT_NOT_REACHED() is touched in WebCore::minimumValueForLength
https://bugs.webkit.org/show_bug.cgi?id=125781


Reviewed by Simon Fraser.

Source/WebCore:

RenderTableSection::calcRowLogicalHeight misused minimumValueForLength to fallback to 0 for non-fixed values.
While this patch fixes the assertion, the table section code needs works to support calc values. See webkit.org/b/161273.

Test: fast/table/assert-on-min-max-content-values.html

* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::calcRowLogicalHeight):

LayoutTests:

* fast/table/assert-on-min-max-content-values-expected.txt: Added.
* fast/table/assert-on-min-max-content-values.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableSection.cpp


Added Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-min-max-content-values-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-min-max-content-values.html




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205439 => 205440)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:17:09 UTC (rev 205439)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:18:47 UTC (rev 205440)
@@ -1,3 +1,14 @@
+2016-08-26  Zalan Bujtas  
+
+ASSERT_NOT_REACHED() is touched in WebCore::minimumValueForLength
+https://bugs.webkit.org/show_bug.cgi?id=125781
+
+
+Reviewed by Simon Fraser.
+
+* fast/table/assert-on-min-max-content-values-expected.txt: Added.
+* fast/table/assert-on-min-max-content-values.html: Added.
+
 2016-08-24  Zalan Bujtas  
 
 ASSERTION FAILED: contentSize >= 0 in WebCore::RenderFlexibleBox::adjustChildSizeForMinAndMax


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-min-max-content-values-expected.txt (0 => 205440)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-min-max-content-values-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-min-max-content-values-expected.txt	2016-09-05 09:18:47 UTC (rev 205440)
@@ -0,0 +1,2 @@
+PASS if no assert in debug.
+


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-min-max-content-values.html (0 => 205440)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-min-max-content-values.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-on-min-max-content-values.html	2016-09-05 09:18:47 UTC (rev 205440)
@@ -0,0 +1,20 @@
+
+
+
+This tests that we don't assert on tables with min/max-content values.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+
+
+PASS if no assert in debug.
+
+
+	 
+	 
+	 
+
+
+
+


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog (205439 => 205440)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:17:09 UTC (rev 205439)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:18:47 UTC (rev 205440)
@@ -1,3 +1,19 @@
+2016-08-26  Zalan Bujtas  
+
+ASSERT_NOT_REACHED() is touched in WebCore::minimumValueForLength
+https://bugs.webkit.org/show_bug.cgi?id=125781
+
+
+Reviewed by Simon Fraser.
+
+RenderTableSection::calcRowLogicalHeight misused minimumValueForLength to fallback to 0 for non-fixed values.
+While this patch fixes the assertion, the table section code needs works to support calc values. See webkit.org/b/161273.   
+
+Test: fast/table/assert-on-min-max-content-values.html
+
+* rendering/RenderTableSection.cpp:
+(WebCore::RenderTableSection::calcRowLogicalHeight):
+
 2016-08-24  Zalan Bujtas  
 
 ASSERTION FAILED: contentSize >= 0 in WebCore::RenderFlexibleBox::adjustChildSizeForMinAndMax


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableSection.cpp (205439 => 205440)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableSection.cpp	2016-09-05 09:17:09 UTC (rev 205439)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableSection.cpp	2016-09-05 09:18:47 UTC (rev 205440)
@@ -251,6 +251,15 @@
 cell->setCol(table()->effColToCol(col));
 }
 
+static LayoutUnit resolveLogicalHeightForRow(const Length& rowLogicalHeight)
+{
+if (rowLogicalHeight.isFixed())
+return rowLogicalHeight.value();
+if (rowLogicalHeight.isCalculated())
+return rowLogicalHeight.nonNanCalculatedValue(0);
+return 0;
+}
+
 LayoutUnit RenderTableSection::calcRowLogicalHeight()
 {
 

[webkit-changes] [205439] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205439] releases/WebKitGTK/webkit-2.12








Revision 205439
Author carlo...@webkit.org
Date 2016-09-05 02:17:09 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r204951 - ASSERTION FAILED: contentSize >= 0 in WebCore::RenderFlexibleBox::adjustChildSizeForMinAndMax
https://bugs.webkit.org/show_bug.cgi?id=155700


Reviewed by David Hyatt.

Source/WebCore:

RenderBox::m_minPreferredLogicalWidth/m_maxPreferredLogicalWidth don't need special initial value since
preferredLogicalWidthsDirty flag guards them. -1 as initial value can cause problems for renderers that don't
override RenderBox::computePreferredLogicalWidths().

Test: fast/ruby/assert-when-content-size-is-negative.html

* rendering/RenderBox.cpp:
(WebCore::RenderBox::RenderBox):
(WebCore::RenderBox::dirtyLineBoxes):
(WebCore::RenderBox::deleteLineBoxWrapper):
* rendering/RenderBox.h:

LayoutTests:

* fast/ruby/assert-when-content-size-is-negative-expected.txt: Added.
* fast/ruby/assert-when-content-size-is-negative.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBox.cpp
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBox.h


Added Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-when-content-size-is-negative-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-when-content-size-is-negative.html




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205438 => 205439)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:09:52 UTC (rev 205438)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:17:09 UTC (rev 205439)
@@ -1,5 +1,16 @@
 2016-08-24  Zalan Bujtas  
 
+ASSERTION FAILED: contentSize >= 0 in WebCore::RenderFlexibleBox::adjustChildSizeForMinAndMax
+https://bugs.webkit.org/show_bug.cgi?id=155700
+
+
+Reviewed by David Hyatt.
+
+* fast/ruby/assert-when-content-size-is-negative-expected.txt: Added.
+* fast/ruby/assert-when-content-size-is-negative.html: Added.
+
+2016-08-24  Zalan Bujtas  
+
 ASSERTION FAILED: childrenInline() in WebCore::RenderBlockFlow::hasLines
 https://bugs.webkit.org/show_bug.cgi?id=139396
 


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-when-content-size-is-negative-expected.txt (0 => 205439)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-when-content-size-is-negative-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-when-content-size-is-negative-expected.txt	2016-09-05 09:17:09 UTC (rev 205439)
@@ -0,0 +1,4 @@
+This tests that inline does not assert when its composite state changes.
+if (window.testRunner) testRunner.dumpAsText();
+* { display: inline-flex; flex-basis: 1; }
+PASS if no assert in debug.


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-when-content-size-is-negative.html (0 => 205439)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-when-content-size-is-negative.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-when-content-size-is-negative.html	2016-09-05 09:17:09 UTC (rev 205439)
@@ -0,0 +1,20 @@
+
+
+
+This tests that inline does not assert when its composite state changes.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+
+* {
+display: inline-flex;
+flex-basis: 1;
+}
+
+
+
+PASS if no assert in debug.
+
+
+


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog (205438 => 205439)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:09:52 UTC (rev 205438)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:17:09 UTC (rev 205439)
@@ -1,5 +1,25 @@
 2016-08-24  Zalan Bujtas  
 
+ASSERTION FAILED: contentSize >= 0 in WebCore::RenderFlexibleBox::adjustChildSizeForMinAndMax
+https://bugs.webkit.org/show_bug.cgi?id=155700
+
+
+Reviewed by David Hyatt.
+
+RenderBox::m_minPreferredLogicalWidth/m_maxPreferredLogicalWidth don't need special initial value since
+preferredLogicalWidthsDirty flag guards them. -1 as initial value can cause problems for renderers that don't
+override RenderBox::computePreferredLogicalWidths().
+
+Test: fast/ruby/assert-when-content-size-is-negative.html
+
+* rendering/RenderBox.cpp:
+(WebCore::RenderBox::RenderBox):
+(WebCore::RenderBox::dirtyLineBoxes):
+(WebCore::RenderBox::deleteLineBoxWrapper):
+* rendering/RenderBox.h:
+
+2016-08-24  Zalan Bujtas  
+
 ASSERTION FAILED: childrenInline() in WebCore::RenderBlockFlow::hasLines
 https://bugs.webkit.org/show_bug.cgi?id=139396
 


Modified: 

[webkit-changes] [205438] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205438] releases/WebKitGTK/webkit-2.12








Revision 205438
Author carlo...@webkit.org
Date 2016-09-05 02:09:52 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r204908 - ASSERTION FAILED: childrenInline() in WebCore::RenderBlockFlow::hasLines
https://bugs.webkit.org/show_bug.cgi?id=139396


Reviewed by David Hyatt.

Source/WebCore:

RenderBlockFlow::hasLines() should be able to handle the block-children case.

Test: fast/ruby/assert-with-block-rt-child-and-emphasis.html

* rendering/RenderBlock.h:
(WebCore::RenderBlock::hasLines): Deleted.
* rendering/RenderBlockFlow.cpp:
(WebCore::RenderBlockFlow::hasLines):
* rendering/RenderBlockFlow.h:

LayoutTests:

* fast/ruby/assert-with-block-rt-child-and-emphasis-expected.txt: Added.
* fast/ruby/assert-with-block-rt-child-and-emphasis.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBlock.h
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBlockFlow.cpp
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBlockFlow.h


Added Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-with-block-rt-child-and-emphasis-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-with-block-rt-child-and-emphasis.html




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205437 => 205438)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:03:40 UTC (rev 205437)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:09:52 UTC (rev 205438)
@@ -1,3 +1,14 @@
+2016-08-24  Zalan Bujtas  
+
+ASSERTION FAILED: childrenInline() in WebCore::RenderBlockFlow::hasLines
+https://bugs.webkit.org/show_bug.cgi?id=139396
+
+
+Reviewed by David Hyatt.
+
+* fast/ruby/assert-with-block-rt-child-and-emphasis-expected.txt: Added.
+* fast/ruby/assert-with-block-rt-child-and-emphasis.html: Added.
+
 2016-08-23  Zalan Bujtas  
 
 ASSERTION FAILED: !view().layoutStateEnabled() || style().styleType() == FIRST_LETTER in WebCore::RenderInline::clippedOverflowRectForRepaint


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-with-block-rt-child-and-emphasis-expected.txt (0 => 205438)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-with-block-rt-child-and-emphasis-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-with-block-rt-child-and-emphasis-expected.txt	2016-09-05 09:09:52 UTC (rev 205438)
@@ -0,0 +1,2 @@
+PASS if no assert in debug
+


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-with-block-rt-child-and-emphasis.html (0 => 205438)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-with-block-rt-child-and-emphasis.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/ruby/assert-with-block-rt-child-and-emphasis.html	2016-09-05 09:09:52 UTC (rev 205438)
@@ -0,0 +1,18 @@
+
+
+
+This tests that we don't assert when rt has a block child with emphasis on.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+
+ruby {
+text-emphasis-style: filled;
+}
+
+
+
+PASS if no assert in debug
+
+


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog (205437 => 205438)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:03:40 UTC (rev 205437)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:09:52 UTC (rev 205438)
@@ -1,3 +1,21 @@
+2016-08-24  Zalan Bujtas  
+
+ASSERTION FAILED: childrenInline() in WebCore::RenderBlockFlow::hasLines
+https://bugs.webkit.org/show_bug.cgi?id=139396
+
+
+Reviewed by David Hyatt.
+
+RenderBlockFlow::hasLines() should be able to handle the block-children case.
+
+Test: fast/ruby/assert-with-block-rt-child-and-emphasis.html
+
+* rendering/RenderBlock.h:
+(WebCore::RenderBlock::hasLines): Deleted.
+* rendering/RenderBlockFlow.cpp:
+(WebCore::RenderBlockFlow::hasLines):
+* rendering/RenderBlockFlow.h:
+
 2016-08-23  Zalan Bujtas  
 
 ASSERTION FAILED: !view().layoutStateEnabled() || style().styleType() == FIRST_LETTER in WebCore::RenderInline::clippedOverflowRectForRepaint


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBlock.h (205437 => 205438)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBlock.h	2016-09-05 09:03:40 UTC (rev 205437)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBlock.h	2016-09-05 09:09:52 UTC (rev 205438)
@@ -415,9 +415,6 @@
 virtual bool isSelfCollapsingBlock() const override;
 virtual bool childrenPreventSelfCollapsing() const;
 
-// FIXME-BLOCKFLOW: Remove virtualizaion when all 

[webkit-changes] [205437] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205437] releases/WebKitGTK/webkit-2.12








Revision 205437
Author carlo...@webkit.org
Date 2016-09-05 02:03:40 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r204880 - ASSERTION FAILED: !view().layoutStateEnabled() || style().styleType() == FIRST_LETTER in WebCore::RenderInline::clippedOverflowRectForRepaint
https://bugs.webkit.org/show_bug.cgi?id=155363


Reviewed by David Hyatt.

Source/WebCore:

Loosen ASSERT to include calls when the inline has a self painting layer. It's valid to end up here during layout when
the layer's composite state changes.

Test: fast/layers/assert-on-self-painting-inline-with-scrolling.html

* rendering/RenderInline.cpp:
(WebCore::RenderInline::clippedOverflowRectForRepaint):

LayoutTests:

* fast/layers/assert-on-self-painting-inline-with-scrolling-expected.txt: Added.
* fast/layers/assert-on-self-painting-inline-with-scrolling.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderInline.cpp


Added Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/fast/layers/assert-on-self-painting-inline-with-scrolling-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/fast/layers/assert-on-self-painting-inline-with-scrolling.html




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205436 => 205437)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:02:37 UTC (rev 205436)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:03:40 UTC (rev 205437)
@@ -1,5 +1,16 @@
 2016-08-23  Zalan Bujtas  
 
+ASSERTION FAILED: !view().layoutStateEnabled() || style().styleType() == FIRST_LETTER in WebCore::RenderInline::clippedOverflowRectForRepaint
+https://bugs.webkit.org/show_bug.cgi?id=155363
+
+
+Reviewed by David Hyatt.
+
+* fast/layers/assert-on-self-painting-inline-with-scrolling-expected.txt: Added.
+* fast/layers/assert-on-self-painting-inline-with-scrolling.html: Added.
+
+2016-08-23  Zalan Bujtas  
+
 ASSERTION FAILED: hasOverflowingCell == this->hasOverflowingCell() in WebCore::RenderTableSection::computeOverflowFromCells
 https://bugs.webkit.org/show_bug.cgi?id=155702
 


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/layers/assert-on-self-painting-inline-with-scrolling-expected.txt (0 => 205437)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/layers/assert-on-self-painting-inline-with-scrolling-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/layers/assert-on-self-painting-inline-with-scrolling-expected.txt	2016-09-05 09:03:40 UTC (rev 205437)
@@ -0,0 +1 @@
+PASS if no assert in debug.


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/layers/assert-on-self-painting-inline-with-scrolling.html (0 => 205437)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/layers/assert-on-self-painting-inline-with-scrolling.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/layers/assert-on-self-painting-inline-with-scrolling.html	2016-09-05 09:03:40 UTC (rev 205437)
@@ -0,0 +1,25 @@
+
+
+
+This tests that inline does not assert when its composite state changes.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+
+
+ html, body {
+overflow-x: scroll;
+will-change: transform;
+}
+
+span {
+mix-blend-mode: exclusion;
+}
+
+
+
+PASS if no assert in debug.
+
+
+


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog (205436 => 205437)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:02:37 UTC (rev 205436)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:03:40 UTC (rev 205437)
@@ -1,5 +1,21 @@
 2016-08-23  Zalan Bujtas  
 
+ASSERTION FAILED: !view().layoutStateEnabled() || style().styleType() == FIRST_LETTER in WebCore::RenderInline::clippedOverflowRectForRepaint
+https://bugs.webkit.org/show_bug.cgi?id=155363
+
+
+Reviewed by David Hyatt.
+
+Loosen ASSERT to include calls when the inline has a self painting layer. It's valid to end up here during layout when
+the layer's composite state changes.
+
+Test: fast/layers/assert-on-self-painting-inline-with-scrolling.html
+
+* rendering/RenderInline.cpp:
+(WebCore::RenderInline::clippedOverflowRectForRepaint):
+
+2016-08-23  Zalan Bujtas  
+
 ASSERTION FAILED: hasOverflowingCell == this->hasOverflowingCell() in WebCore::RenderTableSection::computeOverflowFromCells
 https://bugs.webkit.org/show_bug.cgi?id=155702
 


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderInline.cpp (205436 => 205437)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderInline.cpp	2016-09-05 09:02:37 UTC (rev 205436)

[webkit-changes] [205436] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205436] releases/WebKitGTK/webkit-2.12








Revision 205436
Author carlo...@webkit.org
Date 2016-09-05 02:02:37 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r204860 - ASSERTION FAILED: hasOverflowingCell == this->hasOverflowingCell() in WebCore::RenderTableSection::computeOverflowFromCells
https://bugs.webkit.org/show_bug.cgi?id=155702


Reviewed by David Hyatt.

Source/WebCore:

The simplified layout path for table/section should also clear the slow path flag for visual overflow (since it recomputes
the overflow for the descendant cells).

Test: fast/table/assert-when-visual-overflow-is-cleared.html

* rendering/RenderTableSection.cpp:
(WebCore::RenderTableSection::layout):

LayoutTests:

* fast/table/assert-when-visual-overflow-is-cleared-expected.txt: Added.
* fast/table/assert-when-visual-overflow-is-cleared.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableSection.cpp


Added Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-when-visual-overflow-is-cleared-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-when-visual-overflow-is-cleared.html




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205435 => 205436)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 08:56:49 UTC (rev 205435)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 09:02:37 UTC (rev 205436)
@@ -1,3 +1,14 @@
+2016-08-23  Zalan Bujtas  
+
+ASSERTION FAILED: hasOverflowingCell == this->hasOverflowingCell() in WebCore::RenderTableSection::computeOverflowFromCells
+https://bugs.webkit.org/show_bug.cgi?id=155702
+
+
+Reviewed by David Hyatt.
+
+* fast/table/assert-when-visual-overflow-is-cleared-expected.txt: Added.
+* fast/table/assert-when-visual-overflow-is-cleared.html: Added.
+
 2016-08-24  Chris Dumez  
 
 Assertion hit: ASSERT(!m_parserScheduler) in ~HTMLDocumentParser()


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-when-visual-overflow-is-cleared-expected.txt (0 => 205436)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-when-visual-overflow-is-cleared-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-when-visual-overflow-is-cleared-expected.txt	2016-09-05 09:02:37 UTC (rev 205436)
@@ -0,0 +1,3 @@
+PASS if no assert in debug.
+a 
+  


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-when-visual-overflow-is-cleared.html (0 => 205436)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-when-visual-overflow-is-cleared.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/table/assert-when-visual-overflow-is-cleared.html	2016-09-05 09:02:37 UTC (rev 205436)
@@ -0,0 +1,29 @@
+
+
+
+This tests that table does not assert when child visual overflow is cleared.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+window._onload_ = function() {
+	window.scrollTo(0, document.body.scrollHeight);
+	document.designMode = 'on';
+}
+
+
+pre {
+	margin-top: -1px;
+	position: relative;
+	display: inline-table;
+}
+
+span {
+	position: absolute;
+}
+
+
+
+PASS if no assert in debug.
+a
+
\ No newline at end of file


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog (205435 => 205436)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 08:56:49 UTC (rev 205435)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 09:02:37 UTC (rev 205436)
@@ -1,3 +1,19 @@
+2016-08-23  Zalan Bujtas  
+
+ASSERTION FAILED: hasOverflowingCell == this->hasOverflowingCell() in WebCore::RenderTableSection::computeOverflowFromCells
+https://bugs.webkit.org/show_bug.cgi?id=155702
+
+
+Reviewed by David Hyatt.
+
+The simplified layout path for table/section should also clear the slow path flag for visual overflow (since it recomputes
+the overflow for the descendant cells). 
+
+Test: fast/table/assert-when-visual-overflow-is-cleared.html
+
+* rendering/RenderTableSection.cpp:
+(WebCore::RenderTableSection::layout):
+
 2016-08-24  Chris Dumez  
 
 Assertion hit: ASSERT(!m_parserScheduler) in ~HTMLDocumentParser()


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableSection.cpp (205435 => 205436)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableSection.cpp	2016-09-05 08:56:49 UTC (rev 205435)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderTableSection.cpp	2016-09-05 09:02:37 UTC (rev 205436)
@@ -372,6 +372,7 @@
 ASSERT(!needsCellRecalc());
 ASSERT(!table()->needsSectionRecalc());
 
+

[webkit-changes] [205435] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205435] releases/WebKitGTK/webkit-2.12








Revision 205435
Author carlo...@webkit.org
Date 2016-09-05 01:56:49 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r204936 - Assertion hit: ASSERT(!m_parserScheduler) in ~HTMLDocumentParser()
https://bugs.webkit.org/show_bug.cgi?id=161083


Reviewed by Darin Adler.

Source/WebCore:

Call cancelParsing() right before re-creating the new parser in
Document::implicitOpen(). We would previously call removeChildren()
in between which is was an issue because it can cause JS to be
executed and therefore the parser to be re-created after we have
cancelled parsing.

No new tests, unskipped html/browsers/browsing-the-web/unloading-documents/004.html
which was consistently crashing.

* dom/Document.cpp:
(WebCore::Document::implicitOpen):

LayoutTests:

Unskip html/browsers/browsing-the-web/unloading-documents/004.html as
it is no longer crashing.

* TestExpectations:

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/dom/Document.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205434 => 205435)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 08:55:23 UTC (rev 205434)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 08:56:49 UTC (rev 205435)
@@ -1,3 +1,16 @@
+2016-08-24  Chris Dumez  
+
+Assertion hit: ASSERT(!m_parserScheduler) in ~HTMLDocumentParser()
+https://bugs.webkit.org/show_bug.cgi?id=161083
+
+
+Reviewed by Darin Adler.
+
+Unskip html/browsers/browsing-the-web/unloading-documents/004.html as
+it is no longer crashing.
+
+* TestExpectations:
+
 2016-08-25  Zalan Bujtas  
 
 Infinite recursion crash in WebCore::RenderBlockFlow::layoutBlock


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog (205434 => 205435)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 08:55:23 UTC (rev 205434)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 08:56:49 UTC (rev 205435)
@@ -1,3 +1,23 @@
+2016-08-24  Chris Dumez  
+
+Assertion hit: ASSERT(!m_parserScheduler) in ~HTMLDocumentParser()
+https://bugs.webkit.org/show_bug.cgi?id=161083
+
+
+Reviewed by Darin Adler.
+
+Call cancelParsing() right before re-creating the new parser in
+Document::implicitOpen(). We would previously call removeChildren()
+in between which is was an issue because it can cause JS to be
+executed and therefore the parser to be re-created after we have
+cancelled parsing.
+
+No new tests, unskipped html/browsers/browsing-the-web/unloading-documents/004.html
+which was consistently crashing.
+
+* dom/Document.cpp:
+(WebCore::Document::implicitOpen):
+
 2016-08-25  Zalan Bujtas  
 
 Infinite recursion crash in WebCore::RenderBlockFlow::layoutBlock


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/dom/Document.cpp (205434 => 205435)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/dom/Document.cpp	2016-09-05 08:55:23 UTC (rev 205434)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/dom/Document.cpp	2016-09-05 08:56:49 UTC (rev 205435)
@@ -2561,12 +2561,11 @@
 
 void Document::implicitOpen()
 {
-cancelParsing();
-
 removeChildren();
 
 setCompatibilityMode(DocumentCompatibilityMode::NoQuirksMode);
 
+cancelParsing();
 m_parser = createParser();
 setParsing(true);
 setReadyState(Loading);






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205434] releases/WebKitGTK/webkit-2.12

[carlosgc]

Title: [205434] releases/WebKitGTK/webkit-2.12








Revision 205434
Author carlo...@webkit.org
Date 2016-09-05 01:55:23 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r204980 - Infinite recursion crash in WebCore::RenderBlockFlow::layoutBlock
https://bugs.webkit.org/show_bug.cgi?id=139474


Reviewed by David Hyatt.

Source/WebCore:

We should just give up trying to avoid widow when the page is too small to break line.

Test: fast/multicol/assert-on-small-page-height-with-widow.html

* rendering/RenderBlockFlow.cpp:
(WebCore::RenderBlockFlow::clearShouldBreakAtLineToAvoidWidowIfNeeded):
(WebCore::RenderBlockFlow::adjustLinePositionForPagination):
* rendering/RenderBlockFlow.h:

LayoutTests:

* fast/multicol/assert-on-small-page-height-with-widow-expected.txt: Added.
* fast/multicol/assert-on-small-page-height-with-widow.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBlockFlow.cpp


Added Paths

releases/WebKitGTK/webkit-2.12/LayoutTests/fast/multicol/assert-on-small-page-height-with-widow-expected.txt
releases/WebKitGTK/webkit-2.12/LayoutTests/fast/multicol/assert-on-small-page-height-with-widow.html




Diff

Modified: releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog (205433 => 205434)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 08:49:07 UTC (rev 205433)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/ChangeLog	2016-09-05 08:55:23 UTC (rev 205434)
@@ -1,3 +1,14 @@
+2016-08-25  Zalan Bujtas  
+
+Infinite recursion crash in WebCore::RenderBlockFlow::layoutBlock
+https://bugs.webkit.org/show_bug.cgi?id=139474
+
+
+Reviewed by David Hyatt.
+
+* fast/multicol/assert-on-small-page-height-with-widow-expected.txt: Added.
+* fast/multicol/assert-on-small-page-height-with-widow.html: Added.
+
 2016-05-03  Chris Dumez  
 
 Unreviewed, drop outdated layout test after r200375.


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/multicol/assert-on-small-page-height-with-widow-expected.txt (0 => 205434)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/multicol/assert-on-small-page-height-with-widow-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/multicol/assert-on-small-page-height-with-widow-expected.txt	2016-09-05 08:55:23 UTC (rev 205434)
@@ -0,0 +1,3 @@
+PASS if no assert in debug.
+
+


Added: releases/WebKitGTK/webkit-2.12/LayoutTests/fast/multicol/assert-on-small-page-height-with-widow.html (0 => 205434)

--- releases/WebKitGTK/webkit-2.12/LayoutTests/fast/multicol/assert-on-small-page-height-with-widow.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.12/LayoutTests/fast/multicol/assert-on-small-page-height-with-widow.html	2016-09-05 08:55:23 UTC (rev 205434)
@@ -0,0 +1,28 @@
+
+
+
+This tests that.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+
+html {
+-webkit-column-count: 3;
+}
+
+body {
+max-height: 0px;
+-webkit-column-count: 3;
+widows: 2;   
+}
+span {
+margin-right: -1px;
+}
+
+
+
+PASS if no assert in debug.
+
+
+


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog (205433 => 205434)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 08:49:07 UTC (rev 205433)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/ChangeLog	2016-09-05 08:55:23 UTC (rev 205434)
@@ -1,3 +1,20 @@
+2016-08-25  Zalan Bujtas  
+
+Infinite recursion crash in WebCore::RenderBlockFlow::layoutBlock
+https://bugs.webkit.org/show_bug.cgi?id=139474
+
+
+Reviewed by David Hyatt.
+
+We should just give up trying to avoid widow when the page is too small to break line.
+
+Test: fast/multicol/assert-on-small-page-height-with-widow.html
+
+* rendering/RenderBlockFlow.cpp:
+(WebCore::RenderBlockFlow::clearShouldBreakAtLineToAvoidWidowIfNeeded):
+(WebCore::RenderBlockFlow::adjustLinePositionForPagination):
+* rendering/RenderBlockFlow.h:
+
 2016-05-03  Pranjal Jumde  
 
 WorkerGlobalScope's self, location and navigator attributes should not be replaceable


Modified: releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBlockFlow.cpp (205433 => 205434)

--- releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBlockFlow.cpp	2016-09-05 08:49:07 UTC (rev 205433)
+++ releases/WebKitGTK/webkit-2.12/Source/WebCore/rendering/RenderBlockFlow.cpp	2016-09-05 08:55:23 UTC (rev 205434)
@@ -1648,7 +1648,15 @@
 
 return false;
 }
-
+
+static void clearShouldBreakAtLineToAvoidWidowIfNeeded(RenderBlockFlow& blockFlow)
+{
+if (!blockFlow.shouldBreakAtLineToAvoidWidow())
+return;
+blockFlow.clearShouldBreakAtLineToAvoidWidow();
+blockFlow.setDidBreakAtLineToAvoidWidow();
+}
+
 void 

[webkit-changes] [205433] releases/WebKitGTK/webkit-2.12/Source/WebKit2

[carlosgc]

Title: [205433] releases/WebKitGTK/webkit-2.12/Source/WebKit2








Revision 205433
Author carlo...@webkit.org
Date 2016-09-05 01:49:07 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r205182 - REGRESSION(r194846): [GTK] UI process crash visiting sites protected with HTTP auth when using GTK+ < 3.14
https://bugs.webkit.org/show_bug.cgi?id=161385

Reviewed by Antonio Gomes.

Since r194846, we are chaining up webkitWebViewBaseMotionNotifyEvent, but before GTK+ 3.14
GtkWidget::motion_notify_event was nullptr, so it can't be used unconditionally.

* UIProcess/API/gtk/WebKitWebViewBase.cpp:
(webkitWebViewBaseMotionNotifyEvent):

Modified Paths

releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebKit2/UIProcess/API/gtk/WebKitWebViewBase.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog (205432 => 205433)

--- releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog	2016-09-05 08:31:11 UTC (rev 205432)
+++ releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog	2016-09-05 08:49:07 UTC (rev 205433)
@@ -1,3 +1,16 @@
+2016-08-30  Carlos Garcia Campos  
+
+REGRESSION(r194846): [GTK] UI process crash visiting sites protected with HTTP auth when using GTK+ < 3.14
+https://bugs.webkit.org/show_bug.cgi?id=161385
+
+Reviewed by Antonio Gomes.
+
+Since r194846, we are chaining up webkitWebViewBaseMotionNotifyEvent, but before GTK+ 3.14
+GtkWidget::motion_notify_event was nullptr, so it can't be used unconditionally.
+
+* UIProcess/API/gtk/WebKitWebViewBase.cpp:
+(webkitWebViewBaseMotionNotifyEvent):
+
 2016-07-07  Alex Christensen  
 
 Followup to r202939.


Modified: releases/WebKitGTK/webkit-2.12/Source/WebKit2/UIProcess/API/gtk/WebKitWebViewBase.cpp (205432 => 205433)

--- releases/WebKitGTK/webkit-2.12/Source/WebKit2/UIProcess/API/gtk/WebKitWebViewBase.cpp	2016-09-05 08:31:11 UTC (rev 205432)
+++ releases/WebKitGTK/webkit-2.12/Source/WebKit2/UIProcess/API/gtk/WebKitWebViewBase.cpp	2016-09-05 08:49:07 UTC (rev 205433)
@@ -866,8 +866,10 @@
 WebKitWebViewBase* webViewBase = WEBKIT_WEB_VIEW_BASE(widget);
 WebKitWebViewBasePrivate* priv = webViewBase->priv;
 
-if (priv->authenticationDialog)
-return GTK_WIDGET_CLASS(webkit_web_view_base_parent_class)->motion_notify_event(widget, event);
+if (priv->authenticationDialog) {
+auto* widgetClass = GTK_WIDGET_CLASS(webkit_web_view_base_parent_class);
+return widgetClass->motion_notify_event ? widgetClass->motion_notify_event(widget, event) : FALSE;
+}
 
 priv->pageProxy->handleMouseEvent(NativeWebMouseEvent(reinterpret_cast(event), 0 /* currentClickCount */));
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205432] releases/WebKitGTK/webkit-2.12/Source/WebKit2

[carlosgc]

Title: [205432] releases/WebKitGTK/webkit-2.12/Source/WebKit2








Revision 205432
Author carlo...@webkit.org
Date 2016-09-05 01:31:11 -0700 (Mon, 05 Sep 2016)


Log Message
Merge r202940 - Followup to r202939.
https://bugs.webkit.org/show_bug.cgi?id=159526

* NetworkProcess/NetworkResourceLoader.cpp:
(WebKit::NetworkResourceLoader::didFailLoading):
We need the correct destination id when sending a message.

Modified Paths

releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog
releases/WebKitGTK/webkit-2.12/Source/WebKit2/NetworkProcess/NetworkResourceLoader.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog (205431 => 205432)

--- releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog	2016-09-05 07:57:57 UTC (rev 205431)
+++ releases/WebKitGTK/webkit-2.12/Source/WebKit2/ChangeLog	2016-09-05 08:31:11 UTC (rev 205432)
@@ -1,3 +1,12 @@
+2016-07-07  Alex Christensen  
+
+Followup to r202939.
+https://bugs.webkit.org/show_bug.cgi?id=159526
+
+* NetworkProcess/NetworkResourceLoader.cpp:
+(WebKit::NetworkResourceLoader::didFailLoading):
+We need the correct destination id when sending a message.
+
 2016-08-24  Carlos Garcia Campos  
 
 Unreviewed. Update OptionsGTK.cmake and NEWS for 2.12.4 release.


Modified: releases/WebKitGTK/webkit-2.12/Source/WebKit2/NetworkProcess/NetworkResourceLoader.cpp (205431 => 205432)

--- releases/WebKitGTK/webkit-2.12/Source/WebKit2/NetworkProcess/NetworkResourceLoader.cpp	2016-09-05 07:57:57 UTC (rev 205431)
+++ releases/WebKitGTK/webkit-2.12/Source/WebKit2/NetworkProcess/NetworkResourceLoader.cpp	2016-09-05 08:31:11 UTC (rev 205432)
@@ -378,7 +378,7 @@
 m_synchronousLoadData->error = error;
 sendReplyToSynchronousRequest(*m_synchronousLoadData, nullptr);
 } else if (auto* connection = messageSenderConnection())
-connection->send(Messages::WebResourceLoader::DidFailResourceLoad(error), 0, 0);
+connection->send(Messages::WebResourceLoader::DidFailResourceLoad(error), messageSenderDestinationID(), 0);
 
 cleanup();
 }






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205431] trunk/Source/WebKit2

[carlosgc]

Title: [205431] trunk/Source/WebKit2








Revision 205431
Author carlo...@webkit.org
Date 2016-09-05 00:57:57 -0700 (Mon, 05 Sep 2016)


Log Message
[GTK] SetNativeSurfaceHandleForCompositing should not be compiled in for wayland build
https://bugs.webkit.org/show_bug.cgi?id=161483

Reviewed by Žan Doberšek.

SetNativeSurfaceHandleForCompositing and DestroyNativeSurfaceHandleForCompositing messages should only be used
for GTK+ port when compiled with OpenGL and X11 enabled and redirected XComposite window disabled. So, we should
use USE(TEXTURE_MAPPER) && PLATFORM(GTK) && PLATFORM(X11) && !USE(REDIRECTED_XCOMPOSITE_WINDOW) consistently.
This fixes the build when building the GTK+ port with Wayland enabled and X11 disabled.

* UIProcess/AcceleratedDrawingAreaProxy.cpp:
(WebKit::AcceleratedDrawingAreaProxy::didUpdateBackingStoreState):
* UIProcess/AcceleratedDrawingAreaProxy.h:
* WebProcess/WebPage/AcceleratedDrawingArea.cpp:
(WebKit::AcceleratedDrawingArea::enterAcceleratedCompositingMode):
* WebProcess/WebPage/AcceleratedDrawingArea.h:
* WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.cpp:
* WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.h:
* WebProcess/WebPage/DrawingArea.h:
* WebProcess/WebPage/DrawingArea.messages.in:
* WebProcess/WebPage/gtk/LayerTreeHostGtk.cpp:
* WebProcess/WebPage/gtk/LayerTreeHostGtk.h:

Modified Paths

trunk/Source/WebKit2/ChangeLog
trunk/Source/WebKit2/UIProcess/AcceleratedDrawingAreaProxy.cpp
trunk/Source/WebKit2/UIProcess/AcceleratedDrawingAreaProxy.h
trunk/Source/WebKit2/WebProcess/WebPage/AcceleratedDrawingArea.cpp
trunk/Source/WebKit2/WebProcess/WebPage/AcceleratedDrawingArea.h
trunk/Source/WebKit2/WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.cpp
trunk/Source/WebKit2/WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.h
trunk/Source/WebKit2/WebProcess/WebPage/DrawingArea.h
trunk/Source/WebKit2/WebProcess/WebPage/DrawingArea.messages.in
trunk/Source/WebKit2/WebProcess/WebPage/gtk/LayerTreeHostGtk.cpp
trunk/Source/WebKit2/WebProcess/WebPage/gtk/LayerTreeHostGtk.h




Diff

Modified: trunk/Source/WebKit2/ChangeLog (205430 => 205431)

--- trunk/Source/WebKit2/ChangeLog	2016-09-05 07:15:50 UTC (rev 205430)
+++ trunk/Source/WebKit2/ChangeLog	2016-09-05 07:57:57 UTC (rev 205431)
@@ -1,3 +1,28 @@
+2016-09-05  Carlos Garcia Campos  
+
+[GTK] SetNativeSurfaceHandleForCompositing should not be compiled in for wayland build
+https://bugs.webkit.org/show_bug.cgi?id=161483
+
+Reviewed by Žan Doberšek.
+
+SetNativeSurfaceHandleForCompositing and DestroyNativeSurfaceHandleForCompositing messages should only be used
+for GTK+ port when compiled with OpenGL and X11 enabled and redirected XComposite window disabled. So, we should
+use USE(TEXTURE_MAPPER) && PLATFORM(GTK) && PLATFORM(X11) && !USE(REDIRECTED_XCOMPOSITE_WINDOW) consistently.
+This fixes the build when building the GTK+ port with Wayland enabled and X11 disabled.
+
+* UIProcess/AcceleratedDrawingAreaProxy.cpp:
+(WebKit::AcceleratedDrawingAreaProxy::didUpdateBackingStoreState):
+* UIProcess/AcceleratedDrawingAreaProxy.h:
+* WebProcess/WebPage/AcceleratedDrawingArea.cpp:
+(WebKit::AcceleratedDrawingArea::enterAcceleratedCompositingMode):
+* WebProcess/WebPage/AcceleratedDrawingArea.h:
+* WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.cpp:
+* WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.h:
+* WebProcess/WebPage/DrawingArea.h:
+* WebProcess/WebPage/DrawingArea.messages.in:
+* WebProcess/WebPage/gtk/LayerTreeHostGtk.cpp:
+* WebProcess/WebPage/gtk/LayerTreeHostGtk.h:
+
 2016-09-05  Zan Dobersek  
 
 [GObject bindings] override handleEvent() method in GObjectEventListener


Modified: trunk/Source/WebKit2/UIProcess/AcceleratedDrawingAreaProxy.cpp (205430 => 205431)

--- trunk/Source/WebKit2/UIProcess/AcceleratedDrawingAreaProxy.cpp	2016-09-05 07:15:50 UTC (rev 205430)
+++ trunk/Source/WebKit2/UIProcess/AcceleratedDrawingAreaProxy.cpp	2016-09-05 07:57:57 UTC (rev 205431)
@@ -139,7 +139,7 @@
 else {
 m_hasReceivedFirstUpdate = true;
 
-#if USE(TEXTURE_MAPPER) && PLATFORM(GTK) && !USE(REDIRECTED_XCOMPOSITE_WINDOW)
+#if USE(TEXTURE_MAPPER) && PLATFORM(GTK) && PLATFORM(X11) && !USE(REDIRECTED_XCOMPOSITE_WINDOW)
 if (m_pendingNativeSurfaceHandleForCompositing) {
 setNativeSurfaceHandleForCompositing(m_pendingNativeSurfaceHandleForCompositing);
 m_pendingNativeSurfaceHandleForCompositing = 0;
@@ -261,7 +261,7 @@
 m_webPageProxy.updateAcceleratedCompositingMode(layerTreeContext);
 }
 
-#if USE(TEXTURE_MAPPER) && PLATFORM(GTK) && !USE(REDIRECTED_XCOMPOSITE_WINDOW)
+#if USE(TEXTURE_MAPPER) && PLATFORM(GTK) && PLATFORM(X11) && 

[webkit-changes] [205430] trunk/Source/WebCore

[zandobersek]

Title: [205430] trunk/Source/WebCore








Revision 205430
Author zandober...@gmail.com
Date 2016-09-05 00:15:50 -0700 (Mon, 05 Sep 2016)


Log Message
MediaPlayerPrivateGStreamerBase: improve build guards in nativeImageForCurrentTime()
https://bugs.webkit.org/show_bug.cgi?id=161594

Reviewed by Philippe Normand.

Guard the uses of cairo-gl API in nativeImageForCurrentTime() with the
USE(CAIRO) && ENABLE(ACCELERATED_2D_CANVAS) pair of build guards. This
API is only made available when the ACCELERATED_2D_CANVAS option is
enabled. Placing the guards this way thus avoids compilation errors
when compiling USE_GSTREAMER_GL code without the accelerated 2D canvas
support enabled.

* platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:
(WebCore::MediaPlayerPrivateGStreamerBase::nativeImageForCurrentTime):

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (205429 => 205430)

--- trunk/Source/WebCore/ChangeLog	2016-09-05 07:14:41 UTC (rev 205429)
+++ trunk/Source/WebCore/ChangeLog	2016-09-05 07:15:50 UTC (rev 205430)
@@ -1,3 +1,20 @@
+2016-09-05  Zan Dobersek  
+
+MediaPlayerPrivateGStreamerBase: improve build guards in nativeImageForCurrentTime()
+https://bugs.webkit.org/show_bug.cgi?id=161594
+
+Reviewed by Philippe Normand.
+
+Guard the uses of cairo-gl API in nativeImageForCurrentTime() with the
+USE(CAIRO) && ENABLE(ACCELERATED_2D_CANVAS) pair of build guards. This
+API is only made available when the ACCELERATED_2D_CANVAS option is
+enabled. Placing the guards this way thus avoids compilation errors
+when compiling USE_GSTREAMER_GL code without the accelerated 2D canvas
+support enabled.
+
+* platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:
+(WebCore::MediaPlayerPrivateGStreamerBase::nativeImageForCurrentTime):
+
 2016-09-04  Sam Weinig  
 
 Remove the CREATE_DOM_WRAPPER macro


Modified: trunk/Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp (205429 => 205430)

--- trunk/Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp	2016-09-05 07:14:41 UTC (rev 205429)
+++ trunk/Source/WebCore/platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp	2016-09-05 07:15:50 UTC (rev 205430)
@@ -682,10 +682,7 @@
 #if USE(GSTREAMER_GL)
 NativeImagePtr MediaPlayerPrivateGStreamerBase::nativeImageForCurrentTime()
 {
-#if !USE(CAIRO) || !ENABLE(ACCELERATED_2D_CANVAS)
-return nullptr;
-#endif
-
+#if USE(CAIRO) && ENABLE(ACCELERATED_2D_CANVAS)
 if (m_usingFallbackVideoSink)
 return nullptr;
 
@@ -744,6 +741,9 @@
 gst_video_frame_unmap();
 
 return rotatedSurface;
+#else
+return nullptr;
+#endif
 }
 #endif
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205429] trunk/Source/WebKit2

[zandobersek]

Title: [205429] trunk/Source/WebKit2








Revision 205429
Author zandober...@gmail.com
Date 2016-09-05 00:14:41 -0700 (Mon, 05 Sep 2016)


Log Message
[GObject bindings] override handleEvent() method in GObjectEventListener
https://bugs.webkit.org/show_bug.cgi?id=161593

Reviewed by Carlos Garcia Campos.

* WebProcess/InjectedBundle/API/gtk/DOM/GObjectEventListener.h: Override
the handleMethod() which is inherited from the WebCore EventListener class.
Suppresses the -Winconsistent-missing-override warning that Clang propagates.

Modified Paths

trunk/Source/WebKit2/ChangeLog
trunk/Source/WebKit2/WebProcess/InjectedBundle/API/gtk/DOM/GObjectEventListener.h




Diff

Modified: trunk/Source/WebKit2/ChangeLog (205428 => 205429)

--- trunk/Source/WebKit2/ChangeLog	2016-09-05 06:49:33 UTC (rev 205428)
+++ trunk/Source/WebKit2/ChangeLog	2016-09-05 07:14:41 UTC (rev 205429)
@@ -1,3 +1,14 @@
+2016-09-05  Zan Dobersek  
+
+[GObject bindings] override handleEvent() method in GObjectEventListener
+https://bugs.webkit.org/show_bug.cgi?id=161593
+
+Reviewed by Carlos Garcia Campos.
+
+* WebProcess/InjectedBundle/API/gtk/DOM/GObjectEventListener.h: Override
+the handleMethod() which is inherited from the WebCore EventListener class.
+Suppresses the -Winconsistent-missing-override warning that Clang propagates.
+
 2016-09-04  Joseph Pecoraro  
 
 Remove unnecessary RetainPtr.


Modified: trunk/Source/WebKit2/WebProcess/InjectedBundle/API/gtk/DOM/GObjectEventListener.h (205428 => 205429)

--- trunk/Source/WebKit2/WebProcess/InjectedBundle/API/gtk/DOM/GObjectEventListener.h	2016-09-05 06:49:33 UTC (rev 205428)
+++ trunk/Source/WebKit2/WebProcess/InjectedBundle/API/gtk/DOM/GObjectEventListener.h	2016-09-05 07:14:41 UTC (rev 205429)
@@ -63,7 +63,7 @@
 ~GObjectEventListener();
 void gobjectDestroyed();
 
-virtual void handleEvent(WebCore::ScriptExecutionContext*, WebCore::Event*);
+void handleEvent(WebCore::ScriptExecutionContext*, WebCore::Event*) override;
 
 GObject* m_target;
 // We do not need to keep a reference to the m_coreTarget, because






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205428] trunk/LayoutTests

[commit-queue]

Title: [205428] trunk/LayoutTests








Revision 205428
Author commit-qu...@webkit.org
Date 2016-09-04 23:49:33 -0700 (Sun, 04 Sep 2016)


Log Message
http/tests/fetch/clone-response-body.html uses an incorrect URL
https://bugs.webkit.org/show_bug.cgi?id=161382

Patch by Youenn Fablet  on 2016-09-04
Reviewed by Darin Adler.

* http/tests/fetch/clone-response-body.html: Fixing url, increasing delay to ensure non-flakiness.
Adding assertions to ensure load is successful.

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/http/tests/fetch/clone-response-body.html




Diff

Modified: trunk/LayoutTests/ChangeLog (205427 => 205428)

--- trunk/LayoutTests/ChangeLog	2016-09-05 06:41:59 UTC (rev 205427)
+++ trunk/LayoutTests/ChangeLog	2016-09-05 06:49:33 UTC (rev 205428)
@@ -1,3 +1,13 @@
+2016-09-04  Youenn Fablet  
+
+http/tests/fetch/clone-response-body.html uses an incorrect URL
+https://bugs.webkit.org/show_bug.cgi?id=161382
+
+Reviewed by Darin Adler.
+
+* http/tests/fetch/clone-response-body.html: Fixing url, increasing delay to ensure non-flakiness.
+Adding assertions to ensure load is successful.
+
 2016-09-04  Michael Catanzaro  
 
 Miscellaneous unreviewed GTK test gardening


Modified: trunk/LayoutTests/http/tests/fetch/clone-response-body.html (205427 => 205428)

--- trunk/LayoutTests/http/tests/fetch/clone-response-body.html	2016-09-05 06:41:59 UTC (rev 205427)
+++ trunk/LayoutTests/http/tests/fetch/clone-response-body.html	2016-09-05 06:49:33 UTC (rev 205428)
@@ -11,6 +11,7 @@
 promise_test(function(t) {
 var clone;
 return fetch('/resources/testharnessreport.js').then(function(res) {
+assert_equals(res.status, 200);
 var resolveFunction;
 var promise = new Promise((resolve) => {
 resolveFunction = resolve;
@@ -33,7 +34,8 @@
 
 promise_test(function(t) {
 var clone;
-return fetch('/download-json-with-delay.php?iteration=2=1').then(function(res) {
+return fetch('/resources/download-json-with-delay.php?iteration=2=100').then(function(res) {
+assert_equals(res.status, 200);
 // All data should not be enqueued yet.
 clone = res.clone();
 var string;






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes

[webkit-changes] [205427] trunk/Source

[joepeck]

Title: [205427] trunk/Source








Revision 205427
Author joep...@webkit.org
Date 2016-09-04 23:41:59 -0700 (Sun, 04 Sep 2016)


Log Message
Remove unnecessary RetainPtr.

Review comments on earlier change.

Source/WebKit/mac:

* WebView/WebImmediateActionController.mm:
(-[WebImmediateActionController _defaultAnimationController]):

Source/WebKit2:

* UIProcess/mac/WKImmediateActionController.mm:
(-[WKImmediateActionController _defaultAnimationController]):

Modified Paths

trunk/Source/WebKit/mac/ChangeLog
trunk/Source/WebKit/mac/WebView/WebImmediateActionController.mm
trunk/Source/WebKit2/ChangeLog
trunk/Source/WebKit2/UIProcess/mac/WKImmediateActionController.mm




Diff

Modified: trunk/Source/WebKit/mac/ChangeLog (205426 => 205427)

--- trunk/Source/WebKit/mac/ChangeLog	2016-09-05 05:02:38 UTC (rev 205426)
+++ trunk/Source/WebKit/mac/ChangeLog	2016-09-05 06:41:59 UTC (rev 205427)
@@ -1,3 +1,12 @@
+2016-09-04  Joseph Pecoraro  
+
+Remove unnecessary RetainPtr.
+
+Review comments on earlier change.
+
+* WebView/WebImmediateActionController.mm:
+(-[WebImmediateActionController _defaultAnimationController]):
+
 2016-09-03  Darin Adler  
 
 Streamline DOMImplementation, and move it to our new DOM exception system


Modified: trunk/Source/WebKit/mac/WebView/WebImmediateActionController.mm (205426 => 205427)

--- trunk/Source/WebKit/mac/WebView/WebImmediateActionController.mm	2016-09-05 05:02:38 UTC (rev 205426)
+++ trunk/Source/WebKit/mac/WebView/WebImmediateActionController.mm	2016-09-05 06:41:59 UTC (rev 205427)
@@ -255,10 +255,8 @@
 
 - (id )_defaultAnimationController
 {
-if (_contentPreventsDefault) {
-RetainPtr dummyController = adoptNS([[WebAnimationController alloc] init]);
-return dummyController.autorelease();
-}
+if (_contentPreventsDefault)
+return [[[WebAnimationController alloc] init] autorelease];
 
 NSURL *url = ""
 NSString *absoluteURLString = [url absoluteString];


Modified: trunk/Source/WebKit2/ChangeLog (205426 => 205427)

--- trunk/Source/WebKit2/ChangeLog	2016-09-05 05:02:38 UTC (rev 205426)
+++ trunk/Source/WebKit2/ChangeLog	2016-09-05 06:41:59 UTC (rev 205427)
@@ -1,3 +1,12 @@
+2016-09-04  Joseph Pecoraro  
+
+Remove unnecessary RetainPtr.
+
+Review comments on earlier change.
+
+* UIProcess/mac/WKImmediateActionController.mm:
+(-[WKImmediateActionController _defaultAnimationController]):
+
 2016-09-04  Dan Bernstein  
 
 [Cocoa] Wrapping XPC service execution in os_activity is no longer needed


Modified: trunk/Source/WebKit2/UIProcess/mac/WKImmediateActionController.mm (205426 => 205427)

--- trunk/Source/WebKit2/UIProcess/mac/WKImmediateActionController.mm	2016-09-05 05:02:38 UTC (rev 205426)
+++ trunk/Source/WebKit2/UIProcess/mac/WKImmediateActionController.mm	2016-09-05 06:41:59 UTC (rev 205427)
@@ -269,10 +269,8 @@
 
 - (id )_defaultAnimationController
 {
-if (_contentPreventsDefault) {
-RetainPtr dummyController = adoptNS([[WKAnimationController alloc] init]);
-return dummyController.autorelease();
-}
+if (_contentPreventsDefault)
+return [[[WKAnimationController alloc] init] autorelease];
 
 RefPtr hitTestResult = [self _webHitTestResult];
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes