Timeline



Oct 25, 2018:

10:06 PM Changeset in webkit [237446] by Chris Dumez
  • 6 edits in trunk

[PSON] Navigating cross-site with locked history but unlocked back/forward list fails to create a new BackForwardListItem
https://bugs.webkit.org/show_bug.cgi?id=190915
<rdar://problem/45059069>

Reviewed by Geoffrey Garen.

Source/WebCore:

  • history/PageCache.cpp:

(WebCore::canCacheFrame):
Make sure we do not put into PageCache a page whose main frame is showing the initial empty document.
We usually do not try to put those into PageCache because we do not have a HistoryItem to save the
PageCache entry on. However, when we process-swap on a navigation with the history locked, the new
process has a HistoryItem and is initially showing the initial empty document before continuing
the load from the previous process. Note that saving the initial empty document in PageCache would
lead to crashes later on previous the initial empty document's Window is taken and reused for the
next load.

  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::load):
Stop assuming that we're continuing a client-side redirect when lockHistory is Yes. It is
lockBackForwardList that is actually Yes when we're doing a client-side redirect.

  • loader/PolicyChecker.cpp:

(WebCore::PolicyChecker::checkNavigationPolicy):
Stop using calling the completion handler with an invalid URL when the policy decision is 'Suspend' and
use 'about:blank' instead. Without this change, FrameLoader::continueLoadAfterNavigationPolicy() would
not load 'about:blank' when its AllowNavigationToInvalidURL parameter is No.

Tools:

Add API test coverage.

  • TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
9:51 PM Changeset in webkit [237445] by yusukesuzuki@slowstart.org
  • 13 edits in trunk/Source/JavaScriptCore

Unreviewed, partial rolling in r237254
https://bugs.webkit.org/show_bug.cgi?id=190340

We do not use the added function right now to investigate what is the reason of the regression.
If it causes the regression, it seems that Parser.cpp's inlining decision seems culprit.

  • bytecode/UnlinkedFunctionExecutable.cpp:

(JSC::UnlinkedFunctionExecutable::fromGlobalCode):

  • bytecode/UnlinkedFunctionExecutable.h:
  • parser/Parser.cpp:

(JSC::Parser<LexerType>::parseInner):
(JSC::Parser<LexerType>::parseSingleFunction):
(JSC::Parser<LexerType>::parseFunctionInfo):
(JSC::Parser<LexerType>::parseFunctionDeclaration):
(JSC::Parser<LexerType>::parseAsyncFunctionDeclaration):

  • parser/Parser.h:

(JSC::Parser<LexerType>::parse):
(JSC::parse):
(JSC::parseFunctionForFunctionConstructor):

  • parser/ParserModes.h:
  • parser/ParserTokens.h:

(JSC::JSTextPosition::JSTextPosition):
(JSC::JSTokenLocation::JSTokenLocation): Deleted.

  • parser/SourceCodeKey.h:

(JSC::SourceCodeKey::SourceCodeKey):
(JSC::SourceCodeKey::operator== const):

  • runtime/CodeCache.cpp:

(JSC::CodeCache::getUnlinkedGlobalCodeBlock):
(JSC::CodeCache::getUnlinkedGlobalFunctionExecutable):

  • runtime/CodeCache.h:
  • runtime/FunctionConstructor.cpp:

(JSC::constructFunctionSkippingEvalEnabledCheck):

  • runtime/FunctionExecutable.cpp:

(JSC::FunctionExecutable::fromGlobalCode):

  • runtime/FunctionExecutable.h:
7:03 PM Changeset in webkit [237444] by bshafiei@apple.com
  • 1 copy in tags/Safari-606.3.3

Tag Safari-606.3.3.

5:59 PM Changeset in webkit [237443] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

Layout Test http/tests/websocket/tests/hybi/handshake-ok-with-legacy-sec-websocket-response-headers.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=173041

Unreviewed test gardening.

  • platform/ios/TestExpectations: Mark test as flaky.
5:44 PM Changeset in webkit [237442] by sbarati@apple.com
  • 1 edit
    89 adds in trunk/PerformanceTests

Check in corresponding C code in JetStream 2
https://bugs.webkit.org/show_bug.cgi?id=190733

Rubber-stamped by Filip Pizlo.

  • JetStream2/simple/float-mm.c: Added.
  • JetStream2/wasm/HashSet.cpp: Added.
  • JetStream2/wasm/gcc-loops.cpp: Added.
  • JetStream2/wasm/quicksort.c: Added.
  • JetStream2/wasm/TSF: Added.
  • JetStream2/wasm/TSF/build.sh: Added.
  • JetStream2/wasm/TSF/config.h: Added.
  • JetStream2/wasm/TSF/gpc.h: Added.
  • JetStream2/wasm/TSF/gpc_code_gen_util.c: Added.
  • JetStream2/wasm/TSF/gpc_instruction.c: Added.
  • JetStream2/wasm/TSF/gpc_instruction_dispatch.gen: Added.
  • JetStream2/wasm/TSF/gpc_instruction_size.gen: Added.
  • JetStream2/wasm/TSF/gpc_instruction_stack_effects.gen: Added.
  • JetStream2/wasm/TSF/gpc_instruction_static_size.gen: Added.
  • JetStream2/wasm/TSF/gpc_instruction_to_string.gen: Added.
  • JetStream2/wasm/TSF/gpc_int_common.h: Added.
  • JetStream2/wasm/TSF/gpc_intable.c: Added.
  • JetStream2/wasm/TSF/gpc_internal.h: Added.
  • JetStream2/wasm/TSF/gpc_interpreter.gen: Added.
  • JetStream2/wasm/TSF/gpc_program.c: Added.
  • JetStream2/wasm/TSF/gpc_proto.c: Added.
  • JetStream2/wasm/TSF/gpc_stack_height.c: Added.
  • JetStream2/wasm/TSF/gpc_threaded.c: Added.
  • JetStream2/wasm/TSF/gpc_worklist.h: Added.
  • JetStream2/wasm/TSF/tsf.h: Added.
  • JetStream2/wasm/TSF/tsf_adaptive_reader.c: Added.
  • JetStream2/wasm/TSF/tsf_asprintf.c: Added.
  • JetStream2/wasm/TSF/tsf_atomics.h: Added.
  • JetStream2/wasm/TSF/tsf_buf_reader.c: Added.
  • JetStream2/wasm/TSF/tsf_buf_writer.c: Added.
  • JetStream2/wasm/TSF/tsf_buffer.c: Added.
  • JetStream2/wasm/TSF/tsf_build_defines.h: Added.
  • JetStream2/wasm/TSF/tsf_config.h: Added.
  • JetStream2/wasm/TSF/tsf_config_stub.h: Added.
  • JetStream2/wasm/TSF/tsf_copier.c: Added.
  • JetStream2/wasm/TSF/tsf_define_helpers.c: Added.
  • JetStream2/wasm/TSF/tsf_define_helpers.h: Added.
  • JetStream2/wasm/TSF/tsf_destructor.c: Added.
  • JetStream2/wasm/TSF/tsf_error.c: Added.
  • JetStream2/wasm/TSF/tsf_format.h: Added.
  • JetStream2/wasm/TSF/tsf_fsdb.c: Added.
  • JetStream2/wasm/TSF/tsf_fsdb_protocol.c: Added.
  • JetStream2/wasm/TSF/tsf_fsdb_protocol.h: Added.
  • JetStream2/wasm/TSF/tsf_generator.c: Added.
  • JetStream2/wasm/TSF/tsf_gpc_code_gen.c: Added.
  • JetStream2/wasm/TSF/tsf_indent.h: Added.
  • JetStream2/wasm/TSF/tsf_internal.h: Added.
  • JetStream2/wasm/TSF/tsf_internal_config.h: Added.
  • JetStream2/wasm/TSF/tsf_internal_config_stub.h: Added.
  • JetStream2/wasm/TSF/tsf_inttypes.h: Added.
  • JetStream2/wasm/TSF/tsf_io.c: Added.
  • JetStream2/wasm/TSF/tsf_io_utils.c: Added.
  • JetStream2/wasm/TSF/tsf_ir.c: Added.
  • JetStream2/wasm/TSF/tsf_ir.h: Added.
  • JetStream2/wasm/TSF/tsf_ir_different.c: Added.
  • JetStream2/wasm/TSF/tsf_ir_different.h: Added.
  • JetStream2/wasm/TSF/tsf_ir_speed.c: Added.
  • JetStream2/wasm/TSF/tsf_limits.c: Added.
  • JetStream2/wasm/TSF/tsf_named_type.c: Added.
  • JetStream2/wasm/TSF/tsf_native.c: Added.
  • JetStream2/wasm/TSF/tsf_parser.c: Added.
  • JetStream2/wasm/TSF/tsf_primitive.c: Added.
  • JetStream2/wasm/TSF/tsf_ra_type_man.c: Added.
  • JetStream2/wasm/TSF/tsf_reflect.c: Added.
  • JetStream2/wasm/TSF/tsf_region.h: Added.
  • JetStream2/wasm/TSF/tsf_serial_in_man.c: Added.
  • JetStream2/wasm/TSF/tsf_serial_out_man.c: Added.
  • JetStream2/wasm/TSF/tsf_serial_protocol.h: Added.
  • JetStream2/wasm/TSF/tsf_sha1.c: Added.
  • JetStream2/wasm/TSF/tsf_sha1.h: Added.
  • JetStream2/wasm/TSF/tsf_sha1_writer.c: Added.
  • JetStream2/wasm/TSF/tsf_sort.c: Added.
  • JetStream2/wasm/TSF/tsf_st.c: Added.
  • JetStream2/wasm/TSF/tsf_st.h: Added.
  • JetStream2/wasm/TSF/tsf_st_typetable.c: Added.
  • JetStream2/wasm/TSF/tsf_stream_file_input.c: Added.
  • JetStream2/wasm/TSF/tsf_stream_file_output.c: Added.
  • JetStream2/wasm/TSF/tsf_type.c: Added.
  • JetStream2/wasm/TSF/tsf_type_in_map.c: Added.
  • JetStream2/wasm/TSF/tsf_type_out_map.c: Added.
  • JetStream2/wasm/TSF/tsf_type_table.c: Added.
  • JetStream2/wasm/TSF/tsf_types.h: Added.
  • JetStream2/wasm/TSF/tsf_util.h: Added.
  • JetStream2/wasm/TSF/tsf_version.c: Added.
  • JetStream2/wasm/TSF/tsf_zip_abstract.c: Added.
  • JetStream2/wasm/TSF/tsf_zip_abstract.h: Added.
  • JetStream2/wasm/TSF/tsf_zip_attr.c: Added.
  • JetStream2/wasm/TSF/tsf_zip_reader.c: Added.
  • JetStream2/wasm/TSF/tsf_zip_writer.c: Added.
5:39 PM Changeset in webkit [237441] by Devin Rousso
  • 2 edits in trunk/Source/WebCore

Fix build after r237431 for platforms that don't support FULLSCREEN_API
Need the bug URL (OOPS!).

Reviewed by Joseph Pecoraro.

No new tests. No change in behavior.

  • inspector/agents/InspectorDOMAgent.cpp:

(WebCore::InspectorDOMAgent::addEventListenersToNode):

5:21 PM Changeset in webkit [237440] by Brent Fulgham
  • 2 edits in trunk/Source/JavaScriptCore

Unreviewed build fix for Visual Studio 2017

  • API/tests/testapi.c:

(testMarkingConstraintsAndHeapFinalizers):
(main):

5:11 PM Changeset in webkit [237439] by Kocsen Chung
  • 14 edits in branches/safari-606-branch

Revert r237373. rdar://problem/45567354

5:11 PM Changeset in webkit [237438] by Kocsen Chung
  • 3 edits in branches/safari-606-branch/Source/JavaScriptCore

Revert r236604. rdar://problem/45567354

5:11 PM Changeset in webkit [237437] by Kocsen Chung
  • 2 edits in branches/safari-606-branch/Source/JavaScriptCore

Revert r236606. rdar://problem/45567354

4:59 PM Changeset in webkit [237436] by Devin Rousso
  • 4 edits
    2 adds in trunk

Web Inspector: Canvas Recording loading goes significantly slower when "Frame" tree element is expanded
https://bugs.webkit.org/show_bug.cgi?id=190497

Reviewed by Joseph Pecoraro.

Source/WebInspectorUI:

Save the WI.TreeElement that are in the DOM, and compare that with any WI.TreeElement
that would be added to the DOM. If all of the following are true, don't edit the DOM:

  • no focused WI.TreeElement that isn't currently visible (we would scroll otherwise)
  • the set of previously visible WI.TreeElements (e.g. not used for scroll padding) has at least one of the WI.TreeElements that would be added from this update
    • this means that the user hasn't scrolled beyond the padding WI.TreeElement
  • there are no WI.TreeElements that would be added from this update that were NOT added in a previous update
    • this covers the case that a WI.TreeElement is inserted in the visible area
  • UserInterface/Base/Utilities.js:
  • UserInterface/Views/TreeOutline.js:

(WI.TreeOutline):
(WI.TreeOutline.prototype.registerScrollVirtualizer):
(WI.TreeOutline.prototype.updateVirtualizedElements.walk):
(WI.TreeOutline.prototype.updateVirtualizedElements):

LayoutTests:

  • inspector/unit-tests/set-utilities-expected.txt: Added.
  • inspector/unit-tests/set-utilities.html: Added.
4:55 PM Changeset in webkit [237435] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

JSContext Inspector: Broken frontend, DOMAgent is used but does not exist
https://bugs.webkit.org/show_bug.cgi?id=190922
<rdar://problem/45569827>

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2018-10-25
Reviewed by Devin Rousso.

  • UserInterface/Controllers/DOMManager.js:

(WI.DOMManager.prototype.requestDocument):
Feature check for agents that may not exist in a JSContext.

4:47 PM Changeset in webkit [237434] by Kocsen Chung
  • 7 edits in branches/safari-606-branch/Source

Versioning.

4:42 PM Changeset in webkit [237433] by realdawei@apple.com
  • 2 edits in trunk/LayoutTests

Layout Test compositing/iframes/display-none-subframe.html is a flaky failure.
https://bugs.webkit.org/show_bug.cgi?id=186406

Unreviewed.

Patch by Dawei Fenton <realdawei@apple.com> on 2018-10-25

  • platform/mac-wk1/TestExpectations: Mark test flaky.
4:09 PM Changeset in webkit [237432] by Devin Rousso
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: add corner rounding to the network timing "blocks"
https://bugs.webkit.org/show_bug.cgi?id=190379

Reviewed by Timothy Hatcher.

  • UserInterface/Views/NetworkTableContentView.css:

(.waterfall .block):
(body[dir=ltr] .waterfall .block): Added.
(body[dir=rtl] .waterfall .block): Added.
(.network-table .waterfall .block + .block): Added.
(.network-table .waterfall .block.filler + .block, .network-table .waterfall .block:not(.request, .response) + :matches(.request, .response)): Added.
(.network-table .waterfall .block:last-child): Added.

3:59 PM Changeset in webkit [237431] by Devin Rousso
  • 21 edits in trunk

Web Inspector: display fullscreen enter/exit events in Timelines and Network node waterfalls
https://bugs.webkit.org/show_bug.cgi?id=189874
<rdar://problem/44700000>

Reviewed by Joseph Pecoraro.

Source/JavaScriptCore:

  • inspector/protocol/DOM.json:

Allow data to be passed to the frontend with didFireEvent.

Source/WebCore:

Updated existing test: http/tests/inspector/dom/didFireEvent.html

  • inspector/agents/InspectorDOMAgent.h:
  • inspector/agents/InspectorDOMAgent.cpp:

(WebCore::EventFiredCallback::handleEvent):
(WebCore::InspectorDOMAgent::didCreateFrontendAndBackend):
(WebCore::InspectorDOMAgent::addEventListenersToNode):
(WebCore::InspectorDOMAgent::discardBindings):
(WebCore::InspectorDOMAgent::eventDidResetAfterDispatch): Added.
Prevent the same event from being sent to the frontend more than once.

  • dom/Event.cpp:

(WebCore::Event::resetAfterDispatch):

  • dom/Document.cpp:

(WebCore::Document::Document):

  • inspector/InspectorInstrumentation.h:

(WebCore::InspectorInstrumentation::eventDidResetAfterDispatch): Added.

  • inspector/InspectorInstrumentation.cpp:

(WebCore::InspectorInstrumentation::eventDidResetAfterDispatchImpl): Added.

Source/WebInspectorUI:

  • Localizations/en.lproj/localizedStrings.js:
  • UserInterface/Protocol/DOMObserver.js:

(WI.DOMObserver.prototype.didFireEvent):

  • UserInterface/Controllers/DOMManager.js:

(WI.DOMManager.prototype.didFireEvent):
Allow data to be passed to the frontend with didFireEvent.

  • UserInterface/Models/DOMNode.js:

(WI.DOMNode):
(WI.DOMNode.getFullscreenDOMEvents): Added.
(WI.DOMNode.prototype.didFireEvent):
(WI.DOMNode.prototype._handleDOMNodeDidFireEvent): Added.
(WI.DOMNode.prototype._addDOMEvent):
(WI.DOMNode.prototype._shouldListenForEventListeners): Added.
If an event is fired on an ancestor of this node, also record that event in this node's
domEvents, including the originator node.

  • UserInterface/Views/NetworkTableContentView.js:

(WI.NetworkTableContentView.prototype._populateWaterfallGraph):

  • UserInterface/Views/NetworkTableContentView.css:

(.network-table :not(.header) .cell.waterfall .waterfall-container > .dom-fullscreen): Added.

  • UserInterface/Views/DOMEventsBreakdownView.js:

(WI.DOMEventsBreakdownView.prototype.initialLayout):
(WI.DOMEventsBreakdownView.prototype._populateTable):

  • UserInterface/Views/DOMEventsBreakdownView.css:

(.dom-events-breakdown .graph > .area.fullscreen): Added.
(.dom-events-breakdown .inherited > .name, .dom-events-breakdown .inherited > .graph > .point): Added.
(.dom-events-breakdown:not(.has-inherited) .originator): Added.

LayoutTests:

  • http/tests/inspector/dom/didFireEvent-expected.txt:
  • http/tests/inspector/dom/didFireEvent.html:
3:57 PM Changeset in webkit [237430] by Devin Rousso
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Network: more aggressively snap timing blocks together
https://bugs.webkit.org/show_bug.cgi?id=190439

Reviewed by Timothy Hatcher.

  • UserInterface/Views/NetworkTableContentView.js:

(WI.NetworkTableContentView.prototype._populateWaterfallGraph.appendBlock):
(WI.NetworkTableContentView.prototype._populateWaterfallGraph):
If the time difference between the end of the previous block and the start of this block
would result in less than 2px of space, extend the next block back to the previous block.

3:23 PM Changeset in webkit [237429] by Ross Kirsling
  • 18 edits in trunk/Source

Cleanup: inline constexpr is redundant as constexpr implies inline
https://bugs.webkit.org/show_bug.cgi?id=190819

Reviewed by Mark Lam.

Source/bmalloc:

  • bmalloc/Algorithm.h:

(bmalloc::max):
(bmalloc::min):
(bmalloc::mask):
(bmalloc::test):
(bmalloc::isPowerOfTwo):
(bmalloc::roundDownToMultipleOf):
(bmalloc::sizeOf):
(bmalloc::bitCount):
(bmalloc::log2):

  • bmalloc/Bits.h:

(bmalloc::bitsArrayLength):

  • bmalloc/Sizes.h:

(bmalloc::Sizes::maskSizeClass):

Source/JavaScriptCore:

  • bytecode/ArrayProfile.h:

(JSC::asArrayModes):

  • runtime/IndexingType.h:

(JSC::isCopyOnWrite):

  • runtime/MathCommon.h:

(JSC::maxSafeInteger):
(JSC::minSafeInteger):

  • runtime/StackAlignment.h:

(JSC::stackAlignmentBytes):
(JSC::stackAlignmentRegisters):

Source/WTF:

  • wtf/Bitmap.h:

(WTF::WordType>::Bitmap):

  • wtf/LEBDecoder.h:

(WTF::LEBDecoder::maxByteLength):

  • wtf/MathExtras.h:

(defaultMinimumForClamp):
(defaultMaximumForClamp):
(clampToAccepting64):
(isLessThan):
(isLessThanEqual):
(isGreaterThan):
(isGreaterThanEqual):
(WTF::roundUpToPowerOfTwo):
(WTF::maskForSize):

  • wtf/Optional.h:
  • wtf/PtrTag.h:

(WTF::tagCodePtr):
(WTF::untagCodePtr):
(WTF::retagCodePtr):
(WTF::removeCodePtrTag):

  • wtf/StdLibExtras.h:

(WTF::roundUpToMultipleOf):

  • wtf/Variant.h:

(WTF::operator==):
(WTF::operator!=):
(WTF::operator>=):
(WTF::operator<=):
(WTF::operator>):
(WTF::operator<):

  • wtf/text/StringImpl.h:

(WTF::StringImplShape::StringImplShape):
(WTF::StringImpl::StaticStringImpl::StaticStringImpl):

2:17 PM EnvironmentVariables edited by Michael Catanzaro
Add sandbox envvars (diff)
1:49 PM Changeset in webkit [237428] by Alan Coon
  • 7 edits in tags/Safari-607.1.10.5/Source

Versioning.

1:09 PM Changeset in webkit [237427] by Alan Coon
  • 1 copy in tags/Safari-607.1.10.5

New tag.

1:04 PM Changeset in webkit [237426] by ap@apple.com
  • 2 edits in trunk/LayoutTests

https://bugs.webkit.org/show_bug.cgi?id=175597
XMLHttpRequest should not sniff content encoding

Actually, the discussion in Bugzilla claims that it should always have worked on
iOS. Not sure why the test was not enabled from the start.

  • platform/ios/TestExpectations:
1:01 PM Changeset in webkit [237425] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Update test results for https://bugs.webkit.org/show_bug.cgi?id=175597
XMLHttpRequest should not sniff content encoding

This code doesn't work like planned on macOS High Sierra, and I'm not sure what's
up with iOS and other platforms. But I can to re-enable the test for Mojave.

  • platform/mac/TestExpectations: As expected, this test is passing on macOS Mojave.
12:55 PM Changeset in webkit [237424] by Michael Catanzaro
  • 2 edits in trunk/Source/WebCore

Unreviewed, silence a -Wreturn-type warning

When every return is supposed to be covered by a switch statement, a release assert or CRASH
is required by GCC.

  • Modules/mediastream/libwebrtc/LibWebRTCCertificateGenerator.cpp:

(WebCore::LibWebRTCCertificateGenerator::keyParamsFromCertificateType):

12:53 PM Changeset in webkit [237423] by Michael Catanzaro
  • 2 edits in trunk/Source/WebKit

Unreviewed, another try at fixing the GTK build with sandbox enabled

  • UIProcess/Launcher/glib/BubblewrapLauncher.cpp:

(WebKit::memfd_create): Deleted.

12:51 PM Changeset in webkit [237422] by commit-queue@webkit.org
  • 2 edits in trunk/Tools

Web Inspector: Add a watchlist message rule to warn about feature checking new inspector protocol changes
https://bugs.webkit.org/show_bug.cgi?id=190896

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2018-10-25
Reviewed by Matt Baker.

  • Scripts/webkitpy/common/config/watchlist:
12:17 PM Changeset in webkit [237421] by graouts@webkit.org
  • 8 edits in trunk/Source

[Web Animations] Turn Web Animations CSS Integration off by default
https://bugs.webkit.org/show_bug.cgi?id=190901

Reviewed by Dean Jackson.

Source/WebCore:

  • page/RuntimeEnabledFeatures.h:

Source/WebKit:

  • Shared/WebPreferences.yaml:

Source/WebKitLegacy/mac:

  • WebView/WebPreferences.mm:

(+[WebPreferences initialize]):

Source/WebKitLegacy/win:

  • WebPreferences.cpp:

(WebPreferences::initializeDefaultSettings):

10:44 AM Changeset in webkit [237420] by pvollan@apple.com
  • 3 edits in trunk/LayoutTests

[High Sierra/Sierra Release] Layout Test media/track/track-cue-css.html is a flaky image failure
https://bugs.webkit.org/show_bug.cgi?id=190876

Reviewed by Alexey Proskuryakov.

This is a speculative fix; avoid an unnecessary video seek step in test.

  • media/track/track-cue-css-expected.html:
  • media/track/track-cue-css.html:
10:41 AM Changeset in webkit [237419] by ggaren@apple.com
  • 12 edits in trunk

HashMap should support selecting a random entry
https://bugs.webkit.org/show_bug.cgi?id=190814

Reviewed by Antti Koivisto.

Source/WTF:

In some cases, remove(begin()) is not quite good enough as a random
eviction strategy. (See https://bugs.webkit.org/show_bug.cgi?id=190792.)
So, let's support real random eviction.

(And by "real" I mean "pseudo".)

  • wtf/HashCountedSet.h:
  • wtf/HashMap.h:
  • wtf/HashSet.h:
  • wtf/ListHashSet.h:

(WTF::ListHashSet::random):
(WTF::ListHashSet::random const):

  • wtf/LoggingHashMap.h:
  • wtf/LoggingHashSet.h: Plumb through the random() iterator.
  • wtf/HashTable.h:

(WTF::HashTable::random):
(WTF::HashTable::random const): Implement the random() iterator.
makeIterator() already supports starting from any bucket, so this is
pretty easy.

In the subtle case where we select end(), we choose to wrap around and
return begin(). We expect that clients don't really think of the end()
bucket as being in the domain of the random search. Also, we don't want
to annoy clients who know their tables are non-empty with needless
checks for end().

  • wtf/RandomNumber.cpp:

(WTF::weakRandomUint32):

  • wtf/RandomNumber.h: Added a free function for weak random numbers so

that clients that want cheap random numbers aren't required to allocate
storage for a WeakRandom generator.

Tools:

Unit testing is fun and easy!

  • TestWebKitAPI/Tests/WTF/HashMap.cpp:

(TestWebKitAPI::ZeroHash::hash):
(TestWebKitAPI::TEST):

10:09 AM Changeset in webkit [237418] by Jon Davis
  • 4 edits in trunk

Source/WebCore:
Changed "Under Development" status to use "In Development" instead
https://bugs.webkit.org/show_bug.cgi?id=187615

Reviewed by Joseph Pecoraro.

  • features.json: Updated CSS Painting API Level 1 and CSS Properties and Values API Level 1.

Websites/webkit.org:
Added support for "deprecated" feature status
https://bugs.webkit.org/show_bug.cgi?id=187615

Reviewed by Joseph Pecoraro.

  • wp-content/themes/webkit/status.php:
10:06 AM Changeset in webkit [237417] by Michael Catanzaro
  • 2 edits in trunk/Source/WebKit

Unreviewed, try to fix GTK build with sandbox enabled

  • UIProcess/Launcher/glib/BubblewrapLauncher.cpp:
9:33 AM Changeset in webkit [237416] by Fujii Hironori
  • 2 edits in trunk/Source/WebKitLegacy

[CMake][Ninja][Win] midl.exe fails to create, open and write a temporary file in parallel build in Docker container
https://bugs.webkit.org/show_bug.cgi?id=187725

Reviewed by Michael Catanzaro.

  • PlatformWin.cmake: Added USES_TERMINAL to serialize midl.exe

executions.

8:55 AM Changeset in webkit [237415] by Claudio Saavedra
  • 2 edits in trunk/Source/WebKit

[WPE][GTK] Follow-up fixes to the certificate coders
https://bugs.webkit.org/show_bug.cgi?id=190900

Reviewed by Michael Catanzaro.

Cleanup the GRefPtr mess I created earlier. Also we now bail out
if any certificate in the chain has no data instead of returning a
cropped chain.

  • Shared/soup/WebCoreArgumentCodersSoup.cpp:

(IPC::ArgumentCoder<CertificateInfo>::encode):
(IPC::ArgumentCoder<CertificateInfo>::decode):

7:53 AM Changeset in webkit [237414] by Michael Catanzaro
  • 2 edits in trunk

[GTK] Reenable the sandbox

Previously reviewed by, er, myself. This just reverts the sabotague as our bots should have
the required deps now. Hopefully.

  • Source/cmake/OptionsGTK.cmake:
7:15 AM Changeset in webkit [237413] by Chris Dumez
  • 3 edits
    2 adds in trunk

REGRESSION (236779) scandinaviandesigns.com product pages auto redirect to product image
https://bugs.webkit.org/show_bug.cgi?id=190891
<rdar://problem/45296796>

Reviewed by Antti Koivisto.

Source/WebCore:

When a radio element gets clicked, we should only fire the 'input' and 'change' if the checked state
of the radio element has changed.

Test: fast/dom/HTMLInputElement/radio-element-fires-change-event-only-when-checked-state-changes.html

  • html/RadioInputType.cpp:

(WebCore::RadioInputType::didDispatchClick):

LayoutTests:

Add layout test coverage.

  • fast/dom/HTMLInputElement/radio-element-fires-change-event-only-when-checked-state-changes-expected.txt: Added.
  • fast/dom/HTMLInputElement/radio-element-fires-change-event-only-when-checked-state-changes.html: Added.
7:03 AM Changeset in webkit [237412] by fred.wang@free.fr
  • 2 edits
    3 adds in trunk/LayoutTests

Unreviewed test gardening.

Patch by Frederic Wang <fwang@igalia.com> on 2018-10-25

  • platform/ios-wk1/TestExpectations: Add failures for tests requiring findString.
  • platform/ios-wk1/fast/scrolling/adjust-scroll-offset-on-zoom-expected.txt: Added. Adjust

expectation to handle whitespace difference.

  • platform/ios-wk1/fast/scrolling/scrolling-tree-includes-frame-expected.txt: Added. The

scrolling tree is not displayed on WK1.

5:46 AM Changeset in webkit [237411] by fred.wang@free.fr
  • 2 edits
    8 adds in trunk/LayoutTests

Add more tests for the Find UI
https://bugs.webkit.org/show_bug.cgi?id=190823

Patch by Frederic Wang <fwang@igalia.com> on 2018-10-25
Reviewed by Antonio Gomes.

We add more tests for iOS's find UI to cover cases fixed by r224284. Equivalent tests for
subframes currently fails when frame flattening is disabled, they will be handled in bug
184297.

  • platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position-expected.png: Added.
  • platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position-expected.txt: Added.
  • platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position-limit-expected.png: Added.
  • platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position-limit-expected.txt: Added.
  • platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position-limit.html: Added.

Test with a match at the limit of the scrolling element.

  • platform/ios/fast/scrolling/find-text-in-overflow-node-indicator-position.html: Added. Test

with a match at the middle of the scrolling element.

  • platform/ios/fast/scrolling/find-text-in-overflow-node.html: Use names for parameters to

make it more explicit.

  • platform/ios/fast/scrolling/find-text-in-subframe-expected.txt: Added.
  • platform/ios/fast/scrolling/find-text-in-subframe.html: Added. Test with a subframe instead

of an overflow node.

5:07 AM WebKitGTK/2.22.x edited by Adrian Perez de Castro
(diff)
4:09 AM Changeset in webkit [237410] by zandobersek@gmail.com
  • 13 edits in trunk/Source/WebKit

[GTK][AC] Resizing the window doesn't always update the visible rect
https://bugs.webkit.org/show_bug.cgi?id=189743

Reviewed by Michael Catanzaro.

Disassociate resizing that can be done on an AcceleratedSurface object
between "host resizes" and "client resizes".

The former is done from ThreadedCoordinatedLayerTreeHost directly, and
is currently used only for GTK on X11, where a new pixmap object is
created on each resize, which affects the context ID that is based on
that pixmap object's address.

The latter is done from the composition thread. It's used for GTK on
Wayland and WPE. In both cases, the underlying window object does not
change on each resize, but it's necessary to perform the actual resize
on the composition thread.

So far it hasn't been performed there, which ended up pushing sizes of
the WebKit view and the underlying window object out of sync during
resizing, leaving parts of window unrendered.

  • Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.cpp:

(WebKit::ThreadedCompositor::renderLayerTree):

  • Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.h:
  • WebProcess/WebPage/AcceleratedSurface.cpp:

(WebKit::AcceleratedSurface::hostResize):
(WebKit::AcceleratedSurface::resize): Deleted.

  • WebProcess/WebPage/AcceleratedSurface.h:

(WebKit::AcceleratedSurface::clientResize):

  • WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.cpp:

(WebKit::ThreadedCoordinatedLayerTreeHost::deviceOrPageScaleFactorChanged):
(WebKit::ThreadedCoordinatedLayerTreeHost::sizeDidChange):

  • WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.h:
  • WebProcess/WebPage/gtk/AcceleratedSurfaceWayland.cpp:

(WebKit::AcceleratedSurfaceWayland::clientResize):
(WebKit::AcceleratedSurfaceWayland::resize): Deleted.

  • WebProcess/WebPage/gtk/AcceleratedSurfaceWayland.h:
  • WebProcess/WebPage/gtk/AcceleratedSurfaceX11.cpp:

(WebKit::AcceleratedSurfaceX11::hostResize):
(WebKit::AcceleratedSurfaceX11::resize): Deleted.

  • WebProcess/WebPage/gtk/AcceleratedSurfaceX11.h:
  • WebProcess/WebPage/wpe/AcceleratedSurfaceWPE.cpp:

(WebKit::AcceleratedSurfaceWPE::clientResize):
(WebKit::AcceleratedSurfaceWPE::resize): Deleted.

  • WebProcess/WebPage/wpe/AcceleratedSurfaceWPE.h:
1:34 AM Changeset in webkit [237409] by commit-queue@webkit.org
  • 6 edits in trunk/Source/WebCore

InspectorCanvas is not getting cleared properly for OffscreenCanvas
https://bugs.webkit.org/show_bug.cgi?id=190894
<rdar://problem/45498435>

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2018-10-25
Reviewed by Simon Fraser.

Covered by existing tests not crashing with guard malloc.

InspectorCanvasAgents tracks all CanvasRenderingContexts and needs to
remove its reference when the containing CanvasBase goes away. It does
this by registering as a notification observer, but if it can't map
from the CanvasBase back to the rendering context we were failing to
remove our reference. Enforce CanvasBase classes to notify observers
of destruction while they still have their CanvasRenderingContext.

  • html/CanvasBase.cpp:

(WebCore::CanvasBase::~CanvasBase):
(WebCore::CanvasBase::notifyObserversCanvasDestroyed):

  • html/CanvasBase.h:

Assert that subclasses notify observers of the canvas being destroyed,
since they will need to do this before m_context is cleared.

  • html/CustomPaintCanvas.cpp:

(WebCore::CustomPaintCanvas::~CustomPaintCanvas):

  • html/OffscreenCanvas.cpp:

(WebCore::OffscreenCanvas::~OffscreenCanvas):
Follow the new expected pattern of notifying observers before clearing
the context. HTMLCanvasElement already followed this pattern.

  • inspector/agents/InspectorCanvasAgent.cpp:

(WebCore::InspectorCanvasAgent::canvasDestroyed):
Add an assertion that would catch this earlier.

12:24 AM Changeset in webkit [237408] by jiewen_tan@apple.com
  • 7 edits in trunk/LayoutTests

[WPT] Update resource-timing WPT tests to use the right cross origin
https://bugs.webkit.org/show_bug.cgi?id=190554

Reviewed by Youenn Fablet.

LayoutTests/imported/w3c:

  • web-platform-tests/resource-timing/resource_TAO_cross_origin_redirect_chain-expected.txt:
  • web-platform-tests/resource-timing/resource_TAO_cross_origin_redirect_chain.html:
  • web-platform-tests/resource-timing/resource_timing_cross_origin_redirect_chain-expected.txt:
  • web-platform-tests/resource-timing/resource_timing_cross_origin_redirect_chain.html:

LayoutTests:

Oct 24, 2018:

11:12 PM Changeset in webkit [237407] by timothy_horton@apple.com
  • 2 edits in trunk/Tools

REGRESSION (r237331): DismissingActionSheetShouldNotDismissPresentingViewController times out
https://bugs.webkit.org/show_bug.cgi?id=190897

  • TestWebKitAPI/Tests/ios/ActionSheetTests.mm:

(TestWebKitAPI::TEST):
Disable the test for now.

10:02 PM Changeset in webkit [237406] by mitz@apple.com
  • 9 copies
    1 add in releases/Apple/Safari Technology Preview 68

Added a tag for Safari Technology Preview release 68.

5:19 PM Changeset in webkit [237405] by ap@apple.com
  • 18 edits in trunk

Clean up some obsolete macOS version guards
https://bugs.webkit.org/show_bug.cgi?id=190887

Reviewed by Dan Bernstein.

Source/WebCore:

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

(WebCore::MediaPlayerPrivateAVFoundationObjC::createAVAssetForURL):
(WebCore::MediaPlayerPrivateAVFoundationObjC::maximumDurationToCacheMediaTime const):
(WebCore::MediaPlayerPrivateAVFoundationObjC::didPassCORSAccessCheck const):
(WebCore::MediaPlayerPrivateAVFoundationObjC::wouldTaintOrigin const):

Source/WebCore/PAL:

  • pal/spi/mac/TUCallSPI.h:

Source/WebKit:

  • PluginProcess/mac/com.apple.WebKit.plugin-common.sb.in:
  • UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::WebProcessPool::platformInitializeWebProcess):

Source/WebKitLegacy/mac:

  • Misc/WebNSPasteboardExtras.mm:

(+[NSPasteboard _web_dragTypesForURL]):

  • WebCoreSupport/PopupMenuMac.mm:

(PopupMenuMac::show):

  • WebCoreSupport/WebEditorClient.h:
  • WebCoreSupport/WebEditorClient.mm:

(WebEditorClient::respondToChangedSelection):

  • WebView/WebPDFView.mm:

(-[WebPDFView _clipViewForPDFDocumentView]):

Tools:

  • TestWebKitAPI/cocoa/TestWKWebView.mm:

(simulated_forceClickAssociatedEventsMask):
(-[TestWKWebViewHostWindow _mouseDownAtPoint:simulatePressure:clickCount:]):

  • WebKitTestRunner/cocoa/TestControllerCocoa.mm:

(WTR::initializeWebViewConfiguration):

  • WebKitTestRunner/mac/EventSenderProxy.mm:

(-[EventSenderSyntheticEvent initPressureEventAtLocation:globalLocation:stage:pressure:stageTransition:phase:time:eventNumber:window:]):

4:52 PM Changeset in webkit [237404] by Kocsen Chung
  • 1 copy in tags/Safari-606.3.2

Tag Safari-606.3.2.

3:51 PM Changeset in webkit [237403] by timothy_horton@apple.com
  • 6 edits
    2 adds in trunk

Attachment filenames with RTL characters should format similar to Finder
https://bugs.webkit.org/show_bug.cgi?id=190736
<rdar://problem/44735946>

Reviewed by Dan Bernstein.

Source/WebCore:

Test: fast/attachment/attachment-title-with-rtl.html

  • html/HTMLAttachmentElement.cpp:

(WebCore::HTMLAttachmentElement::attachmentTitleForDisplay const):

  • html/HTMLAttachmentElement.h:

Add attachmentTitleForDisplay(), which wraps the non-extension part of
attachmentTitle in BiDi isolates, matching Finder's behavior.

  • rendering/RenderThemeIOS.mm:

(WebCore::RenderAttachmentInfo::buildWrappedLines):
(WebCore::RenderAttachmentInfo::RenderAttachmentInfo):

  • rendering/RenderThemeMac.mm:

(WebCore::AttachmentLayout::layOutTitle):
Adopt attachmentTitleForDisplay, and ask CoreText to use a subrange of
the original string for the last line, instead of splitting the string
ourselves. This ensures that BiDi control characters are respected
even in the last line of the string.

LayoutTests:

  • fast/attachment/attachment-title-with-rtl-expected.html: Added.
  • fast/attachment/attachment-title-with-rtl.html: Added.

Add a test that directionality marks in the attachment's title are
isolated from the file extension.

3:28 PM Changeset in webkit [237402] by Megan Gardner
  • 24 edits in trunk

Turn on Conic Gradients
https://bugs.webkit.org/show_bug.cgi?id=190810

Reviewed by Tim Horton.

Source/JavaScriptCore:

  • Configurations/FeatureDefines.xcconfig:

Source/WebCore:

Added tests previously, only switching feature from experimental to always avaiable.

  • Configurations/FeatureDefines.xcconfig:
  • css/parser/CSSParserContext.cpp:

(WebCore::CSSParserContext::CSSParserContext):
(WebCore::operator==):

  • css/parser/CSSParserContext.h:

(WebCore::CSSParserContextHash::hash):

  • css/parser/CSSPropertyParserHelpers.cpp:

(WebCore::CSSPropertyParserHelpers::consumeConicGradient):

  • page/Settings.yaml:

Source/WebCore/PAL:

  • Configurations/FeatureDefines.xcconfig:

Source/WebKit:

Remove conic gradients from experimental features and turn it on always.

  • Configurations/FeatureDefines.xcconfig:
  • Shared/WebPreferences.yaml:

Source/WebKitLegacy/mac:

  • Configurations/FeatureDefines.xcconfig:

Source/WTF:

  • wtf/FeatureDefines.h:

Tools:

  • TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

LayoutTests:

No longer necessary to manually turn on conic gradients for testing.

  • http/wpt/css/css-images-4/conic-gradient-parsing.html:
3:12 PM Changeset in webkit [237401] by Devin Rousso
  • 3 edits in trunk/Source/WebInspectorUI

Web Inspector: popovers for function source code are malformed
https://bugs.webkit.org/show_bug.cgi?id=190859

Reviewed by Joseph Pecoraro.

Previously, the WI.Popover would be drawn twice: once when the
WI.CodeMirrorTokenTrackingController determines that the user has hovered on a function
token, and once again when the source for that function is formatted (after being retrieved).
In the case that the formatter is able to return the prettified source within a frame (or
two), the WI.Popover is still in the middle of animating to its new size, meaning that the
changes made by the formatted update will be overridden on the next rAF (creates a flash).

  • UserInterface/Views/Popover.js:

(WI.Popover.prototype):
(WI.Popover.prototype._animateFrame.drawBackground):
(WI.Popover.prototype._drawBackground):
Add a member variable to make sure that there is only ever one rAF firing at a time.
Drive-by: rework the background code to only use one canvas.

  • UserInterface/Views/SourceCodeTextEditor.js:

(WI.SourceCodeTextEditor.prototype._showPopoverForFunction.didGetDetails):
Don't show the WI.Popover until the formatter has finished prettifying the function's
source code, as otherwise there is brief moment that the popover appears and is empty.

2:35 PM Changeset in webkit [237400] by msaboff@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

2018-10-24 Michael Saboff <msaboff@apple.com>

Increase executable memory pool from 64MB to 128MB for ARM64
https://bugs.webkit.org/show_bug.cgi?id=190453

Unreviewed, rolling back in r237024.

The original change did impact ARES-6 performance by 4-8%. That will
be investigated separately.

2:01 PM Changeset in webkit [237399] by ap@apple.com
  • 9 edits in trunk/Source/bmalloc

Add BPLATFORM(IOS_FAMILY)
https://bugs.webkit.org/show_bug.cgi?id=190878

Reviewed by Saam Barati.

  • bmalloc/AvailableMemory.cpp:

(bmalloc::memorySizeAccordingToKernel):
(bmalloc::computeAvailableMemory):

  • bmalloc/AvailableMemory.h:

(bmalloc::isUnderMemoryPressure):

  • bmalloc/BPlatform.h:
  • bmalloc/Gigacage.h:
  • bmalloc/Logging.cpp:

(bmalloc::logVMFailure):

  • bmalloc/VMAllocate.h:

(bmalloc::vmPageSizePhysical):

  • bmalloc/bmalloc.h:
  • bmalloc/darwin/MemoryStatusSPI.h:
1:41 PM Changeset in webkit [237398] by Alan Coon
  • 7 edits in branches/safari-606-branch/Source

Versioning.

1:19 PM Changeset in webkit [237397] by Devin Rousso
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: REGRESSION (r237232): changing resources via up/down blurs the TreeOutline in Resources
https://bugs.webkit.org/show_bug.cgi?id=190862

Reviewed by Matt Baker.

  • UserInterface/Views/TextEditor.js:

(WI.TextEditor.prototype.set selectedTextRange):
Don't focus unless the activeElement is not a WebInspector element (e.g. <body>),
meaning that the previously focused element has been removed from the DOM and no longer
needs to keep the focus.

12:27 PM Changeset in webkit [237396] by Devin Rousso
  • 5 edits in trunk/Source/WebInspectorUI

Web Inspector: CSP request payload on medium.com is unreadable, should be pretty-printed
https://bugs.webkit.org/show_bug.cgi?id=190354
<rdar://problem/45090894>

Reviewed by Joseph Pecoraro.

  • UserInterface/Views/TextEditor.js:

(WI.TextEditor.prototype.set string.update):
(WI.TextEditor.prototype._attemptToDetermineMIMEType): Added.
If the content doesn't already have a MIME type, attempt to determine one by trying to
format it as "javascript" (e.g. request JSON that is simply missing a MIME type).

  • UserInterface/Views/TextContentView.js:

(WI.TextContentView):
(WI.TextContentView.prototype._handleTextEditorMIMETypeChanged): Added.

  • UserInterface/Views/TextResourceContentView.js:

(WI.TextResourceContentView):
(WI.TextResourceContentView.prototype._handleTextEditorMIMETypeChanged): Added.

  • UserInterface/Views/ScriptContentView.js:

(WI.ScriptContentView):
(WI.ScriptContentView.prototype._handleTextEditorMIMETypeChanged): Added.
Enable the "Pretty Print" navigation button if the MIME type changes to something that is
able to be formatted.
Drive-by: reorder the creation of the "Pretty Print" button so that it exists if
_attemptToDetermineMIMEType finishes synchronously.

11:49 AM Changeset in webkit [237395] by jiewen_tan@apple.com
  • 5 edits
    5 adds in trunk

Only report resource timing to parent frame for the first iframe load
https://bugs.webkit.org/show_bug.cgi?id=190498
<rdar://problem/44347398>

Reviewed by Youenn Fablet.

Source/WebCore:

Only the first iframe navigation or the first iframe navigation after about:blank should be reported.
https://www.w3.org/TR/resource-timing-2/#resources-included-in-the-performanceresourcetiming-interface

Test: http/tests/misc/resource-timing-navigation-in-restored-iframe.html

  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::loadWithDocumentLoader):

  • loader/FrameLoader.h:

(WebCore::FrameLoader::shouldReportResourceTimingToParentFrame):
(WebCore::FrameLoader::setShouldReportResourceTimingToParentFrame): Deleted.

  • loader/ResourceTimingInformation.cpp:

(WebCore::ResourceTimingInformation::addResourceTiming):

LayoutTests:

The layout test is from Chromium change:
https://chromium-review.googlesource.com/c/chromium/src/+/1186215.

  • http/tests/misc/resource-timing-navigation-in-restored-iframe-expected.txt: Added.
  • http/tests/misc/resource-timing-navigation-in-restored-iframe.html: Added.
  • http/tests/misc/resources/alert-then-back.html: Added.
  • http/tests/misc/resources/navigate-on-message.html: Added.
  • http/tests/misc/resources/post-message-to-parent.html: Added.
11:46 AM Changeset in webkit [237394] by Brent Fulgham
  • 17 edits in trunk/Source/WebCore

Cure Windows Direct2D Backend of a nasty case of bitrot
https://bugs.webkit.org/show_bug.cgi?id=190875
<rdar://problem/45523268>

Reviewed by Zalan Bujtas.

The Direct2D backend has drifted out of date with the rest of WebKit.
This patch updates things so we can get a runnable build.

  • platform/graphics/FontPlatformData.h:
  • platform/graphics/ImageBuffer.cpp:

(WebCore::ImageBuffer::create):

  • platform/graphics/ImageBuffer.h:
  • platform/graphics/ImageDecoder.h:
  • platform/graphics/ImageSource.cpp:

(WebCore::ImageSource::setTargetContext):

  • platform/graphics/opentype/OpenTypeMathData.cpp:

(WebCore::OpenTypeMathData::OpenTypeMathData): Need non-default constructor to make Visual
Studio happy.

  • platform/graphics/win/GradientDirect2D.cpp:

(WebCore::Gradient::generateGradient):

  • platform/graphics/win/GraphicsContextDirect2D.cpp: Update for new display list implementation.

(WebCore::GraphicsContextPlatformPrivate::setAlpha):
(WebCore::GraphicsContext::savePlatformState):
(WebCore::GraphicsContext::restorePlatformState):
(WebCore::GraphicsContext::drawNativeImage):
(WebCore::GraphicsContext::drawPattern):
(WebCore::GraphicsContext::drawRect):
(WebCore::GraphicsContext::drawLine):
(WebCore::GraphicsContext::drawEllipse):
(WebCore::GraphicsContext::drawPath):
(WebCore::GraphicsContext::fillPath):
(WebCore::GraphicsContext::strokePath):
(WebCore::GraphicsContext::fillRect):
(WebCore::GraphicsContext::platformFillRoundedRect):
(WebCore::GraphicsContext::fillRectWithRoundedHole):
(WebCore::GraphicsContext::clip):
(WebCore::GraphicsContext::clipOut):
(WebCore::GraphicsContext::clipPath):
(WebCore::GraphicsContext::clipBounds const):
(WebCore::GraphicsContext::beginPlatformTransparencyLayer):
(WebCore::GraphicsContext::endPlatformTransparencyLayer):
(WebCore::GraphicsContext::setMiterLimit):
(WebCore::GraphicsContext::clearRect):
(WebCore::GraphicsContext::strokeRect):
(WebCore::GraphicsContext::setLineCap):
(WebCore::GraphicsContext::setLineDash):
(WebCore::GraphicsContext::setLineJoin):
(WebCore::GraphicsContext::scale):
(WebCore::GraphicsContext::rotate):
(WebCore::GraphicsContext::translate):
(WebCore::GraphicsContext::concatCTM):
(WebCore::GraphicsContext::setCTM):
(WebCore::GraphicsContext::getCTM const):
(WebCore::GraphicsContext::roundToDevicePixels):
(WebCore::GraphicsContext::drawLinesForText):
(WebCore::GraphicsContext::setURLForRect):
(WebCore::GraphicsContext::setIsCALayerContext):
(WebCore::GraphicsContext::isCALayerContext const):
(WebCore::GraphicsContext::setIsAcceleratedContext):
(WebCore::GraphicsContext::isAcceleratedContext const):
(WebCore::GraphicsContext::setPlatformShouldAntialias):
(WebCore::GraphicsContext::setPlatformShouldSmoothFonts):
(WebCore::GraphicsContext::setPlatformCompositeOperation):
(WebCore::GraphicsContext::platformFillEllipse):
(WebCore::GraphicsContext::platformStrokeEllipse):

  • platform/graphics/win/GraphicsLayerDirect2D.cpp:

(WebCore::GraphicsLayer::create):
(): Deleted.

  • platform/graphics/win/ImageBufferDataDirect2D.cpp:

(WebCore::ImageBufferData::putData):

  • platform/graphics/win/ImageBufferDataDirect2D.h:
  • platform/graphics/win/ImageBufferDirect2D.cpp:

(WebCore::ImageBuffer::createCompatibleBuffer):
(WebCore::ImageBuffer::ImageBuffer):

  • platform/graphics/win/ImageDecoderDirect2D.cpp:

(WebCore::ImageDecoderDirect2D::bytesDecodedToDetermineProperties const):
(WebCore::ImageDecoderDirect2D::encodedDataStatus const):
(WebCore::ImageDecoderDirect2D::repetitionCount const):
(WebCore::ImageDecoderDirect2D::frameOrientationAtIndex const):
(WebCore::ImageDecoderDirect2D::frameDurationAtIndex const):
(WebCore::ImageDecoderDirect2D::frameAllowSubsamplingAtIndex const):
(WebCore::ImageDecoderDirect2D::frameHasAlphaAtIndex const):
(WebCore::ImageDecoderDirect2D::createFrameImageAtIndex):
(WebCore::ImageDecoderDirect2D::setData):
(WebCore::ImageDecoderDirect2D::bytesDecodedToDetermineProperties): Deleted.
(WebCore::ImageDecoderDirect2D::createFrameImageAtIndex const): Deleted.

  • platform/graphics/win/ImageDecoderDirect2D.h:
  • platform/graphics/win/ImageDirect2D.cpp:

(WebCore::BitmapImage::drawFrameMatchingSourceSize):

  • platform/graphics/win/PatternDirect2D.cpp:

(WebCore::Pattern::createPlatformPattern const):

11:42 AM Changeset in webkit [237393] by commit-queue@webkit.org
  • 17 edits in trunk/Source

Cleanup: MIMETypeRegistry functions
https://bugs.webkit.org/show_bug.cgi?id=190838

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2018-10-24
Reviewed by Simon Fraser.

Source/WebCore:

Modernize the initialization of the lists in MIMETypeRegistry. Make the
functions that return these list be self-contained. Use NeverDestroy<> to
allocate the local static variable. Use std::initializer_list() and
makeNeverDestroyed() to initialize NeverDestroy<> variables only once.

supportedImageResourceMIMETypes will be deleted and all the calls to it
will be replaced by supportedImageMIMETypes because they are identical.

  • loader/archive/ArchiveFactory.cpp:

(WebCore::ArchiveFactory::registerKnownArchiveMIMETypes): This function
is called while initializing the supportedNonImageMIMETypes(). So it
should not have a direct call to it. Instead, supportedNonImageMIMETypes
is passed to it.

  • loader/archive/ArchiveFactory.h:
  • platform/MIMETypeRegistry.cpp:

(WebCore::MIMETypeRegistry::supportedImageMIMETypes):
(WebCore::supportedImageMIMETypesForEncoding):
(WebCore::supportedJavaScriptMIMETypes):
(WebCore::MIMETypeRegistry::supportedNonImageMIMETypes):
(WebCore::MIMETypeRegistry::supportedMediaMIMETypes):
(WebCore::pdfMIMETypes):
(WebCore::MIMETypeRegistry::unsupportedTextMIMETypes):
(WebCore::MIMETypeRegistry::isSupportedImageMIMEType):
(WebCore::MIMETypeRegistry::isSupportedImageMIMETypeForEncoding):
(WebCore::MIMETypeRegistry::isSupportedJavaScriptMIMEType):
(WebCore::MIMETypeRegistry::isSupportedNonImageMIMEType):
(WebCore::MIMETypeRegistry::isSupportedMediaMIMEType):
(WebCore::MIMETypeRegistry::isUnsupportedTextMIMEType):
(WebCore::MIMETypeRegistry::isPDFMIMEType):
(WebCore::MIMETypeRegistry::systemPreviewMIMETypes):
(WebCore::MIMETypeRegistry::isSystemPreviewMIMEType):
(WebCore::initializeSupportedImageMIMETypes): Deleted.
(WebCore::initializeSupportedImageMIMETypesForEncoding): Deleted.
(WebCore::initializeSupportedJavaScriptMIMETypes): Deleted.
(WebCore::initializePDFMIMETypes): Deleted.
(WebCore::initializeSupportedNonImageMimeTypes): Deleted.
(WebCore::initializeSupportedMediaMIMETypes): Deleted.
(WebCore::initializeUnsupportedTextMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::isSupportedImageResourceMIMEType): Deleted.
(WebCore::MIMETypeRegistry::getSupportedImageMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getSupportedImageResourceMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getSupportedNonImageMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getSupportedMediaMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getPDFMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getUnsupportedTextMIMETypes): Deleted.
(WebCore::MIMETypeRegistry::getSystemPreviewMIMETypes): Deleted.

  • platform/MIMETypeRegistry.h:
  • platform/graphics/Image.cpp:

(WebCore::Image::supportsType):

  • platform/graphics/cg/ImageDecoderCG.cpp:

(WebCore::ImageDecoderCG::encodedDataStatus const):

  • platform/graphics/cg/UTIRegistry.cpp:

(WebCore::supportedDefaultImageSourceTypes):
(WebCore::isSupportImageSourceType):
(WebCore::allowedImageUTIs): Deleted.
(WebCore::isAllowedImageUTI): Deleted.

  • platform/graphics/cg/UTIRegistry.h:
  • platform/mac/PasteboardMac.mm:

(WebCore::Pasteboard::write):

Source/WebKit:

  • UIProcess/Cocoa/WKWebViewContentProviderRegistry.mm:

(-[WKWebViewContentProviderRegistry initWithConfiguration:]):

Source/WebKitLegacy/mac:

  • Misc/WebNSPasteboardExtras.mm:

(-[NSPasteboard _web_writePromisedRTFDFromArchive:containsImage:]):
Replace isSupportedImageResourceMIMEType() by isSupportedImageMIMEType()
because they were identical.

  • WebView/WebHTMLRepresentation.mm:

(+[WebHTMLRepresentation supportedMediaMIMETypes]):
(+[WebHTMLRepresentation supportedNonImageMIMETypes]):
(+[WebHTMLRepresentation supportedImageMIMETypes]):
(+[WebHTMLRepresentation unsupportedTextMIMETypes]):

  • WebView/WebView.mm:

(+[WebView _unregisterViewClassAndRepresentationClassForMIMEType:]):
(+[WebView _registerViewClass:representationClass:forURLScheme:]):
(+[WebView registerViewClass:representationClass:forMIMEType:]):

11:26 AM Changeset in webkit [237392] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

[macOS] Layout Test legacy-animation-engine/animations/suspend-resume-animation.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=190883

Unreviewed test gardening.

  • platform/mac/TestExpectations: Mark test as flaky.
11:24 AM Changeset in webkit [237391] by Matt Lewis
  • 2 edits in trunk/LayoutTests

Marked webgl/1.0.2/conformance/glsl/misc/shader-with-reserved-words.html as a flaky timout.
https://bugs.webkit.org/show_bug.cgi?id=170877

Unreviewed test gardening.

  • platform/mac/TestExpectations:
10:44 AM Changeset in webkit [237390] by commit-queue@webkit.org
  • 3 edits in trunk/Tools

[WPE] Fix CMake rules in for TestWebKitAPIBase library building in developer mode
https://bugs.webkit.org/show_bug.cgi?id=190328

Patch by Pablo Saavedra <Pablo Saavedra> on 2018-10-24
Reviewed by Michael Catanzaro.

  • TestWebKitAPI/CMakeLists.txt:
  • TestWebKitAPI/PlatformWPE.cmake:
10:18 AM Changeset in webkit [237389] by Kocsen Chung
  • 4 edits in branches/safari-606-branch/Source/WebKit

Cherry-pick r236227. rdar://problem/45491949

Crash under WebPageProxy::decidePolicyForNavigationAction()
https://bugs.webkit.org/show_bug.cgi?id=189763
<rdar://problem/44597111>

Reviewed by Alex Christensen.

Update WebNavigationState::navigation() / WebNavigationState::takeNavigation()
to return a pointer instead of a reference as we have evidence that they can
return null. I kept the debug assertions to try and catch the cases where we
return null but at least we stop crashing in release builds.

  • UIProcess/WebNavigationState.cpp: (WebKit::WebNavigationState::navigation): (WebKit::WebNavigationState::takeNavigation):
  • UIProcess/WebNavigationState.h:
  • UIProcess/WebPageProxy.cpp: (WebKit::WebPageProxy::didStartProvisionalLoadForFrame): (WebKit::WebPageProxy::didReceiveServerRedirectForProvisionalLoadForFrame): (WebKit::WebPageProxy::didCommitLoadForFrame): (WebKit::WebPageProxy::didFinishDocumentLoadForFrame): (WebKit::WebPageProxy::didFinishLoadForFrame): (WebKit::WebPageProxy::didFailLoadForFrame): (WebKit::WebPageProxy::didSameDocumentNavigationForFrame): (WebKit::WebPageProxy::decidePolicyForNavigationAction): (WebKit::WebPageProxy::decidePolicyForResponse):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236227 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9:56 AM Changeset in webkit [237388] by Kocsen Chung
  • 2 edits in branches/safari-606-branch/Source/WebCore

Cherry-pick r236747. rdar://problem/45491954

Add release assertion to ensure m_owningPointerForClose is null in UniqueIDBDatabase::invokeOperationAndTransactionTimer()
https://bugs.webkit.org/show_bug.cgi?id=190178

Reviewed by Chris Dumez.

This would help debug rdar://problem/44902833.

  • Modules/indexeddb/server/UniqueIDBDatabase.cpp: (WebCore::IDBServer::UniqueIDBDatabase::invokeOperationAndTransactionTimer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9:56 AM Changeset in webkit [237387] by Kocsen Chung
  • 8 edits in branches/safari-606-branch/Source/WebCore

Cherry-pick r234791. rdar://problem/45491958

CrashTracer: com.apple.WebKit.Storage at WebCore::IDBServer::UniqueIDBDatabase::connectionClosedFromClient(WebCore::IDBServer::UniqueIDBDatabaseConnection&)
https://bugs.webkit.org/show_bug.cgi?id=188474

Reviewed by Chris Dumez.

UniqueIDBDatabaseConnection is refcounted by UniqueIDBDatabaseTransaction and it refcounts
UniqueIDBDatabaseTransaction. This cycle could make UniqueIDBDatabaseConnection outlives
UniqueIDBDatabase, so its reference to UniqueIDBDatabase may be stale. Calling a function
on a stale object is probably the reason of recent various storage process crashes in
indexedDB.

This patch makes m_database a WeakPtr and adds assertions that could help us debug the
crashes.

  • Modules/indexeddb/server/IDBServer.cpp: (WebCore::IDBServer::IDBServer::closeAndDeleteDatabasesModifiedSince): (WebCore::IDBServer::IDBServer::closeAndDeleteDatabasesForOrigins):
  • Modules/indexeddb/server/UniqueIDBDatabase.cpp: (WebCore::IDBServer::UniqueIDBDatabase::commitTransaction): (WebCore::IDBServer::UniqueIDBDatabase::abortTransaction):
  • Modules/indexeddb/server/UniqueIDBDatabase.h:
  • Modules/indexeddb/server/UniqueIDBDatabaseConnection.cpp: (WebCore::IDBServer::UniqueIDBDatabaseConnection::UniqueIDBDatabaseConnection): (WebCore::IDBServer::UniqueIDBDatabaseConnection::~UniqueIDBDatabaseConnection): (WebCore::IDBServer::UniqueIDBDatabaseConnection::abortTransactionWithoutCallback): (WebCore::IDBServer::UniqueIDBDatabaseConnection::connectionClosedFromClient): (WebCore::IDBServer::UniqueIDBDatabaseConnection::confirmDidCloseFromServer): (WebCore::IDBServer::UniqueIDBDatabaseConnection::didFireVersionChangeEvent): (WebCore::IDBServer::UniqueIDBDatabaseConnection::didFinishHandlingVersionChange): (WebCore::IDBServer::UniqueIDBDatabaseConnection::createVersionChangeTransaction): (WebCore::IDBServer::UniqueIDBDatabaseConnection::establishTransaction): (WebCore::IDBServer::UniqueIDBDatabaseConnection::didAbortTransaction):
  • Modules/indexeddb/server/UniqueIDBDatabaseConnection.h: (WebCore::IDBServer::UniqueIDBDatabaseConnection::database):
  • Modules/indexeddb/server/UniqueIDBDatabaseTransaction.cpp: (WebCore::IDBServer::UniqueIDBDatabaseTransaction::UniqueIDBDatabaseTransaction): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::~UniqueIDBDatabaseTransaction): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::abort): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::commit): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::createObjectStore): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::deleteObjectStore): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::renameObjectStore): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::clearObjectStore): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::createIndex): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::deleteIndex): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::renameIndex): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::putOrAdd): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::getRecord): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::getAllRecords): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::getCount): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::deleteRecord): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::openCursor): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::iterateCursor): (WebCore::IDBServer::UniqueIDBDatabaseTransaction::objectStoreIdentifiers):
  • Modules/indexeddb/shared/IDBResultData.cpp: (WebCore::IDBResultData::openDatabaseSuccess): (WebCore::IDBResultData::openDatabaseUpgradeNeeded):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234791 268f45cc-cd09-0410-ab3c-d52691b4dbfc

9:39 AM Changeset in webkit [237386] by Chris Dumez
  • 4 edits in trunk

[PSON] When navigating back and forth, 'about:blank' shows up in the back/forward list
https://bugs.webkit.org/show_bug.cgi?id=190846
<rdar://problem/45058938>

Reviewed by Antti Koivisto.

Source/WebCore:

When a page gets suspended after a process-swap, we navigate it to about:blank from inside the navigation
policy handler, by overriding the request URL. This normally works fine because we usually process-swap
on standard navigation. However, when we would process-swap on a back/forward navigation, we would end
up using the back/forward navigation load type to do the about:blank load. This would have repercussions
because history navigations update the current history item with the new URL (in this case 'about:blank').
To avoid the issue, switch to a standard load type whenever the client asks us to suspend and we load
'about:blank' as a result.

  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::continueLoadAfterNavigationPolicy):

Tools:

Add API test coverage.

  • TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
9:33 AM Changeset in webkit [237385] by Truitt Savell
  • 4 edits in trunk/LayoutTests

Marking Multiple tests as flakey at once
https://bugs.webkit.org/show_bug.cgi?id=190847

Unreviewed Test Gardening

  • platform/ios-wk2/TestExpectations:
  • platform/mac-wk1/TestExpectations:
  • platform/mac-wk2/TestExpectations:
8:38 AM Changeset in webkit [237384] by Chris Dumez
  • 2 edits in trunk/Source/WebKit

REGRESSION (r237257): [iOS] Crashes in com.apple.WebKit: WebKit::RemoteScrollingCoordinator::scheduleTreeStateCommit
https://bugs.webkit.org/show_bug.cgi?id=190804

Reviewed by Alex Christensen.

Fix null-dereference of the drawing area in RemoteScrollingCoordinator::scheduleTreeStateCommit.
With process-swap-on-navigation enabled, suspended pages' drawing area gets torn down.

  • WebProcess/WebPage/RemoteLayerTree/RemoteScrollingCoordinator.mm:

(WebKit::RemoteScrollingCoordinator::scheduleTreeStateCommit):

7:32 AM Changeset in webkit [237383] by Claudio Saavedra
  • 2 edits
    1 add in trunk/Tools

[GTK] TestSSL failing because of missing libsoup patch
https://bugs.webkit.org/show_bug.cgi?id=190869

Reviewed by Michael Catanzaro.

  • gtk/jhbuild.modules: Add the patch below to fix the API test TestSSL
  • gtk/patches/libsoup-soup-socket-fix-critical-warning-when-the-peer-certi.patch: Added.
7:27 AM Changeset in webkit [237382] by chris.reid@sony.com
  • 4 edits
    2 adds in trunk/Source/WTF

[Win] Add function call name information to stack traces
https://bugs.webkit.org/show_bug.cgi?id=190761

Reviewed by Fujii Hironori.

Add symbol information to stack traces using dbghelp.dll
This library will use symbols files from these sources:

  • The current working directory
  • The directory containing the application's executable
  • _NT_SYMBOL_PATH environment variable
  • _NT_ALTERNATE_SYMBOL_PATH environment variable

This functionality is currently only enabled in debug mode since dbghelp is not threadsafe.
The DbgHelper class attempts to synchronize calls to dbghelp.dll but external code
can still attempt to call the library at the same time as WebKit.

  • wtf/CMakeLists.txt:
  • wtf/PlatformWin.cmake:
  • wtf/StackTrace.cpp:
  • wtf/win/DbgHelperWin.cpp: Added.
  • wtf/win/DbgHelperWin.h: Added.
7:15 AM Changeset in webkit [237381] by aestes@apple.com
  • 4 edits in trunk

[macOS Debug WK2] Layout Test http/tests/ssl/applepay/ApplePayShippingAddressChangeEventErrorsV3.https.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=190650
<rdar://problem/45341914>

Reviewed by Alex Christensen.

Source/WebCore:

In computeErrors, we intend to ignore errors when computing paymentMethodErrors. But we
weren't clearing exceptions generated from calling convert(), so they would end up being
logged as unhandled promise rejections.

Changed ApplePayPaymentHandler::computeErrors() to use a CatchScope to clear exceptions when
decoding paymentMethodErrors.

Also changed ApplePayShippingAddressChangeEventErrorsV3.https.html to catch promise
rejections from calling PaymentRequest.abort().

Covered by existing test.

  • Modules/applepay/paymentrequest/ApplePayPaymentHandler.cpp:

(WebCore::ApplePayPaymentHandler::computeErrors const):

LayoutTests:

  • http/tests/resources/payment-request.js:

(updateDetailsOnShippingAddressChange):

6:45 AM Changeset in webkit [237380] by Claudio Saavedra
  • 2 edits in trunk/Source/WebKit

[WPE][GTK] Cleanups to the certificate decoder
https://bugs.webkit.org/show_bug.cgi?id=190867

Reviewed by Žan Doberšek.

  • Shared/soup/WebCoreArgumentCodersSoup.cpp:

(IPC::ArgumentCoder<CertificateInfo>::decode):

3:20 AM Changeset in webkit [237379] by Antti Koivisto
  • 2 edits in trunk/Source/WebCore

topPrivatelyControlledDomain is slow
https://bugs.webkit.org/show_bug.cgi?id=190792

Reviewed by Alex Christensen and Chris Dumez.

It calls into some slowish CFNetwork code and ends up showing up in profiles.

  • platform/mac/PublicSuffixMac.mm:

(WebCore::topPrivatelyControlledDomain):

Add a cache that avoids calls into frameworks.

3:03 AM Changeset in webkit [237378] by jer.noble@apple.com
  • 5 edits in trunk/Source/WebCore

Use WeakPtr and GenericTaskQueue within ObjC classes used by MediaPlayerPrivateAVFoundationObjC
https://bugs.webkit.org/show_bug.cgi?id=190790

Reviewed by Alex Christensen.

Move towards using WeakPtr callbacks instead of raw pointers within the utility objects used by
MediaPlayerPrivateAVFoundationObjC. Additionally, accessing WeakPtr off the thread which created
the pointer is not allowed, so use a GenericTaskQueue to schedule callbacks instead. Make
GenericTaskQueue<Timer> thread-safe by locking around access to m_pendingTasks, and by making
incrementing the pending task count atomic.

  • platform/GenericTaskQueue.cpp:

(WebCore::TaskDispatcher<Timer>::postTask):
(WebCore::TaskDispatcher<Timer>::sharedLock):
(WebCore::TaskDispatcher<Timer>::sharedTimerFired):
(WebCore::TaskDispatcher<Timer>::dispatchOneTask):
(WebCore::TaskDispatcher<Timer>::pendingDispatchers):

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

(WebCore::MediaPlayerPrivateAVFoundationObjC::MediaPlayerPrivateAVFoundationObjC):
(WebCore::MediaPlayerPrivateAVFoundationObjC::~MediaPlayerPrivateAVFoundationObjC):
(-[WebCoreAVFMovieObserver initWithPlayer:]):
(-[WebCoreAVFMovieObserver disconnect]):
(-[WebCoreAVFMovieObserver metadataLoaded]):
(-[WebCoreAVFMovieObserver didEnd:]):
(-[WebCoreAVFMovieObserver observeValueForKeyPath:ofObject:change:context:]):
(-[WebCoreAVFMovieObserver legibleOutput:didOutputAttributedStrings:nativeSampleBuffers:forItemTime:]):
(-[WebCoreAVFMovieObserver outputSequenceWasFlushed:]):
(-[WebCoreAVFLoaderDelegate initWithPlayer:]):
(-[WebCoreAVFLoaderDelegate resourceLoader:shouldWaitForLoadingOfRequestedResource:]):
(-[WebCoreAVFLoaderDelegate resourceLoader:didCancelLoadingRequest:]):
(-[WebCoreAVFPullDelegate initWithPlayer:]):
(-[WebCoreAVFPullDelegate outputMediaDataWillChange:]):
(-[WebCoreAVFMovieObserver initWithCallback:]): Deleted.
(-[WebCoreAVFLoaderDelegate initWithCallback:]): Deleted.
(-[WebCoreAVFLoaderDelegate setCallback:]): Deleted.
(-[WebCoreAVFPullDelegate initWithCallback:]): Deleted.
(-[WebCoreAVFPullDelegate setCallback:]): Deleted.

2:10 AM Changeset in webkit [237377] by Claudio Saavedra
  • 2 edits in trunk/Source/WebKit

[WPE][GTK] Cleanups to the certificate encoder
https://bugs.webkit.org/show_bug.cgi?id=190865

Reviewed by Žan Doberšek.

  • Shared/soup/WebCoreArgumentCodersSoup.cpp:

(IPC::ArgumentCoder<CertificateInfo>::encode):

2:05 AM Changeset in webkit [237376] by jer.noble@apple.com
  • 11 edits
    2 adds in trunk

TextTrack cues should be updated more often than every 250ms.
https://bugs.webkit.org/show_bug.cgi?id=190827

Reviewed by Eric Carlson.

Source/WebCore:

Test: media/track/track-cue-timing.html

TextTracks cues are recalculated on the playback timer, which fires at least every 250ms.
In addition to this timer, add a method to MediaPlayer to provide a task which will be
performed at a particular media time, and use this new method to request cues be updated
at the next interesting media time. The next interesting time would be either when the
soonest current cue will end, or when the next non-current cue will start, whichever is
earlier.

(Determining the "next non-current cue" requires new API on PODIntervalTree, as that class
does not have iterators per-se.)

  • html/HTMLMediaElement.cpp:

(WebCore::compareCueIntervalEndTime):
(WebCore::HTMLMediaElement::updateActiveTextTrackCues):

  • platform/PODIntervalTree.h:
  • platform/graphics/MediaPlayer.cpp:

(WebCore::MediaPlayer::performTaskAtMediaTime):

  • platform/graphics/MediaPlayer.h:
  • platform/graphics/MediaPlayerPrivate.h:

(WebCore::MediaPlayerPrivateInterface::performTaskAtMediaTime):

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

(WebCore::MediaPlayerPrivateAVFoundationObjC::performTaskAtMediaTime):

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

(WebCore::MediaPlayerPrivateMediaSourceAVFObjC::performTaskAtMediaTime):

LayoutTests:

Add a LayoutTest which plays back 6s of captions, each 50 ms in duration, and uses
when the "enter" and "exit" were fired (in media time) to check whether they were missed
or not. The test succeeds if fewer than 50 of the 120 cues were missed.

  • media/track/track-cue-missing-expected.txt: Added.
  • media/track/track-cue-missing.html: Added.

Oct 23, 2018:

9:58 PM Changeset in webkit [237375] by Fujii Hironori
  • 2 edits in trunk/Source/WebCore

[Win] Assertion fails while destructing local static AtomicString of FontCache::lastResortFallbackFont
https://bugs.webkit.org/show_bug.cgi?id=190748

Reviewed by Myles C. Maxfield.

FontCache::lastResortFallbackFont had local static AtomicString
variables which are not capsuled by NeverDestroyed. Hence, those
variables are destructed on IPC thread when WebProcess exits.

Changed those AtomicString to NeverDestroyed<AtomicString> not to
be destructed on exit.

This change can't be tested automatically because WebKitTestRunner
doesn't support Windows port yet.

  • platform/graphics/win/FontCacheWin.cpp:

(WebCore::FontCache::lastResortFallbackFont): Changed
fallbackFonts's type from AtomicString array to
NeverDestroyed<AtomicString> array.

4:52 PM Changeset in webkit [237374] by Kocsen Chung
  • 2 edits in branches/safari-606-branch/Source/JavaScriptCore

Cherry-pick r236606. rdar://problem/45285669

Gardening: speculative build fix.
<rdar://problem/44869924>

Not reviewed.

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

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

4:52 PM Changeset in webkit [237373] by Kocsen Chung
  • 14 edits in branches/safari-606-branch

Cherry-pick r236589. rdar://problem/45285669

Verify the contents of AssemblerBuffer on arm64e
https://bugs.webkit.org/show_bug.cgi?id=190057
<rdar://problem/38916630>

Reviewed by Mark Lam.

JSTests:

  • stress/regress-189132.js:

Source/JavaScriptCore:

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

Source/WTF:

  • wtf/PtrTag.h: (WTF::tagInt):

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

4:52 PM Changeset in webkit [237372] by Kocsen Chung
  • 6 edits
    2 adds in branches/safari-606-branch

Cherry-pick r237238. rdar://problem/45363879

[Apple Pay] Increment the API version to 5
https://bugs.webkit.org/show_bug.cgi?id=190686
<rdar://problem/45348523>

Reviewed by Simon Fraser.

Source/WebCore:

Test: http/tests/ssl/applepay/ApplePaySessionV5.html

  • testing/MockPaymentCoordinator.cpp: (WebCore::MockPaymentCoordinator::supportsVersion):

Source/WebKit:

  • WebProcess/ApplePay/WebPaymentCoordinator.cpp: (WebKit::WebPaymentCoordinator::supportsVersion):

LayoutTests:

  • http/tests/ssl/applepay/ApplePaySessionV5-expected.txt: Added.
  • http/tests/ssl/applepay/ApplePaySessionV5.html: Added.
  • platform/mac-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237238 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4:52 PM Changeset in webkit [237371] by Kocsen Chung
  • 3 edits in branches/safari-606-branch/Source/JavaScriptCore

Cherry-pick r236604. rdar://problem/45285669

[JSC] [Armv7] Add a copy function argument to MacroAssemblerARMv7::link() and pass it down to the assembler's linking functions.
https://bugs.webkit.org/show_bug.cgi?id=190080

Reviewed by Mark Lam.

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

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

4:52 PM Changeset in webkit [237370] by Kocsen Chung
  • 5 edits
    2 adds in branches/safari-606-branch

Cherry-pick r236578. rdar://problem/45285613

Resource Load Statistics: Non-redirected top frame navigation should not get captured in statistics
https://bugs.webkit.org/show_bug.cgi?id=190055
<rdar://problem/44843460>

Reviewed by Chris Dumez.

Source/WebCore:

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

  • loader/ResourceLoadStatistics.cpp: (WebCore::ResourceLoadStatistics::decode):

Corrects legacy statistics for frames and triggers a re-classification.

Source/WebKit:

  • UIProcess/ResourceLoadStatisticsMemoryStore.cpp:

Bumped the statisticsModelVersion to 14 to be able to
correct legacy statistics.

(WebKit::ResourceLoadStatisticsMemoryStore::logFrameNavigation):

Now skips capture if it's the main frame.

LayoutTests:

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

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

4:51 PM Changeset in webkit [237369] by Kocsen Chung
  • 2 edits in branches/safari-606-branch/Source/WebKit

Cherry-pick r235674. rdar://problem/45285711

Cancelled fullscreen exit gesture leaves bad state for next exit request.
https://bugs.webkit.org/show_bug.cgi?id=189278

Patch by Jeremy Jones <jeremyj@apple.com> on 2018-09-05
Reviewed by Jer Noble.

  • UIProcess/ios/fullscreen/WKFullScreenWindowControllerIOS.mm: (-[WKFullScreenWindowController _dismissFullscreenViewController]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235674 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3:05 PM Changeset in webkit [237368] by Justin Fan
  • 50 edits
    8 copies
    63 moves
    7 deletes in trunk

[WebGPU] Rename old GPU* implementation files to GPULegacy* for WebMetal
https://bugs.webkit.org/show_bug.cgi?id=190817

Reviewed by Dean Jackson.

Source/WebCore:

No new tests. No change in behavior.

  • [Large list of refactored classes and renamed files]: GPU* is now GPULegacy*

Tools:

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebCore/mac/GPULegacyBuffer.mm: Renamed from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUBuffer.mm.

(TestWebKitAPI::TEST_F):

  • TestWebKitAPI/Tests/WebCore/mac/GPULegacyCommandQueue.mm: Copied from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUFunction.mm.

(TestWebKitAPI::TEST_F):

  • TestWebKitAPI/Tests/WebCore/mac/GPULegacyDevice.mm: Renamed from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUDevice.mm.

(TestWebKitAPI::TEST_F):

  • TestWebKitAPI/Tests/WebCore/mac/GPULegacyFunction.mm: Renamed from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUFunction.mm.

(TestWebKitAPI::TEST_F):

  • TestWebKitAPI/Tests/WebCore/mac/GPULegacyLibrary.mm: Renamed from Tools/TestWebKitAPI/Tests/WebCore/mac/GPULibrary.mm.

(TestWebKitAPI::TEST_F):

  • TestWebKitAPI/Tests/WebCore/mac/GPULegacyTest.h: Renamed from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUTest.h.
2:07 PM Changeset in webkit [237367] by Ryan Haddad
  • 21 edits in trunk

Unreviewed, rolling out r237261.

The layout test for this change crashes under GuardMalloc.

Reverted changeset:

"Handle MDNS resolution of candidates through libwebrtc
directly"
https://bugs.webkit.org/show_bug.cgi?id=190681
https://trac.webkit.org/changeset/237261

1:53 PM Changeset in webkit [237366] by Truitt Savell
  • 2 edits in trunk/Source/WebCore

Unreviewed, rolling out r237357.

API test is now failing on all platforms.

Reverted changeset:

"topPrivatelyControlledDomain is slow"
https://bugs.webkit.org/show_bug.cgi?id=190792
https://trac.webkit.org/changeset/237357

1:36 PM Changeset in webkit [237365] by ddkilzer@apple.com
  • 4 edits in trunk/Source

Fix false positive leaks when using custom -init methods that don't start with -init
<https://webkit.org/b/190818>
<rdar://problem/45476042>

Reviewed by Dan Bernstein.

Source/WebKit:

  • UIProcess/Cocoa/WKWebViewContentProvider.h:

(-[WKWebViewContentProvider web_initWithFrame:webView:mimeType:]):
Add attribute to make this behave like an -init method.

Source/WebKitLegacy/mac:

  • Misc/WebKitErrors.m: Move descriptions static global so that

it is accessible by both categories.
(-[NSError _webkit_initWithDomain:code:URL:]): Declare new
NSError (WebKitInternal) category. Add attribute to make this
behave like an -init method.
(-[NSError _webkit_initWithDomain:code:URL:]): Move into new
NSError (WebKitInternal) category. Modernize method. Replace
constant NSString with NSURLErrorFailingURLStringErrorKey.

1:02 PM Changeset in webkit [237364] by Truitt Savell
  • 5 edits in trunk/Source/WebCore

Unreviewed, rolling out r237350.

Caused several Crashes cross multiple tests and platforms.

Reverted changeset:

"Use WeakPtr and GenericTaskQueue within ObjC classes used by
MediaPlayerPrivateAVFoundationObjC"
https://bugs.webkit.org/show_bug.cgi?id=190790
https://trac.webkit.org/changeset/237350

12:44 PM Changeset in webkit [237363] by commit-queue@webkit.org
  • 1 edit
    1 delete in trunk/LayoutTests

Web Inspector: Remove unused css-rules pretty printing test expectation
https://bugs.webkit.org/show_bug.cgi?id=190834

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2018-10-23
Reviewed by Matt Baker.

  • inspector/codemirror/prettyprinting-css-rules-expected.txt: Removed.
12:29 PM Changeset in webkit [237362] by Alan Coon
  • 2 edits in tags/Safari-607.1.10.4/Source/WebKit

Cherry-pick r237309. rdar://problem/45377609

Allow WebContent process to check some file system features
https://bugs.webkit.org/show_bug.cgi?id=190768
<rdar://problem/45377609>

Reviewed by Alexey Proskuryakov.

This patch unblocks some IOKit properties that are needed by lower level frameworks to make decisions
about how to efficiently use the file system.

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237309 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11:30 AM Changeset in webkit [237361] by ap@apple.com
  • 2 edits in trunk/Source/WTF

Really prevent PLATFORM(IOS) from being used
https://bugs.webkit.org/show_bug.cgi?id=190813

Reviewed by Tim Horton.

  • wtf/Platform.h: This relies on -Wundef, which we have enabled.
11:23 AM Changeset in webkit [237360] by Alan Coon
  • 7 edits in tags/Safari-607.1.10.4/Source

Versioning.

11:17 AM Changeset in webkit [237359] by Ryan Haddad
  • 7 edits in trunk

Unreviewed, rolling out r237280.

Causes fast/box-shadow/box-shadow-with-zero-radius.html to
fail on iOS Simulator.

Reverted changeset:

"[CG] Adopt CG SPI for non-even cornered rounded rects"
https://bugs.webkit.org/show_bug.cgi?id=190155
https://trac.webkit.org/changeset/237280

11:15 AM Changeset in webkit [237358] by Alan Coon
  • 1 copy in tags/Safari-607.1.10.4

New tag.

11:02 AM Changeset in webkit [237357] by Antti Koivisto
  • 2 edits in trunk/Source/WebCore

topPrivatelyControlledDomain is slow
https://bugs.webkit.org/show_bug.cgi?id=190792

Reviewed by Alex Christensen.

It calls into some slowish CFNetwork code and ends up showing up in profiles.

  • platform/mac/PublicSuffixMac.mm:

(WebCore::topPrivatelyControlledDomain):

Add a cache that avoids calls into frameworks.

10:54 AM Changeset in webkit [237356] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

[macOS Debug WK1] Layout Test inspector/codemirror/prettyprinting-css.html is a flaky timeout
https://bugs.webkit.org/show_bug.cgi?id=190833

Unreviewed test gardening.

  • platform/mac-wk1/TestExpectations: Mark test as slow.
10:54 AM Changeset in webkit [237355] by Chris Dumez
  • 25 edits in trunk

[PSON] Add support for cross-site client-side redirects
https://bugs.webkit.org/show_bug.cgi?id=190806
<rdar://problem/45047344>

Reviewed by Geoffrey Garen.

Source/WebCore:

Add support for cross-site client-side redirects so that it swaps process and so that the back
forward list looks as expected. To support this, the following changes had to be done:

  • The NavigationAction now provides additional information so that the WebProcess can communicate things about the client-side redirect to the UIProcess: lockHistory / lockBackForwardList and clientRedirectSourceForHistory.
  • If the UIProcess decides to process-swap on a client-side redirect, we pass the client-side redirect information to the new WebContent process via LoadRequest struct. WebCore then takes care of setting things up using this information so that it recognizes that it is continuing a load that is a client side redirect.
  • We also need to pass the current BackForwardListItem / HistoryItem to the new WebContent process so that the new process can truly lock history and keep updating the *current* HistoryItem, instead of creating a new HistoryItem.
  • After a process swap, when we re-construct the WebFrameProxy for the main frame in the new process, we now set the frame's URL in the UIProcess to the URL it had before we swapped. Clients such as Safari, rely on the main frame's URL being the expected value (the last committed load URL) until the next load is committed when receiving didPerformRedirect calls. Because we are destroying the main frame on process-swapping, we were losing the last committed URL and Safari would hit assertions.

With this model, the willPerformClientRedirect IPC is still sent from the previous WebProcess
and the didPerformClientRedirect IPC is now sent by the new WebProcess. No change should be
observable from the client's point of view.

  • loader/FrameLoadRequest.h:

(WebCore::FrameLoadRequest::setLockHistory):
(WebCore::FrameLoadRequest::setlockBackForwardList):
(WebCore::FrameLoadRequest::clientRedirectSourceForHistory const):
(WebCore::FrameLoadRequest::setClientRedirectSourceForHistory):

  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::loadURL):
(WebCore::FrameLoader::load):

  • loader/HistoryController.cpp:

(WebCore::HistoryController::updateForRedirectWithLockedBackForwardList):

  • loader/HistoryController.h:
  • loader/NavigationAction.h:

(WebCore::NavigationAction::lockHistory const):
(WebCore::NavigationAction::setLockHistory):
(WebCore::NavigationAction::lockBackForwardList const):
(WebCore::NavigationAction::setLockBackForwardList):

Source/WebKit:

  • Shared/LoadParameters.cpp:

(WebKit::LoadParameters::encode const):
(WebKit::LoadParameters::decode):

  • Shared/LoadParameters.h:
  • Shared/NavigationActionData.cpp:

(WebKit::NavigationActionData::encode const):
(WebKit::NavigationActionData::decode):

  • Shared/NavigationActionData.h:
  • Shared/WebBackForwardListItem.cpp:

(WebKit::WebBackForwardListItem::setSuspendedPage):

  • UIProcess/API/APINavigation.h:

(API::Navigation::setLockHistory):
(API::Navigation::lockHistory const):
(API::Navigation::setLockBackForwardList):
(API::Navigation::lockBackForwardList const):
(API::Navigation::setClientRedirectSourceForHistory):
(API::Navigation::clientRedirectSourceForHistory const):

  • UIProcess/API/APINavigationClient.h:

(API::NavigationClient::willPerformClientRedirect):
(API::NavigationClient::didPerformClientRedirect):

  • UIProcess/API/Cocoa/WKNavigationDelegatePrivate.h:
  • UIProcess/Cocoa/NavigationState.h:
  • UIProcess/Cocoa/NavigationState.mm:

(WebKit::NavigationState::setNavigationDelegate):
(WebKit::NavigationState::NavigationClient::didPerformClientRedirect):

  • UIProcess/FrameLoadState.h:

(WebKit::FrameLoadState::setURL):

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::loadRequestWithNavigation):
(WebKit::WebPageProxy::continueNavigationInNewProcess):
(WebKit::WebPageProxy::decidePolicyForNavigationAction):
(WebKit::WebPageProxy::didPerformClientRedirect):

  • WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:

(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction):

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::loadRequest):
(WebKit::WebPage::setCurrentHistoryItemForReattach):

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

Tools:

Add API test coverage.

  • TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:

(-[PSONNavigationDelegate _webView:willPerformClientRedirectToURL:delay:]):
(-[PSONNavigationDelegate _webView:didPerformClientRedirectFromURL:toURL:]):

10:26 AM Changeset in webkit [237354] by Truitt Savell
  • 2 edits in trunk/LayoutTests

[ MacOS Debug WK1 ] Layout Test fast/forms/state-restore-to-non-edited-controls.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=190832

Unreviewed Test Gardening.

  • platform/mac-wk1/TestExpectations:
10:22 AM Changeset in webkit [237353] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

Layout Test imported/w3c/web-platform-tests/html/semantics/embedded-content/media-elements/ready-states/autoplay-with-slow-text-tracks.html is flaky.
https://bugs.webkit.org/show_bug.cgi?id=176929

Unreviewed test gardening.

  • platform/mac/TestExpectations: Mark test as flaky.
9:58 AM Changeset in webkit [237352] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

REGRESSION (r236501?): Layout Test compositing/geometry/limit-layer-bounds-opacity-transition.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=190210

Unreviewed test gardening.

  • platform/mac/TestExpectations: Mark test as flaky.
8:52 AM Changeset in webkit [237351] by Claudio Saavedra
  • 4 edits in trunk

[WPE][GTK] Pass full certificate chain in CertificateInfo coder
https://bugs.webkit.org/show_bug.cgi?id=190789

Reviewed by Michael Catanzaro.

When the network process serializes certificate data to other
processes through the argument coders, the certificate chain, if
present, is lost. In practice this means that applications using
the public API to process certificate info have no details on the
certificate chain, other than the very basics included in the
certificate. Serialize the entire chain if available in the
certificate.

Source/WebKit:

  • Shared/soup/WebCoreArgumentCodersSoup.cpp:

(IPC::ArgumentCoder<CertificateInfo>::encode): Encode the
certificate chain if present.
(IPC::ArgumentCoder<CertificateInfo>::decode): Decode the
entire certificate chain and rebuild it.

Tools:

  • TestWebKitAPI/Tests/WebKitGLib/TestSSL.cpp:

(testSSL): Test that the self-signed certificate has no bogus
issuer certificate.

7:40 AM Changeset in webkit [237350] by jer.noble@apple.com
  • 5 edits in trunk/Source/WebCore

Use WeakPtr and GenericTaskQueue within ObjC classes used by MediaPlayerPrivateAVFoundationObjC
https://bugs.webkit.org/show_bug.cgi?id=190790

Reviewed by Alex Christensen.

Move towards using WeakPtr callbacks instead of raw pointers within the utility objects used by
MediaPlayerPrivateAVFoundationObjC. Additionally, accessing WeakPtr off the thread which created
the pointer is not allowed, so use a GenericTaskQueue to schedule callbacks instead. Make
GenericTaskQueue<Timer> thread-safe by locking around access to m_pendingTasks, and by making
incrementing the pending task count atomic.

  • platform/GenericTaskQueue.cpp:

(WebCore::TaskDispatcher<Timer>::postTask):
(WebCore::TaskDispatcher<Timer>::sharedLock):
(WebCore::TaskDispatcher<Timer>::sharedTimerFired):
(WebCore::TaskDispatcher<Timer>::dispatchOneTask):
(WebCore::TaskDispatcher<Timer>::pendingDispatchers):

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

(WebCore::MediaPlayerPrivateAVFoundationObjC::MediaPlayerPrivateAVFoundationObjC):
(WebCore::MediaPlayerPrivateAVFoundationObjC::~MediaPlayerPrivateAVFoundationObjC):
(-[WebCoreAVFMovieObserver initWithPlayer:]):
(-[WebCoreAVFMovieObserver disconnect]):
(-[WebCoreAVFMovieObserver metadataLoaded]):
(-[WebCoreAVFMovieObserver didEnd:]):
(-[WebCoreAVFMovieObserver observeValueForKeyPath:ofObject:change:context:]):
(-[WebCoreAVFMovieObserver legibleOutput:didOutputAttributedStrings:nativeSampleBuffers:forItemTime:]):
(-[WebCoreAVFMovieObserver outputSequenceWasFlushed:]):
(-[WebCoreAVFLoaderDelegate initWithPlayer:]):
(-[WebCoreAVFLoaderDelegate resourceLoader:shouldWaitForLoadingOfRequestedResource:]):
(-[WebCoreAVFLoaderDelegate resourceLoader:didCancelLoadingRequest:]):
(-[WebCoreAVFPullDelegate initWithPlayer:]):
(-[WebCoreAVFPullDelegate outputMediaDataWillChange:]):
(-[WebCoreAVFMovieObserver initWithCallback:]): Deleted.
(-[WebCoreAVFLoaderDelegate initWithCallback:]): Deleted.
(-[WebCoreAVFLoaderDelegate setCallback:]): Deleted.
(-[WebCoreAVFPullDelegate initWithCallback:]): Deleted.
(-[WebCoreAVFPullDelegate setCallback:]): Deleted.

2:52 AM Changeset in webkit [237349] by Devin Rousso
  • 2 edits in trunk/Source/WebInspectorUI

Uncaught Exception: TypeError: null is not an object (evaluating 'mouseBlock.addEventListener')
https://bugs.webkit.org/show_bug.cgi?id=190766

Reviewed by Brian Burg.

  • UserInterface/Views/NetworkTableContentView.js:

(WI.NetworkTableContentView.prototype._populateWaterfallGraph):
When creating the mouseBlock, if the startTimestamp and endTimestamp are equal, we
don't create an element. If this is true for the total range of the WI.Resouce (meaning
its startTime and responseEnd), don't even try to create blocks.

2:50 AM Changeset in webkit [237348] by Devin Rousso
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: "Queued NaNms" in Network tab
https://bugs.webkit.org/show_bug.cgi?id=190767
<rdar://problem/45420859>

Reviewed by Brian Burg.

  • UserInterface/Models/ResourceTimingData.js:

(WI.ResourceTimingData.prototype.get fetchStart):
If the WI.ResourceTimingData has no fetchStart use the associated WI.Resource's
requestSentTimestamp (just like startTime).

Oct 22, 2018:

11:10 PM Changeset in webkit [237347] by commit-queue@webkit.org
  • 34 edits
    6 adds in trunk

Registered custom properties should support syntax parameter for <length> and *
https://bugs.webkit.org/show_bug.cgi?id=190039

Patch by Justin Michaud <justin_michaud@apple.com> on 2018-10-22
Reviewed by Antti Koivisto.

LayoutTests/imported/w3c:

Update WPT test results to fail in a new way.

  • web-platform-tests/css/css-properties-values-api/register-property-syntax-parsing-expected.txt:
  • web-platform-tests/css/css-properties-values-api/registered-properties-inheritance-expected.txt:
  • web-platform-tests/css/css-properties-values-api/registered-property-cssom-expected.txt:
  • web-platform-tests/css/css-properties-values-api/typedom.tentative-expected.txt:
  • web-platform-tests/css/css-properties-values-api/var-reference-registered-properties-cycles-expected.txt:
  • web-platform-tests/css/css-properties-values-api/var-reference-registered-properties-expected.txt:

Source/WebCore:

Refactor code so that:

  • All properties applied in StyleResolver::applyMatchedProperties are only applied once.
  • Custom properties are only resolved once, in StyleResolver, when they are applied to the RenderStyle. They were previously resolved every time they were referenced, and again in RenderStyle.
  • The font-size property is applied after its variable references, but before custom properties that depend on it.
  • Cycles are detected at the same time as resolution.
  • MutableStyleProperties' custom properties cannot be set from Javascript or WebKitLegacy if they do not parse for the property's type. If they contain var(--...) references, however, then they can be set because we cannot check if the references are valid from setProperty. This behaviour matches chrome, but is not documented in the spec.
  • Custom property values have more explicit resolved/unresolved state.
  • RenderStyle only ever holds resolved custom properties, and StyleResolver::CascadedProperties only holds unresolved properties.

Tests: css-custom-properties-api/crash.html

css-custom-properties-api/cycles.html
css-custom-properties-api/inline.html

  • css/CSSComputedStyleDeclaration.cpp:

(WebCore::ComputedStyleExtractor::customPropertyValue):

  • css/CSSCustomPropertyValue.cpp:

(WebCore::CSSCustomPropertyValue::equals const):
(WebCore::CSSCustomPropertyValue::customCSSText const):
(WebCore::CSSCustomPropertyValue::tokens const):
(WebCore::CSSCustomPropertyValue::checkVariablesForCycles const): Deleted.
(WebCore::CSSCustomPropertyValue::resolveVariableReferences const): Deleted.
(WebCore::CSSCustomPropertyValue::setResolvedTypedValue): Deleted.

  • css/CSSCustomPropertyValue.h:
  • css/CSSRegisteredCustomProperty.cpp:

(WebCore::CSSRegisteredCustomProperty::CSSRegisteredCustomProperty):

  • css/CSSRegisteredCustomProperty.h:
  • css/CSSStyleSheet.h:
  • css/CSSVariableData.cpp:

(WebCore::CSSVariableData::CSSVariableData):
(WebCore::CSSVariableData::consumeAndUpdateTokens): Deleted.
(WebCore::CSSVariableData::checkVariablesForCycles const): Deleted.
(WebCore::CSSVariableData::checkVariablesForCyclesWithRange const): Deleted.
(WebCore::CSSVariableData::resolveVariableFallback const): Deleted.
(WebCore::CSSVariableData::resolveVariableReference const): Deleted.
(WebCore::CSSVariableData::resolveVariableReferences const): Deleted.
(WebCore::CSSVariableData::resolveTokenRange const): Deleted.

  • css/CSSVariableData.h:

(WebCore::CSSVariableData::create):
(WebCore::CSSVariableData::createResolved): Deleted.
(WebCore::CSSVariableData::needsVariableResolution const): Deleted.
(WebCore::CSSVariableData::CSSVariableData): Deleted.

  • css/CSSVariableReferenceValue.cpp:

(WebCore::resolveVariableFallback):
(WebCore::resolveVariableReference):
(WebCore::resolveTokenRange):
(WebCore::CSSVariableReferenceValue::resolveVariableReferences const):
(WebCore::CSSVariableReferenceValue::checkVariablesForCycles const): Deleted.

  • css/CSSVariableReferenceValue.h:

(WebCore::CSSVariableReferenceValue::create):
(WebCore::CSSVariableReferenceValue::equals const):
(WebCore::CSSVariableReferenceValue::variableDataValue const): Deleted.

  • css/DOMCSSRegisterCustomProperty.cpp:

(WebCore::DOMCSSRegisterCustomProperty::registerProperty):

  • css/PropertySetCSSStyleDeclaration.cpp:

(WebCore::PropertySetCSSStyleDeclaration::setProperty):

  • css/StyleBuilderCustom.h:

(WebCore::StyleBuilderCustom::applyInitialCustomProperty):
(WebCore::StyleBuilderCustom::applyValueCustomProperty):

  • css/StyleProperties.cpp:

(WebCore::MutableStyleProperties::setCustomProperty):

  • css/StyleProperties.h:
  • css/StyleResolver.cpp:

(WebCore::StyleResolver::State::setStyle):
(WebCore::StyleResolver::styleForKeyframe):
(WebCore::StyleResolver::styleForPage):
(WebCore::StyleResolver::applyMatchedProperties):
(WebCore::StyleResolver::applyPropertyToCurrentStyle):
(WebCore::StyleResolver::applyProperty):
(WebCore::StyleResolver::resolvedVariableValue const):
(WebCore::StyleResolver::CascadedProperties::applyDeferredProperties):
(WebCore::StyleResolver::CascadedProperties::Property::apply):
(WebCore::StyleResolver::applyCascadedCustomProperty):
(WebCore::StyleResolver::applyCascadedProperties):

  • css/StyleResolver.h:
  • css/parser/CSSParser.cpp:

(WebCore::CSSParser::parseValueWithVariableReferences):

  • css/parser/CSSParser.h:
  • css/parser/CSSPropertyParser.cpp:

(WebCore::CSSPropertyParser::CSSPropertyParser):
(WebCore::CSSPropertyParser::canParseTypedCustomPropertyValue):
(WebCore::CSSPropertyParser::parseTypedCustomPropertyValue):
(WebCore::CSSPropertyParser::collectParsedCustomPropertyValueDependencies):
(WebCore::CSSPropertyParser::parseValueStart):
(WebCore::CSSPropertyParser::parseSingleValue):

  • css/parser/CSSPropertyParser.h:
  • css/parser/CSSVariableParser.cpp:

(WebCore::CSSVariableParser::parseDeclarationValue):

  • dom/ConstantPropertyMap.cpp:

(WebCore::ConstantPropertyMap::setValueForProperty):
(WebCore::variableDataForPositivePixelLength):
(WebCore::variableDataForPositiveDuration):

  • rendering/style/RenderStyle.cpp:

(WebCore::RenderStyle::checkVariablesInCustomProperties): Deleted.

  • rendering/style/RenderStyle.h:

(WebCore::RenderStyle::setInheritedCustomPropertyValue):
(WebCore::RenderStyle::setNonInheritedCustomPropertyValue):

  • rendering/style/StyleCustomPropertyData.h:

(WebCore::StyleCustomPropertyData::operator== const):
(WebCore::StyleCustomPropertyData::setCustomPropertyValue):
(WebCore::StyleCustomPropertyData::StyleCustomPropertyData):
(): Deleted.

LayoutTests:

Add tests for inline styles, font-size cycles with custom properties, and a crash that was reported.

  • css-custom-properties-api/crash-expected.txt: Added.
  • css-custom-properties-api/crash.html: Added.
  • css-custom-properties-api/cycles-expected.txt: Added.
  • css-custom-properties-api/cycles.html: Added.
  • css-custom-properties-api/inline-expected.txt: Added.
  • css-custom-properties-api/inline.html: Added.
6:52 PM Changeset in webkit [237346] by timothy_horton@apple.com
  • 2 edits in trunk/Tools

REGRESSION (r237331): InteractionDeadlockAfterCrash API test fails
https://bugs.webkit.org/show_bug.cgi?id=190801
<rdar://problem/43674361>

  • TestWebKitAPI/Tests/WebKitCocoa/InteractionDeadlockAfterCrash.mm:

(TEST):
Use a parented WKWebView for this test, otherwise none of the assertions
about gesture recognizers work anymore.

5:52 PM Changeset in webkit [237345] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

REGRESSION: [iOS] Layout Test media/media-fullscreen-pause-inline.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=187618

Unreviewed test gardening.

  • platform/ios/TestExpectations: Mark test as flaky.
5:46 PM Changeset in webkit [237344] by commit-queue@webkit.org
  • 21 edits
    5 copies
    1 add in trunk

CSS Paint API should give a 2d rendering context
https://bugs.webkit.org/show_bug.cgi?id=190762

Patch by Justin Michaud <justin_michaud@apple.com> on 2018-10-22
Reviewed by Dean Jackson.

Source/WebCore:

Add a new type of canvas and 2d rendering context to support the CSS Painting API.
Make many of the methods from HTMLCanvasElement virtual functions on CanvasBase, and
remove many of the downcasts in CanvasRenderingContext2DBase as a result.

  • CMakeLists.txt:
  • DerivedSources.make:
  • Sources.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSPaintRenderingContext2DCustom.cpp: Copied from Source/WebCore/css/CSSPaintCallback.h.

(WebCore::root):
(WebCore::JSPaintRenderingContext2DOwner::isReachableFromOpaqueRoots):
(WebCore::JSPaintRenderingContext2D::visitAdditionalChildren):

  • bindings/js/WebCoreBuiltinNames.h:
  • css/CSSPaintCallback.h:
  • css/CSSPaintCallback.idl:
  • html/CanvasBase.cpp:

(WebCore::CanvasBase::~CanvasBase):

  • html/CanvasBase.h:

(WebCore::CanvasBase::isCustomPaintCanvas const):

  • html/CustomPaintCanvas.cpp: Added.

(WebCore::CustomPaintCanvas::create):
(WebCore::CustomPaintCanvas::CustomPaintCanvas):
(WebCore::CustomPaintCanvas::~CustomPaintCanvas):
(WebCore::CustomPaintCanvas::width const):
(WebCore::CustomPaintCanvas::setWidth):
(WebCore::CustomPaintCanvas::height const):
(WebCore::CustomPaintCanvas::setHeight):
(WebCore::CustomPaintCanvas::size const):
(WebCore::CustomPaintCanvas::setSize):
(WebCore::CustomPaintCanvas::getContext):
(WebCore::CustomPaintCanvas::copiedImage const):
(WebCore::CustomPaintCanvas::drawingContext const):
(WebCore::CustomPaintCanvas::existingDrawingContext const):
(WebCore::CustomPaintCanvas::makeRenderingResultsAvailable):

  • html/CustomPaintCanvas.h: Copied from Source/WebCore/html/OffscreenCanvas.h.
  • html/HTMLCanvasElement.h:
  • html/OffscreenCanvas.h:
  • html/canvas/CanvasRenderingContext.cpp:

(WebCore::CanvasRenderingContext::wouldTaintOrigin):

  • html/canvas/CanvasRenderingContext.h:

(WebCore::CanvasRenderingContext::isPaint const):

  • html/canvas/CanvasRenderingContext2DBase.cpp:

(WebCore::DisplayListDrawingContext::DisplayListDrawingContext):
(WebCore::CanvasRenderingContext2DBase::unwindStateStack):
(WebCore::CanvasRenderingContext2DBase::isAccelerated const):
(WebCore::CanvasRenderingContext2DBase::setStrokeStyle):
(WebCore::CanvasRenderingContext2DBase::setFillStyle):
(WebCore::CanvasRenderingContext2DBase::resetTransform):
(WebCore::CanvasRenderingContext2DBase::clearCanvas):
(WebCore::CanvasRenderingContext2DBase::transformAreaToDevice const):
(WebCore::CanvasRenderingContext2DBase::rectContainsCanvas const):
(WebCore::CanvasRenderingContext2DBase::calculateCompositingBufferRect):
(WebCore::CanvasRenderingContext2DBase::compositeBuffer):
(WebCore::CanvasRenderingContext2DBase::createPattern):
(WebCore::CanvasRenderingContext2DBase::didDrawEntireCanvas):
(WebCore::CanvasRenderingContext2DBase::didDraw):
(WebCore::CanvasRenderingContext2DBase::paintRenderingResultsToCanvas):
(WebCore::CanvasRenderingContext2DBase::drawingContext const):

  • html/canvas/CanvasRenderingContext2DBase.h:
  • html/canvas/PaintRenderingContext2D.cpp: Copied from Source/WebCore/css/CSSPaintCallback.h.

(WebCore::PaintRenderingContext2D::create):
(WebCore::PaintRenderingContext2D::PaintRenderingContext2D):

  • html/canvas/PaintRenderingContext2D.h: Copied from Source/WebCore/css/CSSPaintCallback.h.
  • html/canvas/PaintRenderingContext2D.idl: Copied from Source/WebCore/css/CSSPaintCallback.idl.
  • html/canvas/WebMetalRenderPassAttachmentDescriptor.h:
  • platform/graphics/CustomPaintImage.cpp:

(WebCore::CustomPaintImage::doCustomPaint):

LayoutTests:

  • fast/css-custom-paint/basic.html:
5:35 PM Changeset in webkit [237343] by Alan Coon
  • 1 copy in tags/Safari-606.3.1

Tag Safari-606.3.1.

5:21 PM Changeset in webkit [237342] by Ryan Haddad
  • 3 edits in trunk/LayoutTests

Unreviewed, update TestExpectations for imported/w3c/web-platform-tests/fetch/nosniff/importscripts.html.
https://bugs.webkit.org/show_bug.cgi?id=157068

  • TestExpectations: Mark test as a flaky failure for release builds.
  • platform/mac-wk1/TestExpectations: Remove duplicate expectation.
5:15 PM Changeset in webkit [237341] by jiewen_tan@apple.com
  • 5 edits in trunk/LayoutTests/imported/w3c

Update web-platform-tests/resource-timing
https://bugs.webkit.org/show_bug.cgi?id=190550

Reviewed by Youenn Fablet.

  • resources/import-expectations.json:
  • web-platform-tests/resource-timing/resource_initiator_types-expected.txt:
  • web-platform-tests/resource-timing/resource_subframe_self_navigation-expected.txt:
  • web-platform-tests/resource-timing/resource_timing_buffer_full_eventually.html:
  • web-platform-tests/resource-timing/resources/TAOResponse.py:

(main):

5:02 PM WebKitGTK/2.22.x edited by Michael Catanzaro
Propose more backports (diff)
4:58 PM Changeset in webkit [237340] by Keith Rollin
  • 6 edits in trunk/Source

Use Location = "Relative to Build Products" rather than "Relative to Group"
https://bugs.webkit.org/show_bug.cgi?id=190781

Reviewed by Alexey Proskuryakov.

Almost all Derived Files are included in Xcode projects with the
Location attribute set to "Relative to Group". While this currently
works, the Derived Files can no longer be found when enabling XCBuild
(which has stricter requirements). Fix this by setting the Location
attribute to "Relative to Build Products".

Source/JavaScriptCore:

Source/WebCore:

No new tests -- no changed functionality.

  • WebCore.xcodeproj/project.pbxproj:

Source/WebKit:

  • WebKit.xcodeproj/project.pbxproj:
4:56 PM Changeset in webkit [237339] by realdawei@apple.com
  • 2 edits in trunk/LayoutTests

REGRESSION (r234330): [mac-wk1] Layout Test fast/repaint/animation-after-layer-scroll.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=188421

Unreviewed test gardening.

Patch by Dawei Fenton <realdawei@apple.com> on 2018-10-22

  • platform/mac-wk1/TestExpectations: Mark test as flaky.
4:17 PM Changeset in webkit [237338] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

[ Mojave WK1 ] Layout Test storage/indexeddb/database-odd-names.html is failing
https://bugs.webkit.org/show_bug.cgi?id=190350

Unreviewed test gardening.

  • platform/mac-wk1/TestExpectations: Mark test as failing.
4:05 PM Changeset in webkit [237337] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

[macOS WK1] Layout Test http/tests/security/cross-origin-xsl-redirect-BLOCKED.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=189723

Unreviewed test gardening.

  • platform/mac-wk1/TestExpectations: Mark test as flaky.
4:05 PM Changeset in webkit [237336] by Ryan Haddad
  • 2 edits in trunk/Tools

REGRESSION (r234081): TestWebKitAPI.VideoControlsManager.VideoControlsManagerAudioElementFollowingUserInteraction is a flaky timeout
https://bugs.webkit.org/show_bug.cgi?id=187972

Unreviewed test gardening.

  • TestWebKitAPI/Tests/WebKitCocoa/VideoControlsManager.mm:

(TestWebKitAPI::TEST): Disable the flaky test.

3:12 PM Changeset in webkit [237335] by Alan Coon
  • 1 copy in tags/Safari-606.2.104.1.2

Tag Safari-606.2.104.1.2.

3:02 PM Changeset in webkit [237334] by ajuma@chromium.org
  • 2 edits in trunk/LayoutTests

Layout Test imported/w3c/web-platform-tests/intersection-observer/containing-block.html is a flaky failure on Debug builds
https://bugs.webkit.org/show_bug.cgi?id=190808

Unreviewed test gardening.

3:02 PM Changeset in webkit [237333] by realdawei@apple.com
  • 2 edits in trunk/LayoutTests

Some WK1 repaint tests are flaky on Mojave
https://bugs.webkit.org/show_bug.cgi?id=190627

Unreviewed, marked tests as flaky.

Patch by Dawei Fenton <realdawei@apple.com> on 2018-10-22

  • platform/mac-wk1/TestExpectations:
2:52 PM Changeset in webkit [237332] by Kocsen Chung
  • 7 edits in branches/safari-606.2.104.1-branch/Source

Versioning.

2:22 PM Changeset in webkit [237331] by timothy_horton@apple.com
  • 4 edits in trunk/Source/WebKit

Don't waste time under -setupInteraction under -initWithFrame for unparented WKWebViews
https://bugs.webkit.org/show_bug.cgi?id=190801
<rdar://problem/43674361>

Reviewed by Megan Gardner.

  • UIProcess/ios/WKContentView.mm:

(-[WKContentView _commonInitializationWithProcessPool:configuration:]):
(-[WKContentView didMoveToWindow]):
Defer the first call to WKContentViewInteraction's -setupInteraction
until the view is parented. This avoids a few milliseconds of unnecessary
work for views that are never parented.

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

(-[WKContentView setupInteraction]):
(-[WKContentView cleanupInteraction]):
Keep track of the current state of WKContentViewInteraction's gestures.
Use this to make it OK to call -setupInteraction multiple times.

2:05 PM Changeset in webkit [237330] by sihui_liu@apple.com
  • 2 edits in trunk/Source/WebKitLegacy

Regression (r232410): StorageTracker.db file gets unlinked while in use
https://bugs.webkit.org/show_bug.cgi?id=190795

Reviewed by Chris Dumez.

WK2 stopped using StorageTracker.db file in r232410 and would delete
the file for safety.
It turned out WK1 could use the same file path, so WK2 may delete the
file while WK1 is using it.

  • Storage/StorageTracker.cpp:

(WebKit::StorageTracker::trackerDatabasePath):

2:05 PM Changeset in webkit [237329] by Chris Dumez
  • 5 edits in trunk/Source

Deque's contains() and findIf() should be const
https://bugs.webkit.org/show_bug.cgi?id=190796

Reviewed by Antti Koivisto.

Source/WebKit:

Mark method as const now that Deque's implementation allows it to be.

  • UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::hasSuspendedPageProxyFor const):
(WebKit::WebProcessPool::hasSuspendedPageProxyFor): Deleted.

  • UIProcess/WebProcessPool.h:

Source/WTF:

Deque's contains() and findIf() should be const as they do not modify the container.

  • wtf/Deque.h:

(WTF::inlineCapacity>::findIf):
(WTF::inlineCapacity>::findIf const):
(WTF::inlineCapacity>::contains const):
(WTF::inlineCapacity>::contains): Deleted.

1:12 PM Changeset in webkit [237328] by wenson_hsieh@apple.com
  • 9 edits in trunk

[iOS] [Datalist] fast/forms/datalist/datalist-show-hide.html fails
https://bugs.webkit.org/show_bug.cgi?id=190777

Reviewed by Tim Horton.

Tools:

Implement isShowingDataListSuggestions on iOS. See below for more details.

  • DumpRenderTree/ios/UIScriptControllerIOS.mm:

(WTR::UIScriptController::isShowingDataListSuggestions const):

Add a stub implementation for DumpRenderTree.

  • TestRunnerShared/UIScriptContext/UIScriptController.cpp:

(WTR::UIScriptController::isShowingDataListSuggestions const):

Move this stub implementation to !PLATFORM(COCOA).

  • WebKitTestRunner/ios/UIScriptControllerIOS.mm:

(WTR::forEachViewInHierarchy):

Add a helper function to apply a given function to a UIView and each of its subviews, recursively.

(WTR::UIScriptController::isShowingDataListSuggestions const):

For iOS, return whether the UIRemoteKeyboardWindow contains a subview of type WKDataListSuggestionsPickerView.

LayoutTests:

Refactor this test such that it passes on both iOS and macOS. This test verifies that datalist suggestions menu
UI can be shown and hidden. On macOS, we focus and then blur the input field; on iOS, we tap in the datalist
button to show the suggestions UI, and then tap in the text field to bring back the regular keyboard.

  • fast/forms/datalist/datalist-show-hide-expected.txt:
  • fast/forms/datalist/datalist-show-hide.html:
  • platform/ios/TestExpectations:

Unskip the layout test on iOS.

  • resources/ui-helper.js:

(window.UIHelper.isShowingDataListSuggestions):

Tweak this helper function to resolve with either true or false (Boolean types), instead of the strings "true"
and "false".

12:24 PM Changeset in webkit [237327] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Open Quickly dialog doesn't show named scripts that appear in the debugger sidebar
https://bugs.webkit.org/show_bug.cgi?id=190649

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2018-10-22
Reviewed by Devin Rousso.

  • UserInterface/Views/OpenResourceDialog.js:

(WI.OpenResourceDialog.prototype.didPresentDialog):
(WI.OpenResourceDialog.prototype._addResourcesForTarget):
(WI.OpenResourceDialog.prototype._addScriptsForTarget):
Include non-resource named scripts from the main target in
the open quickly dialog.

12:07 PM Changeset in webkit [237326] by mark.lam@apple.com
  • 1 edit
    1 add in trunk/JSTests

DFGAbstractValue::m_arrayModes expects IndexingMode values, not IndexingType.
https://bugs.webkit.org/show_bug.cgi?id=190515
<rdar://problem/45222379>

Rubber-stamped by Saam Barati.

Adding another test.

  • stress/regress-190515-2.js: Added.
10:18 AM Changeset in webkit [237325] by mark.lam@apple.com
  • 7 edits
    1 add in trunk

DFGAbstractValue::m_arrayModes expects IndexingMode values, not IndexingType.
https://bugs.webkit.org/show_bug.cgi?id=190515
<rdar://problem/45222379>

Reviewed by Saam Barati.

JSTests:

  • stress/regress-190515.js: Added.

Source/JavaScriptCore:

  1. Fixes calls to asArrayModes() to take a structure's IndexingMode instead of IndexingType.
  1. DFG's compileNewArrayBuffer()'s HaveABadTime case was previously using the node's indexingType (instead of indexingMode) to choose the array structure to use for creating an array buffer with. This turns out to not be an issue because when the VM is in having a bad time, all the arrayStructureForIndexingTypeDuringAllocation structure pointers will point to the SlowPutArrayStorage structure anyway. However, to be strictly correct, we'll fix it to use the structure for the node's indexingMode.
  • dfg/DFGAbstractValue.cpp:

(JSC::DFG::AbstractValue::set):
(JSC::DFG::AbstractValue::mergeOSREntryValue):

  • dfg/DFGAbstractValue.h:

(JSC::DFG::AbstractValue::validate const):

  • dfg/DFGOSRExit.cpp:

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

  • dfg/DFGRegisteredStructureSet.cpp:

(JSC::DFG::RegisteredStructureSet::arrayModesFromStructures const):

  • dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::compileNewArrayBuffer):

9:27 AM Changeset in webkit [237324] by Alan Bujtas
  • 5 edits in trunk/Source/WebCore

[LFC][IFC] Add justify text-align support.
https://bugs.webkit.org/show_bug.cgi?id=190779

Reviewed by Antti Koivisto.

Collect expansion opportunities and adjust runs accordingly.

  • layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::layoutInlineContent const):

  • layout/inlineformatting/InlineFormattingContext.h:
  • layout/inlineformatting/InlineRun.h:

(WebCore::Layout::InlineRun::expansionOpportunity):
(WebCore::Layout::InlineRun::TextContext::setStart):
(WebCore::Layout::InlineRun::TextContext::setLength):
(WebCore::Layout::InlineRun::setTextContext):
(WebCore::Layout::InlineRun::createRun): Deleted.
(WebCore::Layout::InlineRun::createTextRun): Deleted.

  • layout/inlineformatting/Line.cpp:

(WebCore::Layout::InlineFormattingContext::Line::Line):
(WebCore::Layout::InlineFormattingContext::Line::init):
(WebCore::Layout::InlineFormattingContext::Line::computeExpansionOpportunities):
(WebCore::Layout::InlineFormattingContext::Line::appendContent):
(WebCore::Layout::InlineFormattingContext::Line::justifyRuns):
(WebCore::Layout::InlineFormattingContext::Line::close):
(WebCore::Layout::isNonCollapsedText): Deleted.

9:02 AM Changeset in webkit [237323] by Alan Bujtas
  • 9 edits in trunk/Source/WebCore

[LFC][IFC] Add (right, center)text-align support.
https://bugs.webkit.org/show_bug.cgi?id=190745

Reviewed by Antti Koivisto.

Adjust the logical left of each run while closing the line.

  • layout/Verification.cpp:

(WebCore::Layout::outputMismatchingSimpleLineInformationIfNeeded):
(WebCore::Layout::outputMismatchingComplexLineInformationIfNeeded):

  • layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::trimLeadingRun):
(WebCore::Layout::InlineFormattingContext::layoutInlineContent const):

  • layout/inlineformatting/InlineFormattingContext.h:

(WebCore::Layout::InlineFormattingContext::Line::hasContent const):
(WebCore::Layout::InlineFormattingContext::Line::contentLogicalLeft const): Deleted.

  • layout/inlineformatting/InlineLineBreaker.cpp:

(WebCore::Layout::InlineLineBreaker::nextRun):
(WebCore::Layout::InlineLineBreaker::splitRun):

  • layout/inlineformatting/InlineLineBreaker.h:
  • layout/inlineformatting/InlineRun.h:

(WebCore::Layout::InlineRun::setLogicalLeft):
(WebCore::Layout::InlineRun::TextContext::start const):
(WebCore::Layout::InlineRun::createRun):
(WebCore::Layout::InlineRun::createTextRun):
(WebCore::Layout::InlineRun::InlineRun):
(WebCore::Layout::InlineRun::TextContext::TextContext):
(WebCore::Layout::InlineRun::TextContext::position const): Deleted.

  • layout/inlineformatting/Line.cpp:

(WebCore::Layout::InlineFormattingContext::Line::Line):
(WebCore::Layout::InlineFormattingContext::Line::init):
(WebCore::Layout::adjustedLineLogicalLeft):
(WebCore::Layout::InlineFormattingContext::Line::contentLogicalRight):
(WebCore::Layout::InlineFormattingContext::Line::appendContent):
(WebCore::Layout::InlineFormattingContext::Line::close):
(WebCore::Layout::InlineFormattingContext::Line::setConstraints): Deleted.

  • layout/layouttree/LayoutTreeBuilder.cpp:

(WebCore::Layout::outputInlineRuns):

8:48 AM Changeset in webkit [237322] by Truitt Savell
  • 2 edits in trunk/LayoutTests

[ MacOS WK1 ] Layout Test platform/mac/media/audio-session-category-video-paused.html is flaky Timeout
https://bugs.webkit.org/show_bug.cgi?id=189680

Unreviewed Test Gardening

  • platform/mac-wk1/TestExpectations:
8:42 AM Changeset in webkit [237321] by Alan Bujtas
  • 6 edits in trunk/Source/WebCore

[LFC][IFC] Implement Replaced helper class.
https://bugs.webkit.org/show_bug.cgi?id=190719

Reviewed by Antti Koivisto.

  • layout/layouttree/LayoutBox.cpp:

(WebCore::Layout::Box::Box):

  • layout/layouttree/LayoutBox.h:

(WebCore::Layout::Box::replaced const):

  • layout/layouttree/LayoutReplaced.cpp:

(WebCore::Layout::Replaced::Replaced):
(WebCore::Layout::Replaced::hasIntrinsicWidth const):
(WebCore::Layout::Replaced::hasIntrinsicHeight const):
(WebCore::Layout::Replaced::hasIntrinsicRatio const):
(WebCore::Layout::Replaced::intrinsicWidth const):
(WebCore::Layout::Replaced::intrinsicHeight const):
(WebCore::Layout::Replaced::intrinsicRatio const):

  • layout/layouttree/LayoutReplaced.h:

(WebCore::Layout::Replaced::hasIntrinsicWidth const): Deleted.
(WebCore::Layout::Replaced::hasIntrinsicHeight const): Deleted.
(WebCore::Layout::Replaced::hasIntrinsicRatio const): Deleted.
(WebCore::Layout::Replaced::intrinsicWidth const): Deleted.
(WebCore::Layout::Replaced::intrinsicHeight const): Deleted.
(WebCore::Layout::Replaced::intrinsicRatio const): Deleted.

  • layout/layouttree/LayoutTreeBuilder.cpp:

(WebCore::Layout::TreeBuilder::createSubTree):

8:21 AM Changeset in webkit [237320] by Alan Bujtas
  • 2 edits in trunk/Source/WebCore

[LFC][IFC] Block formatting context roots avoid floats.
https://bugs.webkit.org/show_bug.cgi?id=190723

Reviewed by Antti Koivisto.

Inline formatting context roots don't avoid floats (unless they also establish block formatting context).

  • layout/blockformatting/BlockFormattingContext.cpp:

(WebCore::Layout::BlockFormattingContext::layoutFormattingContextRoot const):

12:19 AM Changeset in webkit [237319] by bshafiei@apple.com
  • 3 edits in branches/safari-606-branch/Source/WebKit

Cherry-pick r235739. rdar://problem/45445194

WebKit/Platform/IPC/mac/ConnectionMac.mm:222: _dispatch_bug_kevent_vanished
https://bugs.webkit.org/show_bug.cgi?id=189314
<rdar://problem/41248286>

Reviewed by Anders Carlsson.

There is a short period in time when m_isServer is true, after open() has been
called, but before we've receive the InitializeConnection IPC, where m_receiveSource
has been initialized but m_isConnected is still false. If platformInvalidate() gets
called during this period of time, we would fail to cancel / release m_receiveSource
and we would forcefully deallocate m_receivePort, leading to the libdispatch simulated
crashes.

To address the issue, platformInvalidate() now properly cancels / releases
m_receiveSource if present, and only deallocates m_receivePort manually if m_receiveSource
has not been initialized (i.e. open() has not been called yet).

  • Platform/IPC/Connection.h:
  • Platform/IPC/mac/ConnectionMac.mm: (IPC::Connection::platformInvalidate): (IPC::Connection::clearReceiveSource):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@235739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12:19 AM Changeset in webkit [237318] by bshafiei@apple.com
  • 2 edits in branches/safari-606-branch/Source/JavaScriptCore

Cherry-pick r237215. rdar://problem/45445113

GetIndexedPropertyStorage can GC.
https://bugs.webkit.org/show_bug.cgi?id=190625
<rdar://problem/45309366>

Reviewed by Saam Barati.

This is because if the ArrayMode type is String, the DFG and FTL will be emitting
a call to operationResolveRope, and operationResolveRope can GC. This patch
updates doesGC() to reflect this.

  • dfg/DFGDoesGC.cpp: (JSC::DFG::doesGC):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@237215 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12:19 AM Changeset in webkit [237317] by bshafiei@apple.com
  • 2 edits in branches/safari-606-branch/Source/WebKit

Apply patch. rdar://problem/45285649

12:19 AM Changeset in webkit [237316] by bshafiei@apple.com
  • 2 edits in branches/safari-606-branch/Source/WTF

Cherry-pick r236969. rdar://problem/45285687

StringTypeAdapter constructor is not properly enforcing String::MaxLength.
https://bugs.webkit.org/show_bug.cgi?id=190392
<rdar://problem/45116210>

Reviewed by Saam Barati.

Previously, the StringTypeAdapter constructor for a UChar* string was summing the
unsigned length of the source string without an overflow check. We now make that
length a size_t which removes this issue, and assert that it's within
String::MaxLength thereafter.

Also made the StringTypeAdapter constructor for a LChar* string behave in an
equivalent manner for consistency. In both cases, we'll crash in a RELEASE_ASSERT
if the source string length exceeds String::MaxLength.

  • wtf/text/StringConcatenate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236969 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12:19 AM Changeset in webkit [237315] by bshafiei@apple.com
  • 10 edits
    1 add in branches/safari-606-branch

Cherry-pick r236804. rdar://problem/45285687

Make string MaxLength for all WTF and JS strings consistently equal to INT_MAX.
https://bugs.webkit.org/show_bug.cgi?id=190187
<rdar://problem/42512909>

Reviewed by Michael Saboff.

JSTests:

  • stress/regress-190187.js: Added.

Source/JavaScriptCore:

Allowing different max string lengths at each level opens up opportunities for
bugs to creep in. With 2 different max length values, it is more difficult to
keep the story straight on how we do overflow / bounds checks at each place in
the code. It's also difficult to tell if a seemingly valid check at the WTF level
will have bad ramifications at the JSC level. Also, it's also not meaningful to
support a max length > INT_MAX. To eliminate this class of bugs, we'll
standardize on a MaxLength of INT_MAX at all levels.

We'll also standardize the way we do length overflow checks on using
CheckedArithmetic, and add some asserts to document the assumptions of the code.

  • runtime/FunctionConstructor.cpp: (JSC::constructFunctionSkippingEvalEnabledCheck):
  • Fix OOM error handling which crashed a test after the new MaxLength was applied.
  • runtime/JSString.h: (JSC::JSString::finishCreation): (JSC::JSString::createHasOtherOwner): (JSC::JSString::setLength):
  • runtime/JSStringInlines.h: (JSC::jsMakeNontrivialString):
  • runtime/Operations.h: (JSC::jsString):

Source/WTF:

  • wtf/text/StringConcatenate.h: (WTF::tryMakeStringFromAdapters): (WTF::sumWithOverflow): Deleted.
  • wtf/text/StringImpl.h:
  • wtf/text/WTFString.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@236804 268f45cc-cd09-0410-ab3c-d52691b4dbfc

12:19 AM Changeset in webkit [237314] by bshafiei@apple.com
  • 18 edits
    2 adds in branches/safari-606-branch

Cherry-pick r234127. rdar://problem/45285391

[INTL] Language tags are not canonicalized
https://bugs.webkit.org/show_bug.cgi?id=185836

Patch by Andy VanWagoner <andy@vanwagoner.family> on 2018-07-23
Reviewed by Keith Miller.

JSTests:

Remove expected failures that have been fixed.

  • test262/expectations.yaml:

Source/JavaScriptCore:

Canonicalize language tags, replacing deprecated tag parts with the
preferred values. Remove broken support for algorithmic numbering systems,
that can cause an error in icu, and are not supported in other engines.

Generate the lookup functions from the language-subtag-registry.

Also initialize the UNumberFormat in initializeNumberFormat so any
failures are thrown immediately instead of failing to format later.

  • CMakeLists.txt:
  • DerivedSources.make:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • Scripts/generateIntlCanonicalizeLanguage.py: Added.
  • runtime/IntlDateTimeFormat.cpp: (JSC::IntlDateTimeFormat::initializeDateTimeFormat):
  • runtime/IntlNumberFormat.cpp: (JSC::IntlNumberFormat::initializeNumberFormat): (JSC::IntlNumberFormat::formatNumber): (JSC::IntlNumberFormat::formatToParts): (JSC::IntlNumberFormat::createNumberFormat): Deleted.
  • runtime/IntlNumberFormat.h:
  • runtime/IntlObject.cpp: (JSC::intlNumberOption): (JSC::intlDefaultNumberOption): (JSC::preferredLanguage): (JSC::preferredRegion): (JSC::canonicalLangTag): (JSC::canonicalizeLanguageTag): (JSC::defaultLocale): (JSC::removeUnicodeLocaleExtension): (JSC::numberingSystemsForLocale): (JSC::grandfatheredLangTag): Deleted.
  • runtime/IntlObject.h:
  • runtime/IntlPluralRules.cpp: (JSC::IntlPluralRules::initializePluralRules):
  • runtime/JSGlobalObject.cpp: (JSC::addMissingScriptLocales): (JSC::JSGlobalObject::intlCollatorAvailableLocales): (JSC::JSGlobalObject::intlDateTimeFormatAvailableLocales): (JSC::JSGlobalObject::intlNumberFormatAvailableLocales): (JSC::JSGlobalObject::intlPluralRulesAvailableLocales):
  • ucd/language-subtag-registry.txt: Added.

LayoutTests:

Use gregory instead of gregorian, matching test262/intl402 and other engines.
Remove tests for algorithmic numbering systems. Add NumberFormat numbering system tests.

  • js/intl-datetimeformat-expected.txt:
  • js/intl-numberformat-expected.txt:
  • js/script-tests/intl-datetimeformat.js:
  • js/script-tests/intl-numberformat.js: (string_appeared_here):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@234127 268f45cc-cd09-0410-ab3c-d52691b4dbfc

Oct 21, 2018:

7:06 PM Changeset in webkit [237313] by Fujii Hironori
  • 3 edits in trunk/Source/WebKitLegacy/win

[Win][Clang] WebView.h: warning: 'QueryInterface' overrides a member function but is not marked 'override' [-Winconsistent-missing-override]
https://bugs.webkit.org/show_bug.cgi?id=190744

Reviewed by Alex Christensen.

clang-cl reports compilation warnings for inconsistent 'override'
keyword usage.

WebView::flushPendingGraphicsLayerChanges is used only if USE(CA).
This can't be marked 'override' if !USE(CA).

  • WebView.cpp:

(WebView::flushPendingGraphicsLayerChanges): Define flushPendingGraphicsLayerChanges only if USE(CA).

  • WebView.h: Marked all overriding member function declarations of WebView 'override'.

Declare flushPendingGraphicsLayerChanges only if USE(CA).

Note: See TracTimeline for information about the timeline view.