WebKit-https.git
3 years ago[SOUP] Remove unused methods toSoupMessage from ResourceRequest/Response
commit-queue@webkit.org [Mon, 3 Oct 2016 19:13:20 +0000 (19:13 +0000)]
[SOUP] Remove unused methods toSoupMessage from ResourceRequest/Response
https://bugs.webkit.org/show_bug.cgi?id=162867

Patch by Carlos Garcia Campos <cgarcia@igalia.com> on 2016-10-03
Reviewed by Sergio Villar Senin.

* platform/network/soup/ResourceRequest.h:
* platform/network/soup/ResourceRequestSoup.cpp:
(WebCore::ResourceRequest::updateSoupMessage):
(WebCore::ResourceRequest::updateFromSoupMessage): Deleted.
* platform/network/soup/ResourceResponse.h:
* platform/network/soup/ResourceResponseSoup.cpp:
(WebCore::ResourceResponse::updateFromSoupMessage): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoB3 trapping memory accesses should be documented
fpizlo@apple.com [Mon, 3 Oct 2016 18:36:55 +0000 (18:36 +0000)]
B3 trapping memory accesses should be documented
https://bugs.webkit.org/show_bug.cgi?id=162845

Reviewed by Geoffrey Garen.

Source/JavaScriptCore:

While writing some documentation, I found some small holes in the code.

* b3/B3Effects.cpp:
(JSC::B3::Effects::operator==): Need this to write tests.
(JSC::B3::Effects::operator!=): Need this to write tests.
* b3/B3Effects.h:
* b3/B3HeapRange.h:
* b3/B3MemoryValue.cpp:
(JSC::B3::MemoryValue::dumpMeta): Sometimes the heap range dump won't show you the memory value's actual range. This makes the dump show you the actual range in that case.
* b3/B3Value.cpp:
(JSC::B3::Value::effects): While documenting this, I remembered that trapping also has to imply reading top. I fixed this.
* b3/testb3.cpp:
(JSC::B3::testTrappingLoad): Added checks for the effects of trapping loads.
(JSC::B3::testTrappingStore): Added checks for the effects of trapping stores.
(JSC::B3::testMoveConstants): Made this not crash with validation.

Websites/webkit.org:

Added documentation for the Traps flag, and factored out the documentation of the Chill flag
to a new flags section.

* docs/b3/intermediate-representation.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[ES6] GeneratorFunction (a.k.a. GeneratorWrapperFunction)'s prototype object does...
utatane.tea@gmail.com [Mon, 3 Oct 2016 18:36:12 +0000 (18:36 +0000)]
[ES6] GeneratorFunction (a.k.a. GeneratorWrapperFunction)'s prototype object does not have constructor property
https://bugs.webkit.org/show_bug.cgi?id=162849

Reviewed by Geoffrey Garen.

JSTests:

* test262.yaml:

Source/JavaScriptCore:

Since GeneratorFunction is not constructible, GeneratorFunction.prototype does not have "constructor" property.

    function* generatorFunction() { }
    generatorFunction.prototype.constructor // undefined

* runtime/JSFunction.cpp:
(JSC::JSFunction::getOwnPropertySlot):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove FetchBody::m_isEmpty
commit-queue@webkit.org [Mon, 3 Oct 2016 17:57:27 +0000 (17:57 +0000)]
Remove FetchBody::m_isEmpty
https://bugs.webkit.org/show_bug.cgi?id=162857

Patch by Youenn Fablet <youenn@apple.com> on 2016-10-03
Reviewed by Alex Christensen.

No change of behavior.

Moving handling of null bodies in FetchBodyOwner, by having an Optional<FetchBody> in FetchBodyOwner.
Moving storage of m_contentType in FetchBodyOwner.

Moving header storage from FetchResponse/FetchRequest to FetchBodyOwner, as it helps updating m_contentType.

* Modules/fetch/FetchBody.cpp: removing m_isEmpty and m_contentType.
(WebCore::FetchBody::extract): Computes the default content-type which is stored in FetchBodyOwner.
(WebCore::FetchBody::blob):
(WebCore::FetchBody::consume):
(WebCore::FetchBody::consumeAsStream):
(WebCore::FetchBody::loadingSucceeded):
(WebCore::FetchBody::bodyForInternalRequest):
(WebCore::FetchBody::clone):
* Modules/fetch/FetchBody.h:
(WebCore::FetchBody::loadingBody):
(WebCore::FetchBody::FetchBody):
* Modules/fetch/FetchBodyOwner.cpp:
(WebCore::FetchBodyOwner::FetchBodyOwner):
(WebCore::FetchBodyOwner::stop):
(WebCore::FetchBodyOwner::arrayBuffer):
(WebCore::FetchBodyOwner::blob):
(WebCore::FetchBodyOwner::cloneBody): Needs to clone m_contentType as well.
(WebCore::FetchBodyOwner::extractBody):
(WebCore::FetchBodyOwner::updateContentType):
(WebCore::FetchBodyOwner::consumeOnceLoadingFinished):
(WebCore::FetchBodyOwner::formData):
(WebCore::FetchBodyOwner::json):
(WebCore::FetchBodyOwner::text):
(WebCore::FetchBodyOwner::loadBlob):
(WebCore::FetchBodyOwner::blobLoadingSucceeded):
(WebCore::FetchBodyOwner::blobLoadingFailed):
(WebCore::FetchBodyOwner::finishBlobLoading):
* Modules/fetch/FetchBodyOwner.h:
(WebCore::FetchBodyOwner::body):
(WebCore::FetchBodyOwner::isBodyNull):
* Modules/fetch/FetchRequest.cpp:
(WebCore::FetchRequest::setBody):
(WebCore::FetchRequest::internalRequest):
(WebCore::FetchRequest::clone):
* Modules/fetch/FetchRequest.h:
(WebCore::FetchRequest::FetchRequest):
* Modules/fetch/FetchResponse.cpp:
(WebCore::FetchResponse::initializeWith):
(WebCore::FetchResponse::FetchResponse):
(WebCore::FetchResponse::cloneForJS):
(WebCore::FetchResponse::fetch):
(WebCore::FetchResponse::BodyLoader::didSucceed):
(WebCore::FetchResponse::BodyLoader::didReceiveResponse):
(WebCore::FetchResponse::BodyLoader::start):
(WebCore::FetchResponse::createReadableStreamSource):
* Modules/fetch/FetchResponse.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206737 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoASSERTION FAILED: result in WebCore::CSSParser::parseURI
aestes@apple.com [Mon, 3 Oct 2016 17:44:29 +0000 (17:44 +0000)]
ASSERTION FAILED: result in WebCore::CSSParser::parseURI
https://bugs.webkit.org/show_bug.cgi?id=141638
<rdar://problem/27709952>

Reviewed by Andreas Kling.

Source/WebCore:

CSSParser::parseURIInternal() failed to parse unquoted URLs with Unicode escape sequences
greater than 0xFF, even when the destination character type was multi-byte. Fixed by
checking the size of DestCharacterType instead of SrcCharacterType.

Updated fast/css/url-with-multi-byte-unicode-escape.html to test for an unquoted URL.

* css/parser/CSSParser.cpp:
(WebCore::CSSParser::parseURIInternal): For code points greater than 0xFF, only returned
false if sizeof(DestCharacterType) == 1.

LayoutTests:

* fast/css/url-with-multi-byte-unicode-escape-expected.txt: Added a test for an unquoted URL.
* fast/css/url-with-multi-byte-unicode-escape.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoURLParser: empty relative URLs should not copy fragment from the base URL
achristensen@apple.com [Mon, 3 Oct 2016 17:28:04 +0000 (17:28 +0000)]
URLParser: empty relative URLs should not copy fragment from the base URL
https://bugs.webkit.org/show_bug.cgi?id=162864

Reviewed by Chris Dumez.

Source/WebCore:

Covered by new API tests.

* platform/URL.cpp:
(WebCore::URL::removeFragmentIdentifier):
Optimize removing fragments, now that it happens more often. We don't need to reparse,
because the result will always be equal to just a substring when removing the fragment at the end.
* platform/URLParser.cpp:
(WebCore::URLParser::copyASCIIStringUntil):
(WebCore::URLParser::copyURLPartsUntil):
(WebCore::containsOnlyC0ControlOrSpace):
(WebCore::URLParser::URLParser):
(WebCore::URLParser::parse):
* platform/URLParser.h:
Because we are not copying the fragment, we can simplify and remove some unreachable code.

Tools:

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoJSStringRef should define JSChar without platform checks
commit-queue@webkit.org [Mon, 3 Oct 2016 16:43:03 +0000 (16:43 +0000)]
JSStringRef should define JSChar without platform checks
https://bugs.webkit.org/show_bug.cgi?id=162808

Patch by Nicolas Breidinger <Nicolas.Breidinger@sony.com> on 2016-10-03
Reviewed by Mark Lam.

* API/JSStringRef.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agotd.scope should only return known values
cdumez@apple.com [Mon, 3 Oct 2016 16:23:24 +0000 (16:23 +0000)]
td.scope should only return known values
https://bugs.webkit.org/show_bug.cgi?id=162755

Reviewed by Sam Weinig.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

* web-platform-tests/html/dom/reflection-tabular-expected.txt:

Source/WebCore:

td.scope should only return known values:
- https://html.spec.whatwg.org/#dom-th-scope

Firefox conforms to the specification.

No new tests, rebaselined existing test.

* html/HTMLTableCellElement.cpp:
(WebCore::HTMLTableCellElement::scope):

LayoutTests:

Update existing test to reflect behavior change.

* fast/dom/element-attribute-js-null-expected.txt:
* fast/dom/element-attribute-js-null.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[SOUP] Simplify AuthenticationChallenge
carlosgc@webkit.org [Mon, 3 Oct 2016 10:42:11 +0000 (10:42 +0000)]
[SOUP] Simplify AuthenticationChallenge
https://bugs.webkit.org/show_bug.cgi?id=162784

Reviewed by Alex Christensen.

We don't really need to keep references to the SoupSession and SoupMessage in AuthenticationChallenge, the
SoupNetworkSession callback already forwards the challenge to the right ResourceHandle.

* platform/network/soup/AuthenticationChallenge.h:
(WebCore::AuthenticationChallenge::authenticationClient):
(WebCore::AuthenticationChallenge::soupSession): Deleted.
(WebCore::AuthenticationChallenge::soupMessage): Deleted.
* platform/network/soup/AuthenticationChallengeSoup.cpp:
(WebCore::AuthenticationChallenge::AuthenticationChallenge):
(WebCore::AuthenticationChallenge::platformCompare):
* platform/network/soup/ResourceHandleSoup.cpp:
(WebCore::ResourceHandle::continueDidReceiveAuthenticationChallenge):
(WebCore::ResourceHandle::didReceiveAuthenticationChallenge):
(WebCore::ResourceHandle::receivedRequestToContinueWithoutCredential):
(WebCore::ResourceHandle::receivedCredential):
(WebCore::ResourceHandle::receivedCancellation):
* platform/network/soup/SoupNetworkSession.cpp:
(WebCore::authenticateCallback):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[SOUP] Cleanup persistent credential storage code
carlosgc@webkit.org [Mon, 3 Oct 2016 10:32:52 +0000 (10:32 +0000)]
[SOUP] Cleanup persistent credential storage code
https://bugs.webkit.org/show_bug.cgi?id=162777

Reviewed by Alex Christensen.

.:

Remove ENABLE(CREDENTIAL_STORAGE) build flag and replace it by USE(LIBSECRET).

* Source/cmake/OptionsGTK.cmake:
* Source/cmake/WebKitFeatures.cmake:

Source/WebCore:

We have this feature behind ENABLE_CREDENTIAL_STORAGE flag, which is confusing, because we use credential
storage unconditionally and this is only about persistent storage. Also the flag assumes libsecret is available,
and since it's only used by GTK sometimes we use GTK ifdefs instead of CREDENTIAL_STORAGE. So, I think we should
use USE(LIBSECRET) instead, and reduce a bit the ifdefs in common soup code. Another problem is that current
implementation is always used, while it should depend on the current network storage session and never used in
ephemeral sessions. This patch moves the code from CredentialBackingStore to NetworkStorageSessionSoup and
modernizes a bit.

* PlatformGTK.cmake: Remove CredentialBackingStore.cpp.
* platform/gtk/GRefPtrGtk.cpp: Use USE(LIBSECRET)
* platform/gtk/GRefPtrGtk.h:
* platform/network/NetworkStorageSession.h:
* platform/network/ResourceHandleInternal.h:
* platform/network/gtk/CredentialBackingStore.cpp: Removed.
* platform/network/gtk/CredentialBackingStore.h: Removed.
* platform/network/soup/NetworkStorageSessionSoup.cpp:
(WebCore::NetworkStorageSession::~NetworkStorageSession):
(WebCore::schemeFromProtectionSpaceServerType):
(WebCore::authTypeFromProtectionSpaceAuthenticationScheme):
(WebCore::NetworkStorageSession::getCredentialFromPersistentStorage):
(WebCore::NetworkStorageSession::saveCredentialToPersistentStorage):
(WebCore::NetworkStorageSession::ensurePrivateBrowsingSession): Deleted.
(WebCore::NetworkStorageSession::switchToNewTestingSession): Deleted.
* platform/network/soup/ResourceHandleSoup.cpp:
(WebCore::gotHeadersCallback):
(WebCore::ResourceHandle::didReceiveAuthenticationChallenge):
(WebCore::ResourceHandle::receivedCredential):
(WebCore::ResourceHandle::continueDidReceiveAuthenticationChallenge): Deleted.
(WebCore::ResourceHandle::receivedRequestToContinueWithoutCredential): Deleted.
(WebCore::ResourceHandle::receivedCancellation): Deleted.

Source/WebKit2:

Use USE(LIBSECRET) instead of ENABLE(CREDENTIAL_STORAGE).

* UIProcess/API/gtk/WebKitAuthenticationRequest.cpp:
(webkit_authentication_request_can_save_credentials):

Tools:

Use USE(LIBSECRET) instead of ENABLE(CREDENTIAL_STORAGE).

* TestWebKitAPI/Tests/WebKit2Gtk/TestAuthentication.cpp:
(testWebViewAuthenticationStorage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed. Fix the build with coordinated graphics enabled after r206712.
carlosgc@webkit.org [Mon, 3 Oct 2016 09:51:18 +0000 (09:51 +0000)]
Unreviewed. Fix the build with coordinated graphics enabled after r206712.

* page/scrolling/ScrollingStateTree.cpp:
(WebCore::ScrollingStateTree::attachNode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Clicking twice on the color swatch square should hide the color picker
commit-queue@webkit.org [Mon, 3 Oct 2016 05:45:15 +0000 (05:45 +0000)]
Web Inspector: Clicking twice on the color swatch square should hide the color picker
https://bugs.webkit.org/show_bug.cgi?id=162759

Patch by Devin Rousso <dcrousso+webkit@gmail.com> on 2016-10-02
Reviewed by Matt Baker.

* UserInterface/Views/InlineSwatch.js:
(WebInspector.InlineSwatch):
(WebInspector.InlineSwatch.prototype.didDismissPopover):
(WebInspector.InlineSwatch.prototype._swatchElementClicked):
Remove the "click" event listener when a popover is presented to prevent improper
interaction.  Add the event listener back when the popover is dismissed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdd support for KeyboardEvent.getModifierState() operation
cdumez@apple.com [Sun, 2 Oct 2016 21:49:44 +0000 (21:49 +0000)]
Add support for KeyboardEvent.getModifierState() operation
https://bugs.webkit.org/show_bug.cgi?id=162855

Reviewed by Darin Adler.

Source/WebCore:

Add support for KeyboardEvent.getModifierState() operation:
- https://w3c.github.io/uievents/#dom-keyboardevent-getmodifierstate

Test: fast/events/constructors/keyboard-event-getModifierState.html

* dom/KeyboardEvent.cpp:
(WebCore::KeyboardEvent::getModifierState):
* dom/KeyboardEvent.idl:

LayoutTests:

Add layout test coverage.

* fast/events/constructors/keyboard-event-getModifierState-expected.txt: Added.
* fast/events/constructors/keyboard-event-getModifierState.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdd support for KeyboardEvent.repeat attribute
cdumez@apple.com [Sun, 2 Oct 2016 21:33:28 +0000 (21:33 +0000)]
Add support for KeyboardEvent.repeat attribute
https://bugs.webkit.org/show_bug.cgi?id=162854

Reviewed by Darin Adler.

Source/WebCore:

Add support for KeyboardEvent.repeat attribute:
- https://w3c.github.io/uievents/#dom-keyboardevent-repeat

No new tests, extended existing test.

* dom/KeyboardEvent.cpp:
(WebCore::KeyboardEvent::KeyboardEvent):
* dom/KeyboardEvent.h:
* dom/KeyboardEvent.idl:

LayoutTests:

Extend layout test coverage.

* fast/events/constructors/keyboard-event-constructor-expected.txt:
* fast/events/constructors/keyboard-event-constructor.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRename ExceptionCode-based exception handling to "legacy"
darin@apple.com [Sun, 2 Oct 2016 21:28:31 +0000 (21:28 +0000)]
Rename ExceptionCode-based exception handling to "legacy"
https://bugs.webkit.org/show_bug.cgi?id=162859

Reviewed by Chris Dumez.

Source/WebCore:

RaisesException is now MayThrowLegacyException.

A follow-up patch soon will add MayThrowException for the new style,
where we use ExceptionOr<> as the return type instead of an ExceptionCode
out argument.

* bindings/scripts/CodeGeneratorJS.pm:
Removed the quotation marks from the syntax object->{PropertyName} so the
renaming script could handle it. Tweaked various comments and logic for clarity.
Changed to use the perl "x if y" syntax for brevity in many places.
Renamed with the script and by hand. Added code to skip including the same header
in both the implementation file and the header, just to make generated bindings
slightly smaller and faster to compile. Used "using" instead of "typedef".

* bindings/scripts/IDLAttributes.txt: Updated from old names to new ones.

* Modules/applepay/ApplePaySession.idl:
* Modules/encryptedmedia/MediaKeySession.idl:
* Modules/encryptedmedia/MediaKeys.idl:
* Modules/fetch/FetchHeaders.idl:
* Modules/fetch/FetchRequest.idl:
* Modules/fetch/FetchResponse.idl:
* Modules/indexeddb/IDBCursor.idl:
* Modules/indexeddb/IDBDatabase.idl:
* Modules/indexeddb/IDBFactory.idl:
* Modules/indexeddb/IDBIndex.idl:
* Modules/indexeddb/IDBKeyRange.idl:
* Modules/indexeddb/IDBObjectStore.idl:
* Modules/indexeddb/IDBRequest.idl:
* Modules/indexeddb/IDBTransaction.idl:
* Modules/mediasource/MediaSource.idl:
* Modules/mediasource/SourceBuffer.idl:
* Modules/mediastream/MediaDevices.idl:
* Modules/mediastream/RTCDTMFSender.idl:
* Modules/mediastream/RTCDataChannel.idl:
* Modules/mediastream/RTCIceCandidate.idl:
* Modules/mediastream/RTCPeerConnection.idl:
* Modules/mediastream/RTCRtpSender.idl:
* Modules/mediastream/RTCSessionDescription.idl:
* Modules/navigatorcontentutils/NavigatorContentUtils.idl:
* Modules/notifications/NotificationCenter.idl:
* Modules/webaudio/AnalyserNode.idl:
* Modules/webaudio/AudioBuffer.idl:
* Modules/webaudio/AudioBufferSourceNode.idl:
* Modules/webaudio/AudioContext.idl:
* Modules/webaudio/AudioNode.idl:
* Modules/webaudio/ConvolverNode.idl:
* Modules/webaudio/OfflineAudioContext.idl:
* Modules/webaudio/OscillatorNode.idl:
* Modules/webdatabase/DOMWindowWebDatabase.idl:
* Modules/webdatabase/SQLResultSet.idl:
* Modules/websockets/WebSocket.idl:
* bindings/scripts/test/TestImplements.idl:
* bindings/scripts/test/TestInterface.idl:
* bindings/scripts/test/TestNamedConstructor.idl:
* bindings/scripts/test/TestNondeterministic.idl:
* bindings/scripts/test/TestObj.idl:
* bindings/scripts/test/TestSupplemental.idl:
* bindings/scripts/test/TestTypedefs.idl:
* css/CSSCharsetRule.idl:
* css/CSSMediaRule.idl:
* css/CSSPrimitiveValue.idl:
* css/CSSRule.idl:
* css/CSSStyleDeclaration.idl:
* css/CSSStyleSheet.idl:
* css/CSSSupportsRule.idl:
* css/CSSValue.idl:
* css/FontFace.idl:
* css/FontFaceSet.idl:
* css/MediaList.idl:
* css/WebKitCSSMatrix.idl:
* dom/CharacterData.idl:
* dom/ChildNode.idl:
* dom/CustomElementRegistry.idl:
* dom/DataTransferItemList.idl:
* dom/Document.idl:
* dom/Element.idl:
* dom/EventTarget.idl:
* dom/MessagePort.idl:
* dom/MutationObserver.idl:
* dom/NamedNodeMap.idl:
* dom/Node.idl:
* dom/ParentNode.idl:
* dom/Range.idl:
* dom/ShadowRoot.idl:
* dom/Text.idl:
* fileapi/FileReader.idl:
* fileapi/FileReaderSync.idl:
* html/DOMTokenList.idl:
* html/DOMURL.idl:
* html/HTMLCanvasElement.idl:
* html/HTMLElement.idl:
* html/HTMLEmbedElement.idl:
* html/HTMLFrameElement.idl:
* html/HTMLIFrameElement.idl:
* html/HTMLInputElement.idl:
* html/HTMLMarqueeElement.idl:
* html/HTMLMediaElement.idl:
* html/HTMLObjectElement.idl:
* html/HTMLOptionElement.idl:
* html/HTMLOptionsCollection.idl:
* html/HTMLProgressElement.idl:
* html/HTMLSelectElement.idl:
* html/HTMLTableElement.idl:
* html/HTMLTableRowElement.idl:
* html/HTMLTableSectionElement.idl:
* html/HTMLTextAreaElement.idl:
* html/HTMLVideoElement.idl:
* html/ImageData.idl:
* html/MediaController.idl:
* html/TimeRanges.idl:
* html/URLUtils.idl:
* html/canvas/CanvasGradient.idl:
* html/canvas/CanvasPath.idl:
* html/canvas/CanvasRenderingContext2D.idl:
* html/canvas/OESVertexArrayObject.idl:
* html/canvas/WebGLRenderingContextBase.idl:
* html/track/DataCue.idl:
* html/track/TextTrack.idl:
* html/track/TextTrackCue.idl:
* html/track/VTTCue.idl:
* html/track/VTTRegion.idl:
* loader/appcache/DOMApplicationCache.idl:
* page/Crypto.idl:
* page/DOMSelection.idl:
* page/DOMWindow.idl:
* page/EventSource.idl:
* page/History.idl:
* page/Location.idl:
* page/Performance.idl:
* page/UserMessageHandler.idl:
* page/WindowOrWorkerGlobalScope.idl:
* storage/Storage.idl:
* svg/SVGAltGlyphElement.idl:
* svg/SVGAngle.idl:
* svg/SVGAnimatedBoolean.idl:
* svg/SVGAnimatedEnumeration.idl:
* svg/SVGAnimatedInteger.idl:
* svg/SVGAnimatedNumber.idl:
* svg/SVGAnimatedString.idl:
* svg/SVGAnimationElement.idl:
* svg/SVGColor.idl:
* svg/SVGGlyphRefElement.idl:
* svg/SVGGraphicsElement.idl:
* svg/SVGLength.idl:
* svg/SVGLengthList.idl:
* svg/SVGMatrix.idl:
* svg/SVGNumberList.idl:
* svg/SVGPaint.idl:
* svg/SVGPathSegList.idl:
* svg/SVGPointList.idl:
* svg/SVGPreserveAspectRatio.idl:
* svg/SVGStringList.idl:
* svg/SVGStyleElement.idl:
* svg/SVGTextContentElement.idl:
* svg/SVGTransformList.idl:
* svg/SVGViewSpec.idl:
* testing/InternalSettings.idl:
* testing/Internals.idl:
* workers/DedicatedWorkerGlobalScope.idl:
* workers/Worker.idl:
* xml/DOMParser.idl:
* xml/XMLHttpRequest.idl:
* xml/XPathEvaluator.idl:
* xml/XPathExpression.idl:
* xml/XPathResult.idl:
Allowed the rename script to rename the exception-related extended attributes.

* bindings/scripts/test/JS/JSInterfaceName.cpp:
* bindings/scripts/test/JS/JSInterfaceName.h:
* bindings/scripts/test/JS/JSTestActiveDOMObject.cpp:
* bindings/scripts/test/JS/JSTestActiveDOMObject.h:
* bindings/scripts/test/JS/JSTestCallback.cpp:
* bindings/scripts/test/JS/JSTestClassWithJSBuiltinConstructor.cpp:
* bindings/scripts/test/JS/JSTestClassWithJSBuiltinConstructor.h:
* bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.cpp:
* bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.h:
* bindings/scripts/test/JS/JSTestCustomNamedGetter.cpp:
* bindings/scripts/test/JS/JSTestCustomNamedGetter.h:
* bindings/scripts/test/JS/JSTestEventConstructor.cpp:
* bindings/scripts/test/JS/JSTestEventConstructor.h:
* bindings/scripts/test/JS/JSTestEventTarget.cpp:
* bindings/scripts/test/JS/JSTestEventTarget.h:
* bindings/scripts/test/JS/JSTestException.cpp:
* bindings/scripts/test/JS/JSTestException.h:
* bindings/scripts/test/JS/JSTestGenerateIsReachable.cpp:
* bindings/scripts/test/JS/JSTestGenerateIsReachable.h:
* bindings/scripts/test/JS/JSTestGlobalObject.cpp:
* bindings/scripts/test/JS/JSTestGlobalObject.h:
* bindings/scripts/test/JS/JSTestInterface.cpp:
* bindings/scripts/test/JS/JSTestInterface.h:
* bindings/scripts/test/JS/JSTestIterable.cpp:
* bindings/scripts/test/JS/JSTestIterable.h:
* bindings/scripts/test/JS/JSTestJSBuiltinConstructor.cpp:
* bindings/scripts/test/JS/JSTestJSBuiltinConstructor.h:
* bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:
* bindings/scripts/test/JS/JSTestMediaQueryListListener.h:
* bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
* bindings/scripts/test/JS/JSTestNamedConstructor.h:
* bindings/scripts/test/JS/JSTestNode.cpp:
* bindings/scripts/test/JS/JSTestNode.h:
* bindings/scripts/test/JS/JSTestNondeterministic.cpp:
* bindings/scripts/test/JS/JSTestNondeterministic.h:
* bindings/scripts/test/JS/JSTestObj.cpp:
* bindings/scripts/test/JS/JSTestObj.h:
* bindings/scripts/test/JS/JSTestOverloadedConstructors.cpp:
* bindings/scripts/test/JS/JSTestOverloadedConstructors.h:
* bindings/scripts/test/JS/JSTestOverloadedConstructorsWithSequence.cpp:
* bindings/scripts/test/JS/JSTestOverloadedConstructorsWithSequence.h:
* bindings/scripts/test/JS/JSTestOverrideBuiltins.cpp:
* bindings/scripts/test/JS/JSTestOverrideBuiltins.h:
* bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:
* bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.h:
* bindings/scripts/test/JS/JSTestTypedefs.cpp:
* bindings/scripts/test/JS/JSTestTypedefs.h:
* bindings/scripts/test/JS/JSattribute.cpp:
* bindings/scripts/test/JS/JSattribute.h:
* bindings/scripts/test/JS/JSreadonly.cpp:
* bindings/scripts/test/JS/JSreadonly.h:
Regenerated.

Source/WebKit/win:

* Interfaces/DOMEvents.idl: Let the script rename here (in a comment).

Tools:

* Scripts/do-webcore-rename: Updated script to do this round of renames.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Exception thrown when hovering network waterfall graphs during reload
mattbaker@apple.com [Sun, 2 Oct 2016 20:59:06 +0000 (20:59 +0000)]
Web Inspector: Exception thrown when hovering network waterfall graphs during reload
https://bugs.webkit.org/show_bug.cgi?id=162850
<rdar://problem/28579653>

Reviewed by Brian Burg.

* UserInterface/Views/ResourceTimelineDataGridNode.js:
(WebInspector.ResourceTimelineDataGridNode.prototype._mouseoverRecordBar):
Check that the node's data grid reference is valid before use.
(WebInspector.ResourceTimelineDataGridNode):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206722 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMedia controls for Soundcloud easily falls out of sync with what's actually playing
wenson_hsieh@apple.com [Sun, 2 Oct 2016 19:22:18 +0000 (19:22 +0000)]
Media controls for Soundcloud easily falls out of sync with what's actually playing
https://bugs.webkit.org/show_bug.cgi?id=162843
<rdar://problem/28176874>

Reviewed by Beth Dakin.

Source/WebCore:

Currently, audio elements are subject to the same main content restrictions as video elements. This is
problematic on sites where audio elements are really the main content, such as soundcloud. We need a different
heuristic for determining whether audio elements such as these should be allowed to show playback controls.

Furthermore, we currently forbid autoplaying audio from showing controls if it has user gesture restrictions
(i.e. it started playing via script). This is, again, problematic on sites where audio elements are played one
after another (such as on soundcloud). Even though a subsequent audio element may have been triggered via
script, we might want still want to show controls for it if the user has explicitly triggered other elements on
the page to begin playing. However, if the user has not gestured on any media in the page, we still should not
show controls for audio elements that begin playing via script.

Adds two new unit tests to TestWebKitAPI simulating audio playback behavior on soundcloud.

* html/MediaElementSession.cpp:
(WebCore::MediaElementSession::removeBehaviorRestriction):
(WebCore::MediaElementSession::canShowControlsManager):
(WebCore::MediaElementSession::allowsPlaybackControlsForAutoplayingAudio):
* html/MediaElementSession.h:
* page/Page.h:
(WebCore::Page::allowsPlaybackControlsForAutoplayingAudio):
(WebCore::Page::setAllowsPlaybackControlsForAutoplayingAudio):

Tools:

Adds two new tests simulating audio playback behavior on soundcloud. Also tweaks some of the WKWebView test
message handling logic to support having multiple message-handler mappings.

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKit2Cocoa/VideoControlsManager.mm:
(TestWebKitAPI::TEST):
* TestWebKitAPI/Tests/WebKit2Cocoa/play-audio-on-click.html: Added.
* TestWebKitAPI/mac/TestWKWebViewMac.h:
* TestWebKitAPI/mac/TestWKWebViewMac.mm:
(-[TestMessageHandler addMessage:withHandler:]):
(-[TestMessageHandler userContentController:didReceiveScriptMessage:]):
(-[TestWKWebView performAfterReceivingMessage:action:]):
(-[TestWKWebView performAfterLoading:]):
(-[TestMessageHandler initWithMessage:handler:]): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r206683.
ryanhaddad@apple.com [Sun, 2 Oct 2016 19:19:29 +0000 (19:19 +0000)]
Unreviewed, rolling out r206683.

This change caused API test WebKit2.FindMatches to fail on
Mac.

Reverted changeset:

"The dragged image should be the current frame only of the
animated image"
https://bugs.webkit.org/show_bug.cgi?id=162109
http://trac.webkit.org/changeset/206683

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r206692 and r206718.
cdumez@apple.com [Sun, 2 Oct 2016 18:19:05 +0000 (18:19 +0000)]
Unreviewed, rolling out r206692 and r206718.

Seems to have caused >200 failures on Sierra

Reverted changesets:

"Implement rendering of font-variation-settings"
https://bugs.webkit.org/show_bug.cgi?id=162782
http://trac.webkit.org/changeset/206692

"Tried to fix the build after r206692."
http://trac.webkit.org/changeset/206718

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoTried to fix the build after r206692.
mitz@apple.com [Sun, 2 Oct 2016 14:49:13 +0000 (14:49 +0000)]
Tried to fix the build after r206692.

* platform/graphics/cocoa/FontPlatformDataCocoa.mm:
(WebCore::cascadeToLastResortAndVariationsFontDescriptor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r206716.
commit-queue@webkit.org [Sun, 2 Oct 2016 14:25:08 +0000 (14:25 +0000)]
Unreviewed, rolling out r206716.
https://bugs.webkit.org/show_bug.cgi?id=162858

It is breaking Mac CMake Debug build (Requested by youenn on
#webkit).

Reverted changeset:

"[Fetch API] Forbid redirection to non-HTTP(s) URL in non-
navigation mode."
https://bugs.webkit.org/show_bug.cgi?id=162785
http://trac.webkit.org/changeset/206716

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206717 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Fetch API] Forbid redirection to non-HTTP(s) URL in non-navigation mode.
commit-queue@webkit.org [Sun, 2 Oct 2016 13:59:18 +0000 (13:59 +0000)]
[Fetch API] Forbid redirection to non-HTTP(s) URL in non-navigation mode.
https://bugs.webkit.org/show_bug.cgi?id=162785

Patch by Youenn Fablet <youenn@apple.com> on 2016-10-02
Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

* web-platform-tests/fetch/api/redirect/redirect-to-dataurl-expected.txt:
* web-platform-tests/fetch/api/redirect/redirect-to-dataurl-worker-expected.txt:

Source/WebCore:

Covered by rebased and existing tests.

Ensuring non-HTTP redirection URLs are not followed at DocumentThreadableLoader level for fetch API only.
This should be applied to all clients at some point, but there is still some uncertainty for data URLs.

Did some refactoring to better separate the case of security checks in case of regular request or redirected request.
This allows in particular to handle more clearly the case of data URLs which are allowed in all modes for regular requests.
But they are not allowed for same-origin redirected requests.

* WebCore.xcodeproj/project.pbxproj:
* loader/DocumentThreadableLoader.cpp:
(WebCore::reportRedirectionWithBadScheme): Reporting bad scheme redirection error.
(WebCore::DocumentThreadableLoader::redirectReceived): Checking that redirection URLs are HTTP(s) in case of Fetch API.
* loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::willSendRequestInternal):
* loader/cache/CachedResourceLoader.cpp:
(WebCore::CachedResourceLoader::requestImage):
(WebCore::CachedResourceLoader::checkInsecureContent):
(WebCore::CachedResourceLoader::allowedByContentSecurityPolicy):
(WebCore::isSameOriginDataURL):
(WebCore::CachedResourceLoader::canRequest):
(WebCore::CachedResourceLoader::canRequestAfterRedirection):
(WebCore::CachedResourceLoader::canRequestInContentDispositionAttachmentSandbox):
(WebCore::CachedResourceLoader::requestResource):
* loader/cache/CachedResourceLoader.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Selection does not show up over execution highlight ranges
commit-queue@webkit.org [Sun, 2 Oct 2016 02:46:37 +0000 (02:46 +0000)]
Web Inspector: Selection does not show up over execution highlight ranges
https://bugs.webkit.org/show_bug.cgi?id=162844
<rdar://problem/28579121>

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2016-10-01
Reviewed by Matt Baker.

* Scripts/update-codemirror-resources.rb:
* UserInterface/External/CodeMirror/mark-selection.js: Added.
* UserInterface/Main.html:
New add-on that makes selection a text marker so it can be styled
at the same level as other text markers.

* UserInterface/Views/TextEditor.css:
(.text-editor > .CodeMirror .execution-range-highlight:not(.CodeMirror-selectedtext)):
Don't use execution-range-highlight styles if the text is selected.

* UserInterface/Views/TextEditor.js:
(WebInspector.TextEditor):
Enable the text selection as text markers addon.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSupport transitions/animations of background-position with right/bottom-relative...
simon.fraser@apple.com [Sun, 2 Oct 2016 01:05:14 +0000 (01:05 +0000)]
Support transitions/animations of background-position with right/bottom-relative values
https://bugs.webkit.org/show_bug.cgi?id=162048

Reviewed by Dean Jackson.
Source/WebCore:

Make transitions between "background-position: 10px 20px" and "background-position: right 10px bottom 20px"
work. We do this by by converting "right 10px" to "calc(100% - 10px)" when blending.

Also improve logging of calculated lengths, and better animation logging for FillLayer properties.

Test: transitions/background-position-transitions.html

* page/animation/CSSPropertyAnimation.cpp:
(WebCore::FillLayerAnimationPropertyWrapperBase::FillLayerAnimationPropertyWrapperBase): Keep the propertyID
around so logging can use it.
(WebCore::FillLayerAnimationPropertyWrapperBase::property):
(WebCore::FillLayerPropertyWrapperGetter::FillLayerPropertyWrapperGetter):
(WebCore::FillLayerPropertyWrapperGetter::value):
(WebCore::FillLayerPropertyWrapper::FillLayerPropertyWrapper):
(WebCore::createCalculatedLength):
(WebCore::FillLayerPositionPropertyWrapper::FillLayerPositionPropertyWrapper):
(WebCore::FillLayerRefCountedPropertyWrapper::FillLayerRefCountedPropertyWrapper):
(WebCore::FillLayerStyleImagePropertyWrapper::FillLayerStyleImagePropertyWrapper):
(WebCore::FillLayersPropertyWrapper::FillLayersPropertyWrapper):
(WebCore::CSSPropertyAnimation::blendProperties): Blend then log, so that the logging
can show the result.
* platform/CalculationValue.cpp:
(WebCore::CalcExpressionNumber::dump):
(WebCore::CalcExpressionBinaryOperation::dump):
(WebCore::CalcExpressionLength::dump):
(WebCore::CalcExpressionBlendLength::dump):
(WebCore::operator<<):
* platform/CalculationValue.h:
* platform/Length.cpp:
(WebCore::operator<<):

LayoutTests:

* transitions/background-position-transitions-expected.txt: Added.
* transitions/background-position-transitions.html: Added.
* transitions/resources/transition-test-helpers.js:
* transitions/svg-transitions-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoBad cast when CSS position programmatically changed from -webkit-sticky to fixed
simon.fraser@apple.com [Sun, 2 Oct 2016 01:05:06 +0000 (01:05 +0000)]
Bad cast when CSS position programmatically changed from -webkit-sticky to fixed
https://bugs.webkit.org/show_bug.cgi?id=160826

Reviewed by Zalan Bujtas.
Source/WebCore:

If a scrolling state tree node changed type (e.g. from sticky to fixed), we'd fail
to recreate the node so keep a node with the wrong type.

Fix by destroying the node and making a new one with a new ID in this case. The
new ID is necessary to ensure that the scrolling tree is updated.

Test: fast/scrolling/sticky-to-fixed.html

* page/scrolling/ScrollingStateTree.cpp:
(WebCore::ScrollingStateTree::nodeTypeAndParentMatch):
(WebCore::ScrollingStateTree::attachNode):
(WebCore::ScrollingStateTree::stateNodeForID):
* page/scrolling/ScrollingStateTree.h:

LayoutTests:

* fast/scrolling/sticky-to-fixed-expected.txt: Added.
* fast/scrolling/sticky-to-fixed.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206712 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[ES6] Align attributes of Generator related properties to spec
utatane.tea@gmail.com [Sat, 1 Oct 2016 23:36:21 +0000 (23:36 +0000)]
[ES6] Align attributes of Generator related properties to spec
https://bugs.webkit.org/show_bug.cgi?id=162839

Reviewed by Saam Barati.

JSTests:

* test262.yaml:

Source/JavaScriptCore:

This patch fixes attributes of Generator related properties.
These fixes are covered by test262.

* runtime/GeneratorFunctionConstructor.cpp:
(JSC::GeneratorFunctionConstructor::finishCreation):
* runtime/GeneratorFunctionConstructor.h:
* runtime/GeneratorFunctionPrototype.cpp:
(JSC::GeneratorFunctionPrototype::finishCreation):
* runtime/GeneratorFunctionPrototype.h:
* runtime/GeneratorPrototype.h:
* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::init):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[ES6] GeneratorFunction constructor should instantiate generator function
utatane.tea@gmail.com [Sat, 1 Oct 2016 23:35:17 +0000 (23:35 +0000)]
[ES6] GeneratorFunction constructor should instantiate generator function
https://bugs.webkit.org/show_bug.cgi?id=162838

Reviewed by Saam Barati.

JSTests:

* stress/generator-function-constructor-is-subclassible.js: Added.
(shouldBe):
(GeneratorFunction):
(prototype.hello):
* stress/generator-function-constructor.js:
* test262.yaml:

Source/JavaScriptCore:

GeneratorFunction's constructor should return an instance of JSGeneratorFunction
instead of JSFunction. In this patch, we fix the following 2 things.

1. GeneratorFunction constructor should use JSGeneratorFunction

    Previously, we used JSFunction to construct a result. It's wrong. We use JSGeneratorFunction.

2. Pass newTarget into GeneratorFunction constructor to make it subclassible

    We did not leverage newTarget when using GeneratorFunction constructor.
    Using it correctly to create the subclass Structure and making GeneratorFunction subclassible.

Test262 test covers (1), but (2) is not covered. We add tests that covers both to stress tests.

* runtime/FunctionConstructor.cpp:
(JSC::constructFunctionSkippingEvalEnabledCheck):
* runtime/GeneratorFunctionConstructor.cpp:
(JSC::constructGeneratorFunctionConstructor):
* runtime/JSGeneratorFunction.cpp:
(JSC::JSGeneratorFunction::JSGeneratorFunction):
(JSC::JSGeneratorFunction::createImpl):
(JSC::JSGeneratorFunction::create):
(JSC::JSGeneratorFunction::createWithInvalidatedReallocationWatchpoint):
* runtime/JSGeneratorFunction.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoGet rid of isMarkedOrNewlyAllocated
fpizlo@apple.com [Sat, 1 Oct 2016 21:14:19 +0000 (21:14 +0000)]
Get rid of isMarkedOrNewlyAllocated
https://bugs.webkit.org/show_bug.cgi?id=162842

Reviewed by Dan Bernstein.

This function has become dead code. This change removes it.

* heap/CellContainer.h:
* heap/CellContainerInlines.h:
(JSC::CellContainer::isMarkedOrNewlyAllocated): Deleted.
* heap/LargeAllocation.h:
(JSC::LargeAllocation::isLive):
(JSC::LargeAllocation::isMarkedOrNewlyAllocated): Deleted.
* heap/MarkedBlock.cpp:
(JSC::MarkedBlock::Handle::isMarkedOrNewlyAllocated): Deleted.
(JSC::MarkedBlock::isMarkedOrNewlyAllocated): Deleted.
* heap/MarkedBlock.h:
(JSC::MarkedBlock::Handle::isMarkedOrNewlyAllocated): Deleted.
(JSC::MarkedBlock::isMarkedOrNewlyAllocated): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoremoving FetchBoyd::m_type
commit-queue@webkit.org [Sat, 1 Oct 2016 16:31:44 +0000 (16:31 +0000)]
removing FetchBoyd::m_type

Remove FetchBody::m_type
https://bugs.webkit.org/show_bug.cgi?id=162841

Patch by Youenn Fablet <youenn@apple.com> on 2016-10-01
Reviewed by Sam Weinig.

No change of behavior.

Removing FetchBody::m_type and using m_data in lieu.
Introducing temporarily FetchBody::m_isEmpty to handle the case of null bodies.
Null body handling should be moved to FetchBodyOwner in a follow-up patch.

* Modules/fetch/FetchBody.cpp:
(WebCore::FetchBody::FetchBody):
(WebCore::FetchBody::extract):
(WebCore::FetchBody::arrayBuffer):
(WebCore::FetchBody::blob):
(WebCore::FetchBody::json):
(WebCore::FetchBody::text):
(WebCore::FetchBody::consumeOnceLoadingFinished):
(WebCore::FetchBody::consume):
(WebCore::FetchBody::consumeAsStream):
(WebCore::FetchBody::loadingSucceeded):
(WebCore::FetchBody::bodyForInternalRequest):
(WebCore::FetchBody::clone):
(WebCore::FetchBody::consumeText): Deleted.
(WebCore::FetchBody::consumeBlob): Deleted.
* Modules/fetch/FetchBody.h:
(WebCore::FetchBody::isEmpty):
(WebCore::FetchBody::isBlob):
(WebCore::FetchBody::isFormData):
(WebCore::FetchBody::isArrayBuffer):
(WebCore::FetchBody::isArrayBufferView):
(WebCore::FetchBody::isURLSearchParams):
(WebCore::FetchBody::isText):
(WebCore::FetchBody::FetchBody):
(WebCore::FetchBody::setContentType): Deleted.
(WebCore::FetchBody::urlSearchParamsBody): Deleted.
* Modules/fetch/FetchBodyOwner.cpp:
(WebCore::FetchBodyOwner::consumeOnceLoadingFinished):
(WebCore::FetchBodyOwner::blobLoadingSucceeded):
(WebCore::FetchBodyOwner::finishBlobLoading): Deleted.
* Modules/fetch/FetchBodyOwner.h:
(WebCore::FetchBodyOwner::body): Deleted.
* Modules/fetch/FetchRequest.cpp:
(WebCore::FetchRequest::setBody):
* Modules/fetch/FetchResponse.cpp:
(WebCore::FetchResponse::fetch):
(WebCore::FetchResponse::BodyLoader::didSucceed):
(WebCore::FetchResponse::consume):
(WebCore::FetchResponse::consumeBodyAsStream):
(WebCore::FetchResponse::createReadableStreamSource):
(WebCore::FetchResponse::finishConsumingStream): Deleted.
* Modules/fetch/FetchResponse.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRename DebugHookID to DebugHookType
commit-queue@webkit.org [Sat, 1 Oct 2016 15:32:59 +0000 (15:32 +0000)]
Rename DebugHookID to DebugHookType
https://bugs.webkit.org/show_bug.cgi?id=162820

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2016-10-01
Reviewed by Alex Christensen.

* bytecode/CodeBlock.cpp:
(JSC::debugHookName):
(JSC::CodeBlock::dumpBytecode):
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitDebugHook):
* bytecompiler/BytecodeGenerator.h:
* interpreter/Interpreter.cpp:
(JSC::Interpreter::debug):
* interpreter/Interpreter.h:
* jit/JITOperations.cpp:
* llint/LLIntSlowPaths.cpp:
(JSC::LLInt::LLINT_SLOW_PATH_DECL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoREGRESSION (r203424): WebCore::ImageBuffer::createCompatibleBuffer() in ImageBufferCG...
ddkilzer@apple.com [Sat, 1 Oct 2016 03:59:23 +0000 (03:59 +0000)]
REGRESSION (r203424): WebCore::ImageBuffer::createCompatibleBuffer() in ImageBufferCG.cpp over-releases CGColorSpaceRef objects
<https://webkit.org/b/162823>
<rdar://problem/27723268>

Reviewed by Joseph Pecoraro.

Code is covered by existing tests, but no crashes have been
observed in practice.  May require running one test multiple
times to reproduce.

* platform/graphics/cg/ImageBufferCG.cpp:
(WebCore::ImageBuffer::createCompatibleBuffer): Don't use
adoptCF() when the function doesn't return a +1 retained
CGColorSpaceRef.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFollow-up patch after r206701
mmaxfield@apple.com [Sat, 1 Oct 2016 03:49:15 +0000 (03:49 +0000)]
Follow-up patch after r206701
https://bugs.webkit.org/show_bug.cgi?id=162818

Unreviewed.

I missed one place required to enable an experimental feature by default.

* UIProcess/Cocoa/WebProcessPoolCocoa.mm:
(WebKit::registerUserDefaultsIfNeeded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206705 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[EFL] Fix build break since r206661. Unreviewed
jh718.park@samsung.com [Sat, 1 Oct 2016 02:48:31 +0000 (02:48 +0000)]
[EFL] Fix build break since r206661. Unreviewed
https://bugs.webkit.org/show_bug.cgi?id=162825

No new tests, no new behaviours.

* rendering/RenderLayer.cpp:
(WebCore::ClipRects::ClipRects):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoDeclare that our variation fonts work is in progress
mmaxfield@apple.com [Sat, 1 Oct 2016 01:28:44 +0000 (01:28 +0000)]
Declare that our variation fonts work is in progress
https://bugs.webkit.org/show_bug.cgi?id=162816

Reviewed by Simon Fraser.

* features.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[iOS] Editing menu omits the Share command if the selected text is longer than 200...
mitz@apple.com [Sat, 1 Oct 2016 01:27:35 +0000 (01:27 +0000)]
[iOS] Editing menu omits the Share command if the selected text is longer than 200 characters
https://bugs.webkit.org/show_bug.cgi?id=162821
<rdar://problem/28409828>

Reviewed by Tim Horton.

* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView canPerformAction:withSender:]): Changed to return YES for the _share:
  action even if the selection is longer than 200 characters. While Define should not have
  been available for longer selections, Share does not have such a limitation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoTurn variation fonts on by default
mmaxfield@apple.com [Sat, 1 Oct 2016 01:26:10 +0000 (01:26 +0000)]
Turn variation fonts on by default
https://bugs.webkit.org/show_bug.cgi?id=162818

Reviewed by Simon Fraser.

Source/WebCore:

No new tests because there is no behavior change.

* page/Settings.in:
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::variationFontsEnabled):

Source/WebKit2:

* Shared/WebPreferencesDefinitions.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Stepping to a line with an autoContinue breakpoint should still pause
joepeck@webkit.org [Sat, 1 Oct 2016 00:52:38 +0000 (00:52 +0000)]
Web Inspector: Stepping to a line with an autoContinue breakpoint should still pause
https://bugs.webkit.org/show_bug.cgi?id=161712
<rdar://problem/28193970>

Reviewed by Brian Burg.

Source/JavaScriptCore:

* debugger/Debugger.cpp:
(JSC::Debugger::pauseIfNeeded):
If we stepped to an auto-continue breakpoint we should continue
stepping, not just continue.

LayoutTests:

* inspector/debugger/stepping/stepping-through-autoContinue-breakpoint-expected.txt: Added.
* inspector/debugger/stepping/stepping-through-autoContinue-breakpoint.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoB3 should support trapping memory accesses
fpizlo@apple.com [Sat, 1 Oct 2016 00:08:28 +0000 (00:08 +0000)]
B3 should support trapping memory accesses
https://bugs.webkit.org/show_bug.cgi?id=162689

Reviewed by Geoffrey Garen.

This adds a traps flag to B3::Kind. It also makes B3::Kind work more like Air::Kind, in the
sense that it's a bag of distinct bits - it doesn't need to be a union unless we get enough
things that it would make a difference.

The only analysis that needs to know about traps is effects. It now knows that traps implies
sideExits, which means that this turns off DCE. The only optimization that needs to know
about traps is eliminateCommonSubexpressions(), which needs to pessimize its store
elimination if the store traps.

The hard part of this change is teaching the instruction selector to faithfully carry the
traps flag down to Air. I got this to work by making ArgPromise a non-copyable object that
knows whether you've used it in an instruction. It knows when you call consume(). If you do
this then ArgPromise cannot be destructed without first passing your inst through it. This,
along with a few other hacks, means that all of the load-op and load-op-store fusions
correctly carry the trap bit: if any of the B3 loads or stores involved traps then you get
traps in Air.

This framework also sets us up to do bug 162688, since the ArgPromise::inst() hook is
powerful enough to allow wrapping the instruction with a Patch.

I added some tests to testb3 that verify that optimizations are appropriately inhibited and
that the traps flag survives until the bitter end of Air.

* b3/B3EliminateCommonSubexpressions.cpp:
* b3/B3Kind.cpp:
(JSC::B3::Kind::dump):
* b3/B3Kind.h:
(JSC::B3::Kind::Kind):
(JSC::B3::Kind::hasExtraBits):
(JSC::B3::Kind::isChill):
(JSC::B3::Kind::setIsChill):
(JSC::B3::Kind::hasTraps):
(JSC::B3::Kind::traps):
(JSC::B3::Kind::setTraps):
(JSC::B3::Kind::operator==):
(JSC::B3::Kind::hash):
(JSC::B3::trapping):
* b3/B3LowerToAir.cpp:
(JSC::B3::Air::LowerToAir::ArgPromise::swap):
(JSC::B3::Air::LowerToAir::ArgPromise::ArgPromise):
(JSC::B3::Air::LowerToAir::ArgPromise::operator=):
(JSC::B3::Air::LowerToAir::ArgPromise::~ArgPromise):
(JSC::B3::Air::LowerToAir::ArgPromise::setTraps):
(JSC::B3::Air::LowerToAir::ArgPromise::consume):
(JSC::B3::Air::LowerToAir::ArgPromise::inst):
(JSC::B3::Air::LowerToAir::trappingInst):
(JSC::B3::Air::LowerToAir::loadPromiseAnyOpcode):
(JSC::B3::Air::LowerToAir::appendUnOp):
(JSC::B3::Air::LowerToAir::appendBinOp):
(JSC::B3::Air::LowerToAir::tryAppendStoreUnOp):
(JSC::B3::Air::LowerToAir::tryAppendStoreBinOp):
(JSC::B3::Air::LowerToAir::appendStore):
(JSC::B3::Air::LowerToAir::append):
(JSC::B3::Air::LowerToAir::createGenericCompare):
(JSC::B3::Air::LowerToAir::createBranch):
(JSC::B3::Air::LowerToAir::createCompare):
(JSC::B3::Air::LowerToAir::createSelect):
(JSC::B3::Air::LowerToAir::lower):
* b3/B3Validate.cpp:
* b3/B3Value.cpp:
(JSC::B3::Value::effects):
* b3/B3Value.h:
* b3/air/AirCode.h:
* b3/testb3.cpp:
(JSC::B3::testTrappingLoad):
(JSC::B3::testTrappingStore):
(JSC::B3::testTrappingLoadAddStore):
(JSC::B3::testTrappingLoadDCE):
(JSC::B3::testTrappingStoreElimination):
(JSC::B3::run):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206694 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Stepping over/out of a function sometimes resumes instead of taking...
joepeck@webkit.org [Fri, 30 Sep 2016 23:26:48 +0000 (23:26 +0000)]
Web Inspector: Stepping over/out of a function sometimes resumes instead of taking you to caller
https://bugs.webkit.org/show_bug.cgi?id=162802
<rdar://problem/28569982>

Reviewed by Mark Lam.

Source/JavaScriptCore:

* debugger/Debugger.cpp:
(JSC::Debugger::stepOverStatement):
(JSC::Debugger::stepOutOfFunction):
Enable stepping mode when we start stepping.

LayoutTests:

* inspector/debugger/resources/log-pause-location.js:
(TestPage.registerInitializer.createLocation):
(TestPage.registerInitializer.window.setBreakpointsOnLinesWithBreakpointComment):
Helper to set breakpoints everywhere in a file that has a BREAKPOINT comment.

* inspector/debugger/stepping/stepping-pause-in-inner-step-to-parent-expected.txt: Added.
* inspector/debugger/stepping/stepping-pause-in-inner-step-to-parent.html: Added.
Test that stepping out and over end up in the parent. Before this change
we would never have made it back into entry and the test failed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoImplement rendering of font-variation-settings
mmaxfield@apple.com [Fri, 30 Sep 2016 23:12:23 +0000 (23:12 +0000)]
Implement rendering of font-variation-settings
https://bugs.webkit.org/show_bug.cgi?id=162782

Reviewed by Zalan Bujtas.

Source/WebCore:

Because the heavy lifting for font-variation-settings is done by CoreText,
this patch is fairly minimal. It simply hooks up the CSS property to CoreText.

There is an existing bug in CoreText where variations inside fonts do not
survive the addition of a cascade list. Therefore, FontPlatformData::ctFont()
needs to work around this (conditionally) by resupplying the variation at
the same time as the cascade list. The CoreText bug is <rdar://problem/28449441>.

Tests: fast/text/variations/duplicate.html
       fast/text/variations/exist.html
       fast/text/variations/inheritance.html
       fast/text/variations/order.html
       fast/text/variations/outofbounds.html

* platform/graphics/FontCache.h:
* platform/graphics/cocoa/FontCacheCoreText.cpp:
(WebCore::preparePlatformFont):
(WebCore::fontWithFamily):
(WebCore::FontCache::createFontPlatformData):
(WebCore::FontCache::systemFallbackForCharacters):
* platform/graphics/cocoa/FontPlatformDataCocoa.mm:
(WebCore::cascadeToLastResortAttributesDictionary):
(WebCore::cascadeToLastResortAndVariationsFontDescriptor):
(WebCore::FontPlatformData::ctFont):
(WebCore::cascadeToLastResortFontDescriptor): Deleted.
* platform/graphics/mac/FontCustomPlatformData.cpp:
(WebCore::FontCustomPlatformData::fontPlatformData):
* rendering/RenderThemeIOS.mm:
(WebCore::RenderThemeIOS::updateCachedSystemFontDescription):

LayoutTests:

* fast/text/variations/duplicate-expected.html: Added.
* fast/text/variations/duplicate.html: Added.
* fast/text/variations/exist-expected-mismatch.html: Added.
* fast/text/variations/exist.html: Added.
* fast/text/variations/getComputedStyle.html:
* fast/text/variations/inheritance-expected.html: Added.
* fast/text/variations/inheritance.html: Added.
* fast/text/variations/order-expected.html: Added.
* fast/text/variations/order.html: Added.
* fast/text/variations/outofbounds-expected.html: Added.
* fast/text/variations/outofbounds.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdd initial support for IDL union conversion
weinig@apple.com [Fri, 30 Sep 2016 23:10:31 +0000 (23:10 +0000)]
Add initial support for IDL union conversion
https://bugs.webkit.org/show_bug.cgi?id=161576

Reviewed by Chris Dumez.

Source/WebCore:

* WebCore.xcodeproj/project.pbxproj:
Add IDLTypes.h

* bindings/generic/IDLTypes.h: Added.
Add a basic type hierarchy for types defined by WebIDL and their corresponding
WebCore implementation types.

* bindings/js/JSDOMConvert.h:
(WebCore::Converter<IDLDOMString>::convert):
(WebCore::Converter<IDLUSVString>::convert):
Add Converters for IDLDOMString and IDLUSVString.

(WebCore::ConditionalConverter<typename ReturnType, typename T, bool enabled>)
Add a struct to allow conditional conversion of types based on a compile boolean
value. This allows the union code to work, by ensuring that code for types that don't
exist doesn't get compiled.

(WebCore::Converter<IDLUnion>::convert):
Add template meta programming based version of the WebIDL conversion algorithm. It uses
introspection of the typelist to conditionally execute the parts of the algorithm that
apply and relies on the compiler eliminate the dead code it produces. This initial cut
doesn't implement the full algorithm, just a subset needed to see if this method will work
out.

* bindings/js/JSDOMWrapper.h:
(WebCore::isJSDOMWrapperType):
Add JSDOMWrapperType to the list of well known types to allow quick checking of wrappers
by the union code.

* bindings/js/JSNodeOrString.cpp:
(WebCore::toNodeOrStringVector):
Switch to new IDLUnion based converter.

Source/WTF:

* WTF.xcodeproj/project.pbxproj:
* wtf/CMakeLists.txt:
Add Brigand.h

* wtf/Brigand.h: Added.
Import a standalone copy of Edouard Alligand and Joel Falcou's
Brigand library for help with for help with list based meta programming

* wtf/StdLibExtras.h:
Add a new three new type traits, IsTemplate, IsBaseOfTemplate and RemoveCVAndReference.
- IsTemplate acts like std::is_class, but works with a generic base.
- IsBaseOfTemplate acts like std::is_base_of, but works with a generic base.
- RemoveCVAndReference combines std::remove_cv and std::remove_reference.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoImplement animation of font-variation-settings
mmaxfield@apple.com [Fri, 30 Sep 2016 23:08:02 +0000 (23:08 +0000)]
Implement animation of font-variation-settings
https://bugs.webkit.org/show_bug.cgi?id=162783

Reviewed by Simon Fraser.

Source/WebCore:

Modify CSSPropertyAnimation to understand FontVariationSettings objects and how
to interpolate them.

If two FontVariationSettings objects are unlike (meaning they specify different
variation axes), for now the interpolation simply returns an empty object. This
might change in the future, but for now, this is a reasonable place to start.

Because CSSPropertyAnimation interacts with RenderStyles instead of
FontDescriptions, this patch adds a transparent accessor from the RenderStyle
to the inner FontDescription.

Tests: animations/font-variation-settings-order.html
       animations/font-variation-settings-unlike.html
       animations/font-variation-settings.html

* page/animation/CSSPropertyAnimation.cpp:
(WebCore::blendFunc):
(WebCore::PropertyWrapperFontVariationSettings::PropertyWrapperFontVariationSettings):
(WebCore::CSSPropertyAnimationWrapperMap::CSSPropertyAnimationWrapperMap):
* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::setFontVariationSettings):
* rendering/style/RenderStyle.h:
(WebCore::RenderStyle::fontVariationSettings):

LayoutTests:

Animation tests need a little infrastructure to be able to tell if two computed
values for font-variation-settings are equivalent.

* animations/font-variation-settings-expected.html: Added.
* animations/font-variation-settings-order-expected.html: Added.
* animations/font-variation-settings-order.html: Added.
* animations/font-variation-settings-unlike-expected.html: Added.
* animations/font-variation-settings-unlike.html: Added.
* animations/font-variation-settings.html: Added.
* animations/resources/animation-test-helpers.js:
(compareFontVariationSettings):
(getPropertyValue):
(comparePropertyValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Modern Media Controls] layout nodes
commit-queue@webkit.org [Fri, 30 Sep 2016 22:45:17 +0000 (22:45 +0000)]
[Modern Media Controls] layout nodes
https://bugs.webkit.org/show_bug.cgi?id=162799
<rdar://problem/28569301>

Patch by Antoine Quint <graouts@apple.com> on 2016-09-30
Reviewed by Dean Jackson.

Source/WebCore:

Modern media controls will be using a tree of LayoutNode objects that commit to the DOM
in coordinated `requestAnimationFrame()` calls to ensure all layouts are done in an efficient
and coordinated manner. As a preamble, we introduced the `scheduler` singleton in
https://webkit.org/b/162726 which is in charge of scheduling callbacks.

A LayoutNode is created by providing an Element to its constructor, or an HTML string. Not
providing a parameter creates a simple <div>.

When we set a property on a LayoutNode, we call `markDirtyProperty(propertyName)` which keeps
track of dirty properties in the `_dirtyProperties` set. When this set is non-empty, the node
is marked as dirty and registered in the global `dirtyNodes` map, asking the shared scheduler
that a layout is needed. When the layout is performed, all nodes in the `dirtyNodes` map are
processed such that `commitProperty(propertyName)` is called to commit dirty properties for
a given node to the DOM, and `layout()` is called to allow subclasses of LayoutNode to conduct
custom layout logic that goes beyond committing a given property.

Another reason why a node may be marked as dirty is when a DOM hierarchy change is needed. A
host of DOM-like methods are exposed to allow flexible manipulations of nodes, with an extra
`children` property which allows wholesale change of a node's subtree with a single array
property assignment. Changes to the DOM hierarchy are performed in the same scheduler callback
as style properties.

Nodes can be marked for layout explicitly with by setting the `needsLayout` property.

Tests: media/modern-media-controls/layout-node/addChild.html
       media/modern-media-controls/layout-node/children.html
       media/modern-media-controls/layout-node/constructor.html
       media/modern-media-controls/layout-node/height.html
       media/modern-media-controls/layout-node/insertAfter.html
       media/modern-media-controls/layout-node/insertBefore.html
       media/modern-media-controls/layout-node/parent.html
       media/modern-media-controls/layout-node/remove.html
       media/modern-media-controls/layout-node/removeChild.html
       media/modern-media-controls/layout-node/subclassing.html
       media/modern-media-controls/layout-node/visible.html
       media/modern-media-controls/layout-node/width.html
       media/modern-media-controls/layout-node/x.html
       media/modern-media-controls/layout-node/y.html

* Modules/modern-media-controls/controls/layout-node.js: Added.
(LayoutNode):
(LayoutNode.prototype.get x):
(LayoutNode.prototype.set x):
(LayoutNode.prototype.get y):
(LayoutNode.prototype.set y):
(LayoutNode.prototype.get width):
(LayoutNode.prototype.set width):
(LayoutNode.prototype.get height):
(LayoutNode.prototype.set height):
(LayoutNode.prototype.get visible):
(LayoutNode.prototype.set visible):
(LayoutNode.prototype.get needsLayout):
(LayoutNode.prototype.set needsLayout):
(LayoutNode.prototype.get parent):
(LayoutNode.prototype.get children):
(LayoutNode.prototype.set children):
(LayoutNode.prototype.addChild):
(LayoutNode.prototype.insertBefore):
(LayoutNode.prototype.insertAfter):
(LayoutNode.prototype.removeChild):
(LayoutNode.prototype.remove):
(LayoutNode.prototype.markDirtyProperty):
(LayoutNode.prototype.commitProperty):
(LayoutNode.prototype.layout):
(LayoutNode.prototype._markNodeManipulation):
(LayoutNode.prototype._updateDirtyState):
(LayoutNode.prototype._updateChildren):
(performScheduledLayout):
(elementFromString):

LayoutTests:

Testing all public properties and methods of the LayoutNode class.

* media/modern-media-controls/layout-node/addChild-expected.txt: Added.
* media/modern-media-controls/layout-node/addChild.html: Added.
* media/modern-media-controls/layout-node/children-expected.txt: Added.
* media/modern-media-controls/layout-node/children.html: Added.
* media/modern-media-controls/layout-node/constructor-expected.txt: Added.
* media/modern-media-controls/layout-node/constructor.html: Added.
* media/modern-media-controls/layout-node/height-expected.txt: Added.
* media/modern-media-controls/layout-node/height.html: Added.
* media/modern-media-controls/layout-node/insertAfter-expected.txt: Added.
* media/modern-media-controls/layout-node/insertAfter.html: Added.
* media/modern-media-controls/layout-node/insertBefore-expected.txt: Added.
* media/modern-media-controls/layout-node/insertBefore.html: Added.
* media/modern-media-controls/layout-node/parent-expected.txt: Added.
* media/modern-media-controls/layout-node/parent.html: Added.
* media/modern-media-controls/layout-node/remove-expected.txt: Added.
* media/modern-media-controls/layout-node/remove.html: Added.
* media/modern-media-controls/layout-node/removeChild-expected.txt: Added.
* media/modern-media-controls/layout-node/removeChild.html: Added.
* media/modern-media-controls/layout-node/subclassing-expected.txt: Added.
* media/modern-media-controls/layout-node/subclassing.html: Added.
* media/modern-media-controls/layout-node/visible-expected.txt: Added.
* media/modern-media-controls/layout-node/visible.html: Added.
* media/modern-media-controls/layout-node/width-expected.txt: Added.
* media/modern-media-controls/layout-node/width.html: Added.
* media/modern-media-controls/layout-node/x-expected.txt: Added.
* media/modern-media-controls/layout-node/x.html: Added.
* media/modern-media-controls/layout-node/y-expected.txt: Added.
* media/modern-media-controls/layout-node/y.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206686 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoREGRESSION (Safari 10 combined with WK changes): Unable to store WebCrypto keys in...
beidson@apple.com [Fri, 30 Sep 2016 22:38:58 +0000 (22:38 +0000)]
REGRESSION (Safari 10 combined with WK changes): Unable to store WebCrypto keys in IndexedDB database.
<rdar://problem/28334440> and https://bugs.webkit.org/show_bug.cgi?id=162554

Reviewed by Alexey Proskuryakov.

Safari 10 has a WKPageNavigation client but does not implement the "copyWebCryptoMasterKey" callback.

WebKit just fails in this case, instead of falling back to "getDefaultWebCryptoMasterKey".
It should fall back.

Additionally there's a WKContextClient callback to get the crypto key, also, but it is completely unused.
So it should be pulled.

* UIProcess/API/APILoaderClient.h:
(API::LoaderClient::webCryptoMasterKey): Deleted.
* UIProcess/API/C/WKContext.h:

* UIProcess/API/C/WKPage.cpp:
(WKPageSetPageLoaderClient):
(WKPageSetPageNavigationClient): If the client doesn't implement this method, fallback to getDefaultWebCryptoMasterKey.

* UIProcess/Cocoa/NavigationState.mm:
(WebKit::NavigationState::NavigationClient::webCryptoMasterKey): If the client doesn't implement this method,
  fallback to getDefaultWebCryptoMasterKey.

* UIProcess/WebContextClient.cpp:
(WebKit::WebContextClient::copyWebCryptoMasterKey): Deleted.
* UIProcess/WebContextClient.h:

* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::wrapCryptoKey): Only ask the NavigationClient before falling back. Don't ask the loader client.
(WebKit::WebPageProxy::unwrapCryptoKey): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206684 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoThe dragged image should be the current frame only of the animated image
commit-queue@webkit.org [Fri, 30 Sep 2016 22:30:20 +0000 (22:30 +0000)]
The dragged image should be the current frame only of the animated image
https://bugs.webkit.org/show_bug.cgi?id=162109

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2016-09-30
Reviewed by Tim Horton.

Source/WebCore:

Instead of creating an NSImage with all the frames for the dragImage,
create an NSImage with the current frame only.

* dom/DataTransferMac.mm:
(WebCore::DataTransfer::createDragImage): Call currentFrameNSImage() to create the dragImage.
* editing/cocoa/HTMLConverter.mm:
(fileWrapperForElement):  Call the Image function with its new name.
* platform/graphics/BitmapImage.h:
* platform/graphics/Image.h:
(WebCore::Image::nsImage): Rename getNSImage() to nsImage().
(WebCore::Image::currentFrameNSImage): Returns the NSImage of the current frame.
(WebCore::Image::tiffRepresentation): Rename getTIFFRepresentation() to tiffRepresentation().
(WebCore::Image::getNSImage): Deleted.
(WebCore::Image::getTIFFRepresentation): Deleted.
* platform/graphics/mac/ImageMac.mm:
(WebCore::BitmapImage::tiffRepresentation): Rename getTIFFRepresentation() to tiffRepresentation().
(WebCore::BitmapImage::nsImage): Rename getNSImage() to nsImage().
(WebCore::BitmapImage::currentFrameNSImage): Returns the NSImage of the current frame.
(WebCore::BitmapImage::getTIFFRepresentation): Deleted.
(WebCore::BitmapImage::getNSImage): Deleted.
* platform/mac/CursorMac.mm:
(WebCore::createCustomCursor): Call currentFrameNSImage() since the cursor does not animate anyway.
* platform/mac/DragImageMac.mm:
(WebCore::createDragImageFromImage): Use currentFrameNSImage() for the dragImage.
* platform/mac/PasteboardMac.mm:
(WebCore::Pasteboard::write): Call the Image function with its new name.

Source/WebKit/mac:

* DOM/DOM.mm:
(-[DOMElement image]): Call the Image function with its new name.
(-[DOMElement _imageTIFFRepresentation]): Ditto.
* Misc/WebElementDictionary.mm:
(-[WebElementDictionary _image]): Call the Image function with its new name.
* Misc/WebIconDatabase.mm:
(-[WebIconDatabase defaultIconWithSize:]): Call currentFrameNSImage() to create the icon image.
(webGetNSImage): Call the Image function with its new name.
* WebCoreSupport/WebContextMenuClient.mm:
(WebContextMenuClient::imageForCurrentSharingServicePickerItem): Call currentFrameNSImage() instead of nsImage()..
(WebContextMenuClient::contextMenuForEvent): Ditto.
* WebView/WebHTMLView.mm:
(-[WebHTMLView pasteboard:provideDataForType:]): Call the Image function with its new name.

Source/WebKit2:

* UIProcess/Cocoa/WebViewImpl.mm:
(WebKit::WebViewImpl::provideDataForPasteboard): Call the Image function with its new name.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206683 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoB3::moveConstants should be able to edit code to minimize the number of constants
fpizlo@apple.com [Fri, 30 Sep 2016 22:29:24 +0000 (22:29 +0000)]
B3::moveConstants should be able to edit code to minimize the number of constants
https://bugs.webkit.org/show_bug.cgi?id=162764

Reviewed by Saam Barati.

Source/JavaScriptCore:

There are some interesting cases where we can reduce the number of constant materializations if
we teach moveConstants() how to edit code. The two examples that this patch supports are:

    - Loads and stores from a constant pointer. Since loads and stores get an offset for free
      and the instruction selector is really good at handling it, and since we can query Air to
      see what kinds of offsets are legal, we can sometimes avoid using a constant pointer that
      is specific to the absolute address of that load and instead pick some other constant
      that is within offset distance of ours.

    - Add and Sub by a constant (x + c, x - c). Since x + c = x - -c and x - c = x + -c, we can
      flip Add to Sub or vice versa if the negated constant is available.

This change makes moveConstants() pick the most dominant constant that works for an value. In
the case of memory accesses, it uses Air::Arg::isValidAddrForm() to work out what other
constants would work. In the case of Add/Sub, it simply looks for the negated constant. This
should result in something like a minimal number of constants since these rules always pick the
most dominant constant that works - so if an Add's constant is already most dominant then
nothing changes, but if the negated one is more dominant then it becomes a Sub.

This is a 0.5% speed-up on LongSpider and neutral elsewhere. It's a speed-up because the
absolute address thing reduces the number of address materializations that we have to do, while
the add/sub thing prevents us from having to materialize 0x1000000000000 to box doubles.
However, this may introduce a pathology, which I've filed a bug for: bug 162796.

* b3/B3MoveConstants.cpp:
* b3/B3MoveConstants.h:
* b3/B3UseCounts.h:
* b3/air/AirFixObviousSpills.cpp:
* b3/testb3.cpp:
(JSC::B3::testMoveConstants):
(JSC::B3::run):

Source/WTF:

I thought it would be a good idea to document the fact that dominator traversal happens in a
particular order for a reason.

* wtf/Dominators.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206682 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[iOS] Allow sequence<Touch> input in TouchEvent constructor
cdumez@apple.com [Fri, 30 Sep 2016 22:21:59 +0000 (22:21 +0000)]
[iOS] Allow sequence<Touch> input in TouchEvent constructor
https://bugs.webkit.org/show_bug.cgi?id=162806
<rdar://problem/28566429>

Reviewed by Ryosuke Niwa.

Allow sequence<Touch> input in TouchEvent constructor in addition to
TouchList objects. It is convenient for developers to pass arrays of
Touch objects.

No new tests, already covered by:
imported/w3c/web-platform-tests/touch-events/touch-touchevent-constructor.html

* bindings/js/JSDOMBinding.h:
(WebCore::toRefNativeArray):
(WebCore::toRefPtrNativeArray):
* bindings/js/JSDictionary.cpp:
(WebCore::JSDictionary::convertValue):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206681 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMarking http/tests/media/hls/hls-video-resize.html as flaky on mac-wk1.
ryanhaddad@apple.com [Fri, 30 Sep 2016 22:21:40 +0000 (22:21 +0000)]
Marking http/tests/media/hls/hls-video-resize.html as flaky on mac-wk1.
https://bugs.webkit.org/show_bug.cgi?id=162507

Unreviewed test gardening.

* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206680 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoParse font-variation-settings
mmaxfield@apple.com [Fri, 30 Sep 2016 21:59:15 +0000 (21:59 +0000)]
Parse font-variation-settings
https://bugs.webkit.org/show_bug.cgi?id=162781

Reviewed by Simon Fraser.

Source/WebCore:

CSS Fonts level 4 [1] details the font-variation-settings property which allows variations
to be applied to fonts which accept it. The property accepts a list of key/value
pairs where the keys are four-character Ascii codes and the values are floating
point values. The implementation uses font-feature-settings as a model, but with one
big difference: font-variation-settings can be animated. The set of variation points
are animated individually.

This font variations work is being done behind a run-time switch, which is settable from
the Safari Develop menu.

Now that FontDescription is growing, I'd like to do an investigation into the possibility of
moving some of the rare pieces of it into their own class to reduce memory. However, upon
advice, this investigation will be fairly involved and should be done in its own bug.

Test: fast/text/variations/getComputedStyle.html

* CMakeLists.txt:
* WebCore.xcodeproj/project.pbxproj:
* css/CSSAllInOne.cpp:
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::ComputedStyleExtractor::propertyValue):
* css/CSSFontVariationValue.cpp: Copied from Source/WebCore/platform/graphics/FontTaggedSettings.cpp.
(WebCore::CSSFontVariationValue::CSSFontVariationValue):
(WebCore::CSSFontVariationValue::customCSSText):
(WebCore::CSSFontVariationValue::equals):
* css/CSSFontVariationValue.h: Copied from Source/WebCore/platform/graphics/FontTaggedSettings.cpp.
* css/CSSPropertyNames.in:
* css/CSSValue.cpp:
(WebCore::CSSValue::equals):
(WebCore::CSSValue::cssText):
(WebCore::CSSValue::destroy):
* css/CSSValue.h:
(WebCore::CSSValue::isFontVariationValue):
(WebCore::CSSValue::isFontFeatureValue): Deleted.
* css/StyleBuilderConverter.h:
(WebCore::StyleBuilderConverter::convertFontVariationSettings):
* css/StyleBuilderCustom.h:
(WebCore::StyleBuilderCustom::applyInitialFontVariationSettings):
(WebCore::StyleBuilderCustom::applyInheritFontVariationSettings):
(WebCore::StyleBuilderCustom::applyInheritFontFeatureSettings): Deleted.
* css/parser/CSSParser.cpp:
(WebCore::CSSParserContext::CSSParserContext):
(WebCore::operator==):
(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::parseFontFeatureSettings):
(WebCore::CSSParser::parseFontVariationTag):
(WebCore::CSSParser::parseFontVariationSettings):
* css/parser/CSSParser.h:
* css/parser/CSSParserMode.h:
* platform/graphics/FontCache.h:
(WebCore::FontDescriptionKey::FontDescriptionKey):
(WebCore::FontDescriptionKey::operator==):
(WebCore::FontDescriptionKey::computeHash):
* platform/graphics/FontDescription.cpp:
* platform/graphics/FontDescription.h:
(WebCore::FontDescription::variationSettings):
(WebCore::FontDescription::setVariationSettings):
(WebCore::FontDescription::operator==):
(WebCore::FontDescription::featureSettings): Deleted.
(WebCore::FontDescription::setFeatureSettings): Deleted.
* platform/graphics/FontTaggedSettings.cpp:
(WebCore::FontVariationSettings::hash):
(WebCore::operator<<):
* platform/graphics/FontTaggedSettings.h:

LayoutTests:

* fast/text/variations/getComputedStyle-expected.txt: Added.
* fast/text/variations/getComputedStyle.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206679 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMark inspector/debugger/breakpoints tests as slow on release, skipped on debug.
ryanhaddad@apple.com [Fri, 30 Sep 2016 21:53:38 +0000 (21:53 +0000)]
Mark inspector/debugger/breakpoints tests as slow on release, skipped on debug.

Unreviewed test gardening.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206678 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRebaseline results after r206654.
joepeck@webkit.org [Fri, 30 Sep 2016 21:43:53 +0000 (21:43 +0000)]
Rebaseline results after r206654.

Better location for unary expressions.

* js/stack-trace-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206677 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFollowup patch to r206664
mmaxfield@apple.com [Fri, 30 Sep 2016 21:41:23 +0000 (21:41 +0000)]
Followup patch to r206664
https://bugs.webkit.org/show_bug.cgi?id=162774

Unreviewed.

No new tests because there is no behavior change.

* platform/graphics/FontTaggedSettings.h:
(WebCore::FourCharacterTagHashTraits::isDeletedValue): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206676 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSecond time going into fullscreen using silverlight, will hide the menu bar and dock...
andersca@apple.com [Fri, 30 Sep 2016 21:32:54 +0000 (21:32 +0000)]
Second time going into fullscreen using silverlight, will hide the menu bar and dock for Safari
https://bugs.webkit.org/show_bug.cgi?id=162805
rdar://problem/28208495

Reviewed by Dan Bernstein.

Turns out that the WindowRef wrappers for full screen NSWindows can end up in the m_windows HashSet,
and never go away.

Fix this by storing the canonical CGWindowIDs in the hash map instead.

* PluginProcess/mac/PluginProcessMac.mm:
(WebKit::cgWindowID):
(WebKit::windowCoversAnyScreen):
(WebKit::FullscreenWindowTracker::windowShown):
(WebKit::FullscreenWindowTracker::windowHidden):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206675 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix modules tests after r206653 handle breakpoint locations in import/export statements
joepeck@webkit.org [Fri, 30 Sep 2016 21:23:34 +0000 (21:23 +0000)]
Fix modules tests after r206653 handle breakpoint locations in import/export statements
https://bugs.webkit.org/show_bug.cgi?id=162807

Reviewed by Mark Lam.

* parser/ASTBuilder.h:
(JSC::ASTBuilder::createExportDefaultDeclaration):
(JSC::ASTBuilder::createExportLocalDeclaration):
Don't record an extra breakpoint location for the statement
within an export statement.

* parser/Parser.cpp:
(JSC::Parser<LexerType>::parseModuleSourceElements):
Record a pause location for import/export statements.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206671 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix fast/text/trak-optimizeLegibility.html
mmaxfield@apple.com [Fri, 30 Sep 2016 21:17:47 +0000 (21:17 +0000)]
Fix fast/text/trak-optimizeLegibility.html
https://bugs.webkit.org/show_bug.cgi?id=162779

Reviewed by Simon Fraser.

Source/WebCore:

I accidentally created this test to allow an unnecessary delta
in the width comparison. This covered the fact that the fonts
were not being selected correctly in our testing code. Update the
font selection code to work with system fonts correctly and
update the test to not use a delta when performing width
comparisons.

Test: fast/text/trak-optimizeLegibility.html

* platform/graphics/cocoa/FontCacheCoreText.cpp:
(WebCore::isSystemFont):
(WebCore::platformFontLookupWithFamily):

LayoutTests:

* fast/text/trak-optimizeLegibility.html:
* platform/ios-simulator/fast/text/trak-optimizeLegibility-expected.txt:
* platform/mac-elcapitan/fast/text/trak-optimizeLegibility-expected.txt: Removed.
* platform/mac/fast/text/trak-optimizeLegibility-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206666 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoCreate runtime flag for variation font work
mmaxfield@apple.com [Fri, 30 Sep 2016 21:16:07 +0000 (21:16 +0000)]
Create runtime flag for variation font work
https://bugs.webkit.org/show_bug.cgi?id=162780

Reviewed by Alex Christensen.

Source/WebCore:

No new tests because there is no behavior change.

Variation font work should be behind a runtime flag.

* page/Settings.in:
* testing/InternalSettings.cpp:
(WebCore::InternalSettings::Backup::Backup):
(WebCore::InternalSettings::Backup::restoreTo):
(WebCore::InternalSettings::variationFontsEnabled):
(WebCore::InternalSettings::setVariationFontsEnabled):
* testing/InternalSettings.h:
* testing/InternalSettings.idl:

Source/WebKit2:

* Shared/WebPreferencesDefinitions.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updatePreferences):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206665 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoTemplatize internal FontFeatureSettings class for variation font support
mmaxfield@apple.com [Fri, 30 Sep 2016 21:11:21 +0000 (21:11 +0000)]
Templatize internal FontFeatureSettings class for variation font support
https://bugs.webkit.org/show_bug.cgi?id=162774

Reviewed by Alex Christensen.

Font features and font variations are similar in that they are identified
by four-character tags and an associated value. However, the value is an
int for font features and is a float for font variations. In order to
simplify code, we can use the same class for both, and use a template
argument to distinguish between the two. Then, typedefs can be used to
name them.

No new tests because there is no behavior change.

* WebCore.xcodeproj/project.pbxproj: Rename FontFeatureSettings.{cpp,h}
to FontTaggedSettings.{cpp,h}
* css/CSSComputedStyleDeclaration.cpp:
(WebCore::ComputedStyleExtractor::propertyValue):
* css/CSSFontFace.h:
* css/CSSFontFaceSource.h:
* css/CSSFontFeatureValue.cpp:
(WebCore::CSSFontFeatureValue::CSSFontFeatureValue):
* css/CSSFontFeatureValue.h:
* css/FontFace.cpp:
(WebCore::FontFace::featureSettings):
* css/parser/CSSParser.cpp:
(WebCore::CSSParser::parseFontFeatureTag):
* loader/cache/CachedFont.h:
* platform/graphics/FontDescription.h:
* platform/graphics/FontFeatureSettings.cpp:
(WebCore::FontFeature::FontFeature): Deleted.
(WebCore::FontFeature::operator==): Deleted.
(WebCore::FontFeature::operator<): Deleted.
(WebCore::FontFeatureSettings::insert): Deleted.
(WebCore::FontFeatureSettings::hash): Deleted.
* platform/graphics/FontFeatureSettings.h: Removed.
(WebCore::fontFeatureTag): Deleted.
(WebCore::FontFeatureTagHash::hash): Deleted.
(WebCore::FontFeatureTagHash::equal): Deleted.
(WebCore::FontFeatureTagHashTraits::constructDeletedValue): Deleted.
(WebCore::FontFeatureTagHashTraits::isDeletedValue): Deleted.
(WebCore::FontFeature::operator!=): Deleted.
(WebCore::FontFeature::tag): Deleted.
(WebCore::FontFeature::value): Deleted.
(WebCore::FontFeature::enabled): Deleted.
(WebCore::FontFeatureSettings::operator==): Deleted.
(WebCore::FontFeatureSettings::operator!=): Deleted.
(WebCore::FontFeatureSettings::size): Deleted.
(WebCore::FontFeatureSettings::operator[]): Deleted.
(WebCore::FontFeatureSettings::at): Deleted.
(WebCore::FontFeatureSettings::begin): Deleted.
(WebCore::FontFeatureSettings::end): Deleted.
* platform/graphics/FontTaggedSettings.cpp: Renamed from Source/WebCore/platform/graphics/FontFeatureSettings.cpp.
(WebCore::FontFeatureSettings::hash):
* platform/graphics/FontTaggedSettings.h: Added.
(WebCore::fontFeatureTag):
(WebCore::FourCharacterTagHash::hash):
(WebCore::FourCharacterTagHash::equal):
(WebCore::FourCharacterTagHashTraits::constructDeletedValue):
(WebCore::FourCharacterTagHashTraits::isDeletedValue):
(WebCore::FontTaggedSetting::operator!=):
(WebCore::FontTaggedSetting::tag):
(WebCore::FontTaggedSetting::value):
(WebCore::FontTaggedSetting::enabled):
(WebCore::FontTaggedSetting<T>::FontTaggedSetting):
(WebCore::=):
(WebCore::FontTaggedSetting<T>::operator<):
(WebCore::FontTaggedSettings::operator==):
(WebCore::FontTaggedSettings::operator!=):
(WebCore::FontTaggedSettings::isEmpty):
(WebCore::FontTaggedSettings::size):
(WebCore::FontTaggedSettings::operator[]):
(WebCore::FontTaggedSettings::at):
(WebCore::FontTaggedSettings::begin):
(WebCore::FontTaggedSettings::end):
(WebCore::FontTaggedSettings<T>::insert):
* platform/graphics/cocoa/FontCacheCoreText.cpp:
(WebCore::tagEquals):
(WebCore::appendOpenTypeFeature):
* platform/graphics/mac/FontCustomPlatformData.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206664 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFragment serialization should always use 'xml:' prefix for attributes in XML namespace
cdumez@apple.com [Fri, 30 Sep 2016 21:03:27 +0000 (21:03 +0000)]
Fragment serialization should always use 'xml:' prefix for attributes in XML namespace
https://bugs.webkit.org/show_bug.cgi?id=162803

Reviewed by Sam Weinig.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

* web-platform-tests/html/syntax/serializing-html-fragments/serializing-expected.txt:

Source/WebCore:

Fragment serialization should always use 'xml:' prefix for attributes in XML namespace:
- https://html.spec.whatwg.org/#serialising-html-fragments
"""
If the attribute is in the XML namespace
-> The attribute's serialized name is the string "xml:" followed by the attribute's local name.
"""

- https://www.w3.org/TR/DOM-Parsing/#dfn-concept-xml-serialization-algorithm
"""
If ns is the XML namespace, then let qualified name be the concatenation of the string "xml:"
and the value of node's localName.
"""

Firefox complies with the specification.

No new tests, rebaselined existing test.

* editing/MarkupAccumulator.cpp:
(WebCore::MarkupAccumulator::appendAttribute):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206663 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r206611.
zalan@apple.com [Fri, 30 Sep 2016 20:58:29 +0000 (20:58 +0000)]
Unreviewed, rolling out r206611.

Scroll perf did not recover.

Reverted changeset:

"Unreviewed, rolling out r206483."
https://bugs.webkit.org/show_bug.cgi?id=162750
http://trac.webkit.org/changeset/206611

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206662 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove ClipRects's custom refcounting.
zalan@apple.com [Fri, 30 Sep 2016 20:52:02 +0000 (20:52 +0000)]
Remove ClipRects's custom refcounting.
https://bugs.webkit.org/show_bug.cgi?id=162798

Reviewed by Simon Fraser.

It's safer to use RefCounted<>.

No change in functionality.

* rendering/RenderLayer.cpp:
(WebCore::ClipRects::ClipRects):
(WebCore::ClipRectsCache::getClipRects):
(WebCore::ClipRectsCache::setClipRects):
(WebCore::ClipRectsCache::getIndex):
(WebCore::RenderLayer::updateClipRects):
(WebCore::ClipRects::ref): Deleted.
(WebCore::ClipRects::deref): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206661 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRebaseline js/dom/stack-trace.html after r206654.
ryanhaddad@apple.com [Fri, 30 Sep 2016 20:47:51 +0000 (20:47 +0000)]
Rebaseline js/dom/stack-trace.html after r206654.

Unreviewed test gardening.

* js/dom/stack-trace-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206660 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFileSaver.js does not work in WebKit
cdumez@apple.com [Fri, 30 Sep 2016 20:08:27 +0000 (20:08 +0000)]
FileSaver.js does not work in WebKit
https://bugs.webkit.org/show_bug.cgi?id=162788

Reviewed by Sam Weinig.

Source/WebCore:

FileSaver.js does not work in WebKit:
- https://eligrey.com/demos/FileSaver.js/

It works in Firefox and Chrome, but in WebKit, we were getting a
"Synthetic clicks on anchors that have a download attribute are
ignored." warning. We were too strict in restricting synthetic clicks.
We now allow synthetic clicks as long as they are triggered by a user
gesture.

Test: fast/dom/HTMLAnchorElement/anchor-download-user-triggered-synthetic-click.html

* html/HTMLAnchorElement.cpp:
(WebCore::HTMLAnchorElement::handleClick):

LayoutTests:

Add layout test coverage.

* fast/dom/HTMLAnchorElement/anchor-download-synthetic-click-expected.txt:
* fast/dom/HTMLAnchorElement/anchor-download-user-triggered-synthetic-click-expected.txt: Added.
* fast/dom/HTMLAnchorElement/anchor-download-user-triggered-synthetic-click.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206659 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove the dumping of the stack back trace in VM::verifyExceptionCheckNeedIsSatisfied().
mark.lam@apple.com [Fri, 30 Sep 2016 19:47:54 +0000 (19:47 +0000)]
Remove the dumping of the stack back trace in VM::verifyExceptionCheckNeedIsSatisfied().
https://bugs.webkit.org/show_bug.cgi?id=162797

Reviewed by Geoffrey Garen.

This is because the RELEASE_ASSERT() that follows immediately after will also
dump the stack back trace.  Hence, the first dump will be redundant.

Also removed an extra space in the dataLog output.

* runtime/VM.cpp:
(JSC::VM::verifyExceptionCheckNeedIsSatisfied):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206658 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMarking fast/images/gif-loop-count.html as flaky on ios-simulator.
ryanhaddad@apple.com [Fri, 30 Sep 2016 19:45:58 +0000 (19:45 +0000)]
Marking fast/images/gif-loop-count.html as flaky on ios-simulator.
https://bugs.webkit.org/show_bug.cgi?id=162739

Unreviewed test gardening.

* platform/ios-simulator-wk1/TestExpectations:
* platform/ios-simulator/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206657 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMarking http/tests/security/cached-cross-origin-preloading-css-stylesheet.html as...
ryanhaddad@apple.com [Fri, 30 Sep 2016 19:43:46 +0000 (19:43 +0000)]
Marking http/tests/security/cached-cross-origin-preloading-css-stylesheet.html as flaky on mac-wk1.
https://bugs.webkit.org/show_bug.cgi?id=162791

Unreviewed test gardening.

* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206656 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Make debugger stepping highlights work in inline <script>s
joepeck@webkit.org [Fri, 30 Sep 2016 19:24:32 +0000 (19:24 +0000)]
Web Inspector: Make debugger stepping highlights work in inline <script>s
https://bugs.webkit.org/show_bug.cgi?id=162753
<rdar://problem/28551332>

Reviewed by Brian Burg.

* UserInterface/Models/TextRange.js:
(WebInspector.TextRange.prototype.contains):
Check if a given line/column falls within this range.

* UserInterface/Views/SourceCodeTextEditor.js:
(WebInspector.SourceCodeTextEditor.prototype._getAssociatedScript):
If we are in a Document resource find the associated script at a given position.

(WebInspector.SourceCodeTextEditor.prototype.textEditorExecutionHighlightRange):
When comparing offsets to SyntaxTree offsets, the SyntaxTree's offset of 0 is the
first character of the Script, which differs from the current SourceCode's offset.
Adjust the offset by the Script's startOffset.

* UserInterface/Views/TextEditor.js:
(WebInspector.TextEditor.prototype.currentPositionToOriginalPosition):
(WebInspector.TextEditor.prototype._updateExecutionRangeHighlight):
Pass both the original offset and original position to the delegate.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206655 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Stepping through `a(); b(); c();` it is unclear where we are and what...
joepeck@webkit.org [Fri, 30 Sep 2016 19:24:29 +0000 (19:24 +0000)]
Web Inspector: Stepping through `a(); b(); c();` it is unclear where we are and what is about to execute
https://bugs.webkit.org/show_bug.cgi?id=161658
<rdar://problem/28181254>

Reviewed by Geoffrey Garen.

Source/JavaScriptCore:

* parser/Parser.cpp:
(JSC::Parser<LexerType>::parseAssignmentExpression):
Updated pause location for unary expressions.

Source/WebInspectorUI:

* UserInterface/Models/Script.js:
(WebInspector.Script.prototype.requestScriptSyntaxTree):
Fix first calls to requestScriptSyntaxTree. They were getting an uncaught
exception because the content argument was missing.

* UserInterface/Models/ScriptSyntaxTree.js:
(WebInspector.ScriptSyntaxTree.prototype.containersOfOffset):
Find all AST nodes that contain a particular offset.

* UserInterface/Views/SourceCodeTextEditor.js:
(WebInspector.SourceCodeTextEditor):
(WebInspector.SourceCodeTextEditor.prototype.close):
(WebInspector.SourceCodeTextEditor.prototype._activeCallFrameDidChange):
(WebInspector.SourceCodeTextEditor.prototype._activeCallFrameSourceCodeLocationChanged):
(WebInspector.SourceCodeTextEditor.prototype.textEditorExecutionHighlightRange):
Provide a good highlight range for a given offset. This is normally the start
of a statement/expression, inside a statement/expression, or the closing brace
of a function (leaving a function). Provide good ranges for each of these.

* UserInterface/Views/TextEditor.js:
(WebInspector.TextEditor):
(WebInspector.TextEditor.set string.update):
(WebInspector.TextEditor.prototype.set string):
(WebInspector.TextEditor.prototype.setExecutionLineAndColumn):
(WebInspector.TextEditor.prototype.revealPosition.revealAndHighlightLine):
(WebInspector.TextEditor.prototype.revealPosition):
(WebInspector.TextEditor.prototype.currentPositionToOriginalOffset):
(WebInspector.TextEditor.prototype._updateAfterFormatting):
(WebInspector.TextEditor.prototype.set executionLineNumber): Deleted.
(WebInspector.TextEditor.prototype.set executionColumnNumber): Deleted.
(WebInspector.TextEditor.prototype._updateExecutionLine.update): Deleted.
Always set the execution line and column together, to simplify how we update highlights.

(WebInspector.TextEditor.prototype._clearMultilineExecutionLineHighlights):
(WebInspector.TextEditor.prototype._updateExecutionLine):
When updating the main highlight clear any multi-line highlights.

(WebInspector.TextEditor.prototype._updateExecutionRangeHighlight):
Ask the delegate for a specific highlight range. If provided use that range,
otherwise just highlight the end of the line. Once we know the range, if it
is multiple lines, give the extra lines the full line highlight as well.
Also make adjustments, such as not highlighting trailing whitespace.

* UserInterface/Views/TextEditor.css:
(.text-editor > .CodeMirror .execution-line.primary .CodeMirror-linenumber::after):
(.text-editor > .CodeMirror .execution-line):
(.text-editor > .CodeMirror .execution-line .CodeMirror-matchingbracket):
(.text-editor > .CodeMirror .execution-range-highlight):
Styles for execution lines and execution range highlights.

LayoutTests:

* inspector/debugger/stepping/stepping-loops-expected.txt:
* inspector/debugger/stepping/stepping-misc-expected.txt:
Updated pause location for unary expressions.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206654 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoBreakpoints on blank lines or comments don't break
joepeck@webkit.org [Fri, 30 Sep 2016 19:22:50 +0000 (19:22 +0000)]
Breakpoints on blank lines or comments don't break
https://bugs.webkit.org/show_bug.cgi?id=9885
<rdar://problem/6134406>

Reviewed by Mark Lam.

Source/JavaScriptCore:

This change introduces a way to perform a Debugger Parse of a script.
This debugger parse gathers a list of breakpoint locations, which
the backend uses to resolve breakpoint locations that came from the
Inspector frontend to the exact location we would actually pause.
We gather this information from the parser so that we can eagerly
get this information without requiring the code to have executed (the
real op_debugs are generated during bytecode generation when code
is actually evaluated).

If an input location was on a line with whitespace or a comment, the
resolved breakpoint location would be before the next statement that
will be executed. That may be the next line, or even later. We also
update our policy when setting breakpoints on and around function
statements to better match user expectations.

For example, when resolving breakpoints in:

    1.  // Comment
    2.  before;
    3.
    4.  function foo() {
    5.      inside;
    6.  }
    7.
    8.  after;

A breakpoint on line 1, a comment, resolves to line 2 the next
statement that will execute.

A breakpoint on line 3 or 7, empty lines, resolves to line 8 the next
statement that will execute. This skips past the definition of foo,
just like stepping would have done. The creation of foo would have
been hoisted, which would have happened before execution of the
other statements.

A breakpoint on line 4, a function signature, resolves to line 5,
inside the function. Users would expect to pause inside of a function
when setting a breakpoint on that function's name or opening brace.

A breakpoint on line 6, a function's closing brace, resolves to
line 6. The debugger will pause whenever execution leaves foo due to
a return and not an exception. This matches stepping behavior. An
explicit or implicit return (the implicit return undefined) will
pause on the closing brace as we leave the function, giving users
an opportunity to inspect the final state before leaving.

--

At this point, op_debug's are still emitted at custom locations during
bytecode generation of other statements / expressions. In order to
ensure the generated op_debugs correspond to locations the Parser
determined were breakpoint locations, the Parser sets a "needs debug
hook" flag on the nodes it will use for breakpoint locations, and
we assert during bytecode generation that op_debugs are only emitted
for nodes that were marked as needing debug hooks.

This still leaves open the possibility that the Parser will mark
some nodes that get missed during bytecode generation, so we might
fail to emit some op_debugs. The next step will be eliminating the
custom emitDebugHooks spread across StatementNode and ExpressionNode
subclasses, and instead always generating op_debugs whenever we
emit a flagged node.

--

* CMakeLists.txt:
* JavaScriptCore.xcodeproj/project.pbxproj:
New DebuggerParseData files.

* API/JSScriptRef.cpp:
(OpaqueJSScript::OpaqueJSScript):
* jsc.cpp:
(functionCheckModuleSyntax):
* parser/SourceCode.h:
(JSC::makeSource):
* parser/SourceProvider.cpp:
(JSC::SourceProvider::SourceProvider):
* parser/SourceProvider.h:
(JSC::SourceProvider::sourceType):
(JSC::StringSourceProvider::create):
(JSC::StringSourceProvider::StringSourceProvider):
(JSC::WebAssemblySourceProvider::WebAssemblySourceProvider):
(JSC::SourceProvider::startPosition): Deleted.
Add a new type on SourceProvider to distinguish if its script was
intended to be a Script, Module, or WebAssembly. This information
will be needed to know how to best parse this file when the
debugger decides to lazily parse.

* runtime/Executable.cpp:
(JSC::EvalExecutable::EvalExecutable):
(JSC::ProgramExecutable::ProgramExecutable):
(JSC::ModuleProgramExecutable::ModuleProgramExecutable):
(JSC::WebAssemblyExecutable::WebAssemblyExecutable):
* runtime/ModuleLoaderPrototype.cpp:
(JSC::moduleLoaderPrototypeParseModule):
ASSERT the SourceProvider type matches the executable type we are
creating for it.

* parser/ASTBuilder.h:
(JSC::ASTBuilder::breakpointLocation):
* parser/SyntaxChecker.h:
(JSC::SyntaxChecker::operatorStackPop):
When gathering breakpoint positions, get the position from the
current node. In the SyntaxChecker, return an invalid position.

* parser/Nodes.h:
(JSC::ExpressionNode::needsDebugHook):
(JSC::ExpressionNode::setNeedsDebugHook):
(JSC::StatementNode::needsDebugHook):
(JSC::StatementNode::setNeedsDebugHook):
When gathering breakpoint positions, mark the node as needing
a debug hook. For now we assert op_debugs generated must come
from these nodes. Later we should just generate op_debugs for
these nodes.

* parser/Parser.cpp:
(JSC::Parser<LexerType>::Parser):
(JSC::Parser<LexerType>::parseStatementListItem):
(JSC::Parser<LexerType>::parseDoWhileStatement):
(JSC::Parser<LexerType>::parseWhileStatement):
(JSC::Parser<LexerType>::parseArrowFunctionSingleExpressionBodySourceElements):
(JSC::Parser<LexerType>::parseForStatement):
(JSC::Parser<LexerType>::parseWithStatement):
(JSC::Parser<LexerType>::parseSwitchStatement):
(JSC::Parser<LexerType>::parseStatement):
(JSC::Parser<LexerType>::parseFunctionBody):
(JSC::Parser<LexerType>::parseFunctionInfo):
(JSC::Parser<LexerType>::parseIfStatement):
(JSC::Parser<LexerType>::parseAssignmentExpression):
* parser/Parser.h:
(JSC::parse):
Add an optional DebuggerParseData struct to the Parser. When available
the Parser will gather debugger data, and parse all functions with the
ASTBuilder instead of SyntaxChecking inner functions.

* debugger/DebuggerParseData.cpp: Added.
(JSC::DebuggerPausePositions::breakpointLocationForLineColumn):
(JSC::DebuggerPausePositions::sort):
(JSC::gatherDebuggerParseData):
(JSC::gatherDebuggerParseDataForSource):
* debugger/DebuggerParseData.h: Copied from Source/JavaScriptCore/debugger/DebuggerPrimitives.h.
(JSC::DebuggerPausePositions::DebuggerPausePositions):
(JSC::DebuggerPausePositions::appendPause):
(JSC::DebuggerPausePositions::appendEntry):
(JSC::DebuggerPausePositions::appendLeave):
The DebuggerParseData struct currently only contains a list of pause positions.
Once populated it can resolve an input location to a pause position.

* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitCall):
(JSC::BytecodeGenerator::emitCallVarargs):
(JSC::BytecodeGenerator::emitDebugHook):
(JSC::BytecodeGenerator::emitEnumeration):
* bytecompiler/BytecodeGenerator.h:
* bytecompiler/NodesCodegen.cpp:
(JSC::EmptyStatementNode::emitBytecode):
(JSC::DebuggerStatementNode::emitBytecode):
(JSC::ExprStatementNode::emitBytecode):
(JSC::DeclarationStatement::emitBytecode):
(JSC::IfElseNode::emitBytecode):
(JSC::DoWhileNode::emitBytecode):
(JSC::WhileNode::emitBytecode):
(JSC::ForNode::emitBytecode):
(JSC::ForInNode::emitBytecode):
(JSC::ContinueNode::emitBytecode):
(JSC::BreakNode::emitBytecode):
(JSC::ReturnNode::emitBytecode):
(JSC::WithNode::emitBytecode):
(JSC::SwitchNode::emitBytecode):
(JSC::ThrowNode::emitBytecode):
Emit op_debugs for the nodes themselves. Assert when we do that the
Parser had marked them as needing a debug hook.

* debugger/Breakpoint.h:
(JSC::Breakpoint::Breakpoint):
A breakpoint may be resolved or unresolved. Debugger::resolveBreakpoint
must be used to resolve the breakpoint. Most methods now require a
resolved breakpoint.

* debugger/Debugger.h:
* debugger/Debugger.cpp:
(JSC::Debugger::detach):
(JSC::Debugger::toggleBreakpoint):
(JSC::Debugger::debuggerParseData):
(JSC::Debugger::resolveBreakpoint):
(JSC::Debugger::setBreakpoint):
(JSC::Debugger::clearParsedData):
Provide a public method to resolve a breakpoint location in a script.
This will gather debugger parse data for the script if none is available.
Ensure clients have resolved a breakpoint before attempting to set it.
Currently we allow only a single breakpoint at a location. This may
need to change if multiple breakpoints resolve to the same location
but have different actions.

* inspector/ScriptDebugListener.h:
ScriptDebugServer::Script is effectively duplicating most of the data from
a SourceProvider. We should eliminate this and just use SourceProvider.

* inspector/ScriptDebugServer.cpp:
(Inspector::ScriptDebugServer::setBreakpointActions):
(Inspector::ScriptDebugServer::removeBreakpointActions):
(Inspector::ScriptDebugServer::getActionsForBreakpoint):
(Inspector::ScriptDebugServer::clearBreakpointActions):
(Inspector::ScriptDebugServer::evaluateBreakpointAction):
(Inspector::ScriptDebugServer::dispatchDidParseSource):
(Inspector::ScriptDebugServer::handleBreakpointHit):
(Inspector::ScriptDebugServer::setBreakpoint): Deleted.
(Inspector::ScriptDebugServer::removeBreakpoint): Deleted.
(Inspector::ScriptDebugServer::clearBreakpoints): Deleted.
* inspector/ScriptDebugServer.h:
Reduce ScriptDebugServer's involvement in breakpoints to just handling
breakpoint actions. Eventually we should eliminate it alltogether and
fold breakpoint logic into Debugger or DebugAgent.

* inspector/agents/InspectorDebuggerAgent.h:
* inspector/agents/InspectorDebuggerAgent.cpp:
(Inspector::buildDebuggerLocation):
(Inspector::parseLocation):
(Inspector::InspectorDebuggerAgent::setBreakpointByUrl):
(Inspector::InspectorDebuggerAgent::setBreakpoint):
(Inspector::InspectorDebuggerAgent::didSetBreakpoint):
(Inspector::InspectorDebuggerAgent::resolveBreakpoint):
(Inspector::InspectorDebuggerAgent::removeBreakpoint):
(Inspector::InspectorDebuggerAgent::continueToLocation):
(Inspector::InspectorDebuggerAgent::didParseSource):
(Inspector::InspectorDebuggerAgent::clearDebuggerBreakpointState):
The Inspector can set breakpoints in multiple ways.
Ensure that once we have the Script that we always
resolve the breakpoint location before setting the
breakpoint. The different paths are:

- setBreakpoint(scriptId, location)
  - Here we know the SourceProvider by its SourceID
    - resolve and set

- setBreakpointByURL(url, location)
  - Search for existing Scripts that match the URL
    - resolve in each and set
  - When new Scripts are parsed that match the URL
    - resolve and set

Source/WebCore:

Tests: inspector/debugger/breakpoints/resolved-dump-all-pause-locations.html
       inspector/debugger/breakpoints/resolved-dump-each-line.html

* bindings/js/CachedScriptSourceProvider.h:
(WebCore::CachedScriptSourceProvider::CachedScriptSourceProvider):

LayoutTests:

* inspector/debugger/breakpoints/resolved-dump-all-pause-locations-expected.txt: Added.
* inspector/debugger/breakpoints/resolved-dump-all-pause-locations.html: Added.
* inspector/debugger/breakpoints/resolved-dump-each-line-expected.txt: Added.
* inspector/debugger/breakpoints/resolved-dump-each-line.html: Added.
* inspector/debugger/breakpoints/resources/dump-functions.js: Added.
* inspector/debugger/breakpoints/resources/dump-general.js: Added.
Test for resolved breakpoint locations in all kinds of different source code.

* inspector/debugger/breakpoints/resources/dump.js: Added.
(TestPage.registerInitializer):
(TestPage.registerInitializer.window.addDumpAllPauseLocationsTestCase):
(TestPage.registerInitializer.window.addDumpEachLinePauseLocationTestCase):
Shared code to run different generalized tests for logging all resolved
breakpoint locations or the resolved breakpoint location if a breakpoint
is set on each individual line.

* inspector/debugger/resources/log-pause-location.js:
(TestPage.registerInitializer.insertCaretIntoStringAtIndex):
(TestPage.registerInitializer.window.findScript):
(TestPage.registerInitializer.window.loadLinesFromSourceCode):
(TestPage.registerInitializer.window.loadMainPageContent):
(TestPage.registerInitializer.window.logResolvedBreakpointLinesWithContext):
(TestPage.registerInitializer.window.logLinesWithContext):
Make some more code shared and provide a way to log two locations,
used to see where a breakpoint was set and where it resolved to.

* inspector/debugger/setBreakpoint-expected.txt:
Update error message. Should not include a period.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206653 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Stepping out of a function finishes the line that called it.
joepeck@webkit.org [Fri, 30 Sep 2016 19:22:37 +0000 (19:22 +0000)]
Web Inspector: Stepping out of a function finishes the line that called it.
https://bugs.webkit.org/show_bug.cgi?id=155325
<rdar://problem/25094578>

Reviewed by Mark Lam.

Source/JavaScriptCore:

Also addresses:
<https://webkit.org/b/161721> Web Inspector: Stepping all the way through program should not cause a pause on the next program that executes
<https://webkit.org/b/161716> Web Inspector: Stepping into a function / program should not require stepping to the first statement

This change introduces a new op_debug hook: WillExecuteExpression.
Currently this new hook is only used for pausing at function calls.
We may decide to add it to other places later where pausing with
finer granularity then statements (or lines) if useful.

This updates the location and behavior of some of the existing debug
hooks, to be more consistent and useful if the exact location of the
pause is displayed. For example, in control flow statements like
`if` and `while`, the pause location is the expression itself that
will be evaluated, not the location of the `if` or `while` keyword.
For example:

    if (|condition)
    while (|condition)

Finally, this change gets rid of some unnecessary / useless pause
locations such as on entering a function and on entering a program.
These pauses are not needed because if there is a statement, we
would pause before the statement and it is equivalent. We continue
to pause when leaving a function via stepping by uniformly jumping
to the closing brace of the function. This gives users a chance
to observe state before leaving the function.

* bytecode/CodeBlock.cpp:
(JSC::debugHookName):
* bytecode/UnlinkedCodeBlock.cpp:
(JSC::dumpLineColumnEntry):
Logging strings for the new debug hook.

* bytecompiler/BytecodeGenerator.h:
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitCallInTailPosition):
(JSC::BytecodeGenerator::emitCallEval):
(JSC::BytecodeGenerator::emitCallVarargsInTailPosition):
(JSC::BytecodeGenerator::emitConstructVarargs):
(JSC::BytecodeGenerator::emitCallForwardArgumentsInTailPosition):
(JSC::BytecodeGenerator::emitCallDefineProperty):
(JSC::BytecodeGenerator::emitConstruct):
(JSC::BytecodeGenerator::emitGetTemplateObject):
(JSC::BytecodeGenerator::emitIteratorNext):
(JSC::BytecodeGenerator::emitIteratorNextWithValue):
(JSC::BytecodeGenerator::emitIteratorClose):
(JSC::BytecodeGenerator::emitDelegateYield):
All emitCall variants now take an enum to decide whether or not to
emit the WillExecuteExpression debug hook.

(JSC::BytecodeGenerator::emitCall):
(JSC::BytecodeGenerator::emitCallVarargs):
In the two real implementations, actually decide to emit the debug
hook or not based on the parameter.

(JSC::BytecodeGenerator::emitEnumeration):
This is shared looping code used by for..of iteration of iterables.
When used by ForOfNode, we want to emit a pause location during
iteration.

(JSC::BytecodeGenerator::emitWillLeaveCallFrameDebugHook):
This is shared call frame leave code to emit a consistent pause
location when leaving a function.

* bytecompiler/NodesCodegen.cpp:
(JSC::EvalFunctionCallNode::emitBytecode):
(JSC::FunctionCallValueNode::emitBytecode):
(JSC::FunctionCallResolveNode::emitBytecode):
(JSC::BytecodeIntrinsicNode::emit_intrinsic_tailCallForwardArguments):
(JSC::FunctionCallBracketNode::emitBytecode):
(JSC::FunctionCallDotNode::emitBytecode):
(JSC::CallFunctionCallDotNode::emitBytecode):
(JSC::ApplyFunctionCallDotNode::emitBytecode):
(JSC::TaggedTemplateNode::emitBytecode):
(JSC::ArrayPatternNode::bindValue):
All tail position calls are the function calls that we want to emit
debug hooks for. All non-tail call calls appear to be internal
implementation details, and these should not have the debug hook.

(JSC::IfElseNode::emitBytecode):
(JSC::WhileNode::emitBytecode):
(JSC::WithNode::emitBytecode):
(JSC::SwitchNode::emitBytecode):
Make the pause location consistent at the expression.

(JSC::DoWhileNode::emitBytecode):
Make the pause location consistent at the expression.
Remove the errant pause at the do's '}' when entering the do block.

(JSC::ForNode::emitBytecode):
(JSC::ForInNode::emitMultiLoopBytecode):
(JSC::ForOfNode::emitBytecode):
Make the pause location consistent at expressions.
Also allow stepping to the traditional for loop's
update expression, which was previously not possible.

(JSC::ReturnNode::emitBytecode):
(JSC::FunctionNode::emitBytecode):
Make the pause location when leaving a function consistently be the
function's closing brace. The two cases are stepping through a return
statement, or the implicit return undefined at the end of a function.

(JSC::LabelNode::emitBytecode):
(JSC::TryNode::emitBytecode):
Remove unnecessary pauses that add no value, as they contain a
statement and we will then pause at that statement.

* parser/Nodes.h:
(JSC::StatementNode::isFunctionNode):
(JSC::StatementNode::isForOfNode):
(JSC::EnumerationNode::lexpr):
(JSC::ForOfNode::isForOfNode):
New virtual methods to distinguish different nodes.

* debugger/Debugger.h:
Rename m_pauseAtNextStatement to m_pauseAtNextOpportunity.
This is the finest granularity of stepping, and it can be
pausing at a location that is not a statement.
Introduce state to properly handle step out and stepping
when there are multiple expressions in a statement.

* debugger/Debugger.cpp:
(JSC::Debugger::Debugger):
(JSC::Debugger::setPauseOnNextStatement):
(JSC::Debugger::breakProgram):
(JSC::Debugger::continueProgram):
(JSC::Debugger::stepIntoStatement):
(JSC::Debugger::exception):
(JSC::Debugger::didReachBreakpoint):

Use new variable names, and clarify if we should attempt
to pause or not.

(JSC::Debugger::stepOutOfFunction):
Set a new state to indicate a step out action.

(JSC::Debugger::updateCallFrame):
(JSC::Debugger::updateCallFrameAndPauseIfNeeded): Deleted.
(JSC::Debugger::updateCallFrameInternal):
(JSC::Debugger::pauseIfNeeded):
Allow updateCallFrame to either attempt a pause or not.

(JSC::Debugger::atStatement):
Attempt pause and reset the at first expression flag.

(JSC::Debugger::atExpression):
Attempt a pause when not stepping over. Also skip
the first expression pause, since that would be
equivalent to when we paused for the expression.

(JSC::Debugger::callEvent):
Do not pause when entering a function.

(JSC::Debugger::returnEvent):
Attempt pause when leaving a function.
If the user did a step-over and is leaving the
function, then behave like step-out.

(JSC::Debugger::unwindEvent):
Behave like return except don't change any
pausing states. If we needed to pause the
Debugger::exception will have handled it.

(JSC::Debugger::willExecuteProgram):
Do not pause when entering a program.

(JSC::Debugger::didExecuteProgram):
Attempt pause when leaving a program that has a caller.
This can be useful for exiting an eval(...) program.
Otherwise treat this like return, and step-over out
of the program should behave like step-out. We use
pause at next opportunity because there may be extra
callframes we do not know about.
When the program doesn't have a parent, clear all
our state so we don't errantly pause on the next
JavaScript microtask that gets executed.

(JSC::Debugger::clearNextPauseState):
Helper to clear all of the pause states now that
it happens in a couple places.

* interpreter/Interpreter.cpp:
(JSC::notifyDebuggerOfUnwinding):
Treat unwinding slightly differently from returning.
We will not want to pause when unwinding callframes.

(JSC::Interpreter::debug):
* interpreter/Interpreter.h:
New debug hook.

* inspector/agents/InspectorDebuggerAgent.cpp:
(Inspector::InspectorDebuggerAgent::stepInto):
(Inspector::InspectorDebuggerAgent::didPause):
* inspector/agents/InspectorDebuggerAgent.h:
Remove unnecessary stepInto code notification for listeners.
The listeners are never notified if the debugger resumes,
so whatever state they were setting by this is going to
get out of date.

Source/WebCore:

Tests: inspector/debugger/stepping/stepInto.html
       inspector/debugger/stepping/stepOut.html
       inspector/debugger/stepping/stepOver.html
       inspector/debugger/stepping/stepping-arrow-functions.html
       inspector/debugger/stepping/stepping-classes.html
       inspector/debugger/stepping/stepping-control-flow.html
       inspector/debugger/stepping/stepping-function-calls.html
       inspector/debugger/stepping/stepping-function-default-parameters.html
       inspector/debugger/stepping/stepping-literal-construction.html
       inspector/debugger/stepping/stepping-loops.html
       inspector/debugger/stepping/stepping-misc.html
       inspector/debugger/stepping/stepping-switch.html
       inspector/debugger/stepping/stepping-template-string.html
       inspector/debugger/stepping/stepping-try-catch-finally.html

* inspector/InspectorDOMDebuggerAgent.h:
* inspector/InspectorDOMDebuggerAgent.cpp:
(WebCore::InspectorDOMDebuggerAgent::stepInto): Deleted.
Setting this state in step-into does not make sense since we do not
know when the debugger resumes and won't know when to clear it.

LayoutTests:

* inspector/debugger/break-on-exception-throw-in-promise.html:
Drive-by remove debug only code that shouldn't have been checked in.

* inspector/debugger/resources/log-pause-location.js: Added.
(TestPage.registerInitializer.String.prototype.myPadStart):
(TestPage.registerInitializer.insertCaretIntoStringAtIndex):
(TestPage.registerInitializer.logLinesWithContext):
(TestPage.registerInitializer.window.logPauseLocation):
(TestPage.registerInitializer.window.step):
(TestPage.registerInitializer.window.initializeSteppingTestSuite):
(TestPage.registerInitializer.window.addSteppingTestCase):
(TestPage.registerInitializer.window.loadMainPageContent):
Shared code for stepping tests that runs in the inspected page.

(global):
When the test page is loaded outside of the test runner,
create buttons for each of the different entry test functions.
This makes it very easy to inspect the test page and run
through an individual test.

* inspector/debugger/stepping/stepInto-expected.txt: Added.
* inspector/debugger/stepping/stepInto.html: Added.
* inspector/debugger/stepping/stepOut-expected.txt: Added.
* inspector/debugger/stepping/stepOut.html: Added.
* inspector/debugger/stepping/stepOver-expected.txt: Added.
* inspector/debugger/stepping/stepOver.html: Added.
* inspector/debugger/stepping/stepping-arrow-functions-expected.txt: Added.
* inspector/debugger/stepping/stepping-arrow-functions.html: Added.
* inspector/debugger/stepping/stepping-classes-expected.txt: Added.
* inspector/debugger/stepping/stepping-classes.html: Added.
* inspector/debugger/stepping/stepping-control-flow-expected.txt: Added.
* inspector/debugger/stepping/stepping-control-flow.html: Added.
* inspector/debugger/stepping/stepping-function-calls-expected.txt: Added.
* inspector/debugger/stepping/stepping-function-calls.html: Added.
* inspector/debugger/stepping/stepping-function-default-parameters-expected.txt: Added.
* inspector/debugger/stepping/stepping-function-default-parameters.html: Added.
* inspector/debugger/stepping/stepping-literal-construction-expected.txt: Added.
* inspector/debugger/stepping/stepping-literal-construction.html: Added.
* inspector/debugger/stepping/stepping-loops-expected.txt: Added.
* inspector/debugger/stepping/stepping-loops.html: Added.
* inspector/debugger/stepping/stepping-misc-expected.txt: Added.
* inspector/debugger/stepping/stepping-misc.html: Added.
* inspector/debugger/stepping/stepping-switch-expected.txt: Added.
* inspector/debugger/stepping/stepping-switch.html: Added.
* inspector/debugger/stepping/stepping-template-string-expected.txt: Added.
* inspector/debugger/stepping/stepping-template-string.html: Added.
* inspector/debugger/stepping/stepping-try-catch-finally-expected.txt: Added.
* inspector/debugger/stepping/stepping-try-catch-finally.html: Added.
Test stepping in different common scenarios.

* inspector/debugger/regress-133182.html:
* inspector/debugger/regress-133182-expected.txt:
* inspector/debugger/tail-deleted-frames-from-vm-entry-expected.txt:
* inspector/debugger/tail-deleted-frames-from-vm-entry.html:
Rebaseline. No need for a double step. And the second pause doesn't make any sense
in the tail deleted frames test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206652 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[WK2][iOS] Add radiusX / radiusY / rotationAngle to WebPlatformTouchPoint
cdumez@apple.com [Fri, 30 Sep 2016 19:19:44 +0000 (19:19 +0000)]
[WK2][iOS] Add radiusX / radiusY / rotationAngle to WebPlatformTouchPoint
https://bugs.webkit.org/show_bug.cgi?id=162787
<rdar://problem/28554292>

Reviewed by Benjamin Poulain.

Source/WebKit2:

Add radiusX / radiusY / rotationAngle to WebPlatformTouchPoint.

* Platform/spi/ios/UIKitSPI.h:
* Shared/WebEvent.h:
(WebKit::WebPlatformTouchPoint::setRadiusX):
(WebKit::WebPlatformTouchPoint::radiusX):
(WebKit::WebPlatformTouchPoint::setRadiusY):
(WebKit::WebPlatformTouchPoint::radiusY):
(WebKit::WebPlatformTouchPoint::setRotationAngle):
(WebKit::WebPlatformTouchPoint::rotationAngle):
* Shared/WebEventConversion.cpp:
(WebKit::WebKit2PlatformTouchPoint::WebKit2PlatformTouchPoint):
* Shared/ios/NativeWebTouchEventIOS.mm:
(WebKit::NativeWebTouchEvent::extractWebTouchPoint):
* Shared/ios/WebPlatformTouchPointIOS.cpp:
(WebKit::WebPlatformTouchPoint::encode):
(WebKit::WebPlatformTouchPoint::decode):

LayoutTests:

Extend Touch constructor test coverage to cover radiusX / radiusY / rotationAngle
attributes.

* fast/events/touch/touch-constructor.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206651 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix off-by-one error in URLParser::parseIPv4Host
achristensen@apple.com [Fri, 30 Sep 2016 18:50:49 +0000 (18:50 +0000)]
Fix off-by-one error in URLParser::parseIPv4Host
https://bugs.webkit.org/show_bug.cgi?id=162789

Reviewed by Tim Horton.

Source/WebCore:

The spec says "If any but the last item in numbers is greater than 255, return failure."
This means check up to size - 1, not size - 2.

Covered by a new API test.

* platform/URLParser.cpp:
(WebCore::URLParser::parseIPv4Host):

Tools:

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206650 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoURLParser: parsing a URL with an empty host and a colon should fail
achristensen@apple.com [Fri, 30 Sep 2016 18:48:28 +0000 (18:48 +0000)]
URLParser: parsing a URL with an empty host and a colon should fail
https://bugs.webkit.org/show_bug.cgi?id=162795

Reviewed by Tim Horton.

Source/WebCore:

Covered by new API tests.

* platform/URLParser.cpp:
(WebCore::URLParser::parseHostAndPort):

Tools:

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206649 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoURLParser: handle syntax violations in non-UTF-8 encoded queries
achristensen@apple.com [Fri, 30 Sep 2016 18:46:28 +0000 (18:46 +0000)]
URLParser: handle syntax violations in non-UTF-8 encoded queries
https://bugs.webkit.org/show_bug.cgi?id=162770

Reviewed by Tim Horton.

Source/WebCore:

There is a fast path for queries of URLs that use UTF-8 encoding, which are quite common.
For non-UTF-8 encoded queries, which are less common, we put the code points in a Vector<UChar>
and encode them all at once.  If there is a syntax violation in the query, we need to copy the
syntax-violation-free string up to the beginning of the query, then encode the query.

Covered by new API tests.

* platform/URLParser.cpp:
(WebCore::URLParser::percentEncodeByte):
(WebCore::URLParser::encodeQuery):
(WebCore::URLParser::parse):
* platform/URLParser.h:

Tools:

* TestWebKitAPI/Tests/WebCore/URLParser.cpp:
(TestWebKitAPI::checkURL):
(TestWebKitAPI::TEST_F):
Tests with emoji change behavior when we insert a tab between the surrogates, so don't do the
insert-tab-at-each-location verification that syntax violations are handled correctly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206648 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoArrow functions should not allow duplicate parameter names
sbarati@apple.com [Fri, 30 Sep 2016 18:44:16 +0000 (18:44 +0000)]
Arrow functions should not allow duplicate parameter names
https://bugs.webkit.org/show_bug.cgi?id=162741

Reviewed by Filip Pizlo.

JSTests:

* test262.yaml:

Source/JavaScriptCore:

This patch makes parsing arrow function parameters throw
a syntax error when there are duplicate parameter names.
It also starts to make some syntax errors for arrow functions
better, however, this is trickier than it seems since we need
to choose between two parsing productions when we decide to
throw a syntax error. I'm going to work on this problem
in another patch specifically devoted to making the error
messages better for parsing arrow functions:
https://bugs.webkit.org/show_bug.cgi?id=162794

* parser/Parser.cpp:
(JSC::Parser<LexerType>::isArrowFunctionParameters):
(JSC::Parser<LexerType>::parseFormalParameters):
(JSC::Parser<LexerType>::parseFunctionParameters):
(JSC::Parser<LexerType>::parseAssignmentExpression):
* parser/Parser.h:

LayoutTests:

* js/parser-syntax-check-expected.txt:
* js/script-tests/parser-syntax-check.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206647 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMake some microbenchmarks run for less time.
sbarati@apple.com [Fri, 30 Sep 2016 18:39:59 +0000 (18:39 +0000)]
Make some microbenchmarks run for less time.

Rubber stamped by Filip Pizlo.

* microbenchmarks/bound-function-construction-performance.js:
(foo):
* microbenchmarks/getter-richards-try-catch.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206646 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMake it possible to test web-related user-interface features
commit-queue@webkit.org [Fri, 30 Sep 2016 18:39:30 +0000 (18:39 +0000)]
Make it possible to test web-related user-interface features
https://bugs.webkit.org/show_bug.cgi?id=162657

Patch by Megan Gardner <megan_gardner@apple.com> on 2016-09-30
Reviewed by Simon Fraser.

Source/WebKit2:

Added the ability to pull the strings from an Action Sheet and pass them to
test scripts to make sure we are giving the correct available actions for
the item we are interacting with.
Added a test for long press on an image, and used the sheet scraping functionality
to make sure that the correct actions were being displayed.
Test for previous patch, Changeset 205915; Bug 161761

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _contentsOfUserInterfaceItem:]):
* UIProcess/API/Cocoa/WKWebViewPrivate.h:
* UIProcess/ios/WKActionSheetAssistant.h:
* UIProcess/ios/WKActionSheetAssistant.mm:
(-[WKActionSheetAssistant currentAvailableActionStrings]):
* UIProcess/ios/WKContentViewInteraction.h:
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView _contentsOfUserInterfaceItem:]):

Tools:

* DumpRenderTree/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::contentsOfUserInterfaceItem):
(WTR::UIScriptController::selectFormAccessoryPickerRow): Deleted.
* TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
* TestRunnerShared/UIScriptContext/UIScriptController.cpp:
(WTR::UIScriptController::contentsOfUserInterfaceItem):
(WTR::UIScriptController::selectFormAccessoryPickerRow): Deleted.
* TestRunnerShared/UIScriptContext/UIScriptController.h:
* WebKitTestRunner/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::contentsOfUserInterfaceItem):
(WTR::UIScriptController::selectFormAccessoryPickerRow): Deleted.

LayoutTests:

* fast/events/touch/ios/long-press-on-image-expected.txt: Added.
* fast/events/touch/ios/long-press-on-image.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206645 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove a couple of unused members from PlatformKeyboardEvent
andersca@apple.com [Fri, 30 Sep 2016 18:29:31 +0000 (18:29 +0000)]
Remove a couple of unused members from PlatformKeyboardEvent
https://bugs.webkit.org/show_bug.cgi?id=162641

Reviewed by Tim Horton.

Source/WebCore:

* platform/PlatformKeyboardEvent.h:
(WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent):
(WebCore::PlatformKeyboardEvent::nativeVirtualKeyCode): Deleted.
(WebCore::PlatformKeyboardEvent::macCharCode): Deleted.
* platform/mac/PlatformEventFactoryMac.mm:
(WebCore::PlatformKeyboardEventBuilder::PlatformKeyboardEventBuilder):
* replay/SerializationMethods.cpp:
(JSC::EncodingTraits<PlatformKeyboardEvent>::encodeValue):
(JSC::EncodingTraits<PlatformKeyboardEvent>::decodeValue):

Source/WebKit2:

* Shared/WebEventConversion.cpp:
(WebKit::WebKit2PlatformKeyboardEvent::WebKit2PlatformKeyboardEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206644 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUse topVMEntryFrame to determine whether to skip the re-throw of a simulated throw.
mark.lam@apple.com [Fri, 30 Sep 2016 18:15:38 +0000 (18:15 +0000)]
Use topVMEntryFrame to determine whether to skip the re-throw of a simulated throw.
https://bugs.webkit.org/show_bug.cgi?id=162793

Reviewed by Saam Barati.

Change the ThrowScope destructor to use topVMEntryFrame (instead of topCallFrame)
in the determination of whether to skip the re-throw of a simulated throw.  This
is needed because the topCallFrame is not updated in operationConstructArityCheck()
(and does not need to be), whereas topVMEntryFrame is always updated properly.
Hence, we should just switch to using the more reliable topVMEntryFrame instead.

This issue was discovered by existing JSC tests when exception check validation
is enabled.

* runtime/ThrowScope.cpp:
(JSC::ThrowScope::~ThrowScope):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206643 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago64-bit LLInt needs to have a concurrency-aware barrier
fpizlo@apple.com [Fri, 30 Sep 2016 17:02:46 +0000 (17:02 +0000)]
64-bit LLInt needs to have a concurrency-aware barrier
https://bugs.webkit.org/show_bug.cgi?id=162790

Reviewed by Mark Lam.

In a concurrent GC the barrier definitely has to be after the store, not before it.

* llint/LowLevelInterpreter64.asm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206642 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove "rem" unit optimization for document element font size changes
antti@apple.com [Fri, 30 Sep 2016 17:00:22 +0000 (17:00 +0000)]
Remove "rem" unit optimization for document element font size changes
https://bugs.webkit.org/show_bug.cgi?id=162778

Reviewed by Alex Christensen.

We awkwardly track from the parser level if any stylesheet in a document uses any rem units. This is only used to minimally
optimize a case where document element's (<html>) font size changes dynamically.

In practice such changes are rare. Browsing around I couldn't find a single case where this optimization got used.
Even if it was used it would be of low value as a full style resolution is likely to happen anyway (as font inherits)
and the only thing really saved is that we don't need to invalidate the matched properties cache.

* css/CSSGrammar.y.in:
* css/StyleSheetContents.cpp:
(WebCore::StyleSheetContents::StyleSheetContents):
* css/StyleSheetContents.h:
* dom/AuthorStyleSheets.cpp:
(WebCore::AuthorStyleSheets::updateActiveStyleSheets):
* dom/AuthorStyleSheets.h:
(WebCore::AuthorStyleSheets::usesRemUnits): Deleted.
(WebCore::AuthorStyleSheets::setUsesRemUnit): Deleted.
* dom/Document.cpp:
(WebCore::Document::recalcStyle):
(WebCore::Document::updateBaseURL):
* style/StyleTreeResolver.cpp:
(WebCore::Style::TreeResolver::resolveElement):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206641 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAir should have a way of expressing additional instruction flags
fpizlo@apple.com [Fri, 30 Sep 2016 16:59:24 +0000 (16:59 +0000)]
Air should have a way of expressing additional instruction flags
https://bugs.webkit.org/show_bug.cgi?id=162699

Reviewed by Mark Lam.
Source/JavaScriptCore:

This follows a similar change in B3 (r206595) and replaces Air::Opcode with Air::Kind,
which holds onto the opcode and some additional flags. Because Air is an orthogonal ISA
(the opcode tells you what the operation does but each operand is allowed to also contain
effectively instructions for what to do to read or write that operand), the flags are
meant to be orthogonal to opcode. This allows us to say things like Add32<Trap>, which
makes sense if any of the operands to the Add32 are addresses.

To demonstrate the flags facility this partly adds a trap flag to Air. B3 doesn't use it
yet, but I made sure that Air respects it. Basically that means blocking DCE when the flag
is set, by making it imply hasNonArgNonControlEffects.

* CMakeLists.txt:
* JavaScriptCore.xcodeproj/project.pbxproj:
* b3/B3CheckSpecial.cpp:
(JSC::B3::Air::numB3Args):
(JSC::B3::CheckSpecial::Key::Key):
(JSC::B3::CheckSpecial::Key::dump):
(JSC::B3::CheckSpecial::CheckSpecial):
(JSC::B3::CheckSpecial::hiddenBranch):
(JSC::B3::CheckSpecial::forEachArg):
(JSC::B3::CheckSpecial::generate):
(JSC::B3::CheckSpecial::dumpImpl):
(JSC::B3::CheckSpecial::deepDumpImpl):
* b3/B3CheckSpecial.h:
(JSC::B3::CheckSpecial::Key::Key):
(JSC::B3::CheckSpecial::Key::operator==):
(JSC::B3::CheckSpecial::Key::kind):
(JSC::B3::CheckSpecial::Key::hash):
(JSC::B3::CheckSpecial::Key::opcode): Deleted.
* b3/B3Kind.cpp:
(JSC::B3::Kind::dump):
* b3/air/AirDumpAsJS.cpp:
(JSC::B3::Air::dumpAsJS):
* b3/air/AirFixObviousSpills.cpp:
* b3/air/AirFixPartialRegisterStalls.cpp:
* b3/air/AirGenerate.cpp:
(JSC::B3::Air::generate):
* b3/air/AirHandleCalleeSaves.cpp:
(JSC::B3::Air::handleCalleeSaves):
* b3/air/AirInst.cpp:
(JSC::B3::Air::Inst::jsHash):
(JSC::B3::Air::Inst::dump):
* b3/air/AirInst.h:
(JSC::B3::Air::Inst::Inst):
(JSC::B3::Air::Inst::kind):
(JSC::B3::Air::Inst::operator bool):
(JSC::B3::Air::Inst::opcode): Deleted.
* b3/air/AirInstInlines.h:
(JSC::B3::Air::Inst::extraClobberedRegs):
(JSC::B3::Air::Inst::extraEarlyClobberedRegs):
(JSC::B3::Air::Inst::forEachDefWithExtraClobberedRegs):
(JSC::B3::Air::Inst::reportUsedRegisters):
(JSC::B3::Air::Inst::shouldTryAliasingDef):
* b3/air/AirIteratedRegisterCoalescing.cpp:
* b3/air/AirKind.cpp: Added.
(JSC::B3::Air::Kind::dump):
* b3/air/AirKind.h: Added.
(JSC::B3::Air::Kind::Kind):
(JSC::B3::Air::Kind::operator==):
(JSC::B3::Air::Kind::operator!=):
(JSC::B3::Air::Kind::hash):
(JSC::B3::Air::Kind::operator bool):
* b3/air/AirLowerAfterRegAlloc.cpp:
(JSC::B3::Air::lowerAfterRegAlloc):
* b3/air/AirLowerEntrySwitch.cpp:
(JSC::B3::Air::lowerEntrySwitch):
* b3/air/AirLowerMacros.cpp:
(JSC::B3::Air::lowerMacros):
* b3/air/AirOptimizeBlockOrder.cpp:
(JSC::B3::Air::optimizeBlockOrder):
* b3/air/AirReportUsedRegisters.cpp:
(JSC::B3::Air::reportUsedRegisters):
* b3/air/AirSimplifyCFG.cpp:
(JSC::B3::Air::simplifyCFG):
* b3/air/AirTmpWidth.cpp:
(JSC::B3::Air::TmpWidth::recompute):
* b3/air/AirUseCounts.h:
(JSC::B3::Air::UseCounts::UseCounts):
* b3/air/AirValidate.cpp:
* b3/air/opcode_generator.rb:
* b3/testb3.cpp:
(JSC::B3::testTernarySubInstructionSelection):
(JSC::B3::testBranchBitAndImmFusion):

Source/WTF:

* wtf/CommaPrinter.h:
(WTF::CommaPrinter::CommaPrinter):
(WTF::CommaPrinter::dump):
(WTF::CommaPrinter::didPrint):

Websites/webkit.org:

* docs/b3/assembly-intermediate-representation.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206640 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRenderLayer::clipRects may return nullptr.
zalan@apple.com [Fri, 30 Sep 2016 16:52:47 +0000 (16:52 +0000)]
RenderLayer::clipRects may return nullptr.
https://bugs.webkit.org/show_bug.cgi?id=162729

Reviewed by Chris Dumez.

This patch refactors RenderLayer::updateClipRects(), parentClipRects() and backgroundClipRect()
so that we don't have to rely on this seemingly unsafe line: clipRects = *parent()->clipRects(clipRectsContext);
Now updateClipRects() returns the computed/cached clip rects as opposed to update and refetch them.
While this patch makes the code look more readable/safer, it also eliminates cached item tripple retrievals.

No change in functionality.

* rendering/RenderLayer.cpp:
(WebCore::ClipRectsCache::getClipRects):
(WebCore::ClipRectsCache::setClipRects):
(WebCore::RenderLayer::updateClipRects):
(WebCore::RenderLayer::clipRects):
(WebCore::RenderLayer::calculateClipRects):
* rendering/RenderLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206639 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdd a way to go from a RefPtr<T> to Ref<const T>
commit-queue@webkit.org [Fri, 30 Sep 2016 16:41:20 +0000 (16:41 +0000)]
Add a way to go from a RefPtr<T> to Ref<const T>
https://bugs.webkit.org/show_bug.cgi?id=162683

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-30
Reviewed by Alex Christensen.

Source/WebCore:

No change of behavior.

* Modules/fetch/FetchBody.cpp:
(WebCore::FetchBody::extract):

Source/WTF:

* wtf/RefPtr.h:
(WTF::RefPtr::releaseConstNonNull): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206638 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, fix 32-bit build.
said@apple.com [Fri, 30 Sep 2016 16:18:25 +0000 (16:18 +0000)]
Unreviewed, fix 32-bit build.

* loader/cache/CachedImage.cpp:
(WebCore::CachedImage::decodedSizeChanged):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206637 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFetchBody should use UTF8Encoding to encode text data
commit-queue@webkit.org [Fri, 30 Sep 2016 16:08:30 +0000 (16:08 +0000)]
FetchBody should use UTF8Encoding to encode text data
https://bugs.webkit.org/show_bug.cgi?id=162775

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-30
Reviewed by Sam Weinig.

LayoutTests/imported/w3c:

* web-platform-tests/fetch/api/basic/text-utf8-expected.txt:
* web-platform-tests/fetch/api/basic/text-utf8.html:
* web-platform-tests/fetch/api/resources/utils.js:
(encode_utf8):

Source/WebCore:

Covered by added tests.

Removing unnecesary vector allocation when extracting UTF8 bytes from a Request/Response text/URLSearchParams body.
Making use of UTF8Encoding for consistency with  the upload code path.

* Modules/fetch/FetchBody.cpp:
(WebCore::FetchBody::consumeAsStream):
(WebCore::FetchBody::consumeText):
(WebCore::extractBytesFromText): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206636 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoChange the MemoryCache and CachedResource adjustSize functions to take a long argument
said@apple.com [Fri, 30 Sep 2016 16:01:36 +0000 (16:01 +0000)]
Change the MemoryCache and CachedResource adjustSize functions to take a long argument
https://bugs.webkit.org/show_bug.cgi?id=162708
<rdar://problem/28555702>

Reviewed by Brent Fulgham.

Source/WebCore:

Because the MemoryCache stores the size of the cached memory in unsigned,
two problems my happen when reporting a change in the size of the memory:

1. Signed integer overflow -- which can happen because MemoryCache::adjustSize()
   takes a signed integer argument. If the allocated or the freed memory size is
   larger than the maximum of a signed integer, an overflow will happen.
   For the image caching code, this can be seen where the unsigned decodedSize
   is casted to an integer before passing it to ImageObserver::decodedSizeChanged().

2. Unsigned integer overflow -- which can happen if the new allocated memory
   size plus the currentSize exceeds the maximum of unsigned.
   This can be seen in MemoryCache::adjustSize() where we add delta to m_liveSize
   or m_deadSize without checking whether this addition will overflow or not. We
   do not assert for overflow although we assert for underflow.

The fix for these two problems can be the following:

1. Make all the adjustSize functions all the way till MemoryCache::adjustSize()
   take a signed long integer argument.

2. Do not create a NativeImagePtr for an ImageFrame if its frameBytes plus the
   ImageFrameCache::decodedSize() will exceed the maximum of an unsigned integer.

* loader/cache/CachedImage.cpp:
(WebCore::CachedImage::decodedSizeChanged): Change the argument to be long. No overflow will happen when casting the argument from unsigned to long.
* loader/cache/CachedImage.h:
* loader/cache/CachedResource.cpp:
(WebCore::CachedResource::setDecodedSize): Use long integer casting when calling MemoryCache::adjustSize().
(WebCore::CachedResource::setEncodedSize): Ditto.
* loader/cache/MemoryCache.cpp:
(WebCore::MemoryCache::MemoryCache): Add as static assert to ensure sizeof(long long) can hold any unsigned or its negation.
(WebCore::MemoryCache::revalidationSucceeded): Use long integer casting when calling MemoryCache::adjustSize().
(WebCore::MemoryCache::remove): Ditto.
(WebCore::MemoryCache::adjustSize): Change the function argument to long integer. No overflow will happen when casting the argument from unsigned to long.
* loader/cache/MemoryCache.h:
* platform/graphics/ImageFrameCache.cpp:
(WebCore::ImageFrameCache::destroyIncompleteDecodedData): Call a function with its new name.
(WebCore::ImageFrameCache::decodedSizeChanged): Change the function argument to long integer. No overflow will happen when casting the argument from unsigned to long.
(WebCore::ImageFrameCache::decodedSizeIncreased): Use long integer casting when calling decodedSizeChanged().
(WebCore::ImageFrameCache::decodedSizeDecreased): Ditto.
(WebCore::ImageFrameCache::decodedSizeReset): Ditto.
(WebCore::ImageFrameCache::didDecodeProperties): Ditto.
(WebCore::ImageFrameCache::frameAtIndex): Do not create the NativeImage if adding its frameByes to the MemoryCache will cause numerical overflow.
(WebCore::ImageFrameCache::decodedSizeIncremented): Deleted. This function is renamed decodedSizeIncreased().
(WebCore::ImageFrameCache::decodedSizeDecremented): Deleted. This function is renamed decodedSizeDecreased().
* platform/graphics/ImageFrameCache.h:
* platform/graphics/ImageObserver.h:
* platform/graphics/IntSize.h:
(WebCore::IntSize::unclampedArea): Returns the area of an IntSize in size_t.
* platform/graphics/cg/PDFDocumentImage.cpp:
(WebCore::PDFDocumentImage::decodedSizeChanged): Use long integer casting when calling ImageObserver::decodedSizeChanged().

LayoutTests:

* TestExpectations: Remove failed tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206635 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdd support for ImageData.data attribute
cdumez@apple.com [Fri, 30 Sep 2016 14:20:25 +0000 (14:20 +0000)]
Add support for ImageData.data attribute
https://bugs.webkit.org/show_bug.cgi?id=162767

Reviewed by Sam Weinig.

LayoutTests/imported/w3c:

Rebaseline W3C test now that one more check is passing.

* web-platform-tests/html/dom/interfaces-expected.txt:

Source/WebCore:

Add support for ImageData.data attribute:
- https://html.spec.whatwg.org/#imagedata

Firefox already supports it.

No new tests, updated existing tests.

* html/ImageData.idl:

LayoutTests:

Add layout test coverage.

* fast/canvas/canvas-imageData-expected.txt:
* fast/canvas/script-tests/canvas-imageData.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206634 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoDumpRenderTree crashed in com.apple.WebCore: WTF::Optional<WebCore::FetchBodyOwner...
commit-queue@webkit.org [Fri, 30 Sep 2016 08:27:02 +0000 (08:27 +0000)]
DumpRenderTree crashed in com.apple.WebCore: WTF::Optional<WebCore::FetchBodyOwner::BlobLoader>::operator bool const + 12
https://bugs.webkit.org/show_bug.cgi?id=162483

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-30
Reviewed by Alex Christensen.

Source/WebCore:

Test: fetch/closing-while-fetching-blob.html
No change of behavior.

* Modules/fetch/FetchBodyOwner.cpp:
(WebCore::FetchBodyOwner::stop): Asserting m_blobLoader is null (meaning that unsetPendingActivity was done)
only in case FetchBodyOwner has no risk being destroyed.

LayoutTests:

* fetch/closing-while-fetching-blob-expected.txt: Added.
* fetch/closing-while-fetching-blob.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206633 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Fetch API] Add support for URLSearchParams body
commit-queue@webkit.org [Fri, 30 Sep 2016 08:04:52 +0000 (08:04 +0000)]
[Fetch API] Add support for URLSearchParams body
https://bugs.webkit.org/show_bug.cgi?id=162667

Patch by Youenn Fablet <youennf@gmail.com> on 2016-09-30
Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

Adding new tests to consume, stream and upload URLSearchParams bodies.

* web-platform-tests/fetch/api/basic/request-headers-expected.txt:
* web-platform-tests/fetch/api/basic/request-headers-worker-expected.txt:
* web-platform-tests/fetch/api/basic/request-headers.js:
(checkContentType):
(requestHeaders):
* web-platform-tests/fetch/api/basic/request-upload-expected.txt: Added.
* web-platform-tests/fetch/api/basic/request-upload-worker-expected.txt: Added.
* web-platform-tests/fetch/api/basic/request-upload-worker.html: Added.
* web-platform-tests/fetch/api/basic/request-upload.html: Added.
* web-platform-tests/fetch/api/basic/request-upload.js: Added.
(testUpload):
* web-platform-tests/fetch/api/request/request-consume-empty-expected.txt:
* web-platform-tests/fetch/api/request/request-consume-empty.html:
* web-platform-tests/fetch/api/request/request-init-002-expected.txt:
* web-platform-tests/fetch/api/resources/echo-content.py: Added.
(main):
* web-platform-tests/fetch/api/response/response-consume-empty-expected.txt:
* web-platform-tests/fetch/api/response/response-consume-empty.html:
* web-platform-tests/fetch/api/response/response-consume-expected.txt:
* web-platform-tests/fetch/api/response/response-consume-stream-expected.txt:
* web-platform-tests/fetch/api/response/response-consume-stream.html:
* web-platform-tests/fetch/api/response/response-consume.html:
* web-platform-tests/fetch/api/response/response-init-002-expected.txt:

Source/WebCore:

Tests: imported/w3c/web-platform-tests/fetch/api/basic/request-upload-worker.html
       imported/w3c/web-platform-tests/fetch/api/basic/request-upload.html

* Modules/fetch/FetchBody.cpp:
(WebCore::extractBytesFromText): Moving to a static function that should be moved to a String method.
(WebCore::FetchBody::FetchBody): Adding constructor for URLSearchParams.
(WebCore::FetchBody::extract): Adding URLSearchParams body initialization.
(WebCore::FetchBody::consume): Consuming URLSearchParams bodies.
(WebCore::FetchBody::consumeAsStream): Ditto.
(WebCore::FetchBody::consumeText): Updated to take a parameter to handle both text and URLSearchParams bodies.
(WebCore::FetchBody::bodyForInternalRequest): Add URLSearchParams body upload.
(WebCore::FetchBody::clone): Add URLSearchParams body cloning
* Modules/fetch/FetchBody.h:
(WebCore::FetchBody::urlSearchParamsBody):
* html/URLSearchParams.cpp:
(WebCore::URLSearchParams::toString):
* html/URLSearchParams.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206632 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRe-order the arguments to drawPattern() functions
simon.fraser@apple.com [Fri, 30 Sep 2016 06:41:08 +0000 (06:41 +0000)]
Re-order the arguments to drawPattern() functions
https://bugs.webkit.org/show_bug.cgi?id=162772

Reviewed by Zalan Bujtas.

Re-order the arguments to have the destination rect come near the front,
leaving CompositeOperator and BlendMode hanging out at the back like
other drawing functions.

No behavior change.

* platform/graphics/BitmapImage.cpp:
(WebCore::BitmapImage::drawPattern):
* platform/graphics/BitmapImage.h:
* platform/graphics/CrossfadeGeneratedImage.cpp:
(WebCore::CrossfadeGeneratedImage::drawPattern):
* platform/graphics/CrossfadeGeneratedImage.h:
* platform/graphics/GeneratedImage.h:
* platform/graphics/GradientImage.cpp:
(WebCore::GradientImage::drawPattern):
* platform/graphics/GradientImage.h:
* platform/graphics/GraphicsContext.h:
* platform/graphics/Image.cpp:
(WebCore::Image::drawTiled):
* platform/graphics/Image.h:
* platform/graphics/ImageBuffer.h:
* platform/graphics/NamedImageGeneratedImage.cpp:
(WebCore::NamedImageGeneratedImage::drawPattern):
* platform/graphics/NamedImageGeneratedImage.h:
* platform/graphics/cairo/GraphicsContextCairo.cpp:
(WebCore::GraphicsContext::drawPattern):
* platform/graphics/cairo/ImageBufferCairo.cpp:
(WebCore::ImageBuffer::drawPattern):
* platform/graphics/cairo/ImageCairo.cpp:
(WebCore::Image::drawPattern):
* platform/graphics/cg/GraphicsContextCG.cpp:
(WebCore::GraphicsContext::drawPattern):
* platform/graphics/cg/ImageBufferCG.cpp:
(WebCore::ImageBuffer::drawPattern):
* platform/graphics/cg/ImageCG.cpp: This file can probably be removed and the code moved, since the
code is almost the same as the cairo code, and not platform-specific.
(WebCore::Image::drawPattern):
* platform/graphics/displaylists/DisplayListItems.cpp:
(WebCore::DisplayList::DrawPattern::DrawPattern):
(WebCore::DisplayList::DrawPattern::apply):
* platform/graphics/displaylists/DisplayListItems.h:
(WebCore::DisplayList::DrawPattern::create):
* platform/graphics/displaylists/DisplayListRecorder.cpp:
(WebCore::DisplayList::Recorder::drawPattern):
* platform/graphics/displaylists/DisplayListRecorder.h:
* svg/graphics/SVGImage.cpp:
(WebCore::SVGImage::drawPatternForContainer):
* svg/graphics/SVGImageForContainer.cpp:
(WebCore::SVGImageForContainer::drawPattern):
* svg/graphics/SVGImageForContainer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206631 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAdd support for download attribute on area elements
cdumez@apple.com [Fri, 30 Sep 2016 05:05:35 +0000 (05:05 +0000)]
Add support for download attribute on area elements
https://bugs.webkit.org/show_bug.cgi?id=162765

Reviewed by Brent Fulgham.

LayoutTests/imported/w3c:

Rebaseline W3C test now that more checks are passing.

* web-platform-tests/html/dom/reflection-embedded-expected.txt:

Source/WebCore:

Add support for download attribute on area elements:
- https://html.spec.whatwg.org/#htmlanchorelement

Because HTMLAreaElement inherits from HTMLAnchorElement on native side,
and because HTMLAnchorElement already supports the download attribute,
all that was needed was to add the download attribute to
HTMLAreaElement.idl file.

Test: http/tests/download/area-download.html

* html/HTMLAreaElement.idl:

LayoutTests:

Add layout test coverage.

* http/tests/download/area-download-expected.txt: Added.
* http/tests/download/area-download.html: Added.
* http/tests/download/resources/FAIL-notify-done.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206630 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoEditor::findStringAndScrollToVisible is unused
rniwa@webkit.org [Fri, 30 Sep 2016 02:04:07 +0000 (02:04 +0000)]
Editor::findStringAndScrollToVisible is unused
https://bugs.webkit.org/show_bug.cgi?id=162762

Reviewed by Simon Fraser.

Deleted the unused function.

* editing/Editor.cpp:
(WebCore::Editor::findStringAndScrollToVisible): Deleted.
* editing/Editor.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206629 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoREGRESSION(r206555): It made Dromaeo/jslib-style-jquery.html crash
fpizlo@apple.com [Fri, 30 Sep 2016 00:48:55 +0000 (00:48 +0000)]
REGRESSION(r206555): It made Dromaeo/jslib-style-jquery.html crash
https://bugs.webkit.org/show_bug.cgi?id=162721

Reviewed by Keith Miller.

The put_by_id-in-put_by_val optimization had the write barrier in the wrong place and
incorrectly filtered on value instead of base.

No reduced test case. You really need to run Dromaeo/jslib to catch it. I love Dromaeo's
ability to catch GC bugs.

* jit/JITPropertyAccess.cpp:
(JSC::JIT::emitPutByValWithCachedId):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206628 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoExpose CryptoKey to web workers
jiewen_tan@apple.com [Fri, 30 Sep 2016 00:47:21 +0000 (00:47 +0000)]
Expose CryptoKey to web workers
https://bugs.webkit.org/show_bug.cgi?id=162640
<rdar://problem/28182204>

Reviewed by Brent Fulgham.

Source/WebCore:

Tests: crypto/workers/aes-postMessage-worker.html
       crypto/workers/hmac-postMessage-worker.html
       crypto/workers/hrsa-postMessage-worker.html
       crypto/workers/multiple-postMessage-worker.html
       crypto/workers/rsa-postMessage-worker.html

* crypto/CryptoKey.idl:
* crypto/SerializedCryptoKeyWrap.h:
* crypto/mac/SerializedCryptoKeyWrapMac.mm:
(WebCore::deleteDefaultWebCryptoMasterKey):
Add a way to delete the default web crypto master key such that we can test
these APIs and cleanup.
* workers/WorkerGlobalScope.cpp:
(WebCore::WorkerGlobalScope::wrapCryptoKey):
(WebCore::WorkerGlobalScope::unwrapCryptoKey):
* workers/WorkerGlobalScope.h:

Tools:

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebCore/SerializedCryptoKeyWrap.mm: Added.
(TestWebKitAPI::TEST_F):
Add API tests for SerializedCryptoKeyWrap APIs.

LayoutTests:

* crypto/resources/common.js: Renamed from LayoutTests/crypto/subtle/resources/common.js.
* crypto/subtle/aes-cbc-192-encrypt-decrypt.html:
* crypto/subtle/aes-cbc-256-encrypt-decrypt.html:
* crypto/subtle/aes-cbc-encrypt-decrypt-with-padding.html:
* crypto/subtle/aes-cbc-encrypt-decrypt.html:
* crypto/subtle/aes-cbc-generate-key.html:
* crypto/subtle/aes-cbc-import-jwk.html:
* crypto/subtle/aes-cbc-invalid-length.html:
* crypto/subtle/aes-cbc-unwrap-failure.html:
* crypto/subtle/aes-cbc-unwrap-rsa.html:
* crypto/subtle/aes-cbc-wrap-rsa-non-extractable.html:
* crypto/subtle/aes-cbc-wrap-rsa.html:
* crypto/subtle/aes-cbc-wrong-key-class.html:
* crypto/subtle/aes-export-key.html:
* crypto/subtle/aes-kw-key-manipulation.html:
* crypto/subtle/aes-kw-wrap-unwrap-aes.html:
* crypto/subtle/aes-postMessage-expected.txt:
* crypto/subtle/aes-postMessage.html:
* crypto/subtle/argument-conversion.html:
* crypto/subtle/array-buffer-view-offset.html:
* crypto/subtle/crypto-key-algorithm-gc.html:
* crypto/subtle/crypto-key-usages-gc.html:
* crypto/subtle/hmac-check-algorithm.html:
* crypto/subtle/hmac-export-key.html:
* crypto/subtle/hmac-generate-key.html:
* crypto/subtle/hmac-import-jwk.html:
* crypto/subtle/hmac-postMessage-expected.txt:
* crypto/subtle/hmac-postMessage.html:
* crypto/subtle/hmac-sign-verify-empty-key.html:
* crypto/subtle/hmac-sign-verify.html:
* crypto/subtle/import-jwk.html:
* crypto/subtle/jwk-export-use-values.html:
* crypto/subtle/jwk-import-use-values.html:
* crypto/subtle/postMessage-worker-expected.txt: Removed.
* crypto/subtle/rsa-export-generated-keys.html:
* crypto/subtle/rsa-export-key.html:
* crypto/subtle/rsa-export-private-key.html:
* crypto/subtle/rsa-indexeddb-non-exportable-private.html:
* crypto/subtle/rsa-indexeddb-non-exportable.html:
* crypto/subtle/rsa-indexeddb-private.html:
* crypto/subtle/rsa-indexeddb.html:
* crypto/subtle/rsa-oaep-generate-non-extractable-key.html:
* crypto/subtle/rsa-oaep-key-manipulation.html:
* crypto/subtle/rsa-oaep-plaintext-length.html:
* crypto/subtle/rsa-oaep-wrap-unwrap-aes.html:
* crypto/subtle/rsa-postMessage.html:
* crypto/subtle/rsaes-pkcs1-v1_5-decrypt.html:
* crypto/subtle/rsaes-pkcs1-v1_5-wrap-unwrap-aes.html:
* crypto/subtle/rsassa-pkcs1-v1_5-generate-key-with-leading-zeroes-in-exponent.html:
* crypto/subtle/rsassa-pkcs1-v1_5-generate-key.html:
* crypto/subtle/rsassa-pkcs1-v1_5-import-jwk-small-key.html:
* crypto/subtle/rsassa-pkcs1-v1_5-import-jwk.html:
* crypto/subtle/rsassa-pkcs1-v1_5-sign-verify.html:
* crypto/subtle/sha-1.html:
* crypto/subtle/sha-224.html:
* crypto/subtle/sha-256.html:
* crypto/subtle/sha-384.html:
* crypto/subtle/sha-512.html:
* crypto/subtle/unimplemented-unwrap-crash.html:
* crypto/subtle/unwrapKey-check-usage.html:
* crypto/subtle/wrapKey-check-usage.html:
* crypto/workers/aes-postMessage-worker-expected.txt: Added.
* crypto/workers/aes-postMessage-worker.html: Added.
* crypto/workers/hmac-postMessage-worker-expected.txt: Added.
* crypto/workers/hmac-postMessage-worker.html: Renamed from LayoutTests/crypto/subtle/postMessage-worker.html.
* crypto/workers/hrsa-postMessage-worker-expected.txt: Added.
* crypto/workers/hrsa-postMessage-worker.html: Added.
* crypto/workers/multiple-postMessage-worker-expected.txt: Added.
* crypto/workers/multiple-postMessage-worker.html: Added.
* crypto/workers/resources/aes-postMessage-worker.js: Added.
* crypto/workers/resources/hmac-postMessage-worker.js: Renamed from LayoutTests/crypto/subtle/resources/postMessage-worker.js.
* crypto/workers/resources/hrsa-postMessage-worker.js: Added.
* crypto/workers/resources/rsa-postMessage-worker.js: Added.
* crypto/workers/rsa-postMessage-worker-expected.txt: Added.
* crypto/workers/rsa-postMessage-worker.html: Added.
* js/dom/global-constructors-attributes-dedicated-worker-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206627 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAX: iOS: Tapping <input> in Safari zooms in a bit when page has max scale = 1
n_wang@apple.com [Fri, 30 Sep 2016 00:38:18 +0000 (00:38 +0000)]
AX: iOS: Tapping <input> in Safari zooms in a bit when page has max scale = 1
https://bugs.webkit.org/show_bug.cgi?id=162471

Reviewed by Simon Fraser.

Source/WebCore:

If the author has defined a maximum scale, we should honor that when keyboard focus moves
to a text field, instead of using the forceAlwaysUserScalableMaximumScale.

Tests: fast/forms/ios/user-scalable-does-not-scale-for-keyboard-focus-with-author-defined-scale.html
       fast/forms/ios/user-scalable-does-not-scale-for-keyboard-focus-with-user-scalable-no.html
       fast/forms/ios/user-scalable-scales-for-keyboard-focus-with-no-author-defined-scale.html

* page/ViewportConfiguration.cpp:
(WebCore::ViewportConfiguration::allowsUserScaling):
(WebCore::ViewportConfiguration::allowsUserScalingIgnoringAlwaysScalable):
(WebCore::ViewportConfiguration::allowsUserScalingIgnoringForceAlwaysScaling): Deleted.
* page/ViewportConfiguration.h:
(WebCore::ViewportConfiguration::maximumScaleIgnoringAlwaysScalable):
(WebCore::ViewportConfiguration::maximumScale): Deleted.

Source/WebKit2:

* Shared/AssistedNodeInformation.cpp:
(WebKit::AssistedNodeInformation::encode):
(WebKit::AssistedNodeInformation::decode):
* Shared/AssistedNodeInformation.h:
* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView _displayFormNodeInputView]):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::maximumPageScaleFactorIgnoringAlwaysScalable):
(WebKit::WebPage::getAssistedNodeInformation):

LayoutTests:

Moved focus input related tests to the right place.

* fast/forms/ios/user-scalable-does-not-scale-for-keyboard-focus-with-author-defined-scale-expected.txt: Added.
* fast/forms/ios/user-scalable-does-not-scale-for-keyboard-focus-with-author-defined-scale.html: Added.
* fast/forms/ios/user-scalable-does-not-scale-for-keyboard-focus-with-user-scalable-no-expected.txt: Added.
* fast/forms/ios/user-scalable-does-not-scale-for-keyboard-focus-with-user-scalable-no.html: Added.
* fast/forms/ios/user-scalable-scales-for-keyboard-focus-with-no-author-defined-scale-expected.txt: Added.
* fast/forms/ios/user-scalable-scales-for-keyboard-focus-with-no-author-defined-scale.html: Added.
* fast/viewport/ios/user-scalable-does-not-scale-for-keyboard-focus-with-author-defined-scale-expected.txt: Removed.
* fast/viewport/ios/user-scalable-does-not-scale-for-keyboard-focus-with-author-defined-scale.html: Removed.
* fast/viewport/ios/user-scalable-scales-for-keyboard-focus-with-no-author-defined-scale-expected.txt: Removed.
* fast/viewport/ios/user-scalable-scales-for-keyboard-focus-with-no-author-defined-scale.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206626 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMark NetworkLoad as final
dbates@webkit.org [Thu, 29 Sep 2016 23:05:35 +0000 (23:05 +0000)]
Mark NetworkLoad as final
https://bugs.webkit.org/show_bug.cgi?id=162681

Reviewed by Alex Christensen.

Mark class NetworkLoad as final as we do not have any derived classes at the moment and it
is unclear if it would make sense to subclass it. We can always unmark the class if it turns
out that it makes sense to subclass it.

* NetworkProcess/NetworkLoad.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206621 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoVersioning.
bshafiei@apple.com [Thu, 29 Sep 2016 22:23:31 +0000 (22:23 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206620 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[iOS] Update Touch constructor test to cover clientX / clientY attributes
cdumez@apple.com [Thu, 29 Sep 2016 22:16:58 +0000 (22:16 +0000)]
[iOS] Update Touch constructor test to cover clientX / clientY attributes
https://bugs.webkit.org/show_bug.cgi?id=162757

Reviewed by Sam Weinig.

Update Touch constructor test to cover clientX / clientY attributes.

* fast/events/touch/touch-constructor.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206619 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: AXI: linkified refs to #document and #text are not usually navigable...
commit-queue@webkit.org [Thu, 29 Sep 2016 22:15:57 +0000 (22:15 +0000)]
Web Inspector: AXI: linkified refs to #document and #text are not usually navigable nodes; consider delinkifying them
https://bugs.webkit.org/show_bug.cgi?id=130600
<rdar://problem/16391333>

Patch by Aaron Chu <aaron_chu@apple.com> on 2016-09-29
Reviewed by Brian Burg.

.:

Added a manual test to test the node links in the Accessibility Inspector
https://bugs.webkit.org/show_bug.cgi?id=130600

* ManualTests/accessibility/delinkified-non-navigable-links.html: Added.

Source/WebInspectorUI:

Removing link style for non-navigable nodes by first
checking nodeType of the node.

* UserInterface/Base/DOMUtilities.js:
(WebInspector.linkifyNodeReference):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206618 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix syntax violation handling in IPv4 address parsing
achristensen@apple.com [Thu, 29 Sep 2016 22:02:04 +0000 (22:02 +0000)]
Fix syntax violation handling in IPv4 address parsing
https://bugs.webkit.org/show_bug.cgi?id=162756

Reviewed by Tim Horton.

Source/WebCore:

When we are parsing the up to 4 numbers in an IPv4 address, if we find a syntax violation
(the canonicalized address would be different than the input string) then wait to report
it until we have determined that this is a valid IPv4 address that will be canonicalized.
If it is not a valid IPv4 address, then we will just treat the characters as the host, and
that could be no syntax violation.

Covered by a new API test and existing API tests.

* platform/URLParser.cpp:
(WebCore::URLParser::URLParser):
(WebCore::URLParser::parseIPv4Number):
(WebCore::URLParser::parseIPv4Host):
* platform/URLParser.h:

Tools:

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@206617 268f45cc-cd09-0410-ab3c-d52691b4dbfc