[webkit-changes] [225519] trunk/Source/JavaScriptCore
Title: [225519] trunk/Source/_javascript_Core Revision 225519 Author jfbast...@apple.com Date 2017-12-04 23:51:33 -0800 (Mon, 04 Dec 2017) Log Message Math: don't redundantly check for exceptions, just release scope https://bugs.webkit.org/show_bug.cgi?id=180395 Rubber stamped by Mark Lam. Two of the exceptions checks could just have been exception scope releases before the return, which is ever-so-slightly more efficient. The same technically applies where we have loops over parameters, but doing the scope release there isn't really more efficient and is way harder to read. * runtime/MathObject.cpp: (JSC::mathProtoFuncATan2): (JSC::mathProtoFuncPow): Modified Paths trunk/Source/_javascript_Core/ChangeLog trunk/Source/_javascript_Core/runtime/MathObject.cpp Diff Modified: trunk/Source/_javascript_Core/ChangeLog (225518 => 225519) --- trunk/Source/_javascript_Core/ChangeLog 2017-12-05 06:25:56 UTC (rev 225518) +++ trunk/Source/_javascript_Core/ChangeLog 2017-12-05 07:51:33 UTC (rev 225519) @@ -1,3 +1,20 @@ +2017-12-04 JF Bastien + +Math: don't redundantly check for exceptions, just release scope +https://bugs.webkit.org/show_bug.cgi?id=180395 + +Rubber stamped by Mark Lam. + +Two of the exceptions checks could just have been exception scope +releases before the return, which is ever-so-slightly more +efficient. The same technically applies where we have loops over +parameters, but doing the scope release there isn't really more +efficient and is way harder to read. + +* runtime/MathObject.cpp: +(JSC::mathProtoFuncATan2): +(JSC::mathProtoFuncPow): + 2017-12-04 David Quesada Add a class for parsing application manifests Modified: trunk/Source/_javascript_Core/runtime/MathObject.cpp (225518 => 225519) --- trunk/Source/_javascript_Core/runtime/MathObject.cpp 2017-12-05 06:25:56 UTC (rev 225518) +++ trunk/Source/_javascript_Core/runtime/MathObject.cpp 2017-12-05 07:51:33 UTC (rev 225519) @@ -153,8 +153,8 @@ auto scope = DECLARE_THROW_SCOPE(vm); double arg0 = exec->argument(0).toNumber(exec); RETURN_IF_EXCEPTION(scope, encodedJSValue()); +scope.release(); double arg1 = exec->argument(1).toNumber(exec); -RETURN_IF_EXCEPTION(scope, encodedJSValue()); return JSValue::encode(jsDoubleNumber(atan2(arg0, arg1))); } @@ -265,8 +265,8 @@ double arg = exec->argument(0).toNumber(exec); RETURN_IF_EXCEPTION(scope, encodedJSValue()); +scope.release(); double arg2 = exec->argument(1).toNumber(exec); -RETURN_IF_EXCEPTION(scope, encodedJSValue()); return JSValue::encode(JSValue(operationMathPow(arg, arg2))); } ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225518] branches/safari-604-branch
Title: [225518] branches/safari-604-branch
Revision 225518
Author jmarc...@apple.com
Date 2017-12-04 22:25:56 -0800 (Mon, 04 Dec 2017)
Log Message
Cherry-pick r225273. rdar://problem/35838163
Modified Paths
branches/safari-604-branch/JSTests/ChangeLog
branches/safari-604-branch/Source/_javascript_Core/ChangeLog
branches/safari-604-branch/Source/_javascript_Core/dfg/DFGAbstractInterpreterInlines.h
branches/safari-604-branch/Source/_javascript_Core/dfg/DFGSpeculativeJIT.cpp
branches/safari-604-branch/Source/_javascript_Core/ftl/FTLLowerDFGToB3.cpp
branches/safari-604-branch/Source/_javascript_Core/runtime/FunctionConstructor.cpp
branches/safari-604-branch/Source/_javascript_Core/runtime/JSFunction.cpp
branches/safari-604-branch/Source/_javascript_Core/runtime/JSFunctionInlines.h
branches/safari-604-branch/Source/_javascript_Core/runtime/JSGlobalObject.cpp
branches/safari-604-branch/Source/_javascript_Core/runtime/JSGlobalObject.h
Added Paths
branches/safari-604-branch/JSTests/stress/get-by-id-strict-arguments.js
branches/safari-604-branch/JSTests/stress/get-by-id-strict-callee.js
branches/safari-604-branch/JSTests/stress/get-by-id-strict-caller.js
branches/safari-604-branch/JSTests/stress/get-by-id-strict-nested-arguments-2.js
branches/safari-604-branch/JSTests/stress/get-by-id-strict-nested-arguments.js
branches/safari-604-branch/JSTests/stress/strict-function-structure.js
branches/safari-604-branch/JSTests/stress/strict-nested-function-structure.js
Diff
Modified: branches/safari-604-branch/JSTests/ChangeLog (225517 => 225518)
--- branches/safari-604-branch/JSTests/ChangeLog 2017-12-05 06:25:50 UTC (rev 225517)
+++ branches/safari-604-branch/JSTests/ChangeLog 2017-12-05 06:25:56 UTC (rev 225518)
@@ -1,5 +1,58 @@
2017-12-04 Jason Marcell
+Cherry-pick r225273. rdar://problem/35838163
+
+2017-11-28 JF Bastien
+
+Strict and sloppy functions shouldn't share structure
+https://bugs.webkit.org/show_bug.cgi?id=180103
+
+
+Reviewed by Saam Barati.
+
+* stress/get-by-id-strict-arguments.js: Added. Used to not throw
+because the IC was wrong.
+(foo):
+(bar):
+(baz):
+(catch):
+* stress/get-by-id-strict-callee.js: Added. Not strictly necessary
+in this patch, but may as well test odd strict mode corner cases.
+(bar):
+(baz):
+(catch):
+* stress/get-by-id-strict-caller.js: Added. Also IC'd wrong.
+(foo):
+(bar):
+(baz):
+(catch):
+* stress/get-by-id-strict-nested-arguments-2.js: Added. Same as
+next file, but with invalidation of the FunctionExecutable's
+singletonFunction() to hit SpeculativeJIT::compileNewFunction's
+slower path.
+(foo):
+(bar.const.x):
+(bar.const.y):
+(bar):
+(catch):
+* stress/get-by-id-strict-nested-arguments.js: Added. Make sure
+strict nesting works correctly.
+(foo):
+(bar.baz):
+(bar):
+* stress/strict-function-structure.js: Added. The test used to
+assert in objectProtoFuncHasOwnProperty.
+(foo):
+(bar):
+(baz):
+* stress/strict-nested-function-structure.js: Added. Nesting.
+(foo):
+(bar):
+(baz.boo):
+(baz):
+
+2017-12-04 Jason Marcell
+
Cherry-pick r225239. rdar://problem/35838157
2017-11-27 JF Bastien
Added: branches/safari-604-branch/JSTests/stress/get-by-id-strict-arguments.js (0 => 225518)
--- branches/safari-604-branch/JSTests/stress/get-by-id-strict-arguments.js (rev 0)
+++ branches/safari-604-branch/JSTests/stress/get-by-id-strict-arguments.js 2017-12-05 06:25:56 UTC (rev 225518)
@@ -0,0 +1,28 @@
+let warm = 1000;
+
+function foo(f) {
+return f.arguments;
+}
+noInline(foo);
+
+function bar() {
+for (let i = 0; i < warm; ++i)
+foo(bar);
+}
+function baz() {
+"use strict";
+foo(baz);
+}
+
+bar();
+
+let caught = false;
+
+try {
+baz();
+} catch (e) {
+caught = true;
+}
+
+if (!caught)
+throw new Error(`bad!`);
Added: branches/safari-604-branch/JSTests/stress/get-by-id-strict-callee.js (0 => 225518)
--- branches/safari-604-branch/JSTests/stress/get-by-id-strict-callee.js (rev 0)
+++ branches/safari-604-branch/JSTests/stress/get-by-id-strict-callee.js 2017-12-05 06:25:56 UTC (rev 225518)
@@ -0,0 +1,24 @@
+let warm = 1000;
+
+function bar() {
+for (let i = 0; i < warm; ++i)
+arguments.callee;
+}
+
+function baz() {
+"use strict";
+arguments.callee;
+}
+
+bar();
+
+let caught = false;
+
+try {
+baz();
+} catch (e) {
+caught = true;
+}
+
+if (!caught)
+throw new Error(`bad!`);
Added: bran
[webkit-changes] [225517] branches/safari-604-branch
Title: [225517] branches/safari-604-branch
Revision 225517
Author jmarc...@apple.com
Date 2017-12-04 22:25:50 -0800 (Mon, 04 Dec 2017)
Log Message
Cherry-pick r225239. rdar://problem/35838157
Modified Paths
branches/safari-604-branch/JSTests/ChangeLog
branches/safari-604-branch/Source/_javascript_Core/ChangeLog
branches/safari-604-branch/Source/_javascript_Core/dfg/DFGAbstractInterpreterInlines.h
Added Paths
branches/safari-604-branch/JSTests/stress/rest-parameter-negative.js
Diff
Modified: branches/safari-604-branch/JSTests/ChangeLog (225516 => 225517)
--- branches/safari-604-branch/JSTests/ChangeLog 2017-12-05 05:23:14 UTC (rev 225516)
+++ branches/safari-604-branch/JSTests/ChangeLog 2017-12-05 06:25:50 UTC (rev 225517)
@@ -1,3 +1,21 @@
+2017-12-04 Jason Marcell
+
+Cherry-pick r225239. rdar://problem/35838157
+
+2017-11-27 JF Bastien
+
+_javascript_ rest function parameter with negative index leads to bad DFG abstract interpretation
+https://bugs.webkit.org/show_bug.cgi?id=180051
+
+
+Reviewed by Saam Barati.
+
+* stress/rest-parameter-negative.js: Added.
+(__f_5484):
+(catch):
+(__f_5485):
+(__v_22598.catch):
+
2017-11-28 Jason Marcell
Cherry-pick r224539. rdar://problem/35698788
Added: branches/safari-604-branch/JSTests/stress/rest-parameter-negative.js (0 => 225517)
--- branches/safari-604-branch/JSTests/stress/rest-parameter-negative.js (rev 0)
+++ branches/safari-604-branch/JSTests/stress/rest-parameter-negative.js 2017-12-05 06:25:50 UTC (rev 225517)
@@ -0,0 +1,21 @@
+function __f_5484(__v_22596) {
+ if (!__v_22596) throw new Error();
+}
+
+try {
+ noInline(__f_5484);
+} catch (e) {}
+
+function __f_5485(...__v_22597) {
+ return __v_22597[-13];
+}
+
+try {
+ noInline(__f_5485);
+} catch (e) {}
+
+for (let __v_22598 = 0; __v_22598 < 1; __v_22598++) {
+ try {
+__f_5484(__f_5485(__v_22598) === __v_22598);
+ } catch (e) {}
+}
Modified: branches/safari-604-branch/Source/_javascript_Core/ChangeLog (225516 => 225517)
--- branches/safari-604-branch/Source/_javascript_Core/ChangeLog 2017-12-05 05:23:14 UTC (rev 225516)
+++ branches/safari-604-branch/Source/_javascript_Core/ChangeLog 2017-12-05 06:25:50 UTC (rev 225517)
@@ -1,3 +1,21 @@
+2017-12-04 Jason Marcell
+
+Cherry-pick r225239. rdar://problem/35838157
+
+2017-11-27 JF Bastien
+
+_javascript_ rest function parameter with negative index leads to bad DFG abstract interpretation
+https://bugs.webkit.org/show_bug.cgi?id=180051
+
+
+Reviewed by Saam Barati.
+
+Checking for int32 isn't sufficient when uint32 is expected
+afterwards. While we're here, also use Checked<>.
+
+* dfg/DFGAbstractInterpreterInlines.h:
+(JSC::DFG::AbstractInterpreter::executeEffects):
+
2017-11-28 Jason Marcell
Cherry-pick r224539. rdar://problem/35698788
Modified: branches/safari-604-branch/Source/_javascript_Core/dfg/DFGAbstractInterpreterInlines.h (225516 => 225517)
--- branches/safari-604-branch/Source/_javascript_Core/dfg/DFGAbstractInterpreterInlines.h 2017-12-05 05:23:14 UTC (rev 225516)
+++ branches/safari-604-branch/Source/_javascript_Core/dfg/DFGAbstractInterpreterInlines.h 2017-12-05 06:25:50 UTC (rev 225517)
@@ -40,6 +40,8 @@
#include "PutByIdStatus.h"
#include "StringObject.h"
+#include
+
namespace JSC { namespace DFG {
template
@@ -1732,25 +1734,29 @@
JSValue index = forNode(node->child2()).m_value;
InlineCallFrame* inlineCallFrame = node->child1()->origin.semantic.inlineCallFrame;
-if (index && index.isInt32()) {
+if (index && index.isUInt32()) {
// This pretends to return TOP for accesses that are actually proven out-of-bounds because
// that's the conservative thing to do. Otherwise we'd need to write more code to mark such
// paths as unreachable, or to return undefined. We could implement that eventually.
-
-unsigned argumentIndex = index.asUInt32() + node->numberOfArgumentsToSkip();
-if (inlineCallFrame) {
-if (argumentIndex < inlineCallFrame->arguments.size() - 1) {
-forNode(node) = m_state.variables().operand(
-virtualRegisterForArgument(argumentIndex + 1) + inlineCallFrame->stackOffset);
-m_state.setFoundConstants(true);
-break;
+
+Checked argumentIndexChecked = index.asUInt32();
+argumentIndexChecked += node->numberOfArgumentsToSkip();
+unsigned argumentIndex;
+if (argumentIndexChecked.safeGet(argumentIndex) != CheckedState::DidOverflow) {
+if (inlineCallFrame) {
+if (argumentIndex < inlineCallFrame->arguments.size() - 1) {
+
[webkit-changes] [225516] releases/Apple
Title: [225516] releases/Apple Revision 225516 Author m...@apple.com Date 2017-12-04 21:23:14 -0800 (Mon, 04 Dec 2017) Log Message Added a tag for iOS 11.2. Added Paths releases/Apple/iOS 11.2/ releases/Apple/iOS 11.2/ANGLE/ releases/Apple/iOS 11.2/_javascript_Core/ releases/Apple/iOS 11.2/WTF/ releases/Apple/iOS 11.2/WebCore/ releases/Apple/iOS 11.2/WebKit/ releases/Apple/iOS 11.2/WebKitLegacy/ releases/Apple/iOS 11.2/bmalloc/ releases/Apple/iOS 11.2/libwebrtc/ Diff Index: releases/Apple/iOS 11.2/ANGLE === --- tags/Safari-604.4.7.0.3/Source/ThirdParty/ANGLE 2017-12-05 05:20:27 UTC (rev 225515) +++ releases/Apple/iOS 11.2/ANGLE 2017-12-05 05:23:14 UTC (rev 225516) Property changes: releases/Apple/iOS 11.2/ANGLE Added: allow-tabs +true \ No newline at end of property Added: svn:mergeinfo +/trunk/Source/ThirdParty/ANGLE:53455 \ No newline at end of property Index: releases/Apple/iOS 11.2/_javascript_Core === --- tags/Safari-604.4.7.0.3/Source/_javascript_Core 2017-12-05 05:20:27 UTC (rev 225515) +++ releases/Apple/iOS 11.2/_javascript_Core 2017-12-05 05:23:14 UTC (rev 225516) Property changes: releases/Apple/iOS 11.2/_javascript_Core Added: svn:mergeinfo +/trunk/Source/_javascript_Core:53455 \ No newline at end of property Index: releases/Apple/iOS 11.2/WTF === --- tags/Safari-604.4.7.0.3/Source/WTF 2017-12-05 05:20:27 UTC (rev 225515) +++ releases/Apple/iOS 11.2/WTF 2017-12-05 05:23:14 UTC (rev 225516) Property changes: releases/Apple/iOS 11.2/WTF Added: svn:mergeinfo +/trunk/Source/WTF:53455 \ No newline at end of property Index: releases/Apple/iOS 11.2/bmalloc === --- tags/Safari-604.4.7.0.3/Source/bmalloc 2017-12-05 05:20:27 UTC (rev 225515) +++ releases/Apple/iOS 11.2/bmalloc 2017-12-05 05:23:14 UTC (rev 225516) Property changes: releases/Apple/iOS 11.2/bmalloc Added: svn:mergeinfo +/trunk/Source/bmalloc:53455 \ No newline at end of property Index: releases/Apple/iOS 11.2/libwebrtc === --- tags/Safari-604.4.7.0.3/Source/ThirdParty/libwebrtc 2017-12-05 05:20:27 UTC (rev 225515) +++ releases/Apple/iOS 11.2/libwebrtc 2017-12-05 05:23:14 UTC (rev 225516) Property changes: releases/Apple/iOS 11.2/libwebrtc Added: svn:mergeinfo +/trunk/Source/ThirdParty/libwebrtc:53455 \ No newline at end of property ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225515] releases/Apple
Title: [225515] releases/Apple Revision 225515 Author m...@apple.com Date 2017-12-04 21:20:27 -0800 (Mon, 04 Dec 2017) Log Message Added a tag for Safari 11.0.1. Added Paths releases/Apple/Safari 11.0.1/ releases/Apple/Safari 11.0.1/ANGLE/ releases/Apple/Safari 11.0.1/_javascript_Core/ releases/Apple/Safari 11.0.1/WTF/ releases/Apple/Safari 11.0.1/WebCore/ releases/Apple/Safari 11.0.1/WebInspectorUI/ releases/Apple/Safari 11.0.1/WebKit/ releases/Apple/Safari 11.0.1/WebKitLegacy/ releases/Apple/Safari 11.0.1/bmalloc/ releases/Apple/Safari 11.0.1/libwebrtc/ Diff Index: releases/Apple/Safari 11.0.1/ANGLE === --- tags/Safari-604.3.5/Source/ThirdParty/ANGLE 2017-12-05 05:18:50 UTC (rev 225514) +++ releases/Apple/Safari 11.0.1/ANGLE 2017-12-05 05:20:27 UTC (rev 225515) Property changes: releases/Apple/Safari 11.0.1/ANGLE Added: allow-tabs +true \ No newline at end of property Added: svn:mergeinfo +/trunk/Source/ThirdParty/ANGLE:53455 \ No newline at end of property Index: releases/Apple/Safari 11.0.1/_javascript_Core === --- tags/Safari-604.3.5/Source/_javascript_Core 2017-12-05 05:18:50 UTC (rev 225514) +++ releases/Apple/Safari 11.0.1/_javascript_Core 2017-12-05 05:20:27 UTC (rev 225515) Property changes: releases/Apple/Safari 11.0.1/_javascript_Core Added: svn:mergeinfo +/trunk/Source/_javascript_Core:53455 \ No newline at end of property Index: releases/Apple/Safari 11.0.1/WTF === --- tags/Safari-604.3.5/Source/WTF 2017-12-05 05:18:50 UTC (rev 225514) +++ releases/Apple/Safari 11.0.1/WTF 2017-12-05 05:20:27 UTC (rev 225515) Property changes: releases/Apple/Safari 11.0.1/WTF Added: svn:mergeinfo +/trunk/Source/WTF:53455 \ No newline at end of property Index: releases/Apple/Safari 11.0.1/WebInspectorUI === --- tags/Safari-604.3.5/Source/WebInspectorUI 2017-12-05 05:18:50 UTC (rev 225514) +++ releases/Apple/Safari 11.0.1/WebInspectorUI 2017-12-05 05:20:27 UTC (rev 225515) Property changes: releases/Apple/Safari 11.0.1/WebInspectorUI Added: svn:mergeinfo +/trunk/Source/WebInspectorUI:53455 \ No newline at end of property Index: releases/Apple/Safari 11.0.1/bmalloc === --- tags/Safari-604.3.5/Source/bmalloc 2017-12-05 05:18:50 UTC (rev 225514) +++ releases/Apple/Safari 11.0.1/bmalloc 2017-12-05 05:20:27 UTC (rev 225515) Property changes: releases/Apple/Safari 11.0.1/bmalloc Added: svn:mergeinfo +/trunk/Source/bmalloc:53455 \ No newline at end of property Index: releases/Apple/Safari 11.0.1/libwebrtc === --- tags/Safari-604.3.5/Source/ThirdParty/libwebrtc 2017-12-05 05:18:50 UTC (rev 225514) +++ releases/Apple/Safari 11.0.1/libwebrtc 2017-12-05 05:20:27 UTC (rev 225515) Property changes: releases/Apple/Safari 11.0.1/libwebrtc Added: svn:mergeinfo +/trunk/Source/ThirdParty/libwebrtc:53455 \ No newline at end of property ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225514] releases/Apple
Title: [225514] releases/Apple Revision 225514 Author m...@apple.com Date 2017-12-04 21:18:50 -0800 (Mon, 04 Dec 2017) Log Message Added a tag for iOS 11.1 Added Paths releases/Apple/iOS 11.1/ releases/Apple/iOS 11.1/ANGLE/ releases/Apple/iOS 11.1/_javascript_Core/ releases/Apple/iOS 11.1/WTF/ releases/Apple/iOS 11.1/WebCore/ releases/Apple/iOS 11.1/WebKit/ releases/Apple/iOS 11.1/WebKitLegacy/ releases/Apple/iOS 11.1/bmalloc/ releases/Apple/iOS 11.1/libwebrtc/ Diff Index: releases/Apple/iOS 11.1/ANGLE === --- tags/Safari-604.3.5.0.3/Source/ThirdParty/ANGLE 2017-12-05 02:06:26 UTC (rev 225513) +++ releases/Apple/iOS 11.1/ANGLE 2017-12-05 05:18:50 UTC (rev 225514) Property changes: releases/Apple/iOS 11.1/ANGLE Added: allow-tabs +true \ No newline at end of property Added: svn:mergeinfo +/trunk/Source/ThirdParty/ANGLE:53455 \ No newline at end of property Index: releases/Apple/iOS 11.1/_javascript_Core === --- tags/Safari-604.3.5.0.3/Source/_javascript_Core 2017-12-05 02:06:26 UTC (rev 225513) +++ releases/Apple/iOS 11.1/_javascript_Core 2017-12-05 05:18:50 UTC (rev 225514) Property changes: releases/Apple/iOS 11.1/_javascript_Core Added: svn:mergeinfo +/trunk/Source/_javascript_Core:53455 \ No newline at end of property Index: releases/Apple/iOS 11.1/WTF === --- tags/Safari-604.3.5.0.3/Source/WTF 2017-12-05 02:06:26 UTC (rev 225513) +++ releases/Apple/iOS 11.1/WTF 2017-12-05 05:18:50 UTC (rev 225514) Property changes: releases/Apple/iOS 11.1/WTF Added: svn:mergeinfo +/trunk/Source/WTF:53455 \ No newline at end of property Index: releases/Apple/iOS 11.1/bmalloc === --- tags/Safari-604.3.5.0.3/Source/bmalloc 2017-12-05 02:06:26 UTC (rev 225513) +++ releases/Apple/iOS 11.1/bmalloc 2017-12-05 05:18:50 UTC (rev 225514) Property changes: releases/Apple/iOS 11.1/bmalloc Added: svn:mergeinfo +/trunk/Source/bmalloc:53455 \ No newline at end of property Index: releases/Apple/iOS 11.1/libwebrtc === --- tags/Safari-604.3.5.0.3/Source/ThirdParty/libwebrtc 2017-12-05 02:06:26 UTC (rev 225513) +++ releases/Apple/iOS 11.1/libwebrtc 2017-12-05 05:18:50 UTC (rev 225514) Property changes: releases/Apple/iOS 11.1/libwebrtc Added: svn:mergeinfo +/trunk/Source/ThirdParty/libwebrtc:53455 \ No newline at end of property ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225513] trunk
Title: [225513] trunk Revision 225513 Author cdu...@apple.com Date 2017-12-04 18:06:26 -0800 (Mon, 04 Dec 2017) Log Message Support container.getRegistration() / getRegistrations() inside service workers https://bugs.webkit.org/show_bug.cgi?id=180360 Reviewed by Youenn Fablet. LayoutTests/imported/w3c: * web-platform-tests/service-workers/service-worker/activation.https-expected.txt: Rebaseline test with slightly different output. * web-platform-tests/service-workers/service-worker/skip-waiting-without-using-registration.https-expected.txt: * web-platform-tests/service-workers/service-worker/skip-waiting-without-using-registration.https.html: With my change, this test started running and passing a check before timing out. When investigating the time out, I found out that this is caused by the test unregistering the worker while the test is still running in the service worker, which causes the worker to terminate early. To address the issue, we no longer add a cleanup step to unregister. The test now passes all checks. Source/WebCore: Support container.getRegistration() / getRegistrations() inside service workers by making sure we hop to the right thread when needed. Test: http/tests/workers/service/ServiceWorkerGlobalScope_getRegistration.html * dom/ScriptExecutionContext.cpp: (WebCore::ScriptExecutionContext::postTaskTo): * dom/ScriptExecutionContext.h: * workers/service/SWClientConnection.cpp: (WebCore::SWClientConnection::jobRejectedInServer): (WebCore::SWClientConnection::registrationJobResolvedInServer): (WebCore::SWClientConnection::unregistrationJobResolvedInServer): (WebCore::SWClientConnection::startScriptFetchForServer): (WebCore::SWClientConnection::clearPendingJobs): * workers/service/SWClientConnection.h: * workers/service/ServiceWorkerContainer.cpp: (WebCore::ServiceWorkerContainer::getRegistration): (WebCore::ServiceWorkerContainer::didFinishGetRegistrationRequest): (WebCore::ServiceWorkerContainer::getRegistrations): (WebCore::ServiceWorkerContainer::didFinishGetRegistrationsRequest): (WebCore::ServiceWorkerContainer::stop): * workers/service/ServiceWorkerContainer.h: Source/WebKit: * WebProcess/Storage/WebSWClientConnection.cpp: (WebKit::WebSWClientConnection::didMatchRegistration): (WebKit::WebSWClientConnection::didGetRegistrations): (WebKit::WebSWClientConnection::matchRegistration): (WebKit::WebSWClientConnection::getRegistrations): LayoutTests: Add layout test coverage. * http/tests/workers/service/ServiceWorkerGlobalScope_getRegistration-expected.txt: Added. * http/tests/workers/service/ServiceWorkerGlobalScope_getRegistration.html: Added. * http/tests/workers/service/resources/ServiceWorkerGlobalScope_getRegistration-worker.js: Added. Modified Paths trunk/LayoutTests/ChangeLog trunk/LayoutTests/TestExpectations trunk/LayoutTests/imported/w3c/ChangeLog trunk/LayoutTests/imported/w3c/web-platform-tests/service-workers/service-worker/activation.https-expected.txt trunk/LayoutTests/imported/w3c/web-platform-tests/service-workers/service-worker/skip-waiting-using-registration.https-expected.txt trunk/LayoutTests/imported/w3c/web-platform-tests/service-workers/service-worker/skip-waiting-using-registration.https.html trunk/LayoutTests/imported/w3c/web-platform-tests/service-workers/service-worker/skip-waiting-without-using-registration.https-expected.txt trunk/LayoutTests/imported/w3c/web-platform-tests/service-workers/service-worker/skip-waiting-without-using-registration.https.html trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/dom/ScriptExecutionContext.cpp trunk/Source/WebCore/dom/ScriptExecutionContext.h trunk/Source/WebCore/workers/service/SWClientConnection.cpp trunk/Source/WebCore/workers/service/SWClientConnection.h trunk/Source/WebCore/workers/service/ServiceWorkerContainer.cpp trunk/Source/WebCore/workers/service/ServiceWorkerContainer.h trunk/Source/WebKit/ChangeLog trunk/Source/WebKit/WebProcess/Storage/WebSWClientConnection.cpp Added Paths trunk/LayoutTests/http/tests/workers/service/ServiceWorkerGlobalScope_getRegistration-expected.txt trunk/LayoutTests/http/tests/workers/service/ServiceWorkerGlobalScope_getRegistration.html trunk/LayoutTests/http/tests/workers/service/resources/ServiceWorkerGlobalScope_getRegistration-worker.js Diff Modified: trunk/LayoutTests/ChangeLog (225512 => 225513) --- trunk/LayoutTests/ChangeLog 2017-12-05 01:37:20 UTC (rev 225512) +++ trunk/LayoutTests/ChangeLog 2017-12-05 02:06:26 UTC (rev 225513) @@ -1,3 +1,16 @@ +2017-12-04 Chris Dumez + +Support container.getRegistration() / getRegistrations() inside service workers +https://bugs.webkit.org/show_bug.cgi?id=180360 + +Reviewed by Youenn Fablet. + +Add layout test coverage. + +* http/tests/workers/service/ServiceWorkerGlobalScope_getRegistration-expected.txt: Added. +* http/tests/workers/service/ServiceWorkerGlobalScope_getRegistration.html: Added. +* http/tests/workers/service/resources/ServiceWorke
[webkit-changes] [225512] trunk/Source/WebCore
Title: [225512] trunk/Source/WebCore Revision 225512 Author simon.fra...@apple.com Date 2017-12-04 17:37:20 -0800 (Mon, 04 Dec 2017) Log Message Cleanup code that computes iframe content offsets in FrameView https://bugs.webkit.org/show_bug.cgi?id=180385 Reviewed by Zalan Bujtas. The coordinate conversion functions in FrameView explicitly added borderLeft/paddingLeft and borderTop/paddingTop to compute the offset from the contents of a frame to its parent renderer. This is equivalent to the call to contextBoxRect() that RenderWidget::updateWidgetGeometry() does, which also takes left-side scrollbars into account. Coordinate offsets don't need height and width, so add RenderBox::contentBoxLocation() that just computes the top left, and use it in FrameView. * page/FrameView.cpp: (WebCore::FrameView::invalidateRect): (WebCore::FrameView::convertToContainingView const): (WebCore::FrameView::convertFromContainingView const): * platform/graphics/FloatPoint.h: (WebCore::FloatPoint::scaled const): (WebCore::FloatPoint::scaled): Deleted. * rendering/RenderBox.cpp: (WebCore::RenderBox::contentBoxLocation const): * rendering/RenderBox.h: Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/page/FrameView.cpp trunk/Source/WebCore/platform/graphics/FloatPoint.h trunk/Source/WebCore/rendering/RenderBox.cpp trunk/Source/WebCore/rendering/RenderBox.h Diff Modified: trunk/Source/WebCore/ChangeLog (225511 => 225512) --- trunk/Source/WebCore/ChangeLog 2017-12-05 01:24:19 UTC (rev 225511) +++ trunk/Source/WebCore/ChangeLog 2017-12-05 01:37:20 UTC (rev 225512) @@ -1,3 +1,30 @@ +2017-12-04 Simon Fraser + +Cleanup code that computes iframe content offsets in FrameView +https://bugs.webkit.org/show_bug.cgi?id=180385 + +Reviewed by Zalan Bujtas. + +The coordinate conversion functions in FrameView explicitly added borderLeft/paddingLeft +and borderTop/paddingTop to compute the offset from the contents of a frame to its parent renderer. + +This is equivalent to the call to contextBoxRect() that RenderWidget::updateWidgetGeometry() does, +which also takes left-side scrollbars into account. + +Coordinate offsets don't need height and width, so add RenderBox::contentBoxLocation() that just +computes the top left, and use it in FrameView. + +* page/FrameView.cpp: +(WebCore::FrameView::invalidateRect): +(WebCore::FrameView::convertToContainingView const): +(WebCore::FrameView::convertFromContainingView const): +* platform/graphics/FloatPoint.h: +(WebCore::FloatPoint::scaled const): +(WebCore::FloatPoint::scaled): Deleted. +* rendering/RenderBox.cpp: +(WebCore::RenderBox::contentBoxLocation const): +* rendering/RenderBox.h: + 2017-12-04 Nan Wang AX: AOM: Implement relation type properties Modified: trunk/Source/WebCore/page/FrameView.cpp (225511 => 225512) --- trunk/Source/WebCore/page/FrameView.cpp 2017-12-05 01:24:19 UTC (rev 225511) +++ trunk/Source/WebCore/page/FrameView.cpp 2017-12-05 01:37:20 UTC (rev 225512) @@ -430,7 +430,7 @@ return; IntRect repaintRect = rect; -repaintRect.move(renderer->borderLeft() + renderer->paddingLeft(), renderer->borderTop() + renderer->paddingTop()); +repaintRect.moveBy(roundedIntPoint(renderer->contentBoxLocation())); renderer->repaintRectangle(repaintRect); } @@ -4384,10 +4384,8 @@ if (!renderer) return localRect; -IntRect rect(localRect); -// Add borders and padding?? -rect.move(renderer->borderLeft() + renderer->paddingLeft(), - renderer->borderTop() + renderer->paddingTop()); +auto rect = localRect; +rect.moveBy(roundedIntPoint(renderer->contentBoxLocation())); return parentView.convertFromRendererToContainingView(renderer, rect); } @@ -4408,10 +4406,8 @@ if (!renderer) return parentRect; -IntRect rect = parentView.convertFromContainingViewToRenderer(renderer, parentRect); -// Subtract borders and padding -rect.move(-renderer->borderLeft() - renderer->paddingLeft(), - -renderer->borderTop() - renderer->paddingTop()); +auto rect = parentView.convertFromContainingViewToRenderer(renderer, parentRect); +rect.moveBy(-roundedIntPoint(renderer->contentBoxLocation())); return rect; } @@ -4432,11 +4428,8 @@ if (!renderer) return localPoint; -IntPoint point(localPoint); - -// Add borders and padding -point.move(renderer->borderLeft() + renderer->paddingLeft(), - renderer->borderTop() + renderer->paddingTop()); +auto point = localPoint; +
[webkit-changes] [225511] trunk
Title: [225511] trunk Revision 225511 Author n_w...@apple.com Date 2017-12-04 17:24:19 -0800 (Mon, 04 Dec 2017) Log Message AX: AOM: Implement relation type properties https://bugs.webkit.org/show_bug.cgi?id=179500 Reviewed by Ryosuke Niwa. Source/WebCore: Accessibility Object Model Explainer: https://wicg.github.io/aom/explainer.html Spec: https://wicg.github.io/aom/spec/ Implemented the AOM support for activeDescendant, details and errorMessage. The corresponding ARIA attributes all take IDREFs, and the AOM properties take references to AccessibleNodes instead. Test: accessibility/mac/AOM-relation-property.html * accessibility/AccessibilityObject.cpp: (WebCore::AccessibilityObject::getAttribute const): (WebCore::AccessibilityObject::hasProperty const): (WebCore::AccessibilityObject::stringValueForProperty const): (WebCore::AccessibilityObject::boolValueForProperty const): (WebCore::AccessibilityObject::intValueForProperty const): (WebCore::AccessibilityObject::unsignedValueForProperty const): (WebCore::AccessibilityObject::doubleValueForProperty const): (WebCore::AccessibilityObject::elementValueForProperty const): (WebCore::AccessibilityObject::supportsARIAAttributes const): (WebCore::AccessibilityObject::elementsFromProperty const): (WebCore::AccessibilityObject::elementsReferencedByProperty const): (WebCore::AccessibilityObject::ariaActiveDescendantReferencingElements const): (WebCore::AccessibilityObject::ariaDetailsElements const): (WebCore::AccessibilityObject::ariaDetailsReferencingElements const): (WebCore::AccessibilityObject::ariaErrorMessageElements const): (WebCore::AccessibilityObject::ariaErrorMessageReferencingElements const): * accessibility/AccessibilityObject.h: * accessibility/AccessibilityRenderObject.cpp: (WebCore::AccessibilityRenderObject::activeDescendant const): * accessibility/AccessibleNode.cpp: (WebCore::ariaAttributeMap): (WebCore::isPropertyValueRelation): (WebCore::AccessibleNode::attributeFromAXPropertyName): (WebCore::AccessibleNode::setRelationProperty): (WebCore::AccessibleNode::singleRelationValueForProperty): (WebCore::AccessibleNode::activeDescendant const): (WebCore::AccessibleNode::setActiveDescendant): (WebCore::AccessibleNode::details const): (WebCore::AccessibleNode::setDetails): (WebCore::AccessibleNode::errorMessage const): (WebCore::AccessibleNode::setErrorMessage): * accessibility/AccessibleNode.h: * accessibility/AccessibleNode.idl: * accessibility/mac/WebAccessibilityObjectWrapperMac.mm: (-[WebAccessibilityObjectWrapper accessibilityAttributeValue:]): Tools: * WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h: (WTR::AccessibilityUIElement::ariaDetailsElementAtIndex): (WTR::AccessibilityUIElement::ariaErrorMessageElementAtIndex): (WTR::AccessibilityUIElement::ariaDescribedByReferencingElementAtIndex): (WTR::AccessibilityUIElement::ariaDetailsReferencingElementAtIndex): * WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm: (WTR::AccessibilityUIElement::ariaDetailsElementAtIndex): (WTR::AccessibilityUIElement::ariaErrorMessageElementAtIndex): LayoutTests: Only enable this test on WK2. * accessibility/mac/AOM-relation-property-expected.txt: Added. * accessibility/mac/AOM-relation-property.html: Added. * platform/mac-wk1/TestExpectations: Modified Paths trunk/LayoutTests/ChangeLog trunk/LayoutTests/platform/mac-wk1/TestExpectations trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/accessibility/AccessibilityObject.cpp trunk/Source/WebCore/accessibility/AccessibilityObject.h trunk/Source/WebCore/accessibility/AccessibilityRenderObject.cpp trunk/Source/WebCore/accessibility/AccessibleNode.cpp trunk/Source/WebCore/accessibility/AccessibleNode.h trunk/Source/WebCore/accessibility/AccessibleNode.idl trunk/Source/WebCore/accessibility/mac/WebAccessibilityObjectWrapperMac.mm trunk/Tools/ChangeLog trunk/Tools/WebKitTestRunner/InjectedBundle/AccessibilityUIElement.h trunk/Tools/WebKitTestRunner/InjectedBundle/mac/AccessibilityUIElementMac.mm Added Paths trunk/LayoutTests/accessibility/mac/AOM-relation-property-expected.txt trunk/LayoutTests/accessibility/mac/AOM-relation-property.html Diff Modified: trunk/LayoutTests/ChangeLog (225510 => 225511) --- trunk/LayoutTests/ChangeLog 2017-12-05 01:19:25 UTC (rev 225510) +++ trunk/LayoutTests/ChangeLog 2017-12-05 01:24:19 UTC (rev 225511) @@ -1,3 +1,16 @@ +2017-12-04 Nan Wang + +AX: AOM: Implement relation type properties +https://bugs.webkit.org/show_bug.cgi?id=179500 + +Reviewed by Ryosuke Niwa. + +Only enable this test on WK2. + +* accessibility/mac/AOM-relation-property-expected.txt: Added. +* accessibility/mac/AOM-relation-property.html: Added. +* platform/mac-wk1/TestExpectations: + 2017-12-04 Zalan Bujtas REGRESSION (r211531): Text flow changes and overlaps other text after double-click selecting paragraph Added: trunk/LayoutTests/accessibility/mac/AOM-relation-property-expected.txt (0 => 225
[webkit-changes] [225510] trunk/Source/WebKit
Title: [225510] trunk/Source/WebKit Revision 225510 Author beid...@apple.com Date 2017-12-04 17:19:25 -0800 (Mon, 04 Dec 2017) Log Message Followup to: Get a directory path to SWServers for storing ServiceWorker registrations https://bugs.webkit.org/show_bug.cgi?id=180362 Unreviewed. * StorageProcess/StorageProcess.cpp: (WebKit::StorageProcess::swServerForSession): Change this ASSERT which is invalid in private browsing sessions. Modified Paths trunk/Source/WebKit/ChangeLog trunk/Source/WebKit/StorageProcess/StorageProcess.cpp Diff Modified: trunk/Source/WebKit/ChangeLog (225509 => 225510) --- trunk/Source/WebKit/ChangeLog 2017-12-05 01:13:27 UTC (rev 225509) +++ trunk/Source/WebKit/ChangeLog 2017-12-05 01:19:25 UTC (rev 225510) @@ -1,3 +1,14 @@ +2017-12-04 Brady Eidson + +Followup to: +Get a directory path to SWServers for storing ServiceWorker registrations +https://bugs.webkit.org/show_bug.cgi?id=180362 + +Unreviewed. + +* StorageProcess/StorageProcess.cpp: +(WebKit::StorageProcess::swServerForSession): Change this ASSERT which is invalid in private browsing sessions. + 2017-12-04 Simon Fraser Minor DisplayRefreshMonitor-related cleanup Modified: trunk/Source/WebKit/StorageProcess/StorageProcess.cpp (225509 => 225510) --- trunk/Source/WebKit/StorageProcess/StorageProcess.cpp 2017-12-05 01:13:27 UTC (rev 225509) +++ trunk/Source/WebKit/StorageProcess/StorageProcess.cpp 2017-12-05 01:19:25 UTC (rev 225510) @@ -387,7 +387,7 @@ auto path = m_swDatabasePaths.get(sessionID); // There should already be a registered path for this PAL::SessionID. // If there's not, then where did this PAL::SessionID come from? -ASSERT(!path.isEmpty()); +ASSERT(sessionID.isEphemeral() || !path.isEmpty()); result.iterator->value = std::make_unique(makeUniqueRef(), path); return *result.iterator->value; ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225509] trunk
Title: [225509] trunk
Revision 225509
Author simon.fra...@apple.com
Date 2017-12-04 17:13:27 -0800 (Mon, 04 Dec 2017)
Log Message
Minor DisplayRefreshMonitor-related cleanup
https://bugs.webkit.org/show_bug.cgi?id=179802
Reviewed by Sam Weinig.
Source/WebCore:
New trace point for when the CVDisplayLink fires on its own thread.
Some #pragma once, and put all the WellKnownRunLoopOrders in one place.
* inspector/agents/InspectorTimelineAgent.cpp:
(WebCore::InspectorTimelineAgent::internalStart):
* platform/cf/RunLoopObserver.h:
* platform/graphics/DisplayRefreshMonitor.cpp:
(WebCore::DisplayRefreshMonitor::displayDidRefresh):
* platform/graphics/ca/LayerFlushScheduler.h:
* platform/graphics/ca/cocoa/LayerFlushSchedulerMac.cpp:
(WebCore::LayerFlushScheduler::LayerFlushScheduler):
Source/WebKit:
Use RunLoopObserver::WellKnownRunLoopOrders. We want to fire before layer flushing.
* UIProcess/WebPageProxy.cpp:
(WebKit::m_configurationPreferenceValues):
Source/WTF:
New trace point for when the CVDisplayLink fires on its own thread.
* wtf/SystemTracing.h:
Tools:
New trace point for when the CVDisplayLink fires on its own thread.
* Tracing/SystemTracePoints.plist:
Modified Paths
trunk/Source/WTF/ChangeLog
trunk/Source/WTF/wtf/SystemTracing.h
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/inspector/agents/InspectorTimelineAgent.cpp
trunk/Source/WebCore/platform/cf/RunLoopObserver.h
trunk/Source/WebCore/platform/graphics/DisplayRefreshMonitor.cpp
trunk/Source/WebCore/platform/graphics/ca/LayerFlushScheduler.h
trunk/Source/WebCore/platform/graphics/ca/cocoa/LayerFlushSchedulerMac.cpp
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/WebPageProxy.cpp
trunk/Tools/ChangeLog
trunk/Tools/Tracing/SystemTracePoints.plist
Diff
Modified: trunk/Source/WTF/ChangeLog (225508 => 225509)
--- trunk/Source/WTF/ChangeLog 2017-12-05 01:11:52 UTC (rev 225508)
+++ trunk/Source/WTF/ChangeLog 2017-12-05 01:13:27 UTC (rev 225509)
@@ -1,3 +1,14 @@
+2017-12-04 Simon Fraser
+
+Minor DisplayRefreshMonitor-related cleanup
+https://bugs.webkit.org/show_bug.cgi?id=179802
+
+Reviewed by Sam Weinig.
+
+New trace point for when the CVDisplayLink fires on its own thread.
+
+* wtf/SystemTracing.h:
+
2017-12-04 JF Bastien
Update std::expected to match libc++ coding style
Modified: trunk/Source/WTF/wtf/SystemTracing.h (225508 => 225509)
--- trunk/Source/WTF/wtf/SystemTracing.h 2017-12-05 01:11:52 UTC (rev 225508)
+++ trunk/Source/WTF/wtf/SystemTracing.h 2017-12-05 01:13:27 UTC (rev 225509)
@@ -73,6 +73,7 @@
UpdateTouchRegionsEnd,
DisplayListRecordStart,
DisplayListRecordEnd,
+DisplayRefreshDispatchingToMainThread,
WebKitRange = 1,
WebHTMLViewPaintStart,
Modified: trunk/Source/WebCore/ChangeLog (225508 => 225509)
--- trunk/Source/WebCore/ChangeLog 2017-12-05 01:11:52 UTC (rev 225508)
+++ trunk/Source/WebCore/ChangeLog 2017-12-05 01:13:27 UTC (rev 225509)
@@ -1,3 +1,23 @@
+2017-12-04 Simon Fraser
+
+Minor DisplayRefreshMonitor-related cleanup
+https://bugs.webkit.org/show_bug.cgi?id=179802
+
+Reviewed by Sam Weinig.
+
+New trace point for when the CVDisplayLink fires on its own thread.
+
+Some #pragma once, and put all the WellKnownRunLoopOrders in one place.
+
+* inspector/agents/InspectorTimelineAgent.cpp:
+(WebCore::InspectorTimelineAgent::internalStart):
+* platform/cf/RunLoopObserver.h:
+* platform/graphics/DisplayRefreshMonitor.cpp:
+(WebCore::DisplayRefreshMonitor::displayDidRefresh):
+* platform/graphics/ca/LayerFlushScheduler.h:
+* platform/graphics/ca/cocoa/LayerFlushSchedulerMac.cpp:
+(WebCore::LayerFlushScheduler::LayerFlushScheduler):
+
2017-12-04 David Quesada
Add a class for parsing application manifests
Modified: trunk/Source/WebCore/inspector/agents/InspectorTimelineAgent.cpp (225508 => 225509)
--- trunk/Source/WebCore/inspector/agents/InspectorTimelineAgent.cpp 2017-12-05 01:11:52 UTC (rev 225508)
+++ trunk/Source/WebCore/inspector/agents/InspectorTimelineAgent.cpp 2017-12-05 01:13:27 UTC (rev 225509)
@@ -67,8 +67,6 @@
using namespace Inspector;
#if PLATFORM(COCOA)
-static const CFIndex frameStopRunLoopOrder = (CFIndex)RunLoopObserver::WellKnownRunLoopOrders::CoreAnimationCommit + 1;
-
static CFRunLoopRef currentRunLoop()
{
#if PLATFORM(IOS)
@@ -174,7 +172,7 @@
// FIXME: Abstract away platform-specific code once https://bugs.webkit.org/show_bug.cgi?id=142748 is fixed.
#if PLATFORM(COCOA)
-m_frameStartObserver = std::make_unique(0, [this]() {
+m_frameStartObserver = std::make_unique(static_cast(RunLoopObserver::WellKnownRunLoopOrders::InspectorFrameBegin), [this]() {
if (!m_enabled || m_environment.scriptDebugServer().isPaused())
return;
@@ -183,7 +181,7 @@
m_runLoopNestingLevel++;
});
-m_frameStopObserver = s
[webkit-changes] [225508] trunk/Source/WebKit
Title: [225508] trunk/Source/WebKit Revision 225508 Author bfulg...@apple.com Date 2017-12-04 17:11:52 -0800 (Mon, 04 Dec 2017) Log Message Don't force creation of process pool when enabling resource load statistics https://bugs.webkit.org/show_bug.cgi?id=180374 Reviewed by Alex Christensen. Pass appropriate arguments to the 'processPool' accessor so that we only get the set of existing process pools. We don't want to pay the cost of launching new process pools just to set this flag. The flag will be properly set when the pools are created as soon as a WebView is instantiated. * UIProcess/WebsiteData/WebsiteDataStore.cpp: (WebKit::WebsiteDataStore::setResourceLoadStatisticsEnabled): Modified Paths trunk/Source/WebKit/ChangeLog trunk/Source/WebKit/UIProcess/WebsiteData/WebsiteDataStore.cpp Diff Modified: trunk/Source/WebKit/ChangeLog (225507 => 225508) --- trunk/Source/WebKit/ChangeLog 2017-12-05 00:57:33 UTC (rev 225507) +++ trunk/Source/WebKit/ChangeLog 2017-12-05 01:11:52 UTC (rev 225508) @@ -1,3 +1,19 @@ +2017-12-04 Brent Fulgham + +Don't force creation of process pool when enabling resource load statistics +https://bugs.webkit.org/show_bug.cgi?id=180374 + + +Reviewed by Alex Christensen. + +Pass appropriate arguments to the 'processPool' accessor so that we only get the +set of existing process pools. We don't want to pay the cost of launching new +process pools just to set this flag. The flag will be properly set when the pools +are created as soon as a WebView is instantiated. + +* UIProcess/WebsiteData/WebsiteDataStore.cpp: +(WebKit::WebsiteDataStore::setResourceLoadStatisticsEnabled): + 2017-12-04 David Quesada Add a class for parsing application manifests Modified: trunk/Source/WebKit/UIProcess/WebsiteData/WebsiteDataStore.cpp (225507 => 225508) --- trunk/Source/WebKit/UIProcess/WebsiteData/WebsiteDataStore.cpp 2017-12-05 00:57:33 UTC (rev 225507) +++ trunk/Source/WebKit/UIProcess/WebsiteData/WebsiteDataStore.cpp 2017-12-05 01:11:52 UTC (rev 225508) @@ -1326,7 +1326,9 @@ } m_resourceLoadStatistics = nullptr; -for (auto& processPool : processPools()) + +auto existingProcessPools = processPools(std::numeric_limits::max(), false); +for (auto& processPool : existingProcessPools) processPool->setResourceLoadStatisticsEnabled(false); } ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225507] trunk
Title: [225507] trunk Revision 225507 Author commit-qu...@webkit.org Date 2017-12-04 16:57:33 -0800 (Mon, 04 Dec 2017) Log Message Add a class for parsing application manifests https://bugs.webkit.org/show_bug.cgi?id=177973 rdar://problem/34747949 Patch by David Quesada on 2017-12-04 Reviewed by Geoffrey Garen. Source/_javascript_Core: * Configurations/FeatureDefines.xcconfig: Add ENABLE_APPLICATION_MANIFEST feature flag. Source/WebCore: Add a new struct ApplicationManifest, along with ApplicationManifestParser to initialize ApplicationManifests from JSON data, according to the App Manifest spec. Just a basic subset of the manifests's properties are implemented to start with. ApplicationManifestParser is tested by new unit tests in TestWebKitAPI. * Configurations/FeatureDefines.xcconfig: Add ENABLE_APPLICATION_MANIFEST feature flag. * Modules/applicationmanifest/ApplicationManifest.h: Added. * Modules/applicationmanifest/ApplicationManifestParser.cpp: Added. (WebCore::ApplicationManifestParser::parse): The main entry point for eventual clients (and currently the unit tests) to invoke ApplicationManifestParser. There is a variant that accepts a ScriptExecutionContext that uses the context exclusively for logging console warnings while parsing the manifest, and not for actually executing any scripts. (WebCore::ApplicationManifestParser::ApplicationManifestParser): (WebCore::ApplicationManifestParser::parseManifest): (WebCore::ApplicationManifestParser::logManifestPropertyNotAString): (WebCore::ApplicationManifestParser::logManifestPropertyInvalidURL): (WebCore::ApplicationManifestParser::logDeveloperWarning): (WebCore::ApplicationManifestParser::parseStartURL): (WebCore::ApplicationManifestParser::parseName): (WebCore::ApplicationManifestParser::parseDescription): (WebCore::ApplicationManifestParser::parseShortName): (WebCore::isInScope): (WebCore::ApplicationManifestParser::parseScope): (WebCore::ApplicationManifestParser::parseGenericString): * Modules/applicationmanifest/ApplicationManifestParser.h: Added. * Sources.txt: * WebCore.xcodeproj/project.pbxproj: * features.json: Change the Web App Manifest feature status to "In Development" Source/WebCore/PAL: * Configurations/FeatureDefines.xcconfig: Add ENABLE_APPLICATION_MANIFEST feature flag. Source/WebKit: * Configurations/FeatureDefines.xcconfig: Add ENABLE_APPLICATION_MANIFEST feature flag. Source/WebKitLegacy/mac: * Configurations/FeatureDefines.xcconfig: Add ENABLE_APPLICATION_MANIFEST feature flag. Tools: Add basic unit tests for ApplicationManifestParser. For each of the implemented top-level properties, attempt to parse manifests with varying types of valid and invalid data to ensure the resulting ApplicationManifest is configured with the appropriate values per the spec. * TestWebKitAPI/Configurations/FeatureDefines.xcconfig: Add ENABLE_APPLICATION_MANIFEST feature flag. * TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj: * TestWebKitAPI/Tests/WebCore/ApplicationManifestParser.cpp: Added. (ApplicationManifestParserTest::SetUp): (ApplicationManifestParserTest::parseString): (ApplicationManifestParserTest::parseTopLevelProperty): (ApplicationManifestParserTest::testStartURL): (ApplicationManifestParserTest::testName): (ApplicationManifestParserTest::testDescription): (ApplicationManifestParserTest::testShortName): (ApplicationManifestParserTest::testScope): (assertManifestHasDefaultValues): (TEST_F): Modified Paths trunk/Source/_javascript_Core/ChangeLog trunk/Source/_javascript_Core/Configurations/FeatureDefines.xcconfig trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/Configurations/FeatureDefines.xcconfig trunk/Source/WebCore/PAL/ChangeLog trunk/Source/WebCore/PAL/Configurations/FeatureDefines.xcconfig trunk/Source/WebCore/Sources.txt trunk/Source/WebCore/WebCore.xcodeproj/project.pbxproj trunk/Source/WebCore/features.json trunk/Source/WebKit/ChangeLog trunk/Source/WebKit/Configurations/FeatureDefines.xcconfig trunk/Source/WebKitLegacy/mac/ChangeLog trunk/Source/WebKitLegacy/mac/Configurations/FeatureDefines.xcconfig trunk/Tools/ChangeLog trunk/Tools/TestWebKitAPI/Configurations/FeatureDefines.xcconfig trunk/Tools/TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj Added Paths trunk/Source/WebCore/Modules/applicationmanifest/ trunk/Source/WebCore/Modules/applicationmanifest/ApplicationManifest.h trunk/Source/WebCore/Modules/applicationmanifest/ApplicationManifestParser.cpp trunk/Source/WebCore/Modules/applicationmanifest/ApplicationManifestParser.h trunk/Tools/TestWebKitAPI/Tests/WebCore/ApplicationManifestParser.cpp Diff Modified: trunk/Source/_javascript_Core/ChangeLog (225506 => 225507) --- trunk/Source/_javascript_Core/ChangeLog 2017-12-05 00:51:01 UTC (rev 225506) +++ trunk/Source/_javascript_Core/ChangeLog 2017-12-05 00:57:33 UTC (rev 225507) @@ -1,3 +1,13 @@ +2017-12-04 David Quesada + +Add a class for parsing application manifests +https://bugs.webkit.org/show_bug.cgi?id=1
[webkit-changes] [225506] trunk/Source/WebCore
Title: [225506] trunk/Source/WebCore
Revision 225506
Author za...@apple.com
Date 2017-12-04 16:51:01 -0800 (Mon, 04 Dec 2017)
Log Message
RenderMultiColumnFlow::fragmentedFlowDescendantInserted should not destroy incoming newDescendant
https://bugs.webkit.org/show_bug.cgi?id=180181
Reviewed by Antti Koivisto.
This is in preparation for having all multicolumn related tree mutation in RenderTreeUpdaterMultiColumn.
Covered by fast/multicol/column-span-range-crash.html
* rendering/RenderMultiColumnFlow.cpp:
(WebCore::RenderMultiColumnFlow::fragmentedFlowDescendantInserted):
* rendering/RenderMultiColumnFlow.h:
* style/RenderTreeUpdaterMultiColumn.cpp:
(WebCore::RenderTreeUpdater::MultiColumn::destroyFragmentedFlow):
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/rendering/RenderMultiColumnFlow.cpp
trunk/Source/WebCore/rendering/RenderMultiColumnFlow.h
trunk/Source/WebCore/style/RenderTreeUpdaterMultiColumn.cpp
Diff
Modified: trunk/Source/WebCore/ChangeLog (225505 => 225506)
--- trunk/Source/WebCore/ChangeLog 2017-12-05 00:13:09 UTC (rev 225505)
+++ trunk/Source/WebCore/ChangeLog 2017-12-05 00:51:01 UTC (rev 225506)
@@ -1,3 +1,20 @@
+2017-12-04 Zalan Bujtas
+
+RenderMultiColumnFlow::fragmentedFlowDescendantInserted should not destroy incoming newDescendant
+https://bugs.webkit.org/show_bug.cgi?id=180181
+
+Reviewed by Antti Koivisto.
+
+This is in preparation for having all multicolumn related tree mutation in RenderTreeUpdaterMultiColumn.
+
+Covered by fast/multicol/column-span-range-crash.html
+
+* rendering/RenderMultiColumnFlow.cpp:
+(WebCore::RenderMultiColumnFlow::fragmentedFlowDescendantInserted):
+* rendering/RenderMultiColumnFlow.h:
+* style/RenderTreeUpdaterMultiColumn.cpp:
+(WebCore::RenderTreeUpdater::MultiColumn::destroyFragmentedFlow):
+
2017-12-04 JF Bastien
Update std::expected to match libc++ coding style
Modified: trunk/Source/WebCore/rendering/RenderMultiColumnFlow.cpp (225505 => 225506)
--- trunk/Source/WebCore/rendering/RenderMultiColumnFlow.cpp 2017-12-05 00:13:09 UTC (rev 225505)
+++ trunk/Source/WebCore/rendering/RenderMultiColumnFlow.cpp 2017-12-05 00:51:01 UTC (rev 225506)
@@ -350,46 +350,26 @@
if (gShiftingSpanner || newDescendant.isInFlowRenderFragmentedFlow())
return;
-Vector> spannersToDelete;
-
-RenderObject* subtreeRoot = &newDescendant;
-for (auto* descendant = &newDescendant; descendant; descendant = (descendant ? descendant->nextInPreOrder(subtreeRoot) : nullptr)) {
+auto* subtreeRoot = &newDescendant;
+auto* descendant = subtreeRoot;
+while (descendant) {
+// Skip nested multicolumn flows.
+if (is(*descendant)) {
+descendant = descendant->nextSibling();
+continue;
+}
if (is(*descendant)) {
// A spanner's placeholder has been inserted. The actual spanner renderer is moved from
// where it would otherwise occur (if it weren't a spanner) to becoming a sibling of the
// column sets.
RenderMultiColumnSpannerPlaceholder& placeholder = downcast(*descendant);
-if (placeholder.fragmentedFlow() != this) {
-// This isn't our spanner! It shifted here from an ancestor multicolumn block. It's going to end up
-// becoming our spanner instead, but for it to do that we first have to nuke the original spanner,
-// and get the spanner content back into this flow thread.
-RenderBox* spanner = placeholder.spanner();
-
-// Insert after the placeholder, but don't let a notification happen.
-gShiftingSpanner = true;
-RenderBlockFlow& ancestorBlock = downcast(*spanner->parent());
-ancestorBlock.moveChildTo(placeholder.parentBox(), spanner, placeholder.nextSibling(), RenderBoxModelObject::NormalizeAfterInsertion::Yes);
-gShiftingSpanner = false;
-
-// We have to nuke the placeholder, since the ancestor already lost the mapping to it when
-// we shifted the placeholder down into this flow thread.
-placeholder.fragmentedFlow()->spannerMap().remove(spanner);
-
-spannersToDelete.append(placeholder.parent()->takeChild(placeholder));
-
-if (subtreeRoot == descendant)
-subtreeRoot = spanner;
-// Now we process the spanner.
-descendant = processPossibleSpannerDescendant(subtreeRoot, *spanner);
-continue;
-}
-
ASSERT(!spannerMap().get(placeholder.spanner()));
-spannerMap().add(placeholder.spanner(), makeWeakPtr(placeholder));
+spannerMap().add(placeholder.spanner(), makeWeakPtr(downcast(descendant)));
ASSERT(!placeholder.
[webkit-changes] [225505] trunk/Source/WebKit
Title: [225505] trunk/Source/WebKit
Revision 225505
Author wenson_hs...@apple.com
Date 2017-12-04 16:13:09 -0800 (Mon, 04 Dec 2017)
Log Message
Rename a static helper in TouchBarMenuItemData.cpp to match style guidelines.
https://bugs.webkit.org/show_bug.cgi?id=180305
Reviewed by Andy Estes.
Style guidelines state that we should use bare words for getters, so getItemType should just be
itemType. Since this collides with the itemType member variable, this patch also renames that
member variable to the more concise "type".
No change in behavior.
* Shared/TouchBarMenuItemData.cpp:
(WebKit::itemType):
(WebKit::TouchBarMenuItemData::TouchBarMenuItemData):
(WebKit::TouchBarMenuItemData::encode const):
(WebKit::TouchBarMenuItemData::decode):
(WebKit::getItemType): Deleted.
* Shared/TouchBarMenuItemData.h:
(WebKit::operator==):
Modified Paths
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/Shared/TouchBarMenuItemData.cpp
trunk/Source/WebKit/Shared/TouchBarMenuItemData.h
Diff
Modified: trunk/Source/WebKit/ChangeLog (225504 => 225505)
--- trunk/Source/WebKit/ChangeLog 2017-12-05 00:11:05 UTC (rev 225504)
+++ trunk/Source/WebKit/ChangeLog 2017-12-05 00:13:09 UTC (rev 225505)
@@ -1,3 +1,25 @@
+2017-12-04 Wenson Hsieh
+
+Rename a static helper in TouchBarMenuItemData.cpp to match style guidelines.
+https://bugs.webkit.org/show_bug.cgi?id=180305
+
+Reviewed by Andy Estes.
+
+Style guidelines state that we should use bare words for getters, so getItemType should just be
+itemType. Since this collides with the itemType member variable, this patch also renames that
+member variable to the more concise "type".
+
+No change in behavior.
+
+* Shared/TouchBarMenuItemData.cpp:
+(WebKit::itemType):
+(WebKit::TouchBarMenuItemData::TouchBarMenuItemData):
+(WebKit::TouchBarMenuItemData::encode const):
+(WebKit::TouchBarMenuItemData::decode):
+(WebKit::getItemType): Deleted.
+* Shared/TouchBarMenuItemData.h:
+(WebKit::operator==):
+
2017-12-04 Brian Burg
Web Automation: add flag to preserve legacy page screenshot behavior
Modified: trunk/Source/WebKit/Shared/TouchBarMenuItemData.cpp (225504 => 225505)
--- trunk/Source/WebKit/Shared/TouchBarMenuItemData.cpp 2017-12-05 00:11:05 UTC (rev 225504)
+++ trunk/Source/WebKit/Shared/TouchBarMenuItemData.cpp 2017-12-05 00:13:09 UTC (rev 225505)
@@ -33,7 +33,7 @@
namespace WebKit {
-static ItemType getItemType(const String&)
+static ItemType itemType(const String&)
{
return ItemType::Button;
}
@@ -40,7 +40,7 @@
TouchBarMenuItemData::TouchBarMenuItemData(const WebCore::HTMLMenuItemElement& element)
{
-itemType = getItemType(element.attributeWithoutSynchronization(WebCore::HTMLNames::typeAttr));
+type = itemType(element.attributeWithoutSynchronization(WebCore::HTMLNames::typeAttr));
identifier = element.attributeWithoutSynchronization(WebCore::HTMLNames::idAttr);
priority = element.attributeWithoutSynchronization(WebCore::HTMLNames::valueAttr).toFloat();
}
@@ -47,7 +47,7 @@
void TouchBarMenuItemData::encode(IPC::Encoder& encoder) const
{
-encoder.encodeEnum(itemType);
+encoder.encodeEnum(type);
encoder << identifier;
encoder << priority;
@@ -56,7 +56,7 @@
std::optional TouchBarMenuItemData::decode(IPC::Decoder& decoder)
{
TouchBarMenuItemData result;
-if (!decoder.decodeEnum(result.itemType))
+if (!decoder.decodeEnum(result.type))
return std::nullopt;
if (!decoder.decode(result.identifier))
Modified: trunk/Source/WebKit/Shared/TouchBarMenuItemData.h (225504 => 225505)
--- trunk/Source/WebKit/Shared/TouchBarMenuItemData.h 2017-12-05 00:11:05 UTC (rev 225504)
+++ trunk/Source/WebKit/Shared/TouchBarMenuItemData.h 2017-12-05 00:13:09 UTC (rev 225505)
@@ -51,7 +51,7 @@
void encode(IPC::Encoder&) const;
static std::optional decode(IPC::Decoder&);
-ItemType itemType { ItemType::Button };
+ItemType type { ItemType::Button };
String identifier;
float priority { 0.0 };
bool validTouchBarDisplay { true };
@@ -80,9 +80,9 @@
inline bool operator==(const TouchBarMenuItemData& lhs, const TouchBarMenuItemData& rhs)
{
-return lhs.itemType == rhs.itemType
-&& lhs.identifier == rhs.identifier
-&& lhs.priority == rhs.priority;
+return lhs.type == rhs.type
+&& lhs.identifier == rhs.identifier
+&& lhs.priority == rhs.priority;
}
inline bool operator!=(const TouchBarMenuItemData& lhs, const TouchBarMenuItemData& rhs)
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225504] tags/Safari-605.1.15.2
Title: [225504] tags/Safari-605.1.15.2
Revision 225504
Author jmarc...@apple.com
Date 2017-12-04 16:11:05 -0800 (Mon, 04 Dec 2017)
Log Message
Cherry-pick r225383. rdar://problem/35709469
Modified Paths
tags/Safari-605.1.15.2/Source/WebKit/ChangeLog
tags/Safari-605.1.15.2/Source/WebKit/NetworkProcess/mac/RemoteNetworkingContext.mm
tags/Safari-605.1.15.2/Tools/ChangeLog
tags/Safari-605.1.15.2/Tools/TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj
Added Paths
tags/Safari-605.1.15.2/Tools/TestWebKitAPI/Tests/WebKitCocoa/CookiePrivateBrowsing.mm
Diff
Modified: tags/Safari-605.1.15.2/Source/WebKit/ChangeLog (225503 => 225504)
--- tags/Safari-605.1.15.2/Source/WebKit/ChangeLog 2017-12-05 00:09:10 UTC (rev 225503)
+++ tags/Safari-605.1.15.2/Source/WebKit/ChangeLog 2017-12-05 00:11:05 UTC (rev 225504)
@@ -1,3 +1,22 @@
+2017-12-04 Jason Marcell
+
+Cherry-pick r225383. rdar://problem/35709469
+
+2017-11-30 Alex Christensen
+
+REGRESSION (r224791): cookies are shared between ephemeral sessions in the same process pool
+https://bugs.webkit.org/show_bug.cgi?id=180235
+
+Reviewed by Joseph Pecoraro.
+
+In r224791 I assumed that uiProcessCookieStorageIdentifier would be empty in the ephemeral session initialization message.
+This is not the case. We did not catch this because almost all of our ephemeral session testing uses the legacyPrivateSessionID.
+I add checks for if the SessionID is ephemeral like we already have in WebFrameNetworkingContext::ensureWebsiteDataStoreSession.
+I also add an API test that makes sure cookies are not shared between ephemeral sessions the way they are made through the API.
+
+* NetworkProcess/mac/RemoteNetworkingContext.mm:
+(WebKit::RemoteNetworkingContext::ensureWebsiteDataStoreSession):
+
2017-11-29 Zan Dobersek
[CoordGraphics] Rename CoordinatedBuffer to Nicosia::Buffer
Modified: tags/Safari-605.1.15.2/Source/WebKit/NetworkProcess/mac/RemoteNetworkingContext.mm (225503 => 225504)
--- tags/Safari-605.1.15.2/Source/WebKit/NetworkProcess/mac/RemoteNetworkingContext.mm 2017-12-05 00:09:10 UTC (rev 225503)
+++ tags/Safari-605.1.15.2/Source/WebKit/NetworkProcess/mac/RemoteNetworkingContext.mm 2017-12-05 00:11:05 UTC (rev 225504)
@@ -97,7 +97,7 @@
SandboxExtension::consumePermanently(parameters.cookieStoragePathExtensionHandle);
RetainPtr uiProcessCookieStorage;
-if (!parameters.uiProcessCookieStorageIdentifier.isEmpty())
+if (!sessionID.isEphemeral() && !parameters.uiProcessCookieStorageIdentifier.isEmpty())
uiProcessCookieStorage = cookieStorageFromIdentifyingData(parameters.uiProcessCookieStorageIdentifier);
NetworkStorageSession::ensureSession(sessionID, base + '.' + String::number(sessionID.sessionID()), WTFMove(uiProcessCookieStorage));
@@ -106,7 +106,7 @@
for (const auto& cookie : parameters.pendingCookies)
session->setCookie(cookie);
-if (!parameters.cacheStorageDirectory.isNull()) {
+if (!sessionID.isEphemeral() && !parameters.cacheStorageDirectory.isNull()) {
SandboxExtension::consumePermanently(parameters.cacheStorageDirectoryExtensionHandle);
session->setCacheStorageDirectory(WTFMove(parameters.cacheStorageDirectory));
session->setCacheStoragePerOriginQuota(parameters.cacheStoragePerOriginQuota);
Modified: tags/Safari-605.1.15.2/Tools/ChangeLog (225503 => 225504)
--- tags/Safari-605.1.15.2/Tools/ChangeLog 2017-12-05 00:09:10 UTC (rev 225503)
+++ tags/Safari-605.1.15.2/Tools/ChangeLog 2017-12-05 00:11:05 UTC (rev 225504)
@@ -1,3 +1,19 @@
+2017-12-04 Jason Marcell
+
+Cherry-pick r225383. rdar://problem/35709469
+
+2017-11-30 Alex Christensen
+
+REGRESSION (r224791): cookies are shared between ephemeral sessions in the same process pool
+https://bugs.webkit.org/show_bug.cgi?id=180235
+
+Reviewed by Joseph Pecoraro.
+
+* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
+* TestWebKitAPI/Tests/WebKitCocoa/CookiePrivateBrowsing.mm: Added.
+(-[CookiePrivateBrowsingDelegate webView:runJavaScriptAlertPanelWithMessage:initiatedByFrame:completionHandler:]):
+(TEST):
+
2017-11-28 Carlos Garcia Campos
WebDriver: add an option to dump test results to a json file
Modified: tags/Safari-605.1.15.2/Tools/TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj (225503 => 225504)
--- tags/Safari-605.1.15.2/Tools/TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj 2017-12-05 00:09:10 UTC (rev 225503)
+++ tags/Safari-605.1.15.2/Tools/TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj 2017-12-05 00:11:05 UTC (rev 225504)
@@ -228,6 +228,7 @@
5C0BF8931DD599BD00B00328 /* IsNavigationActionTrusted.mm in Sources */ = {isa = PBXBuildFile; fileRef = 57F10D921C7E7B3800ECDF30 /* IsNavigationActionTrusted.mm */; };
5C0BF
[webkit-changes] [225503] tags/Safari-605.1.15.2/Source
Title: [225503] tags/Safari-605.1.15.2/Source Revision 225503 Author jmarc...@apple.com Date 2017-12-04 16:09:10 -0800 (Mon, 04 Dec 2017) Log Message Versioning. Modified Paths tags/Safari-605.1.15.2/Source/_javascript_Core/Configurations/Version.xcconfig tags/Safari-605.1.15.2/Source/ThirdParty/libwebrtc/Configurations/Version.xcconfig tags/Safari-605.1.15.2/Source/WebCore/Configurations/Version.xcconfig tags/Safari-605.1.15.2/Source/WebCore/PAL/Configurations/Version.xcconfig tags/Safari-605.1.15.2/Source/WebInspectorUI/Configurations/Version.xcconfig tags/Safari-605.1.15.2/Source/WebKit/Configurations/Version.xcconfig tags/Safari-605.1.15.2/Source/WebKitLegacy/mac/Configurations/Version.xcconfig Diff Modified: tags/Safari-605.1.15.2/Source/_javascript_Core/Configurations/Version.xcconfig (225502 => 225503) --- tags/Safari-605.1.15.2/Source/_javascript_Core/Configurations/Version.xcconfig 2017-12-05 00:07:29 UTC (rev 225502) +++ tags/Safari-605.1.15.2/Source/_javascript_Core/Configurations/Version.xcconfig 2017-12-05 00:09:10 UTC (rev 225503) @@ -24,7 +24,7 @@ MAJOR_VERSION = 605; MINOR_VERSION = 1; TINY_VERSION = 15; -MICRO_VERSION = 1; +MICRO_VERSION = 2; NANO_VERSION = 0; FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION).$(MICRO_VERSION); Modified: tags/Safari-605.1.15.2/Source/ThirdParty/libwebrtc/Configurations/Version.xcconfig (225502 => 225503) --- tags/Safari-605.1.15.2/Source/ThirdParty/libwebrtc/Configurations/Version.xcconfig 2017-12-05 00:07:29 UTC (rev 225502) +++ tags/Safari-605.1.15.2/Source/ThirdParty/libwebrtc/Configurations/Version.xcconfig 2017-12-05 00:09:10 UTC (rev 225503) @@ -24,7 +24,7 @@ MAJOR_VERSION = 605; MINOR_VERSION = 1; TINY_VERSION = 15; -MICRO_VERSION = 1; +MICRO_VERSION = 2; NANO_VERSION = 0; FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION).$(MICRO_VERSION); Modified: tags/Safari-605.1.15.2/Source/WebCore/Configurations/Version.xcconfig (225502 => 225503) --- tags/Safari-605.1.15.2/Source/WebCore/Configurations/Version.xcconfig 2017-12-05 00:07:29 UTC (rev 225502) +++ tags/Safari-605.1.15.2/Source/WebCore/Configurations/Version.xcconfig 2017-12-05 00:09:10 UTC (rev 225503) @@ -24,7 +24,7 @@ MAJOR_VERSION = 605; MINOR_VERSION = 1; TINY_VERSION = 15; -MICRO_VERSION = 1; +MICRO_VERSION = 2; NANO_VERSION = 0; FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION).$(MICRO_VERSION); Modified: tags/Safari-605.1.15.2/Source/WebCore/PAL/Configurations/Version.xcconfig (225502 => 225503) --- tags/Safari-605.1.15.2/Source/WebCore/PAL/Configurations/Version.xcconfig 2017-12-05 00:07:29 UTC (rev 225502) +++ tags/Safari-605.1.15.2/Source/WebCore/PAL/Configurations/Version.xcconfig 2017-12-05 00:09:10 UTC (rev 225503) @@ -24,7 +24,7 @@ MAJOR_VERSION = 605; MINOR_VERSION = 1; TINY_VERSION = 15; -MICRO_VERSION = 1; +MICRO_VERSION = 2; NANO_VERSION = 0; FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION).$(MICRO_VERSION); Modified: tags/Safari-605.1.15.2/Source/WebInspectorUI/Configurations/Version.xcconfig (225502 => 225503) --- tags/Safari-605.1.15.2/Source/WebInspectorUI/Configurations/Version.xcconfig 2017-12-05 00:07:29 UTC (rev 225502) +++ tags/Safari-605.1.15.2/Source/WebInspectorUI/Configurations/Version.xcconfig 2017-12-05 00:09:10 UTC (rev 225503) @@ -1,7 +1,7 @@ MAJOR_VERSION = 605; MINOR_VERSION = 1; TINY_VERSION = 15; -MICRO_VERSION = 1; +MICRO_VERSION = 2; NANO_VERSION = 0; FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION).$(MICRO_VERSION); Modified: tags/Safari-605.1.15.2/Source/WebKit/Configurations/Version.xcconfig (225502 => 225503) --- tags/Safari-605.1.15.2/Source/WebKit/Configurations/Version.xcconfig 2017-12-05 00:07:29 UTC (rev 225502) +++ tags/Safari-605.1.15.2/Source/WebKit/Configurations/Version.xcconfig 2017-12-05 00:09:10 UTC (rev 225503) @@ -24,7 +24,7 @@ MAJOR_VERSION = 605; MINOR_VERSION = 1; TINY_VERSION = 15; -MICRO_VERSION = 1; +MICRO_VERSION = 2; NANO_VERSION = 0; FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION).$(MICRO_VERSION); Modified: tags/Safari-605.1.15.2/Source/WebKitLegacy/mac/Configurations/Version.xcconfig (225502 => 225503) --- tags/Safari-605.1.15.2/Source/WebKitLegacy/mac/Configurations/Version.xcconfig 2017-12-05 00:07:29 UTC (rev 225502) +++ tags/Safari-605.1.15.2/Source/WebKitLegacy/mac/Configurations/Version.xcconfig 2017-12-05 00:09:10 UTC (rev 225503) @@ -24,7 +24,7 @@ MAJOR_VERSION = 605; MINOR_VERSION = 1; TINY_VERSION = 15; -MICRO_VERSION = 1; +MICRO_VERSION = 2; NANO_VERSION = 0; FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION).$(MICRO_VERSION); ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225502] tags/Safari-605.1.15.2/
Title: [225502] tags/Safari-605.1.15.2/ Revision 225502 Author jmarc...@apple.com Date 2017-12-04 16:07:29 -0800 (Mon, 04 Dec 2017) Log Message New tag. Added Paths tags/Safari-605.1.15.2/ Diff ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225501] trunk/Source
Title: [225501] trunk/Source
Revision 225501
Author bb...@apple.com
Date 2017-12-04 16:06:25 -0800 (Mon, 04 Dec 2017)
Log Message
Web Automation: add flag to preserve legacy page screenshot behavior
https://bugs.webkit.org/show_bug.cgi?id=180313
Reviewed by Joseph Pecoraro.
Source/WebDriver:
Set the clipToViewport flag to true when sending Automation.takeScreenshot.
This preserves the current behavior for this driver implementation.
* Session.cpp:
(WebDriver::Session::takeScreenshot):
Source/WebKit:
For compatibility with JSON Wire Protocol implemented by Safari,
we need to retain the ability to perform whole page contents
snapshots using Automation.takeScreenshot. Add an extra flag,
clipToViewport, which can be used by W3C-conforming drivers.
* UIProcess/Automation/Automation.json: Add new flag.
* UIProcess/Automation/WebAutomationSession.h:
* UIProcess/Automation/WebAutomationSession.cpp:
(WebKit::WebAutomationSession::takeScreenshot):
* WebProcess/Automation/WebAutomationSessionProxy.h:
* WebProcess/Automation/WebAutomationSessionProxy.messages.in:
* WebProcess/Automation/WebAutomationSessionProxy.cpp:
(WebKit::snapshotRectForScreenshot):
(WebKit::WebAutomationSessionProxy::takeScreenshot):
If the flag is false, take a screenshot of the whole page contents.
Modified Paths
trunk/Source/WebDriver/ChangeLog
trunk/Source/WebDriver/Session.cpp
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/Automation/Automation.json
trunk/Source/WebKit/UIProcess/Automation/WebAutomationSession.cpp
trunk/Source/WebKit/UIProcess/Automation/WebAutomationSession.h
trunk/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.cpp
trunk/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.h
trunk/Source/WebKit/WebProcess/Automation/WebAutomationSessionProxy.messages.in
Diff
Modified: trunk/Source/WebDriver/ChangeLog (225500 => 225501)
--- trunk/Source/WebDriver/ChangeLog 2017-12-05 00:03:38 UTC (rev 225500)
+++ trunk/Source/WebDriver/ChangeLog 2017-12-05 00:06:25 UTC (rev 225501)
@@ -1,3 +1,17 @@
+2017-12-04 Brian Burg
+
+Web Automation: add flag to preserve legacy page screenshot behavior
+https://bugs.webkit.org/show_bug.cgi?id=180313
+
+
+Reviewed by Joseph Pecoraro.
+
+Set the clipToViewport flag to true when sending Automation.takeScreenshot.
+This preserves the current behavior for this driver implementation.
+
+* Session.cpp:
+(WebDriver::Session::takeScreenshot):
+
2017-12-04 Carlos Garcia Campos
WebDriver: implement element property command
Modified: trunk/Source/WebDriver/Session.cpp (225500 => 225501)
--- trunk/Source/WebDriver/Session.cpp 2017-12-05 00:03:38 UTC (rev 225500)
+++ trunk/Source/WebDriver/Session.cpp 2017-12-05 00:06:25 UTC (rev 225501)
@@ -2123,6 +2123,8 @@
parameters->setString(ASCIILiteral("frameHandle"), m_currentBrowsingContext.value());
if (elementID)
parameters->setString(ASCIILiteral("nodeHandle"), elementID.value());
+else
+parameters->setBoolean(ASCIILiteral("clipToViewport"), true);
if (scrollIntoView.value_or(false))
parameters->setBoolean(ASCIILiteral("scrollIntoViewIfNeeded"), true);
m_host->sendCommandToBackend(ASCIILiteral("takeScreenshot"), WTFMove(parameters), [this, protectedThis = makeRef(*this), completionHandler = WTFMove(completionHandler)](SessionHost::CommandResponse&& response) mutable {
Modified: trunk/Source/WebKit/ChangeLog (225500 => 225501)
--- trunk/Source/WebKit/ChangeLog 2017-12-05 00:03:38 UTC (rev 225500)
+++ trunk/Source/WebKit/ChangeLog 2017-12-05 00:06:25 UTC (rev 225501)
@@ -1,3 +1,27 @@
+2017-12-04 Brian Burg
+
+Web Automation: add flag to preserve legacy page screenshot behavior
+https://bugs.webkit.org/show_bug.cgi?id=180313
+
+
+Reviewed by Joseph Pecoraro.
+
+For compatibility with JSON Wire Protocol implemented by Safari,
+we need to retain the ability to perform whole page contents
+snapshots using Automation.takeScreenshot. Add an extra flag,
+clipToViewport, which can be used by W3C-conforming drivers.
+
+* UIProcess/Automation/Automation.json: Add new flag.
+* UIProcess/Automation/WebAutomationSession.h:
+* UIProcess/Automation/WebAutomationSession.cpp:
+(WebKit::WebAutomationSession::takeScreenshot):
+* WebProcess/Automation/WebAutomationSessionProxy.h:
+* WebProcess/Automation/WebAutomationSessionProxy.messages.in:
+* WebProcess/Automation/WebAutomationSessionProxy.cpp:
+(WebKit::snapshotRectForScreenshot):
+(WebKit::WebAutomationSessionProxy::takeScreenshot):
+If the flag is false, take a screenshot of the whole page contents.
+
2017-12-04 JF Bastien
Update std::expected to match libc++ coding style
Modified: trunk/Source/WebKit/UIProcess/Automation/Automation.json (225
[webkit-changes] [225500] trunk/Tools
Title: [225500] trunk/Tools Revision 225500 Author aakash_j...@apple.com Date 2017-12-04 16:03:38 -0800 (Mon, 04 Dec 2017) Log Message Add docstring in EarlyWarningSystemTask to explain return values https://bugs.webkit.org/show_bug.cgi?id=180303 Reviewed by Daniel Bates. * Scripts/webkitpy/tool/bot/earlywarningsystemtask.py: (EarlyWarningSystemTask.run): Added docstring to explain return values. * Scripts/webkitpy/tool/commands/earlywarningsystem.py: (AbstractEarlyWarningSystem.review_patch): Added comment to explain the result of unlocking a patch. Modified Paths trunk/Tools/ChangeLog trunk/Tools/Scripts/webkitpy/tool/bot/earlywarningsystemtask.py trunk/Tools/Scripts/webkitpy/tool/commands/earlywarningsystem.py Diff Modified: trunk/Tools/ChangeLog (225499 => 225500) --- trunk/Tools/ChangeLog 2017-12-04 23:34:57 UTC (rev 225499) +++ trunk/Tools/ChangeLog 2017-12-05 00:03:38 UTC (rev 225500) @@ -1,3 +1,15 @@ +2017-12-04 Aakash Jain + +Add docstring in EarlyWarningSystemTask to explain return values +https://bugs.webkit.org/show_bug.cgi?id=180303 + +Reviewed by Daniel Bates. + +* Scripts/webkitpy/tool/bot/earlywarningsystemtask.py: +(EarlyWarningSystemTask.run): Added docstring to explain return values. +* Scripts/webkitpy/tool/commands/earlywarningsystem.py: +(AbstractEarlyWarningSystem.review_patch): Added comment to explain the result of unlocking a patch. + 2017-12-04 JF Bastien Update std::expected to match libc++ coding style Modified: trunk/Tools/Scripts/webkitpy/tool/bot/earlywarningsystemtask.py (225499 => 225500) --- trunk/Tools/Scripts/webkitpy/tool/bot/earlywarningsystemtask.py 2017-12-04 23:34:57 UTC (rev 225499) +++ trunk/Tools/Scripts/webkitpy/tool/bot/earlywarningsystemtask.py 2017-12-05 00:03:38 UTC (rev 225500) @@ -54,6 +54,12 @@ return True def run(self): +""" +Returns True if the patch passes EWS. +Raises an exception if the patch fails EWS. +Returns False if the patch status can not be ascertained. AbstractEarlyWarningSystem.review_patch() +will unlock the patch so that it would be retried. +""" if not self._clean(): return False if not self._update(): Modified: trunk/Tools/Scripts/webkitpy/tool/commands/earlywarningsystem.py (225499 => 225500) --- trunk/Tools/Scripts/webkitpy/tool/commands/earlywarningsystem.py 2017-12-04 23:34:57 UTC (rev 225499) +++ trunk/Tools/Scripts/webkitpy/tool/commands/earlywarningsystem.py 2017-12-05 00:03:38 UTC (rev 225500) @@ -102,6 +102,7 @@ succeeded = task.run() if not succeeded: # Caller unlocks when review_patch returns True, so we only need to unlock on transient failure. +# Unlocking the patch would result in patch being re-tried. self._unlock_patch(patch) return succeeded except PatchIsNotValid as error: ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225498] trunk/Tools
Title: [225498] trunk/Tools
Revision 225498
Author ross.kirsl...@sony.com
Date 2017-12-04 15:05:00 -0800 (Mon, 04 Dec 2017)
Log Message
download-latest-github-release.py should use existing release ANY time latest is not found
https://bugs.webkit.org/show_bug.cgi?id=180293
Reviewed by Per Arne Vollan.
* Scripts/download-latest-github-release.py:
(find_latest_release):
Handle all URLErrors, not just HTTPErrors. Stop special-casing 404s.
(main):
Rearrange logic so any failure to detect a latest release falls back to an existing release. Improve logging.
(Status):
(download_release):
(load_version_info): Renamed from existing_version info.
(has_latest_release): Deleted.
Cleanup.
Modified Paths
trunk/Tools/ChangeLog
trunk/Tools/Scripts/download-latest-github-release.py
Diff
Modified: trunk/Tools/ChangeLog (225497 => 225498)
--- trunk/Tools/ChangeLog 2017-12-04 22:52:26 UTC (rev 225497)
+++ trunk/Tools/ChangeLog 2017-12-04 23:05:00 UTC (rev 225498)
@@ -1,3 +1,23 @@
+2017-12-04 Ross Kirsling
+
+download-latest-github-release.py should use existing release ANY time latest is not found
+https://bugs.webkit.org/show_bug.cgi?id=180293
+
+Reviewed by Per Arne Vollan.
+
+* Scripts/download-latest-github-release.py:
+(find_latest_release):
+Handle all URLErrors, not just HTTPErrors. Stop special-casing 404s.
+
+(main):
+Rearrange logic so any failure to detect a latest release falls back to an existing release. Improve logging.
+
+(Status):
+(download_release):
+(load_version_info): Renamed from existing_version info.
+(has_latest_release): Deleted.
+Cleanup.
+
2017-12-04 Joseph Pecoraro
Exclude some more leak callstacks
Modified: trunk/Tools/Scripts/download-latest-github-release.py (225497 => 225498)
--- trunk/Tools/Scripts/download-latest-github-release.py 2017-12-04 22:52:26 UTC (rev 225497)
+++ trunk/Tools/Scripts/download-latest-github-release.py 2017-12-04 23:05:00 UTC (rev 225498)
@@ -46,7 +46,7 @@
DOWNLOADED = 0
UP_TO_DATE = 1
COULD_NOT_FIND = 2
-USE_EXISTING_VERSION = 3
+USING_EXISTING = 3
def parse_args(argv):
@@ -69,9 +69,8 @@
try:
response = urllib2.urlopen(request)
-except urllib2.HTTPError, error:
-if error.code != 404:
-print error.code, error.reason
+except urllib2.URLError as error:
+print error
return None, None
data = ""
@@ -82,7 +81,12 @@
return None, None
-def existing_version_info(version_info_path):
+def download_release(source_url, target_path):
+with open(target_path, 'wb') as file:
+file.write(urllib2.urlopen(source_url).read())
+
+
+def load_version_info(version_info_path):
if not os.path.exists(version_info_path):
return None
@@ -90,15 +94,6 @@
return json.load(file)
-def has_latest_release(version_info_path, version_info):
-return existing_version_info(version_info_path) == version_info
-
-
-def download_release(source_url, target_path):
-with open(target_path, 'wb') as file:
-file.write(urllib2.urlopen(source_url).read())
-
-
def save_version_info(version_info_path, version_info):
with open(version_info_path, 'w') as file:
json.dump(version_info, file)
@@ -110,35 +105,37 @@
binary_path = os.path.join(args.output_dir, args.filename)
version_info_path = binary_path + '.version'
-print 'Seeking latest release of {} from {}...'.format(args.filename, args.repo)
+print 'Updating {}...'.format(args.filename)
-try:
-release_url, version_info = find_latest_release(args)
+existing_version_info = load_version_info(version_info_path)
+if existing_version_info:
+print 'Found existing release:', existing_version_info['tag_name']
+else:
+print 'No existing release found.'
-if not release_url:
-print 'No release found!'
-return Status.COULD_NOT_FIND
+print 'Seeking latest release from {}...'.format(args.repo)
+release_url, latest_version_info = find_latest_release(args)
-if has_latest_release(version_info_path, version_info):
-print 'Already up-to-date:', existing_version_info(version_info_path)
-return Status.UP_TO_DATE
+if not latest_version_info:
+if existing_version_info:
+print 'Falling back to existing release!'
+return Status.USING_EXISTING
-except urllib2.URLError, error:
-print error
+print 'No release found!'
+return Status.COULD_NOT_FIND
-version_info = existing_version_info(version_info_path)
-if version_info:
-print 'Use existing version:', version_info
-return Status.USE_EXISTING_VERSION
-else:
-return Status.COULD_NOT_FIND
+print 'Found latest release:', latest_version_info['tag_name']
+if latest_vers
[webkit-changes] [225497] trunk
Title: [225497] trunk
Revision 225497
Author za...@apple.com
Date 2017-12-04 14:52:26 -0800 (Mon, 04 Dec 2017)
Log Message
REGRESSION (r211531): Text flow changes and overlaps other text after double-click selecting paragraph
https://bugs.webkit.org/show_bug.cgi?id=180377
Reviewed by Antti Koivisto.
Source/WebCore:
Take floats into account when computing the available width for a certain vertical position.
Test: fast/block/float/simple-line-layout-float-shrink-line.html
* rendering/SimpleLineLayout.cpp:
(WebCore::SimpleLineLayout::updateLineConstrains):
LayoutTests:
* fast/block/float/simple-line-layout-float-shrink-line-expected.html: Added.
* fast/block/float/simple-line-layout-float-shrink-line.html: Added.
Modified Paths
trunk/LayoutTests/ChangeLog
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/rendering/SimpleLineLayout.cpp
Added Paths
trunk/LayoutTests/fast/block/float/simple-line-layout-float-shrink-line-expected.html
trunk/LayoutTests/fast/block/float/simple-line-layout-float-shrink-line.html
Diff
Modified: trunk/LayoutTests/ChangeLog (225496 => 225497)
--- trunk/LayoutTests/ChangeLog 2017-12-04 22:51:13 UTC (rev 225496)
+++ trunk/LayoutTests/ChangeLog 2017-12-04 22:52:26 UTC (rev 225497)
@@ -1,3 +1,14 @@
+2017-12-04 Zalan Bujtas
+
+REGRESSION (r211531): Text flow changes and overlaps other text after double-click selecting paragraph
+https://bugs.webkit.org/show_bug.cgi?id=180377
+
+
+Reviewed by Antti Koivisto.
+
+* fast/block/float/simple-line-layout-float-shrink-line-expected.html: Added.
+* fast/block/float/simple-line-layout-float-shrink-line.html: Added.
+
2017-12-04 Antoine Quint
and scale() transforms yield shrunk small or clipped controls
Added: trunk/LayoutTests/fast/block/float/simple-line-layout-float-shrink-line-expected.html (0 => 225497)
--- trunk/LayoutTests/fast/block/float/simple-line-layout-float-shrink-line-expected.html (rev 0)
+++ trunk/LayoutTests/fast/block/float/simple-line-layout-float-shrink-line-expected.html 2017-12-04 22:52:26 UTC (rev 225497)
@@ -0,0 +1,27 @@
+
+
+
+
+.floating {
+width: 40px;
+height: 10px;
+border: 1px solid blue;
+}
+
+.container {
+width: 150px;
+}
+
+
+if (window.internals)
+internals.settings.setSimpleLineLayoutEnabled(false);
+
+
+
+
+
+
+ Pass if "this" wraps to the second line.
+
+
+
Added: trunk/LayoutTests/fast/block/float/simple-line-layout-float-shrink-line.html (0 => 225497)
--- trunk/LayoutTests/fast/block/float/simple-line-layout-float-shrink-line.html (rev 0)
+++ trunk/LayoutTests/fast/block/float/simple-line-layout-float-shrink-line.html 2017-12-04 22:52:26 UTC (rev 225497)
@@ -0,0 +1,23 @@
+
+
+
+
+.floating {
+width: 40px;
+height: 10px;
+border: 1px solid blue;
+}
+
+.container {
+width: 150px;
+}
+
+
+
+
+
+
+ Pass if "this" wraps to the second line.
+
+
+
Modified: trunk/Source/WebCore/ChangeLog (225496 => 225497)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 22:51:13 UTC (rev 225496)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 22:52:26 UTC (rev 225497)
@@ -1,3 +1,18 @@
+2017-12-04 Zalan Bujtas
+
+REGRESSION (r211531): Text flow changes and overlaps other text after double-click selecting paragraph
+https://bugs.webkit.org/show_bug.cgi?id=180377
+
+
+Reviewed by Antti Koivisto.
+
+Take floats into account when computing the available width for a certain vertical position.
+
+Test: fast/block/float/simple-line-layout-float-shrink-line.html
+
+* rendering/SimpleLineLayout.cpp:
+(WebCore::SimpleLineLayout::updateLineConstrains):
+
2017-12-04 Antoine Quint
and scale() transforms yield shrunk small or clipped controls
Modified: trunk/Source/WebCore/rendering/SimpleLineLayout.cpp (225496 => 225497)
--- trunk/Source/WebCore/rendering/SimpleLineLayout.cpp 2017-12-04 22:51:13 UTC (rev 225496)
+++ trunk/Source/WebCore/rendering/SimpleLineLayout.cpp 2017-12-04 22:52:26 UTC (rev 225497)
@@ -601,8 +601,9 @@
{
bool shouldApplyTextIndent = !flow.isAnonymous() || flow.parent()->firstChild() == &flow;
LayoutUnit height = flow.logicalHeight();
-line.setLogicalLeftOffset(flow.logicalLeftOffsetForLine(height, DoNotIndentText) + (shouldApplyTextIndent && isFirstLine ? flow.textIndentOffset() : LayoutUnit(0)));
-float logicalRightOffset = flow.logicalRightOffsetForLine(height, DoNotIndentText);
+LayoutUnit logicalHeight = flow.minLineHeightForReplacedRenderer(false, 0);
+line.setLogicalLeftOffset(flow.logicalLeftOffsetForLine(height, DoNotIndentText, logicalHeight) + (shouldApplyTextIndent && isFirstLine ? flow.textIndentOffset() : LayoutUnit(0)));
+float logicalRightOffset = flow.logicalRightOffsetForLine(height, DoNotIndentText, logicalHeight);
line.setAvailableWidth(std::max(0, logicalRightOffset - line.logicalLeftOffset()
[webkit-changes] [225496] trunk/Tools
Title: [225496] trunk/Tools Revision 225496 Author commit-qu...@webkit.org Date 2017-12-04 14:51:13 -0800 (Mon, 04 Dec 2017) Log Message Exclude some more leak callstacks https://bugs.webkit.org/show_bug.cgi?id=180379 Patch by Joseph Pecoraro on 2017-12-04 Reviewed by Alexey Proskuryakov. * Scripts/webkitpy/port/leakdetector.py: (LeakDetector._types_to_exclude_from_leaks): (LeakDetector._callstacks_to_exclude_from_leaks): (LeakDetector._leaks_args): (LeakDetector._types_to_exlude_from_leaks): Deleted. * Scripts/webkitpy/port/leakdetector_unittest.py: (LeakDetectorTest.test_leaks_args): Modified Paths trunk/Tools/ChangeLog trunk/Tools/Scripts/webkitpy/port/leakdetector.py trunk/Tools/Scripts/webkitpy/port/leakdetector_unittest.py Diff Modified: trunk/Tools/ChangeLog (225495 => 225496) --- trunk/Tools/ChangeLog 2017-12-04 22:49:25 UTC (rev 225495) +++ trunk/Tools/ChangeLog 2017-12-04 22:51:13 UTC (rev 225496) @@ -1,3 +1,18 @@ +2017-12-04 Joseph Pecoraro + +Exclude some more leak callstacks +https://bugs.webkit.org/show_bug.cgi?id=180379 + +Reviewed by Alexey Proskuryakov. + +* Scripts/webkitpy/port/leakdetector.py: +(LeakDetector._types_to_exclude_from_leaks): +(LeakDetector._callstacks_to_exclude_from_leaks): +(LeakDetector._leaks_args): +(LeakDetector._types_to_exlude_from_leaks): Deleted. +* Scripts/webkitpy/port/leakdetector_unittest.py: +(LeakDetectorTest.test_leaks_args): + 2017-12-03 Yusuke Suzuki WTF shouldn't have both Thread and ThreadIdentifier Modified: trunk/Tools/Scripts/webkitpy/port/leakdetector.py (225495 => 225496) --- trunk/Tools/Scripts/webkitpy/port/leakdetector.py 2017-12-04 22:49:25 UTC (rev 225495) +++ trunk/Tools/Scripts/webkitpy/port/leakdetector.py 2017-12-04 22:51:13 UTC (rev 225496) @@ -48,7 +48,7 @@ # This allows us ignore known leaks and only be alerted when new leaks occur. Some leaks are in the old # versions of the system frameworks that are being used by the leaks bots. Even though a leak has been # fixed, it will be listed here until the bot has been updated with the newer frameworks. -def _types_to_exlude_from_leaks(self): +def _types_to_exclude_from_leaks(self): # Currently we don't have any type excludes from OS leaks, but we will likely again in the future. return [] @@ -55,6 +55,9 @@ def _callstacks_to_exclude_from_leaks(self): callstacks = [ 'WTF::BitVector::OutOfLineBits::create', # https://bugs.webkit.org/show_bug.cgi?id=121662 +'WebCore::createPrivateStorageSession', # +'CIDeviceManagerStartMonitoring', # +'NSSpellChecker init', # ] return callstacks @@ -62,7 +65,7 @@ leaks_args = [] for callstack in self._callstacks_to_exclude_from_leaks(): leaks_args += ['--exclude-callstack=%s' % callstack] -for excluded_type in self._types_to_exlude_from_leaks(): +for excluded_type in self._types_to_exclude_from_leaks(): leaks_args += ['--exclude-type=%s' % excluded_type] leaks_args.append(pid) return leaks_args Modified: trunk/Tools/Scripts/webkitpy/port/leakdetector_unittest.py (225495 => 225496) --- trunk/Tools/Scripts/webkitpy/port/leakdetector_unittest.py 2017-12-04 22:49:25 UTC (rev 225495) +++ trunk/Tools/Scripts/webkitpy/port/leakdetector_unittest.py 2017-12-04 22:51:13 UTC (rev 225496) @@ -49,7 +49,7 @@ def test_leaks_args(self): detector = self._make_detector() detector._callstacks_to_exclude_from_leaks = lambda: ['foo bar', 'BAZ'] -detector._types_to_exlude_from_leaks = lambda: ['abcdefg', 'hi jklmno'] +detector._types_to_exclude_from_leaks = lambda: ['abcdefg', 'hi jklmno'] expected_args = ['--exclude-callstack=foo bar', '--exclude-callstack=BAZ', '--exclude-type=abcdefg', '--exclude-type=hi jklmno', 1234] self.assertEqual(detector._leaks_args(1234), expected_args) ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225495] trunk
Title: [225495] trunk
Revision 225495
Author commit-qu...@webkit.org
Date 2017-12-04 14:49:25 -0800 (Mon, 04 Dec 2017)
Log Message
and scale() transforms yield shrunk small or clipped controls
https://bugs.webkit.org/show_bug.cgi?id=180358
Patch by Antoine Quint on 2017-12-04
Reviewed by Dean Jackson.
Source/WebCore:
To compute the bounds of the media controls, we would call getBoundingClientRect()
on the root-most element in the shadow root, which is styled to have width and height
set to 100%, which allows to match the metrics of the host media element, without
the borders.
However, if the host media element has CSS transforms applied, these would be factored
into calls to getBoundingClientRect() and the controls would be too large and clipped,
in case the scale was more than 1, or shrunk, in case the scale was less than 1.
To obtain correct metrics, we computed the complete transformation matrix applied to
the host media element, and apply its inverted matrix to each corner point of the
bounds returned for the root-most container element in the shadow root.
Then we compute the width and height of this transformed rectangle and set those to
be our media controls' metrics.
Test: media/modern-media-controls/css/transformed-media.html
* Modules/modern-media-controls/media/media-controller.js:
(MediaController.prototype._updateControlsSize):
(MediaController.prototype._controlsWidth): Deleted.
LayoutTests:
Write a new test that ensures that we reflect the width and height, not accounting
for borders, of the host media element when sizing media controls.
* media/modern-media-controls/css/transformed-media-expected.txt: Added.
* media/modern-media-controls/css/transformed-media.html: Added.
Modified Paths
trunk/LayoutTests/ChangeLog
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/Modules/modern-media-controls/media/media-controller.js
Added Paths
trunk/LayoutTests/media/modern-media-controls/css/transformed-media-expected.txt
trunk/LayoutTests/media/modern-media-controls/css/transformed-media.html
Diff
Modified: trunk/LayoutTests/ChangeLog (225494 => 225495)
--- trunk/LayoutTests/ChangeLog 2017-12-04 22:47:04 UTC (rev 225494)
+++ trunk/LayoutTests/ChangeLog 2017-12-04 22:49:25 UTC (rev 225495)
@@ -1,5 +1,19 @@
2017-12-04 Antoine Quint
+ and scale() transforms yield shrunk small or clipped controls
+https://bugs.webkit.org/show_bug.cgi?id=180358
+
+
+Reviewed by Dean Jackson.
+
+Write a new test that ensures that we reflect the width and height, not accounting
+for borders, of the host media element when sizing media controls.
+
+* media/modern-media-controls/css/transformed-media-expected.txt: Added.
+* media/modern-media-controls/css/transformed-media.html: Added.
+
+2017-12-04 Antoine Quint
+
Cursor is not visible after exiting full screen video
https://bugs.webkit.org/show_bug.cgi?id=180247
Added: trunk/LayoutTests/media/modern-media-controls/css/transformed-media-expected.txt (0 => 225495)
--- trunk/LayoutTests/media/modern-media-controls/css/transformed-media-expected.txt (rev 0)
+++ trunk/LayoutTests/media/modern-media-controls/css/transformed-media-expected.txt 2017-12-04 22:49:25 UTC (rev 225495)
@@ -0,0 +1,6 @@
+PASS mediaControls.style.width is "560px"
+PASS mediaControls.style.height is "260px"
+PASS successfullyParsed is true
+
+TEST COMPLETE
+
Added: trunk/LayoutTests/media/modern-media-controls/css/transformed-media.html (0 => 225495)
--- trunk/LayoutTests/media/modern-media-controls/css/transformed-media.html (rev 0)
+++ trunk/LayoutTests/media/modern-media-controls/css/transformed-media.html 2017-12-04 22:49:25 UTC (rev 225495)
@@ -0,0 +1,39 @@
+
+
+
+
+* {
+transform-origin: top left;
+}
+
+div {
+transform: scale(0.75);
+}
+
+video {
+width: 600px;
+height: 300px;
+border: 20px solid blue;
+box-sizing: border-box;
+background-color: red;
+transform: scale(0.5);
+}
+
+
+
+
+
+
+
+
+
+const media = document.querySelector("video");
+const shadowRoot = window.internals.shadowRoot(media);
+const mediaControls = shadowRoot.querySelector(".media-controls");
+shouldBeEqualToString("mediaControls.style.width", "560px"); // width - (2 * border)
+shouldBeEqualToString("mediaControls.style.height", "260px"); // height - (2 * border)
+
+
+
[webkit-changes] [225494] trunk
Title: [225494] trunk
Revision 225494
Author commit-qu...@webkit.org
Date 2017-12-04 14:47:04 -0800 (Mon, 04 Dec 2017)
Log Message
Cursor is not visible after exiting full screen video
https://bugs.webkit.org/show_bug.cgi?id=180247
Patch by Antoine Quint on 2017-12-04
Reviewed by Dean Jackson.
Source/WebCore:
There is a dedicated NSCursor method to temporarily hide the mouse cursor while the mouse is idle,
so we use this platform functionality, already exposed through the page chrome, instead which
implements the expected behavior. Now, the mouse cursor is hidden while the mouse is stationary
when a is fullscreen, as before, but as soon as the user exits fullscreen, the mouse cursor
reappears.
No test provided as I don't believe this platform behavior can be tested, there is no API to query
whether the cursor is visible.
* page/EventHandler.cpp:
(WebCore::EventHandler::cancelAutoHideCursorTimer):
(WebCore::EventHandler::autoHideCursorTimerFired):
LayoutTests:
Remove the existing test which would query a cursor state that is no longer relevant.
* fullscreen/video-cursor-auto-hide-expected.txt: Removed.
* fullscreen/video-cursor-auto-hide.html: Removed.
* platform/gtk/TestExpectations:
Modified Paths
trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/gtk/TestExpectations
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/page/EventHandler.cpp
Removed Paths
trunk/LayoutTests/fullscreen/video-cursor-auto-hide-expected.txt
trunk/LayoutTests/fullscreen/video-cursor-auto-hide.html
Diff
Modified: trunk/LayoutTests/ChangeLog (225493 => 225494)
--- trunk/LayoutTests/ChangeLog 2017-12-04 22:06:50 UTC (rev 225493)
+++ trunk/LayoutTests/ChangeLog 2017-12-04 22:47:04 UTC (rev 225494)
@@ -1,3 +1,17 @@
+2017-12-04 Antoine Quint
+
+Cursor is not visible after exiting full screen video
+https://bugs.webkit.org/show_bug.cgi?id=180247
+
+
+Reviewed by Dean Jackson.
+
+Remove the existing test which would query a cursor state that is no longer relevant.
+
+* fullscreen/video-cursor-auto-hide-expected.txt: Removed.
+* fullscreen/video-cursor-auto-hide.html: Removed.
+* platform/gtk/TestExpectations:
+
2017-12-04 Michael Catanzaro
[GTK] Unreviewed layout test gardening
Deleted: trunk/LayoutTests/fullscreen/video-cursor-auto-hide-expected.txt (225493 => 225494)
--- trunk/LayoutTests/fullscreen/video-cursor-auto-hide-expected.txt 2017-12-04 22:06:50 UTC (rev 225493)
+++ trunk/LayoutTests/fullscreen/video-cursor-auto-hide-expected.txt 2017-12-04 22:47:04 UTC (rev 225494)
@@ -1,8 +0,0 @@
-This tests that the cursor hides after hovering over a video element in full screen mode.
-
-
-EVENT(webkitfullscreenchange)
-EXPECTED (window.internals.getCurrentCursorInfo() == 'type=Pointer hotSpot=0,0') OK
-EXPECTED (window.internals.getCurrentCursorInfo() == 'type=None hotSpot=0,0') OK
-END OF TEST
-
Deleted: trunk/LayoutTests/fullscreen/video-cursor-auto-hide.html (225493 => 225494)
--- trunk/LayoutTests/fullscreen/video-cursor-auto-hide.html 2017-12-04 22:06:50 UTC (rev 225493)
+++ trunk/LayoutTests/fullscreen/video-cursor-auto-hide.html 2017-12-04 22:47:04 UTC (rev 225494)
@@ -1,31 +0,0 @@
-
-This tests that the cursor hides after hovering over a video element in full screen mode.
-
-
-
-
-var wrapper = document.getElementById('wrapper');
-
-function checkForHiddenMouse()
-{
-testExpected('window.internals.getCurrentCursorInfo()', 'type=None hotSpot=0,0');
-endTest();
-}
-
-function fullscreenChanged(event)
-{
-if (window.internals) {
-internals.settings.setTimeWithoutMouseMovementBeforeHidingControls(0);
-wrapperBox = internals.boundingBox(wrapper);
-eventSender.mouseMoveTo(wrapperBox.left + wrapperBox.width / 2, wrapperBox.top + wrapperBox.height / 2);
-testExpected('window.internals.getCurrentCursorInfo()', 'type=Pointer hotSpot=0,0');
-setTimeout(checkForHiddenMouse, 0);
-}
-else
-logResult(false, "window.internals == undefined");
-}
-waitForEvent(document, 'webkitfullscreenchange', fullscreenChanged);
-runWithKeyDown(function(){wrapper.webkitRequestFullScreen()});
-
-
Modified: trunk/LayoutTests/platform/gtk/TestExpectations (225493 => 225494)
--- trunk/LayoutTests/platform/gtk/TestExpectations 2017-12-04 22:06:50 UTC (rev 225493)
+++ trunk/LayoutTests/platform/gtk/TestExpectations 2017-12-04 22:47:04 UTC (rev 225494)
@@ -903,9 +903,6 @@
# Requires ENABLE(MEDIA_CAPTURE)
webkit.org/b/85958 fast/forms/file/file-input-capture.html [ Failure ]
-# CURSOR_VISIBILITY is not enabled in gtk
-webkit.org/b/107601 fullscreen/video-cursor-auto-hide.html [ Failure ]
-
webkit.org/b/146718 accessibility/aria-hidden-false-works-in-subtrees.html [ Failure ]
webkit.org/
[webkit-changes] [225493] trunk/JSTests
Title: [225493] trunk/JSTests
Revision 225493
Author jfbast...@apple.com
Date 2017-12-04 14:06:50 -0800 (Mon, 04 Dec 2017)
Log Message
Proxy all functions, except the $ objects
https://bugs.webkit.org/show_bug.cgi?id=180375
Reviewed by Saam Barati.
It looks like this test may have broken some executions because I
call some internal objects. Explicitly ignore objects whose name
starts with "$" because it's a bad idea anyways.
* stress/proxy-all-the-parameters.js:
(generateObjects):
(get throw):
Modified Paths
trunk/JSTests/ChangeLog
trunk/JSTests/stress/proxy-all-the-parameters.js
Diff
Modified: trunk/JSTests/ChangeLog (225492 => 225493)
--- trunk/JSTests/ChangeLog 2017-12-04 22:00:24 UTC (rev 225492)
+++ trunk/JSTests/ChangeLog 2017-12-04 22:06:50 UTC (rev 225493)
@@ -1,3 +1,18 @@
+2017-12-04 JF Bastien
+
+Proxy all functions, except the $ objects
+https://bugs.webkit.org/show_bug.cgi?id=180375
+
+Reviewed by Saam Barati.
+
+It looks like this test may have broken some executions because I
+call some internal objects. Explicitly ignore objects whose name
+starts with "$" because it's a bad idea anyways.
+
+* stress/proxy-all-the-parameters.js:
+(generateObjects):
+(get throw):
+
2017-12-04 Saam Barati
We need to leave room on the top of the stack for the FTL TailCall slow path so it doesn't overwrite things we want to retrieve when doing a stack walk when throwing an exception
Modified: trunk/JSTests/stress/proxy-all-the-parameters.js (225492 => 225493)
--- trunk/JSTests/stress/proxy-all-the-parameters.js 2017-12-04 22:00:24 UTC (rev 225492)
+++ trunk/JSTests/stress/proxy-all-the-parameters.js 2017-12-04 22:06:50 UTC (rev 225493)
@@ -1,7 +1,5 @@
const verbose = false;
-const ignore = ['quit', 'readline', 'waitForReport', 'flashHeapAccess', 'leaving', 'getReport'];
-
function isPropertyOfType(obj, name, type) {
let desc;
desc = Object.getOwnPropertyDescriptor(obj, name)
@@ -22,6 +20,8 @@
return;
let obj_names = getProperties(root, 'object');
for (let obj_name of obj_names) {
+if (obj_name.startsWith('$'))
+continue; // Ignore internal objects.
let obj = root[obj_name];
yield obj;
yield* generateObjects(obj, level + 1);
@@ -44,8 +44,6 @@
for (let o of getObjects()) {
for (let f of getFunctions(o)) {
-if (ignore.includes(f))
-continue;
const arityPlusOne = o[f].length + 1;
if (verbose)
print(`Calling ${o}['${f}'](${Array(arityPlusOne).fill("thrower")})`);
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225492] trunk
Title: [225492] trunk
Revision 225492
Author sbar...@apple.com
Date 2017-12-04 14:00:24 -0800 (Mon, 04 Dec 2017)
Log Message
We need to leave room on the top of the stack for the FTL TailCall slow path so it doesn't overwrite things we want to retrieve when doing a stack walk when throwing an exception
https://bugs.webkit.org/show_bug.cgi?id=180366
Reviewed by Michael Saboff.
JSTests:
* stress/ftl-tail-call-throw-exception-from-slow-path-recover-stack-values.js: Added.
(theParent):
(test1.base.getParentStaticValue):
(test1.base):
(test1.__v_24888.prototype.set prop):
(test1.__v_24888):
(test2.base.getParentStaticValue):
(test2.base):
(test2.__v_24888.prototype.set prop):
(test2.__v_24888):
(test2):
Source/_javascript_Core:
On the TailCall slow path, the CallFrameShuffler will build the frame with
respect to SP instead of FP. However, this may overwrite slots on the stack
that are needed if the slow path C call does a stack walk. The slow path
C call does a stack walk when it throws an exception. This patch fixes
this bug by ensuring that the top of the stack in the FTL always has enough
space to allow CallFrameShuffler to build a frame without overwriting any
items on the stack that are needed when doing a stack walk.
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileTailCall):
Modified Paths
trunk/JSTests/ChangeLog
trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/ftl/FTLLowerDFGToB3.cpp
Added Paths
trunk/JSTests/stress/ftl-tail-call-throw-exception-from-slow-path-recover-stack-values.js
Diff
Modified: trunk/JSTests/ChangeLog (225491 => 225492)
--- trunk/JSTests/ChangeLog 2017-12-04 21:56:25 UTC (rev 225491)
+++ trunk/JSTests/ChangeLog 2017-12-04 22:00:24 UTC (rev 225492)
@@ -1,3 +1,23 @@
+2017-12-04 Saam Barati
+
+We need to leave room on the top of the stack for the FTL TailCall slow path so it doesn't overwrite things we want to retrieve when doing a stack walk when throwing an exception
+https://bugs.webkit.org/show_bug.cgi?id=180366
+
+
+Reviewed by Michael Saboff.
+
+* stress/ftl-tail-call-throw-exception-from-slow-path-recover-stack-values.js: Added.
+(theParent):
+(test1.base.getParentStaticValue):
+(test1.base):
+(test1.__v_24888.prototype.set prop):
+(test1.__v_24888):
+(test2.base.getParentStaticValue):
+(test2.base):
+(test2.__v_24888.prototype.set prop):
+(test2.__v_24888):
+(test2):
+
2017-12-01 JF Bastien
Try proxying all function arguments
Added: trunk/JSTests/stress/ftl-tail-call-throw-exception-from-slow-path-recover-stack-values.js (0 => 225492)
--- trunk/JSTests/stress/ftl-tail-call-throw-exception-from-slow-path-recover-stack-values.js (rev 0)
+++ trunk/JSTests/stress/ftl-tail-call-throw-exception-from-slow-path-recover-stack-values.js 2017-12-04 22:00:24 UTC (rev 225492)
@@ -0,0 +1,31 @@
+var theParent = function () { };
+
+function test1() {
+var base = class C extends theParent {
+static getParentStaticValue() {
+let arrow = (a,b,c) => super.getStaticValue(a,b,c);
+return arrow(1,1,1);
+}
+};
+
+for (let i = 0; i < 1; i++) {
+try { base.getParentStaticValue() } catch (e) {}
+try { base.getParentStaticValue() } catch (e) {}
+}
+}
+test1();
+
+function test2() {
+var base = class C extends theParent {
+static getParentStaticValue() {
+let arrow = () => super.getStaticValue();
+return arrow();
+}
+};
+
+for (let i = 0; i < 1; i++) {
+try { base.getParentStaticValue() } catch (e) {}
+try { base.getParentStaticValue() } catch (e) {}
+}
+}
+test2();
Modified: trunk/Source/_javascript_Core/ChangeLog (225491 => 225492)
--- trunk/Source/_javascript_Core/ChangeLog 2017-12-04 21:56:25 UTC (rev 225491)
+++ trunk/Source/_javascript_Core/ChangeLog 2017-12-04 22:00:24 UTC (rev 225492)
@@ -1,3 +1,22 @@
+2017-12-04 Saam Barati
+
+We need to leave room on the top of the stack for the FTL TailCall slow path so it doesn't overwrite things we want to retrieve when doing a stack walk when throwing an exception
+https://bugs.webkit.org/show_bug.cgi?id=180366
+
+
+Reviewed by Michael Saboff.
+
+On the TailCall slow path, the CallFrameShuffler will build the frame with
+respect to SP instead of FP. However, this may overwrite slots on the stack
+that are needed if the slow path C call does a stack walk. The slow path
+C call does a stack walk when it throws an exception. This patch fixes
+this bug by ensuring that the top of the stack in the FTL always has enough
+space to allow CallFrameShuffler to build a frame without overwriting any
+items on the stack that are needed when doing a stack walk.
+
+* ftl/FTLLowerDFGToB3.cpp:
+
[webkit-changes] [225491] trunk/Source/WebCore
Title: [225491] trunk/Source/WebCore
Revision 225491
Author cdu...@apple.com
Date 2017-12-04 13:56:25 -0800 (Mon, 04 Dec 2017)
Log Message
ServiceWorkerGlobalScope::skipWaiting(Ref&&) is unsafe
https://bugs.webkit.org/show_bug.cgi?id=180372
Reviewed by Youenn Fablet.
Ref the WorkerThread and capture it in the lambda. Keep the pending promises in
a HashMap on the ServiceWorkerGlobalScope so that they stay on the worker thread.
* workers/service/ServiceWorkerGlobalScope.cpp:
(WebCore::ServiceWorkerGlobalScope::skipWaiting):
* workers/service/ServiceWorkerGlobalScope.h:
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/workers/service/ServiceWorkerGlobalScope.cpp
trunk/Source/WebCore/workers/service/ServiceWorkerGlobalScope.h
Diff
Modified: trunk/Source/WebCore/ChangeLog (225490 => 225491)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 21:50:31 UTC (rev 225490)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 21:56:25 UTC (rev 225491)
@@ -1,3 +1,17 @@
+2017-12-04 Chris Dumez
+
+ServiceWorkerGlobalScope::skipWaiting(Ref&&) is unsafe
+https://bugs.webkit.org/show_bug.cgi?id=180372
+
+Reviewed by Youenn Fablet.
+
+Ref the WorkerThread and capture it in the lambda. Keep the pending promises in
+a HashMap on the ServiceWorkerGlobalScope so that they stay on the worker thread.
+
+* workers/service/ServiceWorkerGlobalScope.cpp:
+(WebCore::ServiceWorkerGlobalScope::skipWaiting):
+* workers/service/ServiceWorkerGlobalScope.h:
+
2017-12-04 Brady Eidson
Get a directory path to SWServers for storing ServiceWorker registrations.
Modified: trunk/Source/WebCore/workers/service/ServiceWorkerGlobalScope.cpp (225490 => 225491)
--- trunk/Source/WebCore/workers/service/ServiceWorkerGlobalScope.cpp 2017-12-04 21:50:31 UTC (rev 225490)
+++ trunk/Source/WebCore/workers/service/ServiceWorkerGlobalScope.cpp 2017-12-04 21:56:25 UTC (rev 225491)
@@ -50,11 +50,16 @@
void ServiceWorkerGlobalScope::skipWaiting(Ref&& promise)
{
-callOnMainThread([this, protectedThis = makeRef(*this), threadIdentifier = thread().identifier(), promise = WTFMove(promise)]() mutable {
+uint64_t requestIdentifier = ++m_lastRequestIdentifier;
+m_pendingSkipWaitingPromises.add(requestIdentifier, WTFMove(promise));
+
+callOnMainThread([workerThread = makeRef(thread()), requestIdentifier]() mutable {
if (auto* connection = SWContextManager::singleton().connection()) {
-connection->skipWaiting(threadIdentifier, [this, protectedThis = WTFMove(protectedThis), promise = WTFMove(promise)]() mutable {
-thread().runLoop().postTask([promise = WTFMove(promise), protectedThis = WTFMove(protectedThis)](auto&) {
-promise->resolve();
+connection->skipWaiting(workerThread->identifier(), [workerThread = WTFMove(workerThread), requestIdentifier] {
+workerThread->runLoop().postTask([requestIdentifier](auto& context) {
+auto& scope = downcast(context);
+if (auto promise = scope.m_pendingSkipWaitingPromises.take(requestIdentifier))
+promise->resolve();
});
});
}
Modified: trunk/Source/WebCore/workers/service/ServiceWorkerGlobalScope.h (225490 => 225491)
--- trunk/Source/WebCore/workers/service/ServiceWorkerGlobalScope.h 2017-12-04 21:50:31 UTC (rev 225490)
+++ trunk/Source/WebCore/workers/service/ServiceWorkerGlobalScope.h 2017-12-04 21:56:25 UTC (rev 225491)
@@ -77,6 +77,9 @@
Ref m_clients;
HashMap m_clientMap;
Vector> m_extendedEvents;
+
+uint64_t m_lastRequestIdentifier { 0 };
+HashMap> m_pendingSkipWaitingPromises;
};
} // namespace WebCore
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225490] trunk/Source
Title: [225490] trunk/Source
Revision 225490
Author beid...@apple.com
Date 2017-12-04 13:50:31 -0800 (Mon, 04 Dec 2017)
Log Message
Get a directory path to SWServers for storing ServiceWorker registrations.
https://bugs.webkit.org/show_bug.cgi?id=180362
Reviewed by Chris Dumez.
Source/WebCore:
No new tests (No observable behavior change yet).
* workers/service/server/SWServer.cpp:
(WebCore::SWServer::SWServer):
* workers/service/server/SWServer.h:
Source/WebKit:
This gets all of the plumbing in place for specifying a ServiceWorker registration storage path
in a particular WebsiteDataStore.
* Shared/Storage/StorageProcessCreationParameters.cpp:
(WebKit::StorageProcessCreationParameters::encode const):
(WebKit::StorageProcessCreationParameters::decode):
* Shared/Storage/StorageProcessCreationParameters.h:
* StorageProcess/StorageProcess.cpp:
(WebKit::StorageProcess::initializeWebsiteDataStore):
(WebKit::StorageProcess::swServerForSession):
* StorageProcess/StorageProcess.h:
* UIProcess/API/APIProcessPoolConfiguration.cpp:
(API::ProcessPoolConfiguration::createWithLegacyOptions):
(API::ProcessPoolConfiguration::createWithWebsiteDataStoreConfiguration):
(API::ProcessPoolConfiguration::ProcessPoolConfiguration):
* UIProcess/API/APIProcessPoolConfiguration.h:
* UIProcess/API/APIWebsiteDataStore.h:
* UIProcess/API/Cocoa/APIWebsiteDataStoreCocoa.mm:
(API::WebsiteDataStore::defaultServiceWorkerRegistrationDirectory):
* UIProcess/API/glib/APIWebsiteDataStoreGLib.cpp:
(API::WebsiteDataStore::defaultServiceWorkerRegistrationDirectory):
* UIProcess/Cocoa/WebProcessPoolCocoa.mm:
(WebKit::WebProcessPool::legacyPlatformDefaultIndexedDBDatabaseDirectory):
(WebKit::WebProcessPool::legacyPlatformDefaultServiceWorkerRegistrationDirectory):
* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::ensureStorageProcessAndWebsiteDataStore):
* UIProcess/WebProcessPool.h:
* UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::resolveDirectoriesIfNecessary):
(WebKit::WebsiteDataStore::storageProcessParameters):
* UIProcess/WebsiteData/WebsiteDataStore.h:
(WebKit::WebsiteDataStore::resolvedServiceWorkerRegistrationDirectory const):
* UIProcess/gtk/WebProcessPoolGtk.cpp:
(WebKit::WebProcessPool::legacyPlatformDefaultServiceWorkerRegistrationDirectory):
* UIProcess/wpe/WebProcessPoolWPE.cpp:
(WebKit::WebProcessPool::legacyPlatformDefaultServiceWorkerRegistrationDirectory):
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/workers/service/server/SWServer.cpp
trunk/Source/WebCore/workers/service/server/SWServer.h
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/Shared/Storage/StorageProcessCreationParameters.cpp
trunk/Source/WebKit/Shared/Storage/StorageProcessCreationParameters.h
trunk/Source/WebKit/StorageProcess/StorageProcess.cpp
trunk/Source/WebKit/StorageProcess/StorageProcess.h
trunk/Source/WebKit/UIProcess/API/APIProcessPoolConfiguration.cpp
trunk/Source/WebKit/UIProcess/API/APIProcessPoolConfiguration.h
trunk/Source/WebKit/UIProcess/API/APIWebsiteDataStore.h
trunk/Source/WebKit/UIProcess/API/Cocoa/APIWebsiteDataStoreCocoa.mm
trunk/Source/WebKit/UIProcess/API/glib/APIWebsiteDataStoreGLib.cpp
trunk/Source/WebKit/UIProcess/Cocoa/WebProcessPoolCocoa.mm
trunk/Source/WebKit/UIProcess/WebProcessPool.cpp
trunk/Source/WebKit/UIProcess/WebProcessPool.h
trunk/Source/WebKit/UIProcess/WebsiteData/WebsiteDataStore.cpp
trunk/Source/WebKit/UIProcess/WebsiteData/WebsiteDataStore.h
trunk/Source/WebKit/UIProcess/gtk/WebProcessPoolGtk.cpp
trunk/Source/WebKit/UIProcess/wpe/WebProcessPoolWPE.cpp
Diff
Modified: trunk/Source/WebCore/ChangeLog (225489 => 225490)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 21:43:36 UTC (rev 225489)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 21:50:31 UTC (rev 225490)
@@ -1,3 +1,16 @@
+2017-12-04 Brady Eidson
+
+Get a directory path to SWServers for storing ServiceWorker registrations.
+https://bugs.webkit.org/show_bug.cgi?id=180362
+
+Reviewed by Chris Dumez.
+
+No new tests (No observable behavior change yet).
+
+* workers/service/server/SWServer.cpp:
+(WebCore::SWServer::SWServer):
+* workers/service/server/SWServer.h:
+
2017-12-04 Devin Rousso
Web Inspector: provide method for recording CanvasRenderingContext2D from _javascript_
Modified: trunk/Source/WebCore/workers/service/server/SWServer.cpp (225489 => 225490)
--- trunk/Source/WebCore/workers/service/server/SWServer.cpp 2017-12-04 21:43:36 UTC (rev 225489)
+++ trunk/Source/WebCore/workers/service/server/SWServer.cpp 2017-12-04 21:50:31 UTC (rev 225490)
@@ -204,9 +204,10 @@
m_server.syncTerminateWorker(*worker);
}
-SWServer::SWServer(UniqueRef&& originStore)
+SWServer::SWServer(UniqueRef&& originStore, const String& registrationDatabaseDirectory)
: m_originStore(WTFMove(originStore))
{
+UNUSED_PARAM(registrationDatabaseDirectory);
allServers().add(this);
m_taskThread = Thread::create(ASCI
[webkit-changes] [225489] trunk/LayoutTests
Title: [225489] trunk/LayoutTests Revision 225489 Author mcatanz...@igalia.com Date 2017-12-04 13:43:36 -0800 (Mon, 04 Dec 2017) Log Message [GTK] Unreviewed layout test gardening * platform/gtk/TestExpectations: Modified Paths trunk/LayoutTests/ChangeLog trunk/LayoutTests/platform/gtk/TestExpectations Diff Modified: trunk/LayoutTests/ChangeLog (225488 => 225489) --- trunk/LayoutTests/ChangeLog 2017-12-04 21:40:55 UTC (rev 225488) +++ trunk/LayoutTests/ChangeLog 2017-12-04 21:43:36 UTC (rev 225489) @@ -1,3 +1,9 @@ +2017-12-04 Michael Catanzaro + +[GTK] Unreviewed layout test gardening + +* platform/gtk/TestExpectations: + 2017-12-04 Devin Rousso Web Inspector: provide method for recording CanvasRenderingContext2D from _javascript_ Modified: trunk/LayoutTests/platform/gtk/TestExpectations (225488 => 225489) --- trunk/LayoutTests/platform/gtk/TestExpectations 2017-12-04 21:40:55 UTC (rev 225488) +++ trunk/LayoutTests/platform/gtk/TestExpectations 2017-12-04 21:43:36 UTC (rev 225489) @@ -847,7 +847,6 @@ webkit.org/b/141698 editing/pasteboard/5761530-1.html [ Failure ] webkit.org/b/141699 fast/attachment/attachment-disabled-rendering.html [ Failure ] webkit.org/b/141699 fast/attachment/attachment-rendering.html [ Failure ] -webkit.org/b/141699 fast/attachment/attachment-without-appearance.html [ Failure ] webkit.org/b/141831 svg/animations/animate-text-nested-transforms.html [ Failure ] webkit.org/b/98653 svg/text/lengthAdjust-text-metrics.html [ Failure ] @@ -1176,10 +1175,10 @@ webkit.org/b/175419 imported/w3c/web-platform-tests/streams/readable-byte-streams/detached-buffers.serviceworker.https.html [ Failure ] # Loading MP4 as an image is not supported by GTK -fast/images/animated-image-mp4.html [ Skip ] +webkit.org/b/180370 fast/images/animated-image-mp4.html [ Skip ] +webkit.org/b/180370 http/tests/images/image-supports-video.html [ Skip ] -webkit.org/b/177632 fast/text/flag-codepoint.html [ ImageOnlyFailure ] - +# GTK ought to support the new modern pasteboard API webkit.org/b/177633 editing/pasteboard/data-transfer-get-data-on-paste-rich-text.html [ Failure ] webkit.org/b/177633 editing/pasteboard/data-transfer-get-data-on-paste-plain-text.html [ Failure ] webkit.org/b/177633 editing/pasteboard/data-transfer-get-data-on-paste-custom.html [ Failure ] @@ -1187,26 +1186,28 @@ webkit.org/b/177633 http/tests/security/clipboard/copy-paste-html-cross-origin-iframe-across-origin.html [ Skip ] webkit.org/b/177633 http/tests/security/clipboard/copy-paste-html-cross-origin-iframe-in-same-origin.html [ Skip ] -webkit.org/b/179052 compositing/contents-format/subpixel-antialiased-nested-layer.html [ Failure ] - -webkit.org/b/179989 http/tests/misc/prefetch-purpose.html [ Timeout ] -webkit.org/b/179989 http/tests/misc/link-rel-prefetch-and-subresource.html [ Timeout ] -webkit.org/b/179989 fast/dom/HTMLLinkElement/subresource.html [ Timeout ] -webkit.org/b/179989 fast/dom/HTMLLinkElement/prefetch-onload.html [ Timeout ] -webkit.org/b/179989 fast/dom/HTMLLinkElement/prefetch-onerror.html [ Timeout ] -webkit.org/b/179989 fast/dom/HTMLLinkElement/prefetch.html [ Timeout ] -webkit.org/b/179989 fast/dom/HTMLLinkElement/link-and-subresource-test-nonexistent.html [ Timeout ] -webkit.org/b/179989 fast/dom/HTMLLinkElement/link-and-subresource-test.html [ Timeout ] +# We should consider enabling ENABLE_LINK_PREFETCH and ENABLE_CSS3_TEXT +webkit.org/b/179989 http/tests/misc/prefetch-purpose.html [ Skip ] +webkit.org/b/179989 http/tests/misc/link-rel-prefetch-and-subresource.html [ Skip ] +webkit.org/b/179989 fast/dom/HTMLLinkElement/subresource.html [ Skip ] +webkit.org/b/179989 fast/dom/HTMLLinkElement/prefetch-onload.html [ Skip ] +webkit.org/b/179989 fast/dom/HTMLLinkElement/prefetch-onerror.html [ Skip ] +webkit.org/b/179989 fast/dom/HTMLLinkElement/prefetch.html [ Skip ] +webkit.org/b/179989 fast/dom/HTMLLinkElement/link-and-subresource-test-nonexistent.html [ Skip ] +webkit.org/b/179989 fast/dom/HTMLLinkElement/link-and-subresource-test.html [ Skip ] webkit.org/b/179989 fast/css3-text/css3-text-justify/getComputedStyle/getComputedStyle-text-justify-inherited.html [ Failure ] webkit.org/b/179989 fast/css3-text/css3-text-justify/getComputedStyle/getComputedStyle-text-justify.html [ Failure ] webkit.org/b/179989 fast/css3-text/css3-text-align-last/getComputedStyle/getComputedStyle-text-align-last-inherited.html [ Failure ] webkit.org/b/179989 fast/css3-text/css3-text-align-last/getComputedStyle/getComputedStyle-text-align-last.html [ Failure ] -webkit.org/b/180148 accessibility/alternative-presentation-button-input-type.html [ Failure ] -webkit.org/b/180148 accessibility/alternative-presentation-button.html [ Failure ] +# These tests rely on an iOS feature. +webkit.org/b/180148 accessibility/alternative-presentation-button-input-type.html [ Skip ] +webkit.org/b/180148 accessibility/alternative-presentation-button.html [ Skip ] #//
[webkit-changes] [225487] trunk/Source/WebInspectorUI
Title: [225487] trunk/Source/WebInspectorUI Revision 225487 Author mattba...@apple.com Date 2017-12-04 12:40:27 -0800 (Mon, 04 Dec 2017) Log Message Web Inspector: Canvas Tab initial user interface needs some polish https://bugs.webkit.org/show_bug.cgi?id=179330 Reviewed by Joseph Pecoraro. Update the Canvas overview placeholder text and prevent appearance of the vertical scrollbar when no canvases exist. This patch also includes a bunch of cleanup of code that creates content placeholder text throughout the UI, and makes sure all text is styled consistently. * Localizations/en.lproj/localizedStrings.js: Updated Canvas overview message text. * UserInterface/Base/Main.js: Make the "message text view" a container with child message element, instead of a simple element with text content. This makes it simpler to customize the message (by adding buttons, more text, etc). * UserInterface/Views/CanvasDetailsSidebarPanel.css: (.sidebar > .panel.details.canvas .details-section.canvas-extensions .content > ul): (.sidebar > .panel.details.canvas > .content > .empty-content-placeholder): Deleted. (.sidebar > .panel.details.canvas > .content > .empty-content-placeholder > .message): Deleted. * UserInterface/Views/CanvasDetailsSidebarPanel.js: (WI.CanvasDetailsSidebarPanel.prototype.initialLayout): * UserInterface/Views/CanvasOverviewContentView.js: (WI.CanvasOverviewContentView): Customize the content placeholder to include a description beneath the "No Canvas Contexts" message. * UserInterface/Views/CollectionContentView.css: (.content-view.collection): Change overflow-y to auto, to prevent scrollbar from appearing when "Show scroll bars" system preference is enabled on macOS. * UserInterface/Views/CollectionContentView.js: Add support for custom content placeholders. (WI.CollectionContentView): (WI.CollectionContentView.prototype._showContentPlaceholder): (WI.CollectionContentView.prototype._hideContentPlaceholder): * UserInterface/Views/ContentView.css: (.content-view > .message-text-view): (.content-view > .message-text-view > .message): * UserInterface/Views/DebuggerSidebarPanel.css: (.sidebar > .panel.navigation.debugger > .content): (.sidebar > .panel.navigation.debugger > :matches(.content, .empty-content-placeholder)): Deleted. * UserInterface/Views/DetailsSection.css: (.details-section > .content > .group > .row > .message-text-view): * UserInterface/Views/Main.css: (.message-text-view): (.message-text-view > .message): (.message-text-view.error): * UserInterface/Views/NavigationSidebarPanel.css: (.sidebar > .panel.navigation > .overflow-shadow): (.sidebar > .panel.navigation > .content .empty-content-placeholder): Deleted. (.sidebar > .panel.navigation > .content > .empty-content-placeholder): Deleted. (.sidebar > .panel.navigation > .content .empty-content-placeholder > .message): Deleted. * UserInterface/Views/NavigationSidebarPanel.js: (WI.NavigationSidebarPanel.prototype.showEmptyContentPlaceholder): (WI.NavigationSidebarPanel.prototype._createEmptyContentPlaceholderIfNeeded): * UserInterface/Views/NetworkTableContentView.css: (.content-view.network .message-text-view): (.content-view.network .message-text-view > .message): Workaround to prevent "No Filter Results" message from being styled like "full width" messages displayed in ContentViews (large, bold font). This is needed because the placeholder message element is a immediate child of the network content view, rather than the network table. (.content-view.network .empty-content-placeholder): Deleted. (body[dir=ltr] .content-view.network .empty-content-placeholder): Deleted. (body[dir=rtl] .content-view.network .empty-content-placeholder): Deleted. (.content-view.network .empty-content-placeholder > .message): Deleted. * UserInterface/Views/NetworkTableContentView.js: (WI.NetworkTableContentView.prototype._showEmptyFilterResultsMessage): * UserInterface/Views/RecordingNavigationSidebarPanel.css: (.sidebar > .panel.navigation.recording > .content): (.sidebar > .panel.navigation.recording > :matches(.content, .empty-content-placeholder)): Deleted. * UserInterface/Views/ResourceSidebarPanel.css: (.sidebar > .panel.navigation.resource > .content): (.sidebar > .panel.navigation.resource > :matches(.content, .empty-content-placeholder)): Deleted. * UserInterface/Views/RulesStyleDetailsPanel.css: (.sidebar > .panel.details.css-style > .content > .rules > .message-text-view): (.sidebar > .panel.details.css-style > .content > .rules:not(.filter-non-matching) > .message-text-view): (.sidebar > .panel.details.css-style > .content > .rules:not(.filter-non-matching) > .no-filter-results): Deleted. (.sidebar > .panel.details.css-style > .content.filter-in-progress > .rules.filter-non-matching > .no-filter-results): Deleted. (.sidebar > .panel.details.css-style > .content.filter-in-progress > .rules.filter-non-matching > .no-filter-results > .no-filter-results-message): Deleted. * UserInterface/Views/RulesStyleDetailsPanel.js: (WI.R
[webkit-changes] [225486] trunk/Source/WebCore
Title: [225486] trunk/Source/WebCore
Revision 225486
Author commit-qu...@webkit.org
Date 2017-12-04 12:35:05 -0800 (Mon, 04 Dec 2017)
Log Message
Call "Terminate Service Worker" on all workers when all their clients are gone
https://bugs.webkit.org/show_bug.cgi?id=179552
Patch by Youenn Fablet on 2017-12-04
Reviewed by Chris Dumez.
Tested manually.
When SWServer detects that there is no longer any client for a given client origin,
it iterates through the running workers to terminate them if they have the given client origin.
A timer of 60 seconds is used in case a user closes a tab or a window, and reopens one with the same origin shortly after.
* workers/service/server/SWServer.cpp:
(WebCore::SWServer::findClientByIdentifier):
(WebCore::SWServer::matchAll):
(WebCore::SWServer::registerServiceWorkerClient):
(WebCore::SWServer::unregisterServiceWorkerClient):
* workers/service/server/SWServer.h:
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/workers/service/server/SWServer.cpp
trunk/Source/WebCore/workers/service/server/SWServer.h
Diff
Modified: trunk/Source/WebCore/ChangeLog (225485 => 225486)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 20:31:35 UTC (rev 225485)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 20:35:05 UTC (rev 225486)
@@ -1,3 +1,23 @@
+2017-12-04 Youenn Fablet
+
+Call "Terminate Service Worker" on all workers when all their clients are gone
+https://bugs.webkit.org/show_bug.cgi?id=179552
+
+Reviewed by Chris Dumez.
+
+Tested manually.
+
+When SWServer detects that there is no longer any client for a given client origin,
+it iterates through the running workers to terminate them if they have the given client origin.
+A timer of 60 seconds is used in case a user closes a tab or a window, and reopens one with the same origin shortly after.
+
+* workers/service/server/SWServer.cpp:
+(WebCore::SWServer::findClientByIdentifier):
+(WebCore::SWServer::matchAll):
+(WebCore::SWServer::registerServiceWorkerClient):
+(WebCore::SWServer::unregisterServiceWorkerClient):
+* workers/service/server/SWServer.h:
+
2017-12-04 Antti Koivisto
Enable selector filtering for ::before and ::after pseudo element resolution
Modified: trunk/Source/WebCore/workers/service/server/SWServer.cpp (225485 => 225486)
--- trunk/Source/WebCore/workers/service/server/SWServer.cpp 2017-12-04 20:31:35 UTC (rev 225485)
+++ trunk/Source/WebCore/workers/service/server/SWServer.cpp 2017-12-04 20:35:05 UTC (rev 225486)
@@ -52,6 +52,8 @@
return generateObjectIdentifier();
}
+static Seconds terminationDelay { 60_s };
+
SWServer::Connection::Connection(SWServer& server)
: m_server(server)
, m_identifier(generateObjectIdentifier())
@@ -322,7 +324,7 @@
if (iterator == m_clients.end())
return std::nullopt;
-auto& clients = iterator->value;
+auto& clients = iterator->value.clients;
auto position = clients.findMatching([&] (const auto& client) {
return clientIdentifier == client.identifier;
});
@@ -336,7 +338,7 @@
// FIXME: Support reserved client filtering.
// FIXME: Support WindowClient additional properties.
-auto clients = m_clients.get(worker.origin());
+auto clients = m_clients.find(worker.origin())->value.clients;
if (!options.includeUncontrolled) {
clients.removeAllMatching([&] (const auto& client) {
@@ -615,9 +617,13 @@
void SWServer::registerServiceWorkerClient(ClientOrigin&& clientOrigin, ServiceWorkerClientIdentifier identifier, ServiceWorkerClientData&& data)
{
-m_clients.ensure(WTFMove(clientOrigin), [] {
-return Vector { };
-}).iterator->value.append(ServiceWorkerClientInformation { identifier, WTFMove(data) });
+auto& clientsData = m_clients.ensure(WTFMove(clientOrigin), [] {
+return Clients { };
+}).iterator->value;
+
+clientsData.clients.append(ServiceWorkerClientInformation { identifier, WTFMove(data) });
+if (clientsData.terminateServiceWorkersTimer)
+clientsData.terminateServiceWorkersTimer = nullptr;
}
void SWServer::unregisterServiceWorkerClient(const ClientOrigin& clientOrigin, ServiceWorkerClientIdentifier identifier)
@@ -625,14 +631,23 @@
auto iterator = m_clients.find(clientOrigin);
ASSERT(iterator != m_clients.end());
-auto& clients = iterator->value;
+auto& clients = iterator->value.clients;
clients.removeFirstMatching([&] (const auto& item) {
return identifier == item.identifier;
});
-if (clients.isEmpty()) {
-// FIXME: We might want to terminate any clientOrigin related service worker.
-m_clients.remove(iterator);
-}
+
+if (!clients.isEmpty() || m_runningOrTerminatingWorkers.isEmpty())
+return;
+
+ASSERT(!iterator->value.terminateServiceWorkersTimer);
+iterator->value.terminateServiceWorkersTimer = std::make_unique(
[webkit-changes] [225485] trunk/Source/WebCore
Title: [225485] trunk/Source/WebCore
Revision 225485
Author an...@apple.com
Date 2017-12-04 12:31:35 -0800 (Mon, 04 Dec 2017)
Log Message
Enable selector filtering for ::before and ::after pseudo element resolution
https://bugs.webkit.org/show_bug.cgi?id=180364
Reviewed by Ryosuke Niwa.
Since these pseudo elements are now resolved by TreeResolver we can use selector filter for them.
This is 10%+ progression in StyleBench pseudo element subtest.
* css/StyleResolver.cpp:
(WebCore::StyleResolver::pseudoStyleForElement):
* css/StyleResolver.h:
* style/StyleTreeResolver.cpp:
(WebCore::Style::TreeResolver::resolvePseudoStyle):
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/css/StyleResolver.cpp
trunk/Source/WebCore/css/StyleResolver.h
trunk/Source/WebCore/style/StyleTreeResolver.cpp
Diff
Modified: trunk/Source/WebCore/ChangeLog (225484 => 225485)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 19:37:00 UTC (rev 225484)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 20:31:35 UTC (rev 225485)
@@ -1,3 +1,20 @@
+2017-12-04 Antti Koivisto
+
+Enable selector filtering for ::before and ::after pseudo element resolution
+https://bugs.webkit.org/show_bug.cgi?id=180364
+
+Reviewed by Ryosuke Niwa.
+
+Since these pseudo elements are now resolved by TreeResolver we can use selector filter for them.
+
+This is 10%+ progression in StyleBench pseudo element subtest.
+
+* css/StyleResolver.cpp:
+(WebCore::StyleResolver::pseudoStyleForElement):
+* css/StyleResolver.h:
+* style/StyleTreeResolver.cpp:
+(WebCore::Style::TreeResolver::resolvePseudoStyle):
+
2017-12-04 Youenn Fablet
Add support https://w3c.github.io/ServiceWorker/#on-client-unload-algorithm
Modified: trunk/Source/WebCore/css/StyleResolver.cpp (225484 => 225485)
--- trunk/Source/WebCore/css/StyleResolver.cpp 2017-12-04 19:37:00 UTC (rev 225484)
+++ trunk/Source/WebCore/css/StyleResolver.cpp 2017-12-04 20:31:35 UTC (rev 225485)
@@ -552,9 +552,9 @@
}
}
-std::unique_ptr StyleResolver::pseudoStyleForElement(const Element& element, const PseudoStyleRequest& pseudoStyleRequest, const RenderStyle& parentStyle)
+std::unique_ptr StyleResolver::pseudoStyleForElement(const Element& element, const PseudoStyleRequest& pseudoStyleRequest, const RenderStyle& parentStyle, const SelectorFilter* selectorFilter)
{
-m_state = State(element, &parentStyle);
+m_state = State(element, &parentStyle, m_overrideDocumentElementStyle, selectorFilter);
State& state = m_state;
Modified: trunk/Source/WebCore/css/StyleResolver.h (225484 => 225485)
--- trunk/Source/WebCore/css/StyleResolver.h 2017-12-04 19:37:00 UTC (rev 225484)
+++ trunk/Source/WebCore/css/StyleResolver.h 2017-12-04 20:31:35 UTC (rev 225485)
@@ -133,7 +133,7 @@
void keyframeStylesForAnimation(const Element&, const RenderStyle*, KeyframeList&);
-std::unique_ptr pseudoStyleForElement(const Element&, const PseudoStyleRequest&, const RenderStyle& parentStyle);
+std::unique_ptr pseudoStyleForElement(const Element&, const PseudoStyleRequest&, const RenderStyle& parentStyle, const SelectorFilter* = nullptr);
std::unique_ptr styleForPage(int pageIndex);
std::unique_ptr defaultStyleForElement();
Modified: trunk/Source/WebCore/style/StyleTreeResolver.cpp (225484 => 225485)
--- trunk/Source/WebCore/style/StyleTreeResolver.cpp 2017-12-04 19:37:00 UTC (rev 225484)
+++ trunk/Source/WebCore/style/StyleTreeResolver.cpp 2017-12-04 20:31:35 UTC (rev 225485)
@@ -233,7 +233,7 @@
if (!elementUpdate.style->hasPseudoStyle(pseudoId))
return { };
-auto pseudoStyle = scope().styleResolver.pseudoStyleForElement(element, { pseudoId }, *elementUpdate.style);
+auto pseudoStyle = scope().styleResolver.pseudoStyleForElement(element, { pseudoId }, *elementUpdate.style, &scope().selectorFilter);
if (!pseudoStyle)
return { };
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225484] trunk/Source/WebCore/PAL
Title: [225484] trunk/Source/WebCore/PAL
Revision 225484
Author zandober...@gmail.com
Date 2017-12-04 11:37:00 -0800 (Mon, 04 Dec 2017)
Log Message
Unreviewed WPE build fix.
* pal/PlatformWPE.cmake: Use GLIB_INCLUDE_DIRS for building the PAL library.
Modified Paths
trunk/Source/WebCore/PAL/ChangeLog
trunk/Source/WebCore/PAL/pal/PlatformWPE.cmake
Diff
Modified: trunk/Source/WebCore/PAL/ChangeLog (225483 => 225484)
--- trunk/Source/WebCore/PAL/ChangeLog 2017-12-04 18:55:07 UTC (rev 225483)
+++ trunk/Source/WebCore/PAL/ChangeLog 2017-12-04 19:37:00 UTC (rev 225484)
@@ -1,3 +1,9 @@
+2017-12-04 Zan Dobersek
+
+Unreviewed WPE build fix.
+
+* pal/PlatformWPE.cmake: Use GLIB_INCLUDE_DIRS for building the PAL library.
+
2017-12-04 Michael Catanzaro
[WPE][GTK] Implement PAL::SleepDisabler
Modified: trunk/Source/WebCore/PAL/pal/PlatformWPE.cmake (225483 => 225484)
--- trunk/Source/WebCore/PAL/pal/PlatformWPE.cmake 2017-12-04 18:55:07 UTC (rev 225483)
+++ trunk/Source/WebCore/PAL/pal/PlatformWPE.cmake 2017-12-04 19:37:00 UTC (rev 225484)
@@ -16,3 +16,7 @@
crypto/tasn1/Utilities.cpp
)
endif ()
+
+list(APPEND PAL_SYSTEM_INCLUDE_DIRECTORIES
+${GLIB_INCLUDE_DIRS}
+)
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225483] trunk/Source/WebCore
Title: [225483] trunk/Source/WebCore
Revision 225483
Author commit-qu...@webkit.org
Date 2017-12-04 10:55:07 -0800 (Mon, 04 Dec 2017)
Log Message
Add support https://w3c.github.io/ServiceWorker/#on-client-unload-algorithm
https://bugs.webkit.org/show_bug.cgi?id=180359
Patch by Youenn Fablet on 2017-12-04
Reviewed by Chris Dumez.
This algorithm is needed for implementation of claim.
As per specification, it is run for every client unload, hence why this is put in removeClientUsingRegistration.
* workers/service/server/SWServerRegistration.cpp:
(WebCore::SWServerRegistration::removeClientUsingRegistration):
(WebCore::SWServerRegistration::handleClientUnload):
* workers/service/server/SWServerRegistration.h:
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/workers/service/server/SWServerRegistration.cpp
trunk/Source/WebCore/workers/service/server/SWServerRegistration.h
Diff
Modified: trunk/Source/WebCore/ChangeLog (225482 => 225483)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 18:37:00 UTC (rev 225482)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 18:55:07 UTC (rev 225483)
@@ -1,3 +1,18 @@
+2017-12-04 Youenn Fablet
+
+Add support https://w3c.github.io/ServiceWorker/#on-client-unload-algorithm
+https://bugs.webkit.org/show_bug.cgi?id=180359
+
+Reviewed by Chris Dumez.
+
+This algorithm is needed for implementation of claim.
+As per specification, it is run for every client unload, hence why this is put in removeClientUsingRegistration.
+
+* workers/service/server/SWServerRegistration.cpp:
+(WebCore::SWServerRegistration::removeClientUsingRegistration):
+(WebCore::SWServerRegistration::handleClientUnload):
+* workers/service/server/SWServerRegistration.h:
+
2017-12-04 Antti Koivisto
Remove duplicates from selector filter hashes
Modified: trunk/Source/WebCore/workers/service/server/SWServerRegistration.cpp (225482 => 225483)
--- trunk/Source/WebCore/workers/service/server/SWServerRegistration.cpp 2017-12-04 18:37:00 UTC (rev 225482)
+++ trunk/Source/WebCore/workers/service/server/SWServerRegistration.cpp 2017-12-04 18:55:07 UTC (rev 225483)
@@ -161,6 +161,8 @@
if (iterator->value.isEmpty())
m_clientsUsingRegistration.remove(iterator);
+
+handleClientUnload();
}
// https://w3c.github.io/ServiceWorker/#notify-controller-change
@@ -281,6 +283,16 @@
updateWorkerState(*activeWorker(), ServiceWorkerState::Activated);
}
+// https://w3c.github.io/ServiceWorker/#on-client-unload-algorithm
+void SWServerRegistration::handleClientUnload()
+{
+if (hasClientsUsingRegistration())
+return;
+if (isUninstalling() && tryClear())
+return;
+tryActivate();
+}
+
} // namespace WebCore
#endif // ENABLE(SERVICE_WORKER)
Modified: trunk/Source/WebCore/workers/service/server/SWServerRegistration.h (225482 => 225483)
--- trunk/Source/WebCore/workers/service/server/SWServerRegistration.h 2017-12-04 18:37:00 UTC (rev 225482)
+++ trunk/Source/WebCore/workers/service/server/SWServerRegistration.h 2017-12-04 18:55:07 UTC (rev 225483)
@@ -90,6 +90,7 @@
void forEachConnection(const WTF::Function&);
void activate();
+void handleClientUnload();
ServiceWorkerRegistrationIdentifier m_identifier;
ServiceWorkerRegistrationKey m_registrationKey;
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [225482] trunk/Source/WebCore
Title: [225482] trunk/Source/WebCore
Revision 225482
Author an...@apple.com
Date 2017-12-04 10:37:00 -0800 (Mon, 04 Dec 2017)
Log Message
Remove duplicates from selector filter hashes
https://bugs.webkit.org/show_bug.cgi?id=180354
Reviewed by Simon Fraser.
We have only four slots for hashes in RuleSet and adding more regresses performance. To use the limited slots
better we should eliminate duplicates.
This patch also switches to using std::array instead of a C array for the hashes.
The patch reduces the number of selectors passing through the selector filter in StyleBench by ~0.4%.
* css/ElementRuleCollector.cpp:
(WebCore::ElementRuleCollector::collectMatchingRulesForList):
* css/RuleSet.cpp:
(WebCore::RuleData::RuleData):
* css/RuleSet.h:
(WebCore::RuleData::descendantSelectorIdentifierHashes const):
* css/SelectorFilter.cpp:
(WebCore::collectDescendantSelectorIdentifierHashes):
(WebCore::SelectorFilter::collectIdentifierHashes):
* css/SelectorFilter.h:
(WebCore::SelectorFilter::fastRejectSelector const):
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/css/ElementRuleCollector.cpp
trunk/Source/WebCore/css/RuleSet.cpp
trunk/Source/WebCore/css/RuleSet.h
trunk/Source/WebCore/css/SelectorFilter.cpp
trunk/Source/WebCore/css/SelectorFilter.h
Diff
Modified: trunk/Source/WebCore/ChangeLog (225481 => 225482)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 18:13:45 UTC (rev 225481)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 18:37:00 UTC (rev 225482)
@@ -1,3 +1,29 @@
+2017-12-04 Antti Koivisto
+
+Remove duplicates from selector filter hashes
+https://bugs.webkit.org/show_bug.cgi?id=180354
+
+Reviewed by Simon Fraser.
+
+We have only four slots for hashes in RuleSet and adding more regresses performance. To use the limited slots
+better we should eliminate duplicates.
+
+This patch also switches to using std::array instead of a C array for the hashes.
+
+The patch reduces the number of selectors passing through the selector filter in StyleBench by ~0.4%.
+
+* css/ElementRuleCollector.cpp:
+(WebCore::ElementRuleCollector::collectMatchingRulesForList):
+* css/RuleSet.cpp:
+(WebCore::RuleData::RuleData):
+* css/RuleSet.h:
+(WebCore::RuleData::descendantSelectorIdentifierHashes const):
+* css/SelectorFilter.cpp:
+(WebCore::collectDescendantSelectorIdentifierHashes):
+(WebCore::SelectorFilter::collectIdentifierHashes):
+* css/SelectorFilter.h:
+(WebCore::SelectorFilter::fastRejectSelector const):
+
2017-12-04 Youenn Fablet
WorkerCacheStorageConnection should handle the case of terminated workers
Modified: trunk/Source/WebCore/css/ElementRuleCollector.cpp (225481 => 225482)
--- trunk/Source/WebCore/css/ElementRuleCollector.cpp 2017-12-04 18:13:45 UTC (rev 225481)
+++ trunk/Source/WebCore/css/ElementRuleCollector.cpp 2017-12-04 18:37:00 UTC (rev 225482)
@@ -464,7 +464,7 @@
if (!ruleData.canMatchPseudoElement() && m_pseudoStyleRequest.pseudoId != NOPSEUDO)
continue;
-if (m_selectorFilter && m_selectorFilter->fastRejectSelector(ruleData.descendantSelectorIdentifierHashes()))
+if (m_selectorFilter && m_selectorFilter->fastRejectSelector(ruleData.descendantSelectorIdentifierHashes()))
continue;
StyleRule* rule = ruleData.rule();
Modified: trunk/Source/WebCore/css/RuleSet.cpp (225481 => 225482)
--- trunk/Source/WebCore/css/RuleSet.cpp 2017-12-04 18:13:45 UTC (rev 225481)
+++ trunk/Source/WebCore/css/RuleSet.cpp 2017-12-04 18:37:00 UTC (rev 225482)
@@ -164,7 +164,7 @@
{
ASSERT(m_position == position);
ASSERT(m_selectorIndex == selectorIndex);
-SelectorFilter::collectIdentifierHashes(selector(), m_descendantSelectorIdentifierHashes, maximumIdentifierCount);
+SelectorFilter::collectIdentifierHashes(*selector(), m_descendantSelectorIdentifierHashes);
}
RuleSet::RuleSet() = default;
Modified: trunk/Source/WebCore/css/RuleSet.h (225481 => 225482)
--- trunk/Source/WebCore/css/RuleSet.h 2017-12-04 18:13:45 UTC (rev 225481)
+++ trunk/Source/WebCore/css/RuleSet.h 2017-12-04 18:37:00 UTC (rev 225482)
@@ -23,6 +23,7 @@
#include "RuleFeature.h"
#include "SelectorCompiler.h"
+#include "SelectorFilter.h"
#include "StyleRule.h"
#include
#include
@@ -76,9 +77,7 @@
unsigned linkMatchType() const { return m_linkMatchType; }
bool hasDocumentSecurityOrigin() const { return m_hasDocumentSecurityOrigin; }
PropertyWhitelistType propertyWhitelistType() const { return static_cast(m_propertyWhitelistType); }
-// Try to balance between memory usage (there can be lots of RuleData objects) and good filtering performance.
-static const unsigned maximumIdentifierCount = 4;
-const unsigned* descendantSelectorIdentifierHashes() const { return m_descendantSelectorIdentifierHashes; }
+const SelectorFilter::Hashes& descendantSele
[webkit-changes] [225481] trunk/Source/WebCore
Title: [225481] trunk/Source/WebCore
Revision 225481
Author cdu...@apple.com
Date 2017-12-04 10:13:45 -0800 (Mon, 04 Dec 2017)
Log Message
WorkerCacheStorageConnection should handle the case of terminated workers
https://bugs.webkit.org/show_bug.cgi?id=180304
Patch by Youenn Fablet on 2017-12-04
Reviewed by Chris Dumez.
No web page observable change of behavior.
Reworked WorkerCacheStorageConnection hopping.
Instead of refing/unrefing itself, it refs the worker thread and the main thread connection.
This worker thread is then used on the way back from the main thread.
* Modules/cache/WorkerCacheStorageConnection.cpp:
(WebCore::WorkerCacheStorageConnection::create):
(WebCore::WorkerCacheStorageConnection::WorkerCacheStorageConnection):
(WebCore::WorkerCacheStorageConnection::doOpen):
(WebCore::WorkerCacheStorageConnection::doRemove):
(WebCore::WorkerCacheStorageConnection::doRetrieveCaches):
(WebCore::WorkerCacheStorageConnection::reference):
(WebCore::WorkerCacheStorageConnection::dereference):
(WebCore::WorkerCacheStorageConnection::doRetrieveRecords):
(WebCore::WorkerCacheStorageConnection::doBatchDeleteOperation):
(WebCore::WorkerCacheStorageConnection::doBatchPutOperation):
* Modules/cache/WorkerCacheStorageConnection.h:
* workers/WorkerGlobalScope.cpp:
(WebCore::WorkerGlobalScope::cacheStorageConnection):
* workers/WorkerGlobalScope.h:
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/Modules/cache/WorkerCacheStorageConnection.cpp
trunk/Source/WebCore/Modules/cache/WorkerCacheStorageConnection.h
trunk/Source/WebCore/workers/WorkerGlobalScope.cpp
trunk/Source/WebCore/workers/WorkerGlobalScope.h
Diff
Modified: trunk/Source/WebCore/ChangeLog (225480 => 225481)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 16:45:51 UTC (rev 225480)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 18:13:45 UTC (rev 225481)
@@ -1,3 +1,32 @@
+2017-12-04 Youenn Fablet
+
+WorkerCacheStorageConnection should handle the case of terminated workers
+https://bugs.webkit.org/show_bug.cgi?id=180304
+
+Reviewed by Chris Dumez.
+
+No web page observable change of behavior.
+
+Reworked WorkerCacheStorageConnection hopping.
+Instead of refing/unrefing itself, it refs the worker thread and the main thread connection.
+This worker thread is then used on the way back from the main thread.
+
+* Modules/cache/WorkerCacheStorageConnection.cpp:
+(WebCore::WorkerCacheStorageConnection::create):
+(WebCore::WorkerCacheStorageConnection::WorkerCacheStorageConnection):
+(WebCore::WorkerCacheStorageConnection::doOpen):
+(WebCore::WorkerCacheStorageConnection::doRemove):
+(WebCore::WorkerCacheStorageConnection::doRetrieveCaches):
+(WebCore::WorkerCacheStorageConnection::reference):
+(WebCore::WorkerCacheStorageConnection::dereference):
+(WebCore::WorkerCacheStorageConnection::doRetrieveRecords):
+(WebCore::WorkerCacheStorageConnection::doBatchDeleteOperation):
+(WebCore::WorkerCacheStorageConnection::doBatchPutOperation):
+* Modules/cache/WorkerCacheStorageConnection.h:
+* workers/WorkerGlobalScope.cpp:
+(WebCore::WorkerGlobalScope::cacheStorageConnection):
+* workers/WorkerGlobalScope.h:
+
2017-12-04 Frederic Wang
Make ScrollingTreeNode::enclosingFrameNode return the node itself for frame nodes
Modified: trunk/Source/WebCore/Modules/cache/WorkerCacheStorageConnection.cpp (225480 => 225481)
--- trunk/Source/WebCore/Modules/cache/WorkerCacheStorageConnection.cpp 2017-12-04 16:45:51 UTC (rev 225480)
+++ trunk/Source/WebCore/Modules/cache/WorkerCacheStorageConnection.cpp 2017-12-04 18:13:45 UTC (rev 225481)
@@ -91,17 +91,15 @@
Ref WorkerCacheStorageConnection::create(WorkerGlobalScope& scope)
{
auto connection = adoptRef(*new WorkerCacheStorageConnection(scope));
-callOnMainThread([protectedConnection = connection.copyRef()]() mutable {
-ASSERT(isMainThread());
-protectedConnection->m_mainThreadConnection = protectedConnection->m_proxy.createCacheStorageConnection();
+callOnMainThreadAndWait([workerThread = makeRef(scope.thread()), connection = connection.ptr()]() mutable {
+connection->m_mainThreadConnection = workerThread->workerLoaderProxy().createCacheStorageConnection();
});
+ASSERT(connection->m_mainThreadConnection);
return connection;
}
WorkerCacheStorageConnection::WorkerCacheStorageConnection(WorkerGlobalScope& scope)
: m_scope(scope)
-, m_proxy(m_scope.thread().workerLoaderProxy())
-, m_taskMode(WorkerRunLoop::defaultMode().isolatedCopy())
{
}
@@ -113,15 +111,11 @@
void WorkerCacheStorageConnection::doOpen(uint64_t requestIdentifier, const String& origin, const String& cacheName)
{
-callOnMainThread([this, protectedThis = makeRef(*this), requestIdentifier, origin = origin.isolatedCopy(), cacheName = cacheName.isolatedCopy()]() mutabl
[webkit-changes] [225480] trunk/Source
Title: [225480] trunk/Source
Revision 225480
Author fred.w...@free.fr
Date 2017-12-04 08:45:51 -0800 (Mon, 04 Dec 2017)
Log Message
Make ScrollingTreeNode::enclosingFrameNode return the node itself for frame nodes
https://bugs.webkit.org/show_bug.cgi?id=180353
Patch by Frederic Wang on 2017-12-04
Reviewed by Antonio Gomes.
This function was introduced in bug 175135 to solve a rendering bug with fixed positioned
nodes in overflow nodes. For a frame node, the reference frame to use for fixed nodes is
actually the frame node itself or otherwise we will get the same rendering bug. This patch
makes enclosingFrameNode return the node itself when it is a frame node and renames it to
clarify this behavior. Currently, enclosingFrameNode is not used for frame nodes so the
behavior is unchanged.
Source/WebCore:
No new tests, behavior is unchanged.
* page/scrolling/ScrollingTreeNode.cpp:
(WebCore::ScrollingTreeNode::enclosingFrameNodeIncludingSelf): Start the search from a frame
scrolling node from the node itself. Renamed to make explicit that this function may include
the node itself.
* page/scrolling/ScrollingTreeNode.h: Renamed.
Source/WebKit:
* UIProcess/RemoteLayerTree/ios/ScrollingTreeScrollingNodeDelegateIOS.mm:
(WebKit::ScrollingTreeScrollingNodeDelegateIOS::updateChildNodesAfterScroll): Use auto* and
the new function name.
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/page/scrolling/ScrollingTreeNode.cpp
trunk/Source/WebCore/page/scrolling/ScrollingTreeNode.h
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/RemoteLayerTree/ios/ScrollingTreeScrollingNodeDelegateIOS.mm
Diff
Modified: trunk/Source/WebCore/ChangeLog (225479 => 225480)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 16:41:49 UTC (rev 225479)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 16:45:51 UTC (rev 225480)
@@ -1,3 +1,25 @@
+2017-12-04 Frederic Wang
+
+Make ScrollingTreeNode::enclosingFrameNode return the node itself for frame nodes
+https://bugs.webkit.org/show_bug.cgi?id=180353
+
+Reviewed by Antonio Gomes.
+
+This function was introduced in bug 175135 to solve a rendering bug with fixed positioned
+nodes in overflow nodes. For a frame node, the reference frame to use for fixed nodes is
+actually the frame node itself or otherwise we will get the same rendering bug. This patch
+makes enclosingFrameNode return the node itself when it is a frame node and renames it to
+clarify this behavior. Currently, enclosingFrameNode is not used for frame nodes so the
+behavior is unchanged.
+
+No new tests, behavior is unchanged.
+
+* page/scrolling/ScrollingTreeNode.cpp:
+(WebCore::ScrollingTreeNode::enclosingFrameNodeIncludingSelf): Start the search from a frame
+scrolling node from the node itself. Renamed to make explicit that this function may include
+the node itself.
+* page/scrolling/ScrollingTreeNode.h: Renamed.
+
2017-12-04 Michael Catanzaro
[WPE][GTK] Implement PAL::SleepDisabler
Modified: trunk/Source/WebCore/page/scrolling/ScrollingTreeNode.cpp (225479 => 225480)
--- trunk/Source/WebCore/page/scrolling/ScrollingTreeNode.cpp 2017-12-04 16:41:49 UTC (rev 225479)
+++ trunk/Source/WebCore/page/scrolling/ScrollingTreeNode.cpp 2017-12-04 16:45:51 UTC (rev 225480)
@@ -77,9 +77,9 @@
ts.dumpProperty("nodeID", scrollingNodeID());
}
-ScrollingTreeFrameScrollingNode* ScrollingTreeNode::enclosingFrameNode() const
+ScrollingTreeFrameScrollingNode* ScrollingTreeNode::enclosingFrameNodeIncludingSelf()
{
-auto* node = parent();
+auto* node = this;
while (node && !node->isFrameScrollingNode())
node = node->parent();
Modified: trunk/Source/WebCore/page/scrolling/ScrollingTreeNode.h (225479 => 225480)
--- trunk/Source/WebCore/page/scrolling/ScrollingTreeNode.h 2017-12-04 16:41:49 UTC (rev 225479)
+++ trunk/Source/WebCore/page/scrolling/ScrollingTreeNode.h 2017-12-04 16:45:51 UTC (rev 225480)
@@ -66,7 +66,7 @@
void appendChild(Ref&&);
void removeChild(ScrollingTreeNode&);
-WEBCORE_EXPORT ScrollingTreeFrameScrollingNode* enclosingFrameNode() const;
+WEBCORE_EXPORT ScrollingTreeFrameScrollingNode* enclosingFrameNodeIncludingSelf();
WEBCORE_EXPORT void dump(WTF::TextStream&, ScrollingStateTreeAsTextBehavior) const;
Modified: trunk/Source/WebKit/ChangeLog (225479 => 225480)
--- trunk/Source/WebKit/ChangeLog 2017-12-04 16:41:49 UTC (rev 225479)
+++ trunk/Source/WebKit/ChangeLog 2017-12-04 16:45:51 UTC (rev 225480)
@@ -1,3 +1,21 @@
+2017-12-04 Frederic Wang
+
+Make ScrollingTreeNode::enclosingFrameNode return the node itself for frame nodes
+https://bugs.webkit.org/show_bug.cgi?id=180353
+
+Reviewed by Antonio Gomes.
+
+This function was introduced in bug 175135 to solve a rendering bug with fixed positioned
+nodes in overflow nodes. For a frame node, the refer
[webkit-changes] [225479] trunk/LayoutTests
Title: [225479] trunk/LayoutTests
Revision 225479
Author ms2...@igalia.com
Date 2017-12-04 08:41:49 -0800 (Mon, 04 Dec 2017)
Log Message
[GTK] Rebaseline imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/dialog-showModal.html
https://bugs.webkit.org/show_bug.cgi?id=180357
Unreviewed test gardening.
The difference with the platform-neutral expectation is whitespace-only.
* platform/gtk/TestExpectations:
* platform/gtk/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/dialog-showModal-expected.txt: Added.
Modified Paths
trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/gtk/TestExpectations
Added Paths
trunk/LayoutTests/platform/gtk/imported/w3c/web-platform-tests/html/semantics/interactive-elements/
trunk/LayoutTests/platform/gtk/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/
trunk/LayoutTests/platform/gtk/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/dialog-showModal-expected.txt
Diff
Modified: trunk/LayoutTests/ChangeLog (225478 => 225479)
--- trunk/LayoutTests/ChangeLog 2017-12-04 15:32:41 UTC (rev 225478)
+++ trunk/LayoutTests/ChangeLog 2017-12-04 16:41:49 UTC (rev 225479)
@@ -1,3 +1,15 @@
+2017-12-04 Ms2ger
+
+[GTK] Rebaseline imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/dialog-showModal.html
+https://bugs.webkit.org/show_bug.cgi?id=180357
+
+Unreviewed test gardening.
+
+The difference with the platform-neutral expectation is whitespace-only.
+
+* platform/gtk/TestExpectations:
+* platform/gtk/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/dialog-showModal-expected.txt: Added.
+
2017-12-04 Jer Noble
Include 'video/*' in image request Accept header if browser supports video media in image contexts.
Modified: trunk/LayoutTests/platform/gtk/TestExpectations (225478 => 225479)
--- trunk/LayoutTests/platform/gtk/TestExpectations 2017-12-04 15:32:41 UTC (rev 225478)
+++ trunk/LayoutTests/platform/gtk/TestExpectations 2017-12-04 16:41:49 UTC (rev 225479)
@@ -2924,7 +2924,6 @@
webkit.org/b/148936 imported/w3c/web-platform-tests/html/semantics/forms/textfieldselection/textfieldselection-setSelectionRange.html [ Failure ]
webkit.org/b/148936 imported/w3c/web-platform-tests/html/semantics/forms/the-form-element/form-autocomplete.html [ Failure ]
webkit.org/b/148936 imported/w3c/web-platform-tests/html/semantics/forms/the-input-element/telephone.html [ Failure ]
-webkit.org/b/148936 imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/dialog-showModal.html [ Failure ]
webkit.org/b/149128 fast/text/control-characters [ ImageOnlyFailure ]
Added: trunk/LayoutTests/platform/gtk/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/dialog-showModal-expected.txt (0 => 225479)
--- trunk/LayoutTests/platform/gtk/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/dialog-showModal-expected.txt (rev 0)
+++ trunk/LayoutTests/platform/gtk/imported/w3c/web-platform-tests/html/semantics/interactive-elements/the-dialog-element/dialog-showModal-expected.txt 2017-12-04 16:41:49 UTC (rev 225479)
@@ -0,0 +1,32 @@
+
+FAIL dialog element: showModal() assert_false: expected false got undefined
+FAIL showModal() on a that already has an open attribute throws an InvalidStateError exception assert_throws: function "function () {
+ d2.showModal();
+ this.add_cleanup(function() { d2.close(); });
+}" threw object "TypeError: d2.showModal is not a function. (In 'd2.showModal()', 'd2.showModal' is undefined)" that is not a DOMException INVALID_STATE_ERR: property "code" is equal to undefined, expected 11
+FAIL showModal() on a after initial showModal() and removing the open attribute d9.showModal is not a function. (In 'd9.showModal()', 'd9.showModal' is undefined)
+FAIL showModal() on a not in a Document throws an InvalidStateError exception assert_throws: function "function () {
+ d.showModal();
+ this.add_cleanup(function() { d.close(); });
+}" threw object "TypeError: d.showModal is not a function. (In 'd.showModal()', 'd.showModal' is undefined)" that is not a DOMException INVALID_STATE_ERR: property "code" is equal to undefined, expected 11
+FAIL when opening multiple dialogs, only the newest one is non-inert assert_false: expected false got undefined
+FAIL opening dialog without focusable children assert_false: expected false got undefined
+FAIL opening dialog with multiple focusable children assert_false: expected false got undefined
+FAIL opening dialog with multiple focusable children, one having the autofocus attribute assert_false: expected false got undefined
+FAIL when opening multiple dialogs, the most recently opened is rendered on
[webkit-changes] [225478] trunk/PerformanceTests
Title: [225478] trunk/PerformanceTests Revision 225478 Author an...@apple.com Date 2017-12-04 07:32:41 -0800 (Mon, 04 Dec 2017) Log Message Fix StyleBench/InteractiveRunner.html https://bugs.webkit.org/show_bug.cgi?id=180355 * StyleBench/InteractiveRunner.html: Modified Paths trunk/PerformanceTests/ChangeLog trunk/PerformanceTests/StyleBench/InteractiveRunner.html Diff Modified: trunk/PerformanceTests/ChangeLog (225477 => 225478) --- trunk/PerformanceTests/ChangeLog 2017-12-04 15:15:43 UTC (rev 225477) +++ trunk/PerformanceTests/ChangeLog 2017-12-04 15:32:41 UTC (rev 225478) @@ -1,3 +1,10 @@ +2017-12-04 Antti Koivisto + +Fix StyleBench/InteractiveRunner.html +https://bugs.webkit.org/show_bug.cgi?id=180355 + +* StyleBench/InteractiveRunner.html: + 2017-12-01 Antti Koivisto Add StyleBench Modified: trunk/PerformanceTests/StyleBench/InteractiveRunner.html (225477 => 225478) --- trunk/PerformanceTests/StyleBench/InteractiveRunner.html 2017-12-04 15:15:43 UTC (rev 225477) +++ trunk/PerformanceTests/StyleBench/InteractiveRunner.html 2017-12-04 15:32:41 UTC (rev 225478) @@ -3,8 +3,8 @@ Speedometer 2.0 Interactive Runner - - + +
[webkit-changes] [225477] trunk/Source
Title: [225477] trunk/Source
Revision 225477
Author mcatanz...@igalia.com
Date 2017-12-04 07:15:43 -0800 (Mon, 04 Dec 2017)
Log Message
Source/WebCore:
[WPE][GTK] Implement PAL::SleepDisabler
https://bugs.webkit.org/show_bug.cgi?id=178485
Reviewed by Carlos Garcia Campos.
Add a comment to warn about some odd use of SleepDisabler in HTMLMediaElement.
* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::shouldDisableSleep const):
Source/WebCore/PAL:
[WPE][GTK] Implement PAL::SleepDisabler
https://bugs.webkit.org/show_bug.cgi?id=178485
Reviewed by Carlos Garcia Campos.
* pal/PlatformGTK.cmake:
* pal/PlatformWPE.cmake:
* pal/system/SleepDisabler.cpp:
* pal/system/glib/SleepDisablerGLib.cpp: Added.
(PAL::SleepDisabler::create):
(PAL::SleepDisablerGLib::SleepDisablerGLib):
(PAL::SleepDisablerGLib::~SleepDisablerGLib):
(PAL::SleepDisablerGLib::acquireInhibitor):
(PAL::SleepDisablerGLib::releaseInhibitor):
* pal/system/glib/SleepDisablerGLib.h: Added.
Source/WebKit:
[GTK] Implement PAL::SleepDisabler
https://bugs.webkit.org/show_bug.cgi?id=178485
Reviewed by Carlos Garcia Campos.
* UIProcess/API/gtk/WebKitWebViewBase.cpp:
(webkitWebViewBaseDispose):
(webkitWebViewBaseEnterFullScreen):
(webkitWebViewBaseExitFullScreen):
(screenSaverInhibitedCallback): Deleted.
(webkitWebViewBaseSendInhibitMessageToScreenSaver): Deleted.
(screenSaverProxyCreatedCallback): Deleted.
(webkitWebViewBaseInhibitScreenSaver): Deleted.
(webkitWebViewBaseUninhibitScreenSaver): Deleted.
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/PAL/ChangeLog
trunk/Source/WebCore/PAL/pal/PlatformGTK.cmake
trunk/Source/WebCore/PAL/pal/PlatformWPE.cmake
trunk/Source/WebCore/PAL/pal/system/SleepDisabler.cpp
trunk/Source/WebCore/html/HTMLMediaElement.cpp
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/UIProcess/API/gtk/WebKitWebViewBase.cpp
Added Paths
trunk/Source/WebCore/PAL/pal/system/glib/
trunk/Source/WebCore/PAL/pal/system/glib/SleepDisablerGLib.cpp
trunk/Source/WebCore/PAL/pal/system/glib/SleepDisablerGLib.h
Diff
Modified: trunk/Source/WebCore/ChangeLog (225476 => 225477)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 11:34:41 UTC (rev 225476)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 15:15:43 UTC (rev 225477)
@@ -1,3 +1,15 @@
+2017-12-04 Michael Catanzaro
+
+[WPE][GTK] Implement PAL::SleepDisabler
+https://bugs.webkit.org/show_bug.cgi?id=178485
+
+Reviewed by Carlos Garcia Campos.
+
+Add a comment to warn about some odd use of SleepDisabler in HTMLMediaElement.
+
+* html/HTMLMediaElement.cpp:
+(WebCore::HTMLMediaElement::shouldDisableSleep const):
+
2017-12-04 Frederic Wang
Bug 161300 - Move RenderMathMLRoot:RootType and RenderMathMLScripts:ScriptsType to element classes
Modified: trunk/Source/WebCore/PAL/ChangeLog (225476 => 225477)
--- trunk/Source/WebCore/PAL/ChangeLog 2017-12-04 11:34:41 UTC (rev 225476)
+++ trunk/Source/WebCore/PAL/ChangeLog 2017-12-04 15:15:43 UTC (rev 225477)
@@ -1,3 +1,21 @@
+2017-12-04 Michael Catanzaro
+
+[WPE][GTK] Implement PAL::SleepDisabler
+https://bugs.webkit.org/show_bug.cgi?id=178485
+
+Reviewed by Carlos Garcia Campos.
+
+* pal/PlatformGTK.cmake:
+* pal/PlatformWPE.cmake:
+* pal/system/SleepDisabler.cpp:
+* pal/system/glib/SleepDisablerGLib.cpp: Added.
+(PAL::SleepDisabler::create):
+(PAL::SleepDisablerGLib::SleepDisablerGLib):
+(PAL::SleepDisablerGLib::~SleepDisablerGLib):
+(PAL::SleepDisablerGLib::acquireInhibitor):
+(PAL::SleepDisablerGLib::releaseInhibitor):
+* pal/system/glib/SleepDisablerGLib.h: Added.
+
2017-11-30 Stephan Szabo
Make LegacyCustomProtocolManager optional for network process
Modified: trunk/Source/WebCore/PAL/pal/PlatformGTK.cmake (225476 => 225477)
--- trunk/Source/WebCore/PAL/pal/PlatformGTK.cmake 2017-12-04 11:34:41 UTC (rev 225476)
+++ trunk/Source/WebCore/PAL/pal/PlatformGTK.cmake 2017-12-04 15:15:43 UTC (rev 225477)
@@ -3,6 +3,8 @@
system/ClockGeneric.cpp
+system/glib/SleepDisablerGLib.cpp
+
system/gtk/SoundGtk.cpp
text/KillRing.cpp
Modified: trunk/Source/WebCore/PAL/pal/PlatformWPE.cmake (225476 => 225477)
--- trunk/Source/WebCore/PAL/pal/PlatformWPE.cmake 2017-12-04 11:34:41 UTC (rev 225476)
+++ trunk/Source/WebCore/PAL/pal/PlatformWPE.cmake 2017-12-04 15:15:43 UTC (rev 225477)
@@ -4,6 +4,8 @@
system/ClockGeneric.cpp
system/Sound.cpp
+system/glib/SleepDisablerGLib.cpp
+
text/KillRing.cpp
unix/LoggingUnix.cpp
Modified: trunk/Source/WebCore/PAL/pal/system/SleepDisabler.cpp (225476 => 225477)
--- trunk/Source/WebCore/PAL/pal/system/SleepDisabler.cpp 2017-12-04 11:34:41 UTC (rev 225476)
+++ trunk/Source/WebCore/PAL/pal/system/SleepDisabler.cpp 2017-12-04 15:15:43 UTC (rev 225477)
@@ -28,12 +28,12 @@
namespace PAL {
-#if !PLATFORM(COCOA)
+#if !PLATFORM(COCOA) && !USE(GLIB)
std::un
[webkit-changes] [225476] trunk/WebDriverTests
Title: [225476] trunk/WebDriverTests
Revision 225476
Author carlo...@webkit.org
Date 2017-12-04 03:34:41 -0800 (Mon, 04 Dec 2017)
Log Message
Unreviewed. Update W3C WebDriver imported tests.
* imported/w3c/importer.json:
* imported/w3c/tools/webdriver/webdriver/client.py:
* imported/w3c/tools/webdriver/webdriver/error.py:
* imported/w3c/webdriver/tests/element_retrieval/get_active_element.py:
* imported/w3c/webdriver/tests/interaction/element_clear.py:
* imported/w3c/webdriver/tests/sessions/status.py:
* imported/w3c/webdriver/tests/support/wait.py:
Modified Paths
trunk/WebDriverTests/ChangeLog
trunk/WebDriverTests/imported/w3c/importer.json
trunk/WebDriverTests/imported/w3c/tools/webdriver/webdriver/client.py
trunk/WebDriverTests/imported/w3c/tools/webdriver/webdriver/error.py
trunk/WebDriverTests/imported/w3c/webdriver/tests/element_retrieval/get_active_element.py
trunk/WebDriverTests/imported/w3c/webdriver/tests/interaction/element_clear.py
trunk/WebDriverTests/imported/w3c/webdriver/tests/sessions/status.py
trunk/WebDriverTests/imported/w3c/webdriver/tests/support/wait.py
Diff
Modified: trunk/WebDriverTests/ChangeLog (225475 => 225476)
--- trunk/WebDriverTests/ChangeLog 2017-12-04 11:30:10 UTC (rev 225475)
+++ trunk/WebDriverTests/ChangeLog 2017-12-04 11:34:41 UTC (rev 225476)
@@ -1,3 +1,15 @@
+2017-12-04 Carlos Garcia Campos
+
+Unreviewed. Update W3C WebDriver imported tests.
+
+* imported/w3c/importer.json:
+* imported/w3c/tools/webdriver/webdriver/client.py:
+* imported/w3c/tools/webdriver/webdriver/error.py:
+* imported/w3c/webdriver/tests/element_retrieval/get_active_element.py:
+* imported/w3c/webdriver/tests/interaction/element_clear.py:
+* imported/w3c/webdriver/tests/sessions/status.py:
+* imported/w3c/webdriver/tests/support/wait.py:
+
2017-12-01 Carlos Garcia Campos
WebDriver: auto-install pytest instead of importing it from wpt tools directory
Modified: trunk/WebDriverTests/imported/w3c/importer.json (225475 => 225476)
--- trunk/WebDriverTests/imported/w3c/importer.json 2017-12-04 11:30:10 UTC (rev 225475)
+++ trunk/WebDriverTests/imported/w3c/importer.json 2017-12-04 11:34:41 UTC (rev 225476)
@@ -1,6 +1,6 @@
{
"repository": "https://github.com/w3c/web-platform-tests.git",
-"revision": "2b50389ee72d89dd0be12bc6ca54a6e95c98d163",
+"revision": "267d6b27ffd416577702653c0987b30239daae3f",
"paths_to_import": [
"tools/webdriver",
"tools/wptrunner",
Modified: trunk/WebDriverTests/imported/w3c/tools/webdriver/webdriver/client.py (225475 => 225476)
--- trunk/WebDriverTests/imported/w3c/tools/webdriver/webdriver/client.py 2017-12-04 11:30:10 UTC (rev 225475)
+++ trunk/WebDriverTests/imported/w3c/tools/webdriver/webdriver/client.py 2017-12-04 11:34:41 UTC (rev 225476)
@@ -42,7 +42,7 @@
def _set(self, key, secs):
body = {key: secs * 1000}
timeouts = self.session.send_session_command("POST", "timeouts", body)
-return timeouts[key]
+return None
@property
def script(self):
Modified: trunk/WebDriverTests/imported/w3c/tools/webdriver/webdriver/error.py (225475 => 225476)
--- trunk/WebDriverTests/imported/w3c/tools/webdriver/webdriver/error.py 2017-12-04 11:30:10 UTC (rev 225475)
+++ trunk/WebDriverTests/imported/w3c/tools/webdriver/webdriver/error.py 2017-12-04 11:34:41 UTC (rev 225476)
@@ -8,6 +8,7 @@
def __init__(self, message, stacktrace=None):
super(WebDriverException, self)
+self.message = message
self.stacktrace = stacktrace
def __repr__(self):
@@ -14,11 +15,13 @@
return "<%s http_status=%s>" % (self.__class__.__name__, self.http_status)
def __str__(self):
-return ("%s (%s)\n"
-"\n"
+message = "%s (%s): %s\n" % (self.status_code, self.http_status, self.message)
+if self.stacktrace:
+message += ("\n"
"Remote-end stacktrace:\n"
"\n"
-"%s" % (self.status_code, self.http_status, self.stacktrace))
+"%s" % self.stacktrace)
+return message
class ElementClickInterceptedException(WebDriverException):
Modified: trunk/WebDriverTests/imported/w3c/webdriver/tests/element_retrieval/get_active_element.py (225475 => 225476)
--- trunk/WebDriverTests/imported/w3c/webdriver/tests/element_retrieval/get_active_element.py 2017-12-04 11:30:10 UTC (rev 225475)
+++ trunk/WebDriverTests/imported/w3c/webdriver/tests/element_retrieval/get_active_element.py 2017-12-04 11:34:41 UTC (rev 225476)
@@ -69,7 +69,7 @@
response = get_active_element(session)
assert_is_active_element(session, response)
assert_dialog_handled(session, "dismiss #2")
-assert read_global(session, "dismiss2") is None
+assert read_global(session, "dismiss2") is False
create_dialog(session)("prompt", text="dismiss #3", result_var="dismiss3")
Modified: trunk/WebDriverTes
[webkit-changes] [225475] trunk/Source/WebCore
Title: [225475] trunk/Source/WebCore Revision 225475 Author fred.w...@free.fr Date 2017-12-04 03:30:10 -0800 (Mon, 04 Dec 2017) Log Message Bug 161300 - Move RenderMathMLRoot:RootType and RenderMathMLScripts:ScriptsType to element classes https://bugs.webkit.org/show_bug.cgi?id=161300 Patch by Frederic Wang on 2017-12-04 Reviewed by Darin Adler. RenderMathMLRoot::m_kind and RenderMathMLScripts::m_scriptType are initialized in the constructors of the renderer classes from the tag name of the corresponding elements. This patch moves them into the corresponding element classes and makes them const members parsed in the constructors. It also introduces a MathMLRootElement class deriving from MathMLRowElement to store the new member. Finally, the types are redefined as enum classes. No new tests, behavior unchanged and already covered by existing tests. * Sources.txt: Add MathMLRootElement files. * WebCore.xcodeproj/project.pbxproj: Ditto. * mathml/MathMLAllInOne.cpp: Ditto. * mathml/MathMLRootElement.cpp: (WebCore::element): Override this function to downcast to MathMLRootElement. (WebCore::rootTypeOf): Helper function to parse the root type, moved from RenderMathMLRoot. (WebCore::MathMLRootElement::MathMLRootElement): Initialize the root type. (WebCore::MathMLRootElement::create): (WebCore::MathMLRootElement::createElementRenderer): Moved from MathMLRowElement. * mathml/MathMLRootElement.h: New class deriving from MathMLRowElement, with a RootType member exposed to the renderer class. * mathml/MathMLRowElement.cpp: Remove header for RenderMathMLRoot.h. (WebCore::MathMLRowElement::createElementRenderer): Moved to MathMLRootElement. * mathml/MathMLScriptsElement.cpp: Introduce a script type. (WebCore::scriptTypeOf): Helper function to parse the script type, moved from RenerMathMLScripts. (WebCore::MathMLScriptsElement::MathMLScriptsElement): Initialize the script type. * mathml/MathMLScriptsElement.h: Add new script type member. (WebCore::MathMLScriptsElement::scriptType const): Expose the script type to the renderer class. * mathml/mathtags.in: Map and to MathMLRootElement. * rendering/mathml/RenderMathMLRoot.cpp: (WebCore::RenderMathMLRoot::RenderMathMLRoot): Use MathMLRootElement and remove parsing of the root type. (WebCore::RenderMathMLRoot::rootType const): Helper function to access the root type from the element class. (WebCore::RenderMathMLRoot::isValid const): Use rootType() and add prefix for enum class values. (WebCore::RenderMathMLRoot::getBase const): Ditto. (WebCore::RenderMathMLRoot::getIndex const): Ditto. (WebCore::RenderMathMLRoot::horizontalParameters): Ditto. (WebCore::RenderMathMLRoot::verticalParameters): Ditto. (WebCore::RenderMathMLRoot::computePreferredLogicalWidths): Ditto. (WebCore::RenderMathMLRoot::layoutBlock): Ditto. (WebCore::RenderMathMLRoot::paint): Ditto. * rendering/mathml/RenderMathMLRoot.h: Define root type as an enum class, replace MathMLRowElement with MathMLRootElement, declare and use new rootType() function and remove the m_kind member. * rendering/mathml/RenderMathMLScripts.cpp: (WebCore::RenderMathMLScripts::RenderMathMLScripts): Remove parsing of script type. (WebCore::RenderMathMLScripts::scriptType const): Helper function to access the script type from the element class. (WebCore::RenderMathMLScripts::validateAndGetReferenceChildren): Use scriptType() and add prefix for enum class values. (WebCore::RenderMathMLScripts::computePreferredLogicalWidths): Ditto. (WebCore::RenderMathMLScripts::verticalMetrics): Ditto. (WebCore::RenderMathMLScripts::layoutBlock): Ditto. * rendering/mathml/RenderMathMLScripts.h: Define root type as an enum class, declare the scriptType() function and remove the m_scriptType member. * rendering/mathml/RenderMathMLUnderOver.cpp: (WebCore::RenderMathMLUnderOver::isValid const): Use scriptType() and add prefix for enum class values. (WebCore::RenderMathMLUnderOver::under const): Ditto. (WebCore::RenderMathMLUnderOver::over const): Ditto. (WebCore::RenderMathMLUnderOver::computePreferredLogicalWidths): Ditto. (WebCore::RenderMathMLUnderOver::hasAccent const): Ditto. (WebCore::RenderMathMLUnderOver::layoutBlock): Ditto. Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/Sources.txt trunk/Source/WebCore/WebCore.xcodeproj/project.pbxproj trunk/Source/WebCore/mathml/MathMLAllInOne.cpp trunk/Source/WebCore/mathml/MathMLRowElement.cpp trunk/Source/WebCore/mathml/MathMLScriptsElement.cpp trunk/Source/WebCore/mathml/MathMLScriptsElement.h trunk/Source/WebCore/mathml/mathtags.in trunk/Source/WebCore/rendering/mathml/RenderMathMLRoot.cpp trunk/Source/WebCore/rendering/mathml/RenderMathMLRoot.h trunk/Source/WebCore/rendering/mathml/RenderMathMLScripts.cpp trunk/Source/WebCore/rendering/mathml/RenderMathMLScripts.h trunk/Source/WebCore/rendering/mathml/RenderMathMLUnderOver.cpp Added Paths trunk/Source/WebCore/mathml/MathMLRootElement.cpp trunk/Source/WebCore/mathml/MathMLRootElement.h Diff Modified: trunk/Source
[webkit-changes] [225474] trunk/Source/WebDriver
Title: [225474] trunk/Source/WebDriver
Revision 225474
Author carlo...@webkit.org
Date 2017-12-04 03:29:29 -0800 (Mon, 04 Dec 2017)
Log Message
WebDriver: implement element property command
https://bugs.webkit.org/show_bug.cgi?id=180244
Reviewed by Brian Burg.
13.3 Get Element Property
https://w3c.github.io/webdriver/webdriver-spec.html#get-element-property
Fixes: imported/w3c/webdriver/tests/state/get_element_property.py::test_no_browsing_context
imported/w3c/webdriver/tests/state/get_element_property.py::test_handle_prompt_dismiss
imported/w3c/webdriver/tests/state/get_element_property.py::test_handle_prompt_accept
imported/w3c/webdriver/tests/state/get_element_property.py::test_handle_prompt_missing_value
imported/w3c/webdriver/tests/state/get_element_property.py::test_element_stale
* Session.cpp:
(WebDriver::Session::getElementAttribute):
(WebDriver::Session::getElementProperty):
* Session.h:
* WebDriverService.cpp:
(WebDriver::WebDriverService::getElementProperty):
* WebDriverService.h:
Modified Paths
trunk/Source/WebDriver/ChangeLog
trunk/Source/WebDriver/Session.cpp
trunk/Source/WebDriver/Session.h
trunk/Source/WebDriver/WebDriverService.cpp
trunk/Source/WebDriver/WebDriverService.h
Diff
Modified: trunk/Source/WebDriver/ChangeLog (225473 => 225474)
--- trunk/Source/WebDriver/ChangeLog 2017-12-04 08:52:21 UTC (rev 225473)
+++ trunk/Source/WebDriver/ChangeLog 2017-12-04 11:29:29 UTC (rev 225474)
@@ -1,3 +1,27 @@
+2017-12-04 Carlos Garcia Campos
+
+WebDriver: implement element property command
+https://bugs.webkit.org/show_bug.cgi?id=180244
+
+Reviewed by Brian Burg.
+
+13.3 Get Element Property
+https://w3c.github.io/webdriver/webdriver-spec.html#get-element-property
+
+Fixes: imported/w3c/webdriver/tests/state/get_element_property.py::test_no_browsing_context
+ imported/w3c/webdriver/tests/state/get_element_property.py::test_handle_prompt_dismiss
+ imported/w3c/webdriver/tests/state/get_element_property.py::test_handle_prompt_accept
+ imported/w3c/webdriver/tests/state/get_element_property.py::test_handle_prompt_missing_value
+ imported/w3c/webdriver/tests/state/get_element_property.py::test_element_stale
+
+* Session.cpp:
+(WebDriver::Session::getElementAttribute):
+(WebDriver::Session::getElementProperty):
+* Session.h:
+* WebDriverService.cpp:
+(WebDriver::WebDriverService::getElementProperty):
+* WebDriverService.h:
+
2017-12-02 Carlos Garcia Campos
WebDriver: handle user prompts shown while executing scripts
Modified: trunk/Source/WebDriver/Session.cpp (225473 => 225474)
--- trunk/Source/WebDriver/Session.cpp 2017-12-04 08:52:21 UTC (rev 225473)
+++ trunk/Source/WebDriver/Session.cpp 2017-12-04 11:29:29 UTC (rev 225474)
@@ -1289,6 +1289,47 @@
});
}
+void Session::getElementProperty(const String& elementID, const String& property, Function&& completionHandler)
+{
+if (!m_toplevelBrowsingContext) {
+completionHandler(CommandResult::fail(CommandResult::ErrorCode::NoSuchWindow));
+return;
+}
+
+handleUserPrompts([this, elementID, property, completionHandler = WTFMove(completionHandler)](CommandResult&& result) mutable {
+if (result.isError()) {
+completionHandler(WTFMove(result));
+return;
+}
+RefPtr arguments = JSON::Array::create();
+arguments->pushString(createElement(elementID)->toJSONString());
+
+RefPtr parameters = JSON::Object::create();
+parameters->setString(ASCIILiteral("browsingContextHandle"), m_toplevelBrowsingContext.value());
+if (m_currentBrowsingContext)
+parameters->setString(ASCIILiteral("frameHandle"), m_currentBrowsingContext.value());
+parameters->setString(ASCIILiteral("function"), makeString("function(element) { return element.", property, "; }"));
+parameters->setArray(ASCIILiteral("arguments"), WTFMove(arguments));
+m_host->sendCommandToBackend(ASCIILiteral("evaluateJavaScriptFunction"), WTFMove(parameters), [this, protectedThis = makeRef(*this), completionHandler = WTFMove(completionHandler)](SessionHost::CommandResponse&& response) {
+if (response.isError || !response.responseObject) {
+completionHandler(CommandResult::fail(WTFMove(response.responseObject)));
+return;
+}
+String valueString;
+if (!response.responseObject->getString(ASCIILiteral("result"), valueString)) {
+completionHandler(CommandResult::fail(CommandResult::ErrorCode::UnknownError));
+return;
+}
+RefPtr resultValue;
+if (!JSON::Value::parseJSON(valueString, resultValue)) {
+completionHandler(CommandResult::fail(CommandResult::ErrorCode::UnknownError));
+
[webkit-changes] [225473] trunk/Source/WebCore
Title: [225473] trunk/Source/WebCore
Revision 225473
Author carlo...@webkit.org
Date 2017-12-04 00:52:21 -0800 (Mon, 04 Dec 2017)
Log Message
[GStreamer] More leaks in TextCombinerGStreamer
https://bugs.webkit.org/show_bug.cgi?id=180331
Reviewed by Michael Catanzaro.
I've noticed more leaks reading the code, but I couldn't check them with valgrind, because I don't know how to
trigger that code path. This patch cleans up the file using smart pointer everywhere.
* platform/graphics/gstreamer/TextCombinerGStreamer.cpp:
(webkit_text_combiner_pad_init):
(webkitTextCombinerPadEvent):
(webkitTextCombinerReleasePad):
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/gstreamer/TextCombinerGStreamer.cpp
Diff
Modified: trunk/Source/WebCore/ChangeLog (225472 => 225473)
--- trunk/Source/WebCore/ChangeLog 2017-12-04 08:49:26 UTC (rev 225472)
+++ trunk/Source/WebCore/ChangeLog 2017-12-04 08:52:21 UTC (rev 225473)
@@ -1,3 +1,18 @@
+2017-12-04 Carlos Garcia Campos
+
+[GStreamer] More leaks in TextCombinerGStreamer
+https://bugs.webkit.org/show_bug.cgi?id=180331
+
+Reviewed by Michael Catanzaro.
+
+I've noticed more leaks reading the code, but I couldn't check them with valgrind, because I don't know how to
+trigger that code path. This patch cleans up the file using smart pointer everywhere.
+
+* platform/graphics/gstreamer/TextCombinerGStreamer.cpp:
+(webkit_text_combiner_pad_init):
+(webkitTextCombinerPadEvent):
+(webkitTextCombinerReleasePad):
+
2017-12-04 Jer Noble
Include 'video/*' in image request Accept header if browser supports video media in image contexts.
Modified: trunk/Source/WebCore/platform/graphics/gstreamer/TextCombinerGStreamer.cpp (225472 => 225473)
--- trunk/Source/WebCore/platform/graphics/gstreamer/TextCombinerGStreamer.cpp 2017-12-04 08:49:26 UTC (rev 225472)
+++ trunk/Source/WebCore/platform/graphics/gstreamer/TextCombinerGStreamer.cpp 2017-12-04 08:52:21 UTC (rev 225473)
@@ -94,8 +94,6 @@
static void webkit_text_combiner_pad_init(WebKitTextCombinerPad* pad)
{
-pad->tags = 0;
-
gst_pad_set_event_function(GST_PAD(pad), webkitTextCombinerPadEvent);
}
@@ -137,17 +135,17 @@
gst_event_parse_caps(event, &caps);
ASSERT(caps);
-GstPad* target = gst_ghost_pad_get_target(GST_GHOST_PAD(pad));
+GRefPtr target = adoptGRef(gst_ghost_pad_get_target(GST_GHOST_PAD(pad)));
ASSERT(target);
-GstElement* targetParent = gst_pad_get_parent_element(target);
+GRefPtr targetParent = adoptGRef(gst_pad_get_parent_element(target.get()));
ASSERT(targetParent);
-GstCaps* textCaps = gst_caps_new_empty_simple("text/x-raw");
-if (gst_caps_can_intersect(textCaps, caps)) {
+GRefPtr textCaps = adoptGRef(gst_caps_new_empty_simple("text/x-raw"));
+if (gst_caps_can_intersect(textCaps.get(), caps)) {
/* Caps are plain text, put a WebVTT encoder between the ghostpad and
* the funnel */
-if (targetParent == combiner->funnel) {
+if (targetParent.get() == combiner->funnel) {
/* Setup a WebVTT encoder */
GstElement* encoder = gst_element_factory_make("webvttenc", nullptr);
ASSERT(encoder);
@@ -159,45 +157,38 @@
ASSERT(ret);
/* Switch the ghostpad to target the WebVTT encoder */
-GstPad* sinkPad = gst_element_get_static_pad(encoder, "sink");
+GRefPtr sinkPad = adoptGRef(gst_element_get_static_pad(encoder, "sink"));
ASSERT(sinkPad);
-ret = gst_ghost_pad_set_target(GST_GHOST_PAD(pad), sinkPad);
+ret = gst_ghost_pad_set_target(GST_GHOST_PAD(pad), sinkPad.get());
ASSERT(ret);
-gst_object_unref(sinkPad);
/* Connect the WebVTT encoder to the funnel */
-GstPad* srcPad = gst_element_get_static_pad(encoder, "src");
+GRefPtr srcPad = adoptGRef(gst_element_get_static_pad(encoder, "src"));
ASSERT(srcPad);
-ret = GST_PAD_LINK_SUCCESSFUL(gst_pad_link(srcPad, target));
+ret = GST_PAD_LINK_SUCCESSFUL(gst_pad_link(srcPad.get(), target.get()));
ASSERT(ret);
-gst_object_unref(srcPad);
} /* else: pipeline is already correct */
} else {
/* Caps are not plain text, remove the WebVTT encoder */
-if (targetParent != combiner->funnel) {
+if (targetParent.get() != combiner->funnel) {
/* Get the funnel sink pad */
-GstPad* srcPad = gst_element_get_static_pad(targetParent, "src");
+GRefPtr srcPad = adoptGRef(gst_element_get_static_pad(targetParent.get(), "src"));
ASSERT(srcPad);
-GstP
[webkit-changes] [225472] trunk
Title: [225472] trunk Revision 225472 Author jer.no...@apple.com Date 2017-12-04 00:49:26 -0800 (Mon, 04 Dec 2017) Log Message Include 'video/*' in image request Accept header if browser supports video media in image contexts. https://bugs.webkit.org/show_bug.cgi?id=179178 Reviewed by Alex Christensen. Source/WebCore: Test: http/tests/images/image-supports-video.html Ensure that 'video/*' is included in the Accept header by asking the ImageDecoder whether "Video" media is supported. The ImageDecoder will ask all it's constituent decoders, which aside from ImageDecoderAVFObjC will say only support the "Image" media type. Drive-by fix: Make sure the MIMETypeRegistry includes all the types supported by AVFoundation when ImageDecoderAVFObjC is enabled. Also, now that Image reports it can support video mime types, check whether MediaDocument can support a given mime type before asking ImageDocument. * html/HTMLImageElement.cpp: (WebCore::HTMLImageElement::bestFitSourceFromPictureElement): * loader/LinkLoader.cpp: (WebCore::LinkLoader::isSupportedType): * loader/cache/CachedResourceRequest.cpp: (WebCore::acceptHeaderValueFromType): * platform/MIMETypeRegistry.cpp: (WebCore::MIMETypeRegistry::isSupportedImageVideoOrSVGMIMEType): (WebCore::MIMETypeRegistry::isSupportedImageOrSVGMIMEType): Deleted. * platform/MIMETypeRegistry.h: * platform/graphics/ImageDecoder.cpp: (WebCore::ImageDecoder::supportsMediaType): * platform/graphics/ImageDecoder.h: * platform/graphics/avfoundation/objc/ImageDecoderAVFObjC.h: (WebCore::ImageDecoderAVFObjC::supportsMediaType): * platform/graphics/avfoundation/objc/ImageDecoderAVFObjC.mm: (WebCore::ImageDecoderAVFObjC::supportsContentType): * platform/graphics/cg/ImageDecoderCG.h: * platform/graphics/win/ImageDecoderDirect2D.h: * platform/image-decoders/ScalableImageDecoder.h: (WebCore::ScalableImageDecoder::supportsMediaType): * html/parser/HTMLPreloadScanner.cpp: (WebCore::TokenPreloadScanner::StartTagScanner::processAttribute): LayoutTests: * http/tests/images/image-supports-video-expected.txt: Added. * http/tests/images/image-supports-video.html: Added. * http/tests/misc/resources/image-checks-for-accept.php: * http/tests/resources/redirect-to-video-if-accepted.php: Added. * platform/ios/TestExpectations: * platform/mac/imported/w3c/web-platform-tests/html/semantics/the-img-element/update-the-source-set-expected.txt: Added. Modified Paths trunk/LayoutTests/ChangeLog trunk/LayoutTests/http/tests/misc/resources/image-checks-for-accept.php trunk/LayoutTests/platform/ios/TestExpectations trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/html/HTMLImageElement.cpp trunk/Source/WebCore/html/parser/HTMLPreloadScanner.cpp trunk/Source/WebCore/loader/LinkLoader.cpp trunk/Source/WebCore/loader/cache/CachedResourceRequest.cpp trunk/Source/WebCore/platform/MIMETypeRegistry.cpp trunk/Source/WebCore/platform/MIMETypeRegistry.h trunk/Source/WebCore/platform/graphics/ImageDecoder.cpp trunk/Source/WebCore/platform/graphics/ImageDecoder.h trunk/Source/WebCore/platform/graphics/avfoundation/objc/ImageDecoderAVFObjC.h trunk/Source/WebCore/platform/graphics/avfoundation/objc/ImageDecoderAVFObjC.mm trunk/Source/WebCore/platform/graphics/cg/ImageDecoderCG.h trunk/Source/WebCore/platform/graphics/win/ImageDecoderDirect2D.h trunk/Source/WebCore/platform/image-decoders/ScalableImageDecoder.h Added Paths trunk/LayoutTests/http/tests/images/image-supports-video-expected.txt trunk/LayoutTests/http/tests/images/image-supports-video.html trunk/LayoutTests/http/tests/resources/redirect-to-video-if-accepted.php trunk/LayoutTests/platform/mac/imported/w3c/web-platform-tests/html/semantics/ trunk/LayoutTests/platform/mac/imported/w3c/web-platform-tests/html/semantics/embedded-content/ trunk/LayoutTests/platform/mac/imported/w3c/web-platform-tests/html/semantics/embedded-content/the-img-element/ trunk/LayoutTests/platform/mac/imported/w3c/web-platform-tests/html/semantics/embedded-content/the-img-element/update-the-source-set-expected.txt Diff Modified: trunk/LayoutTests/ChangeLog (225471 => 225472) --- trunk/LayoutTests/ChangeLog 2017-12-04 06:26:13 UTC (rev 225471) +++ trunk/LayoutTests/ChangeLog 2017-12-04 08:49:26 UTC (rev 225472) @@ -1,3 +1,17 @@ +2017-12-04 Jer Noble + +Include 'video/*' in image request Accept header if browser supports video media in image contexts. +https://bugs.webkit.org/show_bug.cgi?id=179178 + +Reviewed by Alex Christensen. + +* http/tests/images/image-supports-video-expected.txt: Added. +* http/tests/images/image-supports-video.html: Added. +* http/tests/misc/resources/image-checks-for-accept.php: +* http/tests/resources/redirect-to-video-if-accepted.php: Added. +* platform/ios/TestExpectations: +* platform/mac/imported/w3c/web-platform-tests/html/semantics/the-img-element/update-the-source-set-expected.txt: Added. + 2017-12-03 Chris Dumez Re-sync Service Workers web-plat
