[webkit-changes] [230472] trunk/LayoutTests

[graouts]

Title: [230472] trunk/LayoutTests








Revision 230472
Author grao...@webkit.org
Date 2018-04-09 22:45:41 -0700 (Mon, 09 Apr 2018)


Log Message
Attempt to fix some test reported test failures in the new legacy-animation-engine directory.

Unreviewed test gardening.

* platform/win/TestExpectations:
* platform/wpe/TestExpectations:

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/win/TestExpectations
trunk/LayoutTests/platform/wpe/TestExpectations




Diff

Modified: trunk/LayoutTests/ChangeLog (230471 => 230472)

--- trunk/LayoutTests/ChangeLog	2018-04-10 05:42:12 UTC (rev 230471)
+++ trunk/LayoutTests/ChangeLog	2018-04-10 05:45:41 UTC (rev 230472)
@@ -1,5 +1,14 @@
 2018-04-09  Antoine Quint  
 
+Attempt to fix some test reported test failures in the new legacy-animation-engine directory. 
+
+Unreviewed test gardening.
+
+* platform/win/TestExpectations:
+* platform/wpe/TestExpectations:
+
+2018-04-09  Antoine Quint  
+
 [Web Animations] Duplicate layout tests exercising CSS Animations and CSS Transitions in a dedicated legacy directory
 https://bugs.webkit.org/show_bug.cgi?id=184420
 


Modified: trunk/LayoutTests/platform/win/TestExpectations (230471 => 230472)

--- trunk/LayoutTests/platform/win/TestExpectations	2018-04-10 05:42:12 UTC (rev 230471)
+++ trunk/LayoutTests/platform/win/TestExpectations	2018-04-10 05:45:41 UTC (rev 230472)
@@ -114,6 +114,7 @@
 webkit.org/b/74716 css3/filters/filter-is-on-subpixel-position.html [ Skip ]
 webkit.org/b/74716 css3/masking/clip-path-filter.html [ Skip ]
 webkit.org/b/74716 css3/filters/backdrop [ Skip ]
+webkit.org/b/74716 legacy-animation-engine/css3/filters/backdrop [ Skip ]
 
 # TODO Should be moved to platform/mac 
 http/tests/misc/willCacheResponse-delegate-callback.html [ Skip ]
@@ -371,6 +372,7 @@
 
 # TODO Full Screen support is not yet enabled.
 fullscreen/ [ Skip ]
+legacy-animation-engine/fullscreen/ [ Skip ]
 
 # viewport meta tag support
 fast/viewport/ [ Skip ]
@@ -2610,6 +2612,7 @@
 
 # Skip tests that fail only in Debug mode 
 [ Debug ] compositing [ Skip ]
+[ Debug ] legacy-animation-engine/compositing [ Skip ]
 [ Debug ] transforms/3d/hit-testing [ Skip ]
 [ Debug ] transforms/3d/point-mapping [ Skip ]
 


Modified: trunk/LayoutTests/platform/wpe/TestExpectations (230471 => 230472)

--- trunk/LayoutTests/platform/wpe/TestExpectations	2018-04-10 05:42:12 UTC (rev 230471)
+++ trunk/LayoutTests/platform/wpe/TestExpectations	2018-04-10 05:45:41 UTC (rev 230472)
@@ -1,4 +1,5 @@
 Bug(WPE) compositing/ [ Skip ]
+Bug(WPE) legacy-animation-engine/compositing/ [ Skip ]
 Bug(WPE) editing/ [ Skip ]
 Bug(WPE) fonts/ [ Skip ]
 Bug(WPE) fullscreen/ [ Skip ]






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

[webkit-changes] [230470] trunk/Source/WebKit

[wenson_hsieh]

Title: [230470] trunk/Source/WebKit








Revision 230470
Author wenson_hs...@apple.com
Date 2018-04-09 22:06:43 -0700 (Mon, 09 Apr 2018)


Log Message
Add missing availability macros after r230462
https://bugs.webkit.org/show_bug.cgi?id=184426

Reviewed by Timothy Hatcher.

Annotate new SPI added in r230462 with the appropriate availability macros.

* UIProcess/API/Cocoa/WKViewPrivate.h:
* UIProcess/API/Cocoa/WKWebViewPrivate.h:

Modified Paths

trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/API/Cocoa/WKViewPrivate.h
trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebViewPrivate.h




Diff

Modified: trunk/Source/WebKit/ChangeLog (230469 => 230470)

--- trunk/Source/WebKit/ChangeLog	2018-04-10 05:06:37 UTC (rev 230469)
+++ trunk/Source/WebKit/ChangeLog	2018-04-10 05:06:43 UTC (rev 230470)
@@ -1,3 +1,15 @@
+2018-04-09  Wenson Hsieh  
+
+Add missing availability macros after r230462
+https://bugs.webkit.org/show_bug.cgi?id=184426
+
+Reviewed by Timothy Hatcher.
+
+Annotate new SPI added in r230462 with the appropriate availability macros.
+
+* UIProcess/API/Cocoa/WKViewPrivate.h:
+* UIProcess/API/Cocoa/WKWebViewPrivate.h:
+
 2018-04-09  Brent Fulgham  
 
 Add ProcessPrivilege assertions to places that access NSApp


Modified: trunk/Source/WebKit/UIProcess/API/Cocoa/WKViewPrivate.h (230469 => 230470)

--- trunk/Source/WebKit/UIProcess/API/Cocoa/WKViewPrivate.h	2018-04-10 05:06:37 UTC (rev 230469)
+++ trunk/Source/WebKit/UIProcess/API/Cocoa/WKViewPrivate.h	2018-04-10 05:06:43 UTC (rev 230470)
@@ -82,7 +82,7 @@
 @property (readonly) NSColor *_pageExtendedBackgroundColor;
 @property (copy, nonatomic) NSColor *underlayColor;
 
-@property (nonatomic, setter=_setBackgroundColor:) NSColor *_backgroundColor;
+@property (nonatomic, setter=_setBackgroundColor:) NSColor *_backgroundColor WK_API_AVAILABLE(macosx(WK_MAC_TBA));
 
 #if WK_API_ENABLED
 @property (strong, nonatomic, setter=_setInspectorAttachmentView:) NSView *_inspectorAttachmentView WK_API_AVAILABLE(macosx(10.11));


Modified: trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebViewPrivate.h (230469 => 230470)

--- trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebViewPrivate.h	2018-04-10 05:06:37 UTC (rev 230469)
+++ trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebViewPrivate.h	2018-04-10 05:06:43 UTC (rev 230470)
@@ -260,7 +260,7 @@
 @property (nonatomic, setter=_setRubberBandingEnabled:) _WKRectEdge _rubberBandingEnabled WK_API_AVAILABLE(macosx(10.13.4));
 
 @property (readonly) NSColor *_pageExtendedBackgroundColor;
-@property (nonatomic, setter=_setBackgroundColor:) NSColor *_backgroundColor;
+@property (nonatomic, setter=_setBackgroundColor:) NSColor *_backgroundColor WK_API_AVAILABLE(macosx(WK_MAC_TBA));
 @property (nonatomic, setter=_setDrawsBackground:) BOOL _drawsBackground;
 @property (nonatomic, setter=_setTopContentInset:) CGFloat _topContentInset;
 






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

[webkit-changes] [230469] trunk/Tools

[zalan]

Title: [230469] trunk/Tools








Revision 230469
Author za...@apple.com
Date 2018-04-09 22:06:37 -0700 (Mon, 09 Apr 2018)


Log Message
[LayoutReloaded] Add support for relatively positioned containers in inline formatting context.
https://bugs.webkit.org/show_bug.cgi?id=184439

Reviewed by Antti Koivisto.

* LayoutReloaded/FormattingContext/BlockFormatting/BlockFormattingContext.js:
(BlockFormattingContext.prototype._placeInFlowPositionedChildren): Deleted.
(BlockFormattingContext.prototype._computeInFlowPositionedPosition): Deleted.
* LayoutReloaded/FormattingContext/FormattingContext.js:
(FormattingContext.prototype._computeInFlowPositionedPosition):
(FormattingContext.prototype._placeInFlowPositionedChildren):
* LayoutReloaded/FormattingContext/InlineFormatting/InlineFormattingContext.js:
(InlineFormattingContext.prototype._placeInFlowPositionedChildren): Deleted.
* LayoutReloaded/Utils.js:
(Utils._dumpBox):
(Utils._dumpLines.):
(Utils._dumpLines):
(Utils.precisionRoundWithDecimals):
(Utils.precisionRound):
(Utils):
* LayoutReloaded/test/index.html:
* LayoutReloaded/test/inline-with-relative-positioning.html: Added.

Modified Paths

trunk/Tools/ChangeLog
trunk/Tools/LayoutReloaded/FormattingContext/BlockFormatting/BlockFormattingContext.js
trunk/Tools/LayoutReloaded/FormattingContext/FormattingContext.js
trunk/Tools/LayoutReloaded/FormattingContext/InlineFormatting/InlineFormattingContext.js
trunk/Tools/LayoutReloaded/Utils.js
trunk/Tools/LayoutReloaded/test/index.html


Added Paths

trunk/Tools/LayoutReloaded/test/inline-with-relative-positioning.html




Diff

Modified: trunk/Tools/ChangeLog (230468 => 230469)

--- trunk/Tools/ChangeLog	2018-04-10 03:50:17 UTC (rev 230468)
+++ trunk/Tools/ChangeLog	2018-04-10 05:06:37 UTC (rev 230469)
@@ -1,5 +1,30 @@
 2018-04-09  Zalan Bujtas  
 
+[LayoutReloaded] Add support for relatively positioned containers in inline formatting context.
+https://bugs.webkit.org/show_bug.cgi?id=184439
+
+Reviewed by Antti Koivisto.
+
+* LayoutReloaded/FormattingContext/BlockFormatting/BlockFormattingContext.js:
+(BlockFormattingContext.prototype._placeInFlowPositionedChildren): Deleted.
+(BlockFormattingContext.prototype._computeInFlowPositionedPosition): Deleted.
+* LayoutReloaded/FormattingContext/FormattingContext.js:
+(FormattingContext.prototype._computeInFlowPositionedPosition):
+(FormattingContext.prototype._placeInFlowPositionedChildren):
+* LayoutReloaded/FormattingContext/InlineFormatting/InlineFormattingContext.js:
+(InlineFormattingContext.prototype._placeInFlowPositionedChildren): Deleted.
+* LayoutReloaded/Utils.js:
+(Utils._dumpBox):
+(Utils._dumpLines.):
+(Utils._dumpLines):
+(Utils.precisionRoundWithDecimals):
+(Utils.precisionRound):
+(Utils):
+* LayoutReloaded/test/index.html:
+* LayoutReloaded/test/inline-with-relative-positioning.html: Added.
+
+2018-04-09  Zalan Bujtas  
+
 [LayoutReloaded] Remove Layout.InitialContainingBlock class
 https://bugs.webkit.org/show_bug.cgi?id=184436
 


Modified: trunk/Tools/LayoutReloaded/FormattingContext/BlockFormatting/BlockFormattingContext.js (230468 => 230469)

--- trunk/Tools/LayoutReloaded/FormattingContext/BlockFormatting/BlockFormattingContext.js	2018-04-10 03:50:17 UTC (rev 230468)
+++ trunk/Tools/LayoutReloaded/FormattingContext/BlockFormatting/BlockFormattingContext.js	2018-04-10 05:06:37 UTC (rev 230469)
@@ -118,20 +118,6 @@
 this.displayBox(layoutBox).setTopLeft(position);
 }
 
-_placeInFlowPositionedChildren(container) {
-if (!container.isContainer())
-return;
-// If this layoutBox also establishes a formatting context, then positioning already has happend at the formatting context.
-if (container.establishesFormattingContext() && container != this.formattingRoot())
-return;
-ASSERT(container.isContainer());
-for (let inFlowChild = container.firstInFlowChild(); inFlowChild; inFlowChild = inFlowChild.nextInFlowSibling()) {
-if (!inFlowChild.isInFlowPositioned())
-continue;
-this._computeInFlowPositionedPosition(inFlowChild);
-}
-}
-
 _layoutOutOfFlowDescendants() {
 // This lays out all the out-of-flow boxes that belong to this formatting context even if
 // the root container is not the containing block.
@@ -279,23 +265,6 @@
 return bottom;
 }
 
-_computeInFlowPositionedPosition(layoutBox) {
-// Start with the original, static position.
-let displayBox = this.displayBox(layoutBox);
-let relativePosition = displayBox.topLeft();
-// Top/bottom
-if (!Utils.isTopAuto(layoutBox))
-relativePosition.shiftTop(Utils.top(layoutBox));
-else if (!Utils.isBottomAuto(layoutBox))
-   

[webkit-changes] [230468] trunk/Source

[bfulgham]

Title: [230468] trunk/Source








Revision 230468
Author bfulg...@apple.com
Date 2018-04-09 20:50:17 -0700 (Mon, 09 Apr 2018)


Log Message
Add ProcessPrivilege assertions to places that access NSApp
https://bugs.webkit.org/show_bug.cgi?id=184322


Reviewed by Per Arne Vollan.

Add ProcessPrivilege assertions to places where we interact with NSApp so
that we can prevent accidentally using them in the WebContent process.

Source/WebCore:

* page/mac/EventHandlerMac.mm:
(WebCore::lastEventIsMouseUp):
(WebCore::EventHandler::sendFakeEventsAfterWidgetTracking):
* platform/mac/EventLoopMac.mm:
(WebCore::EventLoop::cycle):
* platform/mac/PasteboardMac.mm:
(WebCore::Pasteboard::setDragImage):

Source/WebKit:

* Shared/mac/ChildProcessMac.mm:
(WebKit::ChildProcess::stopNSAppRunLoop):
* Shared/mac/HangDetectionDisablerMac.mm:
(WebKit::setClientsMayIgnoreEvents):
* UIProcess/Cocoa/WebProcessPoolCocoa.mm:
(WebKit::WebProcessPool::platformInitializeWebProcess):
* UIProcess/Cocoa/WebViewImpl.mm:
(WebKit::WebViewImpl::WebViewImpl):
(WebKit::WebViewImpl::becomeFirstResponder):
(WebKit::WebViewImpl::pluginFocusOrWindowFocusChanged):
(WebKit::WebViewImpl::validateUserInterfaceItem):
(WebKit::WebViewImpl::startSpeaking):
(WebKit::WebViewImpl::stopSpeaking):
(WebKit::applicationFlagsForDrag):
(WebKit::WebViewImpl::doneWithKeyEvent):
* UIProcess/Gamepad/mac/UIGamepadProviderMac.mm:
(WebKit::UIGamepadProvider::platformWebPageProxyForGamepadInput):
* UIProcess/Plugins/mac/PluginProcessProxyMac.mm:
(WebKit::PluginProcessProxy::enterFullscreen):
(WebKit::PluginProcessProxy::beginModal):
(WebKit::PluginProcessProxy::endModal):
* UIProcess/mac/DisplayLink.cpp:
(WebKit::DisplayLink::DisplayLink):
(WebKit::DisplayLink::~DisplayLink):
* UIProcess/mac/PageClientImplMac.mm:
(WebKit::PageClientImpl::isViewWindowActive):
(WebKit::PageClientImpl::setCursor):
* UIProcess/mac/WebPageProxyMac.mm:
(WebKit::WebPageProxy::getIsSpeaking):
(WebKit::WebPageProxy::speak):
(WebKit::WebPageProxy::stopSpeaking):
(WebKit::WebPageProxy::startDisplayLink):
* UIProcess/mac/WebPopupMenuProxyMac.mm:
(WebKit::WebPopupMenuProxyMac::showPopupMenu):

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/page/mac/EventHandlerMac.mm
trunk/Source/WebCore/platform/mac/EventLoopMac.mm
trunk/Source/WebCore/platform/mac/PasteboardMac.mm
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/Shared/mac/ChildProcessMac.mm
trunk/Source/WebKit/Shared/mac/HangDetectionDisablerMac.mm
trunk/Source/WebKit/UIProcess/Cocoa/WebProcessPoolCocoa.mm
trunk/Source/WebKit/UIProcess/Cocoa/WebViewImpl.mm
trunk/Source/WebKit/UIProcess/Gamepad/mac/UIGamepadProviderMac.mm
trunk/Source/WebKit/UIProcess/Plugins/mac/PluginProcessProxyMac.mm
trunk/Source/WebKit/UIProcess/mac/DisplayLink.cpp
trunk/Source/WebKit/UIProcess/mac/PageClientImplMac.mm
trunk/Source/WebKit/UIProcess/mac/WebPageProxyMac.mm
trunk/Source/WebKit/UIProcess/mac/WebPopupMenuProxyMac.mm




Diff

Modified: trunk/Source/WebCore/ChangeLog (230467 => 230468)

--- trunk/Source/WebCore/ChangeLog	2018-04-10 03:43:46 UTC (rev 230467)
+++ trunk/Source/WebCore/ChangeLog	2018-04-10 03:50:17 UTC (rev 230468)
@@ -1,3 +1,22 @@
+2018-04-09  Brent Fulgham  
+
+Add ProcessPrivilege assertions to places that access NSApp
+https://bugs.webkit.org/show_bug.cgi?id=184322
+
+
+Reviewed by Per Arne Vollan.
+
+Add ProcessPrivilege assertions to places where we interact with NSApp so
+that we can prevent accidentally using them in the WebContent process.
+
+* page/mac/EventHandlerMac.mm:
+(WebCore::lastEventIsMouseUp):
+(WebCore::EventHandler::sendFakeEventsAfterWidgetTracking):
+* platform/mac/EventLoopMac.mm:
+(WebCore::EventLoop::cycle):
+* platform/mac/PasteboardMac.mm:
+(WebCore::Pasteboard::setDragImage):
+
 2018-04-09  John Wilander  
 
 Refactor Ignore HSTS code


Modified: trunk/Source/WebCore/page/mac/EventHandlerMac.mm (230467 => 230468)

--- trunk/Source/WebCore/page/mac/EventHandlerMac.mm	2018-04-10 03:43:46 UTC (rev 230467)
+++ trunk/Source/WebCore/page/mac/EventHandlerMac.mm	2018-04-10 03:50:17 UTC (rev 230468)
@@ -69,6 +69,7 @@
 #include 
 #include 
 #include 
+#include 
 
 #if ENABLE(MAC_GESTURE_EVENTS)
 #import 
@@ -200,6 +201,7 @@
 // It's not clear in what cases this is helpful now -- it's possible it can be removed. 
 
 ASSERT([NSApp isRunning]);
+RELEASE_ASSERT(hasProcessPrivilege(ProcessPrivilege::CanCommunicateWithWindowServer));
 
 BEGIN_BLOCK_OBJC_EXCEPTIONS;
 NSEvent *currentEventAfterHandlingMouseDown = [NSApp currentEvent];
@@ -570,6 +572,8 @@
 if (!view)
 return;
 
+RELEASE_ASSERT(hasProcessPrivilege(ProcessPrivilege::CanCommunicateWithWindowServer));
+
 BEGIN_BLOCK_OBJC_EXCEPTIONS;
 
 m_sendingEventToSubview = false;


Modified: trunk/Source/WebCore/platform/mac/EventLoopMac.mm (230467 => 230468)

--- 

[webkit-changes] [230467] trunk/Source

[wilander]

Title: [230467] trunk/Source








Revision 230467
Author wilan...@apple.com
Date 2018-04-09 20:43:46 -0700 (Mon, 09 Apr 2018)


Log Message
Refactor Ignore HSTS code
https://bugs.webkit.org/show_bug.cgi?id=184433


Reviewed by Darin Adler.

This patch refactors our ignore HSTS code. The convenience functions are moved
out of CFNetwork SPI in PAL, and into where they are used. It also switches
from performSelector: calls to straight function calls, after checking that
there is a responder.

Source/WebCore:

* platform/network/mac/WebCoreURLResponse.mm:
(WebCore::schemeWasUpgradedDueToDynamicHSTS):
Add convenience function here since it was moved out of
CFNetworkSPI.h.

Source/WebCore/PAL:

* pal/spi/cf/CFNetworkSPI.h:
(schemeWasUpgradedDueToDynamicHSTS): Deleted.
(setIgnoreHSTS): Deleted.
(ignoreHSTS): Deleted.

Source/WebKit:

* NetworkProcess/cocoa/NetworkSessionCocoa.mm:
(schemeWasUpgradedDueToDynamicHSTS):
(setIgnoreHSTS):
(ignoreHSTS):
Add convenience functions here since they were moved out of
CFNetworkSPI.h.

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/PAL/ChangeLog
trunk/Source/WebCore/PAL/pal/spi/cf/CFNetworkSPI.h
trunk/Source/WebCore/platform/network/mac/WebCoreURLResponse.mm
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/NetworkProcess/cocoa/NetworkSessionCocoa.mm




Diff

Modified: trunk/Source/WebCore/ChangeLog (230466 => 230467)

--- trunk/Source/WebCore/ChangeLog	2018-04-10 03:03:28 UTC (rev 230466)
+++ trunk/Source/WebCore/ChangeLog	2018-04-10 03:43:46 UTC (rev 230467)
@@ -1,3 +1,21 @@
+2018-04-09  John Wilander  
+
+Refactor Ignore HSTS code
+https://bugs.webkit.org/show_bug.cgi?id=184433
+
+
+Reviewed by Darin Adler.
+
+This patch refactors our ignore HSTS code. The convenience functions are moved
+out of CFNetwork SPI in PAL, and into where they are used. It also switches
+from performSelector: calls to straight function calls, after checking that
+there is a responder.
+
+* platform/network/mac/WebCoreURLResponse.mm:
+(WebCore::schemeWasUpgradedDueToDynamicHSTS):
+Add convenience function here since it was moved out of
+CFNetworkSPI.h.
+
 2018-04-09  Brady Eidson  
 
 Expand WebCore policy concept of "shouldContinue" to allow for more than true/false


Modified: trunk/Source/WebCore/PAL/ChangeLog (230466 => 230467)

--- trunk/Source/WebCore/PAL/ChangeLog	2018-04-10 03:03:28 UTC (rev 230466)
+++ trunk/Source/WebCore/PAL/ChangeLog	2018-04-10 03:43:46 UTC (rev 230467)
@@ -1,3 +1,21 @@
+2018-04-09  John Wilander  
+
+Refactor Ignore HSTS code
+https://bugs.webkit.org/show_bug.cgi?id=184433
+
+
+Reviewed by Darin Adler.
+
+This patch refactors our ignore HSTS code. The convenience functions are moved
+out of CFNetwork SPI in PAL, and into where they are used. It also switches
+from performSelector: calls to straight function calls, after checking that
+there is a responder.
+
+* pal/spi/cf/CFNetworkSPI.h:
+(schemeWasUpgradedDueToDynamicHSTS): Deleted.
+(setIgnoreHSTS): Deleted.
+(ignoreHSTS): Deleted.
+
 2018-04-07  Timothy Hatcher  
 
 Use the system's link color when system appearance is desired for a WebView.


Modified: trunk/Source/WebCore/PAL/pal/spi/cf/CFNetworkSPI.h (230466 => 230467)

--- trunk/Source/WebCore/PAL/pal/spi/cf/CFNetworkSPI.h	2018-04-10 03:03:28 UTC (rev 230466)
+++ trunk/Source/WebCore/PAL/pal/spi/cf/CFNetworkSPI.h	2018-04-10 03:43:46 UTC (rev 230467)
@@ -337,24 +337,4 @@
 - (void)_setMIMEType:(NSString *)type;
 @end
 
-static bool schemeWasUpgradedDueToDynamicHSTS(NSURLRequest *request)
-{
-if ([request respondsToSelector:@selector(_schemeWasUpgradedDueToDynamicHSTS)])
-return [request performSelector:@selector(_schemeWasUpgradedDueToDynamicHSTS)];
-return false;
-}
-
-static void setIgnoreHSTS(NSMutableURLRequest *request, bool ignoreHSTS)
-{
-if ([request respondsToSelector:@selector(_setIgnoreHSTS:)])
-[request performSelector:@selector(_setIgnoreHSTS:) withObject:[NSNumber numberWithBool:ignoreHSTS]];
-}
-
-static bool ignoreHSTS(NSURLRequest *request)
-{
-if ([request respondsToSelector:@selector(_ignoreHSTS)])
-return [request performSelector:@selector(_ignoreHSTS)];
-return false;
-}
-
 #endif // defined(__OBJC__)


Modified: trunk/Source/WebCore/platform/network/mac/WebCoreURLResponse.mm (230466 => 230467)

--- trunk/Source/WebCore/platform/network/mac/WebCoreURLResponse.mm	2018-04-10 03:03:28 UTC (rev 230466)
+++ trunk/Source/WebCore/platform/network/mac/WebCoreURLResponse.mm	2018-04-10 03:43:46 UTC (rev 230467)
@@ -327,6 +327,17 @@
 }
 #endif
 
+static bool schemeWasUpgradedDueToDynamicHSTS(NSURLRequest *request)
+{
+#if !USE(CFNETWORK_IGNORE_HSTS)
+UNUSED_PARAM(request);
+

[webkit-changes] [230466] trunk/Tools

[zalan]

Title: [230466] trunk/Tools








Revision 230466
Author za...@apple.com
Date 2018-04-09 20:03:28 -0700 (Mon, 09 Apr 2018)


Log Message
[LayoutReloaded] Remove Layout.InitialContainingBlock class
https://bugs.webkit.org/show_bug.cgi?id=184436

Reviewed by Antti Koivisto.

* LayoutReloaded/LayoutTree/Box.js:
(Layout.Box.prototype.establishesBlockFormattingContext):
* LayoutReloaded/LayoutTree/InitialBlockContainer.js: Removed.
* LayoutReloaded/TreeBuilder.js:
(TreeBuilder.prototype.createTree):
* LayoutReloaded/test/index.html:

Modified Paths

trunk/Tools/ChangeLog
trunk/Tools/LayoutReloaded/LayoutTree/Box.js
trunk/Tools/LayoutReloaded/TreeBuilder.js
trunk/Tools/LayoutReloaded/test/index.html


Removed Paths

trunk/Tools/LayoutReloaded/LayoutTree/InitialBlockContainer.js




Diff

Modified: trunk/Tools/ChangeLog (230465 => 230466)

--- trunk/Tools/ChangeLog	2018-04-10 02:42:27 UTC (rev 230465)
+++ trunk/Tools/ChangeLog	2018-04-10 03:03:28 UTC (rev 230466)
@@ -1,5 +1,19 @@
 2018-04-09  Zalan Bujtas  
 
+[LayoutReloaded] Remove Layout.InitialContainingBlock class
+https://bugs.webkit.org/show_bug.cgi?id=184436
+
+Reviewed by Antti Koivisto.
+
+* LayoutReloaded/LayoutTree/Box.js:
+(Layout.Box.prototype.establishesBlockFormattingContext):
+* LayoutReloaded/LayoutTree/InitialBlockContainer.js: Removed.
+* LayoutReloaded/TreeBuilder.js:
+(TreeBuilder.prototype.createTree):
+* LayoutReloaded/test/index.html:
+
+2018-04-09  Zalan Bujtas  
+
 [LayoutReloaded] Add support for inline-block.
 https://bugs.webkit.org/show_bug.cgi?id=184434
 


Modified: trunk/Tools/LayoutReloaded/LayoutTree/Box.js (230465 => 230466)

--- trunk/Tools/LayoutReloaded/LayoutTree/Box.js	2018-04-10 02:42:27 UTC (rev 230465)
+++ trunk/Tools/LayoutReloaded/LayoutTree/Box.js	2018-04-10 03:03:28 UTC (rev 230466)
@@ -150,6 +150,9 @@
 }
 
 establishesBlockFormattingContext() {
+// Initial Containing Block always creates a new (inital) block formatting context.
+if (!this.parent())
+return true;
 // 9.4.1 Block formatting contexts
 // Floats, absolutely positioned elements, block containers (such as inline-blocks, table-cells, and table-captions)
 // that are not block boxes, and block boxes with 'overflow' other than 'visible' (except when that value has been propagated to the viewport)


Deleted: trunk/Tools/LayoutReloaded/LayoutTree/InitialBlockContainer.js (230465 => 230466)

--- trunk/Tools/LayoutReloaded/LayoutTree/InitialBlockContainer.js	2018-04-10 02:42:27 UTC (rev 230465)
+++ trunk/Tools/LayoutReloaded/LayoutTree/InitialBlockContainer.js	2018-04-10 03:03:28 UTC (rev 230466)
@@ -1,35 +0,0 @@
-/*
- * Copyright (C) 2018 Apple Inc. All Rights Reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- * 1. Redistributions of source code must retain the above copyright
- *notice, this list of conditions and the following disclaimer.
- * 2. Redistributions in binary form must reproduce the above copyright
- *notice, this list of conditions and the following disclaimer in the
- *documentation and/or other materials provided with the distribution.
- *
- * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY
- * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
- * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
- * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL APPLE INC. OR
- * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
- * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
- * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
- * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
- * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
- * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- */
-
-Layout.InitialBlockContainer = class InitialBlockContainer extends Layout.BlockContainer {
-constructor(document, id, name) {
-super(document, id);
-}
-
-establishesBlockFormattingContext() {
-// ICB always creates a new (inital) block formatting context.
-return true;
-}
-}


Modified: trunk/Tools/LayoutReloaded/TreeBuilder.js (230465 => 230466)

--- trunk/Tools/LayoutReloaded/TreeBuilder.js	2018-04-10 02:42:27 UTC (rev 230465)
+++ trunk/Tools/LayoutReloaded/TreeBuilder.js	2018-04-10 03:03:28 UTC (rev 230466)
@@ -28,7 +28,7 @@
 
 createTree(document, renderTreeDump) {
 // Root.
-let initialBlockContainer = new Layout.InitialBlockContainer(document, parseInt(renderTreeDump.substring(0, renderTreeDump.indexOf("(";
+let initialBlockContainer = new 

[webkit-changes] [230465] trunk/Source/JavaScriptCore

[fpizlo]

Title: [230465] trunk/Source/_javascript_Core








Revision 230465
Author fpi...@apple.com
Date 2018-04-09 19:42:27 -0700 (Mon, 09 Apr 2018)


Log Message
Executing known edge types may reveal a contradiction causing us to emit an exit at a node that is not allowed to exit
https://bugs.webkit.org/show_bug.cgi?id=184372

Reviewed by Saam Barati.

We do a pretty good job of not emitting checks for KnownBlah edges, since those mean that we
have already proved, using techniques that are more precise than AI, that the edge has type
Blah. Unfortunately, we do not handle this case gracefully when AI state becomes bottom,
because we have a bad habit of treating terminate/terminateSpeculativeExecution as something
other than a check - so we think we can call those just because we should have already
bailed. It's better to think of them as the result of folding a check. Therefore, we should
only do it if there had been a check to begin with.

* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::fillSpeculateInt32Internal):
(JSC::DFG::SpeculativeJIT::fillSpeculateInt52):
(JSC::DFG::SpeculativeJIT::fillSpeculateDouble):
(JSC::DFG::SpeculativeJIT::fillSpeculateCell):
(JSC::DFG::SpeculativeJIT::fillSpeculateBoolean):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::lowInt32):
(JSC::FTL::DFG::LowerDFGToB3::lowInt52):
(JSC::FTL::DFG::LowerDFGToB3::lowCell):
(JSC::FTL::DFG::LowerDFGToB3::lowBoolean):
(JSC::FTL::DFG::LowerDFGToB3::lowDouble):
(JSC::FTL::DFG::LowerDFGToB3::speculate):
(JSC::FTL::DFG::LowerDFGToB3::speculateCellOrOther):
(JSC::FTL::DFG::LowerDFGToB3::speculateStringOrOther):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/dfg/DFGSpeculativeJIT64.cpp
trunk/Source/_javascript_Core/ftl/FTLLowerDFGToB3.cpp




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (230464 => 230465)

--- trunk/Source/_javascript_Core/ChangeLog	2018-04-10 01:53:00 UTC (rev 230464)
+++ trunk/Source/_javascript_Core/ChangeLog	2018-04-10 02:42:27 UTC (rev 230465)
@@ -1,3 +1,34 @@
+2018-04-09  Filip Pizlo  
+
+Executing known edge types may reveal a contradiction causing us to emit an exit at a node that is not allowed to exit
+https://bugs.webkit.org/show_bug.cgi?id=184372
+
+Reviewed by Saam Barati.
+
+We do a pretty good job of not emitting checks for KnownBlah edges, since those mean that we
+have already proved, using techniques that are more precise than AI, that the edge has type
+Blah. Unfortunately, we do not handle this case gracefully when AI state becomes bottom,
+because we have a bad habit of treating terminate/terminateSpeculativeExecution as something
+other than a check - so we think we can call those just because we should have already
+bailed. It's better to think of them as the result of folding a check. Therefore, we should
+only do it if there had been a check to begin with.
+
+* dfg/DFGSpeculativeJIT64.cpp:
+(JSC::DFG::SpeculativeJIT::fillSpeculateInt32Internal):
+(JSC::DFG::SpeculativeJIT::fillSpeculateInt52):
+(JSC::DFG::SpeculativeJIT::fillSpeculateDouble):
+(JSC::DFG::SpeculativeJIT::fillSpeculateCell):
+(JSC::DFG::SpeculativeJIT::fillSpeculateBoolean):
+* ftl/FTLLowerDFGToB3.cpp:
+(JSC::FTL::DFG::LowerDFGToB3::lowInt32):
+(JSC::FTL::DFG::LowerDFGToB3::lowInt52):
+(JSC::FTL::DFG::LowerDFGToB3::lowCell):
+(JSC::FTL::DFG::LowerDFGToB3::lowBoolean):
+(JSC::FTL::DFG::LowerDFGToB3::lowDouble):
+(JSC::FTL::DFG::LowerDFGToB3::speculate):
+(JSC::FTL::DFG::LowerDFGToB3::speculateCellOrOther):
+(JSC::FTL::DFG::LowerDFGToB3::speculateStringOrOther):
+
 2018-04-08  Yusuke Suzuki  
 
 [JSC] Introduce @putByIdDirectPrivate


Modified: trunk/Source/_javascript_Core/dfg/DFGSpeculativeJIT64.cpp (230464 => 230465)

--- trunk/Source/_javascript_Core/dfg/DFGSpeculativeJIT64.cpp	2018-04-10 01:53:00 UTC (rev 230464)
+++ trunk/Source/_javascript_Core/dfg/DFGSpeculativeJIT64.cpp	2018-04-10 02:42:27 UTC (rev 230465)
@@ -1088,7 +1088,8 @@
 
 m_interpreter.filter(value, SpecInt32Only);
 if (value.isClear()) {
-terminateSpeculativeExecution(Uncountable, JSValueRegs(), 0);
+if (mayHaveTypeCheck(edge.useKind()))
+terminateSpeculativeExecution(Uncountable, JSValueRegs(), 0);
 returnFormat = DataFormatInt32;
 return allocate();
 }
@@ -1232,7 +1233,8 @@
 
 m_interpreter.filter(value, SpecAnyInt);
 if (value.isClear()) {
-terminateSpeculativeExecution(Uncountable, JSValueRegs(), 0);
+if (mayHaveTypeCheck(edge.useKind()))
+terminateSpeculativeExecution(Uncountable, JSValueRegs(), 0);
 return allocate();
 }
 
@@ -1339,7 +1341,8 @@
 info.fillDouble(*m_stream, fpr);
 return fpr;
 }
-  

[webkit-changes] [230464] trunk/Tools

[zalan]

Title: [230464] trunk/Tools








Revision 230464
Author za...@apple.com
Date 2018-04-09 18:53:00 -0700 (Mon, 09 Apr 2018)


Log Message
[LayoutReloaded] Add support for inline-block.
https://bugs.webkit.org/show_bug.cgi?id=184434

Reviewed by Antti Koivisto.

* LayoutReloaded/FormattingContext/InlineFormatting/InlineFormattingContext.js:
(InlineFormattingContext):
(InlineFormattingContext.prototype.layout):
(InlineFormattingContext.prototype._handleInlineContainer):
(InlineFormattingContext.prototype._handleInlineBlockContainer):
(InlineFormattingContext.prototype._handleInlineContent):
* LayoutReloaded/FormattingContext/InlineFormatting/Line.js:
(Line.prototype.addInlineContainerBox):
(Line.prototype.addTextLineBox):
(Line):
* LayoutReloaded/LayoutTree/Box.js:
(Layout.Box.prototype.isInlineBlockBox):
* LayoutReloaded/Utils.js:
(LayoutRect.prototype.growHorizontally):
(Utils.isBlockContainerElement):
(Utils.isInlineBlockElement):
(Utils._dumpLines.):
(Utils._dumpLines):
* LayoutReloaded/test/index.html:
* LayoutReloaded/test/inline-block-with-fixed-width-height.html: Added.

Modified Paths

trunk/Tools/ChangeLog
trunk/Tools/LayoutReloaded/FormattingContext/InlineFormatting/InlineFormattingContext.js
trunk/Tools/LayoutReloaded/FormattingContext/InlineFormatting/Line.js
trunk/Tools/LayoutReloaded/LayoutTree/Box.js
trunk/Tools/LayoutReloaded/Utils.js
trunk/Tools/LayoutReloaded/test/index.html


Added Paths

trunk/Tools/LayoutReloaded/test/inline-block-with-fixed-width-height.html




Diff

Modified: trunk/Tools/ChangeLog (230463 => 230464)

--- trunk/Tools/ChangeLog	2018-04-10 01:42:45 UTC (rev 230463)
+++ trunk/Tools/ChangeLog	2018-04-10 01:53:00 UTC (rev 230464)
@@ -1,3 +1,31 @@
+2018-04-09  Zalan Bujtas  
+
+[LayoutReloaded] Add support for inline-block.
+https://bugs.webkit.org/show_bug.cgi?id=184434
+
+Reviewed by Antti Koivisto.
+
+* LayoutReloaded/FormattingContext/InlineFormatting/InlineFormattingContext.js:
+(InlineFormattingContext):
+(InlineFormattingContext.prototype.layout):
+(InlineFormattingContext.prototype._handleInlineContainer):
+(InlineFormattingContext.prototype._handleInlineBlockContainer):
+(InlineFormattingContext.prototype._handleInlineContent):
+* LayoutReloaded/FormattingContext/InlineFormatting/Line.js:
+(Line.prototype.addInlineContainerBox):
+(Line.prototype.addTextLineBox):
+(Line):
+* LayoutReloaded/LayoutTree/Box.js:
+(Layout.Box.prototype.isInlineBlockBox):
+* LayoutReloaded/Utils.js:
+(LayoutRect.prototype.growHorizontally):
+(Utils.isBlockContainerElement):
+(Utils.isInlineBlockElement):
+(Utils._dumpLines.):
+(Utils._dumpLines):
+* LayoutReloaded/test/index.html:
+* LayoutReloaded/test/inline-block-with-fixed-width-height.html: Added.
+
 2018-04-09  Timothy Hatcher  
 
 Add support for setting a background color on WKWebView and WKView


Modified: trunk/Tools/LayoutReloaded/FormattingContext/InlineFormatting/InlineFormattingContext.js (230463 => 230464)

--- trunk/Tools/LayoutReloaded/FormattingContext/InlineFormatting/InlineFormattingContext.js	2018-04-10 01:42:45 UTC (rev 230463)
+++ trunk/Tools/LayoutReloaded/FormattingContext/InlineFormatting/InlineFormattingContext.js	2018-04-10 01:53:00 UTC (rev 230464)
@@ -27,6 +27,7 @@
 constructor(inlineFormattingState) {
 super(inlineFormattingState);
 ASSERT(this.formattingRoot().isBlockContainerBox());
+this.m_inlineContainerStack = new Array();
 }
 
 layout() {
@@ -39,27 +40,12 @@
 this._addToLayoutQueue(this.formattingRoot().firstInFlowOrFloatChild());
 while (this._descendantNeedsLayout()) {
 let layoutBox = this._nextInLayoutQueue();
-if (layoutBox.isInlineContainer()) {
-if (inlineContainerStack.indexOf(layoutBox) == -1) {
-inlineContainerStack.push(layoutBox);
-this._adjustLineForInlineContainerStart(layoutBox);
-if (layoutBox.establishesFormattingContext())
-this.layoutState().layout(layoutBox);
-else
-this._addToLayoutQueue(layoutBox.firstInFlowOrFloatChild());
-} else {
-inlineContainerStack.pop(layoutBox);
-this._adjustLineForInlineContainerEnd(layoutBox);
-this._removeFromLayoutQueue(layoutBox);
-this._addToLayoutQueue(layoutBox.nextInFlowOrFloatSibling());
-// Place the inflow positioned children.
-this._placeInFlowPositionedChildren(this.formattingRoot());
-}
-continue;
-}
-this._handleInlineContent(layoutBox);
-this._removeFromLayoutQueue(layoutBox);
-

[webkit-changes] [230463] trunk/Source

[mcatanzaro]

Title: [230463] trunk/Source








Revision 230463
Author mcatanz...@igalia.com
Date 2018-04-09 18:42:45 -0700 (Mon, 09 Apr 2018)


Log Message
Rename UNUSED to BUNUSED
https://bugs.webkit.org/show_bug.cgi?id=184093

Reviewed by Yusuke Suzuki.

Source/bmalloc:

* bmalloc/BAssert.h:
* bmalloc/VMAllocate.h:
(bmalloc::vmValidate):
(bmalloc::vmValidatePhysical):

Source/WebKit:

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _getApplicationManifestWithCompletionHandler:]):

Modified Paths

trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm
trunk/Source/bmalloc/ChangeLog
trunk/Source/bmalloc/bmalloc/BAssert.h
trunk/Source/bmalloc/bmalloc/VMAllocate.h




Diff

Modified: trunk/Source/WebKit/ChangeLog (230462 => 230463)

--- trunk/Source/WebKit/ChangeLog	2018-04-10 01:36:29 UTC (rev 230462)
+++ trunk/Source/WebKit/ChangeLog	2018-04-10 01:42:45 UTC (rev 230463)
@@ -1,3 +1,13 @@
+2018-04-09  Michael Catanzaro  
+
+Rename UNUSED to BUNUSED
+https://bugs.webkit.org/show_bug.cgi?id=184093
+
+Reviewed by Yusuke Suzuki.
+
+* UIProcess/API/Cocoa/WKWebView.mm:
+(-[WKWebView _getApplicationManifestWithCompletionHandler:]):
+
 2018-04-09  Timothy Hatcher  
 
 Add support for setting a background color on WKWebView and WKView


Modified: trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm (230462 => 230463)

--- trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm	2018-04-10 01:36:29 UTC (rev 230462)
+++ trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm	2018-04-10 01:42:45 UTC (rev 230463)
@@ -4477,7 +4477,7 @@
 {
 #if ENABLE(APPLICATION_MANIFEST)
 _page->getApplicationManifest([completionHandler = makeBlockPtr(completionHandler)](const std::optional& manifest, WebKit::CallbackBase::Error error) {
-UNUSED(error);
+UNUSED_PARAM(error);
 if (completionHandler) {
 if (manifest) {
 auto apiManifest = API::ApplicationManifest::create(*manifest);


Modified: trunk/Source/bmalloc/ChangeLog (230462 => 230463)

--- trunk/Source/bmalloc/ChangeLog	2018-04-10 01:36:29 UTC (rev 230462)
+++ trunk/Source/bmalloc/ChangeLog	2018-04-10 01:42:45 UTC (rev 230463)
@@ -1,3 +1,15 @@
+2018-04-09  Michael Catanzaro  
+
+Rename UNUSED to BUNUSED
+https://bugs.webkit.org/show_bug.cgi?id=184093
+
+Reviewed by Yusuke Suzuki.
+
+* bmalloc/BAssert.h:
+* bmalloc/VMAllocate.h:
+(bmalloc::vmValidate):
+(bmalloc::vmValidatePhysical):
+
 2018-04-08  Yusuke Suzuki  
 
 Use alignas instead of compiler-specific attributes


Modified: trunk/Source/bmalloc/bmalloc/BAssert.h (230462 => 230463)

--- trunk/Source/bmalloc/bmalloc/BAssert.h	2018-04-10 01:36:29 UTC (rev 230462)
+++ trunk/Source/bmalloc/bmalloc/BAssert.h	2018-04-10 01:42:45 UTC (rev 230463)
@@ -100,7 +100,7 @@
 } while (0)
 #endif
 
-#define UNUSED(x) ((void)x)
+#define BUNUSED(x) ((void)x)
 
 // = Release build =
 


Modified: trunk/Source/bmalloc/bmalloc/VMAllocate.h (230462 => 230463)

--- trunk/Source/bmalloc/bmalloc/VMAllocate.h	2018-04-10 01:36:29 UTC (rev 230462)
+++ trunk/Source/bmalloc/bmalloc/VMAllocate.h	2018-04-10 01:42:45 UTC (rev 230463)
@@ -76,7 +76,7 @@
 
 inline void vmValidate(size_t vmSize)
 {
-UNUSED(vmSize);
+BUNUSED(vmSize);
 BASSERT(vmSize);
 BASSERT(vmSize == roundUpToMultipleOf(vmPageSize(), vmSize));
 }
@@ -85,7 +85,7 @@
 {
 vmValidate(vmSize);
 
-UNUSED(p);
+BUNUSED(p);
 BASSERT(p);
 BASSERT(p == mask(p, ~(vmPageSize() - 1)));
 }
@@ -104,7 +104,7 @@
 
 inline void vmValidatePhysical(size_t vmSize)
 {
-UNUSED(vmSize);
+BUNUSED(vmSize);
 BASSERT(vmSize);
 BASSERT(vmSize == roundUpToMultipleOf(vmPageSizePhysical(), vmSize));
 }
@@ -113,7 +113,7 @@
 {
 vmValidatePhysical(vmSize);
 
-UNUSED(p);
+BUNUSED(p);
 BASSERT(p);
 BASSERT(p == mask(p, ~(vmPageSizePhysical() - 1)));
 }






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

[webkit-changes] [230462] trunk

[timothy]

Title: [230462] trunk








Revision 230462
Author timo...@apple.com
Date 2018-04-09 18:36:29 -0700 (Mon, 09 Apr 2018)


Log Message
Add support for setting a background color on WKWebView and WKView
https://bugs.webkit.org/show_bug.cgi?id=184426

Reviewed by Wenson Hsieh.

Source/WebKit:

* UIProcess/API/Cocoa/WKViewPrivate.h: Added _backgroundColor property.
* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _backgroundColor]): Added. Call through to WebViewImpl.
(-[WKWebView _setBackgroundColor:]): Added. Call through to WebViewImpl.
* UIProcess/API/Cocoa/WKWebViewPrivate.h: Added _backgroundColor property.
* UIProcess/API/mac/WKView.mm:
(-[WKView _backgroundColor]): Added. Call through to WebViewImpl.
(-[WKView _setBackgroundColor:]): Added. Call through to WebViewImpl.
* UIProcess/Cocoa/WebViewImpl.h:
* UIProcess/Cocoa/WebViewImpl.mm:
(WebKit::WebViewImpl::setBackgroundColor): Added.
(WebKit::WebViewImpl::backgroundColor const): Added.
(WebKit::WebViewImpl::updateLayer): Use m_backgroundColor when set.

Tools:

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/mac/BackgroundColor.mm: Added.

Modified Paths

trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/API/Cocoa/WKViewPrivate.h
trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm
trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebViewPrivate.h
trunk/Source/WebKit/UIProcess/API/mac/WKView.mm
trunk/Source/WebKit/UIProcess/Cocoa/WebViewImpl.h
trunk/Source/WebKit/UIProcess/Cocoa/WebViewImpl.mm
trunk/Tools/ChangeLog
trunk/Tools/TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj


Added Paths

trunk/Tools/TestWebKitAPI/Tests/mac/BackgroundColor.mm




Diff

Modified: trunk/Source/WebKit/ChangeLog (230461 => 230462)

--- trunk/Source/WebKit/ChangeLog	2018-04-10 01:33:34 UTC (rev 230461)
+++ trunk/Source/WebKit/ChangeLog	2018-04-10 01:36:29 UTC (rev 230462)
@@ -1,3 +1,24 @@
+2018-04-09  Timothy Hatcher  
+
+Add support for setting a background color on WKWebView and WKView
+https://bugs.webkit.org/show_bug.cgi?id=184426
+
+Reviewed by Wenson Hsieh.
+
+* UIProcess/API/Cocoa/WKViewPrivate.h: Added _backgroundColor property.
+* UIProcess/API/Cocoa/WKWebView.mm:
+(-[WKWebView _backgroundColor]): Added. Call through to WebViewImpl.
+(-[WKWebView _setBackgroundColor:]): Added. Call through to WebViewImpl.
+* UIProcess/API/Cocoa/WKWebViewPrivate.h: Added _backgroundColor property.
+* UIProcess/API/mac/WKView.mm:
+(-[WKView _backgroundColor]): Added. Call through to WebViewImpl.
+(-[WKView _setBackgroundColor:]): Added. Call through to WebViewImpl.
+* UIProcess/Cocoa/WebViewImpl.h:
+* UIProcess/Cocoa/WebViewImpl.mm:
+(WebKit::WebViewImpl::setBackgroundColor): Added.
+(WebKit::WebViewImpl::backgroundColor const): Added.
+(WebKit::WebViewImpl::updateLayer): Use m_backgroundColor when set.
+
 2018-04-09  Yousuke Kimoto  and Fujii Hironori  
 
 [WinCairo] Add WebKit Shared/win event files for wincairo webkit


Modified: trunk/Source/WebKit/UIProcess/API/Cocoa/WKViewPrivate.h (230461 => 230462)

--- trunk/Source/WebKit/UIProcess/API/Cocoa/WKViewPrivate.h	2018-04-10 01:33:34 UTC (rev 230461)
+++ trunk/Source/WebKit/UIProcess/API/Cocoa/WKViewPrivate.h	2018-04-10 01:36:29 UTC (rev 230462)
@@ -82,6 +82,8 @@
 @property (readonly) NSColor *_pageExtendedBackgroundColor;
 @property (copy, nonatomic) NSColor *underlayColor;
 
+@property (nonatomic, setter=_setBackgroundColor:) NSColor *_backgroundColor;
+
 #if WK_API_ENABLED
 @property (strong, nonatomic, setter=_setInspectorAttachmentView:) NSView *_inspectorAttachmentView WK_API_AVAILABLE(macosx(10.11));
 #endif


Modified: trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm (230461 => 230462)

--- trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm	2018-04-10 01:33:34 UTC (rev 230461)
+++ trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm	2018-04-10 01:36:29 UTC (rev 230462)
@@ -5526,6 +5526,16 @@
 _impl->setDrawsBackground(drawsBackground);
 }
 
+- (NSColor *)_backgroundColor
+{
+return _impl->backgroundColor();
+}
+
+- (void)_setBackgroundColor:(NSColor *)backgroundColor
+{
+_impl->setBackgroundColor(backgroundColor);
+}
+
 - (void)_setDrawsTransparentBackground:(BOOL)drawsTransparentBackground
 {
 static BOOL hasLoggedDeprecationWarning;


Modified: trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebViewPrivate.h (230461 => 230462)

--- trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebViewPrivate.h	2018-04-10 01:33:34 UTC (rev 230461)
+++ trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebViewPrivate.h	2018-04-10 01:36:29 UTC (rev 230462)
@@ -260,6 +260,7 @@
 @property (nonatomic, setter=_setRubberBandingEnabled:) _WKRectEdge _rubberBandingEnabled WK_API_AVAILABLE(macosx(10.13.4));
 
 @property (readonly) NSColor *_pageExtendedBackgroundColor;
+@property (nonatomic, 

[webkit-changes] [230460] trunk/Tools

[cturner]

Title: [230460] trunk/Tools








Revision 230460
Author ctur...@igalia.com
Date 2018-04-09 18:25:19 -0700 (Mon, 09 Apr 2018)


Log Message
[WPE] Add some missing install dependencies
https://bugs.webkit.org/show_bug.cgi?id=184383

Reviewed by Michael Catanzaro.

* wpe/install-dependencies:

Modified Paths

trunk/Tools/ChangeLog
trunk/Tools/wpe/install-dependencies




Diff

Modified: trunk/Tools/ChangeLog (230459 => 230460)

--- trunk/Tools/ChangeLog	2018-04-10 01:05:40 UTC (rev 230459)
+++ trunk/Tools/ChangeLog	2018-04-10 01:25:19 UTC (rev 230460)
@@ -1,3 +1,12 @@
+2018-04-09  Charlie Turner  
+
+[WPE] Add some missing install dependencies
+https://bugs.webkit.org/show_bug.cgi?id=184383
+
+Reviewed by Michael Catanzaro.
+
+* wpe/install-dependencies:
+
 2018-04-09  Sihui Liu  
 
 REGRESSION(r229929): localStorage is broken for WebInspector


Modified: trunk/Tools/wpe/install-dependencies (230459 => 230460)

--- trunk/Tools/wpe/install-dependencies	2018-04-10 01:05:40 UTC (rev 230459)
+++ trunk/Tools/wpe/install-dependencies	2018-04-10 01:25:19 UTC (rev 230460)
@@ -107,6 +107,7 @@
 packages="$packages \
 git \
 gsettings-desktop-schemas-dev \
+libegl1-mesa-dev \
 libexpat1-dev \
 libgbm-dev \
 libgles2-mesa-dev \
@@ -124,7 +125,8 @@
 libxcb-xkb-dev \
 libxml-libxml-perl \
 libwayland-dev \
-luajit"
+luajit \
+yasm"
 
 # These are dependencies necessary for using webkit-patch
 packages="$packages \
@@ -215,7 +217,9 @@
 perl-xml-libxml \
 util-linux \
 v4l-utils \
-wayland"
+xorg-xrandr \
+wayland \
+yasm"
 
 # These are dependencies necessary for using webkit-patch
 packages="$packages \
@@ -228,7 +232,7 @@
 }
 
 function installDependenciesWithDnf {
-# These are dependencies necessary for building WebKitGTK+.
+# These are dependencies necessary for building WebKitWPE.
 packages=" \
 autoconf \
 automake \
@@ -254,6 +258,7 @@
 ninja-build \
 patch \
 perl-File-Copy-Recursive \
+perl-JSON-PP \
 perl-Time-HiRes \
 perl-version \
 ruby \
@@ -290,15 +295,18 @@
 libvorbis-devel \
 libvpx-devel \
 libxcb-devel \
+libXrandr-devel \
 luajit \
 luajit-devel \
 mesa-libgbm-devel \
 mesa-libGLES-devel \
+mesa-libEGL-devel \
 opus-devel \
 orc-devel \
 perl-XML-LibXML \
 pulseaudio-libs-devel \
-wayland-devel"
+wayland-devel \
+yasm"
 
 # These are dependencies necessary for using webkit-patch
 packages="$packages






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

[webkit-changes] [230459] trunk

[utatane . tea]

Title: [230459] trunk








Revision 230459
Author utatane@gmail.com
Date 2018-04-09 18:05:40 -0700 (Mon, 09 Apr 2018)


Log Message
[JSC] Introduce @putByIdDirectPrivate
https://bugs.webkit.org/show_bug.cgi?id=184400

Reviewed by Saam Barati.

This patch adds @putByIdDirectPrivate() to use it for builtin JS.
@getByIdDirectPrivate and @putByIdDirectPrivate are pair of intrinsics
accessing to ECMAScript internal fields.

This change removes accidental [[Put]] operation to an object whose [[Prototype]]
has internal fields (not direct properties). By using @getByIdDirectPrivate() and
@putByIdDirectPrivate(), we strongly keep the semantics of the ECMAScript internal
fields that accessing to the internal fields does not traverse prototype chains.

* builtins/ArrayIteratorPrototype.js:
(globalPrivate.arrayIteratorValueNext):
(globalPrivate.arrayIteratorKeyNext):
(globalPrivate.arrayIteratorKeyValueNext):
* builtins/ArrayPrototype.js:
(globalPrivate.createArrayIterator):
* builtins/AsyncFromSyncIteratorPrototype.js:
(globalPrivate.AsyncFromSyncIteratorConstructor):
* builtins/AsyncFunctionPrototype.js:
(globalPrivate.asyncFunctionResume):
* builtins/AsyncGeneratorPrototype.js:
(globalPrivate.asyncGeneratorQueueEnqueue):
(globalPrivate.asyncGeneratorQueueDequeue):
(asyncGeneratorYieldAwaited):
(globalPrivate.asyncGeneratorYield):
(globalPrivate.doAsyncGeneratorBodyCall):
(globalPrivate.asyncGeneratorResumeNext):
* builtins/GeneratorPrototype.js:
(globalPrivate.generatorResume):
* builtins/MapIteratorPrototype.js:
(globalPrivate.mapIteratorNext):
* builtins/MapPrototype.js:
(globalPrivate.createMapIterator):
* builtins/ModuleLoaderPrototype.js:
(forceFulfillPromise):
* builtins/PromiseOperations.js:
(globalPrivate.newHandledRejectedPromise):
(globalPrivate.rejectPromise):
(globalPrivate.fulfillPromise):
(globalPrivate.initializePromise):
* builtins/PromisePrototype.js:
(then):
* builtins/SetIteratorPrototype.js:
(globalPrivate.setIteratorNext):
* builtins/SetPrototype.js:
(globalPrivate.createSetIterator):
* builtins/StringIteratorPrototype.js:
(next):
* bytecode/BytecodeIntrinsicRegistry.h:
* bytecompiler/NodesCodegen.cpp:
(JSC::BytecodeIntrinsicNode::emit_intrinsic_putByIdDirect):
(JSC::BytecodeIntrinsicNode::emit_intrinsic_putByIdDirectPrivate):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/builtins/ArrayIteratorPrototype.js
trunk/Source/_javascript_Core/builtins/ArrayPrototype.js
trunk/Source/_javascript_Core/builtins/AsyncFromSyncIteratorPrototype.js
trunk/Source/_javascript_Core/builtins/AsyncFunctionPrototype.js
trunk/Source/_javascript_Core/builtins/AsyncGeneratorPrototype.js
trunk/Source/_javascript_Core/builtins/GeneratorPrototype.js
trunk/Source/_javascript_Core/builtins/MapIteratorPrototype.js
trunk/Source/_javascript_Core/builtins/MapPrototype.js
trunk/Source/_javascript_Core/builtins/ModuleLoaderPrototype.js
trunk/Source/_javascript_Core/builtins/PromiseOperations.js
trunk/Source/_javascript_Core/builtins/PromisePrototype.js
trunk/Source/_javascript_Core/builtins/SetIteratorPrototype.js
trunk/Source/_javascript_Core/builtins/SetPrototype.js
trunk/Source/_javascript_Core/builtins/StringIteratorPrototype.js
trunk/Source/_javascript_Core/bytecode/BytecodeIntrinsicRegistry.h
trunk/Source/_javascript_Core/bytecompiler/NodesCodegen.cpp


Added Paths

trunk/JSTests/stress/put-by-id-direct.js




Diff

Added: trunk/JSTests/stress/put-by-id-direct.js (0 => 230459)

--- trunk/JSTests/stress/put-by-id-direct.js	(rev 0)
+++ trunk/JSTests/stress/put-by-id-direct.js	2018-04-10 01:05:40 UTC (rev 230459)
@@ -0,0 +1,23 @@
+var createBuiltin = $vm.createBuiltin;
+
+function shouldBe(actual, expected) {
+if (actual !== expected)
+throw new Error('bad value: ' + actual);
+}
+
+(function () {
+var builtin = createBuiltin(`(function (obj, value) {
+return @putByIdDirect(obj, "hello", value);
+})`);
+noInline(builtin);
+
+var setValue = null;
+var object = {
+__proto__: {
+hello: 30
+}
+};
+builtin(object, 42);
+shouldBe(object.hello, 42);
+shouldBe(object.hasOwnProperty("hello"), true);
+}());


Modified: trunk/Source/_javascript_Core/ChangeLog (230458 => 230459)

--- trunk/Source/_javascript_Core/ChangeLog	2018-04-10 00:31:02 UTC (rev 230458)
+++ trunk/Source/_javascript_Core/ChangeLog	2018-04-10 01:05:40 UTC (rev 230459)
@@ -1,3 +1,62 @@
+2018-04-08  Yusuke Suzuki  
+
+[JSC] Introduce @putByIdDirectPrivate
+https://bugs.webkit.org/show_bug.cgi?id=184400
+
+Reviewed by Saam Barati.
+
+This patch adds @putByIdDirectPrivate() to use it for builtin JS.
+@getByIdDirectPrivate and @putByIdDirectPrivate are pair of intrinsics
+accessing to ECMAScript internal fields.
+
+This change removes accidental [[Put]] operation to an object whose [[Prototype]]
+has internal fields (not direct 

[webkit-changes] [230458] trunk/Source/WebCore

[beidson]

Title: [230458] trunk/Source/WebCore








Revision 230458
Author beid...@apple.com
Date 2018-04-09 17:31:02 -0700 (Mon, 09 Apr 2018)


Log Message
Expand WebCore policy concept of "shouldContinue" to allow for more than true/false
https://bugs.webkit.org/show_bug.cgi?id=184424

Reviewed by Alex Christensen.

No new tests (No behavior change, refactor only)

Specifically this expands the "shouldContinue" bool to be an enum class with:
-Yes
-No
-ForSuspension

* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::willSendRequest):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::loadURL):
(WebCore::FrameLoader::load):
(WebCore::FrameLoader::loadWithDocumentLoader):
(WebCore::FrameLoader::loadPostRequest):
(WebCore::FrameLoader::continueLoadAfterNavigationPolicy):
(WebCore::FrameLoader::continueLoadAfterNewWindowPolicy):
* loader/FrameLoader.h:
* loader/PolicyChecker.cpp:
(WebCore::PolicyChecker::checkNavigationPolicy):
(WebCore::PolicyChecker::checkNewWindowPolicy):
* loader/PolicyChecker.h:

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/loader/DocumentLoader.cpp
trunk/Source/WebCore/loader/FrameLoader.cpp
trunk/Source/WebCore/loader/FrameLoader.h
trunk/Source/WebCore/loader/PolicyChecker.cpp
trunk/Source/WebCore/loader/PolicyChecker.h




Diff

Modified: trunk/Source/WebCore/ChangeLog (230457 => 230458)

--- trunk/Source/WebCore/ChangeLog	2018-04-10 00:28:24 UTC (rev 230457)
+++ trunk/Source/WebCore/ChangeLog	2018-04-10 00:31:02 UTC (rev 230458)
@@ -1,3 +1,32 @@
+2018-04-09  Brady Eidson  
+
+Expand WebCore policy concept of "shouldContinue" to allow for more than true/false
+https://bugs.webkit.org/show_bug.cgi?id=184424
+
+Reviewed by Alex Christensen.
+
+No new tests (No behavior change, refactor only)
+
+Specifically this expands the "shouldContinue" bool to be an enum class with:
+-Yes
+-No
+-ForSuspension
+
+* loader/DocumentLoader.cpp:
+(WebCore::DocumentLoader::willSendRequest):
+* loader/FrameLoader.cpp:
+(WebCore::FrameLoader::loadURL):
+(WebCore::FrameLoader::load):
+(WebCore::FrameLoader::loadWithDocumentLoader):
+(WebCore::FrameLoader::loadPostRequest):
+(WebCore::FrameLoader::continueLoadAfterNavigationPolicy):
+(WebCore::FrameLoader::continueLoadAfterNewWindowPolicy):
+* loader/FrameLoader.h:
+* loader/PolicyChecker.cpp:
+(WebCore::PolicyChecker::checkNavigationPolicy):
+(WebCore::PolicyChecker::checkNewWindowPolicy):
+* loader/PolicyChecker.h:
+
 2018-04-09  Sihui Liu  
 
 REGRESSION(r229929): localStorage is broken for WebInspector


Modified: trunk/Source/WebCore/loader/DocumentLoader.cpp (230457 => 230458)

--- trunk/Source/WebCore/loader/DocumentLoader.cpp	2018-04-10 00:28:24 UTC (rev 230457)
+++ trunk/Source/WebCore/loader/DocumentLoader.cpp	2018-04-10 00:31:02 UTC (rev 230458)
@@ -639,10 +639,19 @@
 
 ASSERT(!m_waitingForNavigationPolicy);
 m_waitingForNavigationPolicy = true;
-frameLoader()->policyChecker().checkNavigationPolicy(ResourceRequest(newRequest), didReceiveRedirectResponse, [this, protectedThis = makeRef(*this), completionHandler = WTFMove(completionHandler)] (ResourceRequest&& request, FormState*, bool shouldContinue) mutable {
+frameLoader()->policyChecker().checkNavigationPolicy(ResourceRequest(newRequest), didReceiveRedirectResponse, [this, protectedThis = makeRef(*this), completionHandler = WTFMove(completionHandler)] (ResourceRequest&& request, FormState*, ShouldContinue shouldContinue) mutable {
 m_waitingForNavigationPolicy = false;
-if (!shouldContinue)
+switch (shouldContinue) {
+case ShouldContinue::ForSuspension:
+FALLTHROUGH;
+// FIXME: Setup this page for suspension (e.g. Page Cache) here.
+case ShouldContinue::No:
 stopLoadingForPolicyChange();
+break;
+case ShouldContinue::Yes:
+break;
+}
+
 completionHandler(WTFMove(request));
 });
 }


Modified: trunk/Source/WebCore/loader/FrameLoader.cpp (230457 => 230458)

--- trunk/Source/WebCore/loader/FrameLoader.cpp	2018-04-10 00:28:24 UTC (rev 230457)
+++ trunk/Source/WebCore/loader/FrameLoader.cpp	2018-04-10 00:31:02 UTC (rev 230458)
@@ -1316,7 +1316,7 @@
 
 if (!targetFrame && !frameName.isEmpty()) {
 action = "" frameLoadRequest));
-policyChecker().checkNewWindowPolicy(WTFMove(action), request, formState, frameName, [this, allowNavigationToInvalidURL, openerPolicy, completionHandler = completionHandlerCaller.release()] (const ResourceRequest& request, FormState* formState, const String& frameName, const NavigationAction& action, bool shouldContinue) {
+policyChecker().checkNewWindowPolicy(WTFMove(action), request, formState, frameName, [this, allowNavigationToInvalidURL, openerPolicy, 

[webkit-changes] [230457] trunk/JSTests/ChangeLog

[commit-queue]

Title: [230457] trunk/JSTests/ChangeLog








Revision 230457
Author commit-qu...@webkit.org
Date 2018-04-09 17:28:24 -0700 (Mon, 09 Apr 2018)


Log Message
Fix errant Test262 files CRLF to LF for consistency with the original source
https://bugs.webkit.org/show_bug.cgi?id=184425

Patch by Leo Balter  on 2018-04-09
Reviewed by Yusuke Suzuki.

* test262/test/built-ins/Math/acosh/nan-returns.js:
* test262/test/built-ins/Math/asinh/asinh-specialVals.js:
* test262/test/built-ins/Math/atanh/atanh-specialVals.js:
* test262/test/built-ins/Math/cbrt/cbrt-specialValues.js:
* test262/test/built-ins/Math/cbrt/prop-desc.js:
* test262/test/built-ins/Math/cosh/cosh-specialVals.js:
* test262/test/built-ins/Math/expm1/expm1-specialVals.js:
* test262/test/built-ins/Math/log10/Log10-specialVals.js:
* test262/test/built-ins/Math/log2/log2-basicTests.js:
* test262/test/built-ins/Math/sign/sign-specialVals.js:
* test262/test/built-ins/Math/sinh/sinh-specialVals.js:
* test262/test/built-ins/Math/tanh/tanh-specialVals.js:
* test262/test/built-ins/Math/trunc/trunc-sampleTests.js:
* test262/test/built-ins/Math/trunc/trunc-specialVals.js:

Modified Paths

trunk/JSTests/ChangeLog




Diff

Modified: trunk/JSTests/ChangeLog (230456 => 230457)

--- trunk/JSTests/ChangeLog	2018-04-10 00:17:49 UTC (rev 230456)
+++ trunk/JSTests/ChangeLog	2018-04-10 00:28:24 UTC (rev 230457)
@@ -1,3 +1,25 @@
+2018-04-09  Leo Balter  
+
+Fix errant Test262 files CRLF to LF for consistency with the original source
+https://bugs.webkit.org/show_bug.cgi?id=184425
+
+Reviewed by Yusuke Suzuki.
+
+* test262/test/built-ins/Math/acosh/nan-returns.js:
+* test262/test/built-ins/Math/asinh/asinh-specialVals.js:
+* test262/test/built-ins/Math/atanh/atanh-specialVals.js:
+* test262/test/built-ins/Math/cbrt/cbrt-specialValues.js:
+* test262/test/built-ins/Math/cbrt/prop-desc.js:
+* test262/test/built-ins/Math/cosh/cosh-specialVals.js:
+* test262/test/built-ins/Math/expm1/expm1-specialVals.js:
+* test262/test/built-ins/Math/log10/Log10-specialVals.js:
+* test262/test/built-ins/Math/log2/log2-basicTests.js:
+* test262/test/built-ins/Math/sign/sign-specialVals.js:
+* test262/test/built-ins/Math/sinh/sinh-specialVals.js:
+* test262/test/built-ins/Math/tanh/tanh-specialVals.js:
+* test262/test/built-ins/Math/trunc/trunc-sampleTests.js:
+* test262/test/built-ins/Math/trunc/trunc-specialVals.js:
+
 2018-04-08  Yusuke Suzuki  
 
 Unreviewed, remove incorrect entry in test262.yaml






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

[webkit-changes] [230456] trunk

[commit-queue]

Title: [230456] trunk








Revision 230456
Author commit-qu...@webkit.org
Date 2018-04-09 17:17:49 -0700 (Mon, 09 Apr 2018)


Log Message
REGRESSION(r229929): localStorage is broken for WebInspector
https://bugs.webkit.org/show_bug.cgi?id=184382


Patch by Sihui Liu  on 2018-04-09
Reviewed by Chris Dumez.

Source/WebCore:

Removed an if condition that caused false positive cases of database error. As per
https://www.sqlite.org/c3ref/errcode.html, return value of sqlite3_errcode() is undefined
on successful API call, so we should not use the code to check if there is an error. We
should only use it when there is an error.
After moving this condition, LocalStorage might return empty string instead of NULL on
sqlite3_column_blob() error.

Modified a test to cover this case:
TestWebKitAPI/Tests/WebKitCocoa/LocalStoragePersistence.mm.

* platform/sql/SQLiteStatement.cpp:
(WebCore::SQLiteStatement::getColumnBlobAsString):

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/LocalStoragePersistence.mm:
(TEST):
* TestWebKitAPI/Tests/WebKitCocoa/localstorage-empty-string-value.html:

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/sql/SQLiteStatement.cpp
trunk/Tools/ChangeLog
trunk/Tools/TestWebKitAPI/Tests/WebKitCocoa/LocalStoragePersistence.mm
trunk/Tools/TestWebKitAPI/Tests/WebKitCocoa/localstorage-empty-string-value.html




Diff

Modified: trunk/Source/WebCore/ChangeLog (230455 => 230456)

--- trunk/Source/WebCore/ChangeLog	2018-04-09 23:44:56 UTC (rev 230455)
+++ trunk/Source/WebCore/ChangeLog	2018-04-10 00:17:49 UTC (rev 230456)
@@ -1,3 +1,24 @@
+2018-04-09  Sihui Liu  
+
+REGRESSION(r229929): localStorage is broken for WebInspector
+https://bugs.webkit.org/show_bug.cgi?id=184382
+
+
+Reviewed by Chris Dumez.
+
+Removed an if condition that caused false positive cases of database error. As per 
+https://www.sqlite.org/c3ref/errcode.html, return value of sqlite3_errcode() is undefined
+on successful API call, so we should not use the code to check if there is an error. We
+should only use it when there is an error.
+After moving this condition, LocalStorage might return empty string instead of NULL on
+sqlite3_column_blob() error.
+
+Modified a test to cover this case: 
+TestWebKitAPI/Tests/WebKitCocoa/LocalStoragePersistence.mm.
+
+* platform/sql/SQLiteStatement.cpp:
+(WebCore::SQLiteStatement::getColumnBlobAsString):
+
 2018-04-09  Brent Fulgham  
 
 WebCore::EventHandler::targetPositionInWindowForSelectionAutoscroll is directly accessing NSScreen


Modified: trunk/Source/WebCore/platform/sql/SQLiteStatement.cpp (230455 => 230456)

--- trunk/Source/WebCore/platform/sql/SQLiteStatement.cpp	2018-04-09 23:44:56 UTC (rev 230455)
+++ trunk/Source/WebCore/platform/sql/SQLiteStatement.cpp	2018-04-10 00:17:49 UTC (rev 230456)
@@ -383,8 +383,6 @@
 return String();
 
 const void* blob = sqlite3_column_blob(m_statement, col);
-if (m_database.lastError() != SQLITE_OK)
-return String();
 if (!blob)
 return emptyString();
 


Modified: trunk/Tools/ChangeLog (230455 => 230456)

--- trunk/Tools/ChangeLog	2018-04-09 23:44:56 UTC (rev 230455)
+++ trunk/Tools/ChangeLog	2018-04-10 00:17:49 UTC (rev 230456)
@@ -1,3 +1,15 @@
+2018-04-09  Sihui Liu  
+
+REGRESSION(r229929): localStorage is broken for WebInspector
+https://bugs.webkit.org/show_bug.cgi?id=184382
+
+
+Reviewed by Chris Dumez.
+
+* TestWebKitAPI/Tests/WebKitCocoa/LocalStoragePersistence.mm:
+(TEST):
+* TestWebKitAPI/Tests/WebKitCocoa/localstorage-empty-string-value.html:
+
 2018-04-09  Zan Dobersek  
 
 Unreviewed follow-up to r230389.


Modified: trunk/Tools/TestWebKitAPI/Tests/WebKitCocoa/LocalStoragePersistence.mm (230455 => 230456)

--- trunk/Tools/TestWebKitAPI/Tests/WebKitCocoa/LocalStoragePersistence.mm	2018-04-09 23:44:56 UTC (rev 230455)
+++ trunk/Tools/TestWebKitAPI/Tests/WebKitCocoa/LocalStoragePersistence.mm	2018-04-10 00:17:49 UTC (rev 230456)
@@ -78,7 +78,7 @@
 TestWebKitAPI::Util::run();
 receivedScriptMessage = false;
 RetainPtr string1 = (NSString *)[lastScriptMessage body];
-EXPECT_WK_STREQ(@"", string1.get());
+EXPECT_WK_STREQ(@"setItem EmptyString", string1.get());
 
 // Ditch this web view (ditch the in-memory local storage).
 webView = nil;


Modified: trunk/Tools/TestWebKitAPI/Tests/WebKitCocoa/localstorage-empty-string-value.html (230455 => 230456)

--- trunk/Tools/TestWebKitAPI/Tests/WebKitCocoa/localstorage-empty-string-value.html	2018-04-09 23:44:56 UTC (rev 230455)
+++ trunk/Tools/TestWebKitAPI/Tests/WebKitCocoa/localstorage-empty-string-value.html	2018-04-10 00:17:49 UTC (rev 230456)
@@ -7,10 +7,10 @@
   if 

[webkit-changes] [230455] trunk/Source

[mark . lam]

Title: [230455] trunk/Source








Revision 230455
Author mark@apple.com
Date 2018-04-09 16:44:56 -0700 (Mon, 09 Apr 2018)


Log Message
Decorate method table entries to support pointer profiling.
https://bugs.webkit.org/show_bug.cgi?id=184430


Reviewed by Saam Barati.

Source/_javascript_Core:

* runtime/ClassInfo.h:

Source/WTF:

* wtf/PointerPreparations.h:

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/runtime/ClassInfo.h
trunk/Source/WTF/ChangeLog
trunk/Source/WTF/wtf/PointerPreparations.h




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (230454 => 230455)

--- trunk/Source/_javascript_Core/ChangeLog	2018-04-09 23:34:17 UTC (rev 230454)
+++ trunk/Source/_javascript_Core/ChangeLog	2018-04-09 23:44:56 UTC (rev 230455)
@@ -1,3 +1,13 @@
+2018-04-09  Mark Lam  
+
+Decorate method table entries to support pointer profiling.
+https://bugs.webkit.org/show_bug.cgi?id=184430
+
+
+Reviewed by Saam Barati.
+
+* runtime/ClassInfo.h:
+
 2018-04-09  Michael Catanzaro  
 
 [WPE] Don't install JSC C API headers


Modified: trunk/Source/_javascript_Core/runtime/ClassInfo.h (230454 => 230455)

--- trunk/Source/_javascript_Core/runtime/ClassInfo.h	2018-04-09 23:34:17 UTC (rev 230454)
+++ trunk/Source/_javascript_Core/runtime/ClassInfo.h	2018-04-09 23:44:56 UTC (rev 230455)
@@ -1,7 +1,7 @@
 /*
  *  Copyright (C) 1999-2001 Harri Porten (por...@kde.org)
  *  Copyright (C) 2001 Peter Kelly (p...@post.com)
- *  Copyright (C) 2003-2017 Apple Inc. All rights reserved.
+ *  Copyright (C) 2003-2018 Apple Inc. All rights reserved.
  *
  *  This library is free software; you can redistribute it and/or
  *  modify it under the terms of the GNU Library General Public
@@ -39,100 +39,100 @@
 
 struct MethodTable {
 using DestroyFunctionPtr = void (*)(JSCell*);
-DestroyFunctionPtr destroy;
+DestroyFunctionPtr WTF_METHOD_TABLE_ENTRY(destroy);
 
 using VisitChildrenFunctionPtr = void (*)(JSCell*, SlotVisitor&);
-VisitChildrenFunctionPtr visitChildren;
+VisitChildrenFunctionPtr WTF_METHOD_TABLE_ENTRY(visitChildren);
 
 using GetCallDataFunctionPtr = CallType (*)(JSCell*, CallData&);
-GetCallDataFunctionPtr getCallData;
+GetCallDataFunctionPtr WTF_METHOD_TABLE_ENTRY(getCallData);
 
 using GetConstructDataFunctionPtr = ConstructType (*)(JSCell*, ConstructData&);
-GetConstructDataFunctionPtr getConstructData;
+GetConstructDataFunctionPtr WTF_METHOD_TABLE_ENTRY(getConstructData);
 
 using PutFunctionPtr = bool (*)(JSCell*, ExecState*, PropertyName propertyName, JSValue, PutPropertySlot&);
-PutFunctionPtr put;
+PutFunctionPtr WTF_METHOD_TABLE_ENTRY(put);
 
 using PutByIndexFunctionPtr = bool (*)(JSCell*, ExecState*, unsigned propertyName, JSValue, bool shouldThrow);
-PutByIndexFunctionPtr putByIndex;
+PutByIndexFunctionPtr WTF_METHOD_TABLE_ENTRY(putByIndex);
 
 using DeletePropertyFunctionPtr = bool (*)(JSCell*, ExecState*, PropertyName);
-DeletePropertyFunctionPtr deleteProperty;
+DeletePropertyFunctionPtr WTF_METHOD_TABLE_ENTRY(deleteProperty);
 
 using DeletePropertyByIndexFunctionPtr = bool (*)(JSCell*, ExecState*, unsigned);
-DeletePropertyByIndexFunctionPtr deletePropertyByIndex;
+DeletePropertyByIndexFunctionPtr WTF_METHOD_TABLE_ENTRY(deletePropertyByIndex);
 
 using GetOwnPropertySlotFunctionPtr = bool (*)(JSObject*, ExecState*, PropertyName, PropertySlot&);
-GetOwnPropertySlotFunctionPtr getOwnPropertySlot;
+GetOwnPropertySlotFunctionPtr WTF_METHOD_TABLE_ENTRY(getOwnPropertySlot);
 
 using GetOwnPropertySlotByIndexFunctionPtr = bool (*)(JSObject*, ExecState*, unsigned, PropertySlot&);
-GetOwnPropertySlotByIndexFunctionPtr getOwnPropertySlotByIndex;
+GetOwnPropertySlotByIndexFunctionPtr WTF_METHOD_TABLE_ENTRY(getOwnPropertySlotByIndex);
 
 using ToThisFunctionPtr = JSValue (*)(JSCell*, ExecState*, ECMAMode);
-ToThisFunctionPtr toThis;
+ToThisFunctionPtr WTF_METHOD_TABLE_ENTRY(toThis);
 
 using DefaultValueFunctionPtr = JSValue (*)(const JSObject*, ExecState*, PreferredPrimitiveType);
-DefaultValueFunctionPtr defaultValue;
+DefaultValueFunctionPtr WTF_METHOD_TABLE_ENTRY(defaultValue);
 
 using GetOwnPropertyNamesFunctionPtr = void (*)(JSObject*, ExecState*, PropertyNameArray&, EnumerationMode);
-GetOwnPropertyNamesFunctionPtr getOwnPropertyNames;
+GetOwnPropertyNamesFunctionPtr WTF_METHOD_TABLE_ENTRY(getOwnPropertyNames);
 
 using GetOwnNonIndexPropertyNamesFunctionPtr = void (*)(JSObject*, ExecState*, PropertyNameArray&, EnumerationMode);
-GetOwnNonIndexPropertyNamesFunctionPtr getOwnNonIndexPropertyNames;
+GetOwnNonIndexPropertyNamesFunctionPtr WTF_METHOD_TABLE_ENTRY(getOwnNonIndexPropertyNames);
 
 using GetPropertyNamesFunctionPtr = void (*)(JSObject*, ExecState*, PropertyNameArray&, EnumerationMode);
-

[webkit-changes] [230454] trunk/Source

[bfulgham]

Title: [230454] trunk/Source








Revision 230454
Author bfulg...@apple.com
Date 2018-04-09 16:34:17 -0700 (Mon, 09 Apr 2018)


Log Message
WebCore::EventHandler::targetPositionInWindowForSelectionAutoscroll is directly accessing NSScreen
https://bugs.webkit.org/show_bug.cgi?id=184344


Reviewed by Per Arne Vollan.

The implementation of targetPositionInWindowForSelectionAutoscroll uses the display ID to get the
screen boundaries of the current display. This causes a bunch of interaction with NSScreen that
we do not want to allow in the WebContent process.

Instead, we should just use the cached screen information the WebContent process already possesses.

This patch makes the following changes:
1. We now retrieve the screen rect of the page's display from the cache, rather than interacting with
   the WindowServer directly.
2. Add a new 'toUserSpaceForPrimaryScreen' so we don't have to deal with a nil NSWindow when computing
   the user space version of the coordinates. A nil Window just means we want to get coordinates in
   terms of the primary display.
3. Keep track of the primary display so we can refer to it later.
4. Modify the IPC messages to include the primary display's ID so we can easily access it later.
5. Modify the PlatformScreen methods to actually use the primary display when appropriate, rather
   than whichever screen happened to hash to the lowest value.

Source/WebCore:

* page/mac/EventHandlerMac.mm:
(WebCore::EventHandler::targetPositionInWindowForSelectionAutoscroll const): Use new methods that
don't require WindowServer access.
* platform/PlatformScreen.h:
* platform/mac/PlatformScreenMac.mm:
(WebCore::displayID): Assert if we hit this code in the WebContent process.
(WebCore::firstScreen): Ditto.
(WebCore::window): Ditto.
(WebCore::screen): Ditto.
(WebCore::primaryScreenID): Added.
(WebCore::getScreenProperties): Modify to return a pair consisting of the primary display ID and
the HashSet of screen settings.
(WebCore::setScreenProperties): Update to also track the primary display ID.
(WebCore::screenProperties): Update to use the primary display ID.
(WebCore::screenHasInvertedColors): Ditto.
(WebCore::toUserSpaceForPrimaryScreen): Added.

Source/WebKit:

Reviewed by Per Arne Vollan.

* UIProcess/WebProcessPool.cpp:
(WebKit::displayReconfigurationCallBack): Update for new getScreenProperties implementation.
(WebKit::WebProcessPool::initializeNewWebProcess): Ditto.
* WebProcess/WebProcess.cpp:
(WebKit::WebProcess::setScreenProperties): Ditto.
* WebProcess/WebProcess.h:
* WebProcess/WebProcess.messages.in: Ditto.

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/page/mac/EventHandlerMac.mm
trunk/Source/WebCore/platform/PlatformScreen.h
trunk/Source/WebCore/platform/mac/PlatformScreenMac.mm
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/WebProcessPool.cpp
trunk/Source/WebKit/WebProcess/WebProcess.cpp
trunk/Source/WebKit/WebProcess/WebProcess.h
trunk/Source/WebKit/WebProcess/WebProcess.messages.in




Diff

Modified: trunk/Source/WebCore/ChangeLog (230453 => 230454)

--- trunk/Source/WebCore/ChangeLog	2018-04-09 23:08:53 UTC (rev 230453)
+++ trunk/Source/WebCore/ChangeLog	2018-04-09 23:34:17 UTC (rev 230454)
@@ -1,3 +1,45 @@
+2018-04-09  Brent Fulgham  
+
+WebCore::EventHandler::targetPositionInWindowForSelectionAutoscroll is directly accessing NSScreen
+https://bugs.webkit.org/show_bug.cgi?id=184344
+
+
+Reviewed by Per Arne Vollan.
+
+The implementation of targetPositionInWindowForSelectionAutoscroll uses the display ID to get the
+screen boundaries of the current display. This causes a bunch of interaction with NSScreen that
+we do not want to allow in the WebContent process.
+
+Instead, we should just use the cached screen information the WebContent process already possesses.
+
+This patch makes the following changes:
+1. We now retrieve the screen rect of the page's display from the cache, rather than interacting with
+   the WindowServer directly.
+2. Add a new 'toUserSpaceForPrimaryScreen' so we don't have to deal with a nil NSWindow when computing
+   the user space version of the coordinates. A nil Window just means we want to get coordinates in
+   terms of the primary display.
+3. Keep track of the primary display so we can refer to it later.
+4. Modify the IPC messages to include the primary display's ID so we can easily access it later.
+5. Modify the PlatformScreen methods to actually use the primary display when appropriate, rather
+   than whichever screen happened to hash to the lowest value.
+
+* page/mac/EventHandlerMac.mm:
+(WebCore::EventHandler::targetPositionInWindowForSelectionAutoscroll const): Use new methods that
+don't require WindowServer access.
+* platform/PlatformScreen.h:
+* platform/mac/PlatformScreenMac.mm:
+

[webkit-changes] [230452] trunk/Source/WebCore

[commit-queue]

Title: [230452] trunk/Source/WebCore








Revision 230452
Author commit-qu...@webkit.org
Date 2018-04-09 16:02:29 -0700 (Mon, 09 Apr 2018)


Log Message
Make InlineTextBox::createTextRun() take a const lvalue reference String
https://bugs.webkit.org/show_bug.cgi?id=184182

Patch by Said Abou-Hallawa  on 2018-04-09
Reviewed by Zalan Bujtas.

InlineTextBox::createTextRun() takes a non-const lvalue reference String.
It is tempting to change the signature of this method to take a const lvalue
reference. But this was done intentionally. TextRun is effectively a StringView:
it does not own the passed string. Having the argument a non-const lvalue
reference makes the compiler prevent calls like createTextRun("abc").

To have a better way to express the lifetime of TextRun, this patch does
the following:

-- It makes TextRun::m_text of type String instead of StringView.
-- It adds a new constructor which takes const String&. This constructor
   will addRef the underlying StringImpl when assigning it to m_text.
-- It keeps the constructor which takes a StringView. The caller of this
   constructor still has to make sure the underlying String outlives the
   TextRun. To avoid copying the underlying buffer of the StringView, we
   will not use  StringView::toString(). Instead we will use
   StringView::toStringWithoutCopying() which makes the returned String
   accesses the same buffer the StringView uses. In this case, the returned
   String is effectively a StringView.

* page/DebugPageOverlays.cpp:
(WebCore::drawRightAlignedText):
* platform/graphics/TextRun.cpp:
* platform/graphics/TextRun.h:
(WebCore::TextRun::TextRun):
(WebCore::TextRun::subRun const):
(WebCore::TextRun::length const):
(WebCore::TextRun::setText):
(WebCore::TextRun::string const): Deleted.
* rendering/InlineTextBox.cpp:
(WebCore::InlineTextBox::localSelectionRect const):
(WebCore::InlineTextBox::paint):
(WebCore::InlineTextBox::paintPlatformDocumentMarker):
(WebCore::InlineTextBox::paintMarkedTextBackground):
(WebCore::InlineTextBox::paintMarkedTextForeground):
(WebCore::InlineTextBox::paintMarkedTextDecoration):
(WebCore::InlineTextBox::offsetForPosition const):
(WebCore::InlineTextBox::positionForOffset const):
(WebCore::InlineTextBox::createTextRun const):
There is no need for this function to take a String argument anymore. The
reason for passing the String was to guarantee its lifetime by keeping
a copy of it in the caller side. Now there is no need for that. The TextRun
itself will keep this copy.

* rendering/InlineTextBox.h:
* rendering/RenderText.cpp:
(WebCore::RenderText::computeCanUseSimplifiedTextMeasuring const):
RenderText::text() returns StringImpl. The compiler wants us to be more
explicit about which constructor of TextRun to call.

* rendering/SimpleLineLayout.cpp:
(WebCore::SimpleLineLayout::canUseForFontAndText):
* rendering/SimpleLineLayoutTextFragmentIterator.cpp:
(WebCore::SimpleLineLayout::TextFragmentIterator::Style::Style):
RenderStyle::hyphenString() returns an AtomicString.

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/page/DebugPageOverlays.cpp
trunk/Source/WebCore/platform/graphics/TextRun.cpp
trunk/Source/WebCore/platform/graphics/TextRun.h
trunk/Source/WebCore/rendering/InlineTextBox.cpp
trunk/Source/WebCore/rendering/InlineTextBox.h
trunk/Source/WebCore/rendering/RenderText.cpp
trunk/Source/WebCore/rendering/SimpleLineLayout.cpp
trunk/Source/WebCore/rendering/SimpleLineLayoutTextFragmentIterator.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (230451 => 230452)

--- trunk/Source/WebCore/ChangeLog	2018-04-09 21:09:54 UTC (rev 230451)
+++ trunk/Source/WebCore/ChangeLog	2018-04-09 23:02:29 UTC (rev 230452)
@@ -1,3 +1,66 @@
+2018-04-09  Said Abou-Hallawa  
+
+Make InlineTextBox::createTextRun() take a const lvalue reference String
+https://bugs.webkit.org/show_bug.cgi?id=184182
+
+Reviewed by Zalan Bujtas.
+
+InlineTextBox::createTextRun() takes a non-const lvalue reference String.
+It is tempting to change the signature of this method to take a const lvalue 
+reference. But this was done intentionally. TextRun is effectively a StringView:
+it does not own the passed string. Having the argument a non-const lvalue
+reference makes the compiler prevent calls like createTextRun("abc").
+
+To have a better way to express the lifetime of TextRun, this patch does
+the following:
+
+-- It makes TextRun::m_text of type String instead of StringView.
+-- It adds a new constructor which takes const String&. This constructor
+   will addRef the underlying StringImpl when assigning it to m_text.
+-- It keeps the constructor which takes a StringView. The caller of this
+   constructor still has to make sure the underlying String outlives the
+   TextRun. To avoid copying the underlying buffer of the StringView, we
+ 

[webkit-changes] [230451] trunk/Source/ThirdParty/libwebrtc

[youenn]

Title: [230451] trunk/Source/ThirdParty/libwebrtc








Revision 230451
Author you...@apple.com
Date 2018-04-09 14:09:54 -0700 (Mon, 09 Apr 2018)


Log Message
Use special software encoder mode in case there is no VCP not hardware encoder
https://bugs.webkit.org/show_bug.cgi?id=183961

Reviewed by Eric Carlson.

In case a compression session is not using a hardware encoder and VCP is not active
use a specific mode if the resolution is standard.

* Source/webrtc/sdk/WebKit/VideoProcessingSoftLink.cpp:
* Source/webrtc/sdk/objc/Framework/Classes/VideoToolbox/RTCVideoEncoderH264.mm:

Modified Paths

trunk/Source/ThirdParty/libwebrtc/ChangeLog
trunk/Source/ThirdParty/libwebrtc/Source/webrtc/sdk/WebKit/VideoProcessingSoftLink.h
trunk/Source/ThirdParty/libwebrtc/Source/webrtc/sdk/objc/Framework/Classes/VideoToolbox/RTCVideoEncoderH264.mm




Diff

Modified: trunk/Source/ThirdParty/libwebrtc/ChangeLog (230450 => 230451)

--- trunk/Source/ThirdParty/libwebrtc/ChangeLog	2018-04-09 20:26:16 UTC (rev 230450)
+++ trunk/Source/ThirdParty/libwebrtc/ChangeLog	2018-04-09 21:09:54 UTC (rev 230451)
@@ -1,3 +1,16 @@
+2018-04-09  Youenn Fablet  
+
+Use special software encoder mode in case there is no VCP not hardware encoder
+https://bugs.webkit.org/show_bug.cgi?id=183961
+
+Reviewed by Eric Carlson.
+
+In case a compression session is not using a hardware encoder and VCP is not active
+use a specific mode if the resolution is standard.
+
+* Source/webrtc/sdk/WebKit/VideoProcessingSoftLink.cpp:
+* Source/webrtc/sdk/objc/Framework/Classes/VideoToolbox/RTCVideoEncoderH264.mm:
+
 2018-04-05  Alejandro G. Castro  
 
 [GTK] Add CMake package search for vpx and libevent libraries


Modified: trunk/Source/ThirdParty/libwebrtc/Source/webrtc/sdk/WebKit/VideoProcessingSoftLink.h (230450 => 230451)

--- trunk/Source/ThirdParty/libwebrtc/Source/webrtc/sdk/WebKit/VideoProcessingSoftLink.h	2018-04-09 20:26:16 UTC (rev 230450)
+++ trunk/Source/ThirdParty/libwebrtc/Source/webrtc/sdk/WebKit/VideoProcessingSoftLink.h	2018-04-09 21:09:54 UTC (rev 230451)
@@ -43,10 +43,6 @@
 
 #endif
 
-#if ENABLE_VCP_ENCODER
-
-#include 
-
 #define ALWAYS_INLINE inline
 
 #ifdef __cplusplus
@@ -83,6 +79,36 @@
 return functionNamespace::softLink##framework##functionName parameterNames; \
 }
 
+#define SOFT_LINK_FRAMEWORK_OPTIONAL(framework) \
+static void* framework##Library() \
+{ \
+static void* frameworkLibrary = dlopen("/System/Library/Frameworks/" #framework ".framework/" #framework, RTLD_NOW); \
+return frameworkLibrary; \
+}
+
+#define SOFT_LINK_POINTER_OPTIONAL(framework, name, type) \
+static type init##name(); \
+static type (*get##name)() = init##name; \
+static type pointer##name; \
+\
+static type name##Function() \
+{ \
+return pointer##name; \
+} \
+\
+static type init##name() \
+{ \
+void** pointer = static_cast(dlsym(framework##Library(), #name)); \
+if (pointer) \
+pointer##name = static_cast(*pointer); \
+get##name = name##Function; \
+return pointer##name; \
+}
+
+#if ENABLE_VCP_ENCODER
+
+#include 
+
 SOFT_LINK_FRAMEWORK_FOR_HEADER(webrtc, VideoProcessing)
 
 SOFT_LINK_FUNCTION_FOR_HEADER(webrtc, VideoProcessing, VCPCompressionSessionSetProperty, OSStatus, (VCPCompressionSessionRef session, CFStringRef key, CFTypeRef value), (session, key, value))


Modified: trunk/Source/ThirdParty/libwebrtc/Source/webrtc/sdk/objc/Framework/Classes/VideoToolbox/RTCVideoEncoderH264.mm (230450 => 230451)

--- trunk/Source/ThirdParty/libwebrtc/Source/webrtc/sdk/objc/Framework/Classes/VideoToolbox/RTCVideoEncoderH264.mm	2018-04-09 20:26:16 UTC (rev 230450)
+++ trunk/Source/ThirdParty/libwebrtc/Source/webrtc/sdk/objc/Framework/Classes/VideoToolbox/RTCVideoEncoderH264.mm	2018-04-09 21:09:54 UTC (rev 230451)
@@ -38,6 +38,40 @@
 #include "sdk/WebKit/EncoderUtilities.h"
 #include "sdk/WebKit/WebKitUtilities.h"
 
+#if !ENABLE_VCP_ENCODER && !defined(WEBRTC_IOS)
+#import 
+#import 
+
+SOFT_LINK_FRAMEWORK_OPTIONAL(VideoToolBox)
+SOFT_LINK_POINTER_OPTIONAL(VideoToolBox, kVTVideoEncoderSpecification_Usage, NSString *)
+
+static inline bool isStandardFrameSize(int32_t width, int32_t height)
+{
+// FIXME: Envision relaxing this rule, something like width and height dividable by 4 or 8 should be good enough.
+if (width == 1280)
+return height == 720;
+if (width == 720)
+return height == 1280;
+if (width == 960)
+return height == 540;
+if (width == 540)
+return height == 960;
+if (width == 640)
+return height == 480;
+if (width == 480)
+return height == 640;
+if (width == 288)
+return height == 352;
+if (width == 352)
+return height == 288;
+if (width == 320)
+return height == 240;
+if (width == 240)
+return 

[webkit-changes] [230450] trunk/Source/JavaScriptCore

[mcatanzaro]

Title: [230450] trunk/Source/_javascript_Core








Revision 230450
Author mcatanz...@igalia.com
Date 2018-04-09 13:26:16 -0700 (Mon, 09 Apr 2018)


Log Message
[WPE] Don't install JSC C API headers
https://bugs.webkit.org/show_bug.cgi?id=184375

Reviewed by Žan Doberšek.

None of the functions declared in these headers are exported in WPE. Use the new jsc API
instead.

* PlatformWPE.cmake:

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/PlatformWPE.cmake




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (230449 => 230450)

--- trunk/Source/_javascript_Core/ChangeLog	2018-04-09 20:03:49 UTC (rev 230449)
+++ trunk/Source/_javascript_Core/ChangeLog	2018-04-09 20:26:16 UTC (rev 230450)
@@ -1,3 +1,15 @@
+2018-04-09  Michael Catanzaro  
+
+[WPE] Don't install JSC C API headers
+https://bugs.webkit.org/show_bug.cgi?id=184375
+
+Reviewed by Žan Doberšek.
+
+None of the functions declared in these headers are exported in WPE. Use the new jsc API
+instead.
+
+* PlatformWPE.cmake:
+
 2018-04-08  Mark Lam  
 
 Add pointer profiling to the FTL and supporting code.


Modified: trunk/Source/_javascript_Core/PlatformWPE.cmake (230449 => 230450)

--- trunk/Source/_javascript_Core/PlatformWPE.cmake	2018-04-09 20:03:49 UTC (rev 230449)
+++ trunk/Source/_javascript_Core/PlatformWPE.cmake	2018-04-09 20:26:16 UTC (rev 230450)
@@ -27,11 +27,6 @@
 inspector/remote/glib/RemoteInspectorUtils.h
 )
 
-install(FILES ${_javascript_Core_PUBLIC_FRAMEWORK_HEADERS}
-DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}/wpe-${WPE_API_VERSION}/WPE/_javascript_Core"
-COMPONENT "Development"
-)
-
 install(FILES ${_javascript_Core_INSTALLED_HEADERS}
 DESTINATION "${CMAKE_INSTALL_INCLUDEDIR}/wpe-${WPE_API_VERSION}/WPE/jsc"
 COMPONENT "Development"






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

[webkit-changes] [230449] trunk/Source/WebKit

[mcatanzaro]

Title: [230449] trunk/Source/WebKit








Revision 230449
Author mcatanz...@igalia.com
Date 2018-04-09 13:03:49 -0700 (Mon, 09 Apr 2018)


Log Message
[WPE] Add API version to library soname and pkg-config files
https://bugs.webkit.org/show_bug.cgi?id=180608

Reviewed by Žan Doberšek.

* PlatformWPE.cmake:
* wpe/wpe-webkit.pc.in:

Modified Paths

trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/PlatformWPE.cmake
trunk/Source/WebKit/wpe/wpe-webkit.pc.in




Diff

Modified: trunk/Source/WebKit/ChangeLog (230448 => 230449)

--- trunk/Source/WebKit/ChangeLog	2018-04-09 19:58:17 UTC (rev 230448)
+++ trunk/Source/WebKit/ChangeLog	2018-04-09 20:03:49 UTC (rev 230449)
@@ -1,3 +1,13 @@
+2018-04-09  Michael Catanzaro  
+
+[WPE] Add API version to library soname and pkg-config files
+https://bugs.webkit.org/show_bug.cgi?id=180608
+
+Reviewed by Žan Doberšek.
+
+* PlatformWPE.cmake:
+* wpe/wpe-webkit.pc.in:
+
 2018-04-09  Jer Noble  
 
 Fix the selection assistant selectionView build


Modified: trunk/Source/WebKit/PlatformWPE.cmake (230448 => 230449)

--- trunk/Source/WebKit/PlatformWPE.cmake	2018-04-09 19:58:17 UTC (rev 230448)
+++ trunk/Source/WebKit/PlatformWPE.cmake	2018-04-09 20:03:49 UTC (rev 230449)
@@ -1,6 +1,6 @@
 include(InspectorGResources.cmake)
 
-set(WebKit_OUTPUT_NAME WPEWebKit)
+set(WebKit_OUTPUT_NAME WPEWebKit-${WPE_API_VERSION})
 set(WebKit_WebProcess_OUTPUT_NAME WPEWebProcess)
 set(WebKit_NetworkProcess_OUTPUT_NAME WPENetworkProcess)
 set(WebKit_StorageProcess_OUTPUT_NAME WPEStorageProcess)
@@ -10,7 +10,7 @@
 file(MAKE_DIRECTORY ${FORWARDING_HEADERS_WPE_EXTENSION_DIR})
 file(MAKE_DIRECTORY ${FORWARDING_HEADERS_WPE_DOM_DIR})
 
-configure_file(wpe/wpe-webkit.pc.in ${CMAKE_BINARY_DIR}/wpe-webkit.pc @ONLY)
+configure_file(wpe/wpe-webkit.pc.in ${CMAKE_BINARY_DIR}/wpe-webkit-${WPE_API_VERSION}.pc @ONLY)
 
 add_definitions(-DWEBKIT2_COMPILATION)
 
@@ -322,7 +322,7 @@
 target_include_directories(WPEInjectedBundle PRIVATE ${WebKit_INCLUDE_DIRECTORIES})
 target_include_directories(WPEInjectedBundle SYSTEM PRIVATE ${WebKit_SYSTEM_INCLUDE_DIRECTORIES})
 
-install(FILES "${CMAKE_BINARY_DIR}/wpe-webkit.pc"
+install(FILES "${CMAKE_BINARY_DIR}/wpe-webkit-${WPE_API_VERSION}.pc"
 DESTINATION "${CMAKE_INSTALL_LIBDIR}/pkgconfig"
 COMPONENT "Development"
 )


Modified: trunk/Source/WebKit/wpe/wpe-webkit.pc.in (230448 => 230449)

--- trunk/Source/WebKit/wpe/wpe-webkit.pc.in	2018-04-09 19:58:17 UTC (rev 230448)
+++ trunk/Source/WebKit/wpe/wpe-webkit.pc.in	2018-04-09 20:03:49 UTC (rev 230449)
@@ -8,5 +8,5 @@
 URL: https://wpewebkit.org
 Version: @PROJECT_VERSION@
 Requires: glib-2.0 libsoup-2.4
-Libs: -L${libdir} -lWPEWebKit
+Libs: -L${libdir} -lWPEWebKit-@WPE_API_VERSION@
 Cflags: -I${includedir}/wpe-@WPE_API_VERSION@ -I${includedir}/wpe-@WPE_API_VERSION@/WPE






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

[webkit-changes] [230448] trunk/Source/WebKit

[jer . noble]

Title: [230448] trunk/Source/WebKit








Revision 230448
Author jer.no...@apple.com
Date 2018-04-09 12:58:17 -0700 (Mon, 09 Apr 2018)


Log Message
Fix the selection assistant selectionView build
https://bugs.webkit.org/show_bug.cgi?id=184423


Reviewed by Wenson Hsieh.

* Platform/spi/ios/UIKitSPI.h:
* UIProcess/ios/WKContentView.mm:
(-[WKContentView _didExitStableState]):
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView _updateChangedSelection:]):

Modified Paths

trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/Platform/spi/ios/UIKitSPI.h
trunk/Source/WebKit/UIProcess/ios/WKContentView.mm
trunk/Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm




Diff

Modified: trunk/Source/WebKit/ChangeLog (230447 => 230448)

--- trunk/Source/WebKit/ChangeLog	2018-04-09 19:31:01 UTC (rev 230447)
+++ trunk/Source/WebKit/ChangeLog	2018-04-09 19:58:17 UTC (rev 230448)
@@ -1,3 +1,17 @@
+2018-04-09  Jer Noble  
+
+Fix the selection assistant selectionView build
+https://bugs.webkit.org/show_bug.cgi?id=184423
+
+
+Reviewed by Wenson Hsieh.
+
+* Platform/spi/ios/UIKitSPI.h:
+* UIProcess/ios/WKContentView.mm:
+(-[WKContentView _didExitStableState]):
+* UIProcess/ios/WKContentViewInteraction.mm:
+(-[WKContentView _updateChangedSelection:]):
+
 2018-04-09  Megan Gardner  
 
 Switch to UIWKTextInteractionAssistant for non-editable text


Modified: trunk/Source/WebKit/Platform/spi/ios/UIKitSPI.h (230447 => 230448)

--- trunk/Source/WebKit/Platform/spi/ios/UIKitSPI.h	2018-04-09 19:31:01 UTC (rev 230447)
+++ trunk/Source/WebKit/Platform/spi/ios/UIKitSPI.h	2018-04-09 19:58:17 UTC (rev 230448)
@@ -497,7 +497,9 @@
 - (void)setGestureRecognizers;
 - (void)willStartScrollingOrZoomingPage;
 - (void)willStartScrollingOverflow;
+#if !ENABLE(MINIMAL_SIMULATOR)
 @property (nonatomic, retain) UIWebSelectionView *selectionView;
+#endif
 @property (nonatomic, readonly) CGRect selectionFrame;
 @end
 


Modified: trunk/Source/WebKit/UIProcess/ios/WKContentView.mm (230447 => 230448)

--- trunk/Source/WebKit/UIProcess/ios/WKContentView.mm	2018-04-09 19:31:01 UTC (rev 230447)
+++ trunk/Source/WebKit/UIProcess/ios/WKContentView.mm	2018-04-09 19:58:17 UTC (rev 230448)
@@ -348,7 +348,9 @@
 return;
 
 [_textSelectionAssistant deactivateSelection];
+#if !ENABLE(MINIMAL_SIMULATOR)
 [[_webSelectionAssistant selectionView] setHidden:YES];
+#endif
 }
 
 - (CGRect)_computeUnobscuredContentRectRespectingInputViewBounds:(CGRect)unobscuredContentRect inputViewBounds:(CGRect)inputViewBounds


Modified: trunk/Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm (230447 => 230448)

--- trunk/Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm	2018-04-09 19:31:01 UTC (rev 230447)
+++ trunk/Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm	2018-04-09 19:58:17 UTC (rev 230448)
@@ -4481,7 +4481,9 @@
 if (!state.isMissingPostLayoutData && state.postLayoutData().isStableStateUpdate && _needsDeferredEndScrollingSelectionUpdate && _page->inStableState()) {
 [[self selectionInteractionAssistant] showSelectionCommands];
 [_webSelectionAssistant didEndScrollingOrZoomingPage];
+#if !ENABLE(MINIMAL_SIMULATOR)
 [[_webSelectionAssistant selectionView] setHidden:NO];
+#endif
 
 if (!self.suppressAssistantSelectionView)
 [_textSelectionAssistant activateSelection];






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

[webkit-changes] [230447] trunk/Source/WebKit

[megan_gardner]

Title: [230447] trunk/Source/WebKit








Revision 230447
Author megan_gard...@apple.com
Date 2018-04-09 12:31:01 -0700 (Mon, 09 Apr 2018)


Log Message
Switch to UIWKTextInteractionAssistant for non-editable text
https://bugs.webkit.org/show_bug.cgi?id=182834

Switch to only using one assistant for text selection.

Reviewed by Timothy Hatcher and Andy Estes.

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

Modified Paths

trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm




Diff

Modified: trunk/Source/WebKit/ChangeLog (230446 => 230447)

--- trunk/Source/WebKit/ChangeLog	2018-04-09 17:58:57 UTC (rev 230446)
+++ trunk/Source/WebKit/ChangeLog	2018-04-09 19:31:01 UTC (rev 230447)
@@ -1,3 +1,15 @@
+2018-04-09  Megan Gardner  
+
+Switch to UIWKTextInteractionAssistant for non-editable text
+https://bugs.webkit.org/show_bug.cgi?id=182834
+
+Switch to only using one assistant for text selection.
+
+Reviewed by Timothy Hatcher and Andy Estes.
+
+* UIProcess/ios/WKContentViewInteraction.mm:
+(-[WKContentView useSelectionAssistantWithGranularity:]):
+
 2018-04-09  Michael Catanzaro  
 
 Unreviewed, move 'using namespace' back to the right place after r230429


Modified: trunk/Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm (230446 => 230447)

--- trunk/Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm	2018-04-09 17:58:57 UTC (rev 230446)
+++ trunk/Source/WebKit/UIProcess/ios/WKContentViewInteraction.mm	2018-04-09 19:31:01 UTC (rev 230447)
@@ -1902,30 +1902,17 @@
 
 - (void)useSelectionAssistantWithGranularity:(WKSelectionGranularity)selectionGranularity
 {
-#if ENABLE(UNIFIED_TEXT_SELECTION)
-selectionGranularity = WKSelectionGranularityCharacter;
-#endif
-if (selectionGranularity == WKSelectionGranularityDynamic) {
-if (_textSelectionAssistant) {
-[_textSelectionAssistant deactivateSelection];
-_textSelectionAssistant = nil;
-}
-if (!_webSelectionAssistant)
-_webSelectionAssistant = adoptNS([[UIWKSelectionAssistant alloc] initWithView:self]);
-} else if (selectionGranularity == WKSelectionGranularityCharacter) {
-if (_webSelectionAssistant)
-_webSelectionAssistant = nil;
+_webSelectionAssistant = nil;
 
-if (!_textSelectionAssistant)
-_textSelectionAssistant = adoptNS([[UIWKTextInteractionAssistant alloc] initWithView:self]);
-else {
-// Reset the gesture recognizers in case editibility has changed.
-[_textSelectionAssistant setGestureRecognizers];
-}
+if (!_textSelectionAssistant)
+_textSelectionAssistant = adoptNS([[UIWKTextInteractionAssistant alloc] initWithView:self]);
+else {
+// Reset the gesture recognizers in case editibility has changed.
+[_textSelectionAssistant setGestureRecognizers];
+}
 
-if (self.isFirstResponder && !self.suppressAssistantSelectionView)
-[_textSelectionAssistant activateSelection];
-}
+if (self.isFirstResponder && !self.suppressAssistantSelectionView)
+[_textSelectionAssistant activateSelection];
 }
 
 - (void)clearSelection






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

[webkit-changes] [230446] tags/Safari-606.1.12.1/Source/WebCore

[kocsen_chung]

Title: [230446] tags/Safari-606.1.12.1/Source/WebCore








Revision 230446
Author kocsen_ch...@apple.com
Date 2018-04-09 10:58:57 -0700 (Mon, 09 Apr 2018)


Log Message
Cherry-pick r230265. rdar://problem/39179126

Fix the !HAVE(AVFOUNDATION_VIDEO_OUTPUT) build
https://bugs.webkit.org/show_bug.cgi?id=184309
-and corresponding-
rdar://problem/39179126

Reviewed by Jer Noble.

* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
(WebCore::MediaPlayerPrivateAVFoundationObjC::updateVideoFullscreenInlineImage):
(WebCore::MediaPlayerPrivateAVFoundationObjC::setVideoFullscreenLayer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@230265 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Modified Paths

tags/Safari-606.1.12.1/Source/WebCore/ChangeLog
tags/Safari-606.1.12.1/Source/WebCore/platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm




Diff

Modified: tags/Safari-606.1.12.1/Source/WebCore/ChangeLog (230445 => 230446)

--- tags/Safari-606.1.12.1/Source/WebCore/ChangeLog	2018-04-09 17:47:54 UTC (rev 230445)
+++ tags/Safari-606.1.12.1/Source/WebCore/ChangeLog	2018-04-09 17:58:57 UTC (rev 230446)
@@ -1,3 +1,35 @@
+2018-04-09  Kocsen Chung  
+
+Cherry-pick r230265. rdar://problem/39179126
+
+Fix the !HAVE(AVFOUNDATION_VIDEO_OUTPUT) build
+https://bugs.webkit.org/show_bug.cgi?id=184309
+-and corresponding-
+rdar://problem/39179126
+
+Reviewed by Jer Noble.
+
+* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
+(WebCore::MediaPlayerPrivateAVFoundationObjC::updateVideoFullscreenInlineImage):
+(WebCore::MediaPlayerPrivateAVFoundationObjC::setVideoFullscreenLayer):
+
+
+
+git-svn-id: https://svn.webkit.org/repository/webkit/trunk@230265 268f45cc-cd09-0410-ab3c-d52691b4dbfc
+
+2018-04-04  Beth Dakin  
+
+Fix the !HAVE(AVFOUNDATION_VIDEO_OUTPUT) build
+https://bugs.webkit.org/show_bug.cgi?id=184309
+-and corresponding-
+rdar://problem/39179126
+
+Reviewed by Jer Noble.
+
+* platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
+(WebCore::MediaPlayerPrivateAVFoundationObjC::updateVideoFullscreenInlineImage):
+(WebCore::MediaPlayerPrivateAVFoundationObjC::setVideoFullscreenLayer):
+
 2018-04-04  Jason Marcell  
 
 Cherry-pick r230272. rdar://problem/39007591


Modified: tags/Safari-606.1.12.1/Source/WebCore/platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm (230445 => 230446)

--- tags/Safari-606.1.12.1/Source/WebCore/platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm	2018-04-09 17:47:54 UTC (rev 230445)
+++ tags/Safari-606.1.12.1/Source/WebCore/platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm	2018-04-09 17:58:57 UTC (rev 230446)
@@ -1168,14 +1168,20 @@
 
 void MediaPlayerPrivateAVFoundationObjC::updateVideoFullscreenInlineImage()
 {
+#if HAVE(AVFOUNDATION_VIDEO_OUTPUT)
 updateLastImage(UpdateType::UpdateSynchronously);
 m_videoFullscreenLayerManager->updateVideoFullscreenInlineImage(m_lastImage);
+#endif
 }
 
 void MediaPlayerPrivateAVFoundationObjC::setVideoFullscreenLayer(PlatformLayer* videoFullscreenLayer, Function&& completionHandler)
 {
+#if HAVE(AVFOUNDATION_VIDEO_OUTPUT)
 updateLastImage(UpdateType::UpdateSynchronously);
 m_videoFullscreenLayerManager->setVideoFullscreenLayer(videoFullscreenLayer, WTFMove(completionHandler), m_lastImage);
+#else
+m_videoFullscreenLayerManager->setVideoFullscreenLayer(videoFullscreenLayer, WTFMove(completionHandler), nil);
+#endif
 updateDisableExternalPlayback();
 }
 






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

[webkit-changes] [230444] trunk/Source/JavaScriptCore

[mark . lam]

Title: [230444] trunk/Source/_javascript_Core








Revision 230444
Author mark@apple.com
Date 2018-04-09 10:42:01 -0700 (Mon, 09 Apr 2018)


Log Message
Add pointer profiling to the FTL and supporting code.
https://bugs.webkit.org/show_bug.cgi?id=184395


Reviewed by Michael Saboff and Filip Pizlo.

* assembler/CodeLocation.h:
(JSC::CodeLocationLabel::retagged):
(JSC::CodeLocationJump::retagged):
* assembler/LinkBuffer.h:
(JSC::LinkBuffer::locationOf):
* dfg/DFGJITCompiler.cpp:
(JSC::DFG::JITCompiler::linkOSRExits):
(JSC::DFG::JITCompiler::link):
* ftl/FTLCompile.cpp:
(JSC::FTL::compile):
* ftl/FTLExceptionTarget.cpp:
(JSC::FTL::ExceptionTarget::label):
(JSC::FTL::ExceptionTarget::jumps):
* ftl/FTLExceptionTarget.h:
* ftl/FTLJITCode.cpp:
(JSC::FTL::JITCode::executableAddressAtOffset):
* ftl/FTLLazySlowPath.cpp:
(JSC::FTL::LazySlowPath::~LazySlowPath):
(JSC::FTL::LazySlowPath::initialize):
(JSC::FTL::LazySlowPath::generate):
(JSC::FTL::LazySlowPath::LazySlowPath): Deleted.
* ftl/FTLLazySlowPath.h:
* ftl/FTLLink.cpp:
(JSC::FTL::link):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::lower):
(JSC::FTL::DFG::LowerDFGToB3::compileCallOrConstruct):
(JSC::FTL::DFG::LowerDFGToB3::compileDirectCallOrConstruct):
(JSC::FTL::DFG::LowerDFGToB3::compileTailCall):
(JSC::FTL::DFG::LowerDFGToB3::compileCallOrConstructVarargsSpread):
(JSC::FTL::DFG::LowerDFGToB3::compileCallOrConstructVarargs):
(JSC::FTL::DFG::LowerDFGToB3::compileCallEval):
(JSC::FTL::DFG::LowerDFGToB3::lazySlowPath):
* ftl/FTLOSRExitCompiler.cpp:
(JSC::FTL::compileStub):
(JSC::FTL::compileFTLOSRExit):
* ftl/FTLOSRExitHandle.cpp:
(JSC::FTL::OSRExitHandle::emitExitThunk):
* ftl/FTLOperations.cpp:
(JSC::FTL::compileFTLLazySlowPath):
* ftl/FTLOutput.h:
(JSC::FTL::Output::callWithoutSideEffects):
(JSC::FTL::Output::operation):
* ftl/FTLPatchpointExceptionHandle.cpp:
(JSC::FTL::PatchpointExceptionHandle::scheduleExitCreationForUnwind):
* ftl/FTLSlowPathCall.cpp:
(JSC::FTL::SlowPathCallContext::makeCall):
* ftl/FTLSlowPathCallKey.h:
(JSC::FTL::SlowPathCallKey::withCallTarget):
(JSC::FTL::SlowPathCallKey::callPtrTag const):
* ftl/FTLThunks.cpp:
(JSC::FTL::genericGenerationThunkGenerator):
(JSC::FTL::osrExitGenerationThunkGenerator):
(JSC::FTL::lazySlowPathGenerationThunkGenerator):
(JSC::FTL::slowPathCallThunkGenerator):
* jit/JITMathIC.h:
(JSC::isProfileEmpty):
* jit/Repatch.cpp:
(JSC::readPutICCallTarget):
(JSC::ftlThunkAwareRepatchCall):
(JSC::tryCacheGetByID):
(JSC::repatchGetByID):
(JSC::tryCachePutByID):
(JSC::repatchPutByID):
(JSC::repatchIn):
(JSC::resetGetByID):
(JSC::resetPutByID):
(JSC::readCallTarget): Deleted.
* jit/Repatch.h:
* runtime/PtrTag.h:

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/assembler/CodeLocation.h
trunk/Source/_javascript_Core/assembler/LinkBuffer.h
trunk/Source/_javascript_Core/dfg/DFGJITCompiler.cpp
trunk/Source/_javascript_Core/ftl/FTLCompile.cpp
trunk/Source/_javascript_Core/ftl/FTLExceptionTarget.cpp
trunk/Source/_javascript_Core/ftl/FTLExceptionTarget.h
trunk/Source/_javascript_Core/ftl/FTLJITCode.cpp
trunk/Source/_javascript_Core/ftl/FTLLazySlowPath.cpp
trunk/Source/_javascript_Core/ftl/FTLLazySlowPath.h
trunk/Source/_javascript_Core/ftl/FTLLink.cpp
trunk/Source/_javascript_Core/ftl/FTLLowerDFGToB3.cpp
trunk/Source/_javascript_Core/ftl/FTLOSRExitCompiler.cpp
trunk/Source/_javascript_Core/ftl/FTLOSRExitHandle.cpp
trunk/Source/_javascript_Core/ftl/FTLOperations.cpp
trunk/Source/_javascript_Core/ftl/FTLOutput.h
trunk/Source/_javascript_Core/ftl/FTLPatchpointExceptionHandle.cpp
trunk/Source/_javascript_Core/ftl/FTLSlowPathCall.cpp
trunk/Source/_javascript_Core/ftl/FTLSlowPathCallKey.h
trunk/Source/_javascript_Core/ftl/FTLThunks.cpp
trunk/Source/_javascript_Core/jit/JITMathIC.h
trunk/Source/_javascript_Core/jit/Repatch.cpp
trunk/Source/_javascript_Core/jit/Repatch.h
trunk/Source/_javascript_Core/runtime/PtrTag.h




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (230443 => 230444)

--- trunk/Source/_javascript_Core/ChangeLog	2018-04-09 17:09:44 UTC (rev 230443)
+++ trunk/Source/_javascript_Core/ChangeLog	2018-04-09 17:42:01 UTC (rev 230444)
@@ -1,3 +1,82 @@
+2018-04-08  Mark Lam  
+
+Add pointer profiling to the FTL and supporting code.
+https://bugs.webkit.org/show_bug.cgi?id=184395
+
+
+Reviewed by Michael Saboff and Filip Pizlo.
+
+* assembler/CodeLocation.h:
+(JSC::CodeLocationLabel::retagged):
+(JSC::CodeLocationJump::retagged):
+* assembler/LinkBuffer.h:
+(JSC::LinkBuffer::locationOf):
+* dfg/DFGJITCompiler.cpp:
+(JSC::DFG::JITCompiler::linkOSRExits):
+(JSC::DFG::JITCompiler::link):
+* ftl/FTLCompile.cpp:
+(JSC::FTL::compile):
+* ftl/FTLExceptionTarget.cpp:
+(JSC::FTL::ExceptionTarget::label):
+(JSC::FTL::ExceptionTarget::jumps):
+* ftl/FTLExceptionTarget.h:
+* ftl/FTLJITCode.cpp:
+

[webkit-changes] [230443] trunk/Source/WebKit

[mcatanzaro]

Title: [230443] trunk/Source/WebKit








Revision 230443
Author mcatanz...@igalia.com
Date 2018-04-09 10:09:44 -0700 (Mon, 09 Apr 2018)


Log Message
Unreviewed, move 'using namespace' back to the right place after r230429

* Shared/glib/ProcessExecutablePathGLib.cpp:

Modified Paths

trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/Shared/glib/ProcessExecutablePathGLib.cpp




Diff

Modified: trunk/Source/WebKit/ChangeLog (230442 => 230443)

--- trunk/Source/WebKit/ChangeLog	2018-04-09 16:53:15 UTC (rev 230442)
+++ trunk/Source/WebKit/ChangeLog	2018-04-09 17:09:44 UTC (rev 230443)
@@ -1,5 +1,11 @@
 2018-04-09  Michael Catanzaro  
 
+Unreviewed, move 'using namespace' back to the right place after r230429
+
+* Shared/glib/ProcessExecutablePathGLib.cpp:
+
+2018-04-09  Michael Catanzaro  
+
 Unreviewed, rolling out r230390.
 
 Broke accelerated compositing


Modified: trunk/Source/WebKit/Shared/glib/ProcessExecutablePathGLib.cpp (230442 => 230443)

--- trunk/Source/WebKit/Shared/glib/ProcessExecutablePathGLib.cpp	2018-04-09 16:53:15 UTC (rev 230442)
+++ trunk/Source/WebKit/Shared/glib/ProcessExecutablePathGLib.cpp	2018-04-09 17:09:44 UTC (rev 230443)
@@ -31,10 +31,9 @@
 #include 
 #include 
 
+namespace WebKit {
 using namespace WebCore;
 
-namespace WebKit {
-
 #if ENABLE(DEVELOPER_MODE)
 static String getExecutablePath()
 {






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

[webkit-changes] [230442] trunk/Source

[mcatanzaro]

Title: [230442] trunk/Source








Revision 230442
Author mcatanz...@igalia.com
Date 2018-04-09 09:53:15 -0700 (Mon, 09 Apr 2018)


Log Message
Unreviewed, rolling out r230390.

Broke accelerated compositing

Reverted changeset:

"[GTK] WaylandCompositorDisplay leaks its wl_display"
https://bugs.webkit.org/show_bug.cgi?id=184406
https://trac.webkit.org/changeset/230390

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (230441 => 230442)

--- trunk/Source/WebCore/ChangeLog	2018-04-09 16:27:52 UTC (rev 230441)
+++ trunk/Source/WebCore/ChangeLog	2018-04-09 16:53:15 UTC (rev 230442)
@@ -1,3 +1,15 @@
+2018-04-09  Michael Catanzaro  
+
+Unreviewed, rolling out r230390.
+
+Broke accelerated compositing
+
+Reverted changeset:
+
+"[GTK] WaylandCompositorDisplay leaks its wl_display"
+https://bugs.webkit.org/show_bug.cgi?id=184406
+https://trac.webkit.org/changeset/230390
+
 2018-04-03  Sergio Villar Senin  
 
 [OpenVR][WebVR] Retrieve FrameData in WebVR's rAF


Modified: trunk/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h (230441 => 230442)

--- trunk/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h	2018-04-09 16:27:52 UTC (rev 230441)
+++ trunk/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h	2018-04-09 16:53:15 UTC (rev 230442)
@@ -50,6 +50,7 @@
 Type type() const override { return PlatformDisplay::Type::Wayland; }
 
 protected:
+PlatformDisplayWayland() = default;
 void initialize(struct wl_display*);
 
 virtual void registryGlobal(const char* interface, uint32_t name);


Modified: trunk/Source/WebKit/ChangeLog (230441 => 230442)

--- trunk/Source/WebKit/ChangeLog	2018-04-09 16:27:52 UTC (rev 230441)
+++ trunk/Source/WebKit/ChangeLog	2018-04-09 16:53:15 UTC (rev 230442)
@@ -1,3 +1,15 @@
+2018-04-09  Michael Catanzaro  
+
+Unreviewed, rolling out r230390.
+
+Broke accelerated compositing
+
+Reverted changeset:
+
+"[GTK] WaylandCompositorDisplay leaks its wl_display"
+https://bugs.webkit.org/show_bug.cgi?id=184406
+https://trac.webkit.org/changeset/230390
+
 2018-04-09  Wenson Hsieh  
 
 [Extra zoom mode] Disable fast clicking by default in extra zoom mode


Modified: trunk/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp (230441 => 230442)

--- trunk/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp	2018-04-09 16:27:52 UTC (rev 230441)
+++ trunk/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp	2018-04-09 16:53:15 UTC (rev 230442)
@@ -44,7 +44,7 @@
 
 struct wl_display* display = wl_display_connect(displayName.utf8().data());
 if (!display) {
-WTFLogAlways("WaylandCompositorDisplay initialization: failed to connect to the Wayland display: %s", displayName.utf8().data());
+WTFLogAlways("PlatformDisplayWayland initialization: failed to connect to the Wayland display: %s", displayName.utf8().data());
 return nullptr;
 }
 
@@ -61,8 +61,8 @@
 }
 
 WaylandCompositorDisplay::WaylandCompositorDisplay(struct wl_display* display)
-: PlatformDisplayWayland(display, NativeDisplayOwned::Yes)
 {
+initialize(display);
 PlatformDisplay::setSharedDisplayForCompositing(*this);
 }
 






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

[webkit-changes] [230441] trunk/Websites/perf.webkit.org

[dewei_zhu]

Title: [230441] trunk/Websites/perf.webkit.org








Revision 230441
Author dewei_...@apple.com
Date 2018-04-09 09:27:52 -0700 (Mon, 09 Apr 2018)


Log Message
Added 'CommitSet.diff' which will be shared between multiple independent incoming changes.
https://bugs.webkit.org/show_bug.cgi?id=184368

Reviewed by Ryosuke Niwa.

'CommitSet.diff' will be used in multiple independent incoming changes.
It would be easier to make this a separate change to parallelize the changes depends on this API.

* public/v3/models/commit-set.js:
(CommitSet.prototype.createNameWithoutCollision): Moved from 'AnalysisTaskPage' and make it more generic.
(CommitSet.prototype.diff): Describe differences between 2 commit sets including commit, root and patch differences.
* public/v3/pages/analysis-task-page.js: Move 'AnalysisTaskPage._createRetryNameForTestGroup' to CommitSet in a more generic form.
(AnalysisTaskPage.prototype._retryCurrentTestGroup): Use 'CommitSet.withoutRootPatchOrOwnedCommit' instead.
(AnalysisTaskPage.prototype._createRetryNameForTestGroup): Moved to CommitSet in a more generic form.
* unit-tests/commit-set-tests.js: Added unit tests for 'CommitSet.diff'.

Modified Paths

trunk/Websites/perf.webkit.org/ChangeLog
trunk/Websites/perf.webkit.org/public/v3/models/commit-set.js
trunk/Websites/perf.webkit.org/public/v3/pages/analysis-task-page.js
trunk/Websites/perf.webkit.org/unit-tests/commit-set-tests.js




Diff

Modified: trunk/Websites/perf.webkit.org/ChangeLog (230440 => 230441)

--- trunk/Websites/perf.webkit.org/ChangeLog	2018-04-09 16:10:52 UTC (rev 230440)
+++ trunk/Websites/perf.webkit.org/ChangeLog	2018-04-09 16:27:52 UTC (rev 230441)
@@ -1,3 +1,21 @@
+2018-04-06  Dewei Zhu  
+
+Added 'CommitSet.diff' which will be shared between multiple independent incoming changes.
+https://bugs.webkit.org/show_bug.cgi?id=184368
+
+Reviewed by Ryosuke Niwa.
+
+'CommitSet.diff' will be used in multiple independent incoming changes.
+It would be easier to make this a separate change to parallelize the changes depends on this API.
+
+* public/v3/models/commit-set.js:
+(CommitSet.prototype.createNameWithoutCollision): Moved from 'AnalysisTaskPage' and make it more generic.
+(CommitSet.prototype.diff): Describe differences between 2 commit sets including commit, root and patch differences.
+* public/v3/pages/analysis-task-page.js: Move 'AnalysisTaskPage._createRetryNameForTestGroup' to CommitSet in a more generic form.
+(AnalysisTaskPage.prototype._retryCurrentTestGroup): Use 'CommitSet.withoutRootPatchOrOwnedCommit' instead.
+(AnalysisTaskPage.prototype._createRetryNameForTestGroup): Moved to CommitSet in a more generic form.
+* unit-tests/commit-set-tests.js: Added unit tests for 'CommitSet.diff'.
+
 2018-04-05  Dewei Zhu  
 
 Fix a bug introduced in r230295 that A/B test result is not shown.


Modified: trunk/Websites/perf.webkit.org/public/v3/models/commit-set.js (230440 => 230441)

--- trunk/Websites/perf.webkit.org/public/v3/models/commit-set.js	2018-04-09 16:10:52 UTC (rev 230440)
+++ trunk/Websites/perf.webkit.org/public/v3/models/commit-set.js	2018-04-09 16:27:52 UTC (rev 230441)
@@ -145,6 +145,82 @@
 }
 return false;
 }
+
+static createNameWithoutCollision(name, existingNameSet)
+{
+console.assert(existingNameSet instanceof Set);
+if (!existingNameSet.has(name))
+return name;
+const nameWithNumberMatch = name.match(/(.+?)\s*\(\s*(\d+)\s*\)\s*$/);
+let number = 1;
+if (nameWithNumberMatch) {
+name = nameWithNumberMatch[1];
+number = parseInt(nameWithNumberMatch[2]);
+}
+
+let newName;
+do {
+number++;
+newName = `${name} (${number})`;
+} while (existingNameSet.has(newName));
+
+return newName;
+}
+
+static diff(firstCommitSet, secondCommitSet)
+{
+console.assert(!firstCommitSet.equals(secondCommitSet));
+const allRepositories = new Set([...firstCommitSet.repositories(), ...secondCommitSet.repositories()]);
+const sortedRepositories = Repository.sortByNamePreferringOnesWithURL([...allRepositories]);
+const nameParts = [];
+const missingCommit = {label: () => 'none'};
+const missingPatch = {filename: () => 'none'};
+const makeNameGenerator = () => {
+const existingNameSet = new Set;
+return (name) => {
+const newName = CommitSet.createNameWithoutCollision(name, existingNameSet);
+existingNameSet.add(newName);
+return newName;
+}
+};
+
+for (const repository of sortedRepositories) {
+const firstCommit = firstCommitSet.commitForRepository(repository) || missingCommit;
+const secondCommit = 

[webkit-changes] [230440] trunk/Source/WebKit

[wenson_hsieh]

Title: [230440] trunk/Source/WebKit








Revision 230440
Author wenson_hs...@apple.com
Date 2018-04-09 09:10:52 -0700 (Mon, 09 Apr 2018)


Log Message
[Extra zoom mode] Disable fast clicking by default in extra zoom mode
https://bugs.webkit.org/show_bug.cgi?id=184411


Reviewed by Andy Estes.

As it turns out, existing fast-clicking heuristics don't work so well in extra zoom mode. Even at device-width,
since the page is scaled to fit within the viewport, having single taps take precedence over double taps leads
to a confusing experience when trying to double tap to zoom further on content that contains links and other
click targets. Revert to legacy behavior here by disabling these heuristics.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _initializeWithConfiguration:]):

Modified Paths

trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm




Diff

Modified: trunk/Source/WebKit/ChangeLog (230439 => 230440)

--- trunk/Source/WebKit/ChangeLog	2018-04-09 15:50:47 UTC (rev 230439)
+++ trunk/Source/WebKit/ChangeLog	2018-04-09 16:10:52 UTC (rev 230440)
@@ -1,3 +1,19 @@
+2018-04-09  Wenson Hsieh  
+
+[Extra zoom mode] Disable fast clicking by default in extra zoom mode
+https://bugs.webkit.org/show_bug.cgi?id=184411
+
+
+Reviewed by Andy Estes.
+
+As it turns out, existing fast-clicking heuristics don't work so well in extra zoom mode. Even at device-width,
+since the page is scaled to fit within the viewport, having single taps take precedence over double taps leads
+to a confusing experience when trying to double tap to zoom further on content that contains links and other
+click targets. Revert to legacy behavior here by disabling these heuristics.
+
+* UIProcess/API/Cocoa/WKWebView.mm:
+(-[WKWebView _initializeWithConfiguration:]):
+
 2018-04-06  Wenson Hsieh  
 
 [Extra zoom mode] Add an SPI hook for clients to opt in to focus overlay UI


Modified: trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm (230439 => 230440)

--- trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm	2018-04-09 15:50:47 UTC (rev 230439)
+++ trunk/Source/WebKit/UIProcess/API/Cocoa/WKWebView.mm	2018-04-09 16:10:52 UTC (rev 230440)
@@ -656,8 +656,17 @@
 
 _viewportMetaTagWidth = WebCore::ViewportArguments::ValueAuto;
 _initialScaleFactor = 1;
-_fastClickingIsDisabled = [[NSUserDefaults standardUserDefaults] boolForKey:@"WebKitFastClickingDisabled"];
 
+if (NSNumber *enabledValue = [[NSUserDefaults standardUserDefaults] objectForKey:@"WebKitFastClickingDisabled"])
+_fastClickingIsDisabled = enabledValue.boolValue;
+else {
+#if ENABLE(EXTRA_ZOOM_MODE)
+_fastClickingIsDisabled = YES;
+#else
+_fastClickingIsDisabled = NO;
+#endif
+}
+
 [self _frameOrBoundsChanged];
 
 NSNotificationCenter *center = [NSNotificationCenter defaultCenter];






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

[webkit-changes] [230439] trunk/Source/WebKit

[wenson_hsieh]

Title: [230439] trunk/Source/WebKit








Revision 230439
Author wenson_hs...@apple.com
Date 2018-04-09 08:50:47 -0700 (Mon, 09 Apr 2018)


Log Message
[Extra zoom mode] Add an SPI hook for clients to opt in to focus overlay UI
https://bugs.webkit.org/show_bug.cgi?id=184370


Reviewed by Timothy Hatcher and Andy Estes.

Add a new SPI hook for internal clients to opt in to showing the focused form control overlay. By default, the
overlay is not shown.

* UIProcess/API/Cocoa/_WKInputDelegate.h:

Modified Paths

trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/API/Cocoa/_WKInputDelegate.h




Diff

Modified: trunk/Source/WebKit/ChangeLog (230438 => 230439)

--- trunk/Source/WebKit/ChangeLog	2018-04-09 15:47:34 UTC (rev 230438)
+++ trunk/Source/WebKit/ChangeLog	2018-04-09 15:50:47 UTC (rev 230439)
@@ -1,3 +1,16 @@
+2018-04-06  Wenson Hsieh  
+
+[Extra zoom mode] Add an SPI hook for clients to opt in to focus overlay UI
+https://bugs.webkit.org/show_bug.cgi?id=184370
+
+
+Reviewed by Timothy Hatcher and Andy Estes.
+
+Add a new SPI hook for internal clients to opt in to showing the focused form control overlay. By default, the
+overlay is not shown.
+
+* UIProcess/API/Cocoa/_WKInputDelegate.h:
+
 2018-04-09  Michael Catanzaro  
 
 [WPE] Use GNU install directories


Modified: trunk/Source/WebKit/UIProcess/API/Cocoa/_WKInputDelegate.h (230438 => 230439)

--- trunk/Source/WebKit/UIProcess/API/Cocoa/_WKInputDelegate.h	2018-04-09 15:47:34 UTC (rev 230438)
+++ trunk/Source/WebKit/UIProcess/API/Cocoa/_WKInputDelegate.h	2018-04-09 15:50:47 UTC (rev 230439)
@@ -47,6 +47,7 @@
 - (void)_webView:(WKWebView *)webView insertTextSuggestion:(UITextSuggestion *)suggestion inInputSession:(id <_WKFormInputSession>)inputSession WK_API_AVAILABLE(ios(10.0));
 - (BOOL)_webView:(WKWebView *)webView focusRequiresStrongPasswordAssistance:(id <_WKFocusedElementInfo>)info WK_API_AVAILABLE(ios(WK_IOS_TBA));
 
+- (BOOL)_webView:(WKWebView *)webView shouldRevealFocusOverlayForInputSession:(id <_WKFormInputSession>)inputSession WK_API_AVAILABLE(ios(WK_IOS_TBA));
 - (CGFloat)_webView:(WKWebView *)webView focusedElementContextViewHeightForFittingSize:(CGSize)fittingSize inputSession:(id <_WKFormInputSession>)inputSession WK_API_AVAILABLE(ios(WK_IOS_TBA));
 - (UIView *)_webView:(WKWebView *)webView focusedElementContextViewForInputSession:(id <_WKFormInputSession>)inputSession WK_API_AVAILABLE(ios(WK_IOS_TBA));
 #endif






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

[webkit-changes] [230433] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230433] releases/WebKitGTK/webkit-2.20








Revision 230433
Author carlo...@webkit.org
Date 2018-04-09 08:47:01 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230287 - REGRESSION(r222563): removed DoubleReal type check causes tons of crashes because CSE has never known how to handle SaneChain
https://bugs.webkit.org/show_bug.cgi?id=184319

Reviewed by Saam Barati.

JSTests:

* stress/array-push-nan-to-double-array-cse-sane-and-insane-chain.js: Added.
(foo):
(bar):
* stress/array-push-nan-to-double-array.js: Added.
(foo):
(bar):

Source/_javascript_Core:

In r222581, we replaced type checks about DoubleReal in ArrayPush in the DFG/FTL backends with
assertions. That's correct because FixupPhase was emitting those checks as Check(DoubleRealRep:) before
the ArrayPush.

But this revealed a longstanding CSE bug: CSE will happily match a SaneChain GetByVal with a InBounds
GetByVal. SaneChain can return NaN while InBounds cannot. This means that if we first use AI to
eliminate the Check(DoubleRealRep:) based on the input being a GetByVal(InBounds) but then replace that
with a GetByVal(SaneChain), then we will hit the assertion.

This teaches CSE to not replace GetByVal(InBounds) with GetByVal(SaneChain) and vice versa. That gets
tricky because PutByVal can match either. So, we use the fact that it's legal for a store to def() more
than once: PutByVal now defs() a HeapLocation for InBounds and a HeapLocation for SaneChain.

* dfg/DFGCSEPhase.cpp:
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* dfg/DFGHeapLocation.cpp:
(WTF::printInternal):
* dfg/DFGHeapLocation.h:
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileArrayPush):

Modified Paths

releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/dfg/DFGCSEPhase.cpp
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/dfg/DFGClobberize.h
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/dfg/DFGHeapLocation.cpp
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/dfg/DFGHeapLocation.h
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/dfg/DFGSpeculativeJIT.cpp
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ftl/FTLLowerDFGToB3.cpp


Added Paths

releases/WebKitGTK/webkit-2.20/JSTests/stress/array-push-nan-to-double-array-cse-sane-and-insane-chain.js
releases/WebKitGTK/webkit-2.20/JSTests/stress/array-push-nan-to-double-array.js




Diff

Modified: releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog (230432 => 230433)

--- releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog	2018-04-09 15:46:50 UTC (rev 230432)
+++ releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog	2018-04-09 15:47:01 UTC (rev 230433)
@@ -1,3 +1,17 @@
+2018-04-04  Filip Pizlo  
+
+REGRESSION(r222563): removed DoubleReal type check causes tons of crashes because CSE has never known how to handle SaneChain
+https://bugs.webkit.org/show_bug.cgi?id=184319
+
+Reviewed by Saam Barati.
+
+* stress/array-push-nan-to-double-array-cse-sane-and-insane-chain.js: Added.
+(foo):
+(bar):
+* stress/array-push-nan-to-double-array.js: Added.
+(foo):
+(bar):
+
 2018-03-31  Filip Pizlo  
 
 JSC crash in JIT code with for-of loop and Array/Set iterators


Added: releases/WebKitGTK/webkit-2.20/JSTests/stress/array-push-nan-to-double-array-cse-sane-and-insane-chain.js (0 => 230433)

--- releases/WebKitGTK/webkit-2.20/JSTests/stress/array-push-nan-to-double-array-cse-sane-and-insane-chain.js	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/JSTests/stress/array-push-nan-to-double-array-cse-sane-and-insane-chain.js	2018-04-09 15:47:01 UTC (rev 230433)
@@ -0,0 +1,23 @@
+function foo(array, otherArray, i)
+{
+var x = otherArray[i];
+var y = otherArray[i];
+array.push(y);
+return x / 42;
+}
+
+function bar()
+{
+return [];
+}
+
+noInline(foo);
+noInline(bar);
+
+for (var i = 0; i < 1; ++i)
+foo(bar(), [42.5], 0);
+
+var result = bar();
+foo(result, [,42.5], 0);
+if (result[0] !== void 0)
+throw "Bad result: " + result;


Added: releases/WebKitGTK/webkit-2.20/JSTests/stress/array-push-nan-to-double-array.js (0 => 230433)

--- releases/WebKitGTK/webkit-2.20/JSTests/stress/array-push-nan-to-double-array.js	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/JSTests/stress/array-push-nan-to-double-array.js	2018-04-09 15:47:01 UTC (rev 230433)
@@ -0,0 +1,20 @@
+function foo(array, num, denum)
+{
+array.push(num/denum);
+}
+
+function bar()
+{
+return [];
+}
+
+noInline(foo);
+noInline(bar);
+
+for (var i = 0; i < 1; ++i)
+foo(bar(), 42.5, 3.1);
+
+var result = bar();
+foo(result, 0, 0);
+if ("" + result[0] !== "NaN")
+throw "Bad result: " + result;


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog (230432 => 230433)

--- 

[webkit-changes] [230437] releases/WebKitGTK/webkit-2.20/Source

[carlosgc]

Title: [230437] releases/WebKitGTK/webkit-2.20/Source








Revision 230437
Author carlo...@webkit.org
Date 2018-04-09 08:47:30 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230390 - [GTK] WaylandCompositorDisplay leaks its wl_display
https://bugs.webkit.org/show_bug.cgi?id=184406

Reviewed by Carlos Garcia Campos.

Source/WebCore:

Remove the protected default constructor, for good measure.

* platform/graphics/wayland/PlatformDisplayWayland.h:

Source/WebKit:

* WebProcess/gtk/WaylandCompositorDisplay.cpp:
(WebKit::WaylandCompositorDisplay::create): Fix a log message (drive-by).
(WebKit::WaylandCompositorDisplay::WaylandCompositorDisplay): Fix the leak.

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h
releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230436 => 230437)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 15:47:22 UTC (rev 230436)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 15:47:30 UTC (rev 230437)
@@ -1,3 +1,14 @@
+2018-04-09  Michael Catanzaro  
+
+[GTK] WaylandCompositorDisplay leaks its wl_display
+https://bugs.webkit.org/show_bug.cgi?id=184406
+
+Reviewed by Carlos Garcia Campos.
+
+Remove the protected default constructor, for good measure.
+
+* platform/graphics/wayland/PlatformDisplayWayland.h:
+
 2018-04-05  Zalan Bujtas  
 
 Folding anonymous blocks should not result in deleting content.


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h (230436 => 230437)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h	2018-04-09 15:47:22 UTC (rev 230436)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h	2018-04-09 15:47:30 UTC (rev 230437)
@@ -50,7 +50,6 @@
 Type type() const override { return PlatformDisplay::Type::Wayland; }
 
 protected:
-PlatformDisplayWayland() = default;
 void initialize(struct wl_display*);
 
 virtual void registryGlobal(const char* interface, uint32_t name);


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog (230436 => 230437)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 15:47:22 UTC (rev 230436)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 15:47:30 UTC (rev 230437)
@@ -1,3 +1,14 @@
+2018-04-09  Michael Catanzaro  
+
+[GTK] WaylandCompositorDisplay leaks its wl_display
+https://bugs.webkit.org/show_bug.cgi?id=184406
+
+Reviewed by Carlos Garcia Campos.
+
+* WebProcess/gtk/WaylandCompositorDisplay.cpp:
+(WebKit::WaylandCompositorDisplay::create): Fix a log message (drive-by).
+(WebKit::WaylandCompositorDisplay::WaylandCompositorDisplay): Fix the leak.
+
 2018-04-06  Brian Burg  
 
 REGRESSION(r228371): WebAutomationSession::deleteAllCookies doesn't delete some cookies


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp (230436 => 230437)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp	2018-04-09 15:47:22 UTC (rev 230436)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp	2018-04-09 15:47:30 UTC (rev 230437)
@@ -44,7 +44,7 @@
 
 struct wl_display* display = wl_display_connect(displayName.utf8().data());
 if (!display) {
-WTFLogAlways("PlatformDisplayWayland initialization: failed to connect to the Wayland display: %s", displayName.utf8().data());
+WTFLogAlways("WaylandCompositorDisplay initialization: failed to connect to the Wayland display: %s", displayName.utf8().data());
 return nullptr;
 }
 
@@ -61,8 +61,8 @@
 }
 
 WaylandCompositorDisplay::WaylandCompositorDisplay(struct wl_display* display)
+: PlatformDisplayWayland(display, NativeDisplayOwned::Yes)
 {
-initialize(display);
 PlatformDisplay::setSharedDisplayForCompositing(*this);
 }
 






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

[webkit-changes] [230429] trunk

[mcatanzaro]

Title: [230429] trunk








Revision 230429
Author mcatanz...@igalia.com
Date 2018-04-09 08:45:48 -0700 (Mon, 09 Apr 2018)


Log Message
[WPE] Use GNU install directories
https://bugs.webkit.org/show_bug.cgi?id=184377

Reviewed by Carlos Garcia Campos.

.:

Notably, this means all the CMake arguments that distributors use to customize install
directories (-DCMAKE_INSTALL_*DIR) will no longer be ignored.

* Source/cmake/OptionsCommon.cmake:
* Source/cmake/OptionsWPE.cmake:

Source/WebKit:

Merge ProcessExecutablePathGtk and ProcessExecutablePathWPE into ProcessExecutablePathGLib.
WPE will now load its secondary processes from PKGLIBEXECDIR, like WebKitGTK+.

* PlatformWPE.cmake:
* Shared/glib/ProcessExecutablePathGLib.cpp: Renamed from Source/WebKit/Shared/gtk/ProcessExecutablePathGtk.cpp.
(WebKit::getExecutablePath):
(WebKit::findWebKitProcess):
(WebKit::executablePathOfWebProcess):
(WebKit::executablePathOfPluginProcess):
(WebKit::executablePathOfNetworkProcess):
(WebKit::executablePathOfStorageProcess):
* Shared/wpe/ProcessExecutablePathWPE.cpp: Removed.
* SourcesGTK.txt:
* SourcesWPE.txt:

Modified Paths

trunk/ChangeLog
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/PlatformGTK.cmake
trunk/Source/WebKit/PlatformWPE.cmake
trunk/Source/WebKit/SourcesGTK.txt
trunk/Source/WebKit/SourcesWPE.txt
trunk/Source/cmake/OptionsCommon.cmake
trunk/Source/cmake/OptionsWPE.cmake


Added Paths

trunk/Source/WebKit/Shared/glib/ProcessExecutablePathGLib.cpp


Removed Paths

trunk/Source/WebKit/Shared/gtk/ProcessExecutablePathGtk.cpp
trunk/Source/WebKit/Shared/wpe/ProcessExecutablePathWPE.cpp




Diff

Modified: trunk/ChangeLog (230428 => 230429)

--- trunk/ChangeLog	2018-04-09 14:40:51 UTC (rev 230428)
+++ trunk/ChangeLog	2018-04-09 15:45:48 UTC (rev 230429)
@@ -1,3 +1,16 @@
+2018-04-09  Michael Catanzaro  
+
+[WPE] Use GNU install directories
+https://bugs.webkit.org/show_bug.cgi?id=184377
+
+Reviewed by Carlos Garcia Campos.
+
+Notably, this means all the CMake arguments that distributors use to customize install
+directories (-DCMAKE_INSTALL_*DIR) will no longer be ignored.
+
+* Source/cmake/OptionsCommon.cmake:
+* Source/cmake/OptionsWPE.cmake:
+
 2018-04-08  Fujii Hironori  
 
 [CMake] WebKit should link to WebCore as a PRIVATE library if WebCore is a static library


Modified: trunk/Source/WebKit/ChangeLog (230428 => 230429)

--- trunk/Source/WebKit/ChangeLog	2018-04-09 14:40:51 UTC (rev 230428)
+++ trunk/Source/WebKit/ChangeLog	2018-04-09 15:45:48 UTC (rev 230429)
@@ -1,5 +1,27 @@
 2018-04-09  Michael Catanzaro  
 
+[WPE] Use GNU install directories
+https://bugs.webkit.org/show_bug.cgi?id=184377
+
+Reviewed by Carlos Garcia Campos.
+
+Merge ProcessExecutablePathGtk and ProcessExecutablePathWPE into ProcessExecutablePathGLib.
+WPE will now load its secondary processes from PKGLIBEXECDIR, like WebKitGTK+.
+
+* PlatformWPE.cmake:
+* Shared/glib/ProcessExecutablePathGLib.cpp: Renamed from Source/WebKit/Shared/gtk/ProcessExecutablePathGtk.cpp.
+(WebKit::getExecutablePath):
+(WebKit::findWebKitProcess):
+(WebKit::executablePathOfWebProcess):
+(WebKit::executablePathOfPluginProcess):
+(WebKit::executablePathOfNetworkProcess):
+(WebKit::executablePathOfStorageProcess):
+* Shared/wpe/ProcessExecutablePathWPE.cpp: Removed.
+* SourcesGTK.txt:
+* SourcesWPE.txt:
+
+2018-04-09  Michael Catanzaro  
+
 [GTK] WaylandCompositorDisplay leaks its wl_display
 https://bugs.webkit.org/show_bug.cgi?id=184406
 


Modified: trunk/Source/WebKit/PlatformGTK.cmake (230428 => 230429)

--- trunk/Source/WebKit/PlatformGTK.cmake	2018-04-09 14:40:51 UTC (rev 230428)
+++ trunk/Source/WebKit/PlatformGTK.cmake	2018-04-09 15:45:48 UTC (rev 230429)
@@ -620,11 +620,12 @@
 
 Shared/cairo/ShareableBitmapCairo.cpp
 
+Shared/glib/ProcessExecutablePathGLib.cpp
+
 Shared/gtk/NativeWebKeyboardEventGtk.cpp
 Shared/gtk/NativeWebMouseEventGtk.cpp
 Shared/gtk/NativeWebTouchEventGtk.cpp
 Shared/gtk/NativeWebWheelEventGtk.cpp
-Shared/gtk/ProcessExecutablePathGtk.cpp
 Shared/gtk/WebEventFactory.cpp
 
 Shared/soup/WebCoreArgumentCodersSoup.cpp


Modified: trunk/Source/WebKit/PlatformWPE.cmake (230428 => 230429)

--- trunk/Source/WebKit/PlatformWPE.cmake	2018-04-09 14:40:51 UTC (rev 230428)
+++ trunk/Source/WebKit/PlatformWPE.cmake	2018-04-09 15:45:48 UTC (rev 230429)
@@ -14,7 +14,7 @@
 
 add_definitions(-DWEBKIT2_COMPILATION)
 
-add_definitions(-DLIBEXECDIR="${LIBEXEC_INSTALL_DIR}")
+add_definitions(-DPKGLIBEXECDIR="${LIBEXEC_INSTALL_DIR}")
 add_definitions(-DLOCALEDIR="${CMAKE_INSTALL_FULL_LOCALEDIR}")
 
 if (NOT DEVELOPER_MODE AND NOT CMAKE_SYSTEM_NAME MATCHES "Darwin")


Copied: 

[webkit-changes] [230436] releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po

[carlosgc]

Title: [230436] releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po








Revision 230436
Author carlo...@webkit.org
Date 2018-04-09 08:47:22 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230384 - Updated Czech translation
https://bugs.webkit.org/show_bug.cgi?id=184403

Patch by Daniel Rusek  on 2018-04-08
Rubber-stamped by Michael Catanzaro.

* cs.po:

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/cs.po




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/ChangeLog (230435 => 230436)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/ChangeLog	2018-04-09 15:47:17 UTC (rev 230435)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/ChangeLog	2018-04-09 15:47:22 UTC (rev 230436)
@@ -1,3 +1,12 @@
+2018-04-08  Daniel Rusek  
+
+Updated Czech translation
+https://bugs.webkit.org/show_bug.cgi?id=184403
+
+Rubber-stamped by Michael Catanzaro.
+
+* cs.po:
+
 2018-03-29  Rafael Fontenelle  
 
 [l10n] [pt_BR] Updated Brazilian Portuguese translation


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/cs.po (230435 => 230436)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/cs.po	2018-04-09 15:47:17 UTC (rev 230435)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/cs.po	2018-04-09 15:47:22 UTC (rev 230436)
@@ -2,6 +2,7 @@
 # Copyright (C) 2010 webkit's COPYRIGHT HOLDER
 # This file is distributed under the same license as the webkit package.
 # Lucas Lommer , 2010.
+# Daniel Rusek , 2018.
 #
 msgid ""
 msgstr ""
@@ -8,8 +9,8 @@
 "Project-Id-Version: webkit HEAD\n"
 "Report-Msgid-Bugs-To: http://bugs.webkit.org/\n"
 "POT-Creation-Date: 2013-03-15 17:34-0300\n"
-"PO-Revision-Date: 2010-03-31 13:07+0100\n"
-"Last-Translator: Lucas Lommer \n"
+"PO-Revision-Date: 2018-04-08 22:56+0200\n"
+"Last-Translator: Daniel Rusek \n"
 "Language-Team: Czech \n"
 "Language: cs\n"
 "MIME-Version: 1.0\n"
@@ -16,6 +17,7 @@
 "Content-Type: text/plain; charset=UTF-8\n"
 "Content-Transfer-Encoding: 8bit\n"
 "Plural-Forms: nplurals=3; plural=(n==1) ? 0 : (n>=2 && n<=4) ? 1 : 2;\n"
+"X-Generator: Poedit 2.0.6\n"
 
 #: ErrorsGtk.cpp:37
 msgid "Load request cancelled"
@@ -27,7 +29,7 @@
 
 #: ErrorsGtk.cpp:49
 msgid "URL cannot be shown"
-msgstr ""
+msgstr "URL nemůže být zobrazen"
 
 #: ErrorsGtk.cpp:55
 msgid "Frame load was interrupted"
@@ -39,7 +41,7 @@
 
 #: ErrorsGtk.cpp:67
 msgid "File does not exist"
-msgstr ""
+msgstr "Soubor neexistuje"
 
 #: ErrorsGtk.cpp:73
 msgid "Plugin will handle load"
@@ -51,7 +53,7 @@
 
 #: ErrorsGtk.cpp:105
 msgid "Printer not found"
-msgstr ""
+msgstr "Tiskárna nebyla nalezena"
 
 #: ErrorsGtk.cpp:112
 #, fuzzy
@@ -61,7 +63,7 @@
 #: GtkAuthenticationDialog.cpp:171
 #, c-format
 msgid "The site %s:%i requests a username and password"
-msgstr ""
+msgstr "Stránka %s:%i požaduje uživatelské jméno a heslo"
 
 #: GtkAuthenticationDialog.cpp:175
 msgid "_Remember password"
@@ -89,7 +91,7 @@
 
 #: LocalizedStringsGtk.cpp:71
 msgid "Details"
-msgstr ""
+msgstr "Podrobnosti"
 
 #: LocalizedStringsGtk.cpp:76
 msgid "This is a searchable index. Enter search keywords: "
@@ -100,9 +102,8 @@
 msgstr "Vybrat soubor"
 
 #: LocalizedStringsGtk.cpp:86
-#, fuzzy
 msgid "Choose Files"
-msgstr "Vybrat soubor"
+msgstr "Vybrat soubory"
 
 #: LocalizedStringsGtk.cpp:91 LocalizedStringsGtk.cpp:96
 msgid "(None)"
@@ -133,54 +134,48 @@
 msgstr "Ko_pírovat obrázek"
 
 #: LocalizedStringsGtk.cpp:131
-#, fuzzy
 msgid "Copy Image _Address"
-msgstr "Ko_pírovat obrázek"
+msgstr "Kopírovat _adresu obrázku"
 
 #: LocalizedStringsGtk.cpp:136
-#, fuzzy
 msgid "Open _Video in New Window"
-msgstr "Otevřít o_brázek v novém okně"
+msgstr "Otevřít _video v novém okně"
 
 #: LocalizedStringsGtk.cpp:141
-#, fuzzy
 msgid "Open _Audio in New Window"
-msgstr "Otevřít odkaz v novém _okně"
+msgstr "Otevřít _zvukový soubor v novém okně"
 
 #: LocalizedStringsGtk.cpp:146
-#, fuzzy
 msgid "Cop_y Video Link Location"
-msgstr "_Kopírovat adresu odkazu"
+msgstr "_Kopírovat adresu videa"
 
 #: LocalizedStringsGtk.cpp:151
-#, fuzzy
 msgid "Cop_y Audio Link Location"
-msgstr "_Kopírovat adresu odkazu"
+msgstr "_Kopírovat adresu zvukového souboru"
 
 #: LocalizedStringsGtk.cpp:156
 msgid "_Toggle Media Controls"
-msgstr ""
+msgstr "_Přepnout ovladače médií"
 
 #: LocalizedStringsGtk.cpp:161
 msgid "Toggle Media _Loop Playback"
-msgstr ""
+msgstr "Přepnout přehrávání média ve _smyčce"
 
 #: LocalizedStringsGtk.cpp:166
 msgid "Switch Video to _Fullscreen"
-msgstr ""
+msgstr "Přepnout video na _celou obrazovku"
 
 #: LocalizedStringsGtk.cpp:171
 msgid "_Play"
-msgstr ""
+msgstr "_Přehrát"
 
 #: LocalizedStringsGtk.cpp:176
-#, fuzzy
 

[webkit-changes] [230438] releases/WebKitGTK/webkit-2.20/Source/WebKit

[carlosgc]

Title: [230438] releases/WebKitGTK/webkit-2.20/Source/WebKit








Revision 230438
Author carlo...@webkit.org
Date 2018-04-09 08:47:34 -0700 (Mon, 09 Apr 2018)


Log Message
Unreviewed. Fix title of 2.20 new symbols section in docs.

* UIProcess/API/gtk/docs/webkit2gtk-docs.sgml:

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/API/gtk/docs/webkit2gtk-docs.sgml




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog (230437 => 230438)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 15:47:30 UTC (rev 230437)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 15:47:34 UTC (rev 230438)
@@ -1,3 +1,9 @@
+2018-04-09  Carlos Garcia Campos  
+
+Unreviewed. Fix title of 2.20 new symbols section in docs.
+
+* UIProcess/API/gtk/docs/webkit2gtk-docs.sgml:
+
 2018-04-09  Michael Catanzaro  
 
 [GTK] WaylandCompositorDisplay leaks its wl_display


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/API/gtk/docs/webkit2gtk-docs.sgml (230437 => 230438)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/API/gtk/docs/webkit2gtk-docs.sgml	2018-04-09 15:47:30 UTC (rev 230437)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/API/gtk/docs/webkit2gtk-docs.sgml	2018-04-09 15:47:34 UTC (rev 230438)
@@ -128,7 +128,7 @@
   
 
   
-Index of new symbols in 2.18
+Index of new symbols in 2.20
 
   
 






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

[webkit-changes] [230434] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230434] releases/WebKitGTK/webkit-2.20








Revision 230434
Author carlo...@webkit.org
Date 2018-04-09 08:47:10 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230313 - Folding anonymous blocks should not result in deleting content.
https://bugs.webkit.org/show_bug.cgi?id=184339


Reviewed by Antti Koivisto.

Source/WebCore:

While folding multiple anonymous blocks (moving the children from next sibling over to previous sibling)
we should ensure that the block we are about to destroy does not gain new descendants.
In case of 4 sibling anonymous blocks (A B C D), while destroying B
1. we move C's children to A and destroy C.
2. While destroying C, we notice B and C as sibling anonymous blocks and we move
D's children over to B (even though B is going to be destroyed as we climb back on the stack).

In this patch, B is detached from the tree before we start moving renderers around so that a subsequent folding won't
find B anymore as a candidate.

Test: fast/block/crash-while-folding-anonymous-blocks.html

* rendering/updating/RenderTreeBuilderBlock.cpp:
(WebCore::RenderTreeBuilder::Block::detach):

LayoutTests:

* fast/block/crash-when-subtree-is-still-attached-expected.txt: Progressing. This test does not
intend to remove "foobar" text at all.
* fast/block/crash-while-folding-anonymous-blocks-expected.txt: Added.
* fast/block/crash-while-folding-anonymous-blocks.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-when-subtree-is-still-attached-expected.txt
releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/updating/RenderTreeBuilderBlock.cpp


Added Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-while-folding-anonymous-blocks-expected.txt
releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-while-folding-anonymous-blocks.html




Diff

Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog (230433 => 230434)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 15:47:01 UTC (rev 230433)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 15:47:10 UTC (rev 230434)
@@ -1,3 +1,16 @@
+2018-04-05  Zalan Bujtas  
+
+Folding anonymous blocks should not result in deleting content.
+https://bugs.webkit.org/show_bug.cgi?id=184339
+
+
+Reviewed by Antti Koivisto.
+
+* fast/block/crash-when-subtree-is-still-attached-expected.txt: Progressing. This test does not 
+intend to remove "foobar" text at all.
+* fast/block/crash-while-folding-anonymous-blocks-expected.txt: Added.
+* fast/block/crash-while-folding-anonymous-blocks.html: Added.
+
 2018-03-27  Fujii Hironori  
 
 [GTK] Layout test editing/deleting/delete-surrogatepair.html crashing with CRITICAL **: enchant_dict_check: assertion 'g_utf8_validate(word, len, NULL)' failed


Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-when-subtree-is-still-attached-expected.txt (230433 => 230434)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-when-subtree-is-still-attached-expected.txt	2018-04-09 15:47:01 UTC (rev 230433)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-when-subtree-is-still-attached-expected.txt	2018-04-09 15:47:10 UTC (rev 230434)
@@ -1,3 +1,4 @@
 PASS if
 no crash.
+foobar
 


Added: releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-while-folding-anonymous-blocks-expected.txt (0 => 230434)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-while-folding-anonymous-blocks-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-while-folding-anonymous-blocks-expected.txt	2018-04-09 15:47:10 UTC (rev 230434)
@@ -0,0 +1 @@
+


Added: releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-while-folding-anonymous-blocks.html (0 => 230434)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-while-folding-anonymous-blocks.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/fast/block/crash-while-folding-anonymous-blocks.html	2018-04-09 15:47:10 UTC (rev 230434)
@@ -0,0 +1,33 @@
+
+if (window.testRunner)
+testRunner.dumpAsText();
+// Setting up 3 continuations in separate blocks.
+document.body.offsetHeight;
+
+let di = document.createElement("div");
+foo2.appendChild(di);
+foo2.style.display = "inline";
+document.body.offsetHeight;
+
+foo1.remove();
+foo3.remove();
+document.body.offsetHeight;
+
+foo2.remove();
+document.body.offsetHeight;
+
+let di2 = document.createElement("div");
+foo5.appendChild(di2);
+foo5.style.display = "inline";
+document.body.offsetHeight;
+
+foo4.remove();
+foo6.remove();
+foo5.remove();
+foo7.remove();
+foo9.remove();
+foo10.remove();
+document.body.offsetHeight;
+foo12.remove();
+document.body.offsetHeight;
+
\ No 

[webkit-changes] [230435] releases/WebKitGTK/webkit-2.20/Source/WebKit

[carlosgc]

Title: [230435] releases/WebKitGTK/webkit-2.20/Source/WebKit








Revision 230435
Author carlo...@webkit.org
Date 2018-04-09 08:47:17 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230367 - REGRESSION(r228371): WebAutomationSession::deleteAllCookies doesn't delete some cookies
https://bugs.webkit.org/show_bug.cgi?id=184334


Reviewed by Timothy Hatcher.

When WebDriver adds a cookie for 'localhost', it actually uses the domain '.localhost' per RFC.
When deleting cookies, we first fetch all cookies matching the document's hostname, and
then delete them one by one. However, this code path does not add the dot prefix. This causes
no cookies to match the requested domain, and thus none of them are deleted.

* UIProcess/Automation/WebAutomationSession.cpp:
(WebKit::domainByAddingDotPrefixIfNeeded): Extract this helper method.
(WebKit::WebAutomationSession::addSingleCookie): Use helper method.
(WebKit::WebAutomationSession::deleteAllCookies): Add a dot prefix when
requesting to delete all cookies for a hostname.

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/Automation/WebAutomationSession.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog (230434 => 230435)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 15:47:10 UTC (rev 230434)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 15:47:17 UTC (rev 230435)
@@ -1,3 +1,22 @@
+2018-04-06  Brian Burg  
+
+REGRESSION(r228371): WebAutomationSession::deleteAllCookies doesn't delete some cookies
+https://bugs.webkit.org/show_bug.cgi?id=184334
+
+
+Reviewed by Timothy Hatcher.
+
+When WebDriver adds a cookie for 'localhost', it actually uses the domain '.localhost' per RFC.
+When deleting cookies, we first fetch all cookies matching the document's hostname, and
+then delete them one by one. However, this code path does not add the dot prefix. This causes
+no cookies to match the requested domain, and thus none of them are deleted.
+
+* UIProcess/Automation/WebAutomationSession.cpp:
+(WebKit::domainByAddingDotPrefixIfNeeded): Extract this helper method.
+(WebKit::WebAutomationSession::addSingleCookie): Use helper method.
+(WebKit::WebAutomationSession::deleteAllCookies): Add a dot prefix when
+requesting to delete all cookies for a hostname.
+
 2018-04-03  Carlos Garcia Campos  
 
 ASSERTION FAILED: !m_mainFrame->coreFrame()->loader().frameHasLoaded() || !m_pendingNavigationID when reloading page while a page is loading


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/Automation/WebAutomationSession.cpp (230434 => 230435)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/Automation/WebAutomationSession.cpp	2018-04-09 15:47:10 UTC (rev 230434)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/Automation/WebAutomationSession.cpp	2018-04-09 15:47:17 UTC (rev 230435)
@@ -1142,6 +1142,18 @@
 callback->sendSuccess();
 }
 
+static String domainByAddingDotPrefixIfNeeded(String domain)
+{
+if (domain[0] != '.') {
+// RFC 2965: If an explicitly specified value does not start with a dot, the user agent supplies a leading dot.
+// Assume that any host that ends with a digit is trying to be an IP address.
+if (!WebCore::URL::hostIsIPAddress(domain))
+return makeString('.', domain);
+}
+
+return domain;
+}
+
 void WebAutomationSession::addSingleCookie(ErrorString& errorString, const String& browsingContextHandle, const JSON::Object& cookieObject, Ref&& callback)
 {
 WebPageProxy* page = webPageProxyForHandle(browsingContextHandle);
@@ -1166,14 +1178,9 @@
 // Inherit the domain/host from the main frame's URL if it is not explicitly set.
 if (domain.isEmpty())
 domain = activeURL.host();
-else if (domain[0] != '.') {
-// RFC 2965: If an explicitly specified value does not start with a dot, the user agent supplies a leading dot.
-// Assume that any host that ends with a digit is trying to be an IP address.
-if (!WebCore::URL::hostIsIPAddress(domain))
-domain = makeString('.', domain);
-}
-cookie.domain = domain;
 
+cookie.domain = domainByAddingDotPrefixIfNeeded(domain);
+
 if (!cookieObject.getString(WTF::ASCIILiteral("path"), cookie.path))
 FAIL_WITH_PREDEFINED_ERROR_AND_DETAILS(MissingParameter, "The parameter 'path' was not found.");
 
@@ -1214,7 +1221,7 @@
 ASSERT(activeURL.isValid());
 
 WebCookieManagerProxy* cookieManager = m_processPool->supplement();
-cookieManager->deleteCookiesForHostname(page->websiteDataStore().sessionID(), activeURL.host());
+cookieManager->deleteCookiesForHostname(page->websiteDataStore().sessionID(), domainByAddingDotPrefixIfNeeded(activeURL.host()));
 }
 
 void 

[webkit-changes] [230430] releases/WebKitGTK/webkit-2.20/Tools

[carlosgc]

Title: [230430] releases/WebKitGTK/webkit-2.20/Tools








Revision 230430
Author carlo...@webkit.org
Date 2018-04-09 08:46:35 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230244 - [GTK] Test /webkit/WebKitWebPage/context-menu is failing since r229831
https://bugs.webkit.org/show_bug.cgi?id=184260

Reviewed by Michael Catanzaro.

This is not a regression introduced in r229831, though, it's a bug in the test itself revealed by r229831. We
should wait for the load to complete before trying to show the context menu.

* TestWebKitAPI/Tests/WebKitGtk/TestContextMenu.cpp:
(testContextMenuWebExtensionMenu):

Modified Paths

releases/WebKitGTK/webkit-2.20/Tools/ChangeLog
releases/WebKitGTK/webkit-2.20/Tools/TestWebKitAPI/Tests/WebKitGtk/TestContextMenu.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Tools/ChangeLog (230429 => 230430)

--- releases/WebKitGTK/webkit-2.20/Tools/ChangeLog	2018-04-09 15:45:48 UTC (rev 230429)
+++ releases/WebKitGTK/webkit-2.20/Tools/ChangeLog	2018-04-09 15:46:35 UTC (rev 230430)
@@ -1,3 +1,16 @@
+2018-04-03  Carlos Garcia Campos  
+
+[GTK] Test /webkit/WebKitWebPage/context-menu is failing since r229831
+https://bugs.webkit.org/show_bug.cgi?id=184260
+
+Reviewed by Michael Catanzaro.
+
+This is not a regression introduced in r229831, though, it's a bug in the test itself revealed by r229831. We
+should wait for the load to complete before trying to show the context menu.
+
+* TestWebKitAPI/Tests/WebKitGtk/TestContextMenu.cpp:
+(testContextMenuWebExtensionMenu):
+
 2018-03-22  Adrian Perez de Castro  
 
 [WPE] Enable WOFF2 support


Modified: releases/WebKitGTK/webkit-2.20/Tools/TestWebKitAPI/Tests/WebKitGtk/TestContextMenu.cpp (230429 => 230430)

--- releases/WebKitGTK/webkit-2.20/Tools/TestWebKitAPI/Tests/WebKitGtk/TestContextMenu.cpp	2018-04-09 15:45:48 UTC (rev 230429)
+++ releases/WebKitGTK/webkit-2.20/Tools/TestWebKitAPI/Tests/WebKitGtk/TestContextMenu.cpp	2018-04-09 15:46:35 UTC (rev 230430)
@@ -975,6 +975,7 @@
 
 // Custom menu.
 test->loadHtml("", "ContextMenuTestCustom");
+test->waitUntilLoadFinished();
 test->showContextMenuAndWaitUntilFinished();
 g_assert_cmpuint(test->m_actions.size(), ==, 4);
 g_assert_cmpuint(test->m_actions[0], ==, WEBKIT_CONTEXT_MENU_ACTION_STOP);
@@ -984,6 +985,7 @@
 
 // Menu cleared by the web process.
 test->loadHtml("", "ContextMenuTestClear");
+test->waitUntilLoadFinished();
 test->showContextMenuAndWaitUntilFinished();
 g_assert_cmpuint(test->m_actions.size(), ==, 0);
 }






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

[webkit-changes] [230432] releases/WebKitGTK/webkit-2.20/Source/JavaScriptCore

[carlosgc]

Title: [230432] releases/WebKitGTK/webkit-2.20/Source/_javascript_Core








Revision 230432
Author carlo...@webkit.org
Date 2018-04-09 08:46:50 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230264 - JSArray::appendMemcpy seems to be missing a barrier
https://bugs.webkit.org/show_bug.cgi?id=184290

Reviewed by Mark Lam.

If you write to an array that may contain pointers and you didn't just allocate it, then you need to
barrier right after.

I don't know if this is really a bug - it's possible that all callers of appendMemcpy do things that
obviate the need for this barrier. But these barriers are cheap, so we should do them if in doubt.

* runtime/JSArray.cpp:
(JSC::JSArray::appendMemcpy):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/JSArray.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog (230431 => 230432)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 15:46:45 UTC (rev 230431)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 15:46:50 UTC (rev 230432)
@@ -1,3 +1,19 @@
+2018-04-03  Filip Pizlo  
+
+JSArray::appendMemcpy seems to be missing a barrier
+https://bugs.webkit.org/show_bug.cgi?id=184290
+
+Reviewed by Mark Lam.
+
+If you write to an array that may contain pointers and you didn't just allocate it, then you need to
+barrier right after.
+
+I don't know if this is really a bug - it's possible that all callers of appendMemcpy do things that
+obviate the need for this barrier. But these barriers are cheap, so we should do them if in doubt.
+
+* runtime/JSArray.cpp:
+(JSC::JSArray::appendMemcpy):
+
 2018-03-31  Filip Pizlo  
 
 JSC crash in JIT code with for-of loop and Array/Set iterators


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/JSArray.cpp (230431 => 230432)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/JSArray.cpp	2018-04-09 15:46:45 UTC (rev 230431)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/JSArray.cpp	2018-04-09 15:46:50 UTC (rev 230432)
@@ -554,8 +554,10 @@
 }
 } else if (type == ArrayWithDouble)
 memcpy(butterfly()->contiguousDouble().data() + startIndex, otherArray->butterfly()->contiguousDouble().data(), sizeof(JSValue) * otherLength);
-else
+else {
 memcpy(butterfly()->contiguous().data() + startIndex, otherArray->butterfly()->contiguous().data(), sizeof(JSValue) * otherLength);
+vm.heap.writeBarrier(this);
+}
 
 return true;
 }






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

[webkit-changes] [230431] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230431] releases/WebKitGTK/webkit-2.20








Revision 230431
Author carlo...@webkit.org
Date 2018-04-09 08:46:45 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230245 - ASSERTION FAILED: !m_mainFrame->coreFrame()->loader().frameHasLoaded() || !m_pendingNavigationID when reloading page while a page is loading
https://bugs.webkit.org/show_bug.cgi?id=153210

Reviewed by Alex Christensen.

Source/WebKit:

The assert happens when WebPage::reload() is called twice and the first time the reload is ignored by
FrameLoader because the document URL is empty. In that case the pending navigation is not reset, because
FrameLoader::reload() returns before creating the document loader.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::reload): Check if the pending navigation has been reset after calling FrameLoader::reload()
and reset it otherwise.

Tools:

Add unit tests to WebKit C API and WebKitGLib.

* TestWebKitAPI/Tests/WebKit/PageLoadBasic.cpp:
(TestWebKitAPI::TEST):
* TestWebKitAPI/Tests/WebKitGLib/TestLoaderClient.cpp:
(LoadTwiceAndReloadTest::reloadOnFinishLoad):
(LoadTwiceAndReloadTest::LoadTwiceAndReloadTest):
(LoadTwiceAndReloadTest::~LoadTwiceAndReloadTest):
(LoadTwiceAndReloadTest::waitUntilFinished):
(testWebViewLoadTwiceAndReload):
(beforeAll):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/WebPage/WebPage.cpp
releases/WebKitGTK/webkit-2.20/Tools/ChangeLog
releases/WebKitGTK/webkit-2.20/Tools/TestWebKitAPI/Tests/WebKit/PageLoadBasic.cpp
releases/WebKitGTK/webkit-2.20/Tools/TestWebKitAPI/Tests/WebKitGLib/TestLoaderClient.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog (230430 => 230431)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 15:46:35 UTC (rev 230430)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 15:46:45 UTC (rev 230431)
@@ -1,5 +1,20 @@
 2018-04-03  Carlos Garcia Campos  
 
+ASSERTION FAILED: !m_mainFrame->coreFrame()->loader().frameHasLoaded() || !m_pendingNavigationID when reloading page while a page is loading
+https://bugs.webkit.org/show_bug.cgi?id=153210
+
+Reviewed by Alex Christensen.
+
+The assert happens when WebPage::reload() is called twice and the first time the reload is ignored by
+FrameLoader because the document URL is empty. In that case the pending navigation is not reset, because
+FrameLoader::reload() returns before creating the document loader.
+
+* WebProcess/WebPage/WebPage.cpp:
+(WebKit::WebPage::reload): Check if the pending navigation has been reset after calling FrameLoader::reload()
+and reset it otherwise.
+
+2018-04-03  Carlos Garcia Campos  
+
 [GTK] NetworkProcess from WebKitGtk+ 2.19.9x SIGSEVs in NetworkStorageSession (secret search callback)
 https://bugs.webkit.org/show_bug.cgi?id=183346
 


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/WebPage/WebPage.cpp (230430 => 230431)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/WebPage/WebPage.cpp	2018-04-09 15:46:35 UTC (rev 230430)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/WebPage/WebPage.cpp	2018-04-09 15:46:45 UTC (rev 230431)
@@ -1349,6 +1349,12 @@
 
 m_sandboxExtensionTracker.beginLoad(m_mainFrame.get(), WTFMove(sandboxExtensionHandle));
 corePage()->userInputBridge().reloadFrame(m_mainFrame->coreFrame(), OptionSet::fromRaw(reloadOptions));
+
+if (m_pendingNavigationID) {
+// This can happen if FrameLoader::reload() returns early because the document URL is empty.
+// The reload does nothing so we need to reset the pending navigation. See webkit.org/b/153210.
+m_pendingNavigationID = 0;
+}
 }
 
 void WebPage::goForward(uint64_t navigationID, uint64_t backForwardItemID)


Modified: releases/WebKitGTK/webkit-2.20/Tools/ChangeLog (230430 => 230431)

--- releases/WebKitGTK/webkit-2.20/Tools/ChangeLog	2018-04-09 15:46:35 UTC (rev 230430)
+++ releases/WebKitGTK/webkit-2.20/Tools/ChangeLog	2018-04-09 15:46:45 UTC (rev 230431)
@@ -1,5 +1,24 @@
 2018-04-03  Carlos Garcia Campos  
 
+ASSERTION FAILED: !m_mainFrame->coreFrame()->loader().frameHasLoaded() || !m_pendingNavigationID when reloading page while a page is loading
+https://bugs.webkit.org/show_bug.cgi?id=153210
+
+Reviewed by Alex Christensen.
+
+Add unit tests to WebKit C API and WebKitGLib.
+
+* TestWebKitAPI/Tests/WebKit/PageLoadBasic.cpp:
+(TestWebKitAPI::TEST):
+* TestWebKitAPI/Tests/WebKitGLib/TestLoaderClient.cpp:
+(LoadTwiceAndReloadTest::reloadOnFinishLoad):
+(LoadTwiceAndReloadTest::LoadTwiceAndReloadTest):
+(LoadTwiceAndReloadTest::~LoadTwiceAndReloadTest):
+(LoadTwiceAndReloadTest::waitUntilFinished):
+(testWebViewLoadTwiceAndReload):
+

[webkit-changes] [230428] trunk/Source/WebCore

[svillar]

Title: [230428] trunk/Source/WebCore








Revision 230428
Author svil...@igalia.com
Date 2018-04-09 07:40:51 -0700 (Mon, 09 Apr 2018)


Log Message
[OpenVR][WebVR] Retrieve FrameData in WebVR's rAF
https://bugs.webkit.org/show_bug.cgi?id=184265

Reviewed by Žan Doberšek.

VRFrameData contains all the required information to properly render a VR scene like view
and projection matrices, pose data (position & orientation) and linear & angular velocity
among others. The getFrameData() call must be issued inside a WebVR's own
requestAnimationFrame.

* Modules/webvr/VRDisplay.cpp:
(WebCore::VRDisplay::getFrameData const):
(WebCore::VRDisplay::getPose const):
(WebCore::VRDisplay::requestAnimationFrame):
(WebCore::VRDisplay::cancelAnimationFrame):
* Modules/webvr/VRDisplay.h:
* Modules/webvr/VREyeParameters.h:
(WebCore::VREyeParameters::rawOffset const): Required to compute view matrices.
* Modules/webvr/VRFrameData.cpp:
(WebCore::matrixToArray):
(WebCore::VRFrameData::leftProjectionMatrix const):
(WebCore::VRFrameData::leftViewMatrix const):
(WebCore::VRFrameData::rightProjectionMatrix const):
(WebCore::VRFrameData::rightViewMatrix const):
(WebCore::projectionMatrixFromFieldOfView):
(WebCore::rotationMatrixFromQuaternion):
(WebCore::applyHeadToEyeTransform):
(WebCore::VRFrameData::update):
(WebCore::VRFrameData::timestamp const): Deleted.
* Modules/webvr/VRFrameData.h:
(WebCore::VRFrameData::timestamp const):
* Modules/webvr/VRPose.cpp:
(WebCore::optionalFloat3ToJSCArray):
(WebCore::VRPose::position const):
(WebCore::VRPose::linearVelocity const):
(WebCore::VRPose::linearAcceleration const):
(WebCore::VRPose::orientation const):
(WebCore::VRPose::angularVelocity const):
(WebCore::VRPose::angularAcceleration const):
* Modules/webvr/VRPose.h:
(WebCore::VRPose::create):
(WebCore::VRPose::update):
(WebCore::VRPose::VRPose):
* platform/vr/VRPlatformDisplay.h:
(WebCore::VRPlatformTrackingInfo::Quaternion::Quaternion):
(WebCore::VRPlatformTrackingInfo::Quaternion::conjugate):
(WebCore::VRPlatformTrackingInfo::Quaternion::operator*):
(WebCore::VRPlatformTrackingInfo::Float3::Float3): Just a group of 3 floats used to store
both velocity and acceleration in a format which is very convenient to later generate JSC
arrays.
(WebCore::VRPlatformTrackingInfo::clear):
* platform/vr/openvr/VRPlatformDisplayOpenVR.cpp:
(WebCore::VRPlatformDisplayOpenVR::VRPlatformDisplayOpenVR):
(WebCore::rotationMatrixToQuaternion):
(WebCore::VRPlatformDisplayOpenVR::getTrackingInfo):
* platform/vr/openvr/VRPlatformDisplayOpenVR.h:

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/Modules/webvr/VRDisplay.cpp
trunk/Source/WebCore/Modules/webvr/VRDisplay.h
trunk/Source/WebCore/Modules/webvr/VREyeParameters.h
trunk/Source/WebCore/Modules/webvr/VRFrameData.cpp
trunk/Source/WebCore/Modules/webvr/VRFrameData.h
trunk/Source/WebCore/Modules/webvr/VRPose.cpp
trunk/Source/WebCore/Modules/webvr/VRPose.h
trunk/Source/WebCore/platform/vr/VRPlatformDisplay.h
trunk/Source/WebCore/platform/vr/openvr/VRPlatformDisplayOpenVR.cpp
trunk/Source/WebCore/platform/vr/openvr/VRPlatformDisplayOpenVR.h




Diff

Modified: trunk/Source/WebCore/ChangeLog (230427 => 230428)

--- trunk/Source/WebCore/ChangeLog	2018-04-09 13:56:21 UTC (rev 230427)
+++ trunk/Source/WebCore/ChangeLog	2018-04-09 14:40:51 UTC (rev 230428)
@@ -1,3 +1,62 @@
+2018-04-03  Sergio Villar Senin  
+
+[OpenVR][WebVR] Retrieve FrameData in WebVR's rAF
+https://bugs.webkit.org/show_bug.cgi?id=184265
+
+Reviewed by Žan Doberšek.
+
+VRFrameData contains all the required information to properly render a VR scene like view
+and projection matrices, pose data (position & orientation) and linear & angular velocity
+among others. The getFrameData() call must be issued inside a WebVR's own
+requestAnimationFrame.
+
+* Modules/webvr/VRDisplay.cpp:
+(WebCore::VRDisplay::getFrameData const):
+(WebCore::VRDisplay::getPose const):
+(WebCore::VRDisplay::requestAnimationFrame):
+(WebCore::VRDisplay::cancelAnimationFrame):
+* Modules/webvr/VRDisplay.h:
+* Modules/webvr/VREyeParameters.h:
+(WebCore::VREyeParameters::rawOffset const): Required to compute view matrices.
+* Modules/webvr/VRFrameData.cpp:
+(WebCore::matrixToArray):
+(WebCore::VRFrameData::leftProjectionMatrix const):
+(WebCore::VRFrameData::leftViewMatrix const):
+(WebCore::VRFrameData::rightProjectionMatrix const):
+(WebCore::VRFrameData::rightViewMatrix const):
+(WebCore::projectionMatrixFromFieldOfView):
+(WebCore::rotationMatrixFromQuaternion):
+(WebCore::applyHeadToEyeTransform):
+(WebCore::VRFrameData::update):
+(WebCore::VRFrameData::timestamp const): Deleted.
+* Modules/webvr/VRFrameData.h:
+(WebCore::VRFrameData::timestamp const):
+* Modules/webvr/VRPose.cpp:
+

[webkit-changes] [230426] releases/WebKitGTK/webkit-2.20/Source/WebCore

[carlosgc]

Title: [230426] releases/WebKitGTK/webkit-2.20/Source/WebCore








Revision 230426
Author carlo...@webkit.org
Date 2018-04-09 06:56:16 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230197 - [GTK] Segfaults in enchant_broker_free_dict()
https://bugs.webkit.org/show_bug.cgi?id=183738

Reviewed by Michael Catanzaro.

Check enchant_broker_request_dict() didn't return nullptr before adding it to the m_enchantDictionaries vector.

* platform/text/enchant/TextCheckerEnchant.cpp:
(WebCore::TextCheckerEnchant::updateSpellCheckingLanguages):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/text/enchant/TextCheckerEnchant.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230425 => 230426)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 13:56:10 UTC (rev 230425)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 13:56:16 UTC (rev 230426)
@@ -1,5 +1,17 @@
 2018-04-03  Carlos Garcia Campos  
 
+[GTK] Segfaults in enchant_broker_free_dict()
+https://bugs.webkit.org/show_bug.cgi?id=183738
+
+Reviewed by Michael Catanzaro.
+
+Check enchant_broker_request_dict() didn't return nullptr before adding it to the m_enchantDictionaries vector.
+
+* platform/text/enchant/TextCheckerEnchant.cpp:
+(WebCore::TextCheckerEnchant::updateSpellCheckingLanguages):
+
+2018-04-03  Carlos Garcia Campos  
+
 [GTK] NetworkProcess from WebKitGtk+ 2.19.9x SIGSEVs in NetworkStorageSession (secret search callback)
 https://bugs.webkit.org/show_bug.cgi?id=183346
 


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/text/enchant/TextCheckerEnchant.cpp (230425 => 230426)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/text/enchant/TextCheckerEnchant.cpp	2018-04-09 13:56:10 UTC (rev 230425)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/text/enchant/TextCheckerEnchant.cpp	2018-04-09 13:56:16 UTC (rev 230426)
@@ -140,8 +140,8 @@
 for (auto& language : languages) {
 CString currentLanguage = language.utf8();
 if (enchant_broker_dict_exists(m_broker, currentLanguage.data())) {
-EnchantDict* dict = enchant_broker_request_dict(m_broker, currentLanguage.data());
-spellDictionaries.append(dict);
+if (auto* dict = enchant_broker_request_dict(m_broker, currentLanguage.data()))
+spellDictionaries.append(dict);
 }
 }
 } else {
@@ -149,15 +149,15 @@
 CString utf8Language = defaultLanguage().utf8();
 const char* language = utf8Language.data();
 if (enchant_broker_dict_exists(m_broker, language)) {
-EnchantDict* dict = enchant_broker_request_dict(m_broker, language);
-spellDictionaries.append(dict);
+if (auto* dict = enchant_broker_request_dict(m_broker, language))
+spellDictionaries.append(dict);
 } else {
 // No dictionaries selected, we get the first one from the list.
 Vector allDictionaries;
 enchant_broker_list_dicts(m_broker, enchantDictDescribeCallback, );
 if (!allDictionaries.isEmpty()) {
-EnchantDict* dict = enchant_broker_request_dict(m_broker, allDictionaries.first().data());
-spellDictionaries.append(dict);
+if (auto* dict = enchant_broker_request_dict(m_broker, allDictionaries.first().data()))
+spellDictionaries.append(dict);
 }
 }
 }






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

[webkit-changes] [230427] releases/WebKitGTK/webkit-2.20/Source/WebCore

[carlosgc]

Title: [230427] releases/WebKitGTK/webkit-2.20/Source/WebCore








Revision 230427
Author carlo...@webkit.org
Date 2018-04-09 06:56:21 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230205 - Make SecurityOrigin safe to create and use from any thread
https://bugs.webkit.org/show_bug.cgi?id=184216

Reviewed by Youenn Fablet.

We found that we have a decent amount of code constructing and using SecurityOrigin
objects from non-main threads. Unfortunately, this was not safe, mostly due to
SecurityOrigin's reliance on the SchemeRegistry.

This patch makes it safe to construct a SecurityOrigin on any thread A and use
it later on the same thread A. However, developers still need to call isolatedCopy()
if they want to pass such object to another thread B.

* page/SecurityOrigin.cpp:
(WebCore::SecurityOrigin::canDisplay const):
* page/SecurityOrigin.h:
* page/SecurityPolicy.cpp:
(WebCore::originAccessMapLock):
(WebCore::originAccessMap):
(WebCore::SecurityPolicy::isAccessWhiteListed):
(WebCore::SecurityPolicy::addOriginAccessWhitelistEntry):
(WebCore::SecurityPolicy::removeOriginAccessWhitelistEntry):
(WebCore::SecurityPolicy::resetOriginAccessWhitelists):
* platform/SchemeRegistry.cpp:
(WebCore::schemeRegistryLock):
(WebCore::allBuiltinSchemes):
(WebCore::builtinLocalURLSchemes):
(WebCore::localURLSchemes):
(WebCore::displayIsolatedURLSchemes):
(WebCore::builtinSecureSchemes):
(WebCore::secureSchemes):
(WebCore::builtinSchemesWithUniqueOrigins):
(WebCore::schemesWithUniqueOrigins):
(WebCore::builtinEmptyDocumentSchemes):
(WebCore::emptyDocumentSchemes):
(WebCore::schemesForbiddenFromDomainRelaxation):
(WebCore::builtinCanDisplayOnlyIfCanRequestSchemes):
(WebCore::canDisplayOnlyIfCanRequestSchemes):
(WebCore::notAllowingJavascriptURLsSchemes):
(WebCore::SchemeRegistry::registerURLSchemeAsLocal):
(WebCore::SchemeRegistry::removeURLSchemeRegisteredAsLocal):
(WebCore::schemesAllowingLocalStorageAccessInPrivateBrowsing):
(WebCore::schemesAllowingDatabaseAccessInPrivateBrowsing):
(WebCore::builtinCORSEnabledSchemes):
(WebCore::CORSEnabledSchemes):
(WebCore::ContentSecurityPolicyBypassingSchemes):
(WebCore::cachePartitioningSchemes):
(WebCore::serviceWorkerSchemes):
(WebCore::alwaysRevalidatedSchemes):
(WebCore::SchemeRegistry::shouldTreatURLSchemeAsLocal):
(WebCore::SchemeRegistry::registerURLSchemeAsNoAccess):
(WebCore::SchemeRegistry::shouldTreatURLSchemeAsNoAccess):
(WebCore::SchemeRegistry::registerURLSchemeAsDisplayIsolated):
(WebCore::SchemeRegistry::shouldTreatURLSchemeAsDisplayIsolated):
(WebCore::SchemeRegistry::registerURLSchemeAsSecure):
(WebCore::SchemeRegistry::shouldTreatURLSchemeAsSecure):
(WebCore::SchemeRegistry::canDisplayOnlyIfCanRequest):
(WebCore::SchemeRegistry::registerAsCanDisplayOnlyIfCanRequest):
(WebCore::SchemeRegistry::registerURLSchemeAsBypassingContentSecurityPolicy):
(WebCore::SchemeRegistry::removeURLSchemeRegisteredAsBypassingContentSecurityPolicy):
(WebCore::SchemeRegistry::schemeShouldBypassContentSecurityPolicy):
(WebCore::SchemeRegistry::registerURLSchemeAsCachePartitioned):
(WebCore::SchemeRegistry::shouldPartitionCacheForURLScheme):
(WebCore::SchemeRegistry::registerURLSchemeServiceWorkersCanHandle):
(WebCore::SchemeRegistry::canServiceWorkersHandleURLScheme):
(WebCore::SchemeRegistry::isServiceWorkerContainerCustomScheme):
* platform/SchemeRegistry.h:

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/page/SecurityOrigin.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/page/SecurityOrigin.h
releases/WebKitGTK/webkit-2.20/Source/WebCore/page/SecurityPolicy.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/SchemeRegistry.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/SchemeRegistry.h




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230426 => 230427)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 13:56:16 UTC (rev 230426)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 13:56:21 UTC (rev 230427)
@@ -1,3 +1,73 @@
+2018-04-03  Chris Dumez  
+
+Make SecurityOrigin safe to create and use from any thread
+https://bugs.webkit.org/show_bug.cgi?id=184216
+
+Reviewed by Youenn Fablet.
+
+We found that we have a decent amount of code constructing and using SecurityOrigin
+objects from non-main threads. Unfortunately, this was not safe, mostly due to
+SecurityOrigin's reliance on the SchemeRegistry.
+
+This patch makes it safe to construct a SecurityOrigin on any thread A and use
+it later on the same thread A. However, developers still need to call isolatedCopy()
+if they want to pass such object to another thread B.
+
+* page/SecurityOrigin.cpp:
+(WebCore::SecurityOrigin::canDisplay const):
+* page/SecurityOrigin.h:
+* page/SecurityPolicy.cpp:
+(WebCore::originAccessMapLock):
+

[webkit-changes] [230425] releases/WebKitGTK/webkit-2.20/Source

[carlosgc]

Title: [230425] releases/WebKitGTK/webkit-2.20/Source








Revision 230425
Author carlo...@webkit.org
Date 2018-04-09 06:56:10 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230196 - [GTK] NetworkProcess from WebKitGtk+ 2.19.9x SIGSEVs in NetworkStorageSession (secret search callback)
https://bugs.webkit.org/show_bug.cgi?id=183346

Reviewed by Michael Catanzaro.

Source/WebCore:

This might happen if a request is cancelled right after the password request starts and before it finishes. We
should cancel the password search when the network request is cancelled, not only when the NetworkStorageSession
is destroyed.

* platform/network/NetworkStorageSession.h:
* platform/network/soup/NetworkStorageSessionSoup.cpp:
(WebCore::NetworkStorageSession::~NetworkStorageSession):
(WebCore::SecretServiceSearchData::SecretServiceSearchData): Helper struct to keep the request cancellable and
completion handler.
(WebCore::NetworkStorageSession::getCredentialFromPersistentStorage): Create a SecretServiceSearchData for the
request.
* platform/network/soup/ResourceHandleSoup.cpp:
(WebCore::ResourceHandle::didReceiveAuthenticationChallenge): Pass the request cancellable to
NetworkStorageSession::getCredentialFromPersistentStorage().

Source/WebKit:

Pass the request cancellable to NetworkStorageSession::getCredentialFromPersistentStorage().

* NetworkProcess/soup/NetworkDataTaskSoup.cpp:
(WebKit::NetworkDataTaskSoup::authenticate):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/network/NetworkStorageSession.h
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/network/soup/NetworkStorageSessionSoup.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/network/soup/ResourceHandleSoup.cpp
releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/NetworkProcess/soup/NetworkDataTaskSoup.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230424 => 230425)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 13:33:18 UTC (rev 230424)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 13:56:10 UTC (rev 230425)
@@ -1,3 +1,25 @@
+2018-04-03  Carlos Garcia Campos  
+
+[GTK] NetworkProcess from WebKitGtk+ 2.19.9x SIGSEVs in NetworkStorageSession (secret search callback)
+https://bugs.webkit.org/show_bug.cgi?id=183346
+
+Reviewed by Michael Catanzaro.
+
+This might happen if a request is cancelled right after the password request starts and before it finishes. We
+should cancel the password search when the network request is cancelled, not only when the NetworkStorageSession
+is destroyed.
+
+* platform/network/NetworkStorageSession.h:
+* platform/network/soup/NetworkStorageSessionSoup.cpp:
+(WebCore::NetworkStorageSession::~NetworkStorageSession):
+(WebCore::SecretServiceSearchData::SecretServiceSearchData): Helper struct to keep the request cancellable and
+completion handler.
+(WebCore::NetworkStorageSession::getCredentialFromPersistentStorage): Create a SecretServiceSearchData for the
+request.
+* platform/network/soup/ResourceHandleSoup.cpp:
+(WebCore::ResourceHandle::didReceiveAuthenticationChallenge): Pass the request cancellable to
+NetworkStorageSession::getCredentialFromPersistentStorage().
+
 2018-03-27  Fujii Hironori  
 
 [GTK] Layout test editing/deleting/delete-surrogatepair.html crashing with CRITICAL **: enchant_dict_check: assertion 'g_utf8_validate(word, len, NULL)' failed


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/network/NetworkStorageSession.h (230424 => 230425)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/network/NetworkStorageSession.h	2018-04-09 13:33:18 UTC (rev 230424)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/network/NetworkStorageSession.h	2018-04-09 13:56:10 UTC (rev 230425)
@@ -121,7 +121,7 @@
 SoupCookieJar* cookieStorage() const;
 void setCookieStorage(SoupCookieJar*);
 void setCookieObserverHandler(Function&&);
-void getCredentialFromPersistentStorage(const ProtectionSpace&, Function completionHandler);
+void getCredentialFromPersistentStorage(const ProtectionSpace&, GCancellable*, Function&& completionHandler);
 void saveCredentialToPersistentStorage(const ProtectionSpace&, const Credential&);
 #elif USE(CURL)
 NetworkStorageSession(PAL::SessionID, NetworkingContext*);
@@ -158,10 +158,6 @@
 mutable std::unique_ptr m_session;
 GRefPtr m_cookieStorage;
 Function m_cookieObserverHandler;
-#if USE(LIBSECRET)
-Function m_persisentStorageCompletionHandler;
-GRefPtr m_persisentStorageCancellable;
-#endif
 #elif USE(CURL)
 UniqueRef m_cookieStorage;
 mutable CookieJarDB m_cookieDatabase;


Modified: 

[webkit-changes] [230424] releases/WebKitGTK/webkit-2.20/Source/WebKit

[carlosgc]

Title: [230424] releases/WebKitGTK/webkit-2.20/Source/WebKit








Revision 230424
Author carlo...@webkit.org
Date 2018-04-09 06:33:18 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230156 - [GTK] DragAndDropHandler.cpp should include GUniquePtrGtk.h
https://bugs.webkit.org/show_bug.cgi?id=184119

Reviewed by Daniel Bates.

* UIProcess/gtk/DragAndDropHandler.cpp:

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/gtk/DragAndDropHandler.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog (230423 => 230424)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 13:33:13 UTC (rev 230423)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 13:33:18 UTC (rev 230424)
@@ -1,3 +1,12 @@
+2018-04-02  Michael Catanzaro  
+
+[GTK] DragAndDropHandler.cpp should include GUniquePtrGtk.h
+https://bugs.webkit.org/show_bug.cgi?id=184119
+
+Reviewed by Daniel Bates.
+
+* UIProcess/gtk/DragAndDropHandler.cpp:
+
 2018-03-29  Brian Burg  
 
 Web Automation: clipToViewport is ignored for element screenshots


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/gtk/DragAndDropHandler.cpp (230423 => 230424)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/gtk/DragAndDropHandler.cpp	2018-04-09 13:33:13 UTC (rev 230423)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/gtk/DragAndDropHandler.cpp	2018-04-09 13:33:18 UTC (rev 230424)
@@ -31,10 +31,10 @@
 #include "WebPageProxy.h"
 #include 
 #include 
+#include 
 #include 
 #include 
 #include 
-#include 
 
 namespace WebKit {
 using namespace WebCore;






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

[webkit-changes] [230421] releases/WebKitGTK/webkit-2.20/Source/WebKit

[carlosgc]

Title: [230421] releases/WebKitGTK/webkit-2.20/Source/WebKit








Revision 230421
Author carlo...@webkit.org
Date 2018-04-09 06:32:58 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230088 - Web Automation: clipToViewport is ignored for element screenshots
https://bugs.webkit.org/show_bug.cgi?id=184158


Reviewed by Timothy Hatcher.

In §19.2 Take Element Screenshot, step 5.2 says that we should clip
the element screenshot rect with the visible viewport rect. We don't
do that right now even though we pass over clipToViewport.

* WebProcess/Automation/WebAutomationSessionProxy.cpp:
(WebKit::snapshotRectForScreenshot):
Clip the rect to viewport if needed.

(WebKit::WebAutomationSessionProxy::takeScreenshot):
This scrollIntoView is misplaced; by this point we have already done
the math to figure out the screenshot rect. Move it before computing the rect.

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog (230420 => 230421)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 13:00:10 UTC (rev 230420)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 13:32:58 UTC (rev 230421)
@@ -1,3 +1,23 @@
+2018-03-29  Brian Burg  
+
+Web Automation: clipToViewport is ignored for element screenshots
+https://bugs.webkit.org/show_bug.cgi?id=184158
+
+
+Reviewed by Timothy Hatcher.
+
+In §19.2 Take Element Screenshot, step 5.2 says that we should clip
+the element screenshot rect with the visible viewport rect. We don't
+do that right now even though we pass over clipToViewport.
+
+* WebProcess/Automation/WebAutomationSessionProxy.cpp:
+(WebKit::snapshotRectForScreenshot):
+Clip the rect to viewport if needed.
+
+(WebKit::WebAutomationSessionProxy::takeScreenshot):
+This scrollIntoView is misplaced; by this point we have already done
+the math to figure out the screenshot rect. Move it before computing the rect.
+
 2018-03-29  Carlos Eduardo Ramalho  
 
 [WPE] Floating point exception in WebEventFactory::createWebWheelEvent


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.cpp (230420 => 230421)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.cpp	2018-04-09 13:00:10 UTC (rev 230420)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.cpp	2018-04-09 13:32:58 UTC (rev 230421)
@@ -663,12 +663,20 @@
 
 static WebCore::IntRect snapshotRectForScreenshot(WebPage& page, WebCore::Element* element, bool clipToViewport)
 {
+auto* frameView = page.mainFrameView();
+if (!frameView)
+return { };
+
 if (element) {
 if (!element->renderer())
 return { };
 
 WebCore::LayoutRect topLevelRect;
-return WebCore::snappedIntRect(element->renderer()->paintingRootRect(topLevelRect));
+WebCore::IntRect elementRect = WebCore::snappedIntRect(element->renderer()->paintingRootRect(topLevelRect));
+if (clipToViewport)
+elementRect.intersect(frameView->visibleContentRect());
+
+return elementRect;
 }
 
 if (auto* frameView = page.mainFrameView())
@@ -705,6 +713,9 @@
 }
 }
 
+if (coreElement && scrollIntoViewIfNeeded)
+coreElement->scrollIntoViewIfNeeded(false);
+
 String screenshotErrorType = Inspector::Protocol::AutomationHelpers::getEnumConstantValue(Inspector::Protocol::Automation::ErrorMessage::ScreenshotError);
 WebCore::IntRect snapshotRect = snapshotRectForScreenshot(*page, coreElement, clipToViewport);
 if (snapshotRect.isEmpty()) {
@@ -712,9 +723,6 @@
 return;
 }
 
-if (coreElement && scrollIntoViewIfNeeded)
-coreElement->scrollIntoViewIfNeeded(false);
-
 RefPtr image = page->scaledSnapshotWithOptions(snapshotRect, 1, SnapshotOptionsShareable);
 if (!image) {
 WebProcess::singleton().parentProcessConnection()->send(Messages::WebAutomationSession::DidTakeScreenshot(callbackID, handle, screenshotErrorType), 0);






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

[webkit-changes] [230423] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230423] releases/WebKitGTK/webkit-2.20








Revision 230423
Author carlo...@webkit.org
Date 2018-04-09 06:33:13 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230143 - JSC crash in JIT code with for-of loop and Array/Set iterators
https://bugs.webkit.org/show_bug.cgi?id=183174

Reviewed by Saam Barati.

JSTests:

* microbenchmarks/hoist-get-by-offset-tower-with-inferred-types.js: Added. This test shows that fixing the bug didn't break hoisting of GetByOffset with inferred types. I confirmed that if I did break it, this test slows down by >7x.
(foo):
* stress/hoist-get-by-offset-with-control-dependent-inferred-type.js: Added. This test shows that the bug is fixed.
(f):

Source/_javascript_Core:

* dfg/DFGSafeToExecute.h:
(JSC::DFG::safeToExecute): Fix the bug by making GetByOffset and friends verify that they are getting the type proof they want at the desired hoisting site.

Modified Paths

releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/dfg/DFGSafeToExecute.h


Added Paths

releases/WebKitGTK/webkit-2.20/JSTests/microbenchmarks/hoist-get-by-offset-tower-with-inferred-types.js
releases/WebKitGTK/webkit-2.20/JSTests/stress/hoist-get-by-offset-with-control-dependent-inferred-type.js




Diff

Modified: releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog (230422 => 230423)

--- releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog	2018-04-09 13:33:06 UTC (rev 230422)
+++ releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog	2018-04-09 13:33:13 UTC (rev 230423)
@@ -1,3 +1,15 @@
+2018-03-31  Filip Pizlo  
+
+JSC crash in JIT code with for-of loop and Array/Set iterators
+https://bugs.webkit.org/show_bug.cgi?id=183174
+
+Reviewed by Saam Barati.
+
+* microbenchmarks/hoist-get-by-offset-tower-with-inferred-types.js: Added. This test shows that fixing the bug didn't break hoisting of GetByOffset with inferred types. I confirmed that if I did break it, this test slows down by >7x.
+(foo):
+* stress/hoist-get-by-offset-with-control-dependent-inferred-type.js: Added. This test shows that the bug is fixed.
+(f):
+
 2018-03-30  Robin Morisset  
 
 Out-of-bounds accesses due to a missing check for MAX_STORAGE_VECTOR_LENGTH in unshiftCountForAnyIndexingType


Added: releases/WebKitGTK/webkit-2.20/JSTests/microbenchmarks/hoist-get-by-offset-tower-with-inferred-types.js (0 => 230423)

--- releases/WebKitGTK/webkit-2.20/JSTests/microbenchmarks/hoist-get-by-offset-tower-with-inferred-types.js	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/JSTests/microbenchmarks/hoist-get-by-offset-tower-with-inferred-types.js	2018-04-09 13:33:13 UTC (rev 230423)
@@ -0,0 +1,19 @@
+function foo(o)
+{
+var result = 0;
+for (var i = 0; i < 100; ++i)
+result += o.f.g.h.i.j.k.l.m.n.o.p.q.r.s.t.u.v.w.x.y.z;
+return result;
+}
+
+noInline(foo);
+
+for (var i = 0; i < 1000; ++i) {
+var o = {f:{g:{h:{i:{j:{k:{l:{m:{n:{o:{p:{q:{r:{s:{t:{u:{v:{w:{x:{y:{z:42};
+for (var j = 0; j < 100; ++j) {
+var result = foo(o);
+if (result != 4200)
+throw "Bad result in loop: " + result;
+}
+}
+


Added: releases/WebKitGTK/webkit-2.20/JSTests/stress/hoist-get-by-offset-with-control-dependent-inferred-type.js (0 => 230423)

--- releases/WebKitGTK/webkit-2.20/JSTests/stress/hoist-get-by-offset-with-control-dependent-inferred-type.js	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/JSTests/stress/hoist-get-by-offset-with-control-dependent-inferred-type.js	2018-04-09 13:33:13 UTC (rev 230423)
@@ -0,0 +1,14 @@
+function f() {
+var a = Array(100).fill(0);
+var ta = new Set(a.map((v,k)=>k));
+var xs = [a, ta];
+var q = 0;
+var t = Date.now();
+for (var i = 0; i < 10; ++i) {
+for (var x of xs[i&1]) q+=x;
+}
+return [Date.now()-t,q];
+}
+noInline(f);
+f();
+


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog (230422 => 230423)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 13:33:06 UTC (rev 230422)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 13:33:13 UTC (rev 230423)
@@ -1,3 +1,13 @@
+2018-03-31  Filip Pizlo  
+
+JSC crash in JIT code with for-of loop and Array/Set iterators
+https://bugs.webkit.org/show_bug.cgi?id=183174
+
+Reviewed by Saam Barati.
+
+* dfg/DFGSafeToExecute.h:
+(JSC::DFG::safeToExecute): Fix the bug by making GetByOffset and friends verify that they are getting the type proof they want at the desired hoisting site.
+
 2018-03-30  Robin Morisset  
 
 Out-of-bounds accesses due to a missing check for MAX_STORAGE_VECTOR_LENGTH in unshiftCountForAnyIndexingType


Modified: 

[webkit-changes] [230422] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230422] releases/WebKitGTK/webkit-2.20








Revision 230422
Author carlo...@webkit.org
Date 2018-04-09 06:33:06 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230101 - Out-of-bounds accesses due to a missing check for MAX_STORAGE_VECTOR_LENGTH in unshiftCountForAnyIndexingType
https://bugs.webkit.org/show_bug.cgi?id=183657
JSTests:

Reviewed by Keith Miller.

* stress/large-unshift-splice.js: Added.
(make_contig_arr):

Source/_javascript_Core:



Reviewed by Keith Miller.

There was just a missing check in unshiftCountForIndexingType.
I've also replaced 'return false' by 'return true' in the case of an 'out-of-memory' exception, because 'return false' means 'please continue to the slow path',
and the slow path has an assert that there is no unhandled exception (line 360 of ArrayPrototype.cpp).
Finally, I made the assert in ensureLength a release assert as it would have caught this bug and prevented it from being a security risk.

* runtime/ArrayPrototype.cpp:
(JSC::unshift):
* runtime/JSArray.cpp:
(JSC::JSArray::unshiftCountWithAnyIndexingType):
* runtime/JSObject.h:
(JSC::JSObject::ensureLength):

Modified Paths

releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/ArrayPrototype.cpp
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/JSArray.cpp
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/JSObject.h


Added Paths

releases/WebKitGTK/webkit-2.20/JSTests/stress/large-unshift-splice.js




Diff

Modified: releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog (230421 => 230422)

--- releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog	2018-04-09 13:32:58 UTC (rev 230421)
+++ releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog	2018-04-09 13:33:06 UTC (rev 230422)
@@ -1,3 +1,13 @@
+2018-03-30  Robin Morisset  
+
+Out-of-bounds accesses due to a missing check for MAX_STORAGE_VECTOR_LENGTH in unshiftCountForAnyIndexingType
+https://bugs.webkit.org/show_bug.cgi?id=183657
+
+Reviewed by Keith Miller.
+
+* stress/large-unshift-splice.js: Added.
+(make_contig_arr):
+
 2018-03-28  Robin Morisset  
 
 appendQuotedJSONString stops on arithmetic overflow instead of propagating it upwards


Added: releases/WebKitGTK/webkit-2.20/JSTests/stress/large-unshift-splice.js (0 => 230422)

--- releases/WebKitGTK/webkit-2.20/JSTests/stress/large-unshift-splice.js	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/JSTests/stress/large-unshift-splice.js	2018-04-09 13:33:06 UTC (rev 230422)
@@ -0,0 +1,16 @@
+//@ skip if $memoryLimited
+
+function make_contig_arr(sz)
+{
+let a = []; 
+while (a.length < sz / 8)
+a.push(3.14); 
+a.length *= 8;
+return a;
+}
+
+let ARRAY_LENGTH = 0x1000;
+let a = make_contig_arr(ARRAY_LENGTH);
+let b = make_contig_arr(0xff00);
+b.unshift(a.length-0x1, 0);
+Array.prototype.splice.apply(a, b);


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog (230421 => 230422)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 13:32:58 UTC (rev 230421)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 13:33:06 UTC (rev 230422)
@@ -1,3 +1,23 @@
+2018-03-30  Robin Morisset  
+
+Out-of-bounds accesses due to a missing check for MAX_STORAGE_VECTOR_LENGTH in unshiftCountForAnyIndexingType
+https://bugs.webkit.org/show_bug.cgi?id=183657
+
+
+Reviewed by Keith Miller.
+
+There was just a missing check in unshiftCountForIndexingType.
+I've also replaced 'return false' by 'return true' in the case of an 'out-of-memory' exception, because 'return false' means 'please continue to the slow path',
+and the slow path has an assert that there is no unhandled exception (line 360 of ArrayPrototype.cpp).
+Finally, I made the assert in ensureLength a release assert as it would have caught this bug and prevented it from being a security risk.
+
+* runtime/ArrayPrototype.cpp:
+(JSC::unshift):
+* runtime/JSArray.cpp:
+(JSC::JSArray::unshiftCountWithAnyIndexingType):
+* runtime/JSObject.h:
+(JSC::JSObject::ensureLength):
+
 2018-03-28  Robin Morisset  
 
 appendQuotedJSONString stops on arithmetic overflow instead of propagating it upwards


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/ArrayPrototype.cpp (230421 => 230422)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/ArrayPrototype.cpp	2018-04-09 13:32:58 UTC (rev 230421)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/ArrayPrototype.cpp	2018-04-09 13:33:06 UTC (rev 230422)
@@ -346,7 +346,7 @@
 RELEASE_ASSERT(currentCount <= (length - header));
 
 // Guard against overflow.
-if (count 

[webkit-changes] [230420] releases/WebKitGTK/webkit-2.20/Source/WebKit

[carlosgc]

Title: [230420] releases/WebKitGTK/webkit-2.20/Source/WebKit








Revision 230420
Author carlo...@webkit.org
Date 2018-04-09 06:00:10 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230074 - [WPE] Floating point exception in WebEventFactory::createWebWheelEvent
https://bugs.webkit.org/show_bug.cgi?id=184037

Patch by Carlos Eduardo Ramalho  on 2018-03-29
Reviewed by Žan Doberšek.

* Shared/wpe/WebEventFactory.cpp:
(WebKit::WebEventFactory::createWebWheelEvent): Use std::copysign() to avoid division by 0.

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/Shared/wpe/WebEventFactory.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog (230419 => 230420)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 13:00:05 UTC (rev 230419)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 13:00:10 UTC (rev 230420)
@@ -1,3 +1,13 @@
+2018-03-29  Carlos Eduardo Ramalho  
+
+[WPE] Floating point exception in WebEventFactory::createWebWheelEvent
+https://bugs.webkit.org/show_bug.cgi?id=184037
+
+Reviewed by Žan Doberšek.
+
+* Shared/wpe/WebEventFactory.cpp:
+(WebKit::WebEventFactory::createWebWheelEvent): Use std::copysign() to avoid division by 0.
+
 2018-03-25  Michael Catanzaro  
 
 [GTK] WebKitWebProcessEnumTypes.h missing from webkit-web-extension.h


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/Shared/wpe/WebEventFactory.cpp (230419 => 230420)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/Shared/wpe/WebEventFactory.cpp	2018-04-09 13:00:05 UTC (rev 230419)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/Shared/wpe/WebEventFactory.cpp	2018-04-09 13:00:10 UTC (rev 230420)
@@ -27,7 +27,7 @@
 #include "WebEventFactory.h"
 
 #include 
-#include 
+#include 
 #include 
 #include 
 
@@ -145,12 +145,12 @@
 WebCore::FloatSize delta;
 switch (event->axis) {
 case Vertical:
-wheelTicks = WebCore::FloatSize(0, event->value / std::abs(event->value));
+wheelTicks = WebCore::FloatSize(0, std::copysign(1, event->value));
 delta = wheelTicks;
 delta.scale(WebCore::Scrollbar::pixelsPerLineStep());
 break;
 case Horizontal:
-wheelTicks = WebCore::FloatSize(event->value / std::abs(event->value), 0);
+wheelTicks = WebCore::FloatSize(std::copysign(1, event->value), 0);
 delta = wheelTicks;
 delta.scale(WebCore::Scrollbar::pixelsPerLineStep());
 break;






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

[webkit-changes] [230419] releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po

[carlosgc]

Title: [230419] releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po








Revision 230419
Author carlo...@webkit.org
Date 2018-04-09 06:00:05 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230073 - [l10n] [pt_BR] Updated Brazilian Portuguese translation
https://bugs.webkit.org/show_bug.cgi?id=184132

Patch by Rafael Fontenelle  on 2018-03-29
Rubber-stamped by Michael Catanzaro.

* pt_BR.po:

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/pt_BR.po




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/ChangeLog (230418 => 230419)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/ChangeLog	2018-04-09 12:59:58 UTC (rev 230418)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/ChangeLog	2018-04-09 13:00:05 UTC (rev 230419)
@@ -1,3 +1,12 @@
+2018-03-29  Rafael Fontenelle  
+
+[l10n] [pt_BR] Updated Brazilian Portuguese translation
+https://bugs.webkit.org/show_bug.cgi?id=184132
+
+Rubber-stamped by Michael Catanzaro.
+
+* pt_BR.po:
+
 2018-03-04  Andika Triwidada  
 
 [l10n] Updated Indonesian translation for WebKitGTK+


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/pt_BR.po (230418 => 230419)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/pt_BR.po	2018-04-09 12:59:58 UTC (rev 230418)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/gtk/po/pt_BR.po	2018-04-09 13:00:05 UTC (rev 230419)
@@ -1,15 +1,15 @@
 # Brazilian Portuguese translations for WebKit package.
-# Copyright © 2017 the webkitgtk authors.
+# Copyright © 2018 the webkitgtk authors.
 # Gustavo Noronha Silva , 2009.
 # Henrique P. Machado , 2011.
 # Enrico Nicoletto , 2013.
-# Rafael Fontenelle , 2016, 2017.
+# Rafael Fontenelle , 2016-2018.
 msgid ""
 msgstr ""
 "Project-Id-Version: webkit HEAD\n"
 "Report-Msgid-Bugs-To: https://bugs.webkit.org/\n"
-"POT-Creation-Date: 2017-08-30 15:26+\n"
-"PO-Revision-Date: 2017-08-18 05:59-0200\n"
+"POT-Creation-Date: 2018-03-29 03:27+\n"
+"PO-Revision-Date: 2018-03-29 08:39-0200\n"
 "Last-Translator: Rafael Fontenelle \n"
 "Language-Team: Brazilian Portuguese \n"
 "Language: pt_BR\n"
@@ -541,14 +541,28 @@
 msgstr "autopreenchimento de informações de contato"
 
 #: ../../LocalizedStrings.cpp:630
+msgid "strong password auto fill"
+msgstr "autopreenchimento de senha forte"
+
+#: ../../LocalizedStrings.cpp:635
+#| msgid "strong password auto fill"
+msgid "strong password confirmation auto fill"
+msgstr "autopreenchimento de confirmação de senha forte"
+
+#: ../../LocalizedStrings.cpp:640
+#| msgid "strong password"
+msgid "Strong Password"
+msgstr "Senha forte"
+
+#: ../../LocalizedStrings.cpp:645
 msgid "Missing Plug-in"
 msgstr "Plug-in em falta"
 
-#: ../../LocalizedStrings.cpp:635
+#: ../../LocalizedStrings.cpp:650
 msgid "Plug-in Failure"
 msgstr "Falha em plug-in"
 
-#: ../../LocalizedStrings.cpp:640
+#: ../../LocalizedStrings.cpp:655
 msgctxt ""
 "Label text to be used if plugin is blocked by a page's Content Security "
 "Policy"
@@ -555,7 +569,7 @@
 msgid "Blocked Plug-in"
 msgstr "Plug-in bloqueado"
 
-#: ../../LocalizedStrings.cpp:645
+#: ../../LocalizedStrings.cpp:660
 msgctxt ""
 "Label text to be used when an insecure plug-in version was blocked from "
 "loading"
@@ -562,328 +576,370 @@
 msgid "Blocked Plug-in"
 msgstr "Plug-in bloqueado"
 
-#: ../../LocalizedStrings.cpp:650
+#: ../../LocalizedStrings.cpp:665
+#| msgid "Snapshotted Plug-In"
+msgctxt ""
+"Label text to be used when an unsupported plug-in was blocked from loading"
+msgid "Unsupported Plug-in"
+msgstr "Plug-in sem suporte"
+
+#: ../../LocalizedStrings.cpp:670
 #, c-format
 msgid "%d files"
 msgstr "%d arquivos"
 
-#: ../../LocalizedStrings.cpp:655
+#: ../../LocalizedStrings.cpp:675
 msgctxt "Unknown filesize FTP directory listing item"
 msgid "Unknown"
 msgstr "Desconhecido"
 
-#: ../../LocalizedStrings.cpp:674
+#: ../../LocalizedStrings.cpp:694
 #, c-format
 msgid "%s %d×%d pixels"
 msgstr "%s %d×%d pixels"
 
-#: ../../LocalizedStrings.cpp:676
+#: ../../LocalizedStrings.cpp:696
 #, c-format
 msgid " %d×%d pixels"
 msgstr " %d×%d pixels"
 
-#: ../../LocalizedStrings.cpp:682
+#: ../../LocalizedStrings.cpp:702
 msgid "Loading…"
 msgstr "Carregando…"
 
-#: ../../LocalizedStrings.cpp:687
+#: ../../LocalizedStrings.cpp:707
 msgid "Live Broadcast"
 msgstr "Transmissão ao vivo"
 
-#: ../../LocalizedStrings.cpp:693
+#: ../../LocalizedStrings.cpp:713
 msgid "audio playback"
 msgstr "Reprodução de áudio"
 
-#: ../../LocalizedStrings.cpp:695
+#: ../../LocalizedStrings.cpp:715
 msgid "video playback"
 msgstr "Reprodução de vídeo"
 
-#: 

[webkit-changes] [230418] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230418] releases/WebKitGTK/webkit-2.20








Revision 230418
Author carlo...@webkit.org
Date 2018-04-09 05:59:58 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r230026 - appendQuotedJSONString stops on arithmetic overflow instead of propagating it upwards
https://bugs.webkit.org/show_bug.cgi?id=183894

Reviewed by Saam Barati.

JSTests:

* stress/json-stringified-overflow.js: Added.
(catch):

Source/_javascript_Core:

Use the return value of appendQuotedJSONString to fail more gracefully when given a string that is too large to handle.

* runtime/JSONObject.cpp:
(JSC::Stringifier::appendStringifiedValue):

Source/WTF:

appendQuotedJSONString now returns a bool indicating whether it succeeded, instead of silently failing when given a string too large
to fit in 4GB.

* wtf/text/StringBuilder.h:
* wtf/text/StringBuilderJSON.cpp:
(WTF::StringBuilder::appendQuotedJSONString):

Modified Paths

releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/JSONObject.cpp
releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/text/StringBuilder.h
releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/text/StringBuilderJSON.cpp


Added Paths

releases/WebKitGTK/webkit-2.20/JSTests/stress/json-stringified-overflow.js




Diff

Modified: releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog (230417 => 230418)

--- releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog	2018-04-09 12:59:46 UTC (rev 230417)
+++ releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog	2018-04-09 12:59:58 UTC (rev 230418)
@@ -1,3 +1,13 @@
+2018-03-28  Robin Morisset  
+
+appendQuotedJSONString stops on arithmetic overflow instead of propagating it upwards
+https://bugs.webkit.org/show_bug.cgi?id=183894
+
+Reviewed by Saam Barati.
+
+* stress/json-stringified-overflow.js: Added.
+(catch):
+
 2018-03-22  Michael Saboff  
 
 Race Condition in arrayProtoFuncReverse() causes wrong results or crash


Added: releases/WebKitGTK/webkit-2.20/JSTests/stress/json-stringified-overflow.js (0 => 230418)

--- releases/WebKitGTK/webkit-2.20/JSTests/stress/json-stringified-overflow.js	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/JSTests/stress/json-stringified-overflow.js	2018-04-09 12:59:58 UTC (rev 230418)
@@ -0,0 +1,3 @@
+try {
+  JSON.stringify("123".padStart(1073741823))
+} catch (e) {}


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog (230417 => 230418)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 12:59:46 UTC (rev 230417)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 12:59:58 UTC (rev 230418)
@@ -1,3 +1,15 @@
+2018-03-28  Robin Morisset  
+
+appendQuotedJSONString stops on arithmetic overflow instead of propagating it upwards
+https://bugs.webkit.org/show_bug.cgi?id=183894
+
+Reviewed by Saam Barati.
+
+Use the return value of appendQuotedJSONString to fail more gracefully when given a string that is too large to handle.
+
+* runtime/JSONObject.cpp:
+(JSC::Stringifier::appendStringifiedValue):
+
 2018-03-25  Saam Barati  
 
 r228149 accidentally removed code that resets m_emptyCursor at the end of a GC


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/JSONObject.cpp (230417 => 230418)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/JSONObject.cpp	2018-04-09 12:59:46 UTC (rev 230417)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/JSONObject.cpp	2018-04-09 12:59:58 UTC (rev 230418)
@@ -357,8 +357,10 @@
 if (value.isString()) {
 const String& string = asString(value)->value(m_exec);
 RETURN_IF_EXCEPTION(scope, StringifyFailed);
-builder.appendQuotedJSONString(string);
-return StringifySucceeded;
+if (builder.appendQuotedJSONString(string))
+return StringifySucceeded;
+throwOutOfMemoryError(m_exec, scope);
+return StringifyFailed;
 }
 
 if (value.isNumber()) {


Modified: releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog (230417 => 230418)

--- releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog	2018-04-09 12:59:46 UTC (rev 230417)
+++ releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog	2018-04-09 12:59:58 UTC (rev 230418)
@@ -1,3 +1,17 @@
+2018-03-28  Robin Morisset  
+
+appendQuotedJSONString stops on arithmetic overflow instead of propagating it upwards
+https://bugs.webkit.org/show_bug.cgi?id=183894
+
+Reviewed by Saam Barati.
+
+appendQuotedJSONString now returns a bool indicating whether it succeeded, instead of silently failing when given a string too large
+to fit in 4GB.
+
+* 

[webkit-changes] [230411] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230411] releases/WebKitGTK/webkit-2.20








Revision 230411
Author carlo...@webkit.org
Date 2018-04-09 05:59:05 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229866 - [WPE] Enable WOFF2 support
https://bugs.webkit.org/show_bug.cgi?id=178158

Reviewed by Frédéric Wang.

.:

* Source/cmake/OptionsWPE.cmake: Add needed logic to detect the Brotli and WOFF2 libraries.

Tools:

* wpe/jhbuild.modules: Add "brotli" and "woff2" to the JHBuild moduleset.

LayoutTests:

* platform/wpe/TestExpectations: Unskip the fast/text/woff2.html and fast/text/woff2-totalsfntsize.html

Modified Paths

releases/WebKitGTK/webkit-2.20/ChangeLog
releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.20/LayoutTests/platform/wpe/TestExpectations
releases/WebKitGTK/webkit-2.20/Source/cmake/OptionsWPE.cmake
releases/WebKitGTK/webkit-2.20/Tools/ChangeLog
releases/WebKitGTK/webkit-2.20/Tools/wpe/jhbuild.modules




Diff

Modified: releases/WebKitGTK/webkit-2.20/ChangeLog (230410 => 230411)

--- releases/WebKitGTK/webkit-2.20/ChangeLog	2018-04-09 12:14:22 UTC (rev 230410)
+++ releases/WebKitGTK/webkit-2.20/ChangeLog	2018-04-09 12:59:05 UTC (rev 230411)
@@ -1,3 +1,12 @@
+2018-03-22  Adrian Perez de Castro  
+
+[WPE] Enable WOFF2 support
+https://bugs.webkit.org/show_bug.cgi?id=178158
+
+Reviewed by Frédéric Wang.
+
+* Source/cmake/OptionsWPE.cmake: Add needed logic to detect the Brotli and WOFF2 libraries.
+
 2018-03-12  Carlos Garcia Campos  
 
 Unreviewed. Update OptionsGTK.cmake and NEWS for 2.20.0 release.


Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog (230410 => 230411)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 12:14:22 UTC (rev 230410)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 12:59:05 UTC (rev 230411)
@@ -1,3 +1,12 @@
+2018-03-22  Adrian Perez de Castro  
+
+[WPE] Enable WOFF2 support
+https://bugs.webkit.org/show_bug.cgi?id=178158
+
+Reviewed by Frédéric Wang.
+
+* platform/wpe/TestExpectations: Unskip the fast/text/woff2.html and fast/text/woff2-totalsfntsize.html
+
 2018-03-09  Nan Wang  
 
 AX: AOM: More accessibility events support


Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/platform/wpe/TestExpectations (230410 => 230411)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/platform/wpe/TestExpectations	2018-04-09 12:14:22 UTC (rev 230410)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/platform/wpe/TestExpectations	2018-04-09 12:59:05 UTC (rev 230411)
@@ -476,6 +476,8 @@
 Bug(WPE) fast/sub-pixel [ Skip ]
 Bug(WPE) fast/table [ Skip ]
 Bug(WPE) fast/text [ Skip ]
+Bug(WPE) fast/text/woff2-totalsfntsize.html [ Pass ]
+Bug(WPE) fast/text/woff2.html [ Pass ]
 Bug(WPE) fast/text-autosizing [ Skip ]
 Bug(WPE) fast/visual-viewport [ Skip ]
 Bug(WPE) fast/writing-mode [ Skip ]


Modified: releases/WebKitGTK/webkit-2.20/Source/cmake/OptionsWPE.cmake (230410 => 230411)

--- releases/WebKitGTK/webkit-2.20/Source/cmake/OptionsWPE.cmake	2018-04-09 12:14:22 UTC (rev 230410)
+++ releases/WebKitGTK/webkit-2.20/Source/cmake/OptionsWPE.cmake	2018-04-09 12:59:05 UTC (rev 230411)
@@ -33,6 +33,11 @@
 WEBKIT_OPTION_DEFAULT_PORT_VALUE(ENABLE_TOUCH_EVENTS PRIVATE ON)
 WEBKIT_OPTION_DEFAULT_PORT_VALUE(ENABLE_WEBGL2 PRIVATE ${ENABLE_EXPERIMENTAL_FEATURES})
 
+# Public options specific to the WPE port. Do not add any options here unless
+# there is a strong reason we should support changing the value of the option,
+# and the option is not relevant to any other WebKit ports.
+WEBKIT_OPTION_DEFINE(USE_WOFF2 "Whether to enable support for WOFF2 Web Fonts." PUBLIC ON)
+
 # Private options specific to the WPE port.
 WEBKIT_OPTION_DEFINE(USE_OPENVR "Whether to use OpenVR as WebVR backend." PRIVATE ${ENABLE_EXPERIMENTAL_FEATURES})
 
@@ -71,6 +76,13 @@
 find_package(WPEBackend REQUIRED)
 find_package(ZLIB REQUIRED)
 
+if (USE_WOFF2)
+find_package(WOFF2Dec 1.0.2)
+if (NOT WOFF2DEC_FOUND)
+message(FATAL_ERROR "liwoff2dec is needed for USE_WOFF2.")
+endif ()
+endif ()
+
 if (ENABLE_ACCELERATED_2D_CANVAS)
 find_package(CairoGL 1.10.2 REQUIRED COMPONENTS cairo-egl)
 endif ()


Modified: releases/WebKitGTK/webkit-2.20/Tools/ChangeLog (230410 => 230411)

--- releases/WebKitGTK/webkit-2.20/Tools/ChangeLog	2018-04-09 12:14:22 UTC (rev 230410)
+++ releases/WebKitGTK/webkit-2.20/Tools/ChangeLog	2018-04-09 12:59:05 UTC (rev 230411)
@@ -1,3 +1,12 @@
+2018-03-22  Adrian Perez de Castro  
+
+[WPE] Enable WOFF2 support
+https://bugs.webkit.org/show_bug.cgi?id=178158
+
+Reviewed by Frédéric Wang.
+
+* wpe/jhbuild.modules: Add "brotli" and "woff2" to the JHBuild moduleset.
+
 2018-03-07  Carlos Garcia Campos  
 
 [GTK][WPE] Leak checker is not working in WebKitGLib web process tests


Modified: 

[webkit-changes] [230417] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230417] releases/WebKitGTK/webkit-2.20








Revision 230417
Author carlo...@webkit.org
Date 2018-04-09 05:59:46 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229994 - [GTK] Layout test editing/deleting/delete-surrogatepair.html crashing with CRITICAL **: enchant_dict_check: assertion 'g_utf8_validate(word, len, NULL)' failed
https://bugs.webkit.org/show_bug.cgi?id=176799

Reviewed by Carlos Garcia Campos.

Source/WebCore:

The length of a surrogate-pair UTF-16 character is 2 even though
the number of characters is 1. An incorrect string length was
passed to enchant_dict_check if the string contains a
surrogate-pair character because the length was calculated by
applying UTF-16 character position to UTF-8 string.

No new tests (Covered by existing tests).

* platform/text/enchant/TextCheckerEnchant.cpp:
(WebCore::TextCheckerEnchant::checkSpellingOfWord): Changed the
type of an argument `word` from CString to String. Convert a
substring of the argument into UTF-8.
(WebCore::TextCheckerEnchant::checkSpellingOfString): Pass the
original UTF-16 string to checkSpellingOfWord instead of a
converted UTF-8 string.
* platform/text/enchant/TextCheckerEnchant.h: Changed the type of
an argument `word` from CString to String.

LayoutTests:

* platform/gtk/TestExpectations: Unmarked editing/deleting/delete-surrogatepair.html.

Modified Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/text/enchant/TextCheckerEnchant.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/text/enchant/TextCheckerEnchant.h




Diff

Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog (230416 => 230417)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 12:59:39 UTC (rev 230416)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 12:59:46 UTC (rev 230417)
@@ -1,3 +1,12 @@
+2018-03-27  Fujii Hironori  
+
+[GTK] Layout test editing/deleting/delete-surrogatepair.html crashing with CRITICAL **: enchant_dict_check: assertion 'g_utf8_validate(word, len, NULL)' failed
+https://bugs.webkit.org/show_bug.cgi?id=176799
+
+Reviewed by Carlos Garcia Campos.
+
+* platform/gtk/TestExpectations: Unmarked editing/deleting/delete-surrogatepair.html.
+
 2018-03-22  Zalan Bujtas  
 
 [Simple line layout] Text with letter spacing is not positioned properly.


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230416 => 230417)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:59:39 UTC (rev 230416)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:59:46 UTC (rev 230417)
@@ -1,3 +1,28 @@
+2018-03-27  Fujii Hironori  
+
+[GTK] Layout test editing/deleting/delete-surrogatepair.html crashing with CRITICAL **: enchant_dict_check: assertion 'g_utf8_validate(word, len, NULL)' failed
+https://bugs.webkit.org/show_bug.cgi?id=176799
+
+Reviewed by Carlos Garcia Campos.
+
+The length of a surrogate-pair UTF-16 character is 2 even though
+the number of characters is 1. An incorrect string length was
+passed to enchant_dict_check if the string contains a
+surrogate-pair character because the length was calculated by
+applying UTF-16 character position to UTF-8 string.
+
+No new tests (Covered by existing tests).
+
+* platform/text/enchant/TextCheckerEnchant.cpp:
+(WebCore::TextCheckerEnchant::checkSpellingOfWord): Changed the
+type of an argument `word` from CString to String. Convert a
+substring of the argument into UTF-8.
+(WebCore::TextCheckerEnchant::checkSpellingOfString): Pass the
+original UTF-16 string to checkSpellingOfWord instead of a
+converted UTF-8 string.
+* platform/text/enchant/TextCheckerEnchant.h: Changed the type of
+an argument `word` from CString to String.
+
 2018-03-22  Zalan Bujtas  
 
 [Simple line layout] Text with letter spacing is not positioned properly.


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/text/enchant/TextCheckerEnchant.cpp (230416 => 230417)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/text/enchant/TextCheckerEnchant.cpp	2018-04-09 12:59:39 UTC (rev 230416)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/text/enchant/TextCheckerEnchant.cpp	2018-04-09 12:59:46 UTC (rev 230417)
@@ -64,14 +64,12 @@
 enchant_dict_add(dictionary, word.utf8().data(), -1);
 }
 
-void TextCheckerEnchant::checkSpellingOfWord(const CString& word, int start, int end, int& misspellingLocation, int& misspellingLength)
+void TextCheckerEnchant::checkSpellingOfWord(const String& word, int start, int end, int& misspellingLocation, int& misspellingLength)
 {
-const char* string = word.data();
-  

[webkit-changes] [230415] releases/WebKitGTK/webkit-2.20/Source/WTF

[carlosgc]

Title: [230415] releases/WebKitGTK/webkit-2.20/Source/WTF








Revision 230415
Author carlo...@webkit.org
Date 2018-04-09 05:59:33 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229965 - WebProcess memory monitor: use %zu format specifier for size_t
https://bugs.webkit.org/show_bug.cgi?id=183997

Reviewed by Michael Catanzaro.

usse %zu format specifier for size_t instead of %lu.

* wtf/MemoryPressureHandler.cpp:
(WTF::MemoryPressureHandler::shrinkOrDie):
(WTF::MemoryPressureHandler::measurementTimerFired):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/MemoryPressureHandler.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog (230414 => 230415)

--- releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog	2018-04-09 12:59:29 UTC (rev 230414)
+++ releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog	2018-04-09 12:59:33 UTC (rev 230415)
@@ -1,3 +1,16 @@
+2018-03-25  Carlos Alberto Lopez Perez  
+
+WebProcess memory monitor: use %zu format specifier for size_t
+https://bugs.webkit.org/show_bug.cgi?id=183997
+
+Reviewed by Michael Catanzaro.
+
+usse %zu format specifier for size_t instead of %lu.
+
+* wtf/MemoryPressureHandler.cpp:
+(WTF::MemoryPressureHandler::shrinkOrDie):
+(WTF::MemoryPressureHandler::measurementTimerFired):
+
 2018-03-23  Carlos Alberto Lopez Perez  
 
 WebProcess memory monitor thresholds should be better tuned for embedded systems.


Modified: releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/MemoryPressureHandler.cpp (230414 => 230415)

--- releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/MemoryPressureHandler.cpp	2018-04-09 12:59:29 UTC (rev 230414)
+++ releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/MemoryPressureHandler.cpp	2018-04-09 12:59:33 UTC (rev 230415)
@@ -141,7 +141,7 @@
 
 auto footprint = memoryFootprint();
 RELEASE_ASSERT(footprint);
-RELEASE_LOG(MemoryPressure, "New memory footprint: %lu MB", footprint.value() / MB);
+RELEASE_LOG(MemoryPressure, "New memory footprint: %zu MB", footprint.value() / MB);
 
 if (footprint.value() < thresholdForMemoryKill()) {
 RELEASE_LOG(MemoryPressure, "Shrank below memory kill threshold. Process gets to live.");
@@ -149,7 +149,7 @@
 return;
 }
 
-WTFLogAlways("Unable to shrink memory footprint of process (%lu MB) below the kill thresold (%lu MB). Killed\n", footprint.value() / MB, thresholdForMemoryKill() / MB);
+WTFLogAlways("Unable to shrink memory footprint of process (%zu MB) below the kill thresold (%zu MB). Killed\n", footprint.value() / MB, thresholdForMemoryKill() / MB);
 RELEASE_ASSERT(m_memoryKillCallback);
 m_memoryKillCallback();
 }
@@ -171,7 +171,7 @@
 if (!footprint)
 return;
 
-RELEASE_LOG(MemoryPressure, "Current memory footprint: %lu MB", footprint.value() / MB);
+RELEASE_LOG(MemoryPressure, "Current memory footprint: %zu MB", footprint.value() / MB);
 if (footprint.value() >= thresholdForMemoryKill()) {
 shrinkOrDie();
 return;






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

[webkit-changes] [230413] releases/WebKitGTK/webkit-2.20/Source/WTF

[carlosgc]

Title: [230413] releases/WebKitGTK/webkit-2.20/Source/WTF








Revision 230413
Author carlo...@webkit.org
Date 2018-04-09 05:59:23 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229894 - WebProcess memory monitor thresholds should be better tuned for embedded systems.
https://bugs.webkit.org/show_bug.cgi?id=183773

Reviewed by Yusuke Suzuki.

Take into account the total system RAM for the thresholds calculation.

For systems with more than 3GB the conservative and strict thresholds remain as they are,
but for systems with less RAM the thresholds are dynamically configured as follows:

- Conservative threshold (release non critical memory) if WebProcess using more than 33% of the total RAM.
- Strict threshold (release all possible memory) if WebProcess using more than 50% of the total RAM.

The Kill threshold is also modified. Now it is capped at 90% of the total RAM.

* wtf/MemoryPressureHandler.cpp:
(WTF::thresholdForMemoryKillWithProcessState):
(WTF::thresholdForPolicy):
(WTF::MemoryPressureHandler::shrinkOrDie):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/MemoryPressureHandler.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog (230412 => 230413)

--- releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog	2018-04-09 12:59:18 UTC (rev 230412)
+++ releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog	2018-04-09 12:59:23 UTC (rev 230413)
@@ -1,3 +1,25 @@
+2018-03-23  Carlos Alberto Lopez Perez  
+
+WebProcess memory monitor thresholds should be better tuned for embedded systems.
+https://bugs.webkit.org/show_bug.cgi?id=183773
+
+Reviewed by Yusuke Suzuki.
+
+Take into account the total system RAM for the thresholds calculation.
+
+For systems with more than 3GB the conservative and strict thresholds remain as they are,
+but for systems with less RAM the thresholds are dynamically configured as follows:
+
+- Conservative threshold (release non critical memory) if WebProcess using more than 33% of the total RAM.
+- Strict threshold (release all possible memory) if WebProcess using more than 50% of the total RAM.
+
+The Kill threshold is also modified. Now it is capped at 90% of the total RAM.
+
+* wtf/MemoryPressureHandler.cpp:
+(WTF::thresholdForMemoryKillWithProcessState):
+(WTF::thresholdForPolicy):
+(WTF::MemoryPressureHandler::shrinkOrDie):
+
 2018-03-08  Zan Dobersek  
 
 [GLib] RunLoop::wakeUp(): use a zero value instead of the monotonic time


Modified: releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/MemoryPressureHandler.cpp (230412 => 230413)

--- releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/MemoryPressureHandler.cpp	2018-04-09 12:59:18 UTC (rev 230412)
+++ releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/MemoryPressureHandler.cpp	2018-04-09 12:59:23 UTC (rev 230413)
@@ -28,6 +28,7 @@
 
 #include 
 #include 
+#include 
 
 #define LOG_CHANNEL_PREFIX Log
 
@@ -84,20 +85,17 @@
 
 static size_t thresholdForMemoryKillWithProcessState(WebsamProcessState processState, unsigned tabCount)
 {
+size_t baseThreshold = 2 * GB;
 #if CPU(X86_64) || CPU(ARM64)
-size_t baseThreshold;
 if (processState == WebsamProcessState::Active)
 baseThreshold = 4 * GB;
-else
-baseThreshold = 2 * GB;
-if (tabCount <= 1)
-return baseThreshold;
-return baseThreshold + (std::min(tabCount - 1, 4u) * 1 * GB);
+if (tabCount > 1)
+baseThreshold += std::min(tabCount - 1, 4u) * 1 * GB;
 #else
-UNUSED_PARAM(processState);
-UNUSED_PARAM(tabCount);
-return 3 * GB;
+if ((tabCount > 1) || (processState == WebsamProcessState::Active))
+baseThreshold = 3 * GB;
 #endif
+return std::min(baseThreshold, static_cast(ramSize() * 0.9));
 }
 
 void MemoryPressureHandler::setPageCount(unsigned pageCount)
@@ -114,11 +112,12 @@
 
 static size_t thresholdForPolicy(MemoryUsagePolicy policy)
 {
+const size_t baseThresholdForPolicy = std::min(3 * GB, ramSize());
 switch (policy) {
 case MemoryUsagePolicy::Conservative:
-return 1 * GB;
+return baseThresholdForPolicy / 3;
 case MemoryUsagePolicy::Strict:
-return 1.5 * GB;
+return baseThresholdForPolicy / 2;
 case MemoryUsagePolicy::Unrestricted:
 default:
 ASSERT_NOT_REACHED();
@@ -150,6 +149,7 @@
 return;
 }
 
+WTFLogAlways("Unable to shrink memory footprint of process (%lu MB) below the kill thresold (%lu MB). Killed\n", footprint.value() / MB, thresholdForMemoryKill() / MB);
 RELEASE_ASSERT(m_memoryKillCallback);
 m_memoryKillCallback();
 }






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

[webkit-changes] [230412] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230412] releases/WebKitGTK/webkit-2.20








Revision 230412
Author carlo...@webkit.org
Date 2018-04-09 05:59:18 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229867 - [Simple line layout] Text with letter spacing is not positioned properly.
https://bugs.webkit.org/show_bug.cgi?id=183079


Reviewed by Antti Koivisto.

Source/WebCore:

We need to recompute RenderText::m_canUseSimplifiedTextMeasuring when the font cascade changes
since we might not be able to use the fast path anymore.

Test: fast/text/simple-line-layout-dynamic-letter-word-spacing.html

* rendering/RenderText.cpp:
(WebCore::RenderText::styleDidChange):

LayoutTests:

* fast/text/simple-line-layout-dynamic-letter-word-spacing-expected.html: Added.
* fast/text/simple-line-layout-dynamic-letter-word-spacing.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/RenderText.cpp


Added Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/fast/text/simple-line-layout-dynamic-letter-word-spacing-expected.html
releases/WebKitGTK/webkit-2.20/LayoutTests/fast/text/simple-line-layout-dynamic-letter-word-spacing.html




Diff

Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog (230411 => 230412)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 12:59:05 UTC (rev 230411)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 12:59:18 UTC (rev 230412)
@@ -1,3 +1,14 @@
+2018-03-22  Zalan Bujtas  
+
+[Simple line layout] Text with letter spacing is not positioned properly.
+https://bugs.webkit.org/show_bug.cgi?id=183079
+
+
+Reviewed by Antti Koivisto.
+
+* fast/text/simple-line-layout-dynamic-letter-word-spacing-expected.html: Added.
+* fast/text/simple-line-layout-dynamic-letter-word-spacing.html: Added.
+
 2018-03-22  Adrian Perez de Castro  
 
 [WPE] Enable WOFF2 support


Added: releases/WebKitGTK/webkit-2.20/LayoutTests/fast/text/simple-line-layout-dynamic-letter-word-spacing-expected.html (0 => 230412)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/fast/text/simple-line-layout-dynamic-letter-word-spacing-expected.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/fast/text/simple-line-layout-dynamic-letter-word-spacing-expected.html	2018-04-09 12:59:18 UTC (rev 230412)
@@ -0,0 +1,21 @@
+
+
+
+
+div {
+width: 200px;
+}
+
+
+
+foobar foobar foobar foobar
+foobar foobar foobar foobar
+
+if (window.internals)
+internals.settings.setSimpleLineLayoutEnabled(false);
+document.body.offsetHeight;
+letterSpacing.style.letterSpacing = "10px";
+wordSpacing.style.wordSpacing = "20px";
+
+
+


Added: releases/WebKitGTK/webkit-2.20/LayoutTests/fast/text/simple-line-layout-dynamic-letter-word-spacing.html (0 => 230412)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/fast/text/simple-line-layout-dynamic-letter-word-spacing.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/fast/text/simple-line-layout-dynamic-letter-word-spacing.html	2018-04-09 12:59:18 UTC (rev 230412)
@@ -0,0 +1,19 @@
+
+
+
+
+div {
+width: 200px;
+}
+
+
+
+foobar foobar foobar foobar
+foobar foobar foobar foobar
+
+document.body.offsetHeight;
+letterSpacing.style.letterSpacing = "10px";
+wordSpacing.style.wordSpacing = "20px";
+
+
+


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230411 => 230412)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:59:05 UTC (rev 230411)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:59:18 UTC (rev 230412)
@@ -1,3 +1,19 @@
+2018-03-22  Zalan Bujtas  
+
+[Simple line layout] Text with letter spacing is not positioned properly.
+https://bugs.webkit.org/show_bug.cgi?id=183079
+
+
+Reviewed by Antti Koivisto.
+
+We need to recompute RenderText::m_canUseSimplifiedTextMeasuring when the font cascade changes
+since we might not be able to use the fast path anymore.
+
+Test: fast/text/simple-line-layout-dynamic-letter-word-spacing.html
+
+* rendering/RenderText.cpp:
+(WebCore::RenderText::styleDidChange):
+
 2018-03-22  Adrian Perez de Castro  
 
 [WPE][GTK] Build failure when ENABLE_VIDEO, ENABLE_WEB_AUDIO and ENABLE_XSLT are disabled


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/RenderText.cpp (230411 => 230412)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/RenderText.cpp	2018-04-09 12:59:05 UTC (rev 230411)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/RenderText.cpp	2018-04-09 12:59:18 UTC (rev 230412)
@@ -263,13 +263,14 @@
 if (!oldStyle) {
 m_useBackslashAsYenSymbol = computeUseBackslashAsYenSymbol();
 needsResetText = m_useBackslashAsYenSymbol;
-

[webkit-changes] [230414] releases/WebKitGTK/webkit-2.20/Source/JavaScriptCore

[carlosgc]

Title: [230414] releases/WebKitGTK/webkit-2.20/Source/_javascript_Core








Revision 230414
Author carlo...@webkit.org
Date 2018-04-09 05:59:29 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229962 - r228149 accidentally removed code that resets m_emptyCursor at the end of a GC
https://bugs.webkit.org/show_bug.cgi?id=183995

Reviewed by Filip Pizlo.

The removal of this line of code was unintended and happened during some
refactoring Fil was doing. The consequence of removing this line of code
is that the m_emptyCursor became a monotonically increasing integer, leading
the cursor to usually being out of bounds of the block range (depending on
what the program is doing). This made the functionality of finding an empty
block to steal almost always fail.

* heap/BlockDirectory.cpp:
(JSC::BlockDirectory::prepareForAllocation):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/heap/BlockDirectory.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog (230413 => 230414)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 12:59:23 UTC (rev 230413)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 12:59:29 UTC (rev 230414)
@@ -1,3 +1,20 @@
+2018-03-25  Saam Barati  
+
+r228149 accidentally removed code that resets m_emptyCursor at the end of a GC
+https://bugs.webkit.org/show_bug.cgi?id=183995
+
+Reviewed by Filip Pizlo.
+
+The removal of this line of code was unintended and happened during some
+refactoring Fil was doing. The consequence of removing this line of code
+is that the m_emptyCursor became a monotonically increasing integer, leading
+the cursor to usually being out of bounds of the block range (depending on
+what the program is doing). This made the functionality of finding an empty
+block to steal almost always fail.
+
+* heap/BlockDirectory.cpp:
+(JSC::BlockDirectory::prepareForAllocation):
+
 2018-03-22  Michael Saboff  
 
 Race Condition in arrayProtoFuncReverse() causes wrong results or crash


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/heap/BlockDirectory.cpp (230413 => 230414)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/heap/BlockDirectory.cpp	2018-04-09 12:59:23 UTC (rev 230413)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/heap/BlockDirectory.cpp	2018-04-09 12:59:29 UTC (rev 230414)
@@ -191,6 +191,7 @@
 });
 
 m_unsweptCursor = 0;
+m_emptyCursor = 0;
 
 m_eden.clearAll();
 






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

[webkit-changes] [230416] releases/WebKitGTK/webkit-2.20/Source/WebKit

[carlosgc]

Title: [230416] releases/WebKitGTK/webkit-2.20/Source/WebKit








Revision 230416
Author carlo...@webkit.org
Date 2018-04-09 05:59:39 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229971 - [GTK] WebKitWebProcessEnumTypes.h missing from webkit-web-extension.h
https://bugs.webkit.org/show_bug.cgi?id=183998

Reviewed by Carlos Garcia Campos.

* WebProcess/InjectedBundle/API/gtk/webkit-web-extension.h:

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/InjectedBundle/API/gtk/webkit-web-extension.h




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog (230415 => 230416)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 12:59:33 UTC (rev 230415)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 12:59:39 UTC (rev 230416)
@@ -1,3 +1,12 @@
+2018-03-25  Michael Catanzaro  
+
+[GTK] WebKitWebProcessEnumTypes.h missing from webkit-web-extension.h
+https://bugs.webkit.org/show_bug.cgi?id=183998
+
+Reviewed by Carlos Garcia Campos.
+
+* WebProcess/InjectedBundle/API/gtk/webkit-web-extension.h:
+
 2018-03-07  Carlos Garcia Campos  
 
 [GTK][WPE] Leak checker is not working in WebKitGLib web process tests


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/InjectedBundle/API/gtk/webkit-web-extension.h (230415 => 230416)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/InjectedBundle/API/gtk/webkit-web-extension.h	2018-04-09 12:59:33 UTC (rev 230415)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/InjectedBundle/API/gtk/webkit-web-extension.h	2018-04-09 12:59:39 UTC (rev 230416)
@@ -39,6 +39,7 @@
 #include 
 #include 
 #include 
+#include 
 
 #include 
 






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

[webkit-changes] [230404] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230404] releases/WebKitGTK/webkit-2.20








Revision 230404
Author carlo...@webkit.org
Date 2018-04-09 05:13:44 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229540 - Add a query and fragment exception to history API's unique origin restriction.
https://bugs.webkit.org/show_bug.cgi?id=183028

Patch by Danyao Wang  on 2018-03-12
Reviewed by Brent Fulgham.

Tests: http/tests/navigation/pushstate-at-unique-origin-denied.php
   Tools/TestWebKitAPI/Tests/WebCore/URL.cpp

* page/History.cpp:
(WebCore::History::stateObjectAdded):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/page/History.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/URL.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/URL.h
releases/WebKitGTK/webkit-2.20/Tools/TestWebKitAPI/Tests/WebCore/URL.cpp


Added Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/navigation/pushstate-at-unique-origin-denied-expected.txt
releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/navigation/pushstate-at-unique-origin-denied.php




Diff

Added: releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/navigation/pushstate-at-unique-origin-denied-expected.txt (0 => 230404)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/navigation/pushstate-at-unique-origin-denied-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/navigation/pushstate-at-unique-origin-denied-expected.txt	2018-04-09 12:13:44 UTC (rev 230404)
@@ -0,0 +1,6 @@
+
+PASS pushState to a new path in unique origin should fail with SecurityError 
+PASS pushState #hash in unique origin should not fail with SecurityError 
+PASS pushState ?query in unique origin should not fail with SecurityError 
+PASS pushState ?query#hash in unique origin should not fail with SecurityError 
+


Added: releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/navigation/pushstate-at-unique-origin-denied.php (0 => 230404)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/navigation/pushstate-at-unique-origin-denied.php	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/navigation/pushstate-at-unique-origin-denied.php	2018-04-09 12:13:44 UTC (rev 230404)
@@ -0,0 +1,40 @@
+
+
+var originURL = document.URL;
+test(function () {
+assert_throws('SecurityError', function () {
+history.pushState(null, null, originURL + "/path");
+});
+}, 'pushState to a new path in unique origin should fail with SecurityError');
+
+test(function () {
+try {
+history.pushState(null, null, originURL + "#hash");
+done();
+} catch (e) {
+assert_unreached("pushState #hash should not fail.");
+}
+}, 'pushState #hash in unique origin should not fail with SecurityError');
+
+test(function () {
+try {
+history.pushState(null, null, originURL + "?query");
+done();
+} catch (e) {
+assert_unreached("pushState ?query should not fail.");
+}
+}, 'pushState ?query in unique origin should not fail with SecurityError');
+
+test(function () {
+try {
+history.pushState(null, null, originURL + "?query#hash");
+done();
+} catch (e) {
+assert_unreached("pushState ?query#hash should not fail.");
+}
+}, 'pushState ?query#hash in unique origin should not fail with SecurityError');
+


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230403 => 230404)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 11:28:02 UTC (rev 230403)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:13:44 UTC (rev 230404)
@@ -1,3 +1,16 @@
+2018-03-12  Danyao Wang  
+
+Add a query and fragment exception to history API's unique origin restriction.
+https://bugs.webkit.org/show_bug.cgi?id=183028
+
+Reviewed by Brent Fulgham.
+
+Tests: http/tests/navigation/pushstate-at-unique-origin-denied.php
+   Tools/TestWebKitAPI/Tests/WebCore/URL.cpp
+
+* page/History.cpp:
+(WebCore::History::stateObjectAdded):
+
 2018-03-07  Zalan Bujtas  
 
 Invalid innerTextRenderer in RenderTextControlSingleLine::styleDidChange()


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/page/History.cpp (230403 => 230404)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/page/History.cpp	2018-04-09 11:28:02 UTC (rev 230403)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/page/History.cpp	2018-04-09 12:13:44 UTC (rev 230404)
@@ -183,7 +183,13 @@
 };
 if (!protocolHostAndPortAreEqual(fullURL, documentURL) || fullURL.user() != documentURL.user() || fullURL.pass() != documentURL.pass())
 return createBlockedURLSecurityErrorWithMessageSuffix("Protocols, domains, ports, usernames, and passwords must match.");
-if (!m_frame->document()->securityOrigin().canRequest(fullURL) && 

[webkit-changes] [230409] releases/WebKitGTK/webkit-2.20/Source/WebCore

[carlosgc]

Title: [230409] releases/WebKitGTK/webkit-2.20/Source/WebCore








Revision 230409
Author carlo...@webkit.org
Date 2018-04-09 05:14:15 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229848 - [WPE][GTK] Build failure when ENABLE_VIDEO, ENABLE_WEB_AUDIO and ENABLE_XSLT are disabled
https://bugs.webkit.org/show_bug.cgi?id=183896

Reviewed by Yusuke Suzuki.

No new tests needed.

* bindings/js/JSWebAnimationCustom.cpp: Add missing #include of Document.h

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/bindings/js/JSWebAnimationCustom.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230408 => 230409)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:14:09 UTC (rev 230408)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:14:15 UTC (rev 230409)
@@ -1,3 +1,14 @@
+2018-03-22  Adrian Perez de Castro  
+
+[WPE][GTK] Build failure when ENABLE_VIDEO, ENABLE_WEB_AUDIO and ENABLE_XSLT are disabled
+https://bugs.webkit.org/show_bug.cgi?id=183896
+
+Reviewed by Yusuke Suzuki.
+
+No new tests needed.
+
+* bindings/js/JSWebAnimationCustom.cpp: Add missing #include of Document.h
+
 2018-03-21  Carlos Alberto Lopez Perez  
 
 [WPE] Build failure with ENABLE_VIDEO=OFF when GStreamer is not available


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/bindings/js/JSWebAnimationCustom.cpp (230408 => 230409)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/bindings/js/JSWebAnimationCustom.cpp	2018-04-09 12:14:09 UTC (rev 230408)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/bindings/js/JSWebAnimationCustom.cpp	2018-04-09 12:14:15 UTC (rev 230409)
@@ -26,6 +26,7 @@
 #include "config.h"
 #include "JSWebAnimation.h"
 
+#include "Document.h"
 #include "JSAnimationEffect.h"
 #include "JSAnimationTimeline.h"
 #include "JSDOMConstructor.h"






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

[webkit-changes] [230408] releases/WebKitGTK/webkit-2.20/Source/WebCore

[carlosgc]

Title: [230408] releases/WebKitGTK/webkit-2.20/Source/WebCore








Revision 230408
Author carlo...@webkit.org
Date 2018-04-09 05:14:09 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229803 - [WPE] Build failure with ENABLE_VIDEO=OFF when GStreamer is not available
https://bugs.webkit.org/show_bug.cgi?id=183811

Reviewed by Philippe Normand.

Add build guards for USE_GSTREAMER.

No new tests, it is a build fix.

* platform/graphics/gstreamer/GStreamerUtilities.h:

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/graphics/gstreamer/GStreamerUtilities.h




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230407 => 230408)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:14:03 UTC (rev 230407)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:14:09 UTC (rev 230408)
@@ -1,3 +1,16 @@
+2018-03-21  Carlos Alberto Lopez Perez  
+
+[WPE] Build failure with ENABLE_VIDEO=OFF when GStreamer is not available
+https://bugs.webkit.org/show_bug.cgi?id=183811
+
+Reviewed by Philippe Normand.
+
+Add build guards for USE_GSTREAMER.
+
+No new tests, it is a build fix.
+
+* platform/graphics/gstreamer/GStreamerUtilities.h:
+
 2018-03-09  Nan Wang  
 
 AX: AOM: More accessibility events support


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/graphics/gstreamer/GStreamerUtilities.h (230407 => 230408)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/graphics/gstreamer/GStreamerUtilities.h	2018-04-09 12:14:03 UTC (rev 230407)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/platform/graphics/gstreamer/GStreamerUtilities.h	2018-04-09 12:14:09 UTC (rev 230408)
@@ -20,6 +20,7 @@
 #pragma once
 
 
+#if USE(GSTREAMER)
 #include "FloatSize.h"
 #include 
 #include 
@@ -84,3 +85,4 @@
 #ifndef GST_BUFFER_DTS_OR_PTS
 #define GST_BUFFER_DTS_OR_PTS(buffer) (GST_BUFFER_DTS_IS_VALID(buffer) ? GST_BUFFER_DTS(buffer) : GST_BUFFER_PTS(buffer))
 #endif
+#endif // USE(GSTREAMER)






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

[webkit-changes] [230405] releases/WebKitGTK/webkit-2.20/Source/bmalloc

[carlosgc]

Title: [230405] releases/WebKitGTK/webkit-2.20/Source/bmalloc








Revision 230405
Author carlo...@webkit.org
Date 2018-04-09 05:13:51 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229669 - Improve error message when Gigacage cannot allocate virtual memory
https://bugs.webkit.org/show_bug.cgi?id=183329

Reviewed by Filip Pizlo.

We've discovered that Deja Dup monitor sets a virtual memory limit, breaking Gigacage. Since
it runs in the background on a fresh out-of-the-box install of Ubuntu, this is not good.
That will have to be fixed by Deja Dup, but there is concern that other applications might
try this, or that users will set a virtual memory limit for the entire desktop session. Of
particular concern is the possibility that users might have copypasted a ulimit line into
a session startup script without understanding it. Let's try to make it slightly easier to
understand what's going wrong.

* bmalloc/Gigacage.cpp:
(Gigacage::ensureGigacage):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/bmalloc/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/bmalloc/bmalloc/Gigacage.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/bmalloc/ChangeLog (230404 => 230405)

--- releases/WebKitGTK/webkit-2.20/Source/bmalloc/ChangeLog	2018-04-09 12:13:44 UTC (rev 230404)
+++ releases/WebKitGTK/webkit-2.20/Source/bmalloc/ChangeLog	2018-04-09 12:13:51 UTC (rev 230405)
@@ -1,3 +1,21 @@
+2018-03-16  Michael Catanzaro  
+
+Improve error message when Gigacage cannot allocate virtual memory
+https://bugs.webkit.org/show_bug.cgi?id=183329
+
+Reviewed by Filip Pizlo.
+
+We've discovered that Deja Dup monitor sets a virtual memory limit, breaking Gigacage. Since
+it runs in the background on a fresh out-of-the-box install of Ubuntu, this is not good.
+That will have to be fixed by Deja Dup, but there is concern that other applications might
+try this, or that users will set a virtual memory limit for the entire desktop session. Of
+particular concern is the possibility that users might have copypasted a ulimit line into
+a session startup script without understanding it. Let's try to make it slightly easier to
+understand what's going wrong.
+
+* bmalloc/Gigacage.cpp:
+(Gigacage::ensureGigacage):
+
 2018-02-05  JF Bastien  
 
 Gigacage: enable only for WebContent process and token executables


Modified: releases/WebKitGTK/webkit-2.20/Source/bmalloc/bmalloc/Gigacage.cpp (230404 => 230405)

--- releases/WebKitGTK/webkit-2.20/Source/bmalloc/bmalloc/Gigacage.cpp	2018-04-09 12:13:44 UTC (rev 230404)
+++ releases/WebKitGTK/webkit-2.20/Source/bmalloc/bmalloc/Gigacage.cpp	2018-04-09 12:13:51 UTC (rev 230405)
@@ -151,6 +151,7 @@
 if (GIGACAGE_ALLOCATION_CAN_FAIL)
 return;
 fprintf(stderr, "FATAL: Could not allocate gigacage memory with maxAlignment = %lu, totalSize = %lu.\n", maxAlignment, totalSize);
+fprintf(stderr, "(Make sure you have not set a virtual memory limit.)\n");
 BCRASH();
 }
 






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

[webkit-changes] [230406] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230406] releases/WebKitGTK/webkit-2.20








Revision 230406
Author carlo...@webkit.org
Date 2018-04-09 05:13:57 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229499 - AX: AOM: More accessibility events support
https://bugs.webkit.org/show_bug.cgi?id=183023


Reviewed by Chris Fleizach.

Source/WebCore:

The test is crashing when we call updateBackingStore when
the AXObjectCache object is gone. Added a check to fix that.

Modified the test by using the right format of setTimeout and extended the delay.

* accessibility/AccessibilityObject.cpp:
(WebCore::AccessibilityObject::updateBackingStore):

LayoutTests:

* accessibility/mac/AOM-events-webarea-crash.html:

Modified Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/accessibility/AccessibilityObject.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog (230405 => 230406)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 12:13:51 UTC (rev 230405)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 12:13:57 UTC (rev 230406)
@@ -1,3 +1,13 @@
+2018-03-09  Nan Wang  
+
+AX: AOM: More accessibility events support
+https://bugs.webkit.org/show_bug.cgi?id=183023
+
+
+Reviewed by Chris Fleizach.
+
+* accessibility/mac/AOM-events-webarea-crash.html:
+
 2018-03-07  Zalan Bujtas  
 
 Invalid innerTextRenderer in RenderTextControlSingleLine::styleDidChange()


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230405 => 230406)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:13:51 UTC (rev 230405)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 12:13:57 UTC (rev 230406)
@@ -1,3 +1,19 @@
+2018-03-09  Nan Wang  
+
+AX: AOM: More accessibility events support
+https://bugs.webkit.org/show_bug.cgi?id=183023
+
+
+Reviewed by Chris Fleizach.
+
+The test is crashing when we call updateBackingStore when 
+the AXObjectCache object is gone. Added a check to fix that.
+
+Modified the test by using the right format of setTimeout and extended the delay.
+
+* accessibility/AccessibilityObject.cpp:
+(WebCore::AccessibilityObject::updateBackingStore):
+
 2018-03-12  Danyao Wang  
 
 Add a query and fragment exception to history API's unique origin restriction.


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/accessibility/AccessibilityObject.cpp (230405 => 230406)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/accessibility/AccessibilityObject.cpp	2018-04-09 12:13:51 UTC (rev 230405)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/accessibility/AccessibilityObject.cpp	2018-04-09 12:13:57 UTC (rev 230406)
@@ -1765,6 +1765,9 @@
 #if HAVE(ACCESSIBILITY)
 void AccessibilityObject::updateBackingStore()
 {
+if (!axObjectCache())
+return;
+
 // Updating the layout may delete this object.
 RefPtr protectedThis(this);
 if (auto* document = this->document()) {






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

[webkit-changes] [230410] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230410] releases/WebKitGTK/webkit-2.20








Revision 230410
Author carlo...@webkit.org
Date 2018-04-09 05:14:22 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229850 - Race Condition in arrayProtoFuncReverse() causes wrong results or crash
https://bugs.webkit.org/show_bug.cgi?id=183901

Reviewed by Keith Miller.

JSTests:

New test.

* stress/array-reverse-doesnt-clobber.js: Added.
(testArrayReverse):
(createArrayOfArrays):
(createArrayStorage):

Source/_javascript_Core:

Added write barriers to ensure the reversed contents are properly marked.

* runtime/ArrayPrototype.cpp:
(JSC::arrayProtoFuncReverse):

Modified Paths

releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/ArrayPrototype.cpp


Added Paths

releases/WebKitGTK/webkit-2.20/JSTests/stress/array-reverse-doesnt-clobber.js




Diff

Modified: releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog (230409 => 230410)

--- releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog	2018-04-09 12:14:15 UTC (rev 230409)
+++ releases/WebKitGTK/webkit-2.20/JSTests/ChangeLog	2018-04-09 12:14:22 UTC (rev 230410)
@@ -1,3 +1,17 @@
+2018-03-22  Michael Saboff  
+
+Race Condition in arrayProtoFuncReverse() causes wrong results or crash
+https://bugs.webkit.org/show_bug.cgi?id=183901
+
+Reviewed by Keith Miller.
+
+New test.
+
+* stress/array-reverse-doesnt-clobber.js: Added.
+(testArrayReverse):
+(createArrayOfArrays):
+(createArrayStorage):
+
 2018-03-01  Yusuke Suzuki  
 
 ASSERTION FAILED: matchContextualKeyword(m_vm->propertyNames->async)


Added: releases/WebKitGTK/webkit-2.20/JSTests/stress/array-reverse-doesnt-clobber.js (0 => 230410)

--- releases/WebKitGTK/webkit-2.20/JSTests/stress/array-reverse-doesnt-clobber.js	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/JSTests/stress/array-reverse-doesnt-clobber.js	2018-04-09 12:14:22 UTC (rev 230410)
@@ -0,0 +1,61 @@
+// This tests that array.Prototype.reverse() doesn't inadvertently clobber indexed properties.
+// This test shouldn't throw or crash.
+
+const outerArrayLength = 1;
+const innerArrayLength = 128;
+
+function testArrayReverse(createArray)
+{
+const limit = 5;
+let save = [0, 0];
+
+for (let at = 0; at < limit; at++) {
+let arr = createArray();
+
+let v = [];
+for (let i = 0; i < 273; i++) {
+for (let j = 0; j < 8; j++)
+arr.reverse();
+
+v.push(new String("X").repeat(123008));
+}
+
+for (let i = 0; i < arr.length; i++) {
+if (arr[i].length != innerArrayLength)
+throw "arr[" + i + "].length has changed from " + innerArrayLength + " to " + arr[i].length;
+}
+
+let f = [];
+for (let i = 0; i < 1000; i++)
+f.push(new Array(16).fill(0x42424242));
+
+save.push(arr);
+save.push(v);
+save.push(f);
+}
+}
+
+function createArrayOfArrays()
+{
+let result = new Array(outerArrayLength);
+
+for (let i = 0; i < result.length; i++)
+result[i] = new Array(innerArrayLength).fill(0x41414141);
+
+return result;
+}
+
+var alt = 0;
+
+function createArrayStorage()
+{
+let result = createArrayOfArrays();
+
+if (!(typeof ensureArrayStorage === undefined) && alt++ % 0)
+ensureArrayStorage(result);
+
+return result;
+}
+
+testArrayReverse(createArrayOfArrays);
+testArrayReverse(createArrayStorage);


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog (230409 => 230410)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 12:14:15 UTC (rev 230409)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 12:14:22 UTC (rev 230410)
@@ -1,3 +1,15 @@
+2018-03-22  Michael Saboff  
+
+Race Condition in arrayProtoFuncReverse() causes wrong results or crash
+https://bugs.webkit.org/show_bug.cgi?id=183901
+
+Reviewed by Keith Miller.
+
+Added write barriers to ensure the reversed contents are properly marked.
+
+* runtime/ArrayPrototype.cpp:
+(JSC::arrayProtoFuncReverse):
+
 2018-03-05  Mark Lam  
 
 JITThunk functions should only be called when the JIT is enabled.


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/ArrayPrototype.cpp (230409 => 230410)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/ArrayPrototype.cpp	2018-04-09 12:14:15 UTC (rev 230409)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/ArrayPrototype.cpp	2018-04-09 12:14:22 UTC (rev 230410)
@@ -837,6 +837,8 @@
 if (containsHole(data, length) && holesMustForwardToPrototype(vm, thisObject))
 break;
 std::reverse(data, data + length);
+  

[webkit-changes] [230407] releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE

[carlosgc]

Title: [230407] releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE








Revision 230407
Author carlo...@webkit.org
Date 2018-04-09 05:14:03 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229721 - REGRESSION(r225340): WPE port should not be getting EGL X11 types on ANGLE
https://bugs.webkit.org/show_bug.cgi?id=183749

Reviewed by Alex Christensen.

Add back the USE_WPE ifdef.

* changes.diff:
* include/EGL/eglplatform.h:

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/changes.diff
releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/include/EGL/eglplatform.h




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/ChangeLog (230406 => 230407)

--- releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/ChangeLog	2018-04-09 12:13:57 UTC (rev 230406)
+++ releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/ChangeLog	2018-04-09 12:14:03 UTC (rev 230407)
@@ -1,3 +1,15 @@
+2018-03-19  Carlos Alberto Lopez Perez  
+
+REGRESSION(r225340): WPE port should not be getting EGL X11 types on ANGLE
+https://bugs.webkit.org/show_bug.cgi?id=183749
+
+Reviewed by Alex Christensen.
+
+Add back the USE_WPE ifdef.
+
+* changes.diff:
+* include/EGL/eglplatform.h:
+
 2018-01-27  Dan Bernstein  
 
 HaveInternalSDK includes should be "#include?"


Modified: releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/changes.diff (230406 => 230407)

--- releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/changes.diff	2018-04-09 12:13:57 UTC (rev 230406)
+++ releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/changes.diff	2018-04-09 12:14:03 UTC (rev 230407)
@@ -8066,3 +8066,16 @@
  bool EndObject(SizeType memberCount = 0) {
  (void)memberCount;
  RAPIDJSON_ASSERT(level_stack_.GetSize() >= sizeof(Level));
+diff --git a/include/EGL/eglplatform.h b/include/EGL/eglplatform.h
+index 333448be348..47d18093c96 100644
+--- a/include/EGL/eglplatform.h
 b/include/EGL/eglplatform.h
+@@ -100,7 +100,7 @@ typedef struct ANativeWindow*   EGLNativeWindowType;
+ typedef struct egl_native_pixmap_t* EGLNativePixmapType;
+ typedef void*   EGLNativeDisplayType;
+ 
+-#elif defined(USE_OZONE)
++#elif defined(USE_OZONE) || defined(USE_WPE)
+ 
+ typedef intptr_t EGLNativeDisplayType;
+ typedef intptr_t EGLNativeWindowType;


Modified: releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/include/EGL/eglplatform.h (230406 => 230407)

--- releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/include/EGL/eglplatform.h	2018-04-09 12:13:57 UTC (rev 230406)
+++ releases/WebKitGTK/webkit-2.20/Source/ThirdParty/ANGLE/include/EGL/eglplatform.h	2018-04-09 12:14:03 UTC (rev 230407)
@@ -100,7 +100,7 @@
 typedef struct egl_native_pixmap_t* EGLNativePixmapType;
 typedef void*   EGLNativeDisplayType;
 
-#elif defined(USE_OZONE)
+#elif defined(USE_OZONE) || defined(USE_WPE)
 
 typedef intptr_t EGLNativeDisplayType;
 typedef intptr_t EGLNativeWindowType;






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

[webkit-changes] [230403] trunk/Source/JavaScriptCore

[utatane . tea]

Title: [230403] trunk/Source/_javascript_Core








Revision 230403
Author utatane@gmail.com
Date 2018-04-09 04:28:02 -0700 (Mon, 09 Apr 2018)


Log Message
Unreviewed, attempt to fix Windows build
https://bugs.webkit.org/show_bug.cgi?id=183508

* jit/JIT.h:

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/jit/JIT.h




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (230402 => 230403)

--- trunk/Source/_javascript_Core/ChangeLog	2018-04-09 10:53:36 UTC (rev 230402)
+++ trunk/Source/_javascript_Core/ChangeLog	2018-04-09 11:28:02 UTC (rev 230403)
@@ -1,5 +1,12 @@
 2018-04-08  Yusuke Suzuki  
 
+Unreviewed, attempt to fix Windows build
+https://bugs.webkit.org/show_bug.cgi?id=183508
+
+* jit/JIT.h:
+
+2018-04-08  Yusuke Suzuki  
+
 Unreviewed, build fix for Windows by suppressing padding warning for JIT
 https://bugs.webkit.org/show_bug.cgi?id=183508
 


Modified: trunk/Source/_javascript_Core/jit/JIT.h (230402 => 230403)

--- trunk/Source/_javascript_Core/jit/JIT.h	2018-04-09 10:53:36 UTC (rev 230402)
+++ trunk/Source/_javascript_Core/jit/JIT.h	2018-04-09 11:28:02 UTC (rev 230403)
@@ -27,6 +27,14 @@
 
 #if ENABLE(JIT)
 
+// We've run into some problems where changing the size of the class JIT leads to
+// performance fluctuations. Try forcing alignment in an attempt to stabilize this.
+#if COMPILER(GCC_OR_CLANG)
+#define JIT_CLASS_ALIGNMENT alignas(32)
+#else
+#define JIT_CLASS_ALIGNMENT
+#endif
+
 #define ASSERT_JIT_OFFSET(actual, expected) ASSERT_WITH_MESSAGE(actual == expected, "JIT Offset \"%s\" should be %d, not %d.\n", #expected, static_cast(expected), static_cast(actual));
 
 #include "CodeBlock.h"
@@ -168,13 +176,7 @@
 
 void ctiPatchCallByReturnAddress(ReturnAddressPtr, FunctionPtr newCalleeFunction);
 
-// We've run into some problems where changing the size of the class JIT leads to
-// performance fluctuations. Try forcing alignment in an attempt to stabilize this.
-#if COMPILER(MSVC)
-#pragma warning(push)
-#pragma warning(disable: 4324) // Disable "structure was padded due to alignment specifier" warning
-#endif
-class alignas(32) JIT : private JSInterfaceJIT {
+class JIT_CLASS_ALIGNMENT JIT : private JSInterfaceJIT {
 friend class JITSlowPathCall;
 friend class JITStubCall;
 
@@ -947,9 +949,6 @@
 bool m_shouldUseIndexMasking;
 unsigned m_loopOSREntryBytecodeOffset { 0 };
 };
-#if COMPILER(MSVC)
-#pragma warning(pop)
-#endif
 
 } // namespace JSC
 






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

[webkit-changes] [230402] releases/WebKitGTK/webkit-2.20/Source/WTF

[carlosgc]

Title: [230402] releases/WebKitGTK/webkit-2.20/Source/WTF








Revision 230402
Author carlo...@webkit.org
Date 2018-04-09 03:53:36 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229406 - [GLib] RunLoop::wakeUp(): use a zero value instead of the monotonic time
https://bugs.webkit.org/show_bug.cgi?id=183447

Reviewed by Carlos Garcia Campos.

* wtf/glib/RunLoopGLib.cpp:
(WTF::RunLoop::wakeUp): Micro-optimize the wake-up schedule by using
a zero value as the GSource ready time, instead of retrieving and using
the current monotonic time.

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/glib/RunLoopGLib.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog (230401 => 230402)

--- releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog	2018-04-09 10:53:30 UTC (rev 230401)
+++ releases/WebKitGTK/webkit-2.20/Source/WTF/ChangeLog	2018-04-09 10:53:36 UTC (rev 230402)
@@ -1,3 +1,15 @@
+2018-03-08  Zan Dobersek  
+
+[GLib] RunLoop::wakeUp(): use a zero value instead of the monotonic time
+https://bugs.webkit.org/show_bug.cgi?id=183447
+
+Reviewed by Carlos Garcia Campos.
+
+* wtf/glib/RunLoopGLib.cpp:
+(WTF::RunLoop::wakeUp): Micro-optimize the wake-up schedule by using
+a zero value as the GSource ready time, instead of retrieving and using
+the current monotonic time.
+
 2018-02-23  Fujii Hironori  
 
 warning: unused variable 'InitialBufferSize' in Assertions.cpp


Modified: releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/glib/RunLoopGLib.cpp (230401 => 230402)

--- releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/glib/RunLoopGLib.cpp	2018-04-09 10:53:30 UTC (rev 230401)
+++ releases/WebKitGTK/webkit-2.20/Source/WTF/wtf/glib/RunLoopGLib.cpp	2018-04-09 10:53:36 UTC (rev 230402)
@@ -120,7 +120,7 @@
 
 void RunLoop::wakeUp()
 {
-g_source_set_ready_time(m_source.get(), g_get_monotonic_time());
+g_source_set_ready_time(m_source.get(), 0);
 }
 
 class DispatchAfterContext {






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

[webkit-changes] [230397] releases/WebKitGTK/webkit-2.20/Source/JavaScriptCore

[carlosgc]

Title: [230397] releases/WebKitGTK/webkit-2.20/Source/_javascript_Core








Revision 230397
Author carlo...@webkit.org
Date 2018-04-09 03:52:49 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229302 - JITThunk functions should only be called when the JIT is enabled.
https://bugs.webkit.org/show_bug.cgi?id=183351


Reviewed by Keith Miller.

* jit/JITThunks.cpp:
(JSC::JITThunks::ctiNativeCall):
(JSC::JITThunks::ctiNativeConstruct):
(JSC::JITThunks::ctiInternalFunctionCall):
(JSC::JITThunks::ctiInternalFunctionConstruct):
* runtime/VM.cpp:
(JSC::VM::VM):
(JSC::VM::getCTIInternalFunctionTrampolineFor):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/jit/JITThunks.cpp
releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/VM.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog (230396 => 230397)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 10:52:43 UTC (rev 230396)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/ChangeLog	2018-04-09 10:52:49 UTC (rev 230397)
@@ -1,3 +1,20 @@
+2018-03-05  Mark Lam  
+
+JITThunk functions should only be called when the JIT is enabled.
+https://bugs.webkit.org/show_bug.cgi?id=183351
+
+
+Reviewed by Keith Miller.
+
+* jit/JITThunks.cpp:
+(JSC::JITThunks::ctiNativeCall):
+(JSC::JITThunks::ctiNativeConstruct):
+(JSC::JITThunks::ctiInternalFunctionCall):
+(JSC::JITThunks::ctiInternalFunctionConstruct):
+* runtime/VM.cpp:
+(JSC::VM::VM):
+(JSC::VM::getCTIInternalFunctionTrampolineFor):
+
 2018-03-01  Yusuke Suzuki  
 
 ASSERTION FAILED: matchContextualKeyword(m_vm->propertyNames->async)


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/jit/JITThunks.cpp (230396 => 230397)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/jit/JITThunks.cpp	2018-04-09 10:52:43 UTC (rev 230396)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/jit/JITThunks.cpp	2018-04-09 10:52:49 UTC (rev 230397)
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2012, 2013, 2015-2016 Apple Inc. All rights reserved.
+ * Copyright (C) 2012-2018 Apple Inc. All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions
@@ -47,15 +47,13 @@
 
 MacroAssemblerCodePtr JITThunks::ctiNativeCall(VM* vm)
 {
-if (!VM::canUseJIT())
-return MacroAssemblerCodePtr::createLLIntCodePtr(llint_native_call_trampoline);
+ASSERT(VM::canUseJIT());
 return ctiStub(vm, nativeCallGenerator).code();
 }
 
 MacroAssemblerCodePtr JITThunks::ctiNativeConstruct(VM* vm)
 {
-if (!VM::canUseJIT())
-return MacroAssemblerCodePtr::createLLIntCodePtr(llint_native_construct_trampoline);
+ASSERT(VM::canUseJIT());
 return ctiStub(vm, nativeConstructGenerator).code();
 }
 
@@ -73,15 +71,13 @@
 
 MacroAssemblerCodePtr JITThunks::ctiInternalFunctionCall(VM* vm)
 {
-if (!VM::canUseJIT())
-return MacroAssemblerCodePtr::createLLIntCodePtr(llint_internal_function_call_trampoline);
+ASSERT(VM::canUseJIT());
 return ctiStub(vm, internalFunctionCallGenerator).code();
 }
 
 MacroAssemblerCodePtr JITThunks::ctiInternalFunctionConstruct(VM* vm)
 {
-if (!VM::canUseJIT())
-return MacroAssemblerCodePtr::createLLIntCodePtr(llint_internal_function_construct_trampoline);
+ASSERT(VM::canUseJIT());
 return ctiStub(vm, internalFunctionConstructGenerator).code();
 }
 


Modified: releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/VM.cpp (230396 => 230397)

--- releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/VM.cpp	2018-04-09 10:52:43 UTC (rev 230396)
+++ releases/WebKitGTK/webkit-2.20/Source/_javascript_Core/runtime/VM.cpp	2018-04-09 10:52:49 UTC (rev 230397)
@@ -454,9 +454,13 @@
 watchdog.setTimeLimit(Seconds::fromMilliseconds(Options::watchdog()));
 }
 
+#if ENABLE(JIT)
 // Make sure that any stubs that the JIT is going to use are initialized in non-compilation threads.
-getCTIInternalFunctionTrampolineFor(CodeForCall);
-getCTIInternalFunctionTrampolineFor(CodeForConstruct);
+if (canUseJIT()) {
+getCTIInternalFunctionTrampolineFor(CodeForCall);
+getCTIInternalFunctionTrampolineFor(CodeForConstruct);
+}
+#endif
 
 VMInspector::instance().add(this);
 }
@@ -698,14 +702,15 @@
 MacroAssemblerCodePtr VM::getCTIInternalFunctionTrampolineFor(CodeSpecializationKind kind)
 {
 #if ENABLE(JIT)
+if (canUseJIT()) {
+if (kind == CodeForCall)
+return jitStubs->ctiInternalFunctionCall(this);
+return jitStubs->ctiInternalFunctionConstruct(this);
+}
+#endif
 if (kind == CodeForCall)
-return jitStubs->ctiInternalFunctionCall(this);
-return 

[webkit-changes] [230399] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230399] releases/WebKitGTK/webkit-2.20








Revision 230399
Author carlo...@webkit.org
Date 2018-04-09 03:53:05 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229375 - replaceState cause back/forward malfunction on html page with 

Patch by Sihui Liu  on 2018-03-07
Reviewed by Chris Dumez.

Source/WebCore:

replaceState should not change URL when the URL argument is NULL, but should change URL when the URL argument is an empty string.

Test: http/tests/history/replacestate-no-url.html

* page/History.cpp:
(WebCore::History::urlForState):

LayoutTests:

* http/tests/history/replacestate-no-url-expected.txt: Added.
* http/tests/history/replacestate-no-url.html: Added.
Add layout test coverage.

* fast/loader/stateobjects/pushstate-with-fragment-urls-and-hashchange-expected.txt:
Rebaseline a layout test as empty string for URL is handled differently.

Modified Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.20/LayoutTests/fast/loader/stateobjects/pushstate-with-fragment-urls-and-hashchange-expected.txt
releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/page/History.cpp


Added Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/history/replacestate-no-url-expected.txt
releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/history/replacestate-no-url.html




Diff

Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog (230398 => 230399)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 10:52:54 UTC (rev 230398)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 10:53:05 UTC (rev 230399)
@@ -1,3 +1,18 @@
+2018-03-07  Sihui Liu  
+
+replaceState cause back/forward malfunction on html page with 
+
+Reviewed by Chris Dumez.
+
+* http/tests/history/replacestate-no-url-expected.txt: Added.
+* http/tests/history/replacestate-no-url.html: Added.
+Add layout test coverage.
+
+* fast/loader/stateobjects/pushstate-with-fragment-urls-and-hashchange-expected.txt:
+Rebaseline a layout test as empty string for URL is handled differently.
+
 2018-03-03  Said Abou-Hallawa  
 
 Delete incorrect version of clampTo() function from SVGToOTFFontConversion.cpp


Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/fast/loader/stateobjects/pushstate-with-fragment-urls-and-hashchange-expected.txt (230398 => 230399)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/fast/loader/stateobjects/pushstate-with-fragment-urls-and-hashchange-expected.txt	2018-04-09 10:52:54 UTC (rev 230398)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/fast/loader/stateobjects/pushstate-with-fragment-urls-and-hashchange-expected.txt	2018-04-09 10:53:05 UTC (rev 230399)
@@ -27,8 +27,9 @@
 Hash change fired and last path component is pushstate-with-fragment-urls-and-hashchange.html#otherhash
 State popped with event null (type object) and last path component pushstate-with-fragment-urls-and-hashchange.html#hash
 Hash change fired and last path component is pushstate-with-fragment-urls-and-hashchange.html#hash
+State popped with event null (type object) and last path component pushstate-with-fragment-urls-and-hashchange.html
+Hash change fired and last path component is pushstate-with-fragment-urls-and-hashchange.html
 State popped with event null (type object) and last path component pushstate-with-fragment-urls-and-hashchange.html#
 Hash change fired and last path component is pushstate-with-fragment-urls-and-hashchange.html#
-State popped with event null (type object) and last path component pushstate-with-fragment-urls-and-hashchange.html#
 State popped with event OriginalEntry (type string) and last path component pushstate-with-fragment-urls-and-hashchange.html
 


Added: releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/history/replacestate-no-url-expected.txt (0 => 230399)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/history/replacestate-no-url-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/history/replacestate-no-url-expected.txt	2018-04-09 10:53:05 UTC (rev 230399)
@@ -0,0 +1,18 @@
+Tests that ReplaceState should not change document URL if URL argument is null.
+
+On success, you will see a series of "PASS" messages, followed by "TEST COMPLETE".
+
+
+Push item one
+Replace item one with null url
+Push item two
+Going back to item one
+PASS document.location.href is "http://127.0.0.1:8000/one"
+Replace item one with empty url
+Push item two
+Going back to item one
+PASS document.location.href is "http://127.0.0.1:8000/"
+PASS successfullyParsed is true
+
+TEST COMPLETE
+


Added: releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/history/replacestate-no-url.html (0 => 230399)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/http/tests/history/replacestate-no-url.html	(rev 0)
+++ 

[webkit-changes] [230395] releases/WebKitGTK/webkit-2.20/Source/WebKit

[carlosgc]

Title: [230395] releases/WebKitGTK/webkit-2.20/Source/WebKit








Revision 230395
Author carlo...@webkit.org
Date 2018-04-09 03:52:37 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229206 - Web Automation: script evaluations via WebDriver should have a user gesture indicator
https://bugs.webkit.org/show_bug.cgi?id=183230


Reviewed by Andy Estes.

APIs that normally require a user gesture should just work when using via WebDriver.
To support cases where tests need to simulate user actions with _javascript_, use a
fake user gesture, similar to how -[WKWebView evaluateJavaScript:] forces a user
gesture when clients evaluate _javascript_ in their web view.

No new tests, this is covered by W3C tests that use the Fullscreen API.
This API does nothing if there is no user gesture; with this patch, it just works.

* WebProcess/Automation/WebAutomationSessionProxy.cpp:
(WebKit::WebAutomationSessionProxy::evaluateJavaScriptFunction):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog (230394 => 230395)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 10:52:32 UTC (rev 230394)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 10:52:37 UTC (rev 230395)
@@ -1,3 +1,22 @@
+2018-03-02  Brian Burg  
+
+Web Automation: script evaluations via WebDriver should have a user gesture indicator
+https://bugs.webkit.org/show_bug.cgi?id=183230
+
+
+Reviewed by Andy Estes.
+
+APIs that normally require a user gesture should just work when using via WebDriver.
+To support cases where tests need to simulate user actions with _javascript_, use a
+fake user gesture, similar to how -[WKWebView evaluateJavaScript:] forces a user
+gesture when clients evaluate _javascript_ in their web view.
+
+No new tests, this is covered by W3C tests that use the Fullscreen API.
+This API does nothing if there is no user gesture; with this patch, it just works.
+
+* WebProcess/Automation/WebAutomationSessionProxy.cpp:
+(WebKit::WebAutomationSessionProxy::evaluateJavaScriptFunction):
+
 2018-03-03  Brent Fulgham  
 
 Notify the NetworkProcess when a session is servicing an automation client


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.cpp (230394 => 230395)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.cpp	2018-04-09 10:52:32 UTC (rev 230394)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.cpp	2018-04-09 10:52:37 UTC (rev 230395)
@@ -271,7 +271,10 @@
 JSValueMakeNumber(context, callbackTimeout)
 };
 
-callPropertyFunction(context, scriptObject, ASCIILiteral("evaluateJavaScriptFunction"), WTF_ARRAY_LENGTH(functionArguments), functionArguments, );
+{
+WebCore::UserGestureIndicator gestureIndicator(WebCore::ProcessingUserGesture, frame->coreFrame()->document());
+callPropertyFunction(context, scriptObject, ASCIILiteral("evaluateJavaScriptFunction"), WTF_ARRAY_LENGTH(functionArguments), functionArguments, );
+}
 
 if (!exception)
 return;






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

[webkit-changes] [230401] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230401] releases/WebKitGTK/webkit-2.20








Revision 230401
Author carlo...@webkit.org
Date 2018-04-09 03:53:30 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229395 - [GTK][WPE] Leak checker is not working in WebKitGLib web process tests
https://bugs.webkit.org/show_bug.cgi?id=183404

Reviewed by Michael Catanzaro.

Source/WebKit:

Add private helper for testing to do a garbage collection when the page is closing.

* WebProcess/InjectedBundle/API/glib/WebKitWebExtension.cpp:
(webkitWebExtensionSetGarbageCollectOnPageDestroy):
* WebProcess/InjectedBundle/API/glib/WebKitWebExtensionPrivate.h:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMClientRectList.h: Fix annotation of
webkit_dom_client_rect_list_item(), it should be transfer full.

Tools:

This might have regressed when we started to use the JSC garbage collector timers. The thing is that we expect
that the WebProcessTester object that we expose to _javascript_ is released when the web frame is destroyed, but
that's no longer the case. On window object cleared a GC is scheduled, but JSC timers do the actual garbage
collection later. In the case of tests that never happens because the web process finishes quickly after the
test. We need to force a garbage collection at some point when the web page is destroyed. We can't use the
WebKitWebPage destroy signal, since we are also checking that WebKitWebPage isn't leaked. The
API::InjectedBundle::Client::willDestroyPage() always happen when the page is closed, even if WebKitWebPage is
still alive, so we can force the GC at that point. The only problem is that the frame is detached right after
that point, so we can't check WebKitFrame leaks. The only frame in the tests is the main one, so we can assume
that if WebKitWebPage is released, the frame is too.

* TestWebKitAPI/Tests/WebKitGLib/FrameTest.cpp:
(WebKitFrameTest::testMainFrame): Stop checking we don't leak WebKitFrame.
(WebKitFrameTest::testURI): Ditto.
(WebKitFrameTest::testJavaScriptContext): Ditto.
* TestWebKitAPI/Tests/WebKitGLib/TestFrame.cpp:
(testWebKitFrameMainFrame): Use new WebViewTest::runWebProcessTest() API.
(testWebKitFrameURI): Ditto.
(testWebKitFrameJavaScriptContext): Ditto.
(webkitFrameTestRun): Deleted.
* TestWebKitAPI/Tests/WebKitGLib/WebProcessTest.cpp:
(runTest): We no longer need the special case for dom-cache test.
(windowObjectClearedCallback): Only expose test runner object to _javascript_ when loading tests.
(webkit_web_extension_initialize): Call webkitWebExtensionSetGarbageCollectOnPageDestroy() to ensure a garbage
collection is performed when the page is closing.
* TestWebKitAPI/Tests/WebKitGtk/DOMClientRectTest.cpp:
(WebKitDOMClientRectTest::testDivClientRectsPositionAndLength): Fix memory leak,
webkit_dom_client_rect_list_item() returns a full reference.
* TestWebKitAPI/Tests/WebKitGtk/TestAutocleanups.cpp:
(testWebProcessAutocleanups):Use new WebViewTest::runWebProcessTest() API.
* TestWebKitAPI/Tests/WebKitGtk/TestDOMClientRect.cpp:
(testWebKitDOMClientRectDivBoundingClientRectPosition): Use new WebViewTest::runWebProcessTest() API.
(testWebKitDOMClientRectDivClientRectsPositionAndLength): Use new WebViewTest::runWebProcessTest() API.
(prepareDOMForClientRectPositionTests): Deleted.
* TestWebKitAPI/Tests/WebKitGtk/TestDOMNode.cpp:
(testWebKitDOMNodeHierarchyNavigation): Use new WebViewTest::runWebProcessTest() API.
(testWebKitDOMNodeInsertion): Ditto.
(testWebKitDOMNodeTagNamesNodeList): Ditto.
(testWebKitDOMNodeTagNamesHTMLCollection): Ditto.
(testWebKitDOMObjectCache): We no longer need to run the test several times, since runWebProcessTest() loads
about blank after every test.
(prepareDOMForTagNamesTests): Deleted.
* TestWebKitAPI/Tests/WebKitGtk/TestDOMNodeFilter.cpp:
(testWebKitDOMNodeFilterTreeWalker): Use new WebViewTest::runWebProcessTest() API.
(testWebKitDOMNodeFilterNodeIterator): Ditto.
(runTest): Deleted.
* TestWebKitAPI/Tests/WebKitGtk/TestDOMXPathNSResolver.cpp:
(testWebKitDOMXPathNSResolverNative): Use new WebViewTest::runWebProcessTest() API.
(testWebKitDOMXPathNSResolverCustom): Ditto.
* TestWebKitAPI/Tests/WebKitGtk/TestEditor.cpp:
(testWebKitWebEditorSelectionChanged): Use new WebViewTest::runWebProcessTest() API.
* TestWebKitAPI/glib/WebKitGLib/WebViewTest.cpp:
(WebViewTest::runWebProcessTest): It now receives the contents, so it automatically loads the view using
"webprocess://test" as base URI, used to detect tests in the web process. It also loads about:blank after every
test to ensure that window object is cleared.
* TestWebKitAPI/glib/WebKitGLib/WebViewTest.h:

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/InjectedBundle/API/glib/WebKitWebExtension.cpp
releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/InjectedBundle/API/glib/WebKitWebExtensionPrivate.h
releases/WebKitGTK/webkit-2.20/Source/WebKit/WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMClientRectList.h

[webkit-changes] [230398] releases/WebKitGTK/webkit-2.20/Source/WebCore

[carlosgc]

Title: [230398] releases/WebKitGTK/webkit-2.20/Source/WebCore








Revision 230398
Author carlo...@webkit.org
Date 2018-04-09 03:52:54 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229328 - Change the type of SVGToOTFFontConverter::m_weight to be not a char
https://bugs.webkit.org/show_bug.cgi?id=183339

Reviewed by Alex Christensen.

No new tests because there is no behavior change.

* svg/SVGToOTFFontConversion.cpp:
(WebCore::SVGToOTFFontConverter::appendOS2Table):
(WebCore::SVGToOTFFontConverter::SVGToOTFFontConverter):

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/svg/SVGToOTFFontConversion.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230397 => 230398)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 10:52:49 UTC (rev 230397)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 10:52:54 UTC (rev 230398)
@@ -1,3 +1,16 @@
+2018-03-06  Myles C. Maxfield  
+
+Change the type of SVGToOTFFontConverter::m_weight to be not a char
+https://bugs.webkit.org/show_bug.cgi?id=183339
+
+Reviewed by Alex Christensen.
+
+No new tests because there is no behavior change.
+
+* svg/SVGToOTFFontConversion.cpp:
+(WebCore::SVGToOTFFontConverter::appendOS2Table):
+(WebCore::SVGToOTFFontConverter::SVGToOTFFontConverter):
+
 2018-03-05  Antti Koivisto  
 
 Don't invalidate all children when doing insertion/deletion in presence of backward positional selectors


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/svg/SVGToOTFFontConversion.cpp (230397 => 230398)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/svg/SVGToOTFFontConversion.cpp	2018-04-09 10:52:49 UTC (rev 230397)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/svg/SVGToOTFFontConversion.cpp	2018-04-09 10:52:54 UTC (rev 230398)
@@ -264,7 +264,7 @@
 int m_descent;
 unsigned m_featureCountGSUB;
 unsigned m_tablesAppendedCount;
-char m_weight;
+uint8_t m_weight;
 bool m_italic;
 bool m_error { false };
 };
@@ -499,7 +499,7 @@
 
 append16(2); // Version
 append16(clampTo(averageAdvance));
-append16(clampTo(m_weight)); // Weight class
+append16(m_weight); // Weight class
 append16(5); // Width class
 append16(0); // Protected font
 // WebKit handles these superscripts and subscripts
@@ -525,7 +525,7 @@
 for (auto& segment : segments) {
 bool ok;
 int value = segment.toInt();
-if (ok && value >= 0 && value <= 0xFF)
+if (ok && value >= std::numeric_limits::min() && value <= std::numeric_limits::max())
 panoseBytes[numPanoseBytes++] = value;
 }
 }
@@ -1461,7 +1461,7 @@
 bool ok;
 int value = segment.toInt(ok);
 if (ok && value >= 0 && value < 1000) {
-m_weight = (value + 50) / 100;
+m_weight = std::max(std::min((value + 50) / 100, static_cast(std::numeric_limits::max())), static_cast(std::numeric_limits::min()));
 break;
 }
 }






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

[webkit-changes] [230396] releases/WebKitGTK/webkit-2.20/Source/WebCore

[carlosgc]

Title: [230396] releases/WebKitGTK/webkit-2.20/Source/WebCore








Revision 230396
Author carlo...@webkit.org
Date 2018-04-09 03:52:43 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229288 - Don't invalidate all children when doing insertion/deletion in presence of backward positional selectors
https://bugs.webkit.org/show_bug.cgi?id=183325


Reviewed by Zalan Bujtas.

It is sufficient to invalidate siblings before the mutation point.

* dom/Element.cpp:
(WebCore::checkForSiblingStyleChanges):

We already do sibling walk in the case of forwards positional rules and sibling combinators. The work
done here is insignifant compared to cost of overinvalidating.

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/dom/Element.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230395 => 230396)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 10:52:37 UTC (rev 230395)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 10:52:43 UTC (rev 230396)
@@ -1,3 +1,19 @@
+2018-03-05  Antti Koivisto  
+
+Don't invalidate all children when doing insertion/deletion in presence of backward positional selectors
+https://bugs.webkit.org/show_bug.cgi?id=183325
+
+
+Reviewed by Zalan Bujtas.
+
+It is sufficient to invalidate siblings before the mutation point.
+
+* dom/Element.cpp:
+(WebCore::checkForSiblingStyleChanges):
+
+We already do sibling walk in the case of forwards positional rules and sibling combinators. The work
+done here is insignifant compared to cost of overinvalidating.
+
 2018-03-03  Said Abou-Hallawa  
 
 Delete incorrect version of clampTo() function from SVGToOTFFontConversion.cpp


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/dom/Element.cpp (230395 => 230396)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/dom/Element.cpp	2018-04-09 10:52:37 UTC (rev 230395)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/dom/Element.cpp	2018-04-09 10:52:43 UTC (rev 230396)
@@ -2077,11 +2077,10 @@
 // Backward positional selectors include nth-last-child, nth-last-of-type, last-of-type and only-of-type.
 // We have to invalidate everything following the insertion point in the forward case, and everything before the insertion point in the
 // backward case.
-// |afterChange| is 0 in the parser callback case, so we won't do any work for the forward case if we don't have to.
-// For performance reasons we just mark the parent node as changed, since we don't want to make childrenChanged O(n^2) by crawling all our kids
-// here.  recalcStyle will then force a walk of the children when it sees that this has happened.
-if (parent.childrenAffectedByBackwardPositionalRules() && elementBeforeChange)
-parent.invalidateStyleForSubtree();
+if (parent.childrenAffectedByBackwardPositionalRules()) {
+for (auto* previous = elementBeforeChange; previous; previous = previous->previousElementSibling())
+previous->invalidateStyleForSubtree();
+}
 }
 
 void Element::childrenChanged(const ChildChange& change)






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

[webkit-changes] [230400] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230400] releases/WebKitGTK/webkit-2.20








Revision 230400
Author carlo...@webkit.org
Date 2018-04-09 03:53:14 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229393 - Invalid innerTextRenderer in RenderTextControlSingleLine::styleDidChange()
https://bugs.webkit.org/show_bug.cgi?id=183385


Reviewed by Antti Koivisto.

Source/WebCore:

When HTMLInputElement::updateType() is called with a dirty value, we eagerly change the m_inputType first
and then we take care of the dirty value by calling setAttributeWithoutSynchronization().
With a DOMSubtreeModified event listener attached, setAttributeWithoutSynchronization() can end up running some
layout code (offsetHeight) with a renderer - m_inputType mismatch.

This patch ensures that we don't change the m_inputType until after we finished setting the new value.

Test: fast/DOM/HTMLInputElement/input-value-and-type-change-crash.html

* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::updateType):

LayoutTests:

* fast/DOM/HTMLInputElement/input-value-and-type-change-crash-expected.txt: Added.
* fast/DOM/HTMLInputElement/input-value-and-type-change-crash.html: Added.

Modified Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/html/HTMLInputElement.cpp


Added Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/fast/DOM/
releases/WebKitGTK/webkit-2.20/LayoutTests/fast/DOM/HTMLInputElement/
releases/WebKitGTK/webkit-2.20/LayoutTests/fast/DOM/HTMLInputElement/input-value-and-type-change-crash-expected.txt
releases/WebKitGTK/webkit-2.20/LayoutTests/fast/DOM/HTMLInputElement/input-value-and-type-change-crash.html




Diff

Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog (230399 => 230400)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 10:53:05 UTC (rev 230399)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 10:53:14 UTC (rev 230400)
@@ -1,3 +1,14 @@
+2018-03-07  Zalan Bujtas  
+
+Invalid innerTextRenderer in RenderTextControlSingleLine::styleDidChange()
+https://bugs.webkit.org/show_bug.cgi?id=183385
+
+
+Reviewed by Antti Koivisto.
+
+* fast/DOM/HTMLInputElement/input-value-and-type-change-crash-expected.txt: Added.
+* fast/DOM/HTMLInputElement/input-value-and-type-change-crash.html: Added.
+
 2018-03-07  Sihui Liu  
 
 replaceState cause back/forward malfunction on html page with 

Added: releases/WebKitGTK/webkit-2.20/LayoutTests/fast/DOM/HTMLInputElement/input-value-and-type-change-crash-expected.txt (0 => 230400)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/fast/DOM/HTMLInputElement/input-value-and-type-change-crash-expected.txt	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/fast/DOM/HTMLInputElement/input-value-and-type-change-crash-expected.txt	2018-04-09 10:53:14 UTC (rev 230400)
@@ -0,0 +1 @@
+ Pass if no crash.


Added: releases/WebKitGTK/webkit-2.20/LayoutTests/fast/DOM/HTMLInputElement/input-value-and-type-change-crash.html (0 => 230400)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/fast/DOM/HTMLInputElement/input-value-and-type-change-crash.html	(rev 0)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/fast/DOM/HTMLInputElement/input-value-and-type-change-crash.html	2018-04-09 10:53:14 UTC (rev 230400)
@@ -0,0 +1,16 @@
+
+Pass if no crash.
+
+if (window.testRunner)
+testRunner.dumpAsText();
+
+function eventhandler() {
+document.body.offsetHeight;
+}
+
+input.addEventListener("DOMSubtreeModified", eventhandler);
+document.execCommand("selectAll", true);
+
+input.value = "foobar";
+input.type = "button";
+


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230399 => 230400)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 10:53:05 UTC (rev 230399)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 10:53:14 UTC (rev 230400)
@@ -1,3 +1,23 @@
+2018-03-07  Zalan Bujtas  
+
+Invalid innerTextRenderer in RenderTextControlSingleLine::styleDidChange()
+https://bugs.webkit.org/show_bug.cgi?id=183385
+
+
+Reviewed by Antti Koivisto.
+
+When HTMLInputElement::updateType() is called with a dirty value, we eagerly change the m_inputType first
+and then we take care of the dirty value by calling setAttributeWithoutSynchronization().
+With a DOMSubtreeModified event listener attached, setAttributeWithoutSynchronization() can end up running some
+layout code (offsetHeight) with a renderer - m_inputType mismatch.
+
+This patch ensures that we don't change the m_inputType until after we finished setting the new value.
+
+Test: fast/DOM/HTMLInputElement/input-value-and-type-change-crash.html
+
+* html/HTMLInputElement.cpp:
+(WebCore::HTMLInputElement::updateType):

[webkit-changes] [230394] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230394] releases/WebKitGTK/webkit-2.20








Revision 230394
Author carlo...@webkit.org
Date 2018-04-09 03:52:32 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229202 - Delete incorrect version of clampTo() function from SVGToOTFFontConversion.cpp
https://bugs.webkit.org/show_bug.cgi?id=183165

Patch by Said Abou-Hallawa  on 2018-03-03
Reviewed by Darin Adler.

Source/WebCore:

Some of the calls in SVGToOTFFontConversion.cpp were directed to the function
in MathExtras.h while the rest were directed to this local static function.
There should not be two versions with the same name while they are supposed
to do the same thing. Besides, the local version does not work correctly
if the type of the argument's max limit is less the max limit of the returned
type, e.g. char -> uint16_t.

* svg/SVGToOTFFontConversion.cpp:
(WebCore::clampTo): Deleted.

LayoutTests:

Re-baseline the results of these two tests since the local version of
clampTo() was not returning the expected results always.

* platform/ios/svg/W3C-SVG-1.1/text-altglyph-01-b-expected.txt:
* platform/ios/svg/text/text-altglyph-01-b-expected.txt:

Modified Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.20/LayoutTests/platform/ios/svg/W3C-SVG-1.1/text-altglyph-01-b-expected.txt
releases/WebKitGTK/webkit-2.20/LayoutTests/platform/ios/svg/text/text-altglyph-01-b-expected.txt
releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/svg/SVGToOTFFontConversion.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog (230393 => 230394)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 10:52:25 UTC (rev 230393)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 10:52:32 UTC (rev 230394)
@@ -1,3 +1,16 @@
+2018-03-03  Said Abou-Hallawa  
+
+Delete incorrect version of clampTo() function from SVGToOTFFontConversion.cpp
+https://bugs.webkit.org/show_bug.cgi?id=183165
+
+Reviewed by Darin Adler.
+
+Re-baseline the results of these two tests since the local version of
+clampTo() was not returning the expected results always.
+
+* platform/ios/svg/W3C-SVG-1.1/text-altglyph-01-b-expected.txt:
+* platform/ios/svg/text/text-altglyph-01-b-expected.txt:
+
 2018-03-03  Zalan Bujtas  
 
 [RenderTreeBuilder] Move styleDidChange mutation logic to RenderTreeUpdater


Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/platform/ios/svg/W3C-SVG-1.1/text-altglyph-01-b-expected.txt (230393 => 230394)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/platform/ios/svg/W3C-SVG-1.1/text-altglyph-01-b-expected.txt	2018-04-09 10:52:25 UTC (rev 230393)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/platform/ios/svg/W3C-SVG-1.1/text-altglyph-01-b-expected.txt	2018-04-09 10:52:32 UTC (rev 230394)
@@ -11,26 +11,26 @@
   chunk 1 text run 1 at (5.00,90.00) startOffset 0 endOffset 32 width 460.21: "and many-to-many chars to glyphs"
   RenderSVGHiddenContainer {defs} at (0,0) size 0x0
   RenderSVGContainer {g} at (45,125) size 428x165
-RenderSVGText {text} at (140,130) size 192x75 contains 1 chunk(s)
+RenderSVGText {text} at (140,130) size 199x75 contains 1 chunk(s)
   RenderSVGTSpan {altGlyph} at (0,0) size 44x75
 RenderSVGInlineText {#text} at (0,0) size 44x75
-  chunk 1 text run 1 at (140.00,190.00) startOffset 0 endOffset 1 width 43.32: "H"
+  chunk 1 text run 1 at (140.00,190.00) startOffset 0 endOffset 1 width 43.33: "H"
   RenderSVGInlineText {#text} at (0,0) size 0x0
   RenderSVGTSpan {altGlyph} at (0,0) size 38x75
 RenderSVGInlineText {#text} at (43,0) size 38x75
-  chunk 1 text run 1 at (183.32,190.00) startOffset 0 endOffset 1 width 37.50: "A"
+  chunk 1 text run 1 at (183.33,190.00) startOffset 0 endOffset 1 width 37.50: "A"
   RenderSVGInlineText {#text} at (0,0) size 0x0
-  RenderSVGTSpan {altGlyph} at (0,0) size 38x75
-RenderSVGInlineText {#text} at (80,0) size 37x75
-  chunk 1 text run 1 at (220.82,190.00) startOffset 0 endOffset 1 width 36.66: "P"
+  RenderSVGTSpan {altGlyph} at (0,0) size 41x75
+RenderSVGInlineText {#text} at (80,0) size 41x75
+  chunk 1 text run 1 at (220.83,190.00) startOffset 0 endOffset 1 width 40.02: "P"
   RenderSVGInlineText {#text} at (0,0) size 0x0
-  RenderSVGTSpan {altGlyph} at (0,0) size 38x75
-RenderSVGInlineText {#text} at (117,0) size 37x75
-  chunk 1 text run 1 at (257.48,190.00) startOffset 0 endOffset 1 width 36.66: "P"
+  RenderSVGTSpan {altGlyph} at (0,0) size 41x75
+RenderSVGInlineText {#text} at (120,0) size 41x75
+  chunk 1 text run 1 at (260.85,190.00) startOffset 0 endOffset 1 width 40.02: "P"

[webkit-changes] [230393] releases/WebKitGTK/webkit-2.20/Source/WebKit

[carlosgc]

Title: [230393] releases/WebKitGTK/webkit-2.20/Source/WebKit








Revision 230393
Author carlo...@webkit.org
Date 2018-04-09 03:52:25 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229201 - Notify the NetworkProcess when a session is servicing an automation client
https://bugs.webkit.org/show_bug.cgi?id=183306


Reviewed by Brian Burg.

Network loads servicing WebDriver are done through an ephemeral session. While this is great
for protecting a developer's machine from sharing state with test runs, it has the unintended
effect of blocking certain logging operations.

We do not log content in ephemeral sessions to protect user privacy. However, ephemeral sessions
generated by WebDriver should participate in logging so that proper testing (with logging) can
be done.

This patch signals the NetworkProcess when an ephemeral session (created for automation purposes)
is created, so that it can allow logging.

* NetworkProcess/NetworkProcess.cpp:
(WebKit::NetworkProcess::destroySession): Remove controlled-by-automation entry.
(WebKit::NetworkProcess::sessionIsControlledByAutomation const): Added.
(WebKit::NetworkProcess::setSessionIsControlledByAutomation): Added.
* NetworkProcess/NetworkProcess.h:
* NetworkProcess/NetworkProcess.messages.in:
* NetworkProcess/NetworkResourceLoader.cpp:
(WebKit::NetworkResourceLoader::isAlwaysOnLoggingAllowed const): Checks if the relevant session
is servicing an automation client, and returns true if it is.
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::WebPageProxy): Signal the network process if this page is being created
for an automation client.

Modified Paths

releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebKit/NetworkProcess/NetworkProcess.cpp
releases/WebKitGTK/webkit-2.20/Source/WebKit/NetworkProcess/NetworkProcess.h
releases/WebKitGTK/webkit-2.20/Source/WebKit/NetworkProcess/NetworkProcess.messages.in
releases/WebKitGTK/webkit-2.20/Source/WebKit/NetworkProcess/NetworkResourceLoader.cpp
releases/WebKitGTK/webkit-2.20/Source/WebKit/UIProcess/WebPageProxy.cpp




Diff

Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog (230392 => 230393)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 10:52:15 UTC (rev 230392)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/ChangeLog	2018-04-09 10:52:25 UTC (rev 230393)
@@ -1,3 +1,35 @@
+2018-03-03  Brent Fulgham  
+
+Notify the NetworkProcess when a session is servicing an automation client
+https://bugs.webkit.org/show_bug.cgi?id=183306
+
+
+Reviewed by Brian Burg.
+
+Network loads servicing WebDriver are done through an ephemeral session. While this is great
+for protecting a developer's machine from sharing state with test runs, it has the unintended
+effect of blocking certain logging operations.
+
+We do not log content in ephemeral sessions to protect user privacy. However, ephemeral sessions
+generated by WebDriver should participate in logging so that proper testing (with logging) can
+be done.
+
+This patch signals the NetworkProcess when an ephemeral session (created for automation purposes)
+is created, so that it can allow logging.
+
+* NetworkProcess/NetworkProcess.cpp:
+(WebKit::NetworkProcess::destroySession): Remove controlled-by-automation entry.
+(WebKit::NetworkProcess::sessionIsControlledByAutomation const): Added.
+(WebKit::NetworkProcess::setSessionIsControlledByAutomation): Added.
+* NetworkProcess/NetworkProcess.h:
+* NetworkProcess/NetworkProcess.messages.in:
+* NetworkProcess/NetworkResourceLoader.cpp:
+(WebKit::NetworkResourceLoader::isAlwaysOnLoggingAllowed const): Checks if the relevant session
+is servicing an automation client, and returns true if it is.
+* UIProcess/WebPageProxy.cpp:
+(WebKit::WebPageProxy::WebPageProxy): Signal the network process if this page is being created
+for an automation client.
+
 2018-03-12  Carlos Garcia Campos  
 
 Unreviewed. Update OptionsGTK.cmake and NEWS for 2.20.0 release.


Modified: releases/WebKitGTK/webkit-2.20/Source/WebKit/NetworkProcess/NetworkProcess.cpp (230392 => 230393)

--- releases/WebKitGTK/webkit-2.20/Source/WebKit/NetworkProcess/NetworkProcess.cpp	2018-04-09 10:52:15 UTC (rev 230392)
+++ releases/WebKitGTK/webkit-2.20/Source/WebKit/NetworkProcess/NetworkProcess.cpp	2018-04-09 10:52:25 UTC (rev 230393)
@@ -314,6 +314,7 @@
 void NetworkProcess::destroySession(PAL::SessionID sessionID)
 {
 SessionTracker::destroySession(sessionID);
+m_sessionsControlledByAutomation.remove(sessionID);
 }
 
 void NetworkProcess::grantSandboxExtensionsToStorageProcessForBlobs(const Vector& filenames, Function&& completionHandler)
@@ -389,6 +390,19 @@
 }
 #endif
 
+bool NetworkProcess::sessionIsControlledByAutomation(PAL::SessionID sessionID) 

[webkit-changes] [230392] releases/WebKitGTK/webkit-2.20

[carlosgc]

Title: [230392] releases/WebKitGTK/webkit-2.20








Revision 230392
Author carlo...@webkit.org
Date 2018-04-09 03:52:15 -0700 (Mon, 09 Apr 2018)


Log Message
Merge r229200  - [RenderTreeBuilder] Move styleDidChange mutation logic to RenderTreeUpdater
https://bugs.webkit.org/show_bug.cgi?id=183273


Reviewed by Antti Koivisto.

Source/WebCore:

Covered by existing tests.

* rendering/RenderBlockFlow.cpp:
(WebCore::RenderBlockFlow::styleDidChange):
* rendering/RenderElement.cpp:
(WebCore::RenderElement::styleDidChange):
(WebCore::RenderElement::noLongerAffectsParentBlock const): Deleted.
* rendering/RenderElement.h:
* rendering/updating/RenderTreeUpdater.cpp:
(WebCore::RenderTreeUpdater::updateElementRenderer):

LayoutTests:

This is just a different repaint order.

* fast/repaint/absolute-position-change-containing-block-expected.txt:

Modified Paths

releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog
releases/WebKitGTK/webkit-2.20/LayoutTests/fast/repaint/absolute-position-change-containing-block-expected.txt
releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/RenderBlockFlow.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/RenderElement.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/RenderElement.h
releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/updating/RenderTreeBuilder.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/updating/RenderTreeBuilder.h
releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/updating/RenderTreeUpdater.cpp
releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/updating/RenderTreeUpdater.h




Diff

Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog (230391 => 230392)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 08:08:50 UTC (rev 230391)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/ChangeLog	2018-04-09 10:52:15 UTC (rev 230392)
@@ -1,3 +1,15 @@
+2018-03-03  Zalan Bujtas  
+
+[RenderTreeBuilder] Move styleDidChange mutation logic to RenderTreeUpdater
+https://bugs.webkit.org/show_bug.cgi?id=183273
+
+
+Reviewed by Antti Koivisto.
+
+This is just a different repaint order.
+
+* fast/repaint/absolute-position-change-containing-block-expected.txt:
+
 2018-03-02  Claudio Saavedra  
 
 [GTK] Unreviewed gardening


Modified: releases/WebKitGTK/webkit-2.20/LayoutTests/fast/repaint/absolute-position-change-containing-block-expected.txt (230391 => 230392)

--- releases/WebKitGTK/webkit-2.20/LayoutTests/fast/repaint/absolute-position-change-containing-block-expected.txt	2018-04-09 08:08:50 UTC (rev 230391)
+++ releases/WebKitGTK/webkit-2.20/LayoutTests/fast/repaint/absolute-position-change-containing-block-expected.txt	2018-04-09 10:52:15 UTC (rev 230392)
@@ -1,10 +1,10 @@
 (repaint rects
   (rect 8 5000 100 100)
   (rect 108 5100 100 100)
-  (rect 8 8 784 2000)
   (rect 8 5000 100 100)
   (rect 108 5100 100 100)
   (rect 100 100 100 100)
+  (rect 8 8 784 2000)
   (rect 16 5008 100 100)
   (rect 16 5008 100 100)
   (rect 8 8 100 100)


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog (230391 => 230392)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 08:08:50 UTC (rev 230391)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/ChangeLog	2018-04-09 10:52:15 UTC (rev 230392)
@@ -1,3 +1,22 @@
+2018-03-03  Zalan Bujtas  
+
+[RenderTreeBuilder] Move styleDidChange mutation logic to RenderTreeUpdater
+https://bugs.webkit.org/show_bug.cgi?id=183273
+
+
+Reviewed by Antti Koivisto.
+
+Covered by existing tests.
+
+* rendering/RenderBlockFlow.cpp:
+(WebCore::RenderBlockFlow::styleDidChange):
+* rendering/RenderElement.cpp:
+(WebCore::RenderElement::styleDidChange):
+(WebCore::RenderElement::noLongerAffectsParentBlock const): Deleted.
+* rendering/RenderElement.h:
+* rendering/updating/RenderTreeUpdater.cpp:
+(WebCore::RenderTreeUpdater::updateElementRenderer):
+
 2018-03-12  Michael Catanzaro  
 
 [GTK] Crash in WebCore::PlatformDisplayWayland::~PlatformDisplayWayland


Modified: releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/RenderBlockFlow.cpp (230391 => 230392)

--- releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/RenderBlockFlow.cpp	2018-04-09 08:08:50 UTC (rev 230391)
+++ releases/WebKitGTK/webkit-2.20/Source/WebCore/rendering/RenderBlockFlow.cpp	2018-04-09 10:52:15 UTC (rev 230392)
@@ -2020,10 +2020,6 @@
 parentBlock->markAllDescendantsWithFloatsForLayout();
 parentBlock->markSiblingsWithFloatsForLayout();
 }
-// Fresh floats need to be reparented if they actually belong to the previous anonymous block.
-// It copies the logic of RenderBlock::addChildIgnoringContinuation
-if (oldStyle && 

[webkit-changes] [230391] trunk/Tools

[zandobersek]

Title: [230391] trunk/Tools








Revision 230391
Author zandober...@gmail.com
Date 2018-04-09 01:08:50 -0700 (Mon, 09 Apr 2018)


Log Message
Unreviewed follow-up to r230389.

* WebKitTestRunner/TestController.cpp:
(WTR::TestController::platformAdjustContext): Use the 'ServiceWorkers'
subdirectory for the temporary testing-purpose SW registration directory
like it's used in Cocoa's TestController, addressing the review comment
that I forgot about.

Modified Paths

trunk/Tools/ChangeLog
trunk/Tools/WebKitTestRunner/TestController.cpp




Diff

Modified: trunk/Tools/ChangeLog (230390 => 230391)

--- trunk/Tools/ChangeLog	2018-04-09 07:14:15 UTC (rev 230390)
+++ trunk/Tools/ChangeLog	2018-04-09 08:08:50 UTC (rev 230391)
@@ -1,3 +1,13 @@
+2018-04-09  Zan Dobersek  
+
+Unreviewed follow-up to r230389.
+
+* WebKitTestRunner/TestController.cpp:
+(WTR::TestController::platformAdjustContext): Use the 'ServiceWorkers'
+subdirectory for the temporary testing-purpose SW registration directory
+like it's used in Cocoa's TestController, addressing the review comment
+that I forgot about.
+
 2018-04-08  Zan Dobersek  
 
 Non-Cocoa ports use default directory for ServiceWorker data during testing


Modified: trunk/Tools/WebKitTestRunner/TestController.cpp (230390 => 230391)

--- trunk/Tools/WebKitTestRunner/TestController.cpp	2018-04-09 07:14:15 UTC (rev 230390)
+++ trunk/Tools/WebKitTestRunner/TestController.cpp	2018-04-09 08:08:50 UTC (rev 230391)
@@ -2401,7 +2401,7 @@
 String temporaryFolder = String::fromUTF8(dumpRenderTreeTemp);
 const char separator = '/';
 
-WKWebsiteDataStoreSetServiceWorkerRegistrationDirectory(dataStore, toWK(temporaryFolder + separator + "ServiceWorkerRegistration").get());
+WKWebsiteDataStoreSetServiceWorkerRegistrationDirectory(dataStore, toWK(temporaryFolder + separator + "ServiceWorkers").get());
 }
 
 return context;






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

[webkit-changes] [230390] trunk/Source

[mcatanzaro]

Title: [230390] trunk/Source








Revision 230390
Author mcatanz...@igalia.com
Date 2018-04-09 00:14:15 -0700 (Mon, 09 Apr 2018)


Log Message
[GTK] WaylandCompositorDisplay leaks its wl_display
https://bugs.webkit.org/show_bug.cgi?id=184406

Reviewed by Carlos Garcia Campos.

Source/WebCore:

Remove the protected default constructor, for good measure.

* platform/graphics/wayland/PlatformDisplayWayland.h:

Source/WebKit:

* WebProcess/gtk/WaylandCompositorDisplay.cpp:
(WebKit::WaylandCompositorDisplay::create): Fix a log message (drive-by).
(WebKit::WaylandCompositorDisplay::WaylandCompositorDisplay): Fix the leak.

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (230389 => 230390)

--- trunk/Source/WebCore/ChangeLog	2018-04-09 06:31:05 UTC (rev 230389)
+++ trunk/Source/WebCore/ChangeLog	2018-04-09 07:14:15 UTC (rev 230390)
@@ -1,3 +1,14 @@
+2018-04-09  Michael Catanzaro  
+
+[GTK] WaylandCompositorDisplay leaks its wl_display
+https://bugs.webkit.org/show_bug.cgi?id=184406
+
+Reviewed by Carlos Garcia Campos.
+
+Remove the protected default constructor, for good measure.
+
+* platform/graphics/wayland/PlatformDisplayWayland.h:
+
 2018-04-08  Fujii Hironori  
 
 [CMake] WebKit should link to WebCore as a PRIVATE library if WebCore is a static library


Modified: trunk/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h (230389 => 230390)

--- trunk/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h	2018-04-09 06:31:05 UTC (rev 230389)
+++ trunk/Source/WebCore/platform/graphics/wayland/PlatformDisplayWayland.h	2018-04-09 07:14:15 UTC (rev 230390)
@@ -50,7 +50,6 @@
 Type type() const override { return PlatformDisplay::Type::Wayland; }
 
 protected:
-PlatformDisplayWayland() = default;
 void initialize(struct wl_display*);
 
 virtual void registryGlobal(const char* interface, uint32_t name);


Modified: trunk/Source/WebKit/ChangeLog (230389 => 230390)

--- trunk/Source/WebKit/ChangeLog	2018-04-09 06:31:05 UTC (rev 230389)
+++ trunk/Source/WebKit/ChangeLog	2018-04-09 07:14:15 UTC (rev 230390)
@@ -1,3 +1,14 @@
+2018-04-09  Michael Catanzaro  
+
+[GTK] WaylandCompositorDisplay leaks its wl_display
+https://bugs.webkit.org/show_bug.cgi?id=184406
+
+Reviewed by Carlos Garcia Campos.
+
+* WebProcess/gtk/WaylandCompositorDisplay.cpp:
+(WebKit::WaylandCompositorDisplay::create): Fix a log message (drive-by).
+(WebKit::WaylandCompositorDisplay::WaylandCompositorDisplay): Fix the leak.
+
 2018-04-08  Zan Dobersek  
 
 Non-Cocoa ports use default directory for ServiceWorker data during testing


Modified: trunk/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp (230389 => 230390)

--- trunk/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp	2018-04-09 06:31:05 UTC (rev 230389)
+++ trunk/Source/WebKit/WebProcess/gtk/WaylandCompositorDisplay.cpp	2018-04-09 07:14:15 UTC (rev 230390)
@@ -44,7 +44,7 @@
 
 struct wl_display* display = wl_display_connect(displayName.utf8().data());
 if (!display) {
-WTFLogAlways("PlatformDisplayWayland initialization: failed to connect to the Wayland display: %s", displayName.utf8().data());
+WTFLogAlways("WaylandCompositorDisplay initialization: failed to connect to the Wayland display: %s", displayName.utf8().data());
 return nullptr;
 }
 
@@ -61,8 +61,8 @@
 }
 
 WaylandCompositorDisplay::WaylandCompositorDisplay(struct wl_display* display)
+: PlatformDisplayWayland(display, NativeDisplayOwned::Yes)
 {
-initialize(display);
 PlatformDisplay::setSharedDisplayForCompositing(*this);
 }
 






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

[webkit-changes] [230389] trunk

[zandobersek]

Title: [230389] trunk








Revision 230389
Author zandober...@gmail.com
Date 2018-04-08 23:31:05 -0700 (Sun, 08 Apr 2018)


Log Message
Non-Cocoa ports use default directory for ServiceWorker data during testing
https://bugs.webkit.org/show_bug.cgi?id=183784

Reviewed by Youenn Fablet.

Source/WebKit:

Add API to WKWebsiteDataStore that enables setting and retrieving the
service worker registration directory for a given data store object.
This enables setting the temporary directory for testing purposes in
WebKitTestRunner.

* UIProcess/API/C/WKWebsiteDataStoreRef.cpp:
(WKWebsiteDataStoreCopyServiceWorkerRegistrationDirectory):
(WKWebsiteDataStoreSetServiceWorkerRegistrationDirectory):
* UIProcess/API/C/WKWebsiteDataStoreRef.h:

Tools:

* WebKitTestRunner/TestController.cpp:
(WTR::TestController::platformAdjustContext): Set the service worker registration
directory through the new WKWebsiteDataStore API for non-Cocoa ports.

Modified Paths

trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/API/C/WKWebsiteDataStoreRef.cpp
trunk/Source/WebKit/UIProcess/API/C/WKWebsiteDataStoreRef.h
trunk/Tools/ChangeLog
trunk/Tools/WebKitTestRunner/TestController.cpp




Diff

Modified: trunk/Source/WebKit/ChangeLog (230388 => 230389)

--- trunk/Source/WebKit/ChangeLog	2018-04-09 04:55:23 UTC (rev 230388)
+++ trunk/Source/WebKit/ChangeLog	2018-04-09 06:31:05 UTC (rev 230389)
@@ -1,3 +1,20 @@
+2018-04-08  Zan Dobersek  
+
+Non-Cocoa ports use default directory for ServiceWorker data during testing
+https://bugs.webkit.org/show_bug.cgi?id=183784
+
+Reviewed by Youenn Fablet.
+
+Add API to WKWebsiteDataStore that enables setting and retrieving the
+service worker registration directory for a given data store object.
+This enables setting the temporary directory for testing purposes in
+WebKitTestRunner.
+
+* UIProcess/API/C/WKWebsiteDataStoreRef.cpp:
+(WKWebsiteDataStoreCopyServiceWorkerRegistrationDirectory):
+(WKWebsiteDataStoreSetServiceWorkerRegistrationDirectory):
+* UIProcess/API/C/WKWebsiteDataStoreRef.h:
+
 2018-04-08  Fujii Hironori  
 
 [CMake] WebKit should link to WebCore as a PRIVATE library if WebCore is a static library


Modified: trunk/Source/WebKit/UIProcess/API/C/WKWebsiteDataStoreRef.cpp (230388 => 230389)

--- trunk/Source/WebKit/UIProcess/API/C/WKWebsiteDataStoreRef.cpp	2018-04-09 04:55:23 UTC (rev 230388)
+++ trunk/Source/WebKit/UIProcess/API/C/WKWebsiteDataStoreRef.cpp	2018-04-09 06:31:05 UTC (rev 230389)
@@ -493,3 +493,13 @@
 callback(0, context);
 });
 }
+
+WKStringRef WKWebsiteDataStoreCopyServiceWorkerRegistrationDirectory(WKWebsiteDataStoreRef dataStoreRef)
+{
+return WebKit::toCopiedAPI(WebKit::toImpl(dataStoreRef)->websiteDataStore().serviceWorkerRegistrationDirectory());
+}
+
+void WKWebsiteDataStoreSetServiceWorkerRegistrationDirectory(WKWebsiteDataStoreRef dataStoreRef, WKStringRef serviceWorkerRegistrationDirectory)
+{
+WebKit::toImpl(dataStoreRef)->websiteDataStore().setServiceWorkerRegistrationDirectory(WebKit::toImpl(serviceWorkerRegistrationDirectory)->string());
+}


Modified: trunk/Source/WebKit/UIProcess/API/C/WKWebsiteDataStoreRef.h (230388 => 230389)

--- trunk/Source/WebKit/UIProcess/API/C/WKWebsiteDataStoreRef.h	2018-04-09 04:55:23 UTC (rev 230388)
+++ trunk/Source/WebKit/UIProcess/API/C/WKWebsiteDataStoreRef.h	2018-04-09 06:31:05 UTC (rev 230389)
@@ -101,6 +101,9 @@
 typedef void (*WKWebsiteDataStoreGetFetchCacheSizeForOriginFunction)(uint64_t, void*);
 WK_EXPORT void WKWebsiteDataStoreGetFetchCacheSizeForOrigin(WKWebsiteDataStoreRef dataStoreRef, WKStringRef origin, void* context, WKWebsiteDataStoreGetFetchCacheSizeForOriginFunction function);
 
+WK_EXPORT WKStringRef WKWebsiteDataStoreCopyServiceWorkerRegistrationDirectory(WKWebsiteDataStoreRef dataStoreRef);
+WK_EXPORT void WKWebsiteDataStoreSetServiceWorkerRegistrationDirectory(WKWebsiteDataStoreRef dataStoreRef, WKStringRef serviceWorkerRegistrationDirectory);
+
 #ifdef __cplusplus
 }
 #endif


Modified: trunk/Tools/ChangeLog (230388 => 230389)

--- trunk/Tools/ChangeLog	2018-04-09 04:55:23 UTC (rev 230388)
+++ trunk/Tools/ChangeLog	2018-04-09 06:31:05 UTC (rev 230389)
@@ -1,3 +1,14 @@
+2018-04-08  Zan Dobersek  
+
+Non-Cocoa ports use default directory for ServiceWorker data during testing
+https://bugs.webkit.org/show_bug.cgi?id=183784
+
+Reviewed by Youenn Fablet.
+
+* WebKitTestRunner/TestController.cpp:
+(WTR::TestController::platformAdjustContext): Set the service worker registration
+directory through the new WKWebsiteDataStore API for non-Cocoa ports.
+
 2018-04-08  Zalan Bujtas  
 
 [LayoutReloaded] Adjust line with inline container's margin border and padding


Modified: trunk/Tools/WebKitTestRunner/TestController.cpp (230388 => 230389)

---