Context Navigation

← Previous Period
Next Period →

Timeline

May 7, 2019:

9:56 PM Changeset in webkit [245051] by ysuzuki@apple.com

3 edits
1 add in trunk

[JSC] DFG_ASSERT failed in lowInt52
https://bugs.webkit.org/show_bug.cgi?id=197569

Reviewed by Saam Barati.

JSTests:

stress/getstack-int52.js: Added.

(opt):
(main):

Source/JavaScriptCore:

GetStack with FlushedInt52 should load the flushed value in Int52 form and put the result in m_int52Values / m_strictInt52Values. Previously,
we load it in JSValue / Int32 form and lowInt52 fails to get appropriate one since GetStack does not put the result in m_int52Values / m_strictInt52Values.

ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileGetStack):

9:35 PM Changeset in webkit [245050] by ysuzuki@apple.com

11 edits
1 copy in trunk/Source

[JSC] LLIntPrototypeLoadAdaptiveStructureWatchpoint does not require Bag<>
https://bugs.webkit.org/show_bug.cgi?id=197645

Reviewed by Saam Barati.

Source/JavaScriptCore:

We are using HashMap<std::tuple<Structure*, const Instruction*>, Bag<LLIntPrototypeLoadAdaptiveStructureWatchpoint>> for LLIntPrototypeLoadAdaptiveStructureWatchpoint,
but this has several memory inefficiency.

Structure* and Instruction* are too large. We can just use StructureID and bytecodeOffset (unsigned).
While we are using Bag<>, we do not add a new LLIntPrototypeLoadAdaptiveStructureWatchpoint after constructing this Bag first. So we can use Vector<LLIntPrototypeLoadAdaptiveStructureWatchpoint> instead. We ensure that new entry won't be added to this Vector by making Watchpoint non-movable.
Instead of having OpGetById::Metadata&, we just hold unsigned bytecodeOffset, and get Metadata& from the owner CodeBlock when needed.

bytecode/CodeBlock.cpp:

(JSC::CodeBlock::finalizeLLIntInlineCaches):

bytecode/CodeBlock.h:
bytecode/LLIntPrototypeLoadAdaptiveStructureWatchpoint.cpp:

(JSC::LLIntPrototypeLoadAdaptiveStructureWatchpoint::LLIntPrototypeLoadAdaptiveStructureWatchpoint):
(JSC::LLIntPrototypeLoadAdaptiveStructureWatchpoint::fireInternal):

bytecode/LLIntPrototypeLoadAdaptiveStructureWatchpoint.h:
bytecode/Watchpoint.h:
llint/LLIntSlowPaths.cpp:

(JSC::LLInt::setupGetByIdPrototypeCache):

Source/WTF:

WTF.xcodeproj/project.pbxproj:
wtf/CMakeLists.txt:
wtf/Nonmovable.h: Copied from Source/JavaScriptCore/bytecode/LLIntPrototypeLoadAdaptiveStructureWatchpoint.h.
wtf/Vector.h:

(WTF::minCapacity>::uncheckedConstructAndAppend):

7:56 PM Changeset in webkit [245049] by don.olmstead@sony.com

4 edits in trunk/Source/WebCore

Fix !HAVE(ACCESSIBILITY) build
https://bugs.webkit.org/show_bug.cgi?id=197680

Reviewed by Fujii Hironori.

accessibility/AXObjectCache.h:

(WebCore::AXObjectCache::focusedUIElementForPage):
Update declaration for !HAVE(ACCESSIBILITY)

accessibility/AccessibilityObject.h:

Add wrapper implementation for !HAVE(ACCESSIBILITY)

accessibility/AccessibilityProgressIndicator.cpp:

(WebCore::AccessibilityProgressIndicator::roleValue const):
Add ENABLE(METER_ELEMENT) guard.

6:28 PM Changeset in webkit [245048] by Chris Dumez

3 edits in trunk/Source/WebKit

Simplify logic to prevent App Nap in WebPage
https://bugs.webkit.org/show_bug.cgi?id=197674

Reviewed by Geoff Garen.

Simplify logic to prevent App Nap in WebPage. We do not need both m_userActivityHysteresis and
m_userActivity since UserActivity is already a HysteresisActivity. We had 2 levels of
HysteresisActivity stacked on top of one another. Also rename "process suppression" to "app nap" as
I find it clearer.

WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::updateThrottleState):
(WebKit::WebPage::mouseEvent):
(WebKit::WebPage::wheelEvent):
(WebKit::WebPage::keyEvent):
(WebKit::WebPage::updatePreferences):
(WebKit::m_userActivityHysteresis): Deleted.
(WebKit::WebPage::updateUserActivity): Deleted.

WebProcess/WebPage/WebPage.h:

6:23 PM Changeset in webkit [245047] by ysuzuki@apple.com

4 edits
1 add in trunk

JSC: A bug in BytecodeGenerator::emitEqualityOpImpl
https://bugs.webkit.org/show_bug.cgi?id=197479

Reviewed by Saam Barati.

JSTests:

stress/do-not-perform-bytecode-peephole-optimization-in-jump-target.js: Added.

(shouldBe):

Source/JavaScriptCore:

Our peephole optimization in BytecodeGenerator is (1) rewinding the previous instruction and (2) emit optimized instruction instead.
If we have jump target between the previous instruction and the subsequent instruction, this peephole optimization breaks the jump target.
To prevent it, we had a mechanism disabling peephole optimization, setting m_lastOpcodeID = op_end and checking m_lastOpcodeID when performing
peephole optimization. However, BytecodeGenerator::emitEqualityOpImpl checks m_lastInstruction->is<OpTypeof> instead of m_lastOpcodeID == op_typeof,
and miss op_end case.

This patch makes the following changes.

Add canDoPeepholeOptimization method to clarify the intent of m_lastInstruction = op_end.
Check canDoPeepholeOptimization status before performing peephole optimization in emitJumpIfTrue, emitJumpIfFalse, and emitEqualityOpImpl.
Add ASSERT(canDoPeepholeOptimization()) in fuseCompareAndJump and fuseTestAndJmp to ensure that peephole optimization is allowed.

bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::fuseCompareAndJump):
(JSC::BytecodeGenerator::fuseTestAndJmp):
(JSC::BytecodeGenerator::emitJumpIfTrue):
(JSC::BytecodeGenerator::emitJumpIfFalse):
(JSC::BytecodeGenerator::emitEqualityOpImpl):

bytecompiler/BytecodeGenerator.h:

(JSC::BytecodeGenerator::canDoPeepholeOptimization const):

6:06 PM Changeset in webkit [245046] by commit-queue@webkit.org

2 edits in trunk/Source/WebKit

REGRESSION(r244230): Web Automation: use null instead of empty string as success result of scripts when alert is open
https://bugs.webkit.org/show_bug.cgi?id=197655

Patch by Carlos Garcia Campos <cgarcia@igalia.com> on 2019-05-07
Reviewed by Devin Rousso.

This regressed in r244230, the case of an alert open while running a script is now handled in WebKit side, but
an empty string is returned as a result, which is not a valid JSON serialized string. That made all user prompts
tests to fail.

UIProcess/Automation/WebAutomationSession.cpp:

(WebKit::WebAutomationSession::willShowJavaScriptDialog): Use null as script result instead of an empty string.

5:06 PM Changeset in webkit [245045] by Oriol Brufau

2 edits in trunk/Tools

Unreviewed. Add myself as a committer.

Scripts/webkitpy/common/config/contributors.json:

4:44 PM Changeset in webkit [245044] by Ryan Haddad

23 edits
6 deletes in trunk

Unreviewed, rolling out r245038.

Breaks internal builds.

Reverted changeset:

"Add SPI to set a list of hosts to which to send custom header
fields cross-origin"
https://bugs.webkit.org/show_bug.cgi?id=197397
https://trac.webkit.org/changeset/245038

4:43 PM Changeset in webkit [245043] by jiewen_tan@apple.com

20 edits
1 add in trunk

[WebAuthN] A new request should always suppress the pending request if any
https://bugs.webkit.org/show_bug.cgi?id=191517
<rdar://problem/46888222>

Reviewed by Brent Fulgham.

Source/WebCore:

Blocking new requests from the same page when there is a pending request could DoS the
WebAuthN API in the period between [the page is refreshed, the pending request is
hanedled/timeout]. Therefore, the policy will be to always cancel any pending requests
whenever a new request is made. This will enforce the policy of handling only one
request at a time.

Covered by new tests in existing files.

Modules/webauthn/AuthenticatorCoordinatorClient.cpp:

(WebCore::AuthenticatorCoordinatorClient::requestReply):
(WebCore::AuthenticatorCoordinatorClient::setRequestCompletionHandler):
(WebCore::AuthenticatorCoordinatorClient::addQueryCompletionHandler):

Modules/webauthn/AuthenticatorCoordinatorClient.h:

Source/WebKit:

Previously we blocked new WebAuthN requests whenever a pending request was in progress
to prevent background tabs from DoS foreground tabs. However, in r244938, the WebAuthN
API was changed to restrict request handling to the focused document. Therefore, we no
longer have a risk of DoS.

Apart from the vanished benefit, this behavoir actually blocks new pages to use
WebAuthN API in the period between [the previous initating page is closed, the pending
request is hanedled/timeout].

Also, it makes sense to have the current focused document preempt the pending request.
Therefore, the policy will be to always cancel any pending requests whenever a new
request is made. This will enforce the policy of handling only one request at a time.

Note that the current implementation doesn't explicitly cancel pending requests in the
Authenticators, which means that we could receive responses from the Authenticator that
were meant for a previous (now cancelled) request. A follow-up patch (see Bug 191523)
will implement an Authenticator feature to support immediate cancellation.

In the meantime, to protect the atomicity of the request/response pair, i.e., preventing an old
response being used for a new request, there are two safeguards:
1) In web process, each request to UI process is paired with an incremental ID, and therefore an old
response from UI process would have a different ID than the current request, which will then be ignored.
2) In UI process, all responses from authenticators will be piped to the main run loop for processing.
Therefore, when the new request comes in, the old response is either processed or waiting in the pipe.
To prevent the latter being processed, the new request will immediately destroy any authenticators bound
to the old response in the current run loop. Hence, in the next run loop when dealing the old response,
the lambda will have no where to hand the response over.

UIProcess/WebAuthentication/AuthenticatorManager.cpp:

(WebKit::AuthenticatorManager::makeCredential):
(WebKit::AuthenticatorManager::getAssertion):
(WebKit::AuthenticatorManager::clearStateAsync):
(WebKit::AuthenticatorManager::clearState):
(WebKit::AuthenticatorManager::timeOutTimerFired):

UIProcess/WebAuthentication/AuthenticatorManager.h:
UIProcess/WebAuthentication/WebAuthenticatorCoordinatorProxy.cpp:

(WebKit::WebAuthenticatorCoordinatorProxy::makeCredential):
(WebKit::WebAuthenticatorCoordinatorProxy::getAssertion):
(WebKit::WebAuthenticatorCoordinatorProxy::requestReply):

UIProcess/WebAuthentication/WebAuthenticatorCoordinatorProxy.h:
UIProcess/WebAuthentication/WebAuthenticatorCoordinatorProxy.messages.in:
WebProcess/WebAuthentication/WebAuthenticatorCoordinator.cpp:

(WebKit::WebAuthenticatorCoordinator::makeCredential):
(WebKit::WebAuthenticatorCoordinator::getAssertion):

WebProcess/WebAuthentication/WebAuthenticatorCoordinator.messages.in:

LayoutTests:

http/wpt/webauthn/public-key-credential-create-failure.https-expected.txt:
http/wpt/webauthn/public-key-credential-create-failure.https.html:
http/wpt/webauthn/public-key-credential-create-success-hid.https-expected.txt:
http/wpt/webauthn/public-key-credential-create-success-hid.https.html:
http/wpt/webauthn/public-key-credential-get-failure.https-expected.txt:
http/wpt/webauthn/public-key-credential-get-failure.https.html:
http/wpt/webauthn/public-key-credential-get-success-hid.https-expected.txt:
http/wpt/webauthn/public-key-credential-get-success-hid.https.html:
http/wpt/webauthn/resources/new-page.html: Added.

4:41 PM Changeset in webkit [245042] by jh718.park@samsung.com

2 edits in trunk/Tools

Unreviewed. Add myself as a committer.

Scripts/webkitpy/common/config/contributors.json:

4:16 PM Changeset in webkit [245041] by pvollan@apple.com

2 edits in trunk/Source/WebKit

[macOS] Crash with sandbox error
https://bugs.webkit.org/show_bug.cgi?id=197675
<rdar://problem/50473330>

Reviewed by Geoffrey Garen.

This crash is caused by missing syscall rules in the sandbox.

WebProcess/com.apple.WebProcess.sb.in:

3:15 PM Changeset in webkit [245040] by ysuzuki@apple.com

39 edits
12 adds in trunk

TemplateObject passed to template literal tags are not always identical for the same source location.
https://bugs.webkit.org/show_bug.cgi?id=190756

Reviewed by Saam Barati.

JSTests:

complex.yaml:
complex/tagged-template-regeneration-after.js: Added.

(shouldBe):

complex/tagged-template-regeneration.js: Added.

(call):
(test):

modules/tagged-template-inside-module.js: Added.

(from.string_appeared_here.call):

modules/tagged-template-inside-module/other-tagged-templates.js: Added.

(call):
(export.otherTaggedTemplates):

stress/call-and-construct-should-return-same-tagged-templates.js: Added.

(shouldBe):
(call):
(poly):

stress/tagged-templates-in-direct-eval-should-not-produce-same-site-object.js: Added.

(shouldBe):
(call):

stress/tagged-templates-in-function-in-direct-eval.js: Added.

(shouldBe):
(call):
(test):

stress/tagged-templates-in-global-function-should-not-produce-same-site-object.js: Added.

(shouldBe):
(call):

stress/tagged-templates-in-indirect-eval-should-not-produce-same-site-object.js: Added.

(shouldBe):
(call):

stress/tagged-templates-in-multiple-functions.js: Added.

(shouldBe):
(call):
(a):
(b):
(c):

stress/tagged-templates-with-same-start-offset.js: Added.

(shouldBe):

Source/JavaScriptCore:

Tagged template literal requires that the site object is allocated per source location. Previously, we create the site object
when linking CodeBlock and cache it in CodeBlock. But this is wrong because,

CodeBlock can be jettisoned and regenerated. So every time CodeBlock is regenerated, we get the different site object.
Call and Construct can have different CodeBlock. Even if the function is called in call-form or construct-form, we should return the same site object.

In this patch, we start caching these site objects in the top-level ScriptExecutable, this matches the spec's per source location since the only one top-level
ScriptExecutable is created for the given script code. Each ScriptExecutable of JSFunction can be created multiple times because CodeBlock creates it.
But the top-level one is not created by CodeBlock. This top-level ScriptExecutable is well-aligned to the Script itself. The top-level ScriptExecutable now has HashMap,
which maps source locations to cached site objects.

This patch threads the top-level ScriptExecutable to each FunctionExecutable creation. Each FunctionExecutable has a reference to the top-level ScriptExecutable.
We put TemplateObjectMap in ScriptExecutable, which manages cached template objects.
We move FunctionExecutable::m_cachedPolyProtoStructure to the FunctionExecutable::RareDate to keep FunctionExecutable 128 bytes.
TemplateObjectMap is indexed with endOffset of TaggedTemplate.

Scripts/tests/builtins/expected/JavaScriptCore-Builtin.Promise-Combined.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-Builtin.Promise-Separate.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-Builtin.prototype-Combined.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-Builtin.prototype-Separate.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-BuiltinConstructor-Combined.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-BuiltinConstructor-Separate.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-InternalClashingNames-Combined.js-result:
Scripts/tests/builtins/expected/WebCore-AnotherGuardedInternalBuiltin-Separate.js-result:
Scripts/tests/builtins/expected/WebCore-ArbitraryConditionalGuard-Separate.js-result:
Scripts/tests/builtins/expected/WebCore-GuardedBuiltin-Separate.js-result:
Scripts/tests/builtins/expected/WebCore-GuardedInternalBuiltin-Separate.js-result:
Scripts/tests/builtins/expected/WebCore-UnguardedBuiltin-Separate.js-result:
Scripts/tests/builtins/expected/WebCore-xmlCasingTest-Separate.js-result:
Scripts/wkbuiltins/builtins_templates.py:
bytecode/CodeBlock.cpp:

(JSC::CodeBlock::finishCreation):
(JSC::CodeBlock::setConstantRegisters):

bytecode/CodeBlock.h:
bytecode/UnlinkedFunctionExecutable.cpp:

(JSC::UnlinkedFunctionExecutable::link):

bytecode/UnlinkedFunctionExecutable.h:
bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::addTemplateObjectConstant):
(JSC::BytecodeGenerator::emitGetTemplateObject):

bytecompiler/BytecodeGenerator.h:
parser/ASTBuilder.h:

(JSC::ASTBuilder::createTaggedTemplate):

runtime/CachedTypes.cpp:

(JSC::CachedTemplateObjectDescriptor::encode):
(JSC::CachedTemplateObjectDescriptor::decode const):
(JSC::CachedJSValue::encode):
(JSC::CachedJSValue::decode const):

runtime/EvalExecutable.cpp:

(JSC::EvalExecutable::ensureTemplateObjectMap):
(JSC::EvalExecutable::visitChildren):

runtime/EvalExecutable.h:
runtime/FunctionExecutable.cpp:

(JSC::FunctionExecutable::finishCreation):
(JSC::FunctionExecutable::visitChildren):
(JSC::FunctionExecutable::fromGlobalCode):
(JSC::FunctionExecutable::ensureRareDataSlow):
(JSC::FunctionExecutable::ensureTemplateObjectMap):

runtime/FunctionExecutable.h:
runtime/JSModuleRecord.cpp:

(JSC::JSModuleRecord::instantiateDeclarations):

runtime/JSTemplateObjectDescriptor.cpp:

(JSC::JSTemplateObjectDescriptor::JSTemplateObjectDescriptor):
(JSC::JSTemplateObjectDescriptor::create):

runtime/JSTemplateObjectDescriptor.h:
runtime/ModuleProgramExecutable.cpp:

(JSC::ModuleProgramExecutable::ensureTemplateObjectMap):
(JSC::ModuleProgramExecutable::visitChildren):

runtime/ModuleProgramExecutable.h:
runtime/ProgramExecutable.cpp:

(JSC::ProgramExecutable::ensureTemplateObjectMap):
(JSC::ProgramExecutable::visitChildren):

runtime/ProgramExecutable.h:
runtime/ScriptExecutable.cpp:

(JSC::ScriptExecutable::topLevelExecutable):
(JSC::ScriptExecutable::createTemplateObject):
(JSC::ScriptExecutable::ensureTemplateObjectMapImpl):
(JSC::ScriptExecutable::ensureTemplateObjectMap):

runtime/ScriptExecutable.h:
tools/JSDollarVM.cpp:

(JSC::functionCreateBuiltin):
(JSC::functionDeleteAllCodeWhenIdle):
(JSC::JSDollarVM::finishCreation):

3:00 PM Changeset in webkit [245039] by eric.carlson@apple.com

22 edits
1 add in trunk

Define media buffering policy
https://bugs.webkit.org/show_bug.cgi?id=196979
<rdar://problem/28383861>

Reviewed by Jer Noble.

Source/WebCore:

Test: MediaBufferingPolicy API test.

html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::HTMLMediaElement):
(WebCore::HTMLMediaElement::suspend):
(WebCore::HTMLMediaElement::resume):
(WebCore::HTMLMediaElement::createMediaPlayer):
(WebCore::HTMLMediaElement::setBufferingPolicy):
(WebCore::HTMLMediaElement::purgeBufferedDataIfPossible):
(WebCore::HTMLMediaElement::bufferingPolicy const):
(WebCore::HTMLMediaElement::setShouldBufferData): Deleted.

html/HTMLMediaElement.h:

(WebCore::HTMLMediaElement::shouldBufferData const): Deleted.

html/MediaElementSession.cpp:

(WebCore::MediaElementSession::updateClientDataBuffering):
(WebCore::MediaElementSession::preferredBufferingPolicy const):
(WebCore::MediaElementSession::dataBufferingPermitted const): Deleted.

html/MediaElementSession.h:
platform/graphics/MediaPlayer.cpp:

(WebCore::MediaPlayer::setBufferingPolicy):
(WebCore::convertEnumerationToString):
(WebCore::MediaPlayer::setShouldBufferData): Deleted.

platform/graphics/MediaPlayer.h:
platform/graphics/MediaPlayerEnums.h:

(WTF::LogArgument<WebCore::MediaPlayerEnums::BufferingPolicy>::toString):

platform/graphics/MediaPlayerPrivate.h:

(WebCore::MediaPlayerPrivateInterface::setBufferingPolicy):
(WebCore::MediaPlayerPrivateInterface::setShouldBufferData): Deleted.

platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.h:
platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:

(WebCore::MediaPlayerPrivateAVFoundationObjC::MediaPlayerPrivateAVFoundationObjC):
(WebCore::MediaPlayerPrivateAVFoundationObjC::setBufferingPolicy):
(WebCore::MediaPlayerPrivateAVFoundationObjC::setShouldBufferData): Deleted.

platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaStreamAVFObjC.h:
platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaStreamAVFObjC.mm:

(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::setBufferingPolicy):
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::flushAndRemoveVideoSampleBuffers): Deleted.
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::setShouldBufferData): Deleted.

testing/Internals.cpp:

(WebCore::Internals::elementShouldBufferData):
(WebCore::Internals::elementBufferingPolicy):

testing/Internals.h:
testing/Internals.idl:

Source/WebCore/PAL:

pal/spi/mac/AVFoundationSPI.h:

Source/WTF:

wtf/Platform.h:

Tools:

TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
TestWebKitAPI/Tests/WebKitCocoa/MediaBufferingPolicy.mm: Added.

(waitUntilBufferingPolicyIsEqualTo):
(TEST):

2:51 PM Changeset in webkit [245038] by commit-queue@webkit.org

23 edits
1 copy
5 adds in trunk

Add SPI to set a list of hosts to which to send custom header fields cross-origin
https://bugs.webkit.org/show_bug.cgi?id=197397

Patch by Alex Christensen <achristensen@webkit.org> on 2019-05-07
Reviewed by Geoff Garen.

Source/WebCore:

In r223001 I added the ability to send custom headers, but with a restriction that they will not be sent except to the origin of the main document.
We need the ability to specify what origins to send these headers to even if they are not first party requests.
We get this information in a list of strings which are the hosts to send the headers to. Some of the strings have an asterisk at the beginning,
indicating that the headers are to be sent to all subdomains.

I repurposed some ObjC SPI that was never adopted, but I keep testing the C API that was to verify no regression.
I also added some new API tests for the new behavior.

Sources.txt:
WebCore.xcodeproj/project.pbxproj:
loader/CustomHeaderFields.cpp: Added.

(WebCore::CustomHeaderFields::thirdPartyDomainsMatch const):

loader/CustomHeaderFields.h: Added.

(WebCore::CustomHeaderFields::encode const):
(WebCore::CustomHeaderFields::decode):

loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::setCustomHeaderFields): Deleted.

loader/DocumentLoader.h:

(WebCore::DocumentLoader::setCustomHeaderFields):
(WebCore::DocumentLoader::customHeaderFields const):
(WebCore::DocumentLoader::customHeaderFields): Deleted.

loader/cache/CachedResourceLoader.cpp:

(WebCore::CachedResourceLoader::requestResource):

Source/WebKit:

Shared/API/APIObject.h:
Shared/Cocoa/APIObject.mm:

(API::Object::newObject):

Shared/WebsitePoliciesData.cpp:

(WebKit::WebsitePoliciesData::decode):

Shared/WebsitePoliciesData.h:
SourcesCocoa.txt:
UIProcess/API/APICustomHeaderFields.h: Added.
UIProcess/API/APIWebsitePolicies.cpp:

(API::WebsitePolicies::WebsitePolicies):
(API::WebsitePolicies::copy const):
(API::WebsitePolicies::data):

UIProcess/API/APIWebsitePolicies.h:
UIProcess/API/C/WKWebsitePolicies.cpp:

(WKWebsitePoliciesCopyCustomHeaderFields):
(WKWebsitePoliciesSetCustomHeaderFields):

UIProcess/API/Cocoa/WKWebpagePreferences.mm:

(-[WKWebpagePreferences _customHeaderFields]):
(-[WKWebpagePreferences _setCustomHeaderFields:]):

UIProcess/API/Cocoa/WKWebpagePreferencesPrivate.h:
UIProcess/API/Cocoa/_WKCustomHeaderFields.h: Added.
UIProcess/API/Cocoa/_WKCustomHeaderFields.mm: Added.

(-[_WKCustomHeaderFields init]):
(-[_WKCustomHeaderFields dealloc]):
(-[_WKCustomHeaderFields fields]):
(-[_WKCustomHeaderFields setFields:]):
(-[_WKCustomHeaderFields thirdPartyDomains]):
(-[_WKCustomHeaderFields setThirdPartyDomains:]):
(-[_WKCustomHeaderFields _apiObject]):

UIProcess/API/Cocoa/_WKCustomHeaderFieldsInternal.h: Added.
UIProcess/API/Cocoa/_WKWebsitePolicies.h:
UIProcess/API/Cocoa/_WKWebsitePolicies.mm:

(-[_WKWebsitePolicies customHeaderFields]): Deleted.
(-[_WKWebsitePolicies setCustomHeaderFields:]): Deleted.

UIProcess/Cocoa/WebViewImpl.h:
UIProcess/Cocoa/WebViewImpl.mm:

(WebKit::WebViewImpl::takeFocus):
(WebKit::WebViewImpl::accessibilityAttributeValue):

WebKit.xcodeproj/project.pbxproj:

Tools:

TestWebKitAPI/Tests/WebKitCocoa/WebsitePolicies.mm:

(TEST):
(expectLegacyHeaders):
(expectHeaders):
(-[CustomHeaderFieldsDelegate webView:decidePolicyForNavigationAction:preferences:decisionHandler:]):
(-[CustomHeaderFieldsDelegate webView:startURLSchemeTask:]):
(-[CustomHeaderFieldsDelegate _webView:decidePolicyForNavigationAction:decisionHandler:]): Deleted.

2:38 PM Changeset in webkit [245037] by Keith Rollin

5 edits in trunk

Add option to build-webkit to control whether or not XCBuild is used
https://bugs.webkit.org/show_bug.cgi?id=197668
<rdar://problem/50549728>

Reviewed by Tim Horton.

Add --[no-]xcbuild to build-webkit to force the use of XCBuild or not.
Also update build-webkit and the makefiles with the foundation for
automatically using XCBuild when the conditions allow it. This latter
facility is currently turned off until Xcode fully supports building
WebKit with XCBuild.

Makefile.shared:

Tools:

Scripts/build-webkit:
Scripts/webkitdirs.pm:

(XcodeOptions):
(canUseXCBuild):

2:33 PM Changeset in webkit [245036] by aestes@apple.com

5 edits in trunk

run-bindings-tests should test global scope constructor generation
https://bugs.webkit.org/show_bug.cgi?id=197669

Reviewed by Alex Christensen.

Source/WebCore:

For interfaces that are exposed on a global object, preprocess-idls.pl generates a partial
interface for the global object defining attributes for the interfaces' constructors. Most
interfaces don't specify a global object, so preprocess-idls.pl defaults to DOMWindow.
Since there is no DOMWindow.idl test case, we never generate the code for exposed interface
constructors when running bindings tests. This means that we can't test changes to how these
constructors are generated.

To fix this, teach preprocess-idls.pl to treat 'TestGlobalObject' as the default global
object when running bindings tests. This means that all exposed interface test cases will
generate their constructors as part of JSTestGlobalObject (unless otherwise specified
by the 'Exposed' extended attribute).

bindings/scripts/preprocess-idls.pl:

Added --testGlobalContextName and --testGlobalScopeConstructorsFile arguments for use by
run-bindings-tests.

bindings/scripts/test/JS/JSTestGlobalObject.cpp:

Updated expected results.

Tools:

Scripts/webkitpy/bindings/main.py:

Updated to specify --testGlobalContextName and --testGlobalScopeConstructorsFile when
invoking preprocess-idls.pl.

2:28 PM Changeset in webkit [245035] by rmorisset@apple.com

10 edits in trunk/Source/JavaScriptCore

[B3] Constants should be hoisted to the root block until moveConstants
https://bugs.webkit.org/show_bug.cgi?id=197265

Reviewed by Saam Barati.

This patch does the following:

B3ReduceStrength now hoists all constants to the root BB, and de-duplicates them along the way
B3PureCSE no longer bothers with constants, since they are already de-duplicated by the time it gets to see them
We now run eliminateDeadCode just after moveConstants, so that the Nops that moveConstants generates are freed instead of staying live throughout Air compilation, reducing memory pressure.
I also took the opportunity to fix typos in comments in various parts of the code base.

Here are a few numbers to justify this patch:

In JetStream2, about 27% of values at the beginning of B3 are constants
In JetStream2, about 11% of values at the end of B3 are Nops
In JetStream2, this patch increases the number of times that tail duplication happens from a bit less than 24k to a bit more than 25k (hoisting constants makes blocks smaller).

When I tried measuring the total effect on JetStream2 I got a tiny and almost certainly non-significant progression.

b3/B3Generate.cpp:

(JSC::B3::generateToAir):

b3/B3MoveConstants.cpp:
b3/B3PureCSE.cpp:

(JSC::B3::PureCSE::process):

b3/B3PureCSE.h:
b3/B3ReduceStrength.cpp:
bytecode/GetByIdStatus.cpp:

(JSC::GetByIdStatus::computeForStubInfoWithoutExitSiteFeedback):

dfg/DFGCSEPhase.cpp:
dfg/DFGOSRAvailabilityAnalysisPhase.h:
dfg/DFGOSRExit.cpp:

(JSC::DFG::OSRExit::executeOSRExit):

2:09 PM Changeset in webkit [245034] by youenn@apple.com

2 edits in trunk/Tools

Flaky API Test: TestWebKitAPI.ServiceWorkers.ServiceWorkerAndCacheStorageDefaultDirectories
https://bugs.webkit.org/show_bug.cgi?id=195997
<rdar://problem/50509884>

Reviewed by Chris Dumez.

Spin test until successful.

TestWebKitAPI/Tests/WebKitCocoa/ServiceWorkerBasic.mm:

2:07 PM Changeset in webkit [245033] by youenn@apple.com

6 edits in trunk

Video stream freeze on front camera orientation changing
https://bugs.webkit.org/show_bug.cgi?id=197227
<rdar://problem/50175498>

Reviewed by Eric Carlson.

Source/WebCore:

Use m_currentRotationSessionAngle instead of m_currentRotation to create or not a new rotation session.
Covered by updated test.

platform/mediastream/mac/RealtimeOutgoingVideoSourceCocoa.h:
platform/mediastream/mac/RealtimeOutgoingVideoSourceCocoa.mm:

(WebCore::RealtimeOutgoingVideoSourceCocoa::rotatePixelBuffer):

LayoutTests:

webrtc/video-rotation-expected.txt:
webrtc/video-rotation.html:

2:04 PM Changeset in webkit [245032] by youenn@apple.com

2 edits in trunk/Source/WebCore

getUserMedia framerate unusable under low light in iOS 12.2
https://bugs.webkit.org/show_bug.cgi?id=196214
<rdar://problem/49232193>

Reviewed by Geoffrey Garen.

When setting the frame rate, set it to the exact value instead of a range.
Otherwise, the capture device might use the lowest frame rate according the light conditions
for best picture quality which is not what is expected by most web pages.

Move frame rate range computation to closer where actually used.
Since frame rate matching is fuzzy, add some checks in case the expected frame rate is slightly out of min/max range.

Manually tested on a real device.

platform/mediastream/mac/AVVideoCaptureSource.mm:

(WebCore::AVVideoCaptureSource::setSizeAndFrameRateWithPreset):

1:59 PM Changeset in webkit [245031] by rmorisset@apple.com

3 edits in trunk

All prototypes should call didBecomePrototype()
https://bugs.webkit.org/show_bug.cgi?id=196315

Reviewed by Saam Barati.

JSTests:

stress/function-prototype-indexed-accessor.js: Added.

Source/JavaScriptCore:

Otherwise we won't remember to run haveABadTime() when someone adds to them an indexed accessor.

I added a check used in both Structure::finishCreation() and Structure::changePrototypeTransition to make sure we don't
create structures with invalid prototypes.
It found a lot of objects that are used as prototypes in JSGlobalObject and yet were missing didBecomePrototype() in their finishCreation().
Somewhat surprisingly, some of them have names like FunctionConstructor and not only FooPrototype.

runtime/BigIntPrototype.cpp:

(JSC::BigIntPrototype::finishCreation):

runtime/BooleanPrototype.cpp:

(JSC::BooleanPrototype::finishCreation):

runtime/DatePrototype.cpp:

(JSC::DatePrototype::finishCreation):

runtime/ErrorConstructor.cpp:

(JSC::ErrorConstructor::finishCreation):

runtime/ErrorPrototype.cpp:

(JSC::ErrorPrototype::finishCreation):

runtime/FunctionConstructor.cpp:

(JSC::FunctionConstructor::finishCreation):

runtime/FunctionPrototype.cpp:

(JSC::FunctionPrototype::finishCreation):

runtime/IntlCollatorPrototype.cpp:

(JSC::IntlCollatorPrototype::finishCreation):

runtime/IntlDateTimeFormatPrototype.cpp:

(JSC::IntlDateTimeFormatPrototype::finishCreation):

runtime/IntlNumberFormatPrototype.cpp:

(JSC::IntlNumberFormatPrototype::finishCreation):

runtime/IntlPluralRulesPrototype.cpp:

(JSC::IntlPluralRulesPrototype::finishCreation):

runtime/JSArrayBufferPrototype.cpp:

(JSC::JSArrayBufferPrototype::finishCreation):

runtime/JSDataViewPrototype.cpp:

(JSC::JSDataViewPrototype::finishCreation):

runtime/JSGenericTypedArrayViewPrototypeInlines.h:

(JSC::JSGenericTypedArrayViewPrototype<ViewClass>::finishCreation):

runtime/JSGlobalObject.cpp:

(JSC::createConsoleProperty):

runtime/JSPromisePrototype.cpp:

(JSC::JSPromisePrototype::finishCreation):

runtime/JSTypedArrayViewConstructor.cpp:

(JSC::JSTypedArrayViewConstructor::finishCreation):

runtime/JSTypedArrayViewPrototype.cpp:

(JSC::JSTypedArrayViewPrototype::finishCreation):

runtime/NumberPrototype.cpp:

(JSC::NumberPrototype::finishCreation):

runtime/RegExpPrototype.cpp:

(JSC::RegExpPrototype::finishCreation):

runtime/StringPrototype.cpp:

(JSC::StringPrototype::finishCreation):

runtime/Structure.cpp:

(JSC::Structure::isValidPrototype):
(JSC::Structure::changePrototypeTransition):

runtime/Structure.h:
runtime/SymbolPrototype.cpp:

(JSC::SymbolPrototype::finishCreation):

wasm/js/WebAssemblyCompileErrorPrototype.cpp:

(JSC::WebAssemblyCompileErrorPrototype::finishCreation):

wasm/js/WebAssemblyInstancePrototype.cpp:

(JSC::WebAssemblyInstancePrototype::finishCreation):

wasm/js/WebAssemblyLinkErrorPrototype.cpp:

(JSC::WebAssemblyLinkErrorPrototype::finishCreation):

wasm/js/WebAssemblyMemoryPrototype.cpp:

(JSC::WebAssemblyMemoryPrototype::finishCreation):

wasm/js/WebAssemblyModulePrototype.cpp:

(JSC::WebAssemblyModulePrototype::finishCreation):

wasm/js/WebAssemblyPrototype.cpp:

(JSC::WebAssemblyPrototype::finishCreation):

wasm/js/WebAssemblyRuntimeErrorPrototype.cpp:

(JSC::WebAssemblyRuntimeErrorPrototype::finishCreation):

wasm/js/WebAssemblyTablePrototype.cpp:

(JSC::WebAssemblyTablePrototype::finishCreation):

Source/WebCore:

It was found by existing tests, with the new assert in JSC::Structure

bindings/js/JSWindowProxy.cpp:

(WebCore::JSWindowProxy::setWindow):

bindings/scripts/CodeGeneratorJS.pm:

(GeneratePrototypeDeclaration):
(GenerateConstructorHelperMethods):

1:46 PM Changeset in webkit [245030] by rmorisset@apple.com

5 edits in trunk/Source

WTF::BitVector should have an isEmpty() method
https://bugs.webkit.org/show_bug.cgi?id=197637

Reviewed by Keith Miller.

Source/JavaScriptCore:

Just replaces some comparison of bitCount() to 0 by calls to isEmpty()

b3/air/AirAllocateRegistersByGraphColoring.cpp:

Source/WTF:

wtf/BitVector.cpp:

(WTF::BitVector::isEmptySlow const):

wtf/BitVector.h:

(WTF::BitVector::isEmpty const):

1:32 PM Changeset in webkit [245029] by Alan Coon

20 edits in tags/Safari-608.1.23

Revert r244952. rdar://problem/50552733

1:30 PM Changeset in webkit [245028] by Brent Fulgham

6 edits
3 adds in trunk

Correct JSON parser to address unterminated escape character
https://bugs.webkit.org/show_bug.cgi?id=197582
<rdar://problem/50459177>

Reviewed by Alex Christensen.

Source/WTF:

Correct JSON parser code to properly deal with unterminated escape
characters.

wtf/JSONValues.cpp:

(WTF::JSONImpl::decodeString):
(WTF::JSONImpl::parseStringToken):

LayoutTests:

applicationmanifest/display-mode-bad-manifest-expected.txt:
applicationmanifest/display-mode-bad-manifest.html:
applicationmanifest/resources/bad.manifest: Added.
js/resources/JSON-parse.js: Add test case for unterminated escape.
js/dom/JSON-parse-expected.txt: Add new test case.
TestWebKitAPI/Tests/WTF/JSONValue.cpp: Add new false test case for unterminated escape character.

1:29 PM Changeset in webkit [245027] by Adrian Perez de Castro

2 edits in trunk/Source/WebKit

[GTK][WPE] Cannot build documentation with gtk-doc
https://bugs.webkit.org/show_bug.cgi?id=197663

Reviewed by Philippe Normand.

UIProcess/API/glib/WebKitSettings.cpp:

(webkit_settings_class_init): Escape content which gets wrongly interpreted as
SGML tags by gtk-doc as Markdown preformatted content, which fits well for the
tag names in the documentation for the WebKitSettings:enable-media property.

1:18 PM WebKitGTK/GCCRequirement edited by Michael Catanzaro

(diff)

12:57 PM Changeset in webkit [245026] by commit-queue@webkit.org

38 edits
10 deletes in trunk

Unreviewed, rolling out r244978.
https://bugs.webkit.org/show_bug.cgi?id=197671

TemplateObject map should use start/end offsets (Requested by
yusukesuzuki on #webkit).

Reverted changeset:

"TemplateObject passed to template literal tags are not always
identical for the same source location."
https://bugs.webkit.org/show_bug.cgi?id=190756
https://trac.webkit.org/changeset/244978

12:55 PM Changeset in webkit [245025] by wilander@apple.com

44 edits
2 copies
2 moves
7 adds in trunk

Storage Access API: Make two changes requested by developers and complete refactoring and cleanup
https://bugs.webkit.org/show_bug.cgi?id=197648
<rdar://problem/50527493>

Reviewed by Chris Dumez.

Source/WebCore:

Developers have requested two minor changes to the Storage Access API:

Only consume the user gesture when the user explicitly denies access.
Make document.hasStorageAccess() return true instead of false when the feature is off.

In addition to this, we have refactoring and cleanup to do. Namely:

Make use of WebCore::RegistrableDomain all the way.
Remove dead code in WebKit::NetworkProcess since the calls now go through NetworkConnectionToWebProcess.
Introduce boolean enums for state handling.
Break out the Storage Access API functionality into a supplement of WebCore::Document.

Reviewed by Chris Dumez.

Tests: http/tests/storageAccess/deny-with-prompt-does-not-preserve-gesture.html

http/tests/storageAccess/deny-without-prompt-preserves-gesture.html
http/tests/storageAccess/grant-with-prompt-preserves-gesture.html
http/tests/storageAccess/has-storage-access-true-if-feature-off.html

DerivedSources-input.xcfilelist:
DerivedSources-output.xcfilelist:
DerivedSources.make:
Headers.cmake:
Sources.txt:
WebCore.xcodeproj/project.pbxproj:
dom/Document.cpp:

(WebCore::Document::hasStorageAccess): Deleted.
(WebCore::Document::requestStorageAccess): Deleted.
(WebCore::Document::enableTemporaryTimeUserGesture): Deleted.
(WebCore::Document::consumeTemporaryTimeUserGesture): Deleted.
(WebCore::Document::hasFrameSpecificStorageAccess const): Deleted.
(WebCore::Document::setHasFrameSpecificStorageAccess): Deleted.

dom/Document.h:

(WebCore::Document::setUserGrantsStorageAccessOverride): Deleted.

All of this has been moved to the supplement WebCore::DocumentStorageAccess.

dom/Document.idl:

The Storage Access API has been moved to DocumentStorageAccess.idl.

dom/DocumentStorageAccess.cpp: Added.

(WebCore::DocumentStorageAccess::from):
(WebCore::DocumentStorageAccess::supplementName):
(WebCore::DocumentStorageAccess::hasStorageAccess):
(WebCore::DocumentStorageAccess::requestStorageAccess):
(WebCore::DocumentStorageAccess::enableTemporaryTimeUserGesture):
(WebCore::DocumentStorageAccess::consumeTemporaryTimeUserGesture):
(WebCore::DocumentStorageAccess::hasFrameSpecificStorageAccess const):
(WebCore::DocumentStorageAccess::setHasFrameSpecificStorageAccess):

dom/DocumentStorageAccess.h: Added.
dom/DocumentStorageAccess.idl: Added.
page/ChromeClient.h:
testing/Internals.cpp:

(WebCore::Internals::setUserGrantsStorageAccess): Deleted.

This was dead code.

testing/Internals.h:
testing/Internals.idl:

Source/WebKit:

Developers have requested two minor changes to the Storage Access API:

Only consume the user gesture when the user explicitly denies access.
Make document.hasStorageAccess() return true instead of false when the feature is off.

In addition to this, we have refactoring and cleanup to do. Namely:

Make use of WebCore::RegistrableDomain all the way.
Remove dead code in WebKit::NetworkProcess since the calls now go through NetworkConnectionToWebProcess.
Introduce boolean enums for state handling.
Break out the Storage Access API functionality into a supplement of WebCore::Document.

NetworkProcess/Classifier/ResourceLoadStatisticsDatabaseStore.cpp:

(WebKit::ResourceLoadStatisticsDatabaseStore::insertDomainRelationships):
(WebKit::ResourceLoadStatisticsDatabaseStore::requestStorageAccess):
(WebKit::ResourceLoadStatisticsDatabaseStore::requestStorageAccessUnderOpener):
(WebKit::ResourceLoadStatisticsDatabaseStore::grantStorageAccess):
(WebKit::ResourceLoadStatisticsDatabaseStore::grantStorageAccessInternal):
(WebKit::ResourceLoadStatisticsDatabaseStore::hasUserGrantedStorageAccessThroughPrompt const):

These changes are due to the new enums WebCore::StorageAccessWasGranted and
WebCore::StorageAccessPromptWasShown.

NetworkProcess/Classifier/ResourceLoadStatisticsDatabaseStore.h:
NetworkProcess/Classifier/ResourceLoadStatisticsMemoryStore.cpp:

(WebKit::ResourceLoadStatisticsMemoryStore::requestStorageAccess):
(WebKit::ResourceLoadStatisticsMemoryStore::requestStorageAccessUnderOpener):
(WebKit::ResourceLoadStatisticsMemoryStore::grantStorageAccess):
(WebKit::ResourceLoadStatisticsMemoryStore::grantStorageAccessInternal):
(WebKit::ResourceLoadStatisticsMemoryStore::hasUserGrantedStorageAccessThroughPrompt):

These changes are due to the new enums WebCore::StorageAccessWasGranted and
WebCore::StorageAccessPromptWasShown.

NetworkProcess/Classifier/ResourceLoadStatisticsMemoryStore.h:
NetworkProcess/Classifier/ResourceLoadStatisticsStore.h:

These changes are due to the new enums WebCore::StorageAccessWasGranted and
WebCore::StorageAccessPromptWasShown.

NetworkProcess/Classifier/WebResourceLoadStatisticsStore.cpp:

(WebKit::WebResourceLoadStatisticsStore::requestStorageAccess):
(WebKit::WebResourceLoadStatisticsStore::grantStorageAccess):
(WebKit::WebResourceLoadStatisticsStore::callGrantStorageAccessHandler):

These changes are due to the new enums WebCore::StorageAccessWasGranted and
WebCore::StorageAccessPromptWasShown.

(WebKit::WebResourceLoadStatisticsStore::requestStorageAccessGranted): Deleted.

This function is now no longer exposed and its functionality could be folded into
the existing WebResourceLoadStatisticsStore::requestStorageAccess() which is more
clearly named.

NetworkProcess/Classifier/WebResourceLoadStatisticsStore.h:
NetworkProcess/NetworkConnectionToWebProcess.cpp:

(WebKit::NetworkConnectionToWebProcess::hasStorageAccess):
(WebKit::NetworkConnectionToWebProcess::requestStorageAccess):

These changes are due to the new enums WebCore::StorageAccessWasGranted and
WebCore::StorageAccessPromptWasShown.

NetworkProcess/NetworkConnectionToWebProcess.h:
NetworkProcess/NetworkConnectionToWebProcess.messages.in:
NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::hasStorageAccessForFrame): Deleted.
(WebKit::NetworkProcess::hasStorageAccess): Deleted.
(WebKit::NetworkProcess::requestStorageAccess): Deleted.
(WebKit::NetworkProcess::requestStorageAccessGranted): Deleted.
(WebKit::NetworkProcess::grantStorageAccess): Deleted.
(WebKit::NetworkProcess::removeAllStorageAccess): Deleted.

These functions were left behind in the move of ITP to the network process.
This communication goes through WebKit::NetworkConnectionToWebProcess since a while back.

NetworkProcess/NetworkProcess.h:
NetworkProcess/NetworkProcess.messages.in:
Scripts/webkit/messages.py:

Instructions for derived IPC code on how to include the new enums
WebCore::StorageAccessWasGranted and WebCore::StorageAccessPromptWasShown.

UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::hasStorageAccessForFrame): Deleted.
(WebKit::NetworkProcessProxy::hasStorageAccess): Deleted.
(WebKit::NetworkProcessProxy::requestStorageAccess): Deleted.
(WebKit::NetworkProcessProxy::grantStorageAccess): Deleted.
(WebKit::NetworkProcessProxy::removeAllStorageAccess): Deleted.

These functions were left behind in the move of ITP to the network process.
This communication goes through WebKit::NetworkConnectionToWebProcess since a while back.

UIProcess/Network/NetworkProcessProxy.h:
UIProcess/WebsiteData/WebsiteDataStore.cpp:

(WebKit::WebsiteDataStore::hasStorageAccess): Deleted.
(WebKit::WebsiteDataStore::requestStorageAccess): Deleted.
(WebKit::WebsiteDataStore::grantStorageAccess): Deleted.

These functions were left behind in the move of ITP to the network process.
This communication goes through WebKit::NetworkConnectionToWebProcess since a while back.

UIProcess/WebsiteData/WebsiteDataStore.h:
WebProcess/WebCoreSupport/WebChromeClient.cpp:

(WebKit::WebChromeClient::hasStorageAccess):
(WebKit::WebChromeClient::requestStorageAccess):

These changes are due to the new enums WebCore::StorageAccessWasGranted and
WebCore::StorageAccessPromptWasShown. They also receive WebCore::RegistrableDomain
objects instead of Strings now.

WebProcess/WebCoreSupport/WebChromeClient.h:
WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::hasStorageAccess):
(WebKit::WebPage::requestStorageAccess):

These changes are due to the new enums WebCore::StorageAccessWasGranted and
WebCore::StorageAccessPromptWasShown. They also receive WebCore::RegistrableDomain
objects instead of Strings now.

WebProcess/WebPage/WebPage.h:

LayoutTests:

The changed test cases have had calls to WebCore::Internals::setUserGrantsStorageAccess() removed
since it was dead code.

One of the new tests, deny-with-prompt-does-not-preserve-gesture.html, is marked [ Skip ] for now since
we lack the ability to click "Don't allow" in the prompt. I wanted to include the test anyway so that
we have it. I have done a manual test to make sure the code does the right thing for this case.

http/tests/storageAccess/deny-with-prompt-does-not-preserve-gesture-expected.txt: Added.
http/tests/storageAccess/deny-with-prompt-does-not-preserve-gesture.html: Copied from LayoutTests/http/tests/storageAccess/request-and-grant-access-cross-origin-non-sandboxed-iframe-pop-window.html.
http/tests/storageAccess/deny-without-prompt-preserves-gesture-expected.txt: Added.
http/tests/storageAccess/deny-without-prompt-preserves-gesture.html: Copied from LayoutTests/http/tests/storageAccess/request-and-grant-access-cross-origin-non-sandboxed-iframe-pop-window.html.
http/tests/storageAccess/grant-with-prompt-preserves-gesture-expected.txt: Renamed from LayoutTests/http/tests/storageAccess/request-and-grant-access-cross-origin-non-sandboxed-iframe-pop-window-expected.txt.
http/tests/storageAccess/grant-with-prompt-preserves-gesture.html: Renamed from LayoutTests/http/tests/storageAccess/request-and-grant-access-cross-origin-non-sandboxed-iframe-pop-window.html.
http/tests/storageAccess/has-storage-access-true-if-feature-off-expected.txt: Added.
http/tests/storageAccess/has-storage-access-true-if-feature-off.html: Added.
http/tests/storageAccess/resources/request-storage-access-and-immediately-postmessage-iframe.html:
http/tests/storageAccess/resources/request-storage-access-iframe-and-pop-window.html:
http/tests/storageAccess/resources/request-storage-access-iframe.html:
http/tests/storageAccess/resources/request-storage-access-without-user-gesture-iframe.html:
http/tests/storageAccess/resources/self-navigating-frame-after-granted-access.html:
platform/mac-wk2/TestExpectations:

Added expectations for the new tests.

12:54 PM Changeset in webkit [245024] by achristensen@apple.com

3 edits in trunk/Source/WTF

Add a release assertion that Functions can only be constructed from non-null CompletionHandlers
https://bugs.webkit.org/show_bug.cgi?id=197641

Reviewed by Chris Dumez.

This will help us find the cause of rdar://problem/48679972 by seeing the crash when the Function is dispatched,
not when it's called with no interesting stack trace. I manually verified this assertion is hit in such a case.
We should also have no legitimate use of creating a Function out of a null CompletionHandler then never calling it.

wtf/CompletionHandler.h:

(WTF::Detail::CallableWrapper<CompletionHandler<Out):

wtf/Function.h:

(WTF::Detail::CallableWrapperBase::~CallableWrapperBase):
(WTF::Detail::CallableWrapper::CallableWrapper):
(WTF::Function<Out):

12:10 PM Changeset in webkit [245023] by wilander@apple.com

2 edits in trunk/Source/WebKit

Change IsITPFirstPartyWebsiteDataRemovalEnabled from DEFAULT_EXPERIMENTAL_FEATURES_ENABLED to true.
https://bugs.webkit.org/show_bug.cgi?id=197667
<rdar://problem/50549288>

Reviewed by Brent Fulgham.

Shared/WebPreferences.yaml:

12:04 PM Changeset in webkit [245022] by timothy_horton@apple.com

3 edits in trunk/Source/WebKit

iOSMac Share Sheet shows up at the wrong place
https://bugs.webkit.org/show_bug.cgi?id=197649
<rdar://problem/49922942>

Reviewed by Megan Gardner.

UIProcess/ios/WKContentViewInteraction.h:
UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView _showShareSheet:inRect:completionHandler:]):
(-[WKContentView _hoverGestureRecognizerChanged:]):
Pass the last mouse-over location as the presentation location for the
share sheet, if we don't have a rect to present from. Centering the popover
like we do on iPad doesn't work so well (at all) with a macOS context menu.

11:52 AM Changeset in webkit [245021] by Chris Dumez

4 edits
5 copies
1 add
1 delete in trunk/Source/WebKit

Rename ConnectionMac.mm to ConnectionCocoa.mm since it is used on both macOS and iOS
https://bugs.webkit.org/show_bug.cgi?id=197661

Reviewed by Alexey Proskuryakov.

Platform/IPC/cocoa/ConnectionCocoa.mm: Renamed from Source/WebKit/Platform/IPC/mac/ConnectionMac.mm.

(IPC::ConnectionTerminationWatchdog::createConnectionTerminationWatchdog):
(IPC::ConnectionTerminationWatchdog::ConnectionTerminationWatchdog):
(IPC::ConnectionTerminationWatchdog::watchdogTimerFired):
(IPC::Connection::platformInvalidate):
(IPC::Connection::cancelReceiveSource):
(IPC::Connection::terminateSoon):
(IPC::Connection::platformInitialize):
(IPC::Connection::open):
(IPC::Connection::sendMessage):
(IPC::Connection::platformCanSendOutgoingMessages const):
(IPC::Connection::sendOutgoingMessage):
(IPC::Connection::initializeSendSource):
(IPC::Connection::resumeSendSource):
(IPC::createMessageDecoder):
(IPC::readFromMachPort):
(IPC::Connection::receiveSourceEventHandler):
(IPC::Connection::identifier const):
(IPC::Connection::getAuditToken):
(IPC::Connection::kill):
(IPC::AccessibilityProcessSuspendedNotification):
(IPC::Connection::willSendSyncMessage):
(IPC::Connection::didReceiveSyncReply):
(IPC::Connection::remoteProcessID const):

Platform/IPC/cocoa/ImportanceAssertion.h: Renamed from Source/WebKit/Platform/IPC/mac/ImportanceAssertion.h.

(IPC::ImportanceAssertion::ImportanceAssertion):
(IPC::ImportanceAssertion::~ImportanceAssertion):

Platform/IPC/cocoa/MachMessage.cpp: Renamed from Source/WebKit/Platform/IPC/mac/MachMessage.cpp.

(IPC::MachMessage::create):
(IPC::MachMessage::MachMessage):
(IPC::MachMessage::~MachMessage):
(IPC::MachMessage::messageSize):
(IPC::MachMessage::leakDescriptors):

Platform/IPC/cocoa/MachMessage.h: Renamed from Source/WebKit/Platform/IPC/mac/MachMessage.h.

(IPC::MachMessage::size const):
(IPC::MachMessage::header):
(IPC::MachMessage::messageReceiverName const):
(IPC::MachMessage::messageName const):

Platform/IPC/cocoa/MachPort.h: Renamed from Source/WebKit/Platform/IPC/mac/MachPort.h.

(IPC::MachPort::MachPort):
(IPC::MachPort::encode const):
(IPC::MachPort::decode):
(IPC::MachPort::port const):
(IPC::MachPort::disposition const):

PlatformMac.cmake:
SourcesCocoa.txt:
WebKit.xcodeproj/project.pbxproj:

11:43 AM Changeset in webkit [245020] by commit-queue@webkit.org

7 edits in trunk

[Pointer Events] isPrimary property of pointercancel events should match previous events for that pointer
https://bugs.webkit.org/show_bug.cgi?id=197665

Patch by Antoine Quint <Antoine Quint> on 2019-05-07
Reviewed by Dean Jackson.

LayoutTests/imported/w3c:

Record WPT progressions.

web-platform-tests/pointerevents/pointerevent_pointercancel_touch-expected.txt:

Source/WebCore:

The test at web-platform-tests/pointerevents/pointerevent_pointercancel_touch.html would fail early because one of the first assertions
would check that isPrimary for a pointercancel event would match the isPrimary property of the previous pointer event dispatched for that
pointer id. This prevented many further assertions from passing and also was the cause of flakiness for the next test since this test was
ended early and the state of touches created using UIScriptController were not in a clean state.

We now track the isPrimary state for a given pointer using the CapturingData and use that value when dispatching a pointercancel event.

dom/PointerEvent.cpp:

(WebCore::PointerEvent::create):
(WebCore::PointerEvent::PointerEvent):

dom/PointerEvent.h:
page/PointerCaptureController.cpp:

(WebCore::PointerCaptureController::pointerEventWasDispatched):
(WebCore::PointerCaptureController::cancelPointer):

page/PointerCaptureController.h:

11:42 AM Changeset in webkit [245019] by Ryan Haddad

6 edits in trunk/Source/WebCore

Unreviewed, rolling out r244900.

Caused media/track/track-cue-missing.html to fail on debug
queues

Reverted changeset:

"Add logging for RenderLayer clip rects"
https://bugs.webkit.org/show_bug.cgi?id=197547
https://trac.webkit.org/changeset/244900

11:39 AM Changeset in webkit [245018] by Tadeu Zagallo

3 edits
6 adds in trunk

tryCachePutByID should not crash if target offset changes
https://bugs.webkit.org/show_bug.cgi?id=197311
<rdar://problem/48033612>

Reviewed by Filip Pizlo.

JSTests:

Add a series of tests related tryCachePutByID. Two of these tests used to crash and were fixed
by this patch: cache-put-by-id-different-attributes.js and cache-put-by-id-different-offset.js

stress/cache-put-by-id-delete-prototype.js: Added.

(A.prototype.set y):
(A):
(B.prototype.set y):
(B):
(C):

stress/cache-put-by-id-different-proto.js: Added.

(A.prototype.set y):
(A):
(B1):
(B2.prototype.set y):
(B2):
(C):
(D):

stress/cache-put-by-id-different-attributes.js: Added.

(Foo):
(set x):

stress/cache-put-by-id-different-offset.js: Added.

(Foo):
(set x):

stress/cache-put-by-id-insert-prototype.js: Added.

(A.prototype.set y):
(A):
(C):

stress/cache-put-by-id-poly-proto.js: Added.

(Foo):
(set _):
(createBar.Bar):
(createBar):

Source/JavaScriptCore:

When tryCachePutID is called with a cacheable setter, if the target object where the setter was
found is still in the prototype chain and there's no poly protos in the chain, we use
generateConditionsForPrototypePropertyHit to validate that the target object remains the same.
It checks for the absence of the property in every object in the prototype chain from the base
down to the target object and checks that the property is still present in the target object. It
also bails if there are any uncacheable objects, proxies or dictionary objects in the prototype
chain. However, it does not consider two edge cases:

It asserts that the property should still be at the same offset in the target object, but this

assertion does not hold if the setter deletes properties of the object and causes the structure
to be flattened after the deletion. Instead of asserting, we just use the updated offset.

It does not check whether the new slot is also a setter, which leads to a crash in case it's not.

jit/Repatch.cpp:

(JSC::tryCachePutByID):

10:41 AM Changeset in webkit [245017] by sbarati@apple.com

7 edits
1 add in trunk

Don't OSR enter into an FTL CodeBlock that has been jettisoned
https://bugs.webkit.org/show_bug.cgi?id=197531
<rdar://problem/50162379>

Reviewed by Yusuke Suzuki.

JSTests:

stress/dont-osr-enter-into-jettisoned-ftl-code-block.js: Added.

Source/JavaScriptCore:

Sometimes we make silly mistakes. This is one of those times. It's invalid to OSR
enter into an FTL OSR entry code block that has been jettisoned already.

dfg/DFGJITCode.cpp:

(JSC::DFG::JITCode::clearOSREntryBlockAndResetThresholds):

dfg/DFGJITCode.h:

(JSC::DFG::JITCode::clearOSREntryBlock): Deleted.

dfg/DFGOSREntry.cpp:

(JSC::DFG::prepareOSREntry):
(JSC::DFG::prepareCatchOSREntry):

dfg/DFGOperations.cpp:
ftl/FTLOSREntry.cpp:

(JSC::FTL::prepareOSREntry):

10:37 AM Changeset in webkit [245016] by Antti Koivisto

3 edits
2 adds in trunk

<body> with overflow:hidden shouldn't be keyboard scrollable on iOS
https://bugs.webkit.org/show_bug.cgi?id=197659
Source/WebKit:

<rdar://problem/50541453>

Reviewed by Antoine Quint.

UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView isScrollableForKeyboardScrollViewAnimator:]):

Disable also keyboard scrolling when touch scrolling is disabled.

LayoutTests:

Reviewed by Antoine Quint.

fast/scrolling/ios/body-overflow-hidden-keyboard-expected.html: Added.
fast/scrolling/ios/body-overflow-hidden-keyboard.html: Added.

10:26 AM Changeset in webkit [245015] by Kocsen Chung

7 edits in trunk/Source

Versioning.

10:23 AM Changeset in webkit [245014] by Kocsen Chung

1 copy in tags/Safari-608.1.23

Tag Safari-608.1.23.

9:37 AM Changeset in webkit [245013] by Ryan Haddad

4 edits in trunk/LayoutTests

media/W3C/video/events/event_progress.html is flaky.
https://bugs.webkit.org/show_bug.cgi?id=175678

Unreviewed test gardening.

platform/ios/TestExpectations: Mark test as flaky.
platform/mac/TestExpectations: Ditto.
platform/wk2/TestExpectations: Remove old WK2-only expectation.

9:28 AM Changeset in webkit [245012] by Chris Dumez

3 edits in trunk/LayoutTests

fast/dom/frame-src-javascript-url-async.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=197664

Unreviewed, tweak test so that it is no longer flaky.

fast/dom/frame-src-javascript-url-async-expected.txt:
fast/dom/frame-src-javascript-url-async.html:

9:15 AM Changeset in webkit [245011] by Adrian Perez de Castro

2 edits in trunk/Tools

[Flatpak] webkit-build fails due to missing GTK2 in build environment
https://bugs.webkit.org/show_bug.cgi?id=197352

Reviewed by Philippe Normand.

The Flatpak runtimes do not include GTK 2.x anymore, which means that we need
to build it ourselves on top in order to be able to build the GTK 2.x plugin
process.

flatpak/org.webkit.GTK.yaml: Add gtk2 package.

9:03 AM Changeset in webkit [245010] by Ryan Haddad

3 edits in trunk/LayoutTests

[macOS / iOS Debug] Layout Test imported/w3c/web-platform-tests/webrtc/RTCPeerConnection-getStats.https.html a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=197662

Unreviewed test gardening.

platform/ios-wk2/TestExpectations: Mark test as flaky.
platform/mac-wk2/TestExpectations: Ditto.

7:39 AM Changeset in webkit [245009] by Wenson Hsieh

5 edits in trunk

[macOS] Avoid crashing the UI process when writing empty data to the pasteboard
https://bugs.webkit.org/show_bug.cgi?id=197644
<rdar://problem/50526364>

Reviewed by Tim Horton.

Source/WebKit:

Test: WebKit.WKWebProcessPlugInDoNotCrashWhenCopyingEmptyClientData

WebProcess/WebCoreSupport/WebPlatformStrategies.cpp:

(WebKit::WebPlatformStrategies::setBufferForType):

Make this function robust by not attempting to create a shared memory buffer in the case where the given data
buffer is empty.

Tools:

Add a new API test to exercise a possible scenario where we may crash while writing data to the pasteboard.

TestWebKitAPI/Tests/WebKitCocoa/BundleEditingDelegate.mm:
TestWebKitAPI/Tests/WebKitCocoa/BundleEditingDelegatePlugIn.mm:

(-[BundleEditingDelegatePlugIn webProcessPlugIn:didCreateBrowserContextController:]):
(-[BundleEditingDelegatePlugIn _webProcessPlugInBrowserContextController:pasteboardDataForRange:]):

7:33 AM Changeset in webkit [245008] by Carlos Garcia Campos

2 edits in trunk/Source/WebKit

[GTK] Crash in webkitWebViewBaseSetEnableBackForwardNavigationGesture
https://bugs.webkit.org/show_bug.cgi?id=197653

Reviewed by Žan Doberšek.

We still need to null-check the ViewGestureController there, since it can be called with a running process but
not yet attached when using a related view.

UIProcess/API/gtk/WebKitWebViewBase.cpp:

(webkitWebViewBaseSetEnableBackForwardNavigationGesture):

6:41 AM WebKitGTK/GCCRequirement edited by Michael Catanzaro

(diff)

6:40 AM WebKitGTK/GCCRequirement edited by Michael Catanzaro

(diff)

6:38 AM WebKitGTK/DependenciesPolicy edited by Michael Catanzaro

(diff)

6:37 AM WebKitGTK/GCCRequirement edited by Michael Catanzaro

(diff)

6:31 AM WebKitGTK/DependenciesPolicy edited by Michael Catanzaro

Update dependencies policy (diff)

5:33 AM Changeset in webkit [245007] by Carlos Garcia Campos

3 edits in trunk/Tools

[GTK][WPE] MiniBrowser: load about:blank for new web views in automation mode
https://bugs.webkit.org/show_bug.cgi?id=197651

Reviewed by Žan Doberšek.

This is required since the process creation was delayed until something is loaded in the view.

MiniBrowser/gtk/BrowserWindow.c:

(browser_window_get_or_create_web_view_for_automation):

MiniBrowser/wpe/main.cpp:

(main):

5:02 AM Changeset in webkit [245006] by Antti Koivisto

7 edits
4 adds in trunk

<body> with overflow:hidden CSS is scrollable on iOS
https://bugs.webkit.org/show_bug.cgi?id=153852
<rdar://problem/38715356>

Reviewed by Antoine Quint.

Source/WebCore:

Tests: fast/scrolling/ios/body-overflow-hidden-frame.html

fast/scrolling/ios/body-overflow-hidden.html

page/scrolling/ScrollingTreeScrollingNode.h:

Source/WebKit:

Disable touch scrolling of the main scroll view when <body> has overflow:hidden.

This already works for subframes where we don't create a scrollview in the first place.
The patch also adds a test for that.

UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _didCommitLayerTree:]):

Update scrollability after scrolling tree commits.

UIProcess/RemoteLayerTree/RemoteScrollingCoordinatorProxy.cpp:

(WebKit::RemoteScrollingCoordinatorProxy::hasScrollableMainFrame const):

Base the decision on root ScrollingTreeScrollingNode::canHaveScrollbars() which is computed from overflow.
This matches Mac where wheel event dispatch is similarly blocked based on this property.

UIProcess/RemoteLayerTree/RemoteScrollingCoordinatorProxy.h:

LayoutTests:

fast/scrolling/ios/body-overflow-hidden-expected.html: Added.
fast/scrolling/ios/body-overflow-hidden-frame-expected.html: Added.
fast/scrolling/ios/body-overflow-hidden-frame.html: Added.
fast/scrolling/ios/body-overflow-hidden.html: Added.

2:50 AM Changeset in webkit [245005] by graouts@webkit.org

2 edits in trunk/Source/WebCore

Mouse event simulation should be limited to the graphing calculator on Desmos.com
https://bugs.webkit.org/show_bug.cgi?id=197652
<rdar://problem/47068176>

Reviewed by Antti Koivisto.

page/Quirks.cpp:

(WebCore::Quirks::shouldDispatchSimulatedMouseEvents const):

2:44 AM Changeset in webkit [245004] by youenn@apple.com

13 edits
7904 adds
2 deletes in trunk/LayoutTests

Import WPT referrer-policy tests
https://bugs.webkit.org/show_bug.cgi?id=197321

Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

resources/import-expectations.json:
web-platform-tests/common/: Refreshed.
web-platform-tests/custom-elements/custom-element-registry/per-global-expected.txt:
web-platform-tests/html/browsers/the-window-object/named-access-on-the-window-object/navigated-named-objects.window-expected.txt:
web-platform-tests/referrer-policy/: Added.

LayoutTests:

fix-197321

TestExpectations:

May 6, 2019:

11:41 PM Changeset in webkit [245003] by bshafiei@apple.com

7 edits in tags/Safari-608.1.22.1/Source

Versioning.

11:17 PM Changeset in webkit [245002] by bshafiei@apple.com

1 copy in tags/Safari-608.1.22.1

New tag.

10:05 PM Changeset in webkit [245001] by dino@apple.com

2 edits in trunk/JSTests

Update test262 expectations for Proxy passes
https://bugs.webkit.org/show_bug.cgi?id=197628

Reviewed by Yusuke Suzuki.

There are two consistent passes in Proxy.ownKeys.

test262/expectations.yaml:

8:51 PM Changeset in webkit [245000] by commit-queue@webkit.org

4 edits in trunk/Source

Improve coordination for creating UIWindow instances.
https://bugs.webkit.org/show_bug.cgi?id=197578.
<rdar://problem/50456965>.

Patch by James Savage <James Savage> on 2019-05-06
Reviewed by Wenson Hsieh.

Source/WebCore:

platform/ios/VideoFullscreenInterfaceAVKit.mm:

(makeWindowFromView): Pull out window creation.
(VideoFullscreenInterfaceAVKit::doSetup): Call new helper function.

Source/WebKit:

UIProcess/ios/fullscreen/WKFullScreenWindowControllerIOS.mm:

(makeWindowFromView): Pull out window creation.
(-[WKFullScreenWindowController enterFullScreen]): Call new helper function.

6:23 PM Changeset in webkit [244999] by keith_miller@apple.com

3 edits in trunk/Source/JavaScriptCore

JSWrapperMap should check if existing prototype properties are wrappers when copying exported methods.
https://bugs.webkit.org/show_bug.cgi?id=197324
<rdar://problem/50253144>

Reviewed by Saam Barati.

The current implementation prevents using JSExport to shadow a
method from a super class. This was because we would only add a
method if the prototype didn't already claim to have the
property. Normally this would only happen if an Objective-C super
class already exported a ObjCCallbackFunction for the method,
however, if the user exports a property that is already on
Object.prototype the overriden method won't be exported.

This patch fixes the object prototype issue by checking if the
property on the prototype chain is an ObjCCallbackFunction, if
it's not then it adds an override.

API/JSWrapperMap.mm:

(copyMethodsToObject):

API/tests/testapi.mm:

(-[ToStringClass toString]):
(-[ToStringClass other]):
(-[ToStringSubclass toString]):
(-[ToStringSubclassNoProtocol toString]):
(testToString):
(testObjectiveCAPI):

6:03 PM Changeset in webkit [244998] by timothy_horton@apple.com

9 edits in trunk

_overrideViewportWithArguments does not work when called before loading
https://bugs.webkit.org/show_bug.cgi?id=197638
<rdar://problem/50505111>

Reviewed by Wenson Hsieh.

Source/WebCore:

dom/Document.cpp:

(WebCore::Document::viewportArguments const):
(WebCore::Document::updateViewportArguments):
(WebCore::Document::setOverrideViewportArguments): Deleted.

dom/Document.h:

(WebCore::Document::viewportArguments const): Deleted.

page/Page.cpp:

(WebCore::Page::setOverrideViewportArguments):

page/Page.h:

(WebCore::Page::overrideViewportArguments const):

page/ViewportConfiguration.cpp:

(WebCore::ViewportConfiguration::setViewportArguments):
Move overrideViewportArguments to Page, since it is view-global in the API.

Source/WebKit:

WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::WebPage::setOverrideViewportArguments):

Tools:

TestWebKitAPI/Tests/WebKitCocoa/OverrideViewportArguments.mm:

(TEST):

5:42 PM Changeset in webkit [244997] by Chris Dumez

2 edits in trunk/Source/WebKit

ASSERT at WebKit::NetworkCache::makeSafeToUseMemoryMapForPath under Storage::StoreBodyAsBlob when running WebDriver tests
https://bugs.webkit.org/show_bug.cgi?id=197635
<rdar://problem/50473149>

Reviewed by Geoffrey Garen.

We need to make sure the the path at blobPathString exists before calling FileSystem::makeSafeToUseMemoryMapForPath().
If the blob does not already exist when BlobStorage::add() is called, what will create it is the call to Data::mapToFile().

NetworkProcess/cache/NetworkCacheBlobStorage.cpp:

(WebKit::NetworkCache::BlobStorage::add):

5:29 PM Changeset in webkit [244996] by ysuzuki@apple.com

8 edits
1 add in trunk

[JSC] We should check OOM for description string of Symbol
https://bugs.webkit.org/show_bug.cgi?id=197634

Reviewed by Keith Miller.

JSTests:

stress/check-symbol-description-oom.js: Added.

(shouldThrow):

Source/JavaScriptCore:

When resoling JSString for description of Symbol, we should check OOM error.
We also change JSValueMakeSymbol(..., nullptr) to returning a symbol value
without description, (1) to simplify the code and (2) give a way for JSC API
to create a symbol value without description.

API/JSValueRef.cpp:

(JSValueMakeSymbol):

API/tests/testapi.cpp:

(TestAPI::symbolsTypeof):
(TestAPI::symbolsDescription):
(testCAPIViaCpp):

dfg/DFGOperations.cpp:
runtime/Symbol.cpp:

(JSC::Symbol::createWithDescription):

runtime/Symbol.h:
runtime/SymbolConstructor.cpp:

(JSC::callSymbol):

5:21 PM Changeset in webkit [244995] by Chris Dumez

5 edits in trunk/Source/WebCore

Add assertions to JSLazyEventListener to help catch the cause of a crash
https://bugs.webkit.org/show_bug.cgi?id=197617

Reviewed by Alexey Proskuryakov.

Add assertions to JSLazyEventListener to help catch the cause of <rdar://problem/24314027>.

bindings/js/JSLazyEventListener.cpp:

(WebCore::JSLazyEventListener::checkValidityForEventTarget):

bindings/js/JSLazyEventListener.h:
dom/EventListener.h:

(WebCore::EventListener::checkValidityForEventTarget):

dom/EventTarget.cpp:

(WebCore::EventTarget::addEventListener):
(WebCore::EventTarget::setAttributeEventListener):
(WebCore::EventTarget::innerInvokeEventListeners):

5:20 PM Changeset in webkit [244994] by chris.reid@sony.com

2 edits in trunk/Source/WTF

[JSC] Respect already defined USE_LLINT_EMBEDDED_OPCODE_ID compiler variable.
https://bugs.webkit.org/show_bug.cgi?id=197633

Reviewed by Don Olmstead.

When the variable USE_LLINT_EMBEDDED_OPCODE_ID is defined, stop defining
its value with platform default one.

wtf/Platform.h:

5:08 PM Changeset in webkit [244993] by Wenson Hsieh

2 edits in trunk/Source/WebKit

Occasional crashes in layout tests when firing the shrink-to-fit-content timer
https://bugs.webkit.org/show_bug.cgi?id=197629
<rdar://problem/50514382>

Reviewed by Tim Horton.

Fixes several crashing layout tests by stopping the shrink-to-fit-content timer when the page closes.

WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::close):

5:04 PM Changeset in webkit [244992] by Justin Fan

2 edits
14 adds in trunk/Websites/webkit.org

[Web GPU] Add demos to webkit.org
https://bugs.webkit.org/show_bug.cgi?id=196951

Reviewed by Dean Jackson.

Add simple triangle and spinning textured cube demos to the website.
Add Web GPU to experimental-features.html.

demos/webgpu/css/style.css: Added.

(body):
(canvas):
(body.error canvas):
(h1):
(#container):
(select):
(div#log):
(body.error):
(body.error #contents):
(#error):
(@media (max-width: 480px)):
(body.error #error):
(#error h2):
(#error p):

demos/webgpu/hello-cube.html: Added.
demos/webgpu/hello-triangle.html: Added.
demos/webgpu/index.html: Added.
demos/webgpu/resources/circle.svg: Added.
demos/webgpu/resources/hello-cube.png: Added.
demos/webgpu/resources/hello-triangle.png: Added.
demos/webgpu/resources/safari-alpha.png: Added.
demos/webgpu/scripts/gl-matrix-min.js: Added.

(return.n):
(return.n.d):
(return.):
(return.get n.n):
(return.n.o):

demos/webgpu/scripts/hello-triangle.js: Added.

(async.helloTriangle):

experimental-features.html:

4:49 PM Changeset in webkit [244991] by Alan Coon

7 edits in trunk/Source

Versioning.

4:46 PM Changeset in webkit [244990] by Alan Coon

7 edits in tags/Safari-608.1.22/Source

Versioning.

4:43 PM Changeset in webkit [244989] by Alan Coon

1 copy in tags/Safari-608.1.22

New tag.

4:29 PM Changeset in webkit [244988] by pvollan@apple.com

7 edits in trunk

-[WKWebsiteDataStore removeDataOfTypes:forDataRecords:completionHandler:] doesn't delete _WKWebsiteDataTypeCredentials
https://bugs.webkit.org/show_bug.cgi?id=197510
Source/WebCore:

<rdar://problem/50372338>

Reviewed by Alex Christensen.

This patch implements deletion of non persistent credentials for a set of origins. In order for this to work, fetching
credentials from the credential storage needs to return a set of SecurityOriginData objects, instead of a set of origin
strings. This is implemented by iterating over all the elements in the credential map, and creating a SecurityOriginData
object for each credential based on the protection space.

API test: WKWebsiteDataStore.RemoveNonPersistentCredentials

platform/network/CredentialStorage.cpp:

(WebCore::CredentialStorage::removeCredentialsWithOrigin):
(WebCore::CredentialStorage::originsWithCredentials const):

platform/network/CredentialStorage.h:

(WebCore::CredentialStorage::originsWithCredentials const): Deleted.

Source/WebKit:

Reviewed by Alex Christensen.

NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::fetchWebsiteData):
(WebKit::NetworkProcess::deleteWebsiteDataForOrigins):
(WebKit::NetworkProcess::registrableDomainsWithWebsiteData):

Tools:

Reviewed by Alex Christensen.

TestWebKitAPI/Tests/WebKitCocoa/WKWebsiteDatastore.mm:

(TestWebKitAPI::TEST):

4:13 PM Changeset in webkit [244987] by Keith Rollin

9 edits in trunk

Temporarily disable generate-xcfilelists
https://bugs.webkit.org/show_bug.cgi?id=197619
<rdar://problem/50507392>

Reviewed by Alex Christensen.

We need to perform a significant update to the generate-xcfilelist
scripts. This work involves coordinated work with another facility. If
the work does not occur in tandem, the build will be broken. To avoid
this, disable the invoking of the scripts during the transition. The
checking will be restored once the new scripts are in place.

Source/JavaScriptCore:

Scripts/check-xcfilelists.sh:

Source/WebCore:

No new tests -- no change in user-visible functionality.

Scripts/check-xcfilelists.sh:

Source/WebKit:

Scripts/check-xcfilelists.sh:

Tools:

DumpRenderTree/Scripts/check-xcfilelists.sh:
WebKitTestRunner/Scripts/check-xcfilelists.sh:

4:10 PM Changeset in webkit [244986] by Basuke Suzuki

4 edits in trunk/Source/JavaScriptCore

[PlayStation] Fix build break since r244919
https://bugs.webkit.org/show_bug.cgi?id=197627

Reviewed by Ross Kirsling.

Bugfix for POSIX socket implementation and suppress warnings.

inspector/remote/socket/RemoteInspectorConnectionClient.h:

(Inspector::RemoteInspectorConnectionClient::didAccept):

inspector/remote/socket/posix/RemoteInspectorSocketPOSIX.cpp:

(Inspector::Socket::getPort):

4:07 PM Changeset in webkit [244985] by Chris Dumez

2 edits in trunk/Source/WebKit

Unreviewed build fix after r244984.

WebProcess/cocoa/WebProcessCocoa.mm:

3:47 PM Changeset in webkit [244984] by Chris Dumez

3 edits in trunk/Source/WebKit

Prewarmed WebContent processes never app nap
https://bugs.webkit.org/show_bug.cgi?id=197630

Reviewed by Geoffrey Garen.

Do not require actually having a WebPage in the WebContent process to App Nap.

WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::updateThrottleState):

WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::platformInitializeWebProcess):

3:33 PM Changeset in webkit [244983] by commit-queue@webkit.org

6 edits
4 adds in trunk

Hitpoint for link which spans two lines in web content is incorrect
https://bugs.webkit.org/show_bug.cgi?id=197511
<rdar://problem/49971483>

Patch by Andres Gonzalez <Andres Gonzalez> on 2019-05-06
Reviewed by Chris Fleizach.

Source/WebCore:

Special case for links to return first char location as clickPoint instead of middle point of bounding rect.
Modified iOS ActivationPoint to use clickPoint. This way all code paths go through the same function.
Made boundsForRects to return content coordinates in all platforms. Adjusted all callers, directly or indirectly, appropriately.

Tests: accessibility/ios-simulator/links-activation.html

accessibility/links-activation.html

accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::clickPoint):
(WebCore::AccessibilityRenderObject::boundsForRects):
(WebCore::AccessibilityRenderObject::boundsForRects const): Deleted.

accessibility/AccessibilityRenderObject.h:
accessibility/ios/WebAccessibilityObjectWrapperIOS.mm:

(-[WebAccessibilityObjectWrapper accessibilityActivationPoint]):

accessibility/mac/WebAccessibilityObjectWrapperMac.mm:

(-[WebAccessibilityObjectWrapper accessibilityAttributeValue:forParameter:]):

LayoutTests:

Added LayoutTest.

accessibility/ios-simulator/links-activation-expected.txt: Added.
accessibility/ios-simulator/links-activation.html: Added.
accessibility/links-activation-expected.txt: Added.
accessibility/links-activation.html: Added.

3:28 PM Changeset in webkit [244982] by dbates@webkit.org

2 edits in trunk/Source/WebKit

Fix the build after r244975
(https://bugs.webkit.org/show_bug.cgi?id=197474)

Actually make use of the Keyboard Event when building with USE(UIKIT_KEYBOARD_ADDITIONS)
and substitute -> for . since it's now passed by reference.

WebProcess/WebCoreSupport/ios/WebEditorClientIOS.mm:

(WebKit::WebEditorClient::handleInputMethodKeydown):

3:26 PM Changeset in webkit [244981] by ysuzuki@apple.com

1 edit
1 add in trunk/JSTests

Unreviewed, land one more test
https://bugs.webkit.org/show_bug.cgi?id=197587

stress/setter-frame-flush.js: Added.

(setter):
(foo):
(bar):

3:22 PM Changeset in webkit [244980] by jer.noble@apple.com

8 edits
2 moves
2 adds in trunk/Source

Adopt AVStreamDataParser.audiovisualMIMETypes
https://bugs.webkit.org/show_bug.cgi?id=197581
<rdar://problem/50458981>

Reviewed by Eric Carlson.

Source/WebCore:

Add a new singleton class, AVStreamDataParserMIMETypeCache, and rename AVFoundationMIMETypeCache to the more precise
AVAssetMIMETypeCache. Update all the old AVFoundationMIMETypeCache with the new name.

SourcesCocoa.txt:
WebCore.xcodeproj/project.pbxproj:
platform/graphics/avfoundation/objc/AVAssetMIMETypeCache.h: Renamed from Source/WebCore/platform/graphics/avfoundation/objc/AVFoundationMIMETypeCache.h.

(WebCore::AVAssetMIMETypeCache::setCacheMIMETypesCallback):

platform/graphics/avfoundation/objc/AVAssetMIMETypeCache.mm: Renamed from Source/WebCore/platform/graphics/avfoundation/objc/AVFoundationMIMETypeCache.mm.

(WebCore::AVAssetMIMETypeCache::singleton):
(WebCore::AVAssetMIMETypeCache::setSupportedTypes):
(WebCore::AVAssetMIMETypeCache::types):
(WebCore::AVAssetMIMETypeCache::supportsContentType):
(WebCore::AVAssetMIMETypeCache::canDecodeType):
(WebCore::AVAssetMIMETypeCache::isAvailable const):
(WebCore::AVAssetMIMETypeCache::loadMIMETypes):

platform/graphics/avfoundation/objc/AVStreamDataParserMIMETypeCache.h: Added.
platform/graphics/avfoundation/objc/AVStreamDataParserMIMETypeCache.mm: Added.

(WebCore::AVStreamDataParserMIMETypeCache::singleton):
(WebCore::AVStreamDataParserMIMETypeCache::types):
(WebCore::AVStreamDataParserMIMETypeCache::supportsContentType):
(WebCore::AVStreamDataParserMIMETypeCache::canDecodeType):
(WebCore::AVStreamDataParserMIMETypeCache::isAvailable const):
(WebCore::AVStreamDataParserMIMETypeCache::loadMIMETypes):

platform/graphics/avfoundation/objc/ImageDecoderAVFObjC.mm:

(WebCore::ImageDecoderAVFObjC::create):
(WebCore::ImageDecoderAVFObjC::supportsMediaType):
(WebCore::ImageDecoderAVFObjC::supportsContentType):
(WebCore::ImageDecoderAVFObjC::canDecodeType):

platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:

(WebCore::MediaPlayerPrivateAVFoundationObjC::registerMediaEngine):
(WebCore::MediaPlayerPrivateAVFoundationObjC::getSupportedTypes):
(WebCore::MediaPlayerPrivateAVFoundationObjC::supportsType):
(WebCore::MediaPlayerPrivateAVFoundationObjC::supportsKeySystem):

platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaSourceAVFObjC.mm:

(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::registerMediaEngine):
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::getSupportedTypes):
(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::supportsType):

Source/WebKit:

WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::platformInitializeWebProcess):
(WebKit::WebProcess::platformTerminate):
(WebKit::WebProcess::setMediaMIMETypes):

2:44 PM Changeset in webkit [244979] by Chris Dumez

4 edits in trunk/Source/WebKit

Terminate service workers that use too much CPU / power
https://bugs.webkit.org/show_bug.cgi?id=197626
<rdar://problem/50374707>

Reviewed by Geoffrey Garen.

Terminate service worker processes that use over 50% CPU on average over the last 8 minutes,
similarly to what we do for background WebContent processes.

UIProcess/WebProcessProxy.cpp:

(WebKit::WebProcessProxy::didExceedCPULimit):

WebProcess/WebProcess.cpp:

(WebKit::WebProcess::initializeProcess):

WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::updateCPULimit):
(WebKit::WebProcess::updateCPUMonitorState):

2:33 PM Changeset in webkit [244978] by ysuzuki@apple.com

38 edits
11 adds in trunk

TemplateObject passed to template literal tags are not always identical for the same source location.
https://bugs.webkit.org/show_bug.cgi?id=190756

Reviewed by Saam Barati.

JSTests:

complex.yaml:
complex/tagged-template-regeneration-after.js: Added.

(shouldBe):

complex/tagged-template-regeneration.js: Added.

(call):
(test):

modules/tagged-template-inside-module.js: Added.

(from.string_appeared_here.call):

modules/tagged-template-inside-module/other-tagged-templates.js: Added.

(call):
(export.otherTaggedTemplates):

stress/call-and-construct-should-return-same-tagged-templates.js: Added.

(shouldBe):
(call):
(poly):

stress/tagged-templates-in-direct-eval-should-not-produce-same-site-object.js: Added.

(shouldBe):
(call):

stress/tagged-templates-in-global-function-should-not-produce-same-site-object.js: Added.

(shouldBe):
(call):

stress/tagged-templates-in-indirect-eval-should-not-produce-same-site-object.js: Added.

(shouldBe):
(call):

stress/tagged-templates-in-multiple-functions.js: Added.

(shouldBe):
(call):
(a):
(b):
(c):

Source/JavaScriptCore:

CodeBlock can be jettisoned and regenerated. So every time CodeBlock is regenerated, we get the different site object.
Call and Construct can have different CodeBlock. Even if the function is called in call-form or construct-form, we should return the same site object.

This patch threads the top-level ScriptExecutable to each FunctionExecutable creation. Each FunctionExecutable has a reference to the top-level ScriptExecutable.
We put TemplateObjectMap in ScriptExecutable, which manages cached template objects.
We move FunctionExecutable::m_cachedPolyProtoStructure to the FunctionExecutable::RareDate to keep FunctionExecutable 128 bytes.

Scripts/tests/builtins/expected/JavaScriptCore-Builtin.Promise-Combined.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-Builtin.Promise-Separate.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-Builtin.prototype-Combined.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-Builtin.prototype-Separate.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-BuiltinConstructor-Combined.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-BuiltinConstructor-Separate.js-result:
Scripts/tests/builtins/expected/JavaScriptCore-InternalClashingNames-Combined.js-result:
Scripts/tests/builtins/expected/WebCore-AnotherGuardedInternalBuiltin-Separate.js-result:
Scripts/tests/builtins/expected/WebCore-ArbitraryConditionalGuard-Separate.js-result:
Scripts/tests/builtins/expected/WebCore-GuardedBuiltin-Separate.js-result:
Scripts/tests/builtins/expected/WebCore-GuardedInternalBuiltin-Separate.js-result:
Scripts/tests/builtins/expected/WebCore-UnguardedBuiltin-Separate.js-result:
Scripts/tests/builtins/expected/WebCore-xmlCasingTest-Separate.js-result:
Scripts/wkbuiltins/builtins_templates.py:
bytecode/CodeBlock.cpp:

(JSC::CodeBlock::finishCreation):
(JSC::CodeBlock::setConstantRegisters):

bytecode/CodeBlock.h:
bytecode/UnlinkedFunctionExecutable.cpp:

(JSC::UnlinkedFunctionExecutable::link):

bytecode/UnlinkedFunctionExecutable.h:
bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::addTemplateObjectConstant):
(JSC::BytecodeGenerator::emitGetTemplateObject):

bytecompiler/BytecodeGenerator.h:
runtime/CachedTypes.cpp:

(JSC::CachedTemplateObjectDescriptor::encode):
(JSC::CachedTemplateObjectDescriptor::decode const):
(JSC::CachedJSValue::encode):
(JSC::CachedJSValue::decode const):

runtime/EvalExecutable.cpp:

(JSC::EvalExecutable::ensureTemplateObjectMap):
(JSC::EvalExecutable::visitChildren):

runtime/EvalExecutable.h:
runtime/FunctionExecutable.cpp:

runtime/FunctionExecutable.h:
runtime/JSModuleRecord.cpp:

(JSC::JSModuleRecord::instantiateDeclarations):

runtime/JSTemplateObjectDescriptor.cpp:

(JSC::JSTemplateObjectDescriptor::JSTemplateObjectDescriptor):
(JSC::JSTemplateObjectDescriptor::create):

runtime/JSTemplateObjectDescriptor.h:
runtime/ModuleProgramExecutable.cpp:

(JSC::ModuleProgramExecutable::ensureTemplateObjectMap):
(JSC::ModuleProgramExecutable::visitChildren):

runtime/ModuleProgramExecutable.h:
runtime/ProgramExecutable.cpp:

(JSC::ProgramExecutable::ensureTemplateObjectMap):
(JSC::ProgramExecutable::visitChildren):

runtime/ProgramExecutable.h:
runtime/ScriptExecutable.cpp:

(JSC::ScriptExecutable::topLevelExecutable):
(JSC::ScriptExecutable::createTemplateObject):
(JSC::ScriptExecutable::ensureTemplateObjectMap):

runtime/ScriptExecutable.h:
tools/JSDollarVM.cpp:

(JSC::functionCreateBuiltin):
(JSC::functionDeleteAllCodeWhenIdle):
(JSC::JSDollarVM::finishCreation):

2:14 PM Changeset in webkit [244977] by youenn@apple.com

6 edits
3 adds in trunk

WebAudio Node JS wrappers should not be collected if events can be fired
https://bugs.webkit.org/show_bug.cgi?id=197533

Reviewed by Jer Noble.

Source/WebCore:

Before the patch, some web audio nodes could fire event listeners, but were not protected from GC.
Use CustomIsReachable to ensure theses nodes can be collected if:

their AudioContext is stopped (typically due to document being navigated away).
their AudioContext is closed.
nodes do not have event listeners.

Covered by WPT mediacapture-streams/MediaStreamTrack-MediaElement-disabled-audio-is-silence.https.html and
WPT webaudio/the-audio-api/the-mediaelementaudiosourcenode-interface/mediaElementAudioSourceToScriptProcessorTest.html
and web audio WebRTC tests.
Specific newly added test: webaudio/webaudio-gc.html

Modules/webaudio/AudioContext.h:

(WebCore::AudioContext::isClosed const):

Modules/webaudio/AudioNode.idl:
Sources.txt:
WebCore.xcodeproj/project.pbxproj:
bindings/js/JSAudioNodeCustom.cpp: Added.

(WebCore::JSAudioNodeOwner::isReachableFromOpaqueRoots):

LayoutTests:

webaudio/webaudio-gc-expected.txt: Added.
webaudio/webaudio-gc.html: Added.

1:44 PM Changeset in webkit [244976] by Ryan Haddad

2 edits in trunk/LayoutTests

REGRESSION: Layout test imported/w3c/web-platform-tests/resource-timing/resource-timing-level1.sub.html is frequently failing on EWS
https://bugs.webkit.org/show_bug.cgi?id=197473

Unreviewed test gardening.

platform/ios/TestExpectations: Mark test as flaky on iOS.

1:41 PM Changeset in webkit [244975] by dbates@webkit.org

16 edits in trunk/Source

Google Docs & Yahoo! Japan: Can’t compose characters with Chinese or Japanese keyboard
https://bugs.webkit.org/show_bug.cgi?id=197474
<rdar://problem/47219324>

Reviewed by Ryosuke Niwa.

Source/WebCore:

Fix up some #if defs to compile more Mac code when building on iOS.

dom/KeyboardEvent.cpp:

(WebCore::KeyboardEvent::KeyboardEvent):

platform/PlatformKeyboardEvent.h:

(WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent):

Source/WebKit:

Adopt UIKit SPI (added in <rdar://problem/50181162>) to ask the Input Manager to handle
the key event. If the Input Manager handled it then we pass along this information to
WebCore, just as we do for Mac, so that it can alter its event handling logic. Otherwise,
we handle the event as we do now.

A large part of this patch is fixing up #if defs to compile more Mac code when building
on iOS.

Platform/spi/ios/UIKitSPI.h: Expose more SPI.
Shared/NativeWebKeyboardEvent.h:
Shared/WebEvent.h:
Shared/WebEventConversion.cpp:

(WebKit::WebKit2PlatformKeyboardEvent::WebKit2PlatformKeyboardEvent):

Shared/WebKeyboardEvent.cpp:

(WebKit::WebKeyboardEvent::WebKeyboardEvent):
(WebKit::WebKeyboardEvent::encode const):
(WebKit::WebKeyboardEvent::decode):

Shared/ios/NativeWebKeyboardEventIOS.mm:

(WebKit::NativeWebKeyboardEvent::NativeWebKeyboardEvent):

Shared/ios/WebIOSEventFactory.h:
Shared/ios/WebIOSEventFactory.mm:

(WebIOSEventFactory::createWebKeyboardEvent):
Compile more Mac code on iOS. Just like on Mac we maintain some bookkeeping on
whether an event was handled by the Input Manager.

UIProcess/Automation/ios/WebAutomationSessionIOS.mm:

(WebKit::WebAutomationSession::sendSynthesizedEventsToPage): Pass NativeWebKeyboardEvent::HandledByInputMethod::No
to keep the behavior we have now.

UIProcess/ios/WKContentViewInteraction.h:
UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView cleanupInteraction]):
(-[WKContentView shouldSuppressUpdateCandidateView]): Added. Used to tell UIKit whether to
suppress updating/showing the candidate view.
(-[WKContentView setMarkedText:selectedRange:]):
Bookkeeping to track whether we need to delay showing/updating the inline candidate view.
The concept in UIKit is deferment, but at the time of writing its simply a request to delay
the update for 0.4 seconds. We opt into this delay only for the first key that begins
marked text (i.e. the transition from no marked text to marked text). We do this because we
may not have up-to-date editor state at the time UIKit is ready to show/update the inline
candidate view for us to answer -textFirstRect and -textLastRect, which UIKit calls as part
of computing the frame rect for the inline candidate view on screen. Once we receive up-to-date
editor state, in -selectionChanged, we tell UIKit to layout the keyboard, which ultimately
causes it to re-compute the frame rect for the inline candidate view and show it.

(-[WKContentView handleKeyWebEvent:]): Pass NativeWebKeyboardEvent::HandledByInputMethod::No
to keep the behavior we have now.
(-[WKContentView handleKeyWebEvent:withCompletionHandler:]): Ask the keyboard to handle the
event using the Input Manager. If it was handled then there is no need to delay calling the
completion handler, call it, then tell the web process about the key event and that it was
already handled by the Input Manager.
(-[WKContentView _selectionChanged]): Tell the keyboard to update the candidate view, if needed.

WebProcess/WebCoreSupport/ios/WebEditorClientIOS.mm:

(WebKit::WebEditorClient::handleInputMethodKeydown): Mark the event as default handled if
the UI process told us that the event was handled by the Input Manager just like we do on Mac.

1:39 PM Changeset in webkit [244974] by graouts@webkit.org

16 edits in trunk/LayoutTests

Tests under pointerevents/ios are flaky
https://bugs.webkit.org/show_bug.cgi?id=197624

Reviewed by Dean Jackson.

Tests under pointerevents/ios generate touches that use UIScriptController may not succeed if ran in multiple iterations or in a specific
order due to not ensuring that all touches are released when the test completes. We now ensure that we do when running swipes, taps, and pinches.

pointerevents/ios/pointer-events-dispatch-on-stylus.html: Use the new ui.tapStylus() method to generate a tap with the stylus which ensures all

touches are removed upon completion.

pointerevents/ios/pointer-events-dispatch-on-touch.html: Use a tap to ensure all touches are removed upon completion.
pointerevents/ios/pointer-events-implicit-capture-has-pointer-capture-in-pointer-down.html: Use a tap to ensure all touches are removed upon completion.
pointerevents/ios/pointer-events-implicit-capture-release-exception.html: Use a tap to ensure all touches are removed upon completion.
pointerevents/ios/pointer-events-implicit-capture-release.html: Use a tap to ensure all touches are removed upon completion.
pointerevents/ios/pointer-events-prevent-default-allows-click-event.html: Ensure both the "click" event and the tap generation have succeeded before

marking the test as complete.

pointerevents/ios/pointer-events-set-pointer-capture-exceptions.html: Use a tap to ensure all touches are removed upon completion.
pointerevents/ios/touch-action-none-link-traversal.html: Ensure both the "load" event and the tap generation have succeeded before marking the test

as complete.

pointerevents/ios/touch-action-pan-x-pan-y.html: Remove the requestAnimationFrame() call since ui.swipe() now resolves its promise once all touches

have completed.

pointerevents/ios/touch-action-pan-x.html: Remove the requestAnimationFrame() call since ui.swipe() now resolves its promise once all touches

have completed.

pointerevents/ios/touch-action-pan-y.html: Remove the requestAnimationFrame() call since ui.swipe() now resolves its promise once all touches

have completed.

pointerevents/ios/touch-action-pinch-zoom-allows-zooming.html: Remove the requestAnimationFrame() call since ui.pinchOut() now resolves its promise

once all touches have completed.

pointerevents/ios/touch-action-pointercancel-pan-x.html: We don't need to track "pointermove" events since dispatch of "pointercancel" is asynchronous

and the number of "pointermove" events prior to its dispatch can legitimately vary.

pointerevents/ios/touch-action-pointercancel-pan-y.html: We don't need to track "pointermove" events since dispatch of "pointercancel" is asynchronous

and the number of "pointermove" events prior to its dispatch can legitimately vary.

pointerevents/utils.js:

(const.ui.new.UIController.prototype.swipe): Wait until the swipe is complete before resolving the promise.
(const.ui.new.UIController.prototype.pinchOut): Use a custom sequence to ensure that the pinch releases touches upon completion.
(const.ui.new.UIController.prototype.tapStylus): Introduce this new method to perform a stylus tap which ensures all touches are complete before resolving
the promise.
(const.ui.new.UIController.prototype.beginTouches): Deleted.
(const.ui.new.UIController.prototype.beginStylus): Deleted.

1:36 PM Changeset in webkit [244973] by sbarati@apple.com

2 edits in trunk/PerformanceTests

JS2 should print scores for different categories
https://bugs.webkit.org/show_bug.cgi?id=197605

Reviewed by Yusuke Suzuki.

JetStream2/JetStreamDriver.js:

(displayCategoryScores):
(isInBrowser.document.onkeydown):
(Driver.prototype.async.start):
(Driver.prototype.async.reportScoreToRunBenchmarkRunner):

1:27 PM Changeset in webkit [244972] by stephan.szabo@sony.com

2 edits
3 adds in trunk/JSTests

[PlayStation] JSC Stress tests failing due to timezone printing
https://bugs.webkit.org/show_bug.cgi?id=197615

PlayStation's strftime does not give timezone strings, which
results in time strings like "Wed Oct 23 1974 11:45:01 GMT-0700"
rather than "Wed Oct 23 1974 11:45:01 GMT-0700 (Pacific Daylight Time)"
which causes diff failures with the expectations. Add expectations
without the timezone string and use those on playstation.

Reviewed by Ross Kirsling.

ChakraCore.yaml: Update these tests to use alternate expectation file on PlayStation
ChakraCore/test/GlobalFunctions/InternalToString.baseline-jsc-playstation: Added.
ChakraCore/test/Operators/equals.baseline-jsc-playstation: Added.
ChakraCore/test/fieldopts/objtypespec-newobj.2.baseline-jsc-playstation: Added.

1:25 PM Changeset in webkit [244971] by Chris Dumez

4 edits in trunk/Source/WebCore

Add assertions to CachedFrame to help figure out crash in CachedFrame constructor
https://bugs.webkit.org/show_bug.cgi?id=197621

Reviewed by Geoffrey Garen.

Add release assertions to try and figure out who is sometimes detaching the document from its
frame while constructing CachedFrames for its descendants.

dom/Document.cpp:

(WebCore::Document::detachFromFrame):

dom/Document.h:

(WebCore::Document::setMayBeDetachedFromFrame):

history/CachedFrame.cpp:

(WebCore::CachedFrame::CachedFrame):

1:24 PM Changeset in webkit [244970] by commit-queue@webkit.org

2 edits in trunk/Source/WebKit

Null check m_mainFrame in WebPageProxy.cpp
https://bugs.webkit.org/show_bug.cgi?id=197618
<rdar://problem/47463054>

Patch by Alex Christensen <achristensen@webkit.org> on 2019-05-06
Reviewed by Geoffrey Garen.

It's already null checked in some places, and the places where it isn't are causing crashes.
Let's fix all of them.

UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::decidePolicyForNavigationAction):
(WebKit::WebPageProxy::decidePolicyForNewWindowAction):
(WebKit::WebPageProxy::createNewPage):

1:14 PM Changeset in webkit [244969] by Brent Fulgham

2 edits in trunk/Source/WebKit

Use more efficient path resolution logic
https://bugs.webkit.org/show_bug.cgi?id=197389
<rdar://problem/50268491>

Reviewed by Maciej Stachowiak.

The code in SandboxExtensionsCocoa.mm 'resolveSymlinksInPath' is pretty inefficient, and tries to reproduce (badly)
logic that is already provided by the operating system.

To make matters worse, 'resolvePathForSandboxExtension' was effectively performing the work of fully resolving
symlinks twice, since NSString's 'stringByStandardizingPath' method does some of this already.

Instead, we should just use NSString's 'stringByResolvingSymlinksInPath', which does the symlink resolution
using more efficient logic than our 'resolveSymlinksInPath' code.

Shared/Cocoa/SandboxExtensionCocoa.mm:

(WebKit::resolveSymlinksInPath): Removed.
(WebKit::resolvePathForSandboxExtension): Remove redundant call to 'resolveSymlinksInPath', and switches from
'stringByStandardizingPath' to 'stringByResolvingSymlinksInPath', which can take the place of both calls.
(WebKit::stringByResolvingSymlinksInPath): Switch to call 'stringByResolvingSymlinksInPath'.

11:32 AM Changeset in webkit [244968] by Alan Coon

7 edits in tags/Safari-608.1.21/Source

Versioning.

11:28 AM Changeset in webkit [244967] by Alan Coon

1 copy in tags/Safari-608.1.21

New tag.

10:50 AM Changeset in webkit [244966] by Wenson Hsieh

6 edits in trunk/Source/WebKit

Introduce SPI to request modern compatibility mode but defer to site-specific quirks
https://bugs.webkit.org/show_bug.cgi?id=197614
<rdar://problem/50208577>

Reviewed by Tim Horton.

Add plumbing for a new _allowSiteSpecificQuirksToOverrideCompatibilityMode SPI property on
WKWebpagePreferences.

UIProcess/API/APIWebsitePolicies.cpp:

(API::WebsitePolicies::copy const):

UIProcess/API/APIWebsitePolicies.h:
UIProcess/API/Cocoa/WKWebpagePreferences.mm:

(-[WKWebpagePreferences _allowSiteSpecificQuirksToOverrideCompatibilityMode]):
(-[WKWebpagePreferences _setAllowSiteSpecificQuirksToOverrideCompatibilityMode:]):

UIProcess/API/Cocoa/WKWebpagePreferencesPrivate.h:
UIProcess/WebPageProxy.cpp:

Use currentRequest instead of originalRequest; see internal change for more details.

(WebKit::WebPageProxy::decidePolicyForNavigationAction):

10:50 AM Changeset in webkit [244965] by Truitt Savell

2 edits in trunk/LayoutTests

Fix typo in https://trac.webkit.org/changeset/244962/webkit
https://bugs.webkit.org/show_bug.cgi?id=136627

Unreviewed test gardening.

platform/mac/TestExpectations:

10:23 AM Changeset in webkit [244964] by Ryan Haddad

2 edits in trunk/Source/WebKit

Unreviewed, rolling out r244917.

Caused
TestWebKitAPI.WKWebView.InitializingWebViewWithEphemeralStorageDoesNotLog
failure on debug bots.

Reverted changeset:

"Use more efficient path resolution logic"
https://bugs.webkit.org/show_bug.cgi?id=197389
https://trac.webkit.org/changeset/244917

10:07 AM Changeset in webkit [244963] by bshafiei@apple.com

7 edits in trunk/Source

Versioning.

9:14 AM Changeset in webkit [244962] by Truitt Savell

2 edits in trunk/LayoutTests

mark media/track/track-cue-rendering-vertical.html as failing for Mojave after r244891 unmarked it.
https://bugs.webkit.org/show_bug.cgi?id=136627

Unreviewed test gardening.

platform/mac/TestExpectations:

2:03 AM Changeset in webkit [244961] by zandobersek@gmail.com

2 edits in trunk/Source/WebCore

[GLib] WebCore::MainThreadSharedTimer should use the appropriate GSource priority, name
https://bugs.webkit.org/show_bug.cgi?id=197606

Reviewed by Carlos Garcia Campos.

platform/MainThreadSharedTimer.cpp:

(WebCore::MainThreadSharedTimer::MainThreadSharedTimer):
Use the MainThreadSharedTimer GLib priority for this timer. The name is
also adjusted accordingly.

12:54 AM Changeset in webkit [244960] by chris.reid@sony.com

2 edits in trunk

[CMake] fuse-ld should also be set in module linker flags
https://bugs.webkit.org/show_bug.cgi?id=197583

Reviewed by Michael Catanzaro.

Add fuse-ld to CMAKE_MODULE_LINKER_FLAGS.

Source/cmake/OptionsCommon.cmake:

12:04 AM Changeset in webkit [244959] by ysuzuki@apple.com

1 edit
1 add in trunk/JSTests

[JSC] Add more tests for DFG SetLocal emission for adhoc SetterCall frame
https://bugs.webkit.org/show_bug.cgi?id=197587

Reviewed by Sam Weinig.

This patch adds more tests to r244939. It also inlines setter calls, and eventually see that no PutStack is emitted because MovHint's KillStack kills it.

stress/adhoc-setter-frame-should-not-be-killed.js: Added.

Note: See TracTimeline for information about the timeline view.

Download in other formats:

RSS Feed