[webkit-changes] [248933] trunk/Source/JavaScriptCore
Title: [248933] trunk/Source/_javascript_Core
Revision 248933
Author drou...@apple.com
Date 2019-08-20 22:25:34 -0700 (Tue, 20 Aug 2019)
Log Message
Unreviewed, speculative build fix for High Sierra after r248925
* inspector/JSInjectedScriptHost.cpp:
(Inspector::HeapHolderFinder::dump):
Modified Paths
trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/inspector/JSInjectedScriptHost.cpp
Diff
Modified: trunk/Source/_javascript_Core/ChangeLog (248932 => 248933)
--- trunk/Source/_javascript_Core/ChangeLog 2019-08-21 01:41:51 UTC (rev 248932)
+++ trunk/Source/_javascript_Core/ChangeLog 2019-08-21 05:25:34 UTC (rev 248933)
@@ -1,3 +1,10 @@
+2019-08-20 Devin Rousso
+
+Unreviewed, speculative build fix for High Sierra after r248925
+
+* inspector/JSInjectedScriptHost.cpp:
+(Inspector::HeapHolderFinder::dump):
+
2019-08-20 Mark Lam
Remove superfluous size argument to allocateCell() for fixed size objects.
Modified: trunk/Source/_javascript_Core/inspector/JSInjectedScriptHost.cpp (248932 => 248933)
--- trunk/Source/_javascript_Core/inspector/JSInjectedScriptHost.cpp 2019-08-21 01:41:51 UTC (rev 248932)
+++ trunk/Source/_javascript_Core/inspector/JSInjectedScriptHost.cpp 2019-08-21 05:25:34 UTC (rev 248933)
@@ -831,6 +831,7 @@
void setWrappedObjectForCell(JSCell*, void*) { }
void setLabelForCell(JSCell*, const String&) { }
+#ifndef NDEBUG
void dump(PrintStream& out) const
{
Indentation<4> indent;
@@ -860,7 +861,7 @@
out.println();
if (isFirstVisit) {
-IndentationScope scope(indent);
+IndentationScope<4> scope(indent);
for (auto* to : m_successors.get(from))
visit(to);
}
@@ -869,6 +870,7 @@
for (auto* from : m_rootsToInclude)
visit(from);
}
+#endif
private:
Lock m_mutex;
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248932] trunk/Source/WebCore
Title: [248932] trunk/Source/WebCore
Revision 248932
Author rn...@webkit.org
Date 2019-08-20 18:41:51 -0700 (Tue, 20 Aug 2019)
Log Message
nextElementWithGreaterTabIndex should use shadowAdjustedTabIndex
https://bugs.webkit.org/show_bug.cgi?id=200943
Reviewed by Wenson Hsieh.
Use shadowAdjustedTabIndex instead of tabIndexForElement in nextElementWithGreaterTabIndex.
Because nextElementWithGreaterTabIndex is only called to find an element with a tab index
set to a greater value than the argument tabIndex, which is always a positive integer,
whether we call shadowAdjustedTabIndex or tabIndexForElement has no effect on an element
with a shadow root or on a slot element. Furthermore, none of the elements with
an isKeyboardFocusable override can have a author defined shadow root attached either.
As a result, this refactoring will have no observable behavioral change.
No new tests since there should be no observable behavior change.
* page/FocusController.cpp:
(WebCore::tabIndexForElement): Merged into shadowAdjustedTabIndex.
(WebCore::shadowAdjustedTabIndex):
(WebCore::nextElementWithGreaterTabIndex):
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/page/FocusController.cpp
Diff
Modified: trunk/Source/WebCore/ChangeLog (248931 => 248932)
--- trunk/Source/WebCore/ChangeLog 2019-08-21 01:39:54 UTC (rev 248931)
+++ trunk/Source/WebCore/ChangeLog 2019-08-21 01:41:51 UTC (rev 248932)
@@ -1,5 +1,29 @@
2019-08-20 Ryosuke Niwa
+nextElementWithGreaterTabIndex should use shadowAdjustedTabIndex
+https://bugs.webkit.org/show_bug.cgi?id=200943
+
+Reviewed by Wenson Hsieh.
+
+Use shadowAdjustedTabIndex instead of tabIndexForElement in nextElementWithGreaterTabIndex.
+
+Because nextElementWithGreaterTabIndex is only called to find an element with a tab index
+set to a greater value than the argument tabIndex, which is always a positive integer,
+whether we call shadowAdjustedTabIndex or tabIndexForElement has no effect on an element
+with a shadow root or on a slot element. Furthermore, none of the elements with
+an isKeyboardFocusable override can have a author defined shadow root attached either.
+
+As a result, this refactoring will have no observable behavioral change.
+
+No new tests since there should be no observable behavior change.
+
+* page/FocusController.cpp:
+(WebCore::tabIndexForElement): Merged into shadowAdjustedTabIndex.
+(WebCore::shadowAdjustedTabIndex):
+(WebCore::nextElementWithGreaterTabIndex):
+
+2019-08-20 Ryosuke Niwa
+
Remove MathMLElement::defaultTabIndex()
https://bugs.webkit.org/show_bug.cgi?id=200944
Modified: trunk/Source/WebCore/page/FocusController.cpp (248931 => 248932)
--- trunk/Source/WebCore/page/FocusController.cpp 2019-08-21 01:39:54 UTC (rev 248931)
+++ trunk/Source/WebCore/page/FocusController.cpp 2019-08-21 01:41:51 UTC (rev 248932)
@@ -325,12 +325,6 @@
return element.isKeyboardFocusable(event) && isFocusScopeOwner(element);
}
-// FIXME: This function should be merged into shadowAdjustedTabIndex.
-static inline int tabIndexForElement(const Element& element)
-{
-return element.shouldBeIgnoredInSequentialFocusNavigation() ? -1 : element.tabIndexSetExplicitly().valueOr(0);
-}
-
static inline int shadowAdjustedTabIndex(Element& element, KeyboardEvent* event)
{
if (isNonFocusableScopeOwner(element, event)) {
@@ -337,7 +331,7 @@
if (!element.tabIndexSetExplicitly())
return 0; // Treat a shadow host without tabindex if it has tabindex=0 even though HTMLElement::tabIndex returns -1 on such an element.
}
-return tabIndexForElement(element);
+return element.shouldBeIgnoredInSequentialFocusNavigation() ? -1 : element.tabIndexSetExplicitly().valueOr(0);
}
FocusController::FocusController(Page& page, OptionSet activityState)
@@ -626,8 +620,7 @@
if (!is(*node))
continue;
Element& candidate = downcast(*node);
-// FIXME: We should be calling shadowAdjustedTabIndex instead.
-int candidateTabIndex = tabIndexForElement(candidate);
+int candidateTabIndex = shadowAdjustedTabIndex(candidate, event);
if (isFocusableElementOrScopeOwner(candidate, event) && candidateTabIndex > tabIndex && (!winner || candidateTabIndex < winningTabIndex)) {
winner =
winningTabIndex = candidateTabIndex;
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248931] trunk/Source/WebCore
Title: [248931] trunk/Source/WebCore
Revision 248931
Author rn...@webkit.org
Date 2019-08-20 18:39:54 -0700 (Tue, 20 Aug 2019)
Log Message
Remove MathMLElement::defaultTabIndex()
https://bugs.webkit.org/show_bug.cgi?id=200944
Reviewed by Frédéric Wang.
Removed MathMLElement::defaultTabIndex since it doesn't have any observable effect since
MathMLElement doesn't expose tabIndex IDL attribute, and sequential (tab key) focus navigation
checks MathMLElement::isKeyboardFocusable, which doesn't rely on defaultTabIndex, anyway.
* mathml/MathMLElement.cpp:
(WebCore::MathMLElement::defaultTabIndex const): Deleted.
* mathml/MathMLElement.h:
Modified Paths
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/mathml/MathMLElement.cpp
trunk/Source/WebCore/mathml/MathMLElement.h
Diff
Modified: trunk/Source/WebCore/ChangeLog (248930 => 248931)
--- trunk/Source/WebCore/ChangeLog 2019-08-21 01:36:41 UTC (rev 248930)
+++ trunk/Source/WebCore/ChangeLog 2019-08-21 01:39:54 UTC (rev 248931)
@@ -1,3 +1,18 @@
+2019-08-20 Ryosuke Niwa
+
+Remove MathMLElement::defaultTabIndex()
+https://bugs.webkit.org/show_bug.cgi?id=200944
+
+Reviewed by Frédéric Wang.
+
+Removed MathMLElement::defaultTabIndex since it doesn't have any observable effect since
+MathMLElement doesn't expose tabIndex IDL attribute, and sequential (tab key) focus navigation
+checks MathMLElement::isKeyboardFocusable, which doesn't rely on defaultTabIndex, anyway.
+
+* mathml/MathMLElement.cpp:
+(WebCore::MathMLElement::defaultTabIndex const): Deleted.
+* mathml/MathMLElement.h:
+
2019-08-20 Devin Rousso
Web Inspector: Implement `queryHolders` Command Line API
Modified: trunk/Source/WebCore/mathml/MathMLElement.cpp (248930 => 248931)
--- trunk/Source/WebCore/mathml/MathMLElement.cpp 2019-08-21 01:36:41 UTC (rev 248930)
+++ trunk/Source/WebCore/mathml/MathMLElement.cpp 2019-08-21 01:39:54 UTC (rev 248931)
@@ -221,12 +221,6 @@
return isLink() || StyledElement::supportsFocus();
}
-int MathMLElement::defaultTabIndex() const
-{
-// FIXME: This seems wrong.
-return 0;
}
-}
-
#endif // ENABLE(MATHML)
Modified: trunk/Source/WebCore/mathml/MathMLElement.h (248930 => 248931)
--- trunk/Source/WebCore/mathml/MathMLElement.h 2019-08-21 01:36:41 UTC (rev 248930)
+++ trunk/Source/WebCore/mathml/MathMLElement.h 2019-08-21 01:39:54 UTC (rev 248931)
@@ -108,7 +108,6 @@
bool isMouseFocusable() const final;
bool isURLAttribute(const Attribute&) const final;
bool supportsFocus() const final;
-int defaultTabIndex() const final;
};
inline bool Node::hasTagName(const MathMLQualifiedName& name) const
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248930] trunk/Source/WTF
Title: [248930] trunk/Source/WTF
Revision 248930
Author mark@apple.com
Date 2019-08-20 18:36:41 -0700 (Tue, 20 Aug 2019)
Log Message
Make it easier to pass pointers to WTFCrashWithInfo.
https://bugs.webkit.org/show_bug.cgi?id=200960
Reviewed by Saam Barati and Yusuke Suzuki.
Now, we don't have to explicitly cast them to uint64_ts first. The template
wrappers will take care of it for us.
* wtf/Assertions.h:
(wtfCrashArg):
(WTFCrashWithInfo):
(WTF::isIntegralOrPointerType):
(WTF::isIntegralType): Deleted.
Modified Paths
trunk/Source/WTF/ChangeLog
trunk/Source/WTF/wtf/Assertions.h
Diff
Modified: trunk/Source/WTF/ChangeLog (248929 => 248930)
--- trunk/Source/WTF/ChangeLog 2019-08-21 01:13:02 UTC (rev 248929)
+++ trunk/Source/WTF/ChangeLog 2019-08-21 01:36:41 UTC (rev 248930)
@@ -1,3 +1,19 @@
+2019-08-20 Mark Lam
+
+Make it easier to pass pointers to WTFCrashWithInfo.
+https://bugs.webkit.org/show_bug.cgi?id=200960
+
+Reviewed by Saam Barati and Yusuke Suzuki.
+
+Now, we don't have to explicitly cast them to uint64_ts first. The template
+wrappers will take care of it for us.
+
+* wtf/Assertions.h:
+(wtfCrashArg):
+(WTFCrashWithInfo):
+(WTF::isIntegralOrPointerType):
+(WTF::isIntegralType): Deleted.
+
2019-08-20 Saam Barati
Unreviewed. Followup to r248903. It's not valid to remove
Modified: trunk/Source/WTF/wtf/Assertions.h (248929 => 248930)
--- trunk/Source/WTF/wtf/Assertions.h 2019-08-21 01:13:02 UTC (rev 248929)
+++ trunk/Source/WTF/wtf/Assertions.h 2019-08-21 01:36:41 UTC (rev 248930)
@@ -1,5 +1,5 @@
/*
- * Copyright (C) 2003-2017 Apple Inc. All rights reserved.
+ * Copyright (C) 2003-2019 Apple Inc. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
@@ -549,6 +549,54 @@
#ifdef __cplusplus
+template
+ALWAYS_INLINE uint64_t wtfCrashArg(T* arg) { return reinterpret_cast(arg); }
+
+template
+ALWAYS_INLINE uint64_t wtfCrashArg(T arg) { return arg; }
+
+template
+NO_RETURN_DUE_TO_CRASH ALWAYS_INLINE void WTFCrashWithInfo(int line, const char* file, const char* function, int counter, T reason)
+{
+WTFCrashWithInfo(line, file, function, counter, wtfCrashArg(reason));
+}
+
+template
+NO_RETURN_DUE_TO_CRASH ALWAYS_INLINE void WTFCrashWithInfo(int line, const char* file, const char* function, int counter, T reason, U misc1)
+{
+WTFCrashWithInfo(line, file, function, counter, wtfCrashArg(reason), wtfCrashArg(misc1));
+}
+
+template
+NO_RETURN_DUE_TO_CRASH ALWAYS_INLINE void WTFCrashWithInfo(int line, const char* file, const char* function, int counter, T reason, U misc1, V misc2)
+{
+WTFCrashWithInfo(line, file, function, counter, wtfCrashArg(reason), wtfCrashArg(misc1), wtfCrashArg(misc2));
+}
+
+template
+NO_RETURN_DUE_TO_CRASH ALWAYS_INLINE void WTFCrashWithInfo(int line, const char* file, const char* function, int counter, T reason, U misc1, V misc2, W misc3)
+{
+WTFCrashWithInfo(line, file, function, counter, wtfCrashArg(reason), wtfCrashArg(misc1), wtfCrashArg(misc2), wtfCrashArg(misc3));
+}
+
+template
+NO_RETURN_DUE_TO_CRASH ALWAYS_INLINE void WTFCrashWithInfo(int line, const char* file, const char* function, int counter, T reason, U misc1, V misc2, W misc3, X misc4)
+{
+WTFCrashWithInfo(line, file, function, counter, wtfCrashArg(reason), wtfCrashArg(misc1), wtfCrashArg(misc2), wtfCrashArg(misc3), wtfCrashArg(misc4));
+}
+
+template
+NO_RETURN_DUE_TO_CRASH ALWAYS_INLINE void WTFCrashWithInfo(int line, const char* file, const char* function, int counter, T reason, U misc1, V misc2, W misc3, X misc4, Y misc5)
+{
+WTFCrashWithInfo(line, file, function, counter, wtfCrashArg(reason), wtfCrashArg(misc1), wtfCrashArg(misc2), wtfCrashArg(misc3), wtfCrashArg(misc4), wtfCrashArg(misc5));
+}
+
+template
+NO_RETURN_DUE_TO_CRASH ALWAYS_INLINE void WTFCrashWithInfo(int line, const char* file, const char* function, int counter, T reason, U misc1, V misc2, W misc3, X misc4, Y misc5, Z misc6)
+{
+WTFCrashWithInfo(line, file, function, counter, wtfCrashArg(reason), wtfCrashArg(misc1), wtfCrashArg(misc2), wtfCrashArg(misc3), wtfCrashArg(misc4), wtfCrashArg(misc5), wtfCrashArg(misc6));
+}
+
// The combination of line, file, function, and counter should be a unique number per call to this crash. This tricks the compiler into not coalescing calls to WTFCrashWithInfo.
// The easiest way to fill these values per translation unit is to pass __LINE__, __FILE__, WTF_PRETTY_FUNCTION, and __COUNTER__.
WTF_EXPORT_PRIVATE NO_RETURN_DUE_TO_CRASH NOT_TAIL_CALLED void WTFCrashWithInfo(int line, const char* file, const char* function, int counter, uint64_t reason, uint64_t misc1, uint64_t misc2, uint64_t misc3, uint64_t misc4, uint64_t misc5, uint64_t misc6);
@@ -569,13 +617,13 @@
}
namespace WTF {
-inline void isIntegralType() { }
+inline void
[webkit-changes] [248929] trunk/Source/JavaScriptCore
Title: [248929] trunk/Source/_javascript_Core
Revision 248929
Author mark@apple.com
Date 2019-08-20 18:13:02 -0700 (Tue, 20 Aug 2019)
Log Message
Remove superfluous size argument to allocateCell() for fixed size objects.
https://bugs.webkit.org/show_bug.cgi?id=200958
Reviewed by Yusuke Suzuki.
The size is already automatically computed by the allocateCell() template's default
arguments. Removing these superfluous arguments will make it easier for us to
grep for cases where we do allocate variable size cells (for later analysis work).
* jsc.cpp:
(JSC::Masquerader::create):
(JSCMemoryFootprint::create):
* tools/JSDollarVM.cpp:
(JSC::JSDollarVMCallFrame::create):
(JSC::Element::create):
(JSC::Root::create):
(JSC::SimpleObject::create):
(JSC::ImpureGetter::create):
(JSC::CustomGetter::create):
(JSC::DOMJITNode::create):
(JSC::DOMJITGetter::create):
(JSC::DOMJITGetterComplex::create):
(JSC::DOMJITFunctionObject::create):
(JSC::DOMJITCheckSubClassObject::create):
(JSC::DOMJITGetterBaseJSObject::create):
(JSC::JSTestCustomGetterSetter::create):
(JSC::WasmStreamingParser::create):
Modified Paths
trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/jsc.cpp
trunk/Source/_javascript_Core/tools/JSDollarVM.cpp
Diff
Modified: trunk/Source/_javascript_Core/ChangeLog (248928 => 248929)
--- trunk/Source/_javascript_Core/ChangeLog 2019-08-21 00:54:48 UTC (rev 248928)
+++ trunk/Source/_javascript_Core/ChangeLog 2019-08-21 01:13:02 UTC (rev 248929)
@@ -1,5 +1,35 @@
2019-08-20 Mark Lam
+Remove superfluous size argument to allocateCell() for fixed size objects.
+https://bugs.webkit.org/show_bug.cgi?id=200958
+
+Reviewed by Yusuke Suzuki.
+
+The size is already automatically computed by the allocateCell() template's default
+arguments. Removing these superfluous arguments will make it easier for us to
+grep for cases where we do allocate variable size cells (for later analysis work).
+
+* jsc.cpp:
+(JSC::Masquerader::create):
+(JSCMemoryFootprint::create):
+* tools/JSDollarVM.cpp:
+(JSC::JSDollarVMCallFrame::create):
+(JSC::Element::create):
+(JSC::Root::create):
+(JSC::SimpleObject::create):
+(JSC::ImpureGetter::create):
+(JSC::CustomGetter::create):
+(JSC::DOMJITNode::create):
+(JSC::DOMJITGetter::create):
+(JSC::DOMJITGetterComplex::create):
+(JSC::DOMJITFunctionObject::create):
+(JSC::DOMJITCheckSubClassObject::create):
+(JSC::DOMJITGetterBaseJSObject::create):
+(JSC::JSTestCustomGetterSetter::create):
+(JSC::WasmStreamingParser::create):
+
+2019-08-20 Mark Lam
+
JSBigInt::m_length should be immutable.
https://bugs.webkit.org/show_bug.cgi?id=200956
Modified: trunk/Source/_javascript_Core/jsc.cpp (248928 => 248929)
--- trunk/Source/_javascript_Core/jsc.cpp 2019-08-21 00:54:48 UTC (rev 248928)
+++ trunk/Source/_javascript_Core/jsc.cpp 2019-08-21 01:13:02 UTC (rev 248929)
@@ -1,6 +1,6 @@
/*
* Copyright (C) 1999-2000 Harri Porten (por...@kde.org)
- * Copyright (C) 2004-2018 Apple Inc. All rights reserved.
+ * Copyright (C) 2004-2019 Apple Inc. All rights reserved.
* Copyright (C) 2006 Bjoern Graf (bjoern.g...@gmail.com)
*
* This library is free software; you can redistribute it and/or
@@ -201,7 +201,7 @@
{
globalObject->masqueradesAsUndefinedWatchpoint()->fireAll(vm, "Masquerading object allocated");
Structure* structure = createStructure(vm, globalObject, jsNull());
-Masquerader* result = new (NotNull, allocateCell(vm.heap, sizeof(Masquerader))) Masquerader(vm, structure);
+Masquerader* result = new (NotNull, allocateCell(vm.heap)) Masquerader(vm, structure);
result->finishCreation(vm);
return result;
}
@@ -1392,7 +1392,7 @@
static JSCMemoryFootprint* create(VM& vm, JSGlobalObject* globalObject)
{
Structure* structure = createStructure(vm, globalObject, jsNull());
-JSCMemoryFootprint* footprint = new (NotNull, allocateCell(vm.heap, sizeof(JSCMemoryFootprint))) JSCMemoryFootprint(vm, structure);
+JSCMemoryFootprint* footprint = new (NotNull, allocateCell(vm.heap)) JSCMemoryFootprint(vm, structure);
footprint->finishCreation(vm);
return footprint;
}
Modified: trunk/Source/_javascript_Core/tools/JSDollarVM.cpp (248928 => 248929)
--- trunk/Source/_javascript_Core/tools/JSDollarVM.cpp 2019-08-21 00:54:48 UTC (rev 248928)
+++ trunk/Source/_javascript_Core/tools/JSDollarVM.cpp 2019-08-21 01:13:02 UTC (rev 248929)
@@ -79,7 +79,7 @@
VM& vm = exec->vm();
JSGlobalObject* globalObject = exec->lexicalGlobalObject();
Structure* structure = createStructure(vm, globalObject, jsNull());
-JSDollarVMCallFrame* frame = new (NotNull, allocateCell(vm.heap, sizeof(JSDollarVMCallFrame))) JSDollarVMCallFrame(vm,
[webkit-changes] [248928] trunk/Source/WebKit
Title: [248928] trunk/Source/WebKit
Revision 248928
Author commit-qu...@webkit.org
Date 2019-08-20 17:54:48 -0700 (Tue, 20 Aug 2019)
Log Message
[iOS] REGRESSION (248501): gmail.com images are still not shown on some versions of iPhone iOS13
https://bugs.webkit.org/show_bug.cgi?id=200948
Patch by Said Abou-Hallawa on 2019-08-20
Reviewed by Tim Horton.
Apply the gmail.com quirk for all iPhone iOS versions till the gmail.com
bug is fixed.
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::platformUserAgent const):
Modified Paths
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/WebProcess/WebPage/ios/WebPageIOS.mm
Diff
Modified: trunk/Source/WebKit/ChangeLog (248927 => 248928)
--- trunk/Source/WebKit/ChangeLog 2019-08-21 00:39:02 UTC (rev 248927)
+++ trunk/Source/WebKit/ChangeLog 2019-08-21 00:54:48 UTC (rev 248928)
@@ -1,3 +1,16 @@
+2019-08-20 Said Abou-Hallawa
+
+[iOS] REGRESSION (248501): gmail.com images are still not shown on some versions of iPhone iOS13
+https://bugs.webkit.org/show_bug.cgi?id=200948
+
+Reviewed by Tim Horton.
+
+Apply the gmail.com quirk for all iPhone iOS versions till the gmail.com
+bug is fixed.
+
+* WebProcess/WebPage/ios/WebPageIOS.mm:
+(WebKit::WebPage::platformUserAgent const):
+
2019-08-20 Chris Dumez
WebSQLiteDatabaseTracker does not ensure it is still alive after dispatching to the main thread
Modified: trunk/Source/WebKit/WebProcess/WebPage/ios/WebPageIOS.mm (248927 => 248928)
--- trunk/Source/WebKit/WebProcess/WebPage/ios/WebPageIOS.mm 2019-08-21 00:39:02 UTC (rev 248927)
+++ trunk/Source/WebKit/WebProcess/WebPage/ios/WebPageIOS.mm 2019-08-21 00:54:48 UTC (rev 248928)
@@ -3755,7 +3755,7 @@
if (!document)
return String();
-if (document->quirks().shouldAvoidUsingIOS13ForGmail() && osNameForUserAgent() == "iPhone OS" && systemMarketingVersionForUserAgentString() == "13_1")
+if (document->quirks().shouldAvoidUsingIOS13ForGmail() && osNameForUserAgent() == "iPhone OS")
return standardUserAgentWithApplicationName({ }, "12_1_3");
return String();
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248927] trunk/Source/JavaScriptCore
Title: [248927] trunk/Source/_javascript_Core
Revision 248927
Author mark@apple.com
Date 2019-08-20 17:39:02 -0700 (Tue, 20 Aug 2019)
Log Message
JSBigInt::m_length should be immutable.
https://bugs.webkit.org/show_bug.cgi?id=200956
Reviewed by Yusuke Suzuki.
This is because the JSBigInt cell size is allocated with that length. Changing
the length after construction does not change the size of the cell, and hence,
makes no sense.
This patch removes the setLength() method, and decorates the m_length field with
const to enforce that it is immutable after construction.
* runtime/JSBigInt.h:
Modified Paths
trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/runtime/JSBigInt.h
Diff
Modified: trunk/Source/_javascript_Core/ChangeLog (248926 => 248927)
--- trunk/Source/_javascript_Core/ChangeLog 2019-08-21 00:26:52 UTC (rev 248926)
+++ trunk/Source/_javascript_Core/ChangeLog 2019-08-21 00:39:02 UTC (rev 248927)
@@ -1,3 +1,19 @@
+2019-08-20 Mark Lam
+
+JSBigInt::m_length should be immutable.
+https://bugs.webkit.org/show_bug.cgi?id=200956
+
+Reviewed by Yusuke Suzuki.
+
+This is because the JSBigInt cell size is allocated with that length. Changing
+the length after construction does not change the size of the cell, and hence,
+makes no sense.
+
+This patch removes the setLength() method, and decorates the m_length field with
+const to enforce that it is immutable after construction.
+
+* runtime/JSBigInt.h:
+
2019-08-20 Devin Rousso
Web Inspector: Implement `queryHolders` Command Line API
Modified: trunk/Source/_javascript_Core/runtime/JSBigInt.h (248926 => 248927)
--- trunk/Source/_javascript_Core/runtime/JSBigInt.h 2019-08-21 00:26:52 UTC (rev 248926)
+++ trunk/Source/_javascript_Core/runtime/JSBigInt.h 2019-08-21 00:39:02 UTC (rev 248927)
@@ -1,5 +1,6 @@
/*
* Copyright (C) 2017 Caio Lima
+ * Copyright (C) 2019 Apple Inc. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
@@ -71,7 +72,6 @@
void setSign(bool sign) { m_sign = sign; }
bool sign() const { return m_sign; }
-void setLength(unsigned length) { m_length = length; }
unsigned length() const { return m_length; }
enum class ErrorParseMode {
@@ -246,8 +246,8 @@
Digit digit(unsigned);
void setDigit(unsigned, Digit);
-
-unsigned m_length;
+
+const unsigned m_length;
bool m_sign { false };
};
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248926] trunk/Tools
Title: [248926] trunk/Tools
Revision 248926
Author zhifei_f...@apple.com
Date 2019-08-20 17:26:52 -0700 (Tue, 20 Aug 2019)
Log Message
[results.webkit.org Webkit.css] Add right sidebar
https://bugs.webkit.org/show_bug.cgi?id=200946
Reviewed by Jonathan Bedard.
* resultsdbpy/resultsdbpy/view/static/library/css/webkit.css:
(@media screen and (min-width: 600px) and (orientation: landscape)):
(.main.left):
(@media screen and (min-width: 768px) and (orientation: landscape)):
(.sidebar.right):
Modified Paths
trunk/Tools/ChangeLog
trunk/Tools/resultsdbpy/resultsdbpy/view/static/library/css/webkit.css
Diff
Modified: trunk/Tools/ChangeLog (248925 => 248926)
--- trunk/Tools/ChangeLog 2019-08-21 00:24:27 UTC (rev 248925)
+++ trunk/Tools/ChangeLog 2019-08-21 00:26:52 UTC (rev 248926)
@@ -1,3 +1,16 @@
+2019-08-20 Zhifei Fang
+
+[results.webkit.org Webkit.css] Add right sidebar
+https://bugs.webkit.org/show_bug.cgi?id=200946
+
+Reviewed by Jonathan Bedard.
+
+* resultsdbpy/resultsdbpy/view/static/library/css/webkit.css:
+(@media screen and (min-width: 600px) and (orientation: landscape)):
+(.main.left):
+(@media screen and (min-width: 768px) and (orientation: landscape)):
+(.sidebar.right):
+
2019-08-20 Jonathan Bedard
results.webkit.org: Add ToolTips
Modified: trunk/Tools/resultsdbpy/resultsdbpy/view/static/library/css/webkit.css (248925 => 248926)
--- trunk/Tools/resultsdbpy/resultsdbpy/view/static/library/css/webkit.css 2019-08-21 00:24:27 UTC (rev 248925)
+++ trunk/Tools/resultsdbpy/resultsdbpy/view/static/library/css/webkit.css 2019-08-21 00:26:52 UTC (rev 248926)
@@ -962,6 +962,12 @@
margin-left: 299px;
width: calc(100% - 299px);
}
+
+ .main.left {
+margin-left: 0px;
+margin-right: 299px;
+ }
+
.main.under-topbar-with-actions {
margin-top: calc(12px * 2 + var(--largeSize) * 1.3);
}
@@ -981,6 +987,12 @@
margin-left: 310px;
width: calc(100% - 310px);
}
+
+ .main.left {
+margin-left: 0px;
+margin-right: 310px;
+ }
+
.main.under-topbar-with-actions {
margin-top: calc(12px * 2 + var(--largeSize) * 1.3);
}
@@ -989,6 +1001,7 @@
.sidebar.right {
border-right: 0px;
border-left: 1px;
+ right: 0;
}
.sidebar>.list {
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248924] trunk/LayoutTests/imported/w3c
Title: [248924] trunk/LayoutTests/imported/w3c Revision 248924 Author rn...@webkit.org Date 2019-08-20 17:10:02 -0700 (Tue, 20 Aug 2019) Log Message Revert the unintended test change in r248914. * web-platform-tests/html/infrastructure/common-dom-interfaces/collections/htmlformcontrolscollection.html: Modified Paths trunk/LayoutTests/imported/w3c/ChangeLog trunk/LayoutTests/imported/w3c/web-platform-tests/html/infrastructure/common-dom-interfaces/collections/htmlformcontrolscollection.html Diff Modified: trunk/LayoutTests/imported/w3c/ChangeLog (248923 => 248924) --- trunk/LayoutTests/imported/w3c/ChangeLog 2019-08-20 23:47:03 UTC (rev 248923) +++ trunk/LayoutTests/imported/w3c/ChangeLog 2019-08-21 00:10:02 UTC (rev 248924) @@ -1,3 +1,9 @@ +2019-08-20 Ryosuke Niwa + +Revert the unintended test change in r248914. + +* web-platform-tests/html/infrastructure/common-dom-interfaces/collections/htmlformcontrolscollection.html: + 2019-08-14 Alexey Shvayka Error thrown during "acceptNode" lookup is overridden Modified: trunk/LayoutTests/imported/w3c/web-platform-tests/html/infrastructure/common-dom-interfaces/collections/htmlformcontrolscollection.html (248923 => 248924) --- trunk/LayoutTests/imported/w3c/web-platform-tests/html/infrastructure/common-dom-interfaces/collections/htmlformcontrolscollection.html 2019-08-20 23:47:03 UTC (rev 248923) +++ trunk/LayoutTests/imported/w3c/web-platform-tests/html/infrastructure/common-dom-interfaces/collections/htmlformcontrolscollection.html 2019-08-21 00:10:02 UTC (rev 248924) @@ -7,8 +7,8 @@
[webkit-changes] [248923] trunk/Source/WebInspectorUI
Title: [248923] trunk/Source/WebInspectorUI
Revision 248923
Author pecor...@apple.com
Date 2019-08-20 16:47:03 -0700 (Tue, 20 Aug 2019)
Log Message
Web Inspector: Address a few Esprima issues preventing pretty printing of resources
https://bugs.webkit.org/show_bug.cgi?id=200935
Reviewed by Timothy Hatcher.
Address a few Esprima issues:
Issue #1991 - Failure to parse template literal with destructuring assignment _expression_
https://github.com/jquery/esprima/issues/1991
Issue #1920 - Invalid Left Hand Side in for-in
https://github.com/jquery/esprima/issues/1920
* UserInterface/External/Esprima/esprima.js:
Modified Paths
trunk/Source/WebInspectorUI/ChangeLog
trunk/Source/WebInspectorUI/UserInterface/External/Esprima/esprima.js
Diff
Modified: trunk/Source/WebInspectorUI/ChangeLog (248922 => 248923)
--- trunk/Source/WebInspectorUI/ChangeLog 2019-08-20 23:44:39 UTC (rev 248922)
+++ trunk/Source/WebInspectorUI/ChangeLog 2019-08-20 23:47:03 UTC (rev 248923)
@@ -1,5 +1,22 @@
2019-08-20 Joseph Pecoraro
+Web Inspector: Address a few Esprima issues preventing pretty printing of resources
+https://bugs.webkit.org/show_bug.cgi?id=200935
+
+Reviewed by Timothy Hatcher.
+
+Address a few Esprima issues:
+
+Issue #1991 - Failure to parse template literal with destructuring assignment _expression_
+https://github.com/jquery/esprima/issues/1991
+
+Issue #1920 - Invalid Left Hand Side in for-in
+https://github.com/jquery/esprima/issues/1920
+
+* UserInterface/External/Esprima/esprima.js:
+
+2019-08-20 Joseph Pecoraro
+
Web Inspector: Update CodeMirror to support numeric separators in _javascript_ numbers
https://bugs.webkit.org/show_bug.cgi?id=200942
Modified: trunk/Source/WebInspectorUI/UserInterface/External/Esprima/esprima.js (248922 => 248923)
--- trunk/Source/WebInspectorUI/UserInterface/External/Esprima/esprima.js 2019-08-20 23:44:39 UTC (rev 248922)
+++ trunk/Source/WebInspectorUI/UserInterface/External/Esprima/esprima.js 2019-08-20 23:47:03 UTC (rev 248923)
@@ -3784,6 +3784,9 @@
}
else {
const initStartToken = this.lookahead;
+const previousIsBindingElement = this.context.isBindingElement;
+const previousIsAssignmentTarget = this.context.isAssignmentTarget;
+const previousFirstCoverInitializedNameError = this.context.firstCoverInitializedNameError;
const previousAllowIn = this.context.allowIn;
this.context.allowIn = false;
init = this.inheritCoverGrammar(this.parseAssignmentExpression);
@@ -3813,6 +3816,10 @@
forIn = false;
}
else {
+// The `init` node was not parsed isolated, but we would have wanted it to.
+this.context.isBindingElement = previousIsBindingElement;
+this.context.isAssignmentTarget = previousIsAssignmentTarget;
+this.context.firstCoverInitializedNameError = previousFirstCoverInitializedNameError;
if (this.match(',')) {
const initSeq = [init];
while (this.match(',')) {
@@ -3827,11 +3834,11 @@
}
if (typeof left === 'undefined') {
if (!this.match(';')) {
-test = this.parseExpression();
+test = this.isolateCoverGrammar(this.parseExpression);
}
this.expect(';');
if (!this.match(')')) {
-update = this.parseExpression();
+update = this.isolateCoverGrammar(this.parseExpression);
}
}
let body;
@@ -5148,7 +5155,8 @@
return {
index: this.index,
lineNumber: this.lineNumber,
-lineStart: this.lineStart
+lineStart: this.lineStart,
+curlyStack: this.curlyStack.slice()
};
}
restoreState(state) {
@@ -5155,6 +5163,7 @@
this.index = state.index;
this.lineNumber = state.lineNumber;
this.lineStart = state.lineStart;
+this.curlyStack = state.curlyStack;
}
eof() {
return this.index >= this.length;
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248922] trunk/Source/WebInspectorUI
Title: [248922] trunk/Source/WebInspectorUI
Revision 248922
Author pecor...@apple.com
Date 2019-08-20 16:44:39 -0700 (Tue, 20 Aug 2019)
Log Message
Web Inspector: Update CodeMirror to support numeric separators in _javascript_ numbers
https://bugs.webkit.org/show_bug.cgi?id=200942
Cherry-pick a few CodeMirror changes:
[_javascript_ mode] Support numeric separators
https://github.com/codemirror/CodeMirror/commit/beab8ed123683416bfec934df73d13401ec086b5#diff-9812850bb71d31e8dd60b476abb2bae8
[_javascript_ mode] fix tokenizing of underscore properties
https://github.com/codemirror/CodeMirror/commit/463ea2c34ab442c0cae1d9732305219ca9b04dfe#diff-9812850bb71d31e8dd60b476abb2bae8
Reviewed by Timothy Hatcher.
* UserInterface/External/CodeMirror/_javascript_.js:
Modified Paths
trunk/Source/WebInspectorUI/ChangeLog
trunk/Source/WebInspectorUI/UserInterface/External/CodeMirror/_javascript_.js
Diff
Modified: trunk/Source/WebInspectorUI/ChangeLog (248921 => 248922)
--- trunk/Source/WebInspectorUI/ChangeLog 2019-08-20 23:13:17 UTC (rev 248921)
+++ trunk/Source/WebInspectorUI/ChangeLog 2019-08-20 23:44:39 UTC (rev 248922)
@@ -1,3 +1,20 @@
+2019-08-20 Joseph Pecoraro
+
+Web Inspector: Update CodeMirror to support numeric separators in _javascript_ numbers
+https://bugs.webkit.org/show_bug.cgi?id=200942
+
+Cherry-pick a few CodeMirror changes:
+
+[_javascript_ mode] Support numeric separators
+https://github.com/codemirror/CodeMirror/commit/beab8ed123683416bfec934df73d13401ec086b5#diff-9812850bb71d31e8dd60b476abb2bae8
+
+[_javascript_ mode] fix tokenizing of underscore properties
+https://github.com/codemirror/CodeMirror/commit/463ea2c34ab442c0cae1d9732305219ca9b04dfe#diff-9812850bb71d31e8dd60b476abb2bae8
+
+Reviewed by Timothy Hatcher.
+
+* UserInterface/External/CodeMirror/_javascript_.js:
+
2019-08-20 Devin Rousso
Web Inspector: Sources: move the resource type scope bar to be next to the filter
Modified: trunk/Source/WebInspectorUI/UserInterface/External/CodeMirror/_javascript_.js (248921 => 248922)
--- trunk/Source/WebInspectorUI/UserInterface/External/CodeMirror/_javascript_.js 2019-08-20 23:13:17 UTC (rev 248921)
+++ trunk/Source/WebInspectorUI/UserInterface/External/CodeMirror/_javascript_.js 2019-08-20 23:44:39 UTC (rev 248922)
@@ -104,7 +104,7 @@
if (ch == '"' || ch == "'") {
state.tokenize = tokenString(ch);
return state.tokenize(stream, state);
-} else if (ch == "." && stream.match(/^\d+(?:[eE][+\-]?\d+)?/)) {
+} else if (ch == "." && stream.match(/^\d[\d_]*(?:[eE][+\-]?[\d_]+)?/)) {
return ret("number", "number");
} else if (ch == "." && stream.match("..")) {
return ret("spread", "meta");
@@ -112,10 +112,10 @@
return ret(ch);
} else if (ch == "=" && stream.eat(">")) {
return ret("=>", "operator");
-} else if (ch == "0" && stream.match(/^(?:x[\da-f]+|o[0-7]+|b[01]+)n?/i)) {
+} else if (ch == "0" && stream.match(/^(?:x[\dA-Fa-f_]+|o[0-7_]+|b[01_]+)n?/)) {
return ret("number", "number");
} else if (/\d/.test(ch)) {
- stream.match(/^\d*(?:n|(?:\.\d*)?(?:[eE][+\-]?\d+)?)?/);
+ stream.match(/^[\d_]*(?:n|(?:\.[\d_]*)?(?:[eE][+\-]?[\d_]+)?)?/);
return ret("number", "number");
} else if (ch == "/") {
if (stream.eat("*")) {
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248921] releases/WebKitGTK/webkit-2.24
Title: [248921] releases/WebKitGTK/webkit-2.24
Revision 248921
Author ape...@igalia.com
Date 2019-08-20 16:13:17 -0700 (Tue, 20 Aug 2019)
Log Message
Merged r248410 - Do not allow navigations of frames about to get replaced by the result of evaluating _javascript_: URLs
and https://bugs.webkit.org/show_bug.cgi?id=198786
Reviewed by Geoff Garen.
Source/WebCore:
Covered by API Test
Add a "willReplaceWithResultOfExecutingJavascriptURL" flag which is respected inside FrameLoader::isNavigationAllowed
* bindings/js/ScriptController.cpp:
(WebCore::ScriptController::executeIfJavaScriptURL):
* bindings/js/ScriptController.h:
(WebCore::ScriptController::willReplaceWithResultOfExecutingJavascriptURL const):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::isNavigationAllowed const):
Tools:
* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/mac/_javascript_URLNavigation.mm: Added.
Modified Paths
releases/WebKitGTK/webkit-2.24/Source/WebCore/ChangeLog
releases/WebKitGTK/webkit-2.24/Source/WebCore/bindings/js/ScriptController.cpp
releases/WebKitGTK/webkit-2.24/Source/WebCore/bindings/js/ScriptController.h
releases/WebKitGTK/webkit-2.24/Source/WebCore/loader/FrameLoader.cpp
releases/WebKitGTK/webkit-2.24/Tools/ChangeLog
releases/WebKitGTK/webkit-2.24/Tools/TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj
Added Paths
releases/WebKitGTK/webkit-2.24/Tools/TestWebKitAPI/Tests/mac/_javascript_URLNavigation.mm
Diff
Modified: releases/WebKitGTK/webkit-2.24/Source/WebCore/ChangeLog (248920 => 248921)
--- releases/WebKitGTK/webkit-2.24/Source/WebCore/ChangeLog 2019-08-20 23:11:41 UTC (rev 248920)
+++ releases/WebKitGTK/webkit-2.24/Source/WebCore/ChangeLog 2019-08-20 23:13:17 UTC (rev 248921)
@@ -1,3 +1,22 @@
+2019-08-08 Brady Eidson
+
+Do not allow navigations of frames about to get replaced by the result of evaluating _javascript_: URLs
+ and https://bugs.webkit.org/show_bug.cgi?id=198786
+
+Reviewed by Geoff Garen.
+
+Covered by API Test
+
+Add a "willReplaceWithResultOfExecutingJavascriptURL" flag which is respected inside FrameLoader::isNavigationAllowed
+
+* bindings/js/ScriptController.cpp:
+(WebCore::ScriptController::executeIfJavaScriptURL):
+* bindings/js/ScriptController.h:
+(WebCore::ScriptController::willReplaceWithResultOfExecutingJavascriptURL const):
+
+* loader/FrameLoader.cpp:
+(WebCore::FrameLoader::isNavigationAllowed const):
+
2019-08-08 Charlie Turner
[GTK] WebKitWebProcess crashes when viewing an HTML with a element referencing unknown file
Modified: releases/WebKitGTK/webkit-2.24/Source/WebCore/bindings/js/ScriptController.cpp (248920 => 248921)
--- releases/WebKitGTK/webkit-2.24/Source/WebCore/bindings/js/ScriptController.cpp 2019-08-20 23:11:41 UTC (rev 248920)
+++ releases/WebKitGTK/webkit-2.24/Source/WebCore/bindings/js/ScriptController.cpp 2019-08-20 23:13:17 UTC (rev 248921)
@@ -622,8 +622,14 @@
if (shouldReplaceDocumentIfJavaScriptURL == ReplaceDocumentIfJavaScriptURL) {
// We're still in a frame, so there should be a DocumentLoader.
ASSERT(m_frame.document()->loader());
-
-// DocumentWriter::replaceDocument can cause the DocumentLoader to get deref'ed and possible destroyed,
+
+// Signal to FrameLoader to disable navigations within this frame while replacing it with the result of executing _javascript_
+// FIXME: https://bugs.webkit.org/show_bug.cgi?id=200523
+// The only reason we do a nestable save/restore of this flag here is because we sometimes nest _javascript_: url loads as
+// some will load synchronously. We'd like to remove those synchronous loads and then change this.
+SetForScope willBeReplaced(m_willReplaceWithResultOfExecutingJavascriptURL, true);
+
+// DocumentWriter::replaceDocumentWithResultOfExecutingJavascriptURL can cause the DocumentLoader to get deref'ed and possible destroyed,
// so protect it with a RefPtr.
if (RefPtr loader = m_frame.document()->loader())
loader->writer().replaceDocumentWithResultOfExecutingJavascriptURL(scriptResult, ownerDocument.get());
Modified: releases/WebKitGTK/webkit-2.24/Source/WebCore/bindings/js/ScriptController.h (248920 => 248921)
--- releases/WebKitGTK/webkit-2.24/Source/WebCore/bindings/js/ScriptController.h 2019-08-20 23:11:41 UTC (rev 248920)
+++ releases/WebKitGTK/webkit-2.24/Source/WebCore/bindings/js/ScriptController.h 2019-08-20 23:13:17 UTC (rev 248921)
@@ -159,6 +159,8 @@
void initScriptForWindowProxy(JSWindowProxy&);
+bool willReplaceWithResultOfExecutingJavascriptURL() const { return m_willReplaceWithResultOfExecutingJavascriptURL; }
+
private:
void setupModuleScriptHandlers(LoadableModuleScript&, JSC::JSInternalPromise&, DOMWrapperWorld&);
@@ -171,6 +173,7 @@
const String* m_sourceURL;
bool m_paused;
+bool
[webkit-changes] [248920] trunk
Title: [248920] trunk
Revision 248920
Author sbar...@apple.com
Date 2019-08-20 16:11:41 -0700 (Tue, 20 Aug 2019)
Log Message
[WHLSL] We need to null check when emitting native code for operator&.
https://bugs.webkit.org/show_bug.cgi?id=200846
Reviewed by Myles C. Maxfield.
Source/WebCore:
Test: webgpu/whlsl/structure-field-access-on-null.html
* Modules/webgpu/WHLSL/Metal/WHLSLNativeFunctionWriter.cpp:
(WebCore::WHLSL::Metal::inlineNativeFunction):
LayoutTests:
* webgpu/whlsl/structure-field-access-on-null-expected.txt: Added.
* webgpu/whlsl/structure-field-access-on-null.html: Added.
Modified Paths
trunk/LayoutTests/ChangeLog
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/Modules/webgpu/WHLSL/Metal/WHLSLNativeFunctionWriter.cpp
Added Paths
trunk/LayoutTests/webgpu/whlsl/structure-field-access-on-null-expected.txt
trunk/LayoutTests/webgpu/whlsl/structure-field-access-on-null.html
Diff
Modified: trunk/LayoutTests/ChangeLog (248919 => 248920)
--- trunk/LayoutTests/ChangeLog 2019-08-20 22:35:28 UTC (rev 248919)
+++ trunk/LayoutTests/ChangeLog 2019-08-20 23:11:41 UTC (rev 248920)
@@ -1,3 +1,13 @@
+2019-08-20 Saam Barati
+
+[WHLSL] We need to null check when emitting native code for operator&.
+https://bugs.webkit.org/show_bug.cgi?id=200846
+
+Reviewed by Myles C. Maxfield.
+
+* webgpu/whlsl/structure-field-access-on-null-expected.txt: Added.
+* webgpu/whlsl/structure-field-access-on-null.html: Added.
+
2019-08-20 Russell Epstein
Updating Expectations for Multiple Newly Passing Tests.
Added: trunk/LayoutTests/webgpu/whlsl/structure-field-access-on-null-expected.txt (0 => 248920)
--- trunk/LayoutTests/webgpu/whlsl/structure-field-access-on-null-expected.txt (rev 0)
+++ trunk/LayoutTests/webgpu/whlsl/structure-field-access-on-null-expected.txt 2019-08-20 23:11:41 UTC (rev 248920)
@@ -0,0 +1,3 @@
+
+PASS fieldShouldBeNull
+
Added: trunk/LayoutTests/webgpu/whlsl/structure-field-access-on-null.html (0 => 248920)
--- trunk/LayoutTests/webgpu/whlsl/structure-field-access-on-null.html (rev 0)
+++ trunk/LayoutTests/webgpu/whlsl/structure-field-access-on-null.html 2019-08-20 23:11:41 UTC (rev 248920)
@@ -0,0 +1,37 @@
+
+
+
+
+field ander on nullptr should return null.
+
+const whlslTests = {};
+
+whlslTests.fieldShouldBeNull = async () => {
+const program = `
+struct Foo {
+int x;
+int y;
+int z;
+}
+bool foo(uint i) {
+Foo[10] foos;
+thread Foo* foo = &foos[i];
+thread int* ptr = &(foo->z);
+return ptr == null && foo == null;
+}
+`;
+
+for (let i = 0; i < 10; ++i)
+assert_equals(await callBoolFunction(program, "foo", [makeUint(i)]), false);
+assert_equals(await callBoolFunction(program, "foo", [makeUint(10)]), true);
+assert_equals(await callBoolFunction(program, "foo", [makeUint(100)]), true);
+assert_equals(await callBoolFunction(program, "foo", [makeUint(0x)]), true);
+};
+
+runTests(whlslTests);
+
+
Modified: trunk/Source/WebCore/ChangeLog (248919 => 248920)
--- trunk/Source/WebCore/ChangeLog 2019-08-20 22:35:28 UTC (rev 248919)
+++ trunk/Source/WebCore/ChangeLog 2019-08-20 23:11:41 UTC (rev 248920)
@@ -1,3 +1,15 @@
+2019-08-20 Saam Barati
+
+[WHLSL] We need to null check when emitting native code for operator&.
+https://bugs.webkit.org/show_bug.cgi?id=200846
+
+Reviewed by Myles C. Maxfield.
+
+Test: webgpu/whlsl/structure-field-access-on-null.html
+
+* Modules/webgpu/WHLSL/Metal/WHLSLNativeFunctionWriter.cpp:
+(WebCore::WHLSL::Metal::inlineNativeFunction):
+
2019-08-20 Justin Fan
Unreviewed comment follow-up for https://bugs.webkit.org/show_bug.cgi?id=200752.
Modified: trunk/Source/WebCore/Modules/webgpu/WHLSL/Metal/WHLSLNativeFunctionWriter.cpp (248919 => 248920)
--- trunk/Source/WebCore/Modules/webgpu/WHLSL/Metal/WHLSLNativeFunctionWriter.cpp 2019-08-20 22:35:28 UTC (rev 248919)
+++ trunk/Source/WebCore/Modules/webgpu/WHLSL/Metal/WHLSLNativeFunctionWriter.cpp 2019-08-20 23:11:41 UTC (rev 248920)
@@ -282,7 +282,7 @@
auto fieldName = nativeFunctionDeclaration.name().substring("operator&."_str.length());
stringBuilder.append(
-indent, returnName, " = &(", args[0], "->");
+indent, returnName, " = ", args[0], " ? &(", args[0], "->");
auto& unnamedType = *nativeFunctionDeclaration.parameters()[0]->type();
auto& unifyNode = downcast(unnamedType).elementType().unifyNode();
@@ -295,7 +295,7 @@
} else
stringBuilder.append(fieldName);
-stringBuilder.append(");\n");
+stringBuilder.append(") : nullptr;\n");
return;
}
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
[webkit-changes] [248919] trunk/Source/JavaScriptCore
Title: [248919] trunk/Source/_javascript_Core Revision 248919 Author justin_mich...@apple.com Date 2019-08-20 15:35:28 -0700 (Tue, 20 Aug 2019) Log Message [WASM-References] Enable by default https://bugs.webkit.org/show_bug.cgi?id=200931 Reviewed by Saam Barati. * runtime/Options.h: Modified Paths trunk/Source/_javascript_Core/ChangeLog trunk/Source/_javascript_Core/runtime/Options.h Diff Modified: trunk/Source/_javascript_Core/ChangeLog (248918 => 248919) --- trunk/Source/_javascript_Core/ChangeLog 2019-08-20 22:18:23 UTC (rev 248918) +++ trunk/Source/_javascript_Core/ChangeLog 2019-08-20 22:35:28 UTC (rev 248919) @@ -1,3 +1,12 @@ +2019-08-20 Justin Michaud + +[WASM-References] Enable by default +https://bugs.webkit.org/show_bug.cgi?id=200931 + +Reviewed by Saam Barati. + +* runtime/Options.h: + 2019-08-20 Yusuke Suzuki [JSC] Array.prototype.toString should not get "join" function each time Modified: trunk/Source/_javascript_Core/runtime/Options.h (248918 => 248919) --- trunk/Source/_javascript_Core/runtime/Options.h 2019-08-20 22:18:23 UTC (rev 248918) +++ trunk/Source/_javascript_Core/runtime/Options.h 2019-08-20 22:35:28 UTC (rev 248919) @@ -498,7 +498,7 @@ v(bool, useWebAssemblyStreamingApi, enableWebAssemblyStreamingApi, Normal, "Allow to run WebAssembly's Streaming API") \ v(bool, useCallICsForWebAssemblyToJSCalls, true, Normal, "If true, we will use CallLinkInfo to inline cache Wasm to JS calls.") \ v(bool, useEagerWebAssemblyModuleHashing, false, Normal, "Unnamed WebAssembly modules are identified in backtraces through their hash, if available.") \ -v(bool, useWebAssemblyReferences, false, Normal, "Allow types from the wasm references spec.") \ +v(bool, useWebAssemblyReferences, true, Normal, "Allow types from the wasm references spec.") \ v(bool, useWeakRefs, false, Normal, "Expose the WeakRef constructor.") \ v(bool, useBigInt, false, Normal, "If true, we will enable BigInt support.") \ v(bool, useNullishAwareOperators, false, Normal, "Enable support for ?. and ?? operators.") \ ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248918] trunk/Source
Title: [248918] trunk/Source
Revision 248918
Author cdu...@apple.com
Date 2019-08-20 15:18:23 -0700 (Tue, 20 Aug 2019)
Log Message
WebSQLiteDatabaseTracker does not ensure it is still alive after dispatching to the main thread
https://bugs.webkit.org/show_bug.cgi?id=200925
Reviewed by Geoffrey Garen.
WebSQLiteDatabaseTracker does not ensure it is still alive after dispatching to the main thread,
which is not safe. Use WeakPtr to address the issue.
* Shared/WebSQLiteDatabaseTracker.cpp:
(WebKit::WebSQLiteDatabaseTracker::WebSQLiteDatabaseTracker):
(WebKit::WebSQLiteDatabaseTracker::willBeginFirstTransaction):
(WebKit::WebSQLiteDatabaseTracker::didFinishLastTransaction):
* Shared/WebSQLiteDatabaseTracker.h:
Modified Paths
trunk/Source/WTF/wtf/WeakPtr.h
trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h
trunk/Source/WebKit/ChangeLog
trunk/Source/WebKit/Shared/WebSQLiteDatabaseTracker.cpp
trunk/Source/WebKit/Shared/WebSQLiteDatabaseTracker.h
Diff
Modified: trunk/Source/WTF/wtf/WeakPtr.h (248917 => 248918)
--- trunk/Source/WTF/wtf/WeakPtr.h 2019-08-20 21:33:06 UTC (rev 248917)
+++ trunk/Source/WTF/wtf/WeakPtr.h 2019-08-20 22:18:23 UTC (rev 248918)
@@ -154,12 +154,19 @@
m_impl->clear();
}
-WeakPtr createWeakPtr(T& object) const
+void initializeIfNeeded(const T& object) const
{
+if (m_impl)
+return;
+
ASSERT(m_wasConstructedOnMainThread == isMainThread());
-if (!m_impl)
-m_impl = WeakPtrImpl::create();
+m_impl = WeakPtrImpl::create(const_cast());
+}
+WeakPtr createWeakPtr(T& object) const
+{
+initializeIfNeeded(object);
+
ASSERT( == m_impl->get());
return WeakPtr(makeRef(*m_impl));
}
@@ -166,9 +173,7 @@
WeakPtr createWeakPtr(const T& object) const
{
-ASSERT(m_wasConstructedOnMainThread == isMainThread());
-if (!m_impl)
-m_impl = WeakPtrImpl::create(const_cast());
+initializeIfNeeded(object);
ASSERT( == m_impl->get());
return WeakPtr(makeRef(*m_impl));
@@ -192,13 +197,24 @@
#endif
};
-template class CanMakeWeakPtr {
+// We use lazy initialization of the WeakPtrFactory by default to avoid unnecessary initialization. Eager
+// initialization is however useful if you plan to call makeWeakPtr() from other threads.
+enum class WeakPtrFactoryInitialization { Lazy, Eager };
+
+template class CanMakeWeakPtr {
public:
-typedef T WeakValueType;
+using WeakValueType = T;
const WeakPtrFactory& weakPtrFactory() const { return m_weakPtrFactory; }
WeakPtrFactory& weakPtrFactory() { return m_weakPtrFactory; }
+protected:
+CanMakeWeakPtr()
+{
+if (initializationMode == WeakPtrFactoryInitialization::Eager)
+m_weakPtrFactory.initializeIfNeeded(static_cast(*this));
+}
+
private:
WeakPtrFactory m_weakPtrFactory;
};
@@ -278,4 +294,5 @@
using WTF::CanMakeWeakPtr;
using WTF::WeakPtr;
using WTF::WeakPtrFactory;
+using WTF::WeakPtrFactoryInitialization;
using WTF::makeWeakPtr;
Modified: trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h (248917 => 248918)
--- trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h 2019-08-20 21:33:06 UTC (rev 248917)
+++ trunk/Source/WebCore/platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h 2019-08-20 22:18:23 UTC (rev 248918)
@@ -40,7 +40,7 @@
class CDMPrivateMediaSourceAVFObjC;
-class CDMSessionAVStreamSession : public CDMSessionMediaSourceAVFObjC, public CanMakeWeakPtr {
+class CDMSessionAVStreamSession : public CDMSessionMediaSourceAVFObjC {
WTF_MAKE_FAST_ALLOCATED;
public:
CDMSessionAVStreamSession(Vector&& protocolVersions, CDMPrivateMediaSourceAVFObjC&, LegacyCDMSessionClient*);
Modified: trunk/Source/WebKit/ChangeLog (248917 => 248918)
--- trunk/Source/WebKit/ChangeLog 2019-08-20 21:33:06 UTC (rev 248917)
+++ trunk/Source/WebKit/ChangeLog 2019-08-20 22:18:23 UTC (rev 248918)
@@ -1,3 +1,19 @@
+2019-08-20 Chris Dumez
+
+WebSQLiteDatabaseTracker does not ensure it is still alive after dispatching to the main thread
+https://bugs.webkit.org/show_bug.cgi?id=200925
+
+Reviewed by Geoffrey Garen.
+
+WebSQLiteDatabaseTracker does not ensure it is still alive after dispatching to the main thread,
+which is not safe. Use WeakPtr to address the issue.
+
+* Shared/WebSQLiteDatabaseTracker.cpp:
+(WebKit::WebSQLiteDatabaseTracker::WebSQLiteDatabaseTracker):
+(WebKit::WebSQLiteDatabaseTracker::willBeginFirstTransaction):
+(WebKit::WebSQLiteDatabaseTracker::didFinishLastTransaction):
+* Shared/WebSQLiteDatabaseTracker.h:
+
2019-08-20 Brent Fulgham
[FTW] Fix scrolling in modern WebKit views
Modified: trunk/Source/WebKit/Shared/WebSQLiteDatabaseTracker.cpp (248917 => 248918)
---
[webkit-changes] [248917] trunk/LayoutTests
Title: [248917] trunk/LayoutTests Revision 248917 Author russel...@apple.com Date 2019-08-20 14:33:06 -0700 (Tue, 20 Aug 2019) Log Message Updating Expectations for Multiple Newly Passing Tests. Unreviewed Test Gardening. * platform/ios-12/TestExpectations: * platform/ios-wk1/TestExpectations: * platform/ios/TestExpectations: Modified Paths trunk/LayoutTests/ChangeLog trunk/LayoutTests/platform/ios/TestExpectations trunk/LayoutTests/platform/ios-12/TestExpectations trunk/LayoutTests/platform/ios-wk1/TestExpectations Diff Modified: trunk/LayoutTests/ChangeLog (248916 => 248917) --- trunk/LayoutTests/ChangeLog 2019-08-20 21:11:11 UTC (rev 248916) +++ trunk/LayoutTests/ChangeLog 2019-08-20 21:33:06 UTC (rev 248917) @@ -1,3 +1,13 @@ +2019-08-20 Russell Epstein + +Updating Expectations for Multiple Newly Passing Tests. + +Unreviewed Test Gardening. + +* platform/ios-12/TestExpectations: +* platform/ios-wk1/TestExpectations: +* platform/ios/TestExpectations: + 2019-08-16 Ryosuke Niwa The default tab index of output and fieldset should be -1 Modified: trunk/LayoutTests/platform/ios/TestExpectations (248916 => 248917) --- trunk/LayoutTests/platform/ios/TestExpectations 2019-08-20 21:11:11 UTC (rev 248916) +++ trunk/LayoutTests/platform/ios/TestExpectations 2019-08-20 21:33:06 UTC (rev 248917) @@ -3358,9 +3358,6 @@ http/tests/misc/iframe-beforeunload-dialog-not-matching-ancestor-securityorigin.html [ Pass Timeout ] http/tests/misc/iframe-beforeunload-dialog-block-modals.html [ Pass Timeout ] -# rdar://54049321 (REGRESSION: editing/pasteboard/paste-does-not-fire-promises-while-sanitizing-web-content.html Failing) -editing/pasteboard/paste-does-not-fire-promises-while-sanitizing-web-content.html [ Failure ] - # Skip until we have the fixes for and . fast/events/ios/multiple-key-press-and-release-ordering.html [ Skip ] Modified: trunk/LayoutTests/platform/ios-12/TestExpectations (248916 => 248917) --- trunk/LayoutTests/platform/ios-12/TestExpectations 2019-08-20 21:11:11 UTC (rev 248916) +++ trunk/LayoutTests/platform/ios-12/TestExpectations 2019-08-20 21:33:06 UTC (rev 248917) @@ -25,6 +25,3 @@ http/tests/misc/iframe-beforeunload-dialog-not-matching-ancestor-securityorigin.html [ Pass ] http/tests/misc/iframe-beforeunload-dialog-block-modals.html [ Pass ] -# rdar://54049321 (REGRESSION: editing/pasteboard/paste-does-not-fire-promises-while-sanitizing-web-content.html Failing) -editing/pasteboard/paste-does-not-fire-promises-while-sanitizing-web-content.html [ Pass ] - Modified: trunk/LayoutTests/platform/ios-wk1/TestExpectations (248916 => 248917) --- trunk/LayoutTests/platform/ios-wk1/TestExpectations 2019-08-20 21:11:11 UTC (rev 248916) +++ trunk/LayoutTests/platform/ios-wk1/TestExpectations 2019-08-20 21:33:06 UTC (rev 248917) @@ -1609,7 +1609,6 @@ fast/forms/input-text-option-delete.html [ Failure Pass ] fast/forms/number/number-input-changeevent.html [ Failure Pass ] fast/forms/onchange-enter-submit.html [ Failure Pass ] -fast/forms/plaintext-mode-2.html [ Failure Pass ] fast/forms/textinput-not-fired-on-enter-in-input.html [ Failure Pass ] fast/history/history-length.html [ Crash Pass ] fast/history/page-cache-geolocation-active-oneshot.html [ Failure Pass ] ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248916] trunk/Source/WebInspectorUI
Title: [248916] trunk/Source/WebInspectorUI Revision 248916 Author drou...@apple.com Date 2019-08-20 14:11:11 -0700 (Tue, 20 Aug 2019) Log Message Web Inspector: Sources: move the resource type scope bar to be next to the filter https://bugs.webkit.org/show_bug.cgi?id=200891 Reviewed by Joseph Pecoraro. It's odd to have UI for controlling the active filters in two different places. Move the resource type `WI.ScopeBar` to the filter bar area, and "promote" the resource grouping mode items from a context menu to an always visible `WI.ScopeBar` in the space left by the resource type `WI.ScopeBar` (switching between grouping modes quickly is a useful workflow). * UserInterface/Views/SourcesNavigationSidebarPanel.js: (WI.SourcesNavigationSidebarPanel): (WI.SourcesNavigationSidebarPanel.prototype.hasCustomFilters): (WI.SourcesNavigationSidebarPanel.prototype._handleResourceGroupingModeScopeBarSelectionChanged): Added. (WI.SourcesNavigationSidebarPanel.prototype._handleResourceGroupingModeChanged): (WI.SourcesNavigationSidebarPanel.prototype._populateResourceGroupingModeContextMenu.addOption): Deleted. (WI.SourcesNavigationSidebarPanel.prototype._populateResourceGroupingModeContextMenu): Deleted. * UserInterface/Views/SourcesNavigationSidebarPanel.css: (.sidebar > .panel.navigation.sources > .filter-bar .sources-resource-type-scope-bar.default-item-selected:not(:hover)): Added. (.sidebar > .panel.navigation.sources > .filter-bar .sources-resource-type-scope-bar.default-item-selected:hover): Added. * UserInterface/Views/FilterBar.js: (WI.FilterBar.prototype.addFilterNavigationItem): Added. (WI.FilterBar.prototype.addFilterBarButton): * UserInterface/Views/FilterBar.css: (.filter-bar > .navigation-bar > .item): (.filter-bar > .navigation-bar > .item.button): Added. (.filter-bar > .navigation-bar > .item.scope-bar): Added. (.filter-bar > input[type="search"] + .navigation-bar > .item.scope-bar:last-child): Added. Provide a way to add arbitrary `WI.NavigationItem` to the contained `WI.NavigationBar`. Slightly adjust the spacing of the items (depending on their type) in the `WI.NavigationBar` so they are all centered. * UserInterface/Views/ScopeBar.css: (.scope-bar): (.scope-bar > li): (.scope-bar > li::after): (.scope-bar > li:not(.selected):hover): Added. (body[dir=ltr] .scope-bar > li.multiple > select): (body[dir=rtl] .scope-bar > li.multiple > select): (.scope-bar > li:not(.selected):hover::after): Added. Introduce CSS variables for `margin`, `padding`, and `opacity` that callers can override to customize the appearance of the `WI.ScopeBar`. * Localizations/en.lproj/localizedStrings.js: Modified Paths trunk/Source/WebInspectorUI/ChangeLog trunk/Source/WebInspectorUI/Localizations/en.lproj/localizedStrings.js trunk/Source/WebInspectorUI/UserInterface/Views/FilterBar.css trunk/Source/WebInspectorUI/UserInterface/Views/FilterBar.js trunk/Source/WebInspectorUI/UserInterface/Views/ScopeBar.css trunk/Source/WebInspectorUI/UserInterface/Views/SourcesNavigationSidebarPanel.css trunk/Source/WebInspectorUI/UserInterface/Views/SourcesNavigationSidebarPanel.js Diff Modified: trunk/Source/WebInspectorUI/ChangeLog (248915 => 248916) --- trunk/Source/WebInspectorUI/ChangeLog 2019-08-20 20:40:32 UTC (rev 248915) +++ trunk/Source/WebInspectorUI/ChangeLog 2019-08-20 21:11:11 UTC (rev 248916) @@ -1,3 +1,51 @@ +2019-08-20 Devin Rousso + +Web Inspector: Sources: move the resource type scope bar to be next to the filter +https://bugs.webkit.org/show_bug.cgi?id=200891 + +Reviewed by Joseph Pecoraro. + +It's odd to have UI for controlling the active filters in two different places. Move the +resource type `WI.ScopeBar` to the filter bar area, and "promote" the resource grouping mode +items from a context menu to an always visible `WI.ScopeBar` in the space left by the +resource type `WI.ScopeBar` (switching between grouping modes quickly is a useful workflow). + +* UserInterface/Views/SourcesNavigationSidebarPanel.js: +(WI.SourcesNavigationSidebarPanel): +(WI.SourcesNavigationSidebarPanel.prototype.hasCustomFilters): +(WI.SourcesNavigationSidebarPanel.prototype._handleResourceGroupingModeScopeBarSelectionChanged): Added. +(WI.SourcesNavigationSidebarPanel.prototype._handleResourceGroupingModeChanged): +(WI.SourcesNavigationSidebarPanel.prototype._populateResourceGroupingModeContextMenu.addOption): Deleted. +(WI.SourcesNavigationSidebarPanel.prototype._populateResourceGroupingModeContextMenu): Deleted. +* UserInterface/Views/SourcesNavigationSidebarPanel.css: +(.sidebar > .panel.navigation.sources > .filter-bar .sources-resource-type-scope-bar.default-item-selected:not(:hover)): Added. +(.sidebar > .panel.navigation.sources > .filter-bar .sources-resource-type-scope-bar.default-item-selected:hover): Added. + +* UserInterface/Views/FilterBar.js: +
[webkit-changes] [248915] trunk/Source/WebCore
Title: [248915] trunk/Source/WebCore Revision 248915 Author justin_...@apple.com Date 2019-08-20 13:40:32 -0700 (Tue, 20 Aug 2019) Log Message Unreviewed comment follow-up for https://bugs.webkit.org/show_bug.cgi?id=200752. * html/canvas/GPUBasedCanvasRenderingContext.h: Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/html/canvas/GPUBasedCanvasRenderingContext.h Diff Modified: trunk/Source/WebCore/ChangeLog (248914 => 248915) --- trunk/Source/WebCore/ChangeLog 2019-08-20 19:59:00 UTC (rev 248914) +++ trunk/Source/WebCore/ChangeLog 2019-08-20 20:40:32 UTC (rev 248915) @@ -1,3 +1,9 @@ +2019-08-20 Justin Fan + +Unreviewed comment follow-up for https://bugs.webkit.org/show_bug.cgi?id=200752. + +* html/canvas/GPUBasedCanvasRenderingContext.h: + 2019-08-16 Ryosuke Niwa The default tab index of output and fieldset should be -1 Modified: trunk/Source/WebCore/html/canvas/GPUBasedCanvasRenderingContext.h (248914 => 248915) --- trunk/Source/WebCore/html/canvas/GPUBasedCanvasRenderingContext.h 2019-08-20 19:59:00 UTC (rev 248914) +++ trunk/Source/WebCore/html/canvas/GPUBasedCanvasRenderingContext.h 2019-08-20 20:40:32 UTC (rev 248915) @@ -53,7 +53,7 @@ virtual void markLayerComposited() = 0; protected: -GPUBasedCanvasRenderingContext(CanvasBase&); +explicit GPUBasedCanvasRenderingContext(CanvasBase&); HTMLCanvasElement* htmlCanvas() const; void notifyCanvasContentChanged(); ___ webkit-changes mailing list webkit-changes@lists.webkit.org https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248914] trunk
Title: [248914] trunk
Revision 248914
Author rn...@webkit.org
Date 2019-08-20 12:59:00 -0700 (Tue, 20 Aug 2019)
Log Message
The default tab index of output and fieldset should be -1
https://bugs.webkit.org/show_bug.cgi?id=200834
Reviewed by Alex Christensen.
Source/WebCore:
This patch updates the default tab index of output and fieldset to -1 to match the behavior of
Chrome and Firefox as well as the latest HTML specification:
https://html.spec.whatwg.org/multipage/interaction.html#the-tabindex-attribute
To do that, this patch replaces HTMLFormControlElement::defaultTabIndex with defaultTabIndex
implementation in each one of its concrete subclass such as HTMLButtonElement to mirror
the language of the specification as follows:
HTMLFormControlElement
HTMLButtonElement -> 0
HTMLFieldSetElement -> -1
HTMLFormControlElementWithState
HTMLKeygenElement -> 0 - Not specified anywhere but preserving the existing behavior.
HTMLSelectElement -> 0
HTMLTextFormControlElement
HTMLInputElement -> 0
HTMLTextAreaElement -> 0
HTMLOutputElement -> -1
Even though Element::shouldBeIgnoredInSequentialFocusNavigation() checks the value of
defaultTabIndex, this patch does not affect the actual sequential (tab) focus navigation:
Beacuse HTMLOutputElement and HTMLFieldSetElement have supportsFocus overrides to call
HTMLElement::supportsFocus instead of HTMLFormControlElement::supportsFocus, these two
elements are focusable only when tabindex is set or it's a root content editable element.
But all root editable elements are focusable anyway and the default tab index does not
matter when tabindex is set.
Test: fast/dom/tabindex-defaults.html
* html/HTMLButtonElement.cpp:
(WebCore::HTMLButtonElement::defaultTabIndex const): Added.
* html/HTMLButtonElement.h:
* html/HTMLFormControlElement.cpp:
(WebCore::HTMLFormControlElement::defaultTabIndex const): Deleted.
* html/HTMLFormControlElement.h:
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::defaultTabIndex const): Added.
* html/HTMLKeygenElement.h:
* html/HTMLKeygenElement.cpp:
(WebCore::HTMLKeygenElement::defaultTabIndex const): Added.
* html/HTMLKeygenElement.h:
* html/HTMLSelectElement.cpp:
(WebCore::HTMLSelectElement::defaultTabIndex const): Added.
* html/HTMLSelectElement.h:
* html/HTMLTextAreaElement.cpp:
(WebCore::HTMLTextAreaElement::defaultTabIndex const): Added.
* html/HTMLTextAreaElement.h:
LayoutTests:
Added test cases for output, fieldset, and keygen.
* fast/dom/tabindex-defaults-expected.txt:
* fast/dom/tabindex-defaults.html:
Modified Paths
trunk/LayoutTests/ChangeLog
trunk/LayoutTests/fast/dom/tabindex-defaults-expected.txt
trunk/LayoutTests/fast/dom/tabindex-defaults.html
trunk/LayoutTests/imported/w3c/web-platform-tests/html/infrastructure/common-dom-interfaces/collections/htmlformcontrolscollection.html
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/html/HTMLButtonElement.cpp
trunk/Source/WebCore/html/HTMLButtonElement.h
trunk/Source/WebCore/html/HTMLFormControlElement.cpp
trunk/Source/WebCore/html/HTMLFormControlElement.h
trunk/Source/WebCore/html/HTMLInputElement.cpp
trunk/Source/WebCore/html/HTMLInputElement.h
trunk/Source/WebCore/html/HTMLKeygenElement.cpp
trunk/Source/WebCore/html/HTMLKeygenElement.h
trunk/Source/WebCore/html/HTMLSelectElement.cpp
trunk/Source/WebCore/html/HTMLSelectElement.h
trunk/Source/WebCore/html/HTMLTextAreaElement.cpp
trunk/Source/WebCore/html/HTMLTextAreaElement.h
Diff
Modified: trunk/LayoutTests/ChangeLog (248913 => 248914)
--- trunk/LayoutTests/ChangeLog 2019-08-20 19:45:43 UTC (rev 248913)
+++ trunk/LayoutTests/ChangeLog 2019-08-20 19:59:00 UTC (rev 248914)
@@ -1,3 +1,15 @@
+2019-08-16 Ryosuke Niwa
+
+The default tab index of output and fieldset should be -1
+https://bugs.webkit.org/show_bug.cgi?id=200834
+
+Reviewed by Alex Christensen.
+
+Added test cases for output, fieldset, and keygen.
+
+* fast/dom/tabindex-defaults-expected.txt:
+* fast/dom/tabindex-defaults.html:
+
2019-08-20 Zalan Bujtas
[ContentChangeObserver] isConsideredClickable should be able to process elements with no renderers
Modified: trunk/LayoutTests/fast/dom/tabindex-defaults-expected.txt (248913 => 248914)
--- trunk/LayoutTests/fast/dom/tabindex-defaults-expected.txt 2019-08-20 19:45:43 UTC (rev 248913)
+++ trunk/LayoutTests/fast/dom/tabindex-defaults-expected.txt 2019-08-20 19:59:00 UTC (rev 248914)
@@ -8,22 +8,31 @@
PASS input.tabIndex is 0
PASS select.tabIndex is 0
PASS textarea.tabIndex is 0
+PASS keygen.tabIndex is 0
PASS editableDiv.tabIndex is 0
PASS normalDiv.tabIndex is -1
+PASS output.tabIndex is -1
+PASS fieldset.tabIndex is -1
PASS anchor.setAttribute("tabindex", "invalid"); anchor.tabIndex is 0
PASS button.setAttribute("tabindex", "invalid"); button.tabIndex is 0
PASS input.setAttribute("tabindex", "invalid"); input.tabIndex is 0
PASS select.setAttribute("tabindex",
[webkit-changes] [248913] trunk/Source/WTF
Title: [248913] trunk/Source/WTF
Revision 248913
Author sbar...@apple.com
Date 2019-08-20 12:45:43 -0700 (Tue, 20 Aug 2019)
Log Message
Unreviewed. Followup to r248903. It's not valid to remove
hasOverflowed() checks from appendCharacters.
* wtf/text/StringBuilder.cpp:
(WTF::StringBuilder::appendCharacters):
Modified Paths
trunk/Source/WTF/ChangeLog
trunk/Source/WTF/wtf/text/StringBuilder.cpp
Diff
Modified: trunk/Source/WTF/ChangeLog (248912 => 248913)
--- trunk/Source/WTF/ChangeLog 2019-08-20 19:17:59 UTC (rev 248912)
+++ trunk/Source/WTF/ChangeLog 2019-08-20 19:45:43 UTC (rev 248913)
@@ -1,3 +1,11 @@
+2019-08-20 Saam Barati
+
+Unreviewed. Followup to r248903. It's not valid to remove
+hasOverflowed() checks from appendCharacters.
+
+* wtf/text/StringBuilder.cpp:
+(WTF::StringBuilder::appendCharacters):
+
2019-08-20 Darin Adler
Variadic StringBuilder::append does not handle upconverting from 8-bit to 16-bit correctly
Modified: trunk/Source/WTF/wtf/text/StringBuilder.cpp (248912 => 248913)
--- trunk/Source/WTF/wtf/text/StringBuilder.cpp 2019-08-20 19:17:59 UTC (rev 248912)
+++ trunk/Source/WTF/wtf/text/StringBuilder.cpp 2019-08-20 19:45:43 UTC (rev 248913)
@@ -325,7 +325,7 @@
void StringBuilder::appendCharacters(const UChar* characters, unsigned length)
{
-if (!length)
+if (!length || hasOverflowed())
return;
ASSERT(characters);
@@ -347,7 +347,7 @@
void StringBuilder::appendCharacters(const LChar* characters, unsigned length)
{
-if (!length)
+if (!length || hasOverflowed())
return;
ASSERT(characters);
___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes
[webkit-changes] [248912] trunk/Source/WebInspectorUI
Title: [248912] trunk/Source/WebInspectorUI Revision 248912 Author pecor...@apple.com Date 2019-08-20 12:17:59 -0700 (Tue, 20 Aug 2019) Log Message Web Inspector: Sources: Give Origins their own icon in the Sources sidebar https://bugs.webkit.org/show_bug.cgi?id=200683 Reviewed by Devin Rousso. * UserInterface/Images/Origin.svg: Added. * UserInterface/Main.html: New resources. * UserInterface/Views/FolderIcon.css: (.origin-icon .icon): Light and Dark appearances for Origin icons. (@media (prefers-color-scheme: dark)): * UserInterface/Views/OriginTreeElement.js: (WI.OriginTreeElement): Very much like a folder with different classes. * UserInterface/Views/SourcesNavigationSidebarPanel.js: (WI.SourcesNavigationSidebarPanel.prototype.matchTreeElementAgainstCustomFilters.match): (WI.SourcesNavigationSidebarPanel.prototype.matchTreeElementAgainstCustomFilters): (WI.SourcesNavigationSidebarPanel.prototype._compareTreeElements): (WI.SourcesNavigationSidebarPanel.prototype._handleResourceGroupingModeChanged): Use OriginTreeElement in a few places. Modified Paths trunk/Source/WebInspectorUI/ChangeLog trunk/Source/WebInspectorUI/UserInterface/Main.html trunk/Source/WebInspectorUI/UserInterface/Views/FolderIcon.css trunk/Source/WebInspectorUI/UserInterface/Views/SourcesNavigationSidebarPanel.js Added Paths trunk/Source/WebInspectorUI/UserInterface/Images/Origin.svg trunk/Source/WebInspectorUI/UserInterface/Views/OriginTreeElement.js Diff Modified: trunk/Source/WebInspectorUI/ChangeLog (248911 => 248912) --- trunk/Source/WebInspectorUI/ChangeLog 2019-08-20 18:58:12 UTC (rev 248911) +++ trunk/Source/WebInspectorUI/ChangeLog 2019-08-20 19:17:59 UTC (rev 248912) @@ -1,5 +1,33 @@ 2019-08-20 Joseph Pecoraro +Web Inspector: Sources: Give Origins their own icon in the Sources sidebar +https://bugs.webkit.org/show_bug.cgi?id=200683 + + +Reviewed by Devin Rousso. + +* UserInterface/Images/Origin.svg: Added. +* UserInterface/Main.html: +New resources. + +* UserInterface/Views/FolderIcon.css: +(.origin-icon .icon): +Light and Dark appearances for Origin icons. + +(@media (prefers-color-scheme: dark)): +* UserInterface/Views/OriginTreeElement.js: +(WI.OriginTreeElement): +Very much like a folder with different classes. + +* UserInterface/Views/SourcesNavigationSidebarPanel.js: +(WI.SourcesNavigationSidebarPanel.prototype.matchTreeElementAgainstCustomFilters.match): +(WI.SourcesNavigationSidebarPanel.prototype.matchTreeElementAgainstCustomFilters): +(WI.SourcesNavigationSidebarPanel.prototype._compareTreeElements): +(WI.SourcesNavigationSidebarPanel.prototype._handleResourceGroupingModeChanged): +Use OriginTreeElement in a few places. + +2019-08-20 Joseph Pecoraro + Web Inspector: Support for _javascript_ BigInt https://bugs.webkit.org/show_bug.cgi?id=180731 Added: trunk/Source/WebInspectorUI/UserInterface/Images/Origin.svg (0 => 248912) --- trunk/Source/WebInspectorUI/UserInterface/Images/Origin.svg (rev 0) +++ trunk/Source/WebInspectorUI/UserInterface/Images/Origin.svg 2019-08-20 19:17:59 UTC (rev 248912) @@ -0,0 +1,13 @@ + + + + + + + + + + + + + Modified: trunk/Source/WebInspectorUI/UserInterface/Main.html (248911 => 248912) --- trunk/Source/WebInspectorUI/UserInterface/Main.html 2019-08-20 18:58:12 UTC (rev 248911) +++ trunk/Source/WebInspectorUI/UserInterface/Main.html 2019-08-20 19:17:59 UTC (rev 248912) @@ -737,6 +737,7 @@
[webkit-changes] [248911] trunk
Title: [248911] trunk
Revision 248911
Author za...@apple.com
Date 2019-08-20 11:58:12 -0700 (Tue, 20 Aug 2019)
Log Message
[ContentChangeObserver] isConsideredClickable should be able to process elements with no renderers
https://bugs.webkit.org/show_bug.cgi?id=200926
Reviewed by Simon Fraser.
Source/WebCore:
This patch fixes the crash when the visible->hidden change triggers an isConsideredClickable() call with a 'display: none' element.
Test: fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash.html
* page/ios/ContentChangeObserver.cpp:
(WebCore::isConsideredClickable):
LayoutTests:
* fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash-expected.txt: Added.
* fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash.html: Added.
Modified Paths
trunk/LayoutTests/ChangeLog
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/page/ios/ContentChangeObserver.cpp
Added Paths
trunk/LayoutTests/fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash-expected.txt
trunk/LayoutTests/fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash.html
Diff
Modified: trunk/LayoutTests/ChangeLog (248910 => 248911)
--- trunk/LayoutTests/ChangeLog 2019-08-20 18:55:34 UTC (rev 248910)
+++ trunk/LayoutTests/ChangeLog 2019-08-20 18:58:12 UTC (rev 248911)
@@ -1,3 +1,14 @@
+2019-08-20 Zalan Bujtas
+
+[ContentChangeObserver] isConsideredClickable should be able to process elements with no renderers
+https://bugs.webkit.org/show_bug.cgi?id=200926
+
+
+Reviewed by Simon Fraser.
+
+* fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash-expected.txt: Added.
+* fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash.html: Added.
+
2019-08-20 Wenson Hsieh
Clicking the search icon on ae.com hangs the web content process
Added: trunk/LayoutTests/fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash-expected.txt (0 => 248911)
--- trunk/LayoutTests/fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash-expected.txt (rev 0)
+++ trunk/LayoutTests/fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash-expected.txt 2019-08-20 18:58:12 UTC (rev 248911)
@@ -0,0 +1,2 @@
+PASS if no crash and 'clicked' text is shown below.
+clicked
Added: trunk/LayoutTests/fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash.html (0 => 248911)
--- trunk/LayoutTests/fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash.html (rev 0)
+++ trunk/LayoutTests/fast/events/touch/ios/content-observation/going-from-hidden-to-visible-and-to-hidden-crash.html 2019-08-20 18:58:12 UTC (rev 248911)
@@ -0,0 +1,50 @@
+
+
+
+This tests the case when visible and non-clickable content gets removed.
+
+#tapThis {
+width: 400px;
+height: 400px;
+border: 1px solid green;
+}
+
+#alreadyVisible {
+display: block;
+width: 100px;
+height: 100px;
+background-color: green;
+}
+
+