WebKit-https.git
12 months agoUnreviewed, rebaseline tests after r236632.
ryanhaddad@apple.com [Mon, 1 Oct 2018 21:25:21 +0000 (21:25 +0000)]
Unreviewed, rebaseline tests after r236632.

* platform/ios/css3/blending/repaint/blend-mode-isolate-stacking-context-expected.txt:
* platform/ios/css3/blending/repaint/blend-mode-turn-off-isolation-expected.txt:
* platform/ios/css3/blending/repaint/blend-mode-turn-off-isolation-no-effect-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Add SPI to customize the input accessory view when focusing an element
wenson_hsieh@apple.com [Mon, 1 Oct 2018 20:59:19 +0000 (20:59 +0000)]
[iOS] Add SPI to customize the input accessory view when focusing an element
https://bugs.webkit.org/show_bug.cgi?id=190152
<rdar://problem/42754975>

Reviewed by Dan Bernstein.

Source/WebKit:

Adds SPI on WKFormInputSession to customize the input accessory view, alongside the input view. See below for
more details.

Test: KeyboardInputTests.CustomInputViewAndInputAccessoryView

* UIProcess/API/Cocoa/_WKFormInputSession.h:
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKFormInputSession customInputAccessoryView]):
(-[WKFormInputSession setCustomInputAccessoryView:]):

Reload input views when the custom input accessory view changes.

(-[WKContentView requiresAccessoryView]):

If a custom input accessory view is specified, return YES.

(-[WKContentView inputAccessoryView]):

Return the custom input accessory view if present, and fall back to the default web form accessory view.

Tools:

Add an API test to verify that setting a custom input accessory view and custom input view on the form input
session when focusing an element overrides the first responder's (i.e. WKContentView's) `-inputView` and
`-inputAccessoryView`.

* TestWebKitAPI/Tests/ios/KeyboardInputTestsIOS.mm:
(webViewWithAutofocusedInput):
(TestWebKitAPI::TEST):
* TestWebKitAPI/Tests/ios/TestInputDelegate.h:
* TestWebKitAPI/Tests/ios/TestInputDelegate.mm:
(-[TestInputDelegate setWillStartInputSessionHandler:]):
(-[TestInputDelegate willStartInputSessionHandler]):
(-[TestInputDelegate _webView:willStartInputSession:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRemove StorageProcess
sihui_liu@apple.com [Mon, 1 Oct 2018 20:44:06 +0000 (20:44 +0000)]
Remove StorageProcess
https://bugs.webkit.org/show_bug.cgi?id=189975

Reviewed by Geoffrey Garen.

Source/WebCore:

Clean up code. No behavior change.

* English.lproj/Localizable.strings:

Source/WebKit:

* CMakeLists.txt:
* Configurations/Storage-OSX-sandbox.entitlements: Removed.
* Configurations/StorageService.xcconfig: Removed.
* Configurations/WebKit.xcconfig:
* DerivedSources.make:
* NetworkProcess/IndexedDB/WebIDBConnectionToClient.cpp:
(WebKit::WebIDBConnectionToClient::putOrAdd):
* NetworkProcess/NetworkConnectionToWebProcess.cpp:
(WebKit::NetworkConnectionToWebProcess::establishIDBConnectionToServer):
* PlatformGTK.cmake:
* PlatformMac.cmake:
* PlatformWPE.cmake:
* PlatformWin.cmake:
* Shared/ProcessExecutablePath.h:
* Shared/Storage/StorageProcessCreationParameters.cpp: Removed.
* Shared/Storage/StorageProcessCreationParameters.h: Removed.
* Shared/glib/ProcessExecutablePathGLib.cpp:
(WebKit::executablePathOfStorageProcess): Deleted.
* Sources.txt:
* SourcesCocoa.txt:
* SourcesGTK.txt:
* SourcesWPE.txt:
* StorageProcess/EntryPoint/mac/XPCService/StorageService/Info.plist: Removed.
* StorageProcess/EntryPoint/mac/XPCService/StorageServiceEntryPoint.mm: Removed.
* StorageProcess/EntryPoint/unix/StorageProcessMain.cpp: Removed.
* StorageProcess/EntryPoint/win/StorageProcessMain.cpp: Removed.
* StorageProcess/StorageProcess.cpp: Removed.
* StorageProcess/StorageProcess.h: Removed.
* StorageProcess/StorageProcess.messages.in: Removed.
* StorageProcess/StorageToWebProcessConnection.cpp: Removed.
* StorageProcess/StorageToWebProcessConnection.h: Removed.
* StorageProcess/StorageToWebProcessConnection.messages.in: Removed.
* StorageProcess/glib/StorageProcessMainGLib.cpp: Removed.
* StorageProcess/ios/StorageProcessIOS.mm: Removed.
* StorageProcess/mac/StorageProcessMac.mm: Removed.
* StorageProcess/mac/com.apple.WebKit.Storage.sb.in: Removed.
* StorageProcess/unix/StorageProcessMainUnix.h: Removed.
* StorageProcess/win/StorageProcessMainWin.cpp: Removed.
* StorageProcess/win/StorageProcessMainWin.h: Removed.
* UIProcess/API/C/WKContext.cpp:
(WKContextTerminateStorageProcess): Deleted.
(WKContextGetDatabaseProcessIdentifier): Deleted.
* UIProcess/API/C/WKContext.h:
* UIProcess/API/C/WKContextPrivate.h:
* UIProcess/API/Cocoa/WKProcessPool.mm:
(-[WKProcessPool _terminateStorageProcess]): Deleted.
(-[WKProcessPool _storageProcessIdentifier]): Deleted.
* UIProcess/API/Cocoa/WKProcessPoolPrivate.h:
* UIProcess/Launcher/glib/ProcessLauncherGLib.cpp:
(WebKit::ProcessLauncher::launchProcess):
* UIProcess/Launcher/win/ProcessLauncherWin.cpp:
(WebKit::processName):
* UIProcess/Storage/StorageProcessProxy.cpp: Removed.
* UIProcess/Storage/StorageProcessProxy.h: Removed.
* UIProcess/Storage/StorageProcessProxy.messages.in: Removed.
* UIProcess/WebContextClient.cpp:
(WebKit::WebContextClient::storageProcessDidCrash): Deleted.
* UIProcess/WebContextClient.h:
* UIProcess/WebContextSupplement.h:
* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::sendMemoryPressureEvent):
(WebKit::WebProcessPool::setAnyPageGroupMightHavePrivateBrowsingEnabled):
(WebKit::WebProcessPool::ensureStorageProcessAndWebsiteDataStore): Deleted.
(WebKit::WebProcessPool::getStorageProcessConnection): Deleted.
(WebKit::WebProcessPool::storageProcessCrashed): Deleted.
(WebKit::WebProcessPool::storageProcessIdentifier): Deleted.
(WebKit::WebProcessPool::terminateStorageProcessForTesting): Deleted.
* UIProcess/WebProcessPool.h:
(WebKit::WebProcessPool::sendToStorageProcess): Deleted.
(WebKit::WebProcessPool::sendToStorageProcessRelaunchingIfNecessary): Deleted.
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::getStorageProcessConnection): Deleted.
* UIProcess/WebProcessProxy.h:
* UIProcess/WebProcessProxy.messages.in:
* UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::~WebsiteDataStore):
(WebKit::WebsiteDataStore::storageProcessParameters): Deleted.
* UIProcess/WebsiteData/WebsiteDataStore.h:
* WebKit.xcodeproj/project.pbxproj:
* WebProcess/Databases/IndexedDB/WebIDBConnectionToServer.h:
* WebProcess/Storage/WebServiceWorkerProvider.cpp:
* WebProcess/Storage/WebToStorageProcessConnection.cpp: Removed.
* WebProcess/Storage/WebToStorageProcessConnection.h: Removed.
* WebProcess/WebProcess.cpp:
(WebKit::WebProcess::establishWorkerContextConnectionToNetworkProcess):
(WebKit::WebProcess::webToStorageProcessConnectionClosed): Deleted.
(WebKit::WebProcess::ensureWebToStorageProcessConnection): Deleted.
* WebProcess/WebProcess.h:
* webkitglib-symbols.map:

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/ResourceLoadStatistics.mm:
(TEST):
* TestWebKitAPI/Tests/WebKitCocoa/ServiceWorkerBasic.mm:
* TestWebKitAPI/Tests/WebKitCocoa/WebProcessKillIDBCleanup.mm:
(TEST):
* TestWebKitAPI/Tests/WebKitCocoa/WebsiteDataStoreCustomPaths.mm:
(TEST):
* WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::terminateStorageProcess): Deleted.
* WebKitTestRunner/InjectedBundle/TestRunner.h:
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::generatePageConfiguration):
(WTR::TestController::databaseProcessName): Deleted.
(WTR::TestController::databaseProcessDidCrash): Deleted.
(WTR::TestController::terminateStorageProcess): Deleted.
* WebKitTestRunner/TestController.h:
* WebKitTestRunner/TestInvocation.cpp:
(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):

LayoutTests:

Since Service Worker management is in netowrk process now, we should terminate network
process instead of storage process.

* http/wpt/service-workers/persistent-importScripts.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoAdd missing Objective-C versions of downloads SPI
david_quesada@apple.com [Mon, 1 Oct 2018 20:38:47 +0000 (20:38 +0000)]
Add missing Objective-C versions of downloads SPI
https://bugs.webkit.org/show_bug.cgi?id=190149
rdar://problem/44915592

Reviewed by Alex Christensen.

* UIProcess/API/Cocoa/WKProcessPool.mm:
(-[WKProcessPool _downloadURLRequest:]):
(-[WKProcessPool _resumeDownloadFromData:path:]):
* UIProcess/API/Cocoa/WKProcessPoolPrivate.h:
* UIProcess/API/Cocoa/_WKDownload.h:
* UIProcess/API/Cocoa/_WKDownload.mm:
(-[_WKDownload resumeData]):
* UIProcess/API/Cocoa/_WKDownloadInternal.h:
* UIProcess/Cocoa/DownloadClient.mm:
    Move the DownloadProxy's WrapperTraits out of DownloadClient, and into the internal
    header so it can be used from other classes (i.e. WKProcessPool here).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoFix internal build after r236665
achristensen@apple.com [Mon, 1 Oct 2018 19:57:52 +0000 (19:57 +0000)]
Fix internal build after r236665
https://bugs.webkit.org/show_bug.cgi?id=189850

* UIProcess/API/Cocoa/WKNSURLAuthenticationChallenge.h:
* WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WebAuthN] Import a JS CBOR coder
jiewen_tan@apple.com [Mon, 1 Oct 2018 19:47:05 +0000 (19:47 +0000)]
[WebAuthN] Import a JS CBOR coder
https://bugs.webkit.org/show_bug.cgi?id=189877
<rdar://problem/44701124>

Reviewed by Chris Dumez.

Source/WebKit:

Update MockWebAuthenticationConfiguration to have userCertificateBase64 and intermediateCACertificateBase64
as Local's memeber such that tests can pass those certificates to MockLocalConnection instead of letting it
holds some static ones.

* UIProcess/API/C/WKWebsiteDataStoreRef.cpp:
(WKWebsiteDataStoreSetWebAuthenticationMockConfiguration):
* UIProcess/WebAuthentication/Mock/MockLocalConnection.mm:
(WebKit::MockLocalConnection::getAttestation const):
* UIProcess/WebAuthentication/Mock/MockWebAuthenticationConfiguration.h:

Tools:

Add logic to only process privateKeyBase64, userCertificateBase64 and intermediateCACertificateBase64
only if acceptAttestation is true.

* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::setWebAuthenticationMockConfiguration):

LayoutTests:

This patch import a 3rd party JS CBOR coder from https://github.com/paroga/cbor-js.
The library is MIT licensed, which should be fine to use within WebKit.

As a benefit from the library, tests are updated to check CBOR binaries.

* http/wpt/credential-management/credentialscontainer-store-basics.https.html:
* http/wpt/webauthn/idl.https.html:
* http/wpt/webauthn/public-key-credential-create-failure-local.https.html:
* http/wpt/webauthn/public-key-credential-create-success-local.https.html:
* http/wpt/webauthn/public-key-credential-get-failure-local.https.html:
* http/wpt/webauthn/public-key-credential-get-success-local.https.html:
* http/wpt/webauthn/resources/cbor.js: Added.
* http/wpt/webauthn/resources/util.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236686 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoSplit NaN-check into separate test
commit-queue@webkit.org [Mon, 1 Oct 2018 19:40:49 +0000 (19:40 +0000)]
Split NaN-check into separate test
https://bugs.webkit.org/show_bug.cgi?id=190010

Patch by Dominik Infuehr <dinfuehr@igalia.com> on 2018-10-01
Reviewed by Saam Barati.

DataView exposes NaN-representation, which is not necessarily the same on each
architecture. Therefore move the check of the NaN-representation into its own
file such that we can disable this test on MIPS where NaN-representation can be
different on older CPUs.

* stress/dataview-jit-set-nan.js: Added.
(assert):
(test.storeLittleEndian):
(test.storeBigEndian):
(test.store):
(test):
* stress/dataview-jit-set.js:
(test5):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236684 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRegression(r236512): http/tests/navigation/keyboard-events-during-provisional-navigat...
cdumez@apple.com [Mon, 1 Oct 2018 18:57:02 +0000 (18:57 +0000)]
Regression(r236512): http/tests/navigation/keyboard-events-during-provisional-navigation.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=190052

Reviewed by Ryosuke Niwa.

Source/WebKit:

* Platform/IPC/Connection.cpp:
(IPC::Connection::sendMessage):
* Platform/IPC/Connection.h:
* WebProcess/InjectedBundle/API/c/WKBundlePage.cpp:
(WKBundlePagePostMessageIgnoringFullySynchronousMode):
* WebProcess/InjectedBundle/API/c/WKBundlePage.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::postMessageIgnoringFullySynchronousMode):
* WebProcess/WebPage/WebPage.h:

Tools:

The test relies on EventSender to send events to the page synchronously to the page and then uses console.log
to log those events. It also uses console.log() before sending those events to indicate what the test is about
to do. Note that console.log() normally causes the WebKitTestRunner to send an asynchronous IPC to the UIProcess
so that it can log the message.
The issue is that EventSender uses IPC::SendOption::UseFullySynchronousModeForTesting when sending the
sync IPC to the UIProcess. This option causes follow-up *asynchronous* IPC sent from the synchronous IPC reply
handler to be transformed into synchronous IPC.
As a result, some of the console.log IPC ended up being asynchronous and some other ended up being synchronous.
Because synchronous and asynchronous IPC is not necessarily processed in-order by the UIProcess, the logged
messages may end up being out of order, leading to flakiness.

To address the issue, we now make sure that InjectedBundle::outputText() uses a new IPC::SendOption indicated
that the IPC should always be sent asynchronously, even if the connection is in fully synchronous mode. As a
result, all text outputing IPC to the UIProcess will be asynchronous, and thus in order.

* WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
(WTR::InjectedBundle::outputText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236680 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[MSE][GStreamer] Set a minimum sample duration
aboya@igalia.com [Mon, 1 Oct 2018 18:38:49 +0000 (18:38 +0000)]
[MSE][GStreamer] Set a minimum sample duration
https://bugs.webkit.org/show_bug.cgi?id=190125

Reviewed by Xabier Rodriguez-Calvar.

The last sample of the audio track in the asset used in this test
player has a tiny duration (100 ns):

http://orange-opensource.github.io/hasplayer.js/1.2.0/player.html?url=http://playready.directtaps.net/smoothstreaming/SSWSS720H264/SuperSpeedway_720.ism/Manifest

So small, we were truncating it to zero. We're not supposed to have
frames with zero duration. Instead, lets set a minimum frame duration
for those fringe cases.

* platform/graphics/gstreamer/MediaSampleGStreamer.cpp:
(WebCore::MediaSampleGStreamer::MediaSampleGStreamer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236679 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Special keys are misidentified in DOM keyboard events
dbates@webkit.org [Mon, 1 Oct 2018 18:35:37 +0000 (18:35 +0000)]
[iOS] Special keys are misidentified in DOM keyboard events
https://bugs.webkit.org/show_bug.cgi?id=189974

Reviewed by Wenson Hsieh.

Source/WebCore:

This patch fixes two issues:
    1. Special keyboard keys would be misidentified in dispatched DOM keyboard events.
    2. DOM keypress events may not be dispatched for some special keys.

UIKit uses special input strings to identify the Page Up, Page Down, Escape, Up Arrow, Down Arrow,
Left Arrow, and Right Arrow keys. It also uses ASCII control characters to represent some other
special keys, including Num Lock / Clear, Home, End, Forward Delete, and F1, ..., F24. We need
to explicitly handle these special keyboard keys in order to be able to identify the key that
was pressed as well as to correctly disambiguate a key down to know whether to dispatch a DOM
keypress event for the key.

Unlike UIKit, AppKit reserves Unicode Private Use Area (PUA) code points in 0xF700–0xF8FF to
represent special keyboard keys. This makes it straightforward to disambiguate such keys using
the input string of the keyboard event alone. To simplify the implementation for iOS
we normalize the input string be AppKit compatible. See the explaination for WebCore::windowsKeyCodeForCharCode()
below for more details on why this is done.

Tests: fast/events/ios/keydown-keyup-arrow-keys-in-non-editable-element.html
       fast/events/ios/keypress-keys-in-non-editable-element.html

* SourcesCocoa.txt:
* WebCore.xcodeproj/project.pbxproj:
Do not use unified source build strategy when building WebEvent.mm as it makes
use of SoftLinking macros that are incompatible with this strategy.

* platform/ios/KeyEventIOS.mm:
(WebCore::windowsKeyCodeForCharCode): Recognize some special AppKit special char codes.
These special char codes are generated by WebKit. WebKit uses the same special char codes
as AppKit as a convenience instead of defining our own constants for the same purpose.
Encoding the special UIKit input strings (e.g. up arrow) as distinct char codes allows us
to use integer arithmetic and switch blocks to map characters to Windows virtual key
codes as opposed to special cased branches to perform pointer or string comparisions.
The latter would be necessary in Modern WebKit in order for key down events to be properly
disambiguated to dispatch a DOM keypress event because pointers are not perserved, though
what they point to is, when sending the WebEvent from UIProcess to the WebProcess and
vice versa.
(WebCore::isFunctionKey): Convenience function that determines whether the specified char
code corresponds to a function key on the keyboard. The term "function key" is taken from
AppKit parlance to describe a special keyboard key. These keys include F1, F2, ..., F24,
and cursor keys among other special keyboard keys.
(WebCore::PlatformKeyboardEvent::disambiguateKeyDownEvent): Write in terms of isFunctionKey().
* platform/ios/PlatformEventFactoryIOS.h:
* platform/ios/PlatformEventFactoryIOS.mm:
(WebCore::keyIdentifierForKeyEvent): Remove code to handle UIKit special input strings as
we now map such special input strings to char codes and hence can use the default code path.
(WebCore::keyForKeyEvent): Ditto.
(WebCore::codeForKeyEvent): Remove code to compute the Window virtual key code corresponding
to a UIKit special key command now that we map such special input strings to char codes and
subsequently map the char codes to the Windows virtual key code (see -[WebEvent initWithKeyEventType:...]
constructors). So, we can now use WebEvent.keyCode directly to compute the DOM UIEvents code
for the event.
(WebCore::PlatformKeyboardEventBuilder::PlatformKeyboardEventBuilder): Remove code to fix up
WebEvent.keyCode to account for UIKit special input strings now that we map such special key
commands to char codes and subsequently map the char codes to the Windows virtual key code (see -[WebEvent initWithKeyEventType:...]
constructors). So, we can now take WebEvent.keyCode verbatim to be the Window virtual key code.
(WebCore::convertSpecialKeyToCharCode): Deleted.
(WebCore::keyCodeForEvent): Deleted.
* platform/ios/WebEvent.mm:
(normalizedStringWithAppKitCompatibilityMapping): Added; converts a UIKit character string
to the corresponding AppKit-compatible one (if not already compatible). See the explaination
for WebCore::windowsKeyCodeForCharCode() above for more details on why this is done.

(-[WebEvent initWithKeyEventType:timeStamp:characters:charactersIgnoringModifiers:modifiers:isRepeating:withFlags:keyCode:isTabKey:characterSet:]):
(-[WebEvent initWithKeyEventType:timeStamp:characters:charactersIgnoringModifiers:modifiers:isRepeating:withFlags:withInputManagerHint:keyCode:isTabKey:]):
Normalize the character strings to be AppKit compatible.

Source/WebCore/PAL:

Forward declare or define more SPI.

* pal/spi/cocoa/IOKitSPI.h:
* pal/spi/ios/UIKitSPI.h:

Source/WebKit:

Take the key code of WebEvent to be the key code for the new WebKeyboardEvent verbatim
now that we normalize the character strings of the WebEvent to account for the special
UIKit input strings.

* Shared/ios/WebIOSEventFactory.mm:
(WebIOSEventFactory::createWebKeyboardEvent):

Tools:

Add support for testing keys Forward Delete and Num Lock / Clear.

* WebKitTestRunner/ios/HIDEventGenerator.mm:
(hidUsageCodeForCharacter):

LayoutTests:

Add tests to ensure that we do not regress key identification for special keys.

Update the expected results for test fast/events/ios/keydown-keyup-special-keys-in-non-editable-element.html
now that we correctly identify some more keys.

* fast/events/ios/keydown-keyup-arrow-keys-in-non-editable-element-expected.txt: Added.
* fast/events/ios/keydown-keyup-arrow-keys-in-non-editable-element.html: Added.
* fast/events/ios/keydown-keyup-special-keys-in-non-editable-element-expected.txt:
* fast/events/ios/keypress-keys-in-non-editable-element-expected.txt: Added.
* fast/events/ios/keypress-keys-in-non-editable-element.html: Added.
* resources/ui-helper.js:
(window.UIHelper.typeCharacter): Actually type the specified character in DumpRenderTree.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236678 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoOptimize RenderStyle::diff() and clean up the code
simon.fraser@apple.com [Mon, 1 Oct 2018 18:32:32 +0000 (18:32 +0000)]
Optimize RenderStyle::diff() and clean up the code
https://bugs.webkit.org/show_bug.cgi?id=190104

Reviewed by Dan Bernstein.

RenderStyle::changeRequiresLayout() and related should only check values on
m_rareNonInheritedData and m_rareInheritedData after checking for pointer equality.
To reduce the chances of future changes regressing this, move code comparing values
on StyleRare[Non]InheritedData into dedication functions.

In addition, the transform comparison double-compared the transformOperations,
because m_rareNonInheritedData->transform != other.m_rareNonInheritedData->transform
is a deep comparison, and it was followed by *m_rareNonInheritedData->transform != *other.m_rareNonInheritedData->transform.
Change the first to be a pointer comparison.

* rendering/style/RenderStyle.cpp:
(WebCore::rareNonInheritedDataChangeRequiresLayout):
(WebCore::rareInheritedDataChangeRequiresLayout):
(WebCore::RenderStyle::changeRequiresLayout const):
(WebCore::rareNonInheritedDataChangeRequiresLayerRepaint):
(WebCore::RenderStyle::changeRequiresLayerRepaint const):
(WebCore::rareNonInheritedDataChangeRequiresRepaint):
(WebCore::rareInheritedDataChangeRequiresRepaint):
(WebCore::RenderStyle::changeRequiresRepaint const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236677 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoVersioning.
kocsen_chung@apple.com [Mon, 1 Oct 2018 18:27:55 +0000 (18:27 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236676 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoFix an internal build after r236665
achristensen@apple.com [Mon, 1 Oct 2018 18:27:30 +0000 (18:27 +0000)]
Fix an internal build after r236665
https://bugs.webkit.org/show_bug.cgi?id=189850

* WebKit.xcodeproj/project.pbxproj:
Make WKNSURLAuthenticationChallenge.h a private header.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236675 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoURL should not use TextEncoding internally
achristensen@apple.com [Mon, 1 Oct 2018 18:16:33 +0000 (18:16 +0000)]
URL should not use TextEncoding internally
https://bugs.webkit.org/show_bug.cgi?id=190111

Reviewed by Andy Estes.

Source/WebCore:

That dependency makes it impossible to move or use elsewhere.
Using TextEncoding was overkill because we know the credentials are UTF-8 percent-encoded in a parsed URL.
No change in behavior as verified by new API tests.

* page/SecurityOrigin.cpp:
* page/csp/ContentSecurityPolicySourceList.cpp:
* platform/URL.cpp:
(WebCore::decodeEscapeSequencesFromParsedURL):
(WebCore::URL::user const):
(WebCore::URL::pass const):
(WebCore::URL::fileSystemPath const):
(WebCore::decodeURLEscapeSequences): Deleted.
* platform/URL.h:
* platform/network/DataURLDecoder.cpp:
* platform/text/TextEncoding.cpp:
(WebCore::decodeURLEscapeSequences):
* platform/text/TextEncoding.h:

Source/WebKit:

* UIProcess/WebInspectorProxy.cpp:

Source/WebKitLegacy/mac:

* Misc/WebNSURLExtras.mm:

Tools:

* TestWebKitAPI/Tests/WebCore/URLParser.cpp:
(TestWebKitAPI::testUserPass):
(TestWebKitAPI::TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236674 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago<form> in quirks mode should have margin-block-end: 1em
commit-queue@webkit.org [Mon, 1 Oct 2018 18:08:35 +0000 (18:08 +0000)]
<form> in quirks mode should have margin-block-end: 1em
https://bugs.webkit.org/show_bug.cgi?id=157788

Patch by Simon Pieters <zcorpan@gmail.com> on 2018-10-01
Reviewed by Simon Fraser.

LayoutTests/imported/w3c:

* web-platform-tests/html/rendering/non-replaced-elements/flow-content-0/form-margin-quirk.html: Added.
* web-platform-tests/html/rendering/non-replaced-elements/flow-content-0/form-margin-quirk-expected.txt: Added.

Source/WebCore:

Change the default style for forms to take writing-mode into account
in quirks mode. Matches the behavior of Gecko and Edge and the HTML
standard.

Spec: https://html.spec.whatwg.org/multipage/rendering.html#flow-content-3

Test: imported/w3c/web-platform-tests/html/rendering/non-replaced-elements/flow-content-0/form-margin-quirk.html

* css/quirks.css:
(form):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236673 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnify implementation in VideoFullscreenInterfaceAVKit
commit-queue@webkit.org [Mon, 1 Oct 2018 17:59:55 +0000 (17:59 +0000)]
Unify implementation in VideoFullscreenInterfaceAVKit
https://bugs.webkit.org/show_bug.cgi?id=190091
rdar://problem/44734523

Patch by Jeremy Jones <jeremyj@apple.com> on 2018-10-01
Reviewed by Jer Noble.

Source/WebCore:

No new tests because no behavior change.

Unified code in VideoFullscreenInterfaceAVKit now that new code path is proven and include
any changes that had been made in the old path.

* platform/ios/VideoFullscreenInterfaceAVKit.h:
* platform/ios/VideoFullscreenInterfaceAVKit.mm:
(-[WebAVPlayerViewControllerDelegate playerViewControllerShouldStartPictureInPictureFromInlineWhenEnteringBackground:]):
(VideoFullscreenInterfaceAVKit::preparedToExitFullscreen):
(VideoFullscreenInterfaceAVKit::shouldExitFullscreenWithReason):
* platform/ios/WebVideoFullscreenControllerAVKit.mm:
(VideoFullscreenControllerContext::requestUpdateInlineRect):
(VideoFullscreenControllerContext::requestVideoContentLayer):
(VideoFullscreenControllerContext::returnVideoContentLayer):
(VideoFullscreenControllerContext::didSetupFullscreen):
(VideoFullscreenControllerContext::didExitFullscreen):

Source/WebKit:

Unified code in VideoFullscreenInterfaceAVKit now that new code path is proven and include
any changes that had been made in the old path.

* UIProcess/Cocoa/VideoFullscreenManagerProxy.mm:
* WebProcess/cocoa/VideoFullscreenManager.mm:
(WebKit::VideoFullscreenManager::supportsVideoFullscreenStandby const):
(WebKit::VideoFullscreenManager::didSetupFullscreen):
(WebKit::VideoFullscreenManager::didExitFullscreen):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236671 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Web Animations] Ensure renderers with accelerated animations have layers
graouts@webkit.org [Mon, 1 Oct 2018 17:44:39 +0000 (17:44 +0000)]
[Web Animations] Ensure renderers with accelerated animations have layers
https://bugs.webkit.org/show_bug.cgi?id=189990

Reviewed by Simon Fraser.

In r236501 we added code that would make a RenderBox and a RenderInline query the document timeline for whether a given element has
accelerated animations running on it. Since the calls to requiresLayer() are in a hot path, we instead keep a list of elements with
exclusively accelerated animations running.

No new tests, this is already covered by webanimations/accelerated-animation-with-delay.html and webanimations/opacity-animation-yields-compositing-span.html
which respectively check that we can apply an accelerated animation to a non-positioned block and an inline element.

* animation/AnimationTimeline.h:
* animation/DocumentTimeline.cpp:
(WebCore::DocumentTimeline::detachFromDocument):
(WebCore::DocumentTimeline::animationWasAddedToElement):
(WebCore::DocumentTimeline::animationWasRemovedFromElement):
(WebCore::DocumentTimeline::animationAcceleratedRunningStateDidChange):
(WebCore::DocumentTimeline::updateListOfElementsWithRunningAcceleratedAnimationsForElement): Iterate over an element's animations to determine
whether all of its animations are running accelerated, then update the HashSet containing elements running accelerated animations to remove or
add this element.
(WebCore::DocumentTimeline::runningAnimationsForElementAreAllAccelerated const): Make a simple contains() call on the HashSet containing elements
running accelerated animations.
* animation/DocumentTimeline.h:
* animation/KeyframeEffectReadOnly.cpp:
(WebCore::KeyframeEffectReadOnly::updateAcceleratedAnimationState):
(WebCore::KeyframeEffectReadOnly::applyPendingAcceleratedActions):
* rendering/RenderBoxModelObject.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236670 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Wrong key event may be sent to UIKit
dbates@webkit.org [Mon, 1 Oct 2018 17:41:25 +0000 (17:41 +0000)]
[iOS] Wrong key event may be sent to UIKit
https://bugs.webkit.org/show_bug.cgi?id=189992

Reviewed by Simon Fraser.

Retain a clone of a received UIEvent if it is for a hardware key event so as to ensure that we
notify the UIKit keyboard code of the correct keyboard event.

Currently the UIProcess retains the UIEvent associated with a keyboard event so as to defer
notifying the UIKit keyboard code (via -_handleKeyUIEvent) about a received keyboard event until
after the WebProcess has processed the raw key event. If this UIEvent is for a hardware keyboard
event then it is not sufficient to retain it to preserve its value because UIKit uses a singleton
UIEvent for all hardware keyboard events ;=> its value will be clobbered as each hardware keyboard
event is received. Instead we need to explicitly clone a UIEvent for a hardware key event before
retaining it.

* Platform/spi/ios/UIKitSPI.h: Forward declare SPI.
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView handleKeyEvent:]): Clone the UIEvent if it is for a hardware key event.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236669 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[GStreamer] Fix abort in gst_sample_get_info()
aboya@igalia.com [Mon, 1 Oct 2018 17:40:48 +0000 (17:40 +0000)]
[GStreamer] Fix abort in gst_sample_get_info()
https://bugs.webkit.org/show_bug.cgi?id=190135

Reviewed by Philippe Normand.

A flush can occur before any frame has finished decoding -- especially
in tests, where actions on the player often occur in quick succession.

Therefore, the code must not assume by the time a flush occurs any
frame has reached the sink. This patch fixes a case when such wrong
assumption was causing gst_sample_get_info() to abort (crashing
WebKit).

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236668 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoLLDB tests may use wrong configuration of lldbWebKitTester
dbates@webkit.org [Mon, 1 Oct 2018 17:22:54 +0000 (17:22 +0000)]
LLDB tests may use wrong configuration of lldbWebKitTester
https://bugs.webkit.org/show_bug.cgi?id=189011

Reviewed by Dean Jackson.

Fixes an issue where the LLDB unit tests may use the wrong configuration of lldbWebKitTester.
Additionally, add back the unit tests originally added in r235376 (https://bugs.webkit.org/show_bug.cgi?id=188936).

Currently the unit tests pick the LLDBWebKitTester binary based on the configuration set by
script set-webkit-configuration. This may not be the same configuration specified to test-webkitpy.
So, the unit tests may use the wrong binary. Instead we have test-webkitpy store the path
to the correct lldbWebKitTester binary that the unit tests should use in an environment variable,
called LLDB_WEBKIT_TESTER_EXECUTABLE. (We use an environment variable because the test driver,
test-webkitpy, and the unit tests cannot talk to each other directly due the limitations of
the Python unittest module, the Python multiprocessing module, and the general principle of unit
tests - to test code in isolation). The unit tests make use of the value of this environment
variable to launch lldbWebKitTester binary.

* Scripts/build-lldbwebkittester: We only support building lldbWebKitTester on Mac for now. Error
out if this script is invoked for a non-Mac platform.
* Scripts/webkitpy/test/main.py:
(Tester._run_tests): Build lldbWebKitTester and store the path to it in the environment variable
LLDB_WEBKIT_TESTER_EXECUTABLE.
(_build_lldb_webkit_tester): Deleted.
* lldb/dump_class_layout_unittest.py:
(TestDumpClassLayout.shouldSkip): Use SystemHost() here and remove the _host global variable as
this is the only call site that needs the Host object now.
(TestDumpClassLayout.setUpClass): Take the path to the lldbWebKitTester binary from the value of
the environment variable LLDB_WEBKIT_TESTER_EXECUTABLE.
* lldb/lldbWebKitTester/main.cpp:
(testSummaryProviders): Add back unit test support infrastructure originally added in r235376 (https://bugs.webkit.org/show_bug.cgi?id=188936).
* lldb/lldb_webkit_unittest.py:
(LLDBDebugSession.setup): Take the path to the lldbWebKitTester binary from the value of the environment
variable LLDB_WEBKIT_TESTER_EXECUTABLE.
(TestSummaryProviders):
(TestSummaryProviders.shouldSkip): Skip the tests on non-Mac platforms.
(TestSummaryProviders.serial_test_WTFHashSet_tablesize_and_size):
(TestSummaryProviders.serial_test_WTFOptionSet_SummaryProvider_empty):
(TestSummaryProviders.serial_test_WTFOptionSet_SummaryProvider_simple):
(TestSummaryProviders.serial_test_WTFOptionSetProvider_empty):
(TestSummaryProviders.serial_test_WTFOptionSetProvider_simple):
Add back the unit tests originally added in r235376 (https://bugs.webkit.org/show_bug.cgi?id=188936).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236667 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WPE] fix buffer over-read in RenderThemeWPE::mediaControlsStyleSheet()
commit-queue@webkit.org [Mon, 1 Oct 2018 17:19:47 +0000 (17:19 +0000)]
[WPE] fix buffer over-read in RenderThemeWPE::mediaControlsStyleSheet()
https://bugs.webkit.org/show_bug.cgi?id=190139

Patch by Olivier Blin <olivier.blin@softathome.com> on 2018-10-01
Reviewed by Michael Catanzaro.

Like done upstream for EFL in r210213
https://bugs.webkit.org/show_bug.cgi?id=166622

This has been detected by a charactersAreAllASCII() assert failure.

This is because ASCIILiteral() is wrongly used in mediaControlsStyleSheet().
mediaControlsBaseUserAgentStyleSheet is a char array, not a null-terminated string.
It is thus incorrect to use StringImpl::createFromLiteral() that calls
strlen() to get the string length.

The String::ConstructFromLiteral constructor can not be used, since it
skips the last character.

* platform/wpe/RenderThemeWPE.cpp:
(WebCore::RenderThemeWPE::mediaControlsStyleSheet):
Explicitely pass the size to the String constructor.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236666 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoDeprecate ObjC SPI in Deprecated Xcode group
achristensen@apple.com [Mon, 1 Oct 2018 17:16:54 +0000 (17:16 +0000)]
Deprecate ObjC SPI in Deprecated Xcode group
https://bugs.webkit.org/show_bug.cgi?id=189850

Reviewed by Andy Estes.

* Shared/Cocoa/APIObject.mm:
(API::Object::newObject):
* Shared/mac/ObjCObjectGraph.mm:
(WebKit::typeFromObject):
(WebKit::ObjCObjectGraph::encode):
(WebKit::ObjCObjectGraph::decode):
* UIProcess/API/Cocoa/WKBrowsingContextController.h:
* UIProcess/API/Cocoa/WKBrowsingContextController.mm:
(didStartProvisionalNavigation):
(didReceiveServerRedirectForProvisionalNavigation):
(didFailProvisionalNavigation):
(didCommitNavigation):
(didFinishNavigation):
(didFailNavigation):
(canAuthenticateAgainstProtectionSpace):
(didReceiveAuthenticationChallenge):
(processDidCrash):
(setUpPagePolicyClient):
* UIProcess/API/Cocoa/WKBrowsingContextGroup.h:
* UIProcess/API/Cocoa/WKBrowsingContextGroupInternal.h:
* UIProcess/API/Cocoa/WKBrowsingContextHistoryDelegate.h:
* UIProcess/API/Cocoa/WKBrowsingContextLoadDelegate.h:
* UIProcess/API/Cocoa/WKBrowsingContextLoadDelegatePrivate.h:
* UIProcess/API/Cocoa/WKBrowsingContextPolicyDelegate.h:
* UIProcess/API/Cocoa/WKNSURLAuthenticationChallenge.h:
* UIProcess/API/Cocoa/WKNSURLAuthenticationChallenge.mm:
(-[WKNSURLAuthenticationChallengeSender cancelAuthenticationChallenge:]):
(-[WKNSURLAuthenticationChallengeSender continueWithoutCredentialForAuthenticationChallenge:]):
(-[WKNSURLAuthenticationChallengeSender useCredential:forAuthenticationChallenge:]):
(-[WKNSURLAuthenticationChallengeSender performDefaultHandlingForAuthenticationChallenge:]):
(-[WKNSURLAuthenticationChallengeSender rejectProtectionSpaceAndContinueWithChallenge:]):
* UIProcess/API/Cocoa/WKNavigationData.h:
* UIProcess/API/Cocoa/WKNavigationDataInternal.h:
* UIProcess/API/Cocoa/WKProcessGroup.h:
* UIProcess/API/Cocoa/WKProcessGroup.mm:
(didCreateConnection):
(getInjectedBundleInitializationUserData):
(didNavigateWithNavigationData):
(didPerformClientRedirect):
(didPerformServerRedirect):
(didUpdateHistoryTitle):
* UIProcess/API/Cocoa/WKTypeRefWrapper.h:
* UIProcess/API/Cocoa/WKView.h:
(WK_CLASS_DEPRECATED_WITH_REPLACEMENT):
* UIProcess/API/Cocoa/_WKThumbnailView.h:
* UIProcess/API/Cocoa/_WKThumbnailView.mm:
* UIProcess/Cocoa/WebProcessProxyCocoa.mm:
(WebKit::WebProcessProxy::transformHandlesToObjects):
(WebKit::WebProcessProxy::transformObjectsToHandles):
* UIProcess/Cocoa/WebViewImpl.h:
* UIProcess/Cocoa/WebViewImpl.mm:
* UIProcess/Gamepad/mac/UIGamepadProviderMac.mm:
(WebKit::UIGamepadProvider::platformWebPageProxyForGamepadInput):
* UIProcess/ios/WKContentView.h:
* UIProcess/mac/WebInspectorProxyMac.mm:
(WebKit::WebInspectorProxy::inspectedViewFrameDidChange):
* WebProcess/cocoa/WebProcessCocoa.mm:
(WebKit::WebProcess::transformHandlesToObjects):
(WebKit::WebProcess::transformObjectsToHandles):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236665 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoREGRESSION (r235948?): Layout Test media/media-controls-accessibility.html is a flaky...
ryanhaddad@apple.com [Mon, 1 Oct 2018 16:41:02 +0000 (16:41 +0000)]
REGRESSION (r235948?): Layout Test media/media-controls-accessibility.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=189674

Unreviewed test gardening..

* platform/mac/TestExpectations: Remove flaky expectation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236664 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoPatch for landing
rwlbuis@webkit.org [Mon, 1 Oct 2018 16:17:36 +0000 (16:17 +0000)]
Patch for landing

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236663 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[macOS Sierra] Layout Test http/wpt/cache-storage/cache-put-keys.https.any.worker...
youenn@apple.com [Mon, 1 Oct 2018 16:17:29 +0000 (16:17 +0000)]
[macOS Sierra] Layout Test http/wpt/cache-storage/cache-put-keys.https.any.worker.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=184204

Reviewed by Chris Dumez.

Source/WebKit:

NetworkCache::Storage by default limits the length to read to 1500 milliseconds.
This is good for the HTTP cache since networking might be faster.
It does not work for DOM cache which needs to get these resources even if it takes a long time.

Since this is disabled by a Mode::Testing option, use it for DOMCache after renaming it to Mode::AvoidRandomness.

Add a bunch of release logging to help debugging error cases.

* NetworkProcess/cache/CacheStorageEngineCaches.cpp:
(WebKit::CacheStorage::Caches::retrieveOriginFromDirectory):
(WebKit::CacheStorage::Caches::initialize):
(WebKit::CacheStorage::Caches::writeCachesToDisk):
(WebKit::CacheStorage::Caches::readRecord):
* NetworkProcess/cache/NetworkCache.cpp:
(WebKit::NetworkCache::Cache::open):
* NetworkProcess/cache/NetworkCacheStorage.cpp:
(WebKit::NetworkCache::Storage::dispatchReadOperation):
(WebKit::NetworkCache::Storage::shrinkIfNeeded):
* NetworkProcess/cache/NetworkCacheStorage.h:

LayoutTests:

* platform/mac-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236662 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoMake crossOriginObject.then undefined for promises
cdumez@apple.com [Mon, 1 Oct 2018 16:10:52 +0000 (16:10 +0000)]
Make crossOriginObject.then undefined for promises
https://bugs.webkit.org/show_bug.cgi?id=190094

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Rebaseline WPT test now that more checks are passing.

* web-platform-tests/html/browsers/origin/cross-origin-objects/cross-origin-objects-expected.txt:

Source/WebCore:

Make crossOriginObject.then undefined for promises. This allows promises to work better with cross-origin WindowProxy
and Location objects.

Specification:
- https://github.com/whatwg/html/pull/3242
- https://github.com/whatwg/dom/issues/536

This aligns our behavior with Blink and Gecko.

No new tests, rebaselined existing test.

* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::jsDOMWindowGetOwnPropertySlotRestrictedAccess):
(WebCore::addCrossOriginWindowOwnPropertyNames):
* bindings/js/JSLocationCustom.cpp:
(WebCore::getOwnPropertySlotCommon):
(WebCore::addCrossOriginLocationOwnPropertyNames):

LayoutTests:

Update existing tests to reflect behavior change.

* http/tests/navigation/process-swap-window-open-expected.txt:
* http/tests/navigation/process-swap-window-open.html:
* http/wpt/cross-origin-window-policy/resources/utils.js:
(testCrossOriginOption):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236661 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[JSC] Add a JSONStringify overload that receives a JSValue space
commit-queue@webkit.org [Mon, 1 Oct 2018 15:12:24 +0000 (15:12 +0000)]
[JSC] Add a JSONStringify overload that receives a JSValue space
https://bugs.webkit.org/show_bug.cgi?id=190131

Patch by Koby Boyango <koby.b@mce-sys.com> on 2018-10-01
Reviewed by Yusuke Suzuki.

* runtime/JSONObject.cpp:
* runtime/JSONObject.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236660 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WPE] Remove WebKit2InspectorGResourceBundle.xml
commit-queue@webkit.org [Mon, 1 Oct 2018 13:49:48 +0000 (13:49 +0000)]
[WPE] Remove WebKit2InspectorGResourceBundle.xml
https://bugs.webkit.org/show_bug.cgi?id=190132

Patch by Olivier Blin <olivier.blin@softathome.com> on 2018-10-01
Reviewed by Michael Catanzaro.

* UIProcess/API/wpe/WebKit2InspectorGResourceBundle.xml: Removed.
This is unused since legacy INSPECTOR_SERVER implementation has
been removed in r217924.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236659 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[SOUP] Fix the build for libsoup > 2.61.90
commit-queue@webkit.org [Mon, 1 Oct 2018 13:47:53 +0000 (13:47 +0000)]
[SOUP] Fix the build for libsoup > 2.61.90
https://bugs.webkit.org/show_bug.cgi?id=190126

Patch by Xan Lopez <xan@igalia.com> on 2018-10-01
Reviewed by Michael Catanzaro.

* platform/network/soup/SocketStreamHandleImplSoup.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236658 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WPE] Update install-dependencies
commit-queue@webkit.org [Mon, 1 Oct 2018 13:26:50 +0000 (13:26 +0000)]
[WPE] Update install-dependencies
https://bugs.webkit.org/show_bug.cgi?id=190127

Patch by Olivier Blin <olivier.blin@softathome.com> on 2018-10-01
Reviewed by Michael Catanzaro.

* wpe/install-dependencies:
libxrandr-dev is needed for gstreamer-vaapi in jhbuild
It is used by gstvaapidisplay_x11.c.
Add it only for Apt, already ok for Pacman and DNF.

Install libevent-dev for WebRTC (like GTK)

Install libasound2-dev for WebRTC.
Add it only for Apt, already ok for Pacam and DNF.

Install libgbm-dev for HeadlessViewBackend, which uses gbm.h

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236657 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[MSE][GStreamer] Reset running time in PlaybackPipeline::flush()
aboya@igalia.com [Mon, 1 Oct 2018 13:11:26 +0000 (13:11 +0000)]
[MSE][GStreamer] Reset running time in PlaybackPipeline::flush()
https://bugs.webkit.org/show_bug.cgi?id=190076

Reviewed by Philippe Normand.

Source/WebCore:

Test: media/media-source/media-source-seek-redundant-append.html

PlaybackPipeline::flush() is called when already enqueued frames are
appended again. This may be caused by a quality change or just a
redundant append. Either way, the pipeline has to be flushed and
playback begin again, but without changing the player position by
much.

There are two kinds of time to consider here: stream time (i.e. the
time of a frame as written in the file, e.g. a frame may have stream
time 0:01:00), and running time (i.e. how much time since playback
started should pass before the frame should be played, e.g. if we
started playing at 0:00:59 that same frame would have a running time
of just 1 second).

Notice how running time depends on where and when playback starts.
Running time can also be optionally resetted after a flush. (This is
indeed done currently by most demuxers after a seek.)

Instead of resetting running time, PlaybackPipeline used to modify the
first GstSegment emitted after the flush. A GstSegment declares the
mapping between stream time and running time for the following frames.
There, PlaybackPipeline used to set `base` (the running time at which
the segment starts) to the position reported by a position query
(which is stream time).

This, of course, only worked when playback (or the last seek) started
at stream time 0:00:00, since that's the only case where running time
equals stream time. In other cases delays as long as the difference
between these timelines would appear. This is demonstrated in the
attached test, where seeks and appends are made in such an order that
the difference is more than 5 minutes, making the playback stall for
>5 minutes before playing 1 second of audio.

This patch fixes the problem by resetting running time with the flush
and not modifying GstSegment.base anymore (it will be left as zero,
which is now correct since the running time has been reset).

* platform/graphics/gstreamer/mse/PlaybackPipeline.cpp:
(WebCore::PlaybackPipeline::flush):
(WebCore::segmentFixerProbe): Deleted.

LayoutTests:

A test where a seek is followed by a redundant append is added. This
test timed out in the GStreamer MSE implementation before the
accompanying patch fixed it.

The MIME type declared in test-48khz-manifest.json has also been
changed, from non-standard `audio/x-m4a` to `audio/mp4;
codecs="mp4a.40.2"`, as implied by the MSE specs. This should not
affect other tests because no other tests were reading this type
string before.

* media/media-source/content/test-48khz-manifest.json:
* media/media-source/media-source-seek-redundant-append-expected.txt: Added.
* media/media-source/media-source-seek-redundant-append.html: Added.
* media/video-test.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236656 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WTF][JSCONLY] Use MainThreadWin on Windows in the JSCOnly port
commit-queue@webkit.org [Mon, 1 Oct 2018 12:33:05 +0000 (12:33 +0000)]
[WTF][JSCONLY] Use MainThreadWin on Windows in the JSCOnly port
https://bugs.webkit.org/show_bug.cgi?id=190121

Patch by Koby Boyango <koby.b@mce-sys.com> on 2018-10-01
Reviewed by Yusuke Suzuki.

This fixes JSCOnly build on Windows after r236617.

* wtf/PlatformJSCOnly.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236655 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago'HAVE_PTHREAD_MAIN_NP' macro redefined warning when building --jsc-only on macOS
ticaiolima@gmail.com [Mon, 1 Oct 2018 11:56:20 +0000 (11:56 +0000)]
'HAVE_PTHREAD_MAIN_NP' macro redefined warning when building --jsc-only on macOS
https://bugs.webkit.org/show_bug.cgi?id=190118

Reviewed by Yusuke Suzuki.

We define HAVE_PTHREAD_MAIN_NP in WTF/wtf/Platform.h when the OS is
Darwin. In such case, we don't need to check symbols for
pthread_main_np when generating build for this system.

* Source/cmake/OptionsCommon.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236654 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Flatpak] Add libunwind in the sandbox
commit-queue@webkit.org [Mon, 1 Oct 2018 11:41:00 +0000 (11:41 +0000)]
[Flatpak] Add libunwind in the sandbox
https://bugs.webkit.org/show_bug.cgi?id=190123

Allowing us to get proper traces in GStreamer trace grabing
debug functions.

Patch by Thibault Saunier <tsaunier@igalia.com> on 2018-10-01
Reviewed by Philippe Normand.

* flatpak/org.webkit.WebKit.yaml:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236653 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, rolling out r236647.
commit-queue@webkit.org [Mon, 1 Oct 2018 11:26:08 +0000 (11:26 +0000)]
Unreviewed, rolling out r236647.
https://bugs.webkit.org/show_bug.cgi?id=190124

Breaking test stress/big-int-to-string.js (Requested by
caiolima_ on #webkit).

Reverted changeset:

"[BigInt] BigInt.proptotype.toString is broken when radix is
power of 2"
https://bugs.webkit.org/show_bug.cgi?id=190033
https://trac.webkit.org/changeset/236647

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236652 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WebAssembly] Move type conversion code of JSToWasm return type to JS wasm wrapper
yusukesuzuki@slowstart.org [Mon, 1 Oct 2018 09:25:41 +0000 (09:25 +0000)]
[WebAssembly] Move type conversion code of JSToWasm return type to JS wasm wrapper
https://bugs.webkit.org/show_bug.cgi?id=189498

Reviewed by Saam Barati.

To call JS-to-Wasm code we need to convert the result value from wasm function to
the JS type. Previously this is done by callWebAssemblyFunction by using swtich
over signature.returnType(). But since we know the value of `signature.returnType()`
at compiling phase, we can emit a small conversion code directly to JSToWasm glue
and remove this switch from callWebAssemblyFunction.

In JSToWasm glue code, we do not have tag registers. So we use DoNotHaveTagRegisters
in boxInt32 and boxDouble. Since boxDouble does not have DoNotHaveTagRegisters version,
we add an implementation for that.

* jit/AssemblyHelpers.h:
(JSC::AssemblyHelpers::boxDouble):
* wasm/js/JSToWasm.cpp:
(JSC::Wasm::createJSToWasmWrapper):
* wasm/js/WebAssemblyFunction.cpp:
(JSC::callWebAssemblyFunction):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236651 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Web Animations] Fix test regressions after enabling Web Animations with CSS Integration
graouts@webkit.org [Mon, 1 Oct 2018 08:04:52 +0000 (08:04 +0000)]
[Web Animations] Fix test regressions after enabling Web Animations with CSS Integration
https://bugs.webkit.org/show_bug.cgi?id=190032

Unreviewed. We need to opt into the legacy animation engine for these tests to fix some crashes in other
tests due to webkit.org/b/186946.

* legacy-animation-engine/animations/cross-fade-webkit-mask-image.html:
* legacy-animation-engine/transitions/delay.html:
* legacy-animation-engine/transitions/transition-timing-function.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236650 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUse Position instead of Range in createMarkupInternal
rniwa@webkit.org [Mon, 1 Oct 2018 05:30:02 +0000 (05:30 +0000)]
Use Position instead of Range in createMarkupInternal
https://bugs.webkit.org/show_bug.cgi?id=190107

Reviewed by Darin Adler.

Use two Position's indicating start and end instead of Range in createMarkupInternal and StylizedMarkupAccumulator
in order to support copy & paste across shadow boundaries in the bug 157443. This patch also removes the use of
Range in MarkupAccumulator since all uses of range is via StylizedMarkupAccumulator.

Also renamed createMarkupInternal to serializePreservingVisualAppearanceInternal to match the rename in r236612.

* dom/Position.cpp:
(WebCore::Position::firstNode const):  Added.
* dom/Position.h:
* editing/MarkupAccumulator.cpp:
(WebCore::MarkupAccumulator::MarkupAccumulator): No longer takes Range.
(WebCore::MarkupAccumulator::appendText): Removed the code to truncate string at the boundary points of the range.
* editing/MarkupAccumulator.h:
(WebCore::MarkupAccumulator): Made this class non-copyable.
* editing/markup.cpp:
(WebCore::StyledMarkupAccumulator::StyledMarkupAccumulator): Now takes and stores two positions.

(WebCore::StyledMarkupAccumulator::appendText): Use textContentRespectingRange in the case annotation is disabled
instead of calling to MarkupAccumulator::appendText, which no longer respects boundary offsets.

(WebCore::StyledMarkupAccumulator::renderedTextRespectingRange): Renamed from renderedText. Updated to respect
boundary offsets defined by m_start and m_end Positions instead of m_range Range.

(WebCore::StyledMarkupAccumulator::textContentRespectingRange): Renamed from stringValueForRange. Ditto.

(WebCore::StyledMarkupAccumulator::serializeNodes): Now computes startNode and pastEnd nodes from start and end
Positions. Note that the end position is always the next node in the tree order  for a character node
and computeNodeAfterPosition returns nullptr for a character data.

(WebCore::highestAncestorToWrapMarkup): Now takes two positions instead of a range.

(WebCore::serializePreservingVisualAppearanceInternal): Renamed from createMarkupInternal. Removed the obsolete
comments which were added for DOMRange in WebKitLegacy.

(WebCore::serializePreservingVisualAppearance):

(WebCore::sanitizedMarkupForFragmentInDocument): Create positions instead of a range to pass to
serializePreservingVisualAppearanceInternal.

(WebCore::serializeFragment):

* editing/markup.h:
* page/PageSerializer.cpp:
(WebCore::PageSerializer::SerializerMarkupAccumulator): Removed the unnecessary WebCore namespace qualifier.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236649 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoAudioNode.connect should return passed destination node
commit-queue@webkit.org [Sun, 30 Sep 2018 22:07:45 +0000 (22:07 +0000)]
AudioNode.connect should return passed destination node
https://bugs.webkit.org/show_bug.cgi?id=188834

Patch by Walker Henderson <wjahenderson@gmail.com> on 2018-09-30
Reviewed by Eric Carlson.

LayoutTests/imported/w3c:

* web-platform-tests/webaudio/the-audio-api/the-audionode-interface/audionode-connect-return-value-expected.txt:

Source/WebCore:

No new tests, rebaselined existing test.

* Modules/webaudio/AudioBasicInspectorNode.cpp:
(WebCore::AudioBasicInspectorNode::connect): Deleted.
* Modules/webaudio/AudioBasicInspectorNode.h:
* Modules/webaudio/AudioNode.cpp:
* Modules/webaudio/AudioNode.h:
* Modules/webaudio/AudioNode.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236648 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[BigInt] BigInt.proptotype.toString is broken when radix is power of 2
ticaiolima@gmail.com [Sun, 30 Sep 2018 18:18:55 +0000 (18:18 +0000)]
[BigInt] BigInt.proptotype.toString is broken when radix is power of 2
https://bugs.webkit.org/show_bug.cgi?id=190033

Reviewed by Yusuke Suzuki.

JSTests:

* stress/big-int-to-string.js:

Source/JavaScriptCore:

The implementation of JSBigInt::toStringToGeneric doesn't handle power
of 2 radix when JSBigInt length is >= 2. To handle such cases, we
implemented JSBigInt::toStringBasePowerOfTwo that follows the
algorithm that groups bits using mask of (2 ^ n) - 1 to extract every
digit.

* runtime/JSBigInt.cpp:
(JSC::JSBigInt::toString):
(JSC::JSBigInt::toStringBasePowerOfTwo):
* runtime/JSBigInt.h:

Source/WTF:

* wtf/MathExtras.h:
(WTF::ctz32):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236647 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[MediaStream] Clean up RealtimeMediaSource settings change handling
eric.carlson@apple.com [Sun, 30 Sep 2018 14:32:15 +0000 (14:32 +0000)]
[MediaStream] Clean up RealtimeMediaSource settings change handling
https://bugs.webkit.org/show_bug.cgi?id=189998
<rdar://problem/44797884>

Reviewed by Youenn Fablet.

Source/WebCore:

No new tests, updated webrtc/video-disabled-black.html.

* Modules/mediastream/CanvasCaptureMediaStreamTrack.cpp:
(WebCore::CanvasCaptureMediaStreamTrack::Source::Source):
(WebCore::CanvasCaptureMediaStreamTrack::Source::settings):
(WebCore::CanvasCaptureMediaStreamTrack::Source::settingsDidChange):
(WebCore::CanvasCaptureMediaStreamTrack::Source::canvasResized):
* Modules/mediastream/CanvasCaptureMediaStreamTrack.h:
* platform/mediastream/RealtimeIncomingVideoSource.cpp:
(WebCore::RealtimeIncomingVideoSource::RealtimeIncomingVideoSource):
(WebCore::RealtimeIncomingVideoSource::settings):
(WebCore::RealtimeIncomingVideoSource::settingsDidChange):
* platform/mediastream/RealtimeIncomingVideoSource.h:
* platform/mediastream/RealtimeMediaSource.cpp:
(WebCore::RealtimeMediaSource::settingsDidChange):
(WebCore::RealtimeMediaSource::notifySettingsDidChangeObservers):
(WebCore::RealtimeMediaSource::setSize):
(WebCore::RealtimeMediaSource::setFrameRate):
(WebCore::RealtimeMediaSource::setAspectRatio):
(WebCore::RealtimeMediaSource::setFacingMode):
(WebCore::RealtimeMediaSource::setVolume):
(WebCore::RealtimeMediaSource::setSampleRate):
(WebCore::RealtimeMediaSource::setSampleSize):
(WebCore::RealtimeMediaSource::setEchoCancellation):
* platform/mediastream/RealtimeMediaSource.h:
* platform/mediastream/gstreamer/GStreamerAudioCaptureSource.cpp:
(WebCore::GStreamerAudioCaptureSource::settingsDidChange):
* platform/mediastream/gstreamer/GStreamerVideoCaptureSource.cpp:
(WebCore::GStreamerVideoCaptureSource::settingsDidChange):
* platform/mediastream/mac/AVVideoCaptureSource.mm:
(WebCore::AVVideoCaptureSource::settingsDidChange):
* platform/mediastream/mac/CoreAudioCaptureSource.cpp:
(WebCore::CoreAudioCaptureSource::settingsDidChange):
* platform/mediastream/mac/DisplayCaptureSourceCocoa.cpp:
(WebCore::DisplayCaptureSourceCocoa::settingsDidChange):
* platform/mediastream/mac/RealtimeIncomingVideoSourceCocoa.mm:
(WebCore::RealtimeIncomingVideoSourceCocoa::processNewSample):
* platform/mock/MockRealtimeAudioSource.cpp:
(WebCore::MockRealtimeAudioSource::settingsDidChange):
* platform/mock/MockRealtimeVideoSource.cpp:
(WebCore::MockRealtimeVideoSource::settingsDidChange):

Source/WebKit:

* WebProcess/cocoa/UserMediaCaptureManager.cpp:
(WebKit::UserMediaCaptureManager::Source::setSettings):

LayoutTests:

* webrtc/video-disabled-black.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236646 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[MediaStream] Use display-specific capture factories
eric.carlson@apple.com [Sun, 30 Sep 2018 07:02:39 +0000 (07:02 +0000)]
[MediaStream] Use display-specific capture factories
https://bugs.webkit.org/show_bug.cgi?id=190043
<rdar://problem/44834412>

Reviewed by Youenn Fablet.

Source/WebCore:

No new tests, no change in functionality.

* Sources.txt:
* WebCore.xcodeproj/project.pbxproj:
* platform/mediastream/RealtimeMediaSource.cpp:
(WebCore::RealtimeMediaSource::AudioCaptureFactory::~AudioCaptureFactory): Deleted.
(WebCore::RealtimeMediaSource::VideoCaptureFactory::~VideoCaptureFactory): Deleted.
* platform/mediastream/RealtimeMediaSource.h:
* platform/mediastream/RealtimeMediaSourceCenter.cpp:
(WebCore::RealtimeMediaSourceCenter::createMediaStream):
(WebCore::RealtimeMediaSourceCenter::getDisplayMediaDevices):
* platform/mediastream/RealtimeMediaSourceCenter.h:
(WebCore::RealtimeMediaSourceCenter::setAudioFactory):
(WebCore::RealtimeMediaSourceCenter::unsetAudioFactory):
* platform/mediastream/RealtimeMediaSourceFactory.cpp: Added.
(WebCore::SingleSourceFactory::setActiveSource):
(WebCore::SingleSourceFactory::unsetActiveSource):
* platform/mediastream/RealtimeMediaSourceFactory.h: Added.
(WebCore::SingleSourceFactory::activeSource):
(WebCore::VideoCaptureFactory::setVideoCapturePageState):
(WebCore::DisplayCaptureFactory::setDisplayCapturePageState):
* platform/mediastream/gstreamer/GStreamerVideoCaptureSource.cpp:
(WebCore::libWebRTCVideoCaptureSourceFactory):
(WebCore::libWebRTCDisplayCaptureSourceFactory):
(WebCore::GStreamerVideoCaptureSource::factory):
(WebCore::GStreamerVideoCaptureSource::displayFactory):
* platform/mediastream/gstreamer/GStreamerVideoCaptureSource.h:
* platform/mediastream/gstreamer/RealtimeMediaSourceCenterLibWebRTC.cpp:
(WebCore::RealtimeMediaSourceCenterLibWebRTC::audioCaptureSourceFactory):
(WebCore::RealtimeMediaSourceCenterLibWebRTC::audioFactory):
(WebCore::RealtimeMediaSourceCenterLibWebRTC::videoFactory):
(WebCore::RealtimeMediaSourceCenterLibWebRTC::displayCaptureFactory):
* platform/mediastream/gstreamer/RealtimeMediaSourceCenterLibWebRTC.h:
* platform/mediastream/mac/AVVideoCaptureSource.mm:
(WebCore::AVVideoCaptureSource::~AVVideoCaptureSource):
(WebCore::AVVideoCaptureSource::setupCaptureSession):
* platform/mediastream/mac/CoreAudioCaptureSource.cpp:
(WebCore::CoreAudioCaptureSource::factory):
* platform/mediastream/mac/CoreAudioCaptureSource.h:
* platform/mediastream/mac/RealtimeMediaSourceCenterMac.cpp:
(WebCore::RealtimeMediaSourceCenterMac::videoCaptureSourceFactory):
(WebCore::RealtimeMediaSourceCenterMac::displayCaptureSourceFactory):
(WebCore::RealtimeMediaSourceCenterMac::audioCaptureSourceFactory):
(WebCore::RealtimeMediaSourceCenterMac::audioFactory):
(WebCore::RealtimeMediaSourceCenterMac::videoFactory):
(WebCore::RealtimeMediaSourceCenterMac::displayCaptureFactory):
* platform/mediastream/mac/RealtimeMediaSourceCenterMac.h:
* platform/mock/MockRealtimeAudioSource.cpp:
(WebCore::MockRealtimeAudioSource::~MockRealtimeAudioSource):
(WebCore::MockRealtimeAudioSource::startProducingData):
(): Deleted.
(WebCore::mockAudioCaptureSourceFactory): Deleted.
(WebCore::MockRealtimeAudioSource::factory): Deleted.
* platform/mock/MockRealtimeAudioSource.h:
* platform/mock/MockRealtimeMediaSourceCenter.cpp:
(WebCore::MockRealtimeVideoSourceFactory::setVideoCapturePageState):
(WebCore::MockRealtimeMediaSourceCenter::audioFactory):
(WebCore::MockRealtimeMediaSourceCenter::videoFactory):
(WebCore::MockRealtimeMediaSourceCenter::displayCaptureFactory):
* platform/mock/MockRealtimeMediaSourceCenter.h:
* platform/mock/MockRealtimeVideoSource.cpp:
(): Deleted.
(WebCore::MockRealtimeVideoSourceFactory::setVideoCapturePageState): Deleted.
(WebCore::mockVideoCaptureSourceFactory): Deleted.
(WebCore::MockRealtimeVideoSource::factory): Deleted.
* platform/mock/MockRealtimeVideoSource.h:

Source/WebKit:

* WebProcess/cocoa/UserMediaCaptureManager.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236645 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, rolling out r236631.
commit-queue@webkit.org [Sat, 29 Sep 2018 17:40:17 +0000 (17:40 +0000)]
Unreviewed, rolling out r236631.
https://bugs.webkit.org/show_bug.cgi?id=190112

Caused various eventSender tests, including fast/css/pseudo-
active-style-sharing*, to fail (Requested by smfr on #webkit).

Reverted changeset:

"Regression(r236512): http/tests/navigation/keyboard-events-
during-provisional-navigation.html is flaky"
https://bugs.webkit.org/show_bug.cgi?id=190052
https://trac.webkit.org/changeset/236631

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236644 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[css-grid] Properly align items next to collapsed tracks with gutters
commit-queue@webkit.org [Sat, 29 Sep 2018 12:11:17 +0000 (12:11 +0000)]
[css-grid] Properly align items next to collapsed tracks with gutters
https://bugs.webkit.org/show_bug.cgi?id=190089

Patch by Oriol Brufau <obrufau@igalia.com> on 2018-09-29
Reviewed by Manuel Rego Casasnovas.

LayoutTests/imported/w3c:

Import grid-gutters-013.html and reference from WPT.

Test: imported/w3c/web-platform-tests/css/css-grid/alignment/grid-gutters-013.html

* web-platform-tests/css/css-grid/alignment/grid-gutters-013-expected.xht: Added.
* web-platform-tests/css/css-grid/alignment/grid-gutters-013.html: Added.

Source/WebCore:

gridAreaPositionForInFlowChild could return a wrong end position for
grid items adjacent to a collapsed track, because it didn't take into
account that gutters collapse in that case. Therefore, "center" or
"end" alignments displayed the item at the wrong position.

Test: imported/w3c/web-platform-tests/css/css-grid/alignment/grid-gutters-013.html

* rendering/RenderGrid.cpp:
(WebCore::RenderGrid::gridAreaPositionForInFlowChild const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236643 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoAdd yasm as third party tool for libwebrtc compilation
youenn@apple.com [Sat, 29 Sep 2018 11:52:20 +0000 (11:52 +0000)]
Add yasm as third party tool for libwebrtc compilation
https://bugs.webkit.org/show_bug.cgi?id=190025

Reviewed by Eric Carlson.

Add yasm source code and build the yasm executable as it is needed for libvpx compilati

* Source/third_party/yasm-1.3.0: Added.
* libwebrtc.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236642 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[PSON] Deal with the drawing area potentially being null in WebFrameLoaderClient...
cdumez@apple.com [Sat, 29 Sep 2018 09:49:31 +0000 (09:49 +0000)]
[PSON] Deal with the drawing area potentially being null in WebFrameLoaderClient::transitionToCommittedForNewPage()
https://bugs.webkit.org/show_bug.cgi?id=190084
<rdar://problem/44844561>

Reviewed by Antti Koivisto.

Deal with the drawing area potentially being null in WebFrameLoaderClient::transitionToCommittedForNewPage(), as a
result of calling WebPage::tearDownDrawingAreaForSuspend().

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::transitionToCommittedForNewPage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236641 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[GStreamer][MSE] Use GObject for GST_TRACE_OBJECT
aboya@igalia.com [Sat, 29 Sep 2018 09:43:00 +0000 (09:43 +0000)]
[GStreamer][MSE] Use GObject for GST_TRACE_OBJECT
https://bugs.webkit.org/show_bug.cgi?id=190045

Reviewed by Philippe Normand.

Passing a non-GObject object to GST_TRACE_OBJECT() can be
theoretically misunderstood by the GStreamer logging function, so this
patch avoids that.

* platform/graphics/gstreamer/mse/AppendPipeline.cpp:
(WebCore::AppendPipeline::appsrcEndOfAppendCheckerProbe):
(WebCore::AppendPipeline::handleEndOfAppend):
(WebCore::AppendPipeline::consumeAppsinkAvailableSamples):
(WebCore::AppendPipeline::pushNewBuffer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236640 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WHLSL] Fix build after r236635
mmaxfield@apple.com [Sat, 29 Sep 2018 06:55:01 +0000 (06:55 +0000)]
[WHLSL] Fix build after r236635
https://bugs.webkit.org/show_bug.cgi?id=189210

Unreviewed.

* WebGPUShadingLanguageRI/Intrinsics.js:
(Intrinsics.):
* WebGPUShadingLanguageRI/Test.js:
(tests.atomicsNull):
(tests.numThreads):
(tests.textureDimensionsNull):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236639 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[ESNext][BigInt] Implement support for "&"
ticaiolima@gmail.com [Sat, 29 Sep 2018 04:55:40 +0000 (04:55 +0000)]
[ESNext][BigInt] Implement support for "&"
https://bugs.webkit.org/show_bug.cgi?id=186228

Reviewed by Yusuke Suzuki.

JSTests:

* stress/big-int-bitwise-and-general.js: Added.
(assert):
(assert.sameValue):
* stress/big-int-bitwise-and-jit.js: Added.
(let.assert.sameValue):
(bigIntBitAnd):
* stress/big-int-bitwise-and-memory-stress.js: Added.
(assert):
* stress/big-int-bitwise-and-to-primitive-precedence.js: Added.
(assert.sameValue):
(let.o.Symbol.toPrimitive):
(catch):
* stress/big-int-bitwise-and-type-error.js: Added.
(assert):
(assertThrowTypeError):
(let.o.valueOf):
(o.valueOf):
(o.toString):
(o.Symbol.toPrimitive):
* stress/big-int-bitwise-and-wrapped-value.js: Added.
(assert.sameValue):
(testBitAnd):
(let.o.Symbol.toPrimitive):
(o.valueOf):
(o.toString):

Source/JavaScriptCore:

This patch introduces support of BigInt into bitwise "&" operation.
We are also introducing the ValueBitAnd DFG node, that is responsible
to take care of JIT for non-Int32 operands. With the introduction of this
new node, we renamed the BitAnd node to ArithBitAnd. The ArithBitAnd
follows the behavior of ArithAdd and other arithmetic nodes, where
the Arith<op> version always results in Number (in the case of
ArithBitAnd, its is always an Int32).

* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::finishCreation):
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitBinaryOp):
* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):
* dfg/DFGBackwardsPropagationPhase.cpp:
(JSC::DFG::BackwardsPropagationPhase::isWithinPowerOfTwo):
(JSC::DFG::BackwardsPropagationPhase::propagate):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* dfg/DFGDoesGC.cpp:
(JSC::DFG::doesGC):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
* dfg/DFGNodeType.h:
* dfg/DFGOperations.cpp:
* dfg/DFGOperations.h:
* dfg/DFGPredictionPropagationPhase.cpp:
* dfg/DFGSafeToExecute.h:
(JSC::DFG::safeToExecute):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileValueBitwiseOp):
(JSC::DFG::SpeculativeJIT::compileBitwiseOp):
* dfg/DFGSpeculativeJIT.h:
(JSC::DFG::SpeculativeJIT::bitOp):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGStrengthReductionPhase.cpp:
(JSC::DFG::StrengthReductionPhase::handleNode):
* ftl/FTLCapabilities.cpp:
(JSC::FTL::canCompile):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileNode):
(JSC::FTL::DFG::LowerDFGToB3::compileValueBitAnd):
(JSC::FTL::DFG::LowerDFGToB3::compileArithBitAnd):
(JSC::FTL::DFG::LowerDFGToB3::compileBitAnd): Deleted.
* jit/JIT.h:
* jit/JITArithmetic.cpp:
(JSC::JIT::emitBitBinaryOpFastPath):
(JSC::JIT::emit_op_bitand):
* llint/LowLevelInterpreter32_64.asm:
* llint/LowLevelInterpreter64.asm:
* runtime/CommonSlowPaths.cpp:
(JSC::SLOW_PATH_DECL):
* runtime/JSBigInt.cpp:
(JSC::JSBigInt::JSBigInt):
(JSC::JSBigInt::initialize):
(JSC::JSBigInt::createZero):
(JSC::JSBigInt::createFrom):
(JSC::JSBigInt::bitwiseAnd):
(JSC::JSBigInt::absoluteBitwiseOp):
(JSC::JSBigInt::absoluteAnd):
(JSC::JSBigInt::absoluteOr):
(JSC::JSBigInt::absoluteAndNot):
(JSC::JSBigInt::absoluteAddOne):
(JSC::JSBigInt::absoluteSubOne):
* runtime/JSBigInt.h:
* runtime/JSCJSValue.h:
* runtime/JSCJSValueInlines.h:
(JSC::JSValue::toBigIntOrInt32 const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236637 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoAngled gradient backgrounds in body render vertically when body height is 0
commit-queue@webkit.org [Sat, 29 Sep 2018 03:29:04 +0000 (03:29 +0000)]
Angled gradient backgrounds in body render vertically when body height is 0
https://bugs.webkit.org/show_bug.cgi?id=177232
<rdar://problem/34548230>.

Patch by Zamiul Haque <zhaque@apple.com> on 2018-09-28
Reviewed by Tim Horton.

Source/WebCore:

Specifically, gradients displayed at an angle (ie. 45 degrees) are rendered
as if they are vertical when the body tag containing the gradient
has a height of 0. Other browsers do not render under these circumstances,
so WebKit was modified to follow in suit. The problem was due to layout sizes for
fill tiles being calculated with a minimum height of 1px. A simple change of the
minimum height and width to 0px was enough to bring about the desired behavior.

Tests: angled-background-repeating-gradient-rendering-vertical.html

* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::calculateFillTileSize const):

LayoutTests:

Added tests to make sure that angled background gradients inside of a
body tag of height 0 render as empty white screens.

* angled-background-repeating-gradient-rendering-vertical.html: Added.
* angled-background-repeating-gradient-rendering-vertical-expected.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236636 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WHLSL] Allow uniform buffers to be used in the interpreter
mmaxfield@apple.com [Sat, 29 Sep 2018 02:27:54 +0000 (02:27 +0000)]
[WHLSL] Allow uniform buffers to be used in the interpreter
https://bugs.webkit.org/show_bug.cgi?id=189210

Reviewed by Filip Pizlo.

Every lvalue is associated with an address space, and the propagation of these address spaces
matches the propagation of lvalues. Luckily, there was already the infrastructure to do most
of this, so this patch just goes the last few yards. It also updates the standard library to
allow for writing into all writable address spaces for out-params.

* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitAssignment):
(Checker.prototype._finishVisitingPropertyAccess):
* WebGPUShadingLanguageRI/Intrinsics.js:
(Intrinsics.):
(Intrinsics.checkFalse):
(Intrinsics):
* WebGPUShadingLanguageRI/StandardLibrary.js:
(let.standardLibrary):
* WebGPUShadingLanguageRI/Test.js:
(tests.threadArrayRefLoad):
(tests.threadArrayRefLoadIntLiteral):
(tests.deviceArrayRefLoad):
(tests.threadArrayRefStore):
(tests.deviceArrayRefStore):
(tests.deviceArrayRefStoreIntLiteral):
(tests.threadPointerLoad):
(tests.threadPointerStore):
(tests.devicePointerLoad):
(tests.devicePointerStore):
(tests.arrayLoad):
(tests.constantAddressSpace):
(tests.standardLibraryDevicePointers):
(tests.threadArrayLoad): Deleted.
(tests.threadArrayLoadIntLiteral): Deleted.
(tests.deviceArrayLoad): Deleted.
(tests.threadArrayStore): Deleted.
(tests.deviceArrayStore): Deleted.
(tests.deviceArrayStoreIntLiteral): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236635 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoNo DOM API to instantiate an attachment for an img element
wenson_hsieh@apple.com [Sat, 29 Sep 2018 02:12:57 +0000 (02:12 +0000)]
No DOM API to instantiate an attachment for an img element
https://bugs.webkit.org/show_bug.cgi?id=189934
<rdar://problem/44743222>

Reviewed by Ryosuke Niwa.

Source/WebCore:

Adds support for HTMLAttachmentElement.getAttachmentIdentifier, a function that internal WebKit clients can use
to ensure that an image element is backed by a unique _WKAttachment. See below for more details.

Tests:  WKAttachmentTests.AddAttachmentToConnectedImageElement
        WKAttachmentTests.ChangeFileWrapperForPastedImage
        WKAttachmentTests.ConnectImageWithAttachmentToDocument

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

Add a new hook to register an empty _WKAttachment in the UI process with a given identifier. Used when creating
a new empty attachment to back an image element.

* dom/Document.h:
* editing/Editor.cpp:
(WebCore::Editor::registerAttachmentIdentifier):
(WebCore::Editor::notifyClientOfAttachmentUpdates):
* editing/Editor.h:
* html/HTMLAttachmentElement.cpp:
(WebCore::HTMLAttachmentElement::getAttachmentIdentifier):

Creates an attachment element to back the image element, if an attachment does not already exist, and returns
the unique identifier. This also causes an empty corresponding _WKAttachment to be created in the client, whose
file wrapper determines the contents of the image.

(WebCore::HTMLAttachmentElement::ensureUniqueIdentifier):
(WebCore::HTMLAttachmentElement::hasEnclosingImage const):
(WebCore::HTMLAttachmentElement::updateEnclosingImageWithData):

Add a helper that updates the source of the enclosing image element given a content type and image data, by
creating a new blob and blob URL.

* html/HTMLAttachmentElement.h:
* html/HTMLAttachmentElement.idl:
* html/HTMLImageElement.idl:

Rename webkitAttachmentIdentifier to just attachmentIdentifier.

* page/EditorClient.h:
(WebCore::EditorClient::registerAttachmentIdentifier):
(WebCore::EditorClient::didInsertAttachmentWithIdentifier):

Source/WebKit:

Makes some adjustments to support using _WKAttachment's file wrapper to change the contents of any image element
hosting the attachment in its shadow root. To do this, we add some plumbing to allow the UI process to update an
attachment element's enclosing image with data from its file wrapper.

* UIProcess/API/APIAttachment.cpp:
(API::Attachment::isEmpty const):
(API::Attachment::enclosingImageData const):

Helper method that creates a SharedBuffer representing image data for the attachment. Only returns a non-null
value for attachment elements that are enclosed within an image.

* UIProcess/API/APIAttachment.h:
* UIProcess/API/Cocoa/APIAttachmentCocoa.mm:
(API::Attachment::enclosingImageData const):
(API::Attachment::isEmpty const):
* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _attachmentForIdentifier:]):

Add new SPI to request a _WKAttachment for a given unique identifier. Currently, this is only used for testing.

* UIProcess/API/Cocoa/WKWebViewPrivate.h:
* UIProcess/Cocoa/WebPageProxyCocoa.mm:
(WebKit::WebPageProxy::platformRegisterAttachment):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::updateAttachmentAttributes):
(WebKit::WebPageProxy::registerAttachmentIdentifier):
(WebKit::WebPageProxy::didInsertAttachmentWithIdentifier):

Plumb whether or not the attachment is enclosed by an image from the web process to the UI process.

(WebKit::WebPageProxy::didRemoveAttachmentWithIdentifier):
(WebKit::WebPageProxy::didInsertAttachment): Deleted.
(WebKit::WebPageProxy::didRemoveAttachment):
* UIProcess/WebPageProxy.h:
* UIProcess/WebPageProxy.messages.in:
* WebProcess/WebCoreSupport/WebEditorClient.cpp:
(WebKit::WebEditorClient::registerAttachmentIdentifier):
(WebKit::WebEditorClient::didInsertAttachmentWithIdentifier):

Update attachment attributes after inserting an attachment. This ensures that an attachment that was created and
later inserted via script into the document will be synced with state in the UI process, if the client has
changed the contents of the attachment.

* WebProcess/WebCoreSupport/WebEditorClient.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updateAttachmentAttributes):

Plumb attachment data from the UI process to the web process.

* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/WebPage.messages.in:

Tools:

Adds 3 new API tests in WKAttachmentTests:

`AddAttachmentToConnectedImageElement` verifies that an image element that's already in the document can gain an
attachment element via `HTMLAttachmentElement.getAttachmentIdentifier`.

`ChangeFileWrapperForPastedImage` verifies that an image that has been pasted produces a _WKAttachment in the UI
process, and changing the file wrapper of that _WKAttachment changes the pasted image.

`ConnectImageWithAttachmentToDocument` verifies that script can create an image element, ensure that it has an
attachment, and set a file wrapper for the generated _WKAttachment. Connecting the image to the document should
then result in an image element with the contents of the _WKAttachment's file wrapper.

* TestWebKitAPI/Tests/WebKitCocoa/WKAttachmentTests.mm:
(-[TestWKWebView imageElementSize]):
(-[TestWKWebView waitForImageElementSizeToBecome:]):
(TestWebKitAPI::TEST):
(-[TestWKWebView waitForAttachmentElementSizeToBecome:]): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236634 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoThe return value of an OnBeforeUnloadEventHandler should always be coerced into a...
cdumez@apple.com [Sat, 29 Sep 2018 02:03:51 +0000 (02:03 +0000)]
The return value of an OnBeforeUnloadEventHandler should always be coerced into a DOMString
https://bugs.webkit.org/show_bug.cgi?id=190090

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

Rebaseline WPT test now that one more check is passing.

* web-platform-tests/html/browsers/browsing-the-web/unloading-documents/beforeunload-canceling-expected.txt:

Source/WebCore:

The return value of an OnBeforeUnloadEventHandler should always be coerced into a DOMString:
- https://html.spec.whatwg.org/#onbeforeunloadeventhandler
- https://html.spec.whatwg.org/#the-event-handler-processing-algorithm (Step 5)

In particular, this means that returning false in an OnBeforeUnloadEventHandler should NOT
cancel the event when the event is a CustomEvent (and not a BeforeUnloadEvent). This is
because the return value cannot be false at:
- https://html.spec.whatwg.org/#the-event-handler-processing-algorithm (Step 5. Otherwise case).

No new tests, rebaselined existing test.

* bindings/js/JSEventListener.cpp:
(WebCore::JSEventListener::handleEvent):

LayoutTests:

Update test that was returning a value in a beforeunload event listener instead of using an
event handler. The test needs to use an event handler (window.onbeforeunload) as an event
listener does not have a return value. I have verified that our behavior is consistent with
Chrome and Firefox on this test, both with an event listener and an event handler.

* fast/loader/form-submission-after-beforeunload-cancel.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236633 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRenderLayer::removeOnlyThisLayer() should not call updateLayerPositions()
simon.fraser@apple.com [Sat, 29 Sep 2018 01:53:34 +0000 (01:53 +0000)]
RenderLayer::removeOnlyThisLayer() should not call updateLayerPositions()
https://bugs.webkit.org/show_bug.cgi?id=190093

Reviewed by Dean Jackson and Zalan Bujtas.

Source/WebCore:

It's wrong for RenderLayer::removeOnlyThisLayer() to call updateLayerPositions(),
because this is called at style update time, and layout will be stale.

It was added (see webkit.org/b/25252) so that opacity changes, which can destroy layers, correctly update
descendants. However, RenderStyle::changeRequiresLayout() checks for opacity <=> no opacity
changes and triggers layout accordingly, which will result in a full post-layout
updateLayerPositions().

This also revealed that changes to the "isolate" property fail to trigger any kind of style recalc or layout;
we need it to trigger layout (for now) because it affects z-order.

Covered by existing tests.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::removeOnlyThisLayer):
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::changeRequiresLayout const):

LayoutTests:

The ordering of the repaints changes.

blend-mode-turn-off-isolation-no-effect.html now issues a repaint, which is expected now.
Isolation is so rare that this is not a problem.

* css3/blending/repaint/blend-mode-isolate-stacking-context-expected.txt:
* css3/blending/repaint/blend-mode-turn-off-isolation-expected.txt:
* css3/blending/repaint/blend-mode-turn-off-isolation-no-effect-expected.txt:
* fast/repaint/absolute-position-change-containing-block-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236632 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRegression(r236512): http/tests/navigation/keyboard-events-during-provisional-navigat...
cdumez@apple.com [Sat, 29 Sep 2018 01:38:26 +0000 (01:38 +0000)]
Regression(r236512): http/tests/navigation/keyboard-events-during-provisional-navigation.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=190052

Reviewed by Ryosuke Niwa.

Source/WebKit:

Introduce a new IPC::SendOption indicating that the IPC should always be sent asynchronously,
even if the connection is in fully synchronous mode. This is used by WebKitTestRunner for
all text outputting (e.g. console.log) is asynchronous, and thus in order.

* Platform/IPC/Connection.cpp:
(IPC::Connection::sendMessage):
* Platform/IPC/Connection.h:
* WebProcess/InjectedBundle/API/c/WKBundlePage.cpp:
(WKBundlePagePostMessageIgnoringFullySynchronousMode):
* WebProcess/InjectedBundle/API/c/WKBundlePage.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::postMessageIgnoringFullySynchronousMode):
* WebProcess/WebPage/WebPage.h:

Tools:

The test relies on EventSender to send events to the page synchronously to the page and then uses console.log
to log those events. It also uses console.log() before sending those events to indicate what the test is about
to do. Note that console.log() normally causes the WebKitTestRunner to send an asynchronous IPC to the UIProcess
so that it can log the message.
The issue is that EventSender uses IPC::SendOption::UseFullySynchronousModeForTesting when sending the
sync IPC to the UIProcess. This option causes follow-up *asynchronous* IPC sent from the synchronous IPC reply
handler to be transformed into synchronous IPC.
As a result, some of the console.log IPC ended up being asynchronous and some other ended up being synchronous.
Because synchronous and asynchronous IPC is not necessarily processed in-order by the UIProcess, the logged
messages may end up being out of order, leading to flakiness.

To address the issue, we now make sure that InjectedBundle::outputText() uses a new IPC::SendOption indicating
that the IPC should always be sent asynchronously, even if the connection is in fully synchronous mode. As a
result, all text outputing IPC to the UIProcess will be asynchronous, and thus in order.

* WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
(WTR::InjectedBundle::outputText):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236631 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoSkip debug assertion in ResourceLoadStatisticsMemoryStore::recursivelyGetAllDomainsTh...
wilander@apple.com [Sat, 29 Sep 2018 01:36:27 +0000 (01:36 +0000)]
Skip debug assertion in ResourceLoadStatisticsMemoryStore::recursivelyGetAllDomainsThatHaveRedirectedToThisDomain()
https://bugs.webkit.org/show_bug.cgi?id=190097
<rdar://problem/44877080>

Reviewed by Chris Dumez.

The change in https://bugs.webkit.org/show_bug.cgi?id=190055 causes a
re-classification that may trigger a debug assertion in
ResourceLoadStatisticsMemoryStore::recursivelyGetAllDomainsThatHaveRedirectedToThisDomain().
We should skip the assertion for statistics model version 14 to avoid
this.

* UIProcess/ResourceLoadStatisticsMemoryStore.cpp:
(WebKit::ResourceLoadStatisticsMemoryStore::recursivelyGetAllDomainsThatHaveRedirectedToThisDomain const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236630 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agocmake cannot run if python 2 isn't available
commit-queue@webkit.org [Sat, 29 Sep 2018 01:00:37 +0000 (01:00 +0000)]
cmake cannot run if python 2 isn't available
https://bugs.webkit.org/show_bug.cgi?id=190075

Patch by Mike Gorse <mgorse@alum.wpi.edu> on 2018-09-28
Reviewed by Michael Catanzaro.

* Source/cmake/WebKitCommon.cmake: Set Python_ADDITIONAL_VERSIONS
to 3, so that a python3 binary will be found.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236629 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WHLSL] Pointers should have automatically-generated equality checks
mmaxfield@apple.com [Sat, 29 Sep 2018 00:48:34 +0000 (00:48 +0000)]
[WHLSL] Pointers should have automatically-generated equality checks
https://bugs.webkit.org/show_bug.cgi?id=189986

Reviewed by Filip Pizlo.

C allows for pointer equality, and we need it so people can do null checks.
This is generated the same way all our other NativeFuncs are generated - by Checker
creating NativeFuncs inside CallExpression.resolve().

This patch also does some general cleanup.

* WebGPUShadingLanguageRI/All.js: Everything the late checker does is no longer necessary.
The last thing it was doing was making sure that only primitive types are in resources, but
it's totally reasonable to put structs and arrays in resources, so I removed this pass. We
still have to add a check to make sure resources can't live within resources, but I expect
that will be done in the same place that semantics are checked.
* WebGPUShadingLanguageRI/AllocateAtEntryPoints.js:
(allocateAtEntryPoints.updateFunction.UpdateFunctions.prototype._addVariableDeclaration):
Name the global struct for debugging purposes.
* WebGPUShadingLanguageRI/CallExpression.js:
(CallExpression.prototype._resolveByInstantiation): Cleanup
(CallExpression.prototype._resolveWithOperatorAnderIndexer): Ditto
(CallExpression.prototype._resolveWithOperatorLength): Ditto
(CallExpression.prototype._resolveWithReferenceComparator): Add support to automatically
generate pointer equality NativeFuncs.
* WebGPUShadingLanguageRI/EPtr.js: Implement pointer equality in the interpreter.
* WebGPUShadingLanguageRI/LateChecker.js: Removed.
* WebGPUShadingLanguageRI/LayoutBuffers.js: Renamed from Tools/WebGPUShadingLanguageRI/LateCheckAndLayoutBuffers.js.
(layoutBuffers):
* WebGPUShadingLanguageRI/Metal/WHLSL Tests/WHLSL Tests.xcodeproj/project.pbxproj:
* WebGPUShadingLanguageRI/NativeFunc.js:
* WebGPUShadingLanguageRI/OperatorArrayRefLength.js:
(OperatorArrayRefLength.prototype.instantiateImplementation):
(OperatorArrayRefLength):
* WebGPUShadingLanguageRI/Prepare.js:
(let.prepare):
* WebGPUShadingLanguageRI/SPIRV.html:
* WebGPUShadingLanguageRI/Test.html:
* WebGPUShadingLanguageRI/Test.js:
(tests.ternaryExpression):
(tests.break): Speed up testing time
(tests.doWhile): Ditto
(tests.forLoop): Ditto
(tests.atomics): Ditto
(tests.atomicsNull): Ditto
(tests.pointerEquality): Test pointer equality
(tests.standardLibraryDevicePointers):
(tests.devicePtrPtr): Deleted.
(tests.threadgroupPtrPtr): Deleted.
(tests.constantPtrPtr): Deleted.
* WebGPUShadingLanguageRI/index.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236628 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoWeb Inspector: crash in InspectorNetworkAgent::didReceiveResponse when loading denied...
drousso@apple.com [Sat, 29 Sep 2018 00:22:08 +0000 (00:22 +0000)]
Web Inspector: crash in InspectorNetworkAgent::didReceiveResponse when loading denied x-frame resources
https://bugs.webkit.org/show_bug.cgi?id=190046

Reviewed by Joseph Pecoraro.

Source/WebKit:

* NetworkProcess/NetworkResourceLoader.cpp:
(WebKit::NetworkResourceLoader::didReceiveResponse):
(WebKit::NetworkResourceLoader::didRetrieveCacheEntry):
Send the sanitized `ResourceResponse` with the message so WebInspector is able to access it.

* WebProcess/Network/WebResourceLoader.h:
* WebProcess/Network/WebResourceLoader.cpp:
* WebProcess/Network/WebResourceLoader.messages.in:
(WebKit::WebResourceLoader::stopLoadingAfterXFrameOptionsOrContentSecurityPolicyDenied):

LayoutTests:

* http/tests/inspector/network/resources/x-frame-options.php: Added.
* http/tests/inspector/network/x-frame-options-expected.txt: Added.
* http/tests/inspector/network/x-frame-options.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236627 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, rolling out r236620.
realdawei@apple.com [Sat, 29 Sep 2018 00:15:57 +0000 (00:15 +0000)]
Unreviewed, rolling out r236620.

broke internal Mac and iOS builds

Reverted changeset:

"Add yasm as third party tool for libwebrtc compilation"
https://bugs.webkit.org/show_bug.cgi?id=190025
https://trac.webkit.org/changeset/236620

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236626 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WebAuthN] Polish WebAuthN auto-test environment
jiewen_tan@apple.com [Sat, 29 Sep 2018 00:02:39 +0000 (00:02 +0000)]
[WebAuthN] Polish WebAuthN auto-test environment
https://bugs.webkit.org/show_bug.cgi?id=189283
<rdar://problem/44117828>

Reviewed by Chris Dumez.

Source/WebCore:

This patch removes the old mocking mechanism.

Tests: http/wpt/webauthn/public-key-credential-create-with-invalid-parameters.https.html
       http/wpt/webauthn/public-key-credential-get-with-invalid-parameters.https.html
       http/wpt/webauthn/public-key-credential-same-origin-with-ancestors.https.html

* DerivedSources.make:
* WebCore.xcodeproj/project.pbxproj:
* testing/Internals.cpp:
(WebCore::Internals::Internals):
(WebCore::Internals::mockAuthenticatorCoordinator const): Deleted.
* testing/Internals.h:
* testing/Internals.idl:
* testing/MockAuthenticatorCoordinator.cpp: Removed.
* testing/MockAuthenticatorCoordinator.h: Removed.
* testing/MockAuthenticatorCoordinator.idl: Removed.

Source/WebKit:

This patch changes MockWebAuthenticationConfiguration.local to optional such that tests can express
absence of local authenticators.

* UIProcess/API/C/WKWebsiteDataStoreRef.cpp:
(WKWebsiteDataStoreSetWebAuthenticationMockConfiguration):
* UIProcess/WebAuthentication/Mock/MockLocalConnection.mm:
(WebKit::MockLocalConnection::getUserConsent const):
(WebKit::MockLocalConnection::getAttestation const):
* UIProcess/WebAuthentication/Mock/MockLocalService.cpp:
(WebKit::MockLocalService::platformStartDiscovery const):
* UIProcess/WebAuthentication/Mock/MockWebAuthenticationConfiguration.h:

Tools:

* WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::setWebAuthenticationMockConfiguration):

LayoutTests:

This patch does the following few things:
1. Update tests that employ old mocking mechnism to the new one.
2. Move tests from http/tests/webauthn to http/wpt/webauthn. As a result, we have one unified place for
all WebAuthN tests: http/wpt/webauthn.
3. Add a helper function to check DOMException message as well.

* TestExpectations:
* http/tests/webauthn/public-key-credential-create-with-invalid-parameters.https-expected.txt: Removed.
* http/tests/webauthn/public-key-credential-get-with-invalid-parameters.https-expected.txt: Removed.
* http/tests/webauthn/public-key-credential-same-origin-with-ancestors-2.https-expected.txt: Removed.
* http/tests/webauthn/public-key-credential-same-origin-with-ancestors-2.https.html: Removed.
* http/tests/webauthn/public-key-credential-same-origin-with-ancestors.https-expected.txt: Removed.
* http/tests/webauthn/public-key-credential-same-origin-with-ancestors.https.html: Removed.
* http/tests/webauthn/resources/last-layer-frame.https.html: Removed.
* http/tests/webauthn/resources/util.js: Removed.
* http/wpt/credential-management/credentialscontainer-store-basics.https.html:
* http/wpt/webauthn/idl.https.html:
* http/wpt/webauthn/public-key-credential-create-failure-local.https.html:
* 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.https-expected.txt: Removed.
* http/wpt/webauthn/public-key-credential-create-success.https.html: Removed.
* http/wpt/webauthn/public-key-credential-create-with-invalid-parameters.https-expected.txt: Added.
* http/wpt/webauthn/public-key-credential-create-with-invalid-parameters.https.html: Renamed from LayoutTests/http/tests/webauthn/public-key-credential-create-with-invalid-parameters.https.html.
* http/wpt/webauthn/public-key-credential-get-failure-local.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.https-expected.txt: Removed.
* http/wpt/webauthn/public-key-credential-get-success.https.html: Removed.
* http/wpt/webauthn/public-key-credential-get-with-invalid-parameters.https-expected.txt: Added.
* http/wpt/webauthn/public-key-credential-get-with-invalid-parameters.https.html: Renamed from LayoutTests/http/tests/webauthn/public-key-credential-get-with-invalid-parameters.https.html.
* http/wpt/webauthn/public-key-credential-same-origin-with-ancestors.https-expected.txt: Added.
* http/wpt/webauthn/public-key-credential-same-origin-with-ancestors.https.html: Added.
* http/wpt/webauthn/resources/last-layer-frame.https.html: Added.
* http/wpt/webauthn/resources/second-layer-frame.https.html: Renamed from LayoutTests/http/tests/webauthn/resources/second-layer-frame.https.html.
* http/wpt/webauthn/resources/util.js:
* platform/gtk/TestExpectations:
* platform/ios-wk1/TestExpectations:
* platform/mac-wk1/TestExpectations:
* platform/mac-wk2/TestExpectations:
* platform/win/TestExpectations:
* platform/wincairo/TestExpectations:
* platform/wpe/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236625 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRefactoring: eliminate raw pointer usage in Fullscreen code
jer.noble@apple.com [Fri, 28 Sep 2018 23:31:29 +0000 (23:31 +0000)]
Refactoring: eliminate raw pointer usage in Fullscreen code
https://bugs.webkit.org/show_bug.cgi?id=188747
<rdar://problem/43541164>

Reviewed by Alex Christensen.

Source/WebCore:

Two sources of raw pointers in the Fullscreen code:
- Model classes (PlaybackSessionModel and VideoFullscreenModel) aren't ref-able, so
  they are passed around as raw references.
- Observer classes (PlaybackSessionModelClient and VideoFullscreenModelClient, and
  VideoFullscreenChangeObserver) are also passed around as raw pointers, but shouldn't
  be ref-able.

Make Model classes ref-able by adding ref() and deref() which call virtual refModel and
derefModel methods, overridden by implementing subclasses. Make every concrete observer
inherit from CanMakeWeakPtr, and every registration method take WeakPtr wrappers around
the client interface.

Since every Interface class now holds a strong reference to its Model classes, and each
Model class holds a weak reference to all its clients, no explicit invalidate() method
is necessary.

Notes:

- Since the weak pointer methods need to be able to downcast to the abstract base class,
  observers need to inherit publically (rather than privately) from those base classes.
- Media element Models should compose EventListener rather than inheriting from it, since
  EventListener has its own RefCount.
- WeakPtrs can't be held in HashSets (because they change value, and therefore hash, when
  their underlying object is destroyed), so clients should be stored in a Vector instead.
- Interfaces should be given all required Refs at creation time, so that they can store
  those parameters as Refs instead of RefPtrs.

* platform/cocoa/PlaybackSessionInterface.h:
(WebCore::PlaybackSessionInterface::~PlaybackSessionInterface): Deleted.
* platform/cocoa/PlaybackSessionModel.h:
(WebCore::PlaybackSessionModel::ref):
(WebCore::PlaybackSessionModel::deref):
(WebCore::PlaybackSessionModel::~PlaybackSessionModel): Deleted.
* platform/cocoa/PlaybackSessionModelMediaElement.h:
* platform/cocoa/PlaybackSessionModelMediaElement.mm:
(WebCore::PlaybackSessionModelMediaElement::PlaybackSessionModelMediaElement):
(WebCore::PlaybackSessionModelMediaElement::~PlaybackSessionModelMediaElement):
(WebCore::PlaybackSessionModelMediaElement::setMediaElement):
(WebCore::PlaybackSessionModelMediaElement::updateForEventName):
(WebCore::PlaybackSessionModelMediaElement::addClient):
(WebCore::PlaybackSessionModelMediaElement::removeClient):
(WebCore::PlaybackSessionModelMediaElement::updateMediaSelectionOptions):
(WebCore::PlaybackSessionModelMediaElement::updateMediaSelectionIndices):
(WebCore::PlaybackSessionModelMediaElement::handleEvent): Deleted.
* platform/cocoa/VideoFullscreenChangeObserver.h:
(WebCore::VideoFullscreenChangeObserver::~VideoFullscreenChangeObserver): Deleted.
* platform/cocoa/VideoFullscreenModel.h:
(WebCore::VideoFullscreenModel::ref):
(WebCore::VideoFullscreenModel::deref):
(WebCore::VideoFullscreenModel::~VideoFullscreenModel): Deleted.
* platform/cocoa/VideoFullscreenModelVideoElement.h:
* platform/cocoa/VideoFullscreenModelVideoElement.mm:
(VideoFullscreenModelVideoElement::VideoFullscreenModelVideoElement):
(VideoFullscreenModelVideoElement::setVideoElement):
(VideoFullscreenModelVideoElement::addClient):
(VideoFullscreenModelVideoElement::removeClient):
(VideoFullscreenModelVideoElement::setHasVideo):
(VideoFullscreenModelVideoElement::setVideoDimensions):
(VideoFullscreenModelVideoElement::willEnterPictureInPicture):
(VideoFullscreenModelVideoElement::didEnterPictureInPicture):
(VideoFullscreenModelVideoElement::failedToEnterPictureInPicture):
(VideoFullscreenModelVideoElement::willExitPictureInPicture):
(VideoFullscreenModelVideoElement::didExitPictureInPicture):
(VideoFullscreenModelVideoElement::handleEvent): Deleted.
* platform/ios/PlaybackSessionInterfaceAVKit.h:
(WebCore::PlaybackSessionInterfaceAVKit::create):
(WebCore::PlaybackSessionInterfaceAVKit::playbackSessionModel const):
(): Deleted.
* platform/ios/PlaybackSessionInterfaceAVKit.mm:
(WebCore::PlaybackSessionInterfaceAVKit::PlaybackSessionInterfaceAVKit):
(WebCore::PlaybackSessionInterfaceAVKit::~PlaybackSessionInterfaceAVKit):
(WebCore::PlaybackSessionInterfaceAVKit::invalidate): Deleted.
* platform/ios/VideoFullscreenInterfaceAVKit.h:
* platform/ios/VideoFullscreenInterfaceAVKit.mm:
(-[WebAVPlayerLayer layoutSublayers]):
(-[WebAVPlayerLayer resolveBounds]):
(-[WebAVPlayerLayer setVideoGravity:]):
(VideoFullscreenInterfaceAVKit::create):
(VideoFullscreenInterfaceAVKit::VideoFullscreenInterfaceAVKit):
(VideoFullscreenInterfaceAVKit::~VideoFullscreenInterfaceAVKit):
(VideoFullscreenInterfaceAVKit::setVideoFullscreenChangeObserver):
(VideoFullscreenInterfaceAVKit::applicationDidBecomeActive):
(VideoFullscreenInterfaceAVKit::setupFullscreen):
(VideoFullscreenInterfaceAVKit::presentingViewController):
(VideoFullscreenInterfaceAVKit::requestHideAndExitFullscreen):
(VideoFullscreenInterfaceAVKit::preparedToExitFullscreen):
(VideoFullscreenInterfaceAVKit::willStartPictureInPicture):
(VideoFullscreenInterfaceAVKit::didStartPictureInPicture):
(VideoFullscreenInterfaceAVKit::failedToStartPictureInPicture):
(VideoFullscreenInterfaceAVKit::willStopPictureInPicture):
(VideoFullscreenInterfaceAVKit::didStopPictureInPicture):
(VideoFullscreenInterfaceAVKit::shouldExitFullscreenWithReason):
(VideoFullscreenInterfaceAVKit::doSetup):
(VideoFullscreenInterfaceAVKit::setMode):
(VideoFullscreenInterfaceAVKit::clearMode):
(VideoFullscreenInterfaceAVKit::setVideoFullscreenModel): Deleted.
(VideoFullscreenInterfaceAVKit::invalidate): Deleted.
* platform/ios/WebAVPlayerController.h:
* platform/ios/WebAVPlayerController.mm:
(-[WebAVPlayerController delegate]):
(-[WebAVPlayerController playbackSessionInterface]):
(-[WebAVPlayerController setPlaybackSessionInterface:]):
* platform/ios/WebVideoFullscreenControllerAVKit.mm:
(VideoFullscreenControllerContext::didCleanupFullscreen):
(VideoFullscreenControllerContext::addClient):
(VideoFullscreenControllerContext::removeClient):
(VideoFullscreenControllerContext::willEnterPictureInPicture):
(VideoFullscreenControllerContext::didEnterPictureInPicture):
(VideoFullscreenControllerContext::failedToEnterPictureInPicture):
(VideoFullscreenControllerContext::willExitPictureInPicture):
(VideoFullscreenControllerContext::didExitPictureInPicture):
(VideoFullscreenControllerContext::setUpFullscreen):
* platform/mac/PlaybackSessionInterfaceMac.h:
* platform/mac/PlaybackSessionInterfaceMac.mm:
(WebCore::PlaybackSessionInterfaceMac::create):
(WebCore::PlaybackSessionInterfaceMac::PlaybackSessionInterfaceMac):
(WebCore::PlaybackSessionInterfaceMac::playbackSessionModel const):
(WebCore::PlaybackSessionInterfaceMac::rateChanged):
(WebCore::PlaybackSessionInterfaceMac::beginScrubbing):
(WebCore::PlaybackSessionInterfaceMac::endScrubbing):
(WebCore::PlaybackSessionInterfaceMac::setPlayBackControlsManager):
(WebCore::PlaybackSessionInterfaceMac::updatePlaybackControlsManagerTiming):
(WebCore::PlaybackSessionInterfaceMac::~PlaybackSessionInterfaceMac): Deleted.
(WebCore::PlaybackSessionInterfaceMac::invalidate): Deleted.
* platform/mac/VideoFullscreenInterfaceMac.h:
(WebCore::VideoFullscreenInterfaceMac::create):
(WebCore::VideoFullscreenInterfaceMac::videoFullscreenModel const):
(WebCore::VideoFullscreenInterfaceMac::playbackSessionModel const):
(WebCore::VideoFullscreenInterfaceMac::videoFullscreenChangeObserver const):
* platform/mac/VideoFullscreenInterfaceMac.mm:
(-[WebVideoFullscreenInterfaceMacObjC setUpPIPForVideoView:withFrame:inWindow:]):
(-[WebVideoFullscreenInterfaceMacObjC boundsDidChangeForVideoViewContainer:]):
(-[WebVideoFullscreenInterfaceMacObjC pipDidClose:]):
(-[WebVideoFullscreenInterfaceMacObjC pipActionPlay:]):
(-[WebVideoFullscreenInterfaceMacObjC pipActionPause:]):
(-[WebVideoFullscreenInterfaceMacObjC pipActionStop:]):
(WebCore::VideoFullscreenInterfaceMac::VideoFullscreenInterfaceMac):
(WebCore::VideoFullscreenInterfaceMac::~VideoFullscreenInterfaceMac):
(WebCore::VideoFullscreenInterfaceMac::setVideoFullscreenChangeObserver):
(WebCore::VideoFullscreenInterfaceMac::setMode):
(WebCore::VideoFullscreenInterfaceMac::clearMode):
(WebCore::VideoFullscreenInterfaceMac::invalidate):
(WebCore::VideoFullscreenInterfaceMac::requestHideAndExitPiP):
(WebCore::VideoFullscreenInterfaceMac::setVideoFullscreenModel): Deleted.
* platform/mac/WebPlaybackControlsManager.mm:
(-[WebPlaybackControlsManager seekToTime:toleranceBefore:toleranceAfter:]):
(-[WebPlaybackControlsManager setCurrentAudioTouchBarMediaSelectionOption:]):
(-[WebPlaybackControlsManager setCurrentLegibleTouchBarMediaSelectionOption:]):
(-[WebPlaybackControlsManager togglePlayback]):
(-[WebPlaybackControlsManager setPlaying:]):
(-[WebPlaybackControlsManager isPlaying]):
(-[WebPlaybackControlsManager togglePictureInPicture]):

Source/WebKit:

Adopt those Ref and WeakPtr changes made in WebCore.

* UIProcess/Cocoa/PlaybackSessionManagerProxy.h:
* UIProcess/Cocoa/PlaybackSessionManagerProxy.mm:
(WebKit::PlaybackSessionModelContext::addClient):
(WebKit::PlaybackSessionModelContext::removeClient):
(WebKit::PlaybackSessionModelContext::durationChanged):
(WebKit::PlaybackSessionModelContext::currentTimeChanged):
(WebKit::PlaybackSessionModelContext::bufferedTimeChanged):
(WebKit::PlaybackSessionModelContext::rateChanged):
(WebKit::PlaybackSessionModelContext::seekableRangesChanged):
(WebKit::PlaybackSessionModelContext::canPlayFastReverseChanged):
(WebKit::PlaybackSessionModelContext::audioMediaSelectionOptionsChanged):
(WebKit::PlaybackSessionModelContext::legibleMediaSelectionOptionsChanged):
(WebKit::PlaybackSessionModelContext::audioMediaSelectionIndexChanged):
(WebKit::PlaybackSessionModelContext::legibleMediaSelectionIndexChanged):
(WebKit::PlaybackSessionModelContext::externalPlaybackChanged):
(WebKit::PlaybackSessionModelContext::wirelessVideoPlaybackDisabledChanged):
(WebKit::PlaybackSessionModelContext::mutedChanged):
(WebKit::PlaybackSessionModelContext::volumeChanged):
(WebKit::PlaybackSessionModelContext::pictureInPictureActiveChanged):
(WebKit::PlaybackSessionManagerProxy::invalidate):
(WebKit::PlaybackSessionManagerProxy::createModelAndInterface):
(WebKit::PlaybackSessionManagerProxy::removeClientForContext):
* UIProcess/Cocoa/VideoFullscreenManagerProxy.h:
* UIProcess/Cocoa/VideoFullscreenManagerProxy.mm:
(WebKit::VideoFullscreenModelContext::create):
(WebKit::VideoFullscreenModelContext::VideoFullscreenModelContext):
(WebKit::VideoFullscreenModelContext::addClient):
(WebKit::VideoFullscreenModelContext::removeClient):
(WebKit::VideoFullscreenModelContext::willEnterPictureInPicture):
(WebKit::VideoFullscreenModelContext::didEnterPictureInPicture):
(WebKit::VideoFullscreenModelContext::failedToEnterPictureInPicture):
(WebKit::VideoFullscreenModelContext::willExitPictureInPicture):
(WebKit::VideoFullscreenModelContext::didExitPictureInPicture):
(WebKit::VideoFullscreenManagerProxy::invalidate):
(WebKit::VideoFullscreenManagerProxy::createModelAndInterface):
(WebKit::VideoFullscreenManagerProxy::removeClientForContext):
(WebKit::VideoFullscreenModelContext::~VideoFullscreenModelContext): Deleted.
* UIProcess/ios/fullscreen/WKFullScreenViewController.mm:
(WKFullScreenViewControllerPlaybackSessionModelClient::setInterface):
(WKFullScreenViewControllerVideoFullscreenModelClient::setInterface):
(-[WKFullScreenViewController videoControlsManagerDidChange]):
(-[WKFullScreenViewController _togglePiPAction:]):
* UIProcess/mac/WKFullScreenWindowController.mm:
(WebKit::WKFullScreenWindowControllerVideoFullscreenModelClient::setInterface):
* WebProcess/cocoa/PlaybackSessionManager.h:
* WebProcess/cocoa/PlaybackSessionManager.mm:
(WebKit::PlaybackSessionInterfaceContext::PlaybackSessionInterfaceContext):
(WebKit::PlaybackSessionManager::~PlaybackSessionManager):
(WebKit::PlaybackSessionManager::createModelAndInterface):
(WebKit::PlaybackSessionManager::removeContext):
(WebKit::PlaybackSessionInterfaceContext::~PlaybackSessionInterfaceContext): Deleted.
* WebProcess/cocoa/VideoFullscreenManager.h:
(WebKit::VideoFullscreenInterfaceContext::create):
(WebKit::VideoFullscreenInterfaceContext::createWeakPtr):
* WebProcess/cocoa/VideoFullscreenManager.mm:
(WebKit::VideoFullscreenInterfaceContext::VideoFullscreenInterfaceContext):
(WebKit::VideoFullscreenManager::~VideoFullscreenManager):
(WebKit::VideoFullscreenManager::createModelAndInterface):
(WebKit::VideoFullscreenManager::removeContext):

Source/WebKitLegacy/mac:

PlaybackSessionInterface no longer has an invalidate() method.

* WebView/WebView.mm:
(-[WebView _clearPlaybackControlsManager]):

Source/WTF:

* WTF.xcodeproj/project.pbxproj:
* wtf/WeakPtrContainer.h: Added.

LayoutTests:

Test was causing false pass results due to the webkitPresentationMode being correctly set
to "inline" during the close() operation; modify the test to only perform the close()
operation when the presetation mode is changed to "picture-in-picture".

* media/controls/ipad/close-page-with-picture-in-picture-video-assertion-failure.html:
* media/controls/ipad/resources/picture-in-picture.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236624 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoDrop support for cross-origin-window-policy header
cdumez@apple.com [Fri, 28 Sep 2018 23:18:28 +0000 (23:18 +0000)]
Drop support for cross-origin-window-policy header
https://bugs.webkit.org/show_bug.cgi?id=190081

Reviewed by Ryosuke Niwa.

Source/WebCore:

Drop support for cross-origin-window-policy header as this was never enabled and its design has
some issues we have not resolved. An alternative is being worked on but will be substantially
different so there is not much value in keeping this code around.

* bindings/js/JSDOMBindingSecurity.cpp:
* bindings/js/JSDOMBindingSecurity.h:
* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::jsDOMWindowGetOwnPropertySlotRestrictedAccess):
(WebCore::JSDOMWindow::getOwnPropertySlotByIndex):
(WebCore::addCrossOriginWindowPropertyNames):
(WebCore::addScopedChildrenIndexes):
(WebCore::addCrossOriginWindowOwnPropertyNames):
(WebCore::JSDOMWindow::getOwnPropertyNames):
* bindings/js/JSDOMWindowCustom.h:
* bindings/js/JSRemoteDOMWindowCustom.cpp:
* bindings/scripts/CodeGeneratorJS.pm:
(GenerateAttributeGetterBodyDefinition):
(GenerateAttributeSetterBodyDefinition):
(GenerateOperationBodyDefinition):
* bindings/scripts/IDLAttributes.json:
* dom/Document.cpp:
(WebCore::Document::canNavigate):
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::didBeginDocument):
* page/AbstractDOMWindow.cpp:
(WebCore::AbstractDOMWindow::AbstractDOMWindow):
* page/AbstractDOMWindow.h:
* page/DOMWindow.idl:
* page/Settings.yaml:
* platform/network/HTTPParsers.cpp:
* platform/network/HTTPParsers.h:

Source/WebKit:

* Shared/WebPreferences.yaml:
* UIProcess/API/C/WKPreferences.cpp:
* UIProcess/API/C/WKPreferencesRef.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::frameBecameRemote):

Source/WebKitLegacy/mac:

* WebView/WebPreferenceKeysPrivate.h:
* WebView/WebPreferences.mm:
(+[WebPreferences initialize]):
* WebView/WebPreferencesPrivate.h:
* WebView/WebView.mm:
(-[WebView _preferencesChanged:]):

Source/WebKitLegacy/win:

* WebPreferenceKeysPrivate.h:
* WebPreferences.cpp:
(WebPreferences::initializeDefaultSettings):
(WebPreferences::crossOriginWindowPolicySupportEnabled):
(WebPreferences::setCrossOriginWindowPolicySupportEnabled):
* WebView.cpp:
(WebView::notifyPreferencesChanged):

Tools:

190081_DropCrossOriginWindowPolicy

* DumpRenderTree/mac/DumpRenderTree.mm:
(enableExperimentalFeatures):
* DumpRenderTree/win/DumpRenderTree.cpp:
(enableExperimentalFeatures):
* WebKitTestRunner/TestController.cpp:
(WTR::TestController::resetPreferencesToConsistentValues):

LayoutTests:

* http/wpt/cross-origin-window-policy/allow-postmessage-expected.txt: Removed.
* http/wpt/cross-origin-window-policy/allow-postmessage-from-deny-expected.txt: Removed.
* http/wpt/cross-origin-window-policy/allow-postmessage-from-deny.html: Removed.
* http/wpt/cross-origin-window-policy/allow-postmessage-from-deny.html.headers: Removed.
* http/wpt/cross-origin-window-policy/allow-postmessage.html: Removed.
* http/wpt/cross-origin-window-policy/cross-origin-window-policy-header-expected.txt: Removed.
* http/wpt/cross-origin-window-policy/cross-origin-window-policy-header.html: Removed.
* http/wpt/cross-origin-window-policy/navigation-from-opener-via-open-target-expected.txt: Removed.
* http/wpt/cross-origin-window-policy/navigation-from-opener-via-open-target.html: Removed.
* http/wpt/cross-origin-window-policy/navigation-from-subframe-via-anchor-target-expected.txt: Removed.
* http/wpt/cross-origin-window-policy/navigation-from-subframe-via-anchor-target.html: Removed.
* http/wpt/cross-origin-window-policy/resources/cross-origin-window-policy-allow-postmessage-pong.html: Removed.
* http/wpt/cross-origin-window-policy/resources/cross-origin-window-policy-allow-postmessage-pong.html.headers: Removed.
* http/wpt/cross-origin-window-policy/resources/destination.html: Removed.
* http/wpt/cross-origin-window-policy/resources/navigate-parent-via-anchor.html: Removed.
* http/wpt/cross-origin-window-policy/resources/navigation-from-subframe-frame.py: Removed.
* http/wpt/cross-origin-window-policy/resources/serve-cross-origin-window-policy-header.py: Removed.
* http/wpt/cross-origin-window-policy/resources/utils.js: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236623 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoREGRESSION (r236091): CSSKeywordCompletions.js has "checkout" instead of "check-out...
aestes@apple.com [Fri, 28 Sep 2018 23:13:38 +0000 (23:13 +0000)]
REGRESSION (r236091): CSSKeywordCompletions.js has "checkout" instead of "check-out" as a value for -apple-pay-button-type
https://bugs.webkit.org/show_bug.cgi?id=190096

Reviewed by Ryosuke Niwa.

I changed "checkout" to "check-out" in r235754, but it somehow got flipped back to "checkout" in r236091.

* UserInterface/Models/CSSKeywordCompletions.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236622 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoDo not do automatic process prewarming while under memory pressure
cdumez@apple.com [Fri, 28 Sep 2018 23:09:32 +0000 (23:09 +0000)]
Do not do automatic process prewarming while under memory pressure
https://bugs.webkit.org/show_bug.cgi?id=190082
<rdar://problem/39771424>

Reviewed by Geoffrey Garen.

Do not do automatic process prewarming while under memory pressure and
also terminate any prewarmed process when receiving a memory warning.

* UIProcess/WebProcessPool.cpp:
(WebKit::WebProcessPool::prewarmProcess):
(WebKit::WebProcessPool::didReachGoodTimeToPrewarm):
(WebKit::WebProcessPool::handleMemoryPressureWarning):
* UIProcess/WebProcessPool.h:
* UIProcess/ios/WebMemoryPressureHandlerIOS.mm:
(WebKit::installMemoryPressureHandler):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236621 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoAdd yasm as third party tool for libwebrtc compilation
youenn@apple.com [Fri, 28 Sep 2018 23:05:36 +0000 (23:05 +0000)]
Add yasm as third party tool for libwebrtc compilation
https://bugs.webkit.org/show_bug.cgi?id=190025

Reviewed by Eric Carlson.

Add yasm source code and build the yasm executable as it is needed for libvpx compilation.

* Source/third_party/yasm-1.3.0: Added.
* libwebrtc.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236620 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Allow programmatic focus when hardware keyboard is attached
dbates@webkit.org [Fri, 28 Sep 2018 23:02:41 +0000 (23:02 +0000)]
[iOS] Allow programmatic focus when hardware keyboard is attached
https://bugs.webkit.org/show_bug.cgi?id=190017
<rdar://problem/42270463>

Reviewed by Wenson Hsieh.

Source/WebCore:

Add support for checking if the embedding client is WebKitTestRunner and export isDumpRenderTree()
so that we can make use of it from WebKit. We will make use of these functions to keep the current
behavior of disallowing programmatic focus when running tests in these apps. This is needed to
keep testing deterministic. Otherwise, test results would be dependent on whether a hardware
keyboard is attached. When running tests in Simulator.app the hardware keyboard may also not be
connected (i.e. Hardware > Keyboard > Connect Hardware Keyboard is disabled).

* platform/RuntimeApplicationChecks.h:
* platform/cocoa/RuntimeApplicationChecksCocoa.mm:
(WebCore::IOSApplication::isWebKitTestRunner): Added.

Source/WebKit:

Make the experience of using iOS with a hardware keyboard more desktop-like by allowing
programmatic focusing of editable elements.

* Platform/spi/ios/UIKitSPI.h: Forward declare SPI.
* Shared/NativeWebKeyboardEvent.h:
* Shared/ios/NativeWebKeyboardEventIOS.mm:
(WebKit::isInHardwareKeyboardMode): Returns whether we are in hardware keyboard mode. In DumpRenderTree
and WebKitTestRunner this function always returns false to keep test results deterministic.
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView _startAssistingNode:userIsInteracting:blurPreviousNode:changingActivityState:userObject:]):
Allow starting an input session if we are in hardware keyboard mode.
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::platformEditorState const): Send the full editor state if we are in hardware
keyboard mode regardless of whether layout has been performed so that UIProcess can update UI,
including the position of the caret, immediately.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236619 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoREGRESSION(r236609): API tests for mso list preservation are failing
rniwa@webkit.org [Fri, 28 Sep 2018 22:32:56 +0000 (22:32 +0000)]
REGRESSION(r236609): API tests for mso list preservation are failing
https://bugs.webkit.org/show_bug.cgi?id=190095

Reviewed by Wenson Hsieh.

The regression was caused by appendNodeToPreserveMSOList called after an early return for not having renderer.
Clearly, comment & style elements coming from a MS word document wouldn't have a renderer.

Fixed the bug by changing the order.

* editing/markup.cpp:
(WebCore::StyledMarkupAccumulator::traverseNodesForSerialization):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236618 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WTF] Make isMainThread more reliable
yusukesuzuki@slowstart.org [Fri, 28 Sep 2018 22:32:34 +0000 (22:32 +0000)]
[WTF] Make isMainThread more reliable
https://bugs.webkit.org/show_bug.cgi?id=189880

Reviewed by Mark Lam.

.:

* Source/cmake/OptionsCommon.cmake:

Source/WTF:

isMainThread() relied on Thread::current(). This API becomes broken in Windows
when the Thread is about to be destroyed since TLS is already cleared. This causes
a bug since `isMainThread()` is called in Thread::didExit in Windows.

This patch makes this `isMainThread` more reliable in all the platforms. In Windows,
we use `Thread::currentID()` instead of `Thread::current()` since `Thread::currentID`
uses Win32 GetCurrentThreadId directly. In the other system, we use `pthread_main_np`
or `pthread_self` instead.

We also move `holdLock` code inside `if (shouldRemoveThreadFromThreadGroup())`. If
the other thread takes a mutex and destroyed, this `holdLock` waits forever. This problem
only happens in Windows since Windows calls TLS destructor for the main thread.

* WTF.xcodeproj/project.pbxproj:
* wtf/MainThread.cpp:
(WTF::initializeMainThread):
(): Deleted.
(WTF::isMainThread): Deleted.
(WTF::isMainThreadIfInitialized): Deleted.
* wtf/Platform.h:
* wtf/PlatformMac.cmake:
* wtf/Threading.cpp:
(WTF::Thread::didExit):
* wtf/cocoa/MainThreadCocoa.mm: Renamed from Source/WTF/wtf/mac/MainThreadMac.mm.
* wtf/generic/MainThreadGeneric.cpp:
(WTF::initializeMainThreadPlatform):
(WTF::isMainThread):
(WTF::isMainThreadIfInitialized):
* wtf/win/MainThreadWin.cpp:
(WTF::initializeMainThreadPlatform):
(WTF::isMainThread):
(WTF::isMainThreadIfInitialized):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236617 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoBuild fix after r236612.
rniwa@webkit.org [Fri, 28 Sep 2018 22:12:07 +0000 (22:12 +0000)]
Build fix after r236612.

* platform/win/PasteboardWin.cpp:
(WebCore::Pasteboard::writeSelection):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236616 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Apple Pay] Remove the "in-store" button type
aestes@apple.com [Fri, 28 Sep 2018 22:06:02 +0000 (22:06 +0000)]
[Apple Pay] Remove the "in-store" button type
https://bugs.webkit.org/show_bug.cgi?id=190079

Reviewed by Tim Horton.

Source/WebCore:

According to <https://developer.apple.com/design/human-interface-guidelines/apple-pay/buttons-and-marks/buttons/>,
this button is meant only for certain kinds of native apps. It shouldn't be available on the web.

Updated http/tests/ssl/applepay/ApplePayButton.html.

* css/CSSPrimitiveValueMappings.h:
(WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
(WebCore::CSSPrimitiveValue::operator ApplePayButtonType const):
* css/CSSValueKeywords.in:
* css/parser/CSSParserFastPaths.cpp:
(WebCore::CSSParserFastPaths::isValidKeywordPropertyAndValue):
* rendering/RenderThemeCocoa.mm:
(WebCore::toPKPaymentButtonType):
* rendering/style/RenderStyleConstants.h:

Source/WebInspectorUI:

* UserInterface/Models/CSSKeywordCompletions.js:

LayoutTests:

* http/tests/ssl/applepay/ApplePayButton.html:
* platform/mac/http/tests/ssl/applepay/ApplePayButton-expected.png:
* platform/mac/http/tests/ssl/applepay/ApplePayButton-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236615 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoWebGL conformance: Failures and Timeouts in suite 2.0.0/conformance
realdawei@apple.com [Fri, 28 Sep 2018 22:01:21 +0000 (22:01 +0000)]
WebGL conformance: Failures and Timeouts in suite 2.0.0/conformance
https://bugs.webkit.org/show_bug.cgi?id=189641

Unreviewed, skip tests that are timing out.

Patch by Dawei Fenton <realdawei@apple.com> on 2018-09-28

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236614 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agodocument.open() should throw errors for cross-origin calls
cdumez@apple.com [Fri, 28 Sep 2018 21:56:33 +0000 (21:56 +0000)]
document.open() should throw errors for cross-origin calls
https://bugs.webkit.org/show_bug.cgi?id=189371
<rdar://problem/44282700>

Reviewed by Youenn Fablet.

LayoutTests/imported/w3c:

Rebaseline existing WPT tests now that more checks are passing.

* web-platform-tests/html/webappapis/dynamic-markup-insertion/opening-the-input-stream/bailout-exception-vs-return-origin.sub.window-expected.txt:
* web-platform-tests/html/webappapis/dynamic-markup-insertion/opening-the-input-stream/bailout-side-effects-same-origin-domain.sub.window-expected.txt:

Source/WebCore:

document.open() / document.write() should throw errors for cross-origin calls as per:
- https://html.spec.whatwg.org/#document-open-steps (Step 4)

No new tests, rebaselined existing tests.

* dom/Document.cpp:
(WebCore::Document::open):
(WebCore::Document::write):
(WebCore::Document::writeln):
* dom/Document.h:

LayoutTests:

Tweak a couple of existing tests to reflect behavior change.

* fast/dom/HTMLDocument/document-open-return-value.html:
* fast/parser/tokenizer-close-during-document-write.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236613 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRename createMarkup to serializePreservingVisualAppearance
rniwa@webkit.org [Fri, 28 Sep 2018 21:08:54 +0000 (21:08 +0000)]
Rename createMarkup to serializePreservingVisualAppearance
https://bugs.webkit.org/show_bug.cgi?id=190086

Reviewed by Wenson Hsieh.

Source/WebCore:

Renamed the function to clarify what it does. Also removed the unused Range::toHTML.

* dom/Range.cpp:
(WebCore::Range::toHTML const): Deleted.
* dom/Range.h:
* editing/CompositeEditCommand.cpp:
(WebCore::CompositeEditCommand::moveParagraphs):
* editing/cocoa/EditorCocoa.mm:
(WebCore::Editor::selectionInHTMLFormat):
* editing/gtk/EditorGtk.cpp:
(WebCore::Editor::writeSelectionToPasteboard):
* editing/markup.cpp:
(WebCore::serializePreservingVisualAppearance):
(WebCore::createMarkup): Deleted.
* editing/markup.h:
* editing/wpe/EditorWPE.cpp:
(WebCore::Editor::writeSelectionToPasteboard):
* loader/archive/cf/LegacyWebArchive.cpp:
(WebCore::LegacyWebArchive::create):
(WebCore::LegacyWebArchive::createFromSelection):
* platform/win/PasteboardWin.cpp:
(WebCore::Pasteboard::writeRangeToDataObject):

Source/WebKit:

* WebProcess/WebCoreSupport/gtk/WebEditorClientGtk.cpp:
(WebKit::WebEditorClient::updateGlobalSelection):

Source/WebKitLegacy/mac:

* DOM/WebDOMOperations.mm:
(-[DOMRange markupString]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236612 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRemove some unused RenderLayer code
simon.fraser@apple.com [Fri, 28 Sep 2018 21:03:23 +0000 (21:03 +0000)]
Remove some unused RenderLayer code
https://bugs.webkit.org/show_bug.cgi?id=190078

Reviewed by Zalan Bujtas.

The 'outOfFlowDescendantContainingBlocks' code was related to the accelerated overflow scrolling code that
I removed recently.

updateDescendantsLayerListsIfNeeded() is never called.

* rendering/RenderLayer.cpp:
(WebCore::RenderLayer::updateDescendantDependentFlags):
(WebCore::RenderLayer::calculateClipRects const):
* rendering/RenderLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236611 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUnreviewed, rolling out r236605.
commit-queue@webkit.org [Fri, 28 Sep 2018 20:38:32 +0000 (20:38 +0000)]
Unreviewed, rolling out r236605.
https://bugs.webkit.org/show_bug.cgi?id=190087

caused three API test timeouts (Requested by jernoble on
#webkit).

Reverted changeset:

"Refactoring: eliminate raw pointer usage in Fullscreen code"
https://bugs.webkit.org/show_bug.cgi?id=188747
https://trac.webkit.org/changeset/236605

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236610 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoSimplify StyledMarkupAccumulator::traverseNodesForSerialization
rniwa@webkit.org [Fri, 28 Sep 2018 20:03:45 +0000 (20:03 +0000)]
Simplify StyledMarkupAccumulator::traverseNodesForSerialization
https://bugs.webkit.org/show_bug.cgi?id=190073

Reviewed by Antti Koivisto.

Simplified the range traversal algorithm in traverseNodesForSerialization as it was too complicated
to support shadow DOM for copy and paste.

Instead of using NodeTraversal::next to traverse past ancestors and then figuring out which ancestor
must be closed or to wrap the existing markup with, new code collects the list of ancestors as we
traverse out of them.

Also extracted lambdas for generating markup and deciding whether to skip a node as well as keeping
track of the depth of the current markup. This further reduces the code complexity of the actual
node traversal algorithm. Keeping track of the depth allows us to now generate ancestor elements'
closing tags without keeping a stack of ancestor nodes we opened at all times.

* editing/markup.cpp:
(WebCore::StyledMarkupAccumulator::traverseNodesForSerialization):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236609 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoReplace recently added line comments in Compiler.h
bburg@apple.com [Fri, 28 Sep 2018 20:02:55 +0000 (20:02 +0000)]
Replace recently added line comments in Compiler.h
https://bugs.webkit.org/show_bug.cgi?id=190062
<rdar://problem/44838618>

Reviewed by Joseph Pecoraro.

This breaks some Apple-internal tooling. For now, work around it by
changing the comment style. On the other side, the issue will be fixed
more permanently by adopting the approach from r230213.

* wtf/Compiler.h: Use multiline comments.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236608 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoReplace every use of Node::offsetInCharacters() by Node::isCharacterDataNode()
rniwa@webkit.org [Fri, 28 Sep 2018 19:47:22 +0000 (19:47 +0000)]
Replace every use of Node::offsetInCharacters() by Node::isCharacterDataNode()
https://bugs.webkit.org/show_bug.cgi?id=190069

Reviewed by Zalan Bujtas.

Removed Node::offsetInCharacters() and replaced every use of it by isCharacterDataNode()
because their implementations are identical.

Note that offsetInCharacters() sounds like a function which returns some kind of an offset
but it doesn't. It returns true when called on a CharacterData and false elsewhere.

* accessibility/AXObjectCache.cpp:
(WebCore::characterOffsetsInOrder):
(WebCore::AXObjectCache::startOrEndCharacterOffsetForRange):
(WebCore::AXObjectCache::characterOffsetFromVisiblePosition):
* dom/CharacterData.cpp:
(WebCore::CharacterData::offsetInCharacters const): Deleted.
* dom/CharacterData.h:
* dom/Node.cpp:
(WebCore::Node::offsetInCharacters const): Deleted.
* dom/Node.h:
* dom/Position.cpp:
(WebCore::Position::parentAnchoredEquivalent const):
* dom/Position.h:
(WebCore::lastOffsetInNode):
(WebCore::minOffsetForNode):
(WebCore::offsetIsBeforeLastNodeOffset):
* dom/Range.cpp:
(WebCore::Range::firstNode const):
(WebCore::Range::pastLastNode const):
* dom/RangeBoundaryPoint.h:
(WebCore::RangeBoundaryPoint::setOffset):
(WebCore::RangeBoundaryPoint::setToEndOfNode):
* editing/Editing.cpp:
(WebCore::lastOffsetForEditing):
* editing/TextIterator.cpp:
(WebCore::nextInPreOrderCrossingShadowBoundaries):
(WebCore::TextIterator::node const):
(WebCore::SimplifiedBackwardsTextIterator::SimplifiedBackwardsTextIterator):
* page/DOMSelection.cpp:
(WebCore::DOMSelection::extend):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236607 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoGardening: speculative build fix.
mark.lam@apple.com [Fri, 28 Sep 2018 18:46:52 +0000 (18:46 +0000)]
Gardening: speculative build fix.
<rdar://problem/44869924>

Not reviewed.

* assembler/LinkBuffer.cpp:
(JSC::LinkBuffer::copyCompactAndLinkCode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236606 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoRefactoring: eliminate raw pointer usage in Fullscreen code
jer.noble@apple.com [Fri, 28 Sep 2018 18:39:51 +0000 (18:39 +0000)]
Refactoring: eliminate raw pointer usage in Fullscreen code
https://bugs.webkit.org/show_bug.cgi?id=188747
<rdar://problem/43541164>

Reviewed by Alex Christensen.

Source/WebCore:

Two sources of raw pointers in the Fullscreen code:
- Model classes (PlaybackSessionModel and VideoFullscreenModel) aren't ref-able, so
  they are passed around as raw references.
- Observer classes (PlaybackSessionModelClient and VideoFullscreenModelClient, and
  VideoFullscreenChangeObserver) are also passed around as raw pointers, but shouldn't
  be ref-able.

Make Model classes ref-able by adding ref() and deref() which call virtual refModel and
derefModel methods, overridden by implementing subclasses. Make every concrete observer
inherit from CanMakeWeakPtr, and every registration method take WeakPtr wrappers around
the client interface.

Since every Interface class now holds a strong reference to its Model classes, and each
Model class holds a weak reference to all its clients, no explicit invalidate() method
is necessary.

Notes:

- Since the weak pointer methods need to be able to downcast to the abstract base class,
  observers need to inherit publically (rather than privately) from those base classes.
- Media element Models should compose EventListener rather than inheriting from it, since
  EventListener has its own RefCount.
- WeakPtrs can't be held in HashSets (because they change value, and therefore hash, when
  their underlying object is destroyed), so clients should be stored in a Vector instead.
- Interfaces should be given all required Refs at creation time, so that they can store
  those parameters as Refs instead of RefPtrs.

* platform/cocoa/PlaybackSessionInterface.h:
(WebCore::PlaybackSessionInterface::~PlaybackSessionInterface): Deleted.
* platform/cocoa/PlaybackSessionModel.h:
(WebCore::PlaybackSessionModel::ref):
(WebCore::PlaybackSessionModel::deref):
(WebCore::PlaybackSessionModel::~PlaybackSessionModel): Deleted.
* platform/cocoa/PlaybackSessionModelMediaElement.h:
* platform/cocoa/PlaybackSessionModelMediaElement.mm:
(WebCore::PlaybackSessionModelMediaElement::PlaybackSessionModelMediaElement):
(WebCore::PlaybackSessionModelMediaElement::~PlaybackSessionModelMediaElement):
(WebCore::PlaybackSessionModelMediaElement::setMediaElement):
(WebCore::PlaybackSessionModelMediaElement::updateForEventName):
(WebCore::PlaybackSessionModelMediaElement::addClient):
(WebCore::PlaybackSessionModelMediaElement::removeClient):
(WebCore::PlaybackSessionModelMediaElement::updateMediaSelectionOptions):
(WebCore::PlaybackSessionModelMediaElement::updateMediaSelectionIndices):
(WebCore::PlaybackSessionModelMediaElement::handleEvent): Deleted.
* platform/cocoa/VideoFullscreenChangeObserver.h:
(WebCore::VideoFullscreenChangeObserver::~VideoFullscreenChangeObserver): Deleted.
* platform/cocoa/VideoFullscreenModel.h:
(WebCore::VideoFullscreenModel::ref):
(WebCore::VideoFullscreenModel::deref):
(WebCore::VideoFullscreenModel::~VideoFullscreenModel): Deleted.
* platform/cocoa/VideoFullscreenModelVideoElement.h:
* platform/cocoa/VideoFullscreenModelVideoElement.mm:
(VideoFullscreenModelVideoElement::VideoFullscreenModelVideoElement):
(VideoFullscreenModelVideoElement::setVideoElement):
(VideoFullscreenModelVideoElement::addClient):
(VideoFullscreenModelVideoElement::removeClient):
(VideoFullscreenModelVideoElement::setHasVideo):
(VideoFullscreenModelVideoElement::setVideoDimensions):
(VideoFullscreenModelVideoElement::willEnterPictureInPicture):
(VideoFullscreenModelVideoElement::didEnterPictureInPicture):
(VideoFullscreenModelVideoElement::failedToEnterPictureInPicture):
(VideoFullscreenModelVideoElement::willExitPictureInPicture):
(VideoFullscreenModelVideoElement::didExitPictureInPicture):
(VideoFullscreenModelVideoElement::handleEvent): Deleted.
* platform/ios/PlaybackSessionInterfaceAVKit.h:
(WebCore::PlaybackSessionInterfaceAVKit::create):
(WebCore::PlaybackSessionInterfaceAVKit::playbackSessionModel const):
(): Deleted.
* platform/ios/PlaybackSessionInterfaceAVKit.mm:
(WebCore::PlaybackSessionInterfaceAVKit::PlaybackSessionInterfaceAVKit):
(WebCore::PlaybackSessionInterfaceAVKit::~PlaybackSessionInterfaceAVKit):
(WebCore::PlaybackSessionInterfaceAVKit::invalidate): Deleted.
* platform/ios/VideoFullscreenInterfaceAVKit.h:
* platform/ios/VideoFullscreenInterfaceAVKit.mm:
(-[WebAVPlayerLayer layoutSublayers]):
(-[WebAVPlayerLayer resolveBounds]):
(-[WebAVPlayerLayer setVideoGravity:]):
(VideoFullscreenInterfaceAVKit::create):
(VideoFullscreenInterfaceAVKit::VideoFullscreenInterfaceAVKit):
(VideoFullscreenInterfaceAVKit::~VideoFullscreenInterfaceAVKit):
(VideoFullscreenInterfaceAVKit::setVideoFullscreenChangeObserver):
(VideoFullscreenInterfaceAVKit::applicationDidBecomeActive):
(VideoFullscreenInterfaceAVKit::setupFullscreen):
(VideoFullscreenInterfaceAVKit::presentingViewController):
(VideoFullscreenInterfaceAVKit::requestHideAndExitFullscreen):
(VideoFullscreenInterfaceAVKit::preparedToExitFullscreen):
(VideoFullscreenInterfaceAVKit::willStartPictureInPicture):
(VideoFullscreenInterfaceAVKit::didStartPictureInPicture):
(VideoFullscreenInterfaceAVKit::failedToStartPictureInPicture):
(VideoFullscreenInterfaceAVKit::willStopPictureInPicture):
(VideoFullscreenInterfaceAVKit::didStopPictureInPicture):
(VideoFullscreenInterfaceAVKit::shouldExitFullscreenWithReason):
(VideoFullscreenInterfaceAVKit::doSetup):
(VideoFullscreenInterfaceAVKit::setMode):
(VideoFullscreenInterfaceAVKit::clearMode):
(VideoFullscreenInterfaceAVKit::setVideoFullscreenModel): Deleted.
(VideoFullscreenInterfaceAVKit::invalidate): Deleted.
* platform/ios/WebAVPlayerController.h:
* platform/ios/WebAVPlayerController.mm:
(-[WebAVPlayerController delegate]):
(-[WebAVPlayerController playbackSessionInterface]):
(-[WebAVPlayerController setPlaybackSessionInterface:]):
* platform/ios/WebVideoFullscreenControllerAVKit.mm:
(VideoFullscreenControllerContext::didCleanupFullscreen):
(VideoFullscreenControllerContext::addClient):
(VideoFullscreenControllerContext::removeClient):
(VideoFullscreenControllerContext::willEnterPictureInPicture):
(VideoFullscreenControllerContext::didEnterPictureInPicture):
(VideoFullscreenControllerContext::failedToEnterPictureInPicture):
(VideoFullscreenControllerContext::willExitPictureInPicture):
(VideoFullscreenControllerContext::didExitPictureInPicture):
(VideoFullscreenControllerContext::setUpFullscreen):
* platform/mac/PlaybackSessionInterfaceMac.h:
* platform/mac/PlaybackSessionInterfaceMac.mm:
(WebCore::PlaybackSessionInterfaceMac::create):
(WebCore::PlaybackSessionInterfaceMac::PlaybackSessionInterfaceMac):
(WebCore::PlaybackSessionInterfaceMac::playbackSessionModel const):
(WebCore::PlaybackSessionInterfaceMac::rateChanged):
(WebCore::PlaybackSessionInterfaceMac::beginScrubbing):
(WebCore::PlaybackSessionInterfaceMac::endScrubbing):
(WebCore::PlaybackSessionInterfaceMac::setPlayBackControlsManager):
(WebCore::PlaybackSessionInterfaceMac::updatePlaybackControlsManagerTiming):
(WebCore::PlaybackSessionInterfaceMac::~PlaybackSessionInterfaceMac): Deleted.
(WebCore::PlaybackSessionInterfaceMac::invalidate): Deleted.
* platform/mac/VideoFullscreenInterfaceMac.h:
(WebCore::VideoFullscreenInterfaceMac::create):
(WebCore::VideoFullscreenInterfaceMac::videoFullscreenModel const):
(WebCore::VideoFullscreenInterfaceMac::playbackSessionModel const):
(WebCore::VideoFullscreenInterfaceMac::videoFullscreenChangeObserver const):
* platform/mac/VideoFullscreenInterfaceMac.mm:
(-[WebVideoFullscreenInterfaceMacObjC setUpPIPForVideoView:withFrame:inWindow:]):
(-[WebVideoFullscreenInterfaceMacObjC boundsDidChangeForVideoViewContainer:]):
(-[WebVideoFullscreenInterfaceMacObjC pipDidClose:]):
(-[WebVideoFullscreenInterfaceMacObjC pipActionPlay:]):
(-[WebVideoFullscreenInterfaceMacObjC pipActionPause:]):
(-[WebVideoFullscreenInterfaceMacObjC pipActionStop:]):
(WebCore::VideoFullscreenInterfaceMac::VideoFullscreenInterfaceMac):
(WebCore::VideoFullscreenInterfaceMac::~VideoFullscreenInterfaceMac):
(WebCore::VideoFullscreenInterfaceMac::setVideoFullscreenChangeObserver):
(WebCore::VideoFullscreenInterfaceMac::setMode):
(WebCore::VideoFullscreenInterfaceMac::clearMode):
(WebCore::VideoFullscreenInterfaceMac::invalidate):
(WebCore::VideoFullscreenInterfaceMac::requestHideAndExitPiP):
(WebCore::VideoFullscreenInterfaceMac::setVideoFullscreenModel): Deleted.
* platform/mac/WebPlaybackControlsManager.mm:
(-[WebPlaybackControlsManager seekToTime:toleranceBefore:toleranceAfter:]):
(-[WebPlaybackControlsManager setCurrentAudioTouchBarMediaSelectionOption:]):
(-[WebPlaybackControlsManager setCurrentLegibleTouchBarMediaSelectionOption:]):
(-[WebPlaybackControlsManager togglePlayback]):
(-[WebPlaybackControlsManager setPlaying:]):
(-[WebPlaybackControlsManager isPlaying]):
(-[WebPlaybackControlsManager togglePictureInPicture]):

Source/WebKit:

Adopt those Ref and WeakPtr changes made in WebCore.

* UIProcess/Cocoa/PlaybackSessionManagerProxy.h:
* UIProcess/Cocoa/PlaybackSessionManagerProxy.mm:
(WebKit::PlaybackSessionModelContext::addClient):
(WebKit::PlaybackSessionModelContext::removeClient):
(WebKit::PlaybackSessionModelContext::durationChanged):
(WebKit::PlaybackSessionModelContext::currentTimeChanged):
(WebKit::PlaybackSessionModelContext::bufferedTimeChanged):
(WebKit::PlaybackSessionModelContext::rateChanged):
(WebKit::PlaybackSessionModelContext::seekableRangesChanged):
(WebKit::PlaybackSessionModelContext::canPlayFastReverseChanged):
(WebKit::PlaybackSessionModelContext::audioMediaSelectionOptionsChanged):
(WebKit::PlaybackSessionModelContext::legibleMediaSelectionOptionsChanged):
(WebKit::PlaybackSessionModelContext::audioMediaSelectionIndexChanged):
(WebKit::PlaybackSessionModelContext::legibleMediaSelectionIndexChanged):
(WebKit::PlaybackSessionModelContext::externalPlaybackChanged):
(WebKit::PlaybackSessionModelContext::wirelessVideoPlaybackDisabledChanged):
(WebKit::PlaybackSessionModelContext::mutedChanged):
(WebKit::PlaybackSessionModelContext::volumeChanged):
(WebKit::PlaybackSessionModelContext::pictureInPictureActiveChanged):
(WebKit::PlaybackSessionManagerProxy::invalidate):
(WebKit::PlaybackSessionManagerProxy::createModelAndInterface):
(WebKit::PlaybackSessionManagerProxy::removeClientForContext):
* UIProcess/Cocoa/VideoFullscreenManagerProxy.h:
* UIProcess/Cocoa/VideoFullscreenManagerProxy.mm:
(WebKit::VideoFullscreenModelContext::create):
(WebKit::VideoFullscreenModelContext::VideoFullscreenModelContext):
(WebKit::VideoFullscreenModelContext::addClient):
(WebKit::VideoFullscreenModelContext::removeClient):
(WebKit::VideoFullscreenModelContext::willEnterPictureInPicture):
(WebKit::VideoFullscreenModelContext::didEnterPictureInPicture):
(WebKit::VideoFullscreenModelContext::failedToEnterPictureInPicture):
(WebKit::VideoFullscreenModelContext::willExitPictureInPicture):
(WebKit::VideoFullscreenModelContext::didExitPictureInPicture):
(WebKit::VideoFullscreenManagerProxy::invalidate):
(WebKit::VideoFullscreenManagerProxy::createModelAndInterface):
(WebKit::VideoFullscreenManagerProxy::removeClientForContext):
(WebKit::VideoFullscreenModelContext::~VideoFullscreenModelContext): Deleted.
* UIProcess/ios/fullscreen/WKFullScreenViewController.mm:
(WKFullScreenViewControllerPlaybackSessionModelClient::setInterface):
(WKFullScreenViewControllerVideoFullscreenModelClient::setInterface):
(-[WKFullScreenViewController videoControlsManagerDidChange]):
(-[WKFullScreenViewController _togglePiPAction:]):
* UIProcess/mac/WKFullScreenWindowController.mm:
(WebKit::WKFullScreenWindowControllerVideoFullscreenModelClient::setInterface):
* WebProcess/cocoa/PlaybackSessionManager.h:
* WebProcess/cocoa/PlaybackSessionManager.mm:
(WebKit::PlaybackSessionInterfaceContext::PlaybackSessionInterfaceContext):
(WebKit::PlaybackSessionManager::~PlaybackSessionManager):
(WebKit::PlaybackSessionManager::createModelAndInterface):
(WebKit::PlaybackSessionManager::removeContext):
(WebKit::PlaybackSessionInterfaceContext::~PlaybackSessionInterfaceContext): Deleted.
* WebProcess/cocoa/VideoFullscreenManager.h:
(WebKit::VideoFullscreenInterfaceContext::create):
(WebKit::VideoFullscreenInterfaceContext::createWeakPtr):
* WebProcess/cocoa/VideoFullscreenManager.mm:
(WebKit::VideoFullscreenInterfaceContext::VideoFullscreenInterfaceContext):
(WebKit::VideoFullscreenManager::~VideoFullscreenManager):
(WebKit::VideoFullscreenManager::createModelAndInterface):
(WebKit::VideoFullscreenManager::removeContext):

Source/WebKitLegacy/mac:

PlaybackSessionInterface no longer has an invalidate() method.

* WebView/WebView.mm:
(-[WebView _clearPlaybackControlsManager]):

Source/WTF:

* WTF.xcodeproj/project.pbxproj:
* wtf/WeakPtrContainer.h: Added.

LayoutTests:

Test was causing false pass results due to the webkitPresentationMode being correctly set
to "inline" during the close() operation; modify the test to only perform the close()
operation when the presetation mode is changed to "picture-in-picture".

* media/controls/ipad/close-page-with-picture-in-picture-video-assertion-failure.html:
* media/controls/ipad/resources/picture-in-picture.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236605 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[JSC] [Armv7] Add a copy function argument to MacroAssemblerARMv7::link() and pass...
guijemont@igalia.com [Fri, 28 Sep 2018 18:18:14 +0000 (18:18 +0000)]
[JSC] [Armv7] Add a copy function argument to MacroAssemblerARMv7::link() and pass it down to the assembler's linking functions.
https://bugs.webkit.org/show_bug.cgi?id=190080

Reviewed by Mark Lam.

* assembler/ARMv7Assembler.h:
(JSC::ARMv7Assembler::link):
(JSC::ARMv7Assembler::linkJumpT1):
(JSC::ARMv7Assembler::linkJumpT2):
(JSC::ARMv7Assembler::linkJumpT3):
(JSC::ARMv7Assembler::linkJumpT4):
(JSC::ARMv7Assembler::linkConditionalJumpT4):
(JSC::ARMv7Assembler::linkBX):
(JSC::ARMv7Assembler::linkConditionalBX):
* assembler/MacroAssemblerARMv7.h:
(JSC::MacroAssemblerARMv7::link):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236604 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoDrop iOS specific quirk in SettingsBase::scriptEnabledChanged()
cdumez@apple.com [Fri, 28 Sep 2018 17:04:04 +0000 (17:04 +0000)]
Drop iOS specific quirk in SettingsBase::scriptEnabledChanged()
https://bugs.webkit.org/show_bug.cgi?id=190077
<rdar://problem/44812613>

Reviewed by Zalan Bujtas.

Drop iOS specific quirk in SettingsBase::scriptEnabledChanged() that would dirty style after the
"JavaScriptEnabled" setting's state is toggled. I do not see a good reason to do this given that
scripts would not get executed until a reload.

If we find out after dropping this that this is actually useful for some reason, then we can
always bring it back and consider making this non-iOS specific, as well as documenting why this
it is needed.

* page/Settings.yaml:
* page/SettingsBase.cpp:
(WebCore::SettingsBase::scriptEnabledChanged): Deleted.
* page/SettingsBase.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236600 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WTF] Add ExternalStringImpl, a StringImpl for user controlled buffers
commit-queue@webkit.org [Fri, 28 Sep 2018 15:56:54 +0000 (15:56 +0000)]
[WTF] Add ExternalStringImpl, a StringImpl for user controlled buffers
https://bugs.webkit.org/show_bug.cgi?id=189991

Patch by Koby Boyango <koby.b@mce.systems> on 2018-09-28
Reviewed by Yusuke Suzuki.

Source/WTF:

* WTF.xcodeproj/project.pbxproj:
* wtf/CMakeLists.txt:
* wtf/text/ExternalStringImpl.cpp: Added.
* wtf/text/ExternalStringImpl.h: Added.
* wtf/text/StringImpl.cpp:
* wtf/text/StringImpl.h:

Tools:

* TestWebKitAPI/Tests/WTF/StringImpl.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236599 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoREGRESSION (r236573): [iOS] Layout test editing/pasteboard/emacs-ctrl-a-k-y.html...
cdumez@apple.com [Fri, 28 Sep 2018 15:43:17 +0000 (15:43 +0000)]
REGRESSION (r236573): [iOS] Layout test editing/pasteboard/emacs-ctrl-a-k-y.html is failing
https://bugs.webkit.org/show_bug.cgi?id=190067

Unreviewed, re-land ios-specific expected result for this test as they were inadvertently
dropped in r236573.

* platform/ios-wk2/editing/pasteboard/emacs-ctrl-a-k-y-expected.txt: Added.
* platform/ios/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236598 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoJSC test stress/jsc-read.js doesn't support CRLF
ross.kirsling@sony.com [Fri, 28 Sep 2018 15:31:47 +0000 (15:31 +0000)]
JSC test stress/jsc-read.js doesn't support CRLF
https://bugs.webkit.org/show_bug.cgi?id=190063

Reviewed by Yusuke Suzuki.

In order to run this test via Windows command prompt, we can't assume that the final newline will be LF.

* stress/jsc-read.js:
(test):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236597 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoWeb Inspector: image resources without content are still shown when the Images folder...
drousso@apple.com [Fri, 28 Sep 2018 08:04:10 +0000 (08:04 +0000)]
Web Inspector: image resources without content are still shown when the Images folder is selected
https://bugs.webkit.org/show_bug.cgi?id=190072

Reviewed by Matt Baker.

* UserInterface/Views/ResourceContentView.js:
(WI.ResourceContentView.prototype.showGenericNoContentMessage):
Treat paths that result in a "Resource has no content" message as a "content error".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236596 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[WinCairo] error C2027: use of undefined type 'WTF::EnumTraits<E>' while compiling...
Hironori.Fujii@sony.com [Fri, 28 Sep 2018 07:06:04 +0000 (07:06 +0000)]
[WinCairo] error C2027: use of undefined type 'WTF::EnumTraits<E>' while compiling AuthenticationChallengeProxy.cpp
https://bugs.webkit.org/show_bug.cgi?id=190071

Unreviewed build fix for WinCairo port.

AuthenticationChallengeDisposition was not defined in this file.

* UIProcess/Authentication/AuthenticationChallengeProxy.cpp: Include "AuthenticationChallengeDisposition.h".

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236595 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[Curl] Fix priority issue with multiple cookies with different level of path.
Basuke.Suzuki@sony.com [Fri, 28 Sep 2018 06:55:03 +0000 (06:55 +0000)]
[Curl] Fix priority issue with multiple cookies with different level of path.
https://bugs.webkit.org/show_bug.cgi?id=189920

Reviewed by Fujii Hironori.

Source/WebCore:

When multiple cookies are stored in the database for same site, the priority of
multiple cookies which matches path criteria was not defined. The backend
implementation `sqlite` then returns the first matching result, which is the one
stored earlier.

Test: http/tests/cookies/cookie-with-multiple-level-path.html

* platform/network/curl/CookieJarDB.cpp:
(WebCore::CookieJarDB::searchCookies):

LayoutTests:

* http/tests/cookies/cookie-with-multiple-level-path-expected.txt: Added.
* http/tests/cookies/cookie-with-multiple-level-path.html: Added.
* http/tests/cookies/resources/cookie-with-multiple-level-path.php: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236593 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUpdate flakiness dashboard configuration for Mojave queues
ryanhaddad@apple.com [Fri, 28 Sep 2018 05:47:59 +0000 (05:47 +0000)]
Update flakiness dashboard configuration for Mojave queues
https://bugs.webkit.org/show_bug.cgi?id=190068

Reviewed by Alexey Proskuryakov.

* TestResultServer/static-dashboards/builders.jsonp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236590 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoVerify the contents of AssemblerBuffer on arm64e
sbarati@apple.com [Fri, 28 Sep 2018 05:34:38 +0000 (05:34 +0000)]
Verify the contents of AssemblerBuffer on arm64e
https://bugs.webkit.org/show_bug.cgi?id=190057
<rdar://problem/38916630>

Reviewed by Mark Lam.

JSTests:

* stress/regress-189132.js:

Source/JavaScriptCore:

* assembler/ARM64Assembler.h:
(JSC::ARM64Assembler::ARM64Assembler):
(JSC::ARM64Assembler::fillNops):
(JSC::ARM64Assembler::link):
(JSC::ARM64Assembler::linkJumpOrCall):
(JSC::ARM64Assembler::linkCompareAndBranch):
(JSC::ARM64Assembler::linkConditionalBranch):
(JSC::ARM64Assembler::linkTestAndBranch):
(JSC::ARM64Assembler::unlinkedCode): Deleted.
* assembler/ARMAssembler.h:
(JSC::ARMAssembler::fillNops):
* assembler/ARMv7Assembler.h:
(JSC::ARMv7Assembler::unlinkedCode): Deleted.
* assembler/AbstractMacroAssembler.h:
(JSC::AbstractMacroAssembler::emitNops):
(JSC::AbstractMacroAssembler::AbstractMacroAssembler):
* assembler/AssemblerBuffer.h:
(JSC::ARM64EHash::ARM64EHash):
(JSC::ARM64EHash::update):
(JSC::ARM64EHash::hash const):
(JSC::ARM64EHash::randomSeed const):
(JSC::AssemblerBuffer::AssemblerBuffer):
(JSC::AssemblerBuffer::putShort):
(JSC::AssemblerBuffer::putIntUnchecked):
(JSC::AssemblerBuffer::putInt):
(JSC::AssemblerBuffer::hash const):
(JSC::AssemblerBuffer::data const):
(JSC::AssemblerBuffer::putIntegralUnchecked):
(JSC::AssemblerBuffer::append): Deleted.
* assembler/LinkBuffer.cpp:
(JSC::LinkBuffer::copyCompactAndLinkCode):
* assembler/MIPSAssembler.h:
(JSC::MIPSAssembler::fillNops):
* assembler/MacroAssemblerARM64.h:
(JSC::MacroAssemblerARM64::jumpsToLink):
(JSC::MacroAssemblerARM64::link):
(JSC::MacroAssemblerARM64::unlinkedCode): Deleted.
* assembler/MacroAssemblerARMv7.h:
(JSC::MacroAssemblerARMv7::jumpsToLink):
(JSC::MacroAssemblerARMv7::unlinkedCode): Deleted.
* assembler/X86Assembler.h:
(JSC::X86Assembler::fillNops):

Source/WTF:

* wtf/PtrTag.h:
(WTF::tagInt):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236589 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoByValInfo should not use integer offsets.
mark.lam@apple.com [Fri, 28 Sep 2018 05:07:03 +0000 (05:07 +0000)]
ByValInfo should not use integer offsets.
https://bugs.webkit.org/show_bug.cgi?id=190070
<rdar://problem/44803430>

Reviewed by Saam Barati.

Also moved some fields around to allow the ByValInfo struct to be more densely packed.

* bytecode/ByValInfo.h:
(JSC::ByValInfo::ByValInfo):
* jit/JIT.cpp:
(JSC::JIT::link):
* jit/JITOpcodes.cpp:
(JSC::JIT::privateCompileHasIndexedProperty):
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::privateCompileHasIndexedProperty):
* jit/JITPropertyAccess.cpp:
(JSC::JIT::privateCompileGetByVal):
(JSC::JIT::privateCompileGetByValWithCachedId):
(JSC::JIT::privateCompilePutByVal):
(JSC::JIT::privateCompilePutByValWithCachedId):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236587 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoDFG::OSRExit::m_patchableCodeOffset should not be an int
sbarati@apple.com [Fri, 28 Sep 2018 04:46:50 +0000 (04:46 +0000)]
DFG::OSRExit::m_patchableCodeOffset should not be an int
https://bugs.webkit.org/show_bug.cgi?id=190066
<rdar://problem/39498244>

Reviewed by Mark Lam.

* dfg/DFGJITCompiler.cpp:
(JSC::DFG::JITCompiler::linkOSRExits):
(JSC::DFG::JITCompiler::link):
* dfg/DFGOSRExit.cpp:
(JSC::DFG::OSRExit::codeLocationForRepatch const):
(JSC::DFG::OSRExit::compileOSRExit):
(JSC::DFG::OSRExit::setPatchableCodeOffset): Deleted.
(JSC::DFG::OSRExit::getPatchableCodeOffsetAsJump const): Deleted.
(JSC::DFG::OSRExit::correctJump): Deleted.
* dfg/DFGOSRExit.h:
* dfg/DFGOSRExitCompilationInfo.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236585 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoDon't use int offsets in StructureStubInfo
sbarati@apple.com [Fri, 28 Sep 2018 04:27:10 +0000 (04:27 +0000)]
Don't use int offsets in StructureStubInfo
https://bugs.webkit.org/show_bug.cgi?id=190064
<rdar://problem/44784719>

Reviewed by Mark Lam.

* bytecode/InlineAccess.cpp:
(JSC::linkCodeInline):
* bytecode/StructureStubInfo.h:
(JSC::StructureStubInfo::slowPathCallLocation):
(JSC::StructureStubInfo::doneLocation):
(JSC::StructureStubInfo::slowPathStartLocation):
* jit/JITInlineCacheGenerator.cpp:
(JSC::JITInlineCacheGenerator::finalize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236584 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoUse enum class in createMarkup arguments
rniwa@webkit.org [Fri, 28 Sep 2018 04:22:50 +0000 (04:22 +0000)]
Use enum class in createMarkup arguments
https://bugs.webkit.org/show_bug.cgi?id=190028

Reviewed by Wenson Hsieh.

Source/WebCore:

Replaced enums used by createMarkup with equivalent enum classes: EChildrenOnly with SerializedNodes,
EAbsoluteURLs with ResolveURLs, and EFragmentSerialization with SerializationSyntax.

Also replaced the boolean convertBlocksToInlines with an enum class of the same name.

Finally, renamed the createMarkup variant which doesn't serialize style and used for innerHTML and XMLSerializer
to serializeFragment.

* dom/Element.cpp:
(WebCore::Element::innerHTML const):
(WebCore::Element::outerHTML const):
* dom/ShadowRoot.cpp:
(WebCore::ShadowRoot::innerHTML const):
* editing/CompositeEditCommand.cpp:
(WebCore::CompositeEditCommand::moveParagraphs):
* editing/HTMLInterchange.h:
(WebCore::AnnotateForInterchange): Renamed from EAnnotateForInterchange.
* editing/MarkupAccumulator.cpp:
(WebCore::MarkupAccumulator::MarkupAccumulator):
(WebCore::MarkupAccumulator::serializeNodes):
(WebCore::MarkupAccumulator::serializeNodesWithNamespaces):
(WebCore::MarkupAccumulator::resolveURLIfNeeded const):
* editing/MarkupAccumulator.h:
(WebCore::MarkupAccumulator::inXMLFragmentSerialization const):
* editing/ReplaceRangeWithTextCommand.cpp:
(WebCore::ReplaceRangeWithTextCommand::inputEventDataTransfer const):
* editing/ReplaceSelectionCommand.cpp:
(WebCore::ReplaceSelectionCommand::willApplyCommand):
* editing/SpellingCorrectionCommand.cpp:
(WebCore::SpellingCorrectionCommand::inputEventDataTransfer const):
* editing/cocoa/EditorCocoa.mm:
(WebCore::Editor::selectionInHTMLFormat):
* editing/cocoa/WebContentReaderCocoa.mm:
(WebCore::WebContentMarkupReader::readRTFD):
(WebCore::WebContentMarkupReader::readRTF):
* editing/gtk/EditorGtk.cpp:
(WebCore::Editor::writeImageToPasteboard):
(WebCore::Editor::writeSelectionToPasteboard):
* editing/markup.cpp:
(WebCore::StyledMarkupAccumulator::StyledMarkupAccumulator):
(WebCore::StyledMarkupAccumulator::serializeNodes):
(WebCore::StyledMarkupAccumulator::traverseNodesForSerialization):
(WebCore::highestAncestorToWrapMarkup):
(WebCore::createMarkupInternal):
(WebCore::createMarkup):
(WebCore::sanitizedMarkupForFragmentInDocument):
(WebCore::serializeFragment): Renamed from createMarkup. This is used for innerHTML and XMLSerializer,
which faithfully serializes the fragment without any computed style as inline styles.
(WebCore::documentTypeString):
(WebCore::createFullMarkup): Deleted two varinats used in WebKitLegacy.
* editing/markup.h:
(WebCore::ResolveURLs): Renamed from EAbsoluteURLs.
(WebCore::ConvertBlocksToInlines): Added.
(WebCore::SerializedNodes): Renamed from EChildrenOnly.
(WebCore::SerializationSyntax): Renamed from EFragmentSerialization.
* editing/wpe/EditorWPE.cpp:
(WebCore::Editor::writeSelectionToPasteboard):
* inspector/DOMEditor.cpp:
* inspector/agents/InspectorDOMAgent.cpp:
(WebCore::InspectorDOMAgent::getOuterHTML):
* loader/archive/cf/LegacyWebArchive.cpp:
(WebCore::LegacyWebArchive::create):
(WebCore::LegacyWebArchive::createFromSelection):
* page/PageSerializer.cpp:
(WebCore::PageSerializer::SerializerMarkupAccumulator::SerializerMarkupAccumulator):
(WebCore::PageSerializer::serializeFrame):
* page/win/DragControllerWin.cpp:
(WebCore::DragController::declareAndWriteDragImage):
* Source/WebCore/platform/win/PasteboardWin.cpp:
(WebCore::Pasteboard::writeRangeToDataObject):
(WebCore::Pasteboard::writeSelection):
* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::send):
* xml/XMLSerializer.cpp:
(WebCore::XMLSerializer::serializeToString):
* xml/XSLTProcessorLibxslt.cpp:
(WebCore::xsltStylesheetPointer):
(WebCore::xmlDocPtrFromNode):

Source/WebKit:

* WebProcess/WebCoreSupport/gtk/WebEditorClientGtk.cpp:
(WebKit::WebEditorClient::updateGlobalSelection):

Source/WebKitLegacy/mac:

* DOM/WebDOMOperations.mm:
(-[DOMNode markupString]): Moved the code from WebCore/editing/markup.cpp
(-[DOMRange markupString]): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236583 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months ago[iOS] Layout test editing/pasteboard/emacs-ctrl-a-k-y.html is failing
ryanhaddad@apple.com [Fri, 28 Sep 2018 03:22:04 +0000 (03:22 +0000)]
[iOS] Layout test editing/pasteboard/emacs-ctrl-a-k-y.html is failing
https://bugs.webkit.org/show_bug.cgi?id=190067

Unreviewed test gardening.

* platform/ios/TestExpectations: Mark test as failing.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236582 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoBring up queues for Mojave
ryanhaddad@apple.com [Fri, 28 Sep 2018 01:34:36 +0000 (01:34 +0000)]
Bring up queues for Mojave
https://bugs.webkit.org/show_bug.cgi?id=189935

Unreviewed infrastructure fix.

* BuildSlaveSupport/build.webkit.org-config/config.json: Remove trigger, update bot assignment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236581 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoBring up queues for Mojave
ryanhaddad@apple.com [Fri, 28 Sep 2018 01:25:35 +0000 (01:25 +0000)]
Bring up queues for Mojave
https://bugs.webkit.org/show_bug.cgi?id=189935

Reviewed by Alexey Proskuryakov.

* BuildSlaveSupport/build.webkit.org-config/config.json:
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/WebKitBuildbot.js:
(WebKitBuildbot):
* BuildSlaveSupport/build.webkit.org-config/steps_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236580 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoIncrease the timeout for iOS Simulator data migration
ryanhaddad@apple.com [Fri, 28 Sep 2018 00:50:28 +0000 (00:50 +0000)]
Increase the timeout for iOS Simulator data migration
https://bugs.webkit.org/show_bug.cgi?id=190059

Reviewed by Aakash Jain.

3 minutes isn't always enough time for the data migrator to complete
when booting up multiple iOS Simulators. Change the timeout to 10 minutes.

* Scripts/webkitpy/xcode/simulated_device.py:
(SimulatedDeviceManager): Create a constant for the default timeout.
(SimulatedDeviceManager.initialize_devices): Use new constant.
(SimulatedDeviceManager.swap): Ditto.
(SimulatedDeviceManager.wait_until_data_migration_is_done): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236579 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12 months agoResource Load Statistics: Non-redirected top frame navigation should not get captured...
wilander@apple.com [Fri, 28 Sep 2018 00:23:37 +0000 (00:23 +0000)]
Resource Load Statistics: Non-redirected top frame navigation should not get captured in statistics
https://bugs.webkit.org/show_bug.cgi?id=190055
<rdar://problem/44843460>

Reviewed by Chris Dumez.

Source/WebCore:

Test: http/tests/resourceLoadStatistics/do-not-capture-statistics-for-simple-top-navigations.html

* loader/ResourceLoadStatistics.cpp:
(WebCore::ResourceLoadStatistics::decode):
    Corrects legacy statistics for frames and triggers a re-classification.

Source/WebKit:

* UIProcess/ResourceLoadStatisticsMemoryStore.cpp:
    Bumped the statisticsModelVersion to 14 to be able to
    correct legacy statistics.
(WebKit::ResourceLoadStatisticsMemoryStore::logFrameNavigation):
    Now skips capture if it's the main frame.

LayoutTests:

* http/tests/resourceLoadStatistics/do-not-capture-statistics-for-simple-top-navigations-expected.txt: Added.
* http/tests/resourceLoadStatistics/do-not-capture-statistics-for-simple-top-navigations.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236578 268f45cc-cd09-0410-ab3c-d52691b4dbfc