⚠ Archived content — this site is no longer maintained.   Current WebKit documentation is at docs.webkit.org.

Timeline



Jul 25, 2017:

11:19 PM Changeset in webkit [219905] by Carlos Garcia Campos
  • 4 edits in trunk/Source

Icon loader error on startup
​https://bugs.webkit.org/show_bug.cgi?id=174787

Reviewed by Brady Eidson.

Source/WebCore:

This is a regression of the new icon loading, it happens with pages that shouldn't have a favicon, like about
pages. IconController::startLoader() did several checks before starting the load that
DocumentLoader::startIconLoading() is not doing. It checked that the frame is the main one, the document can have
an icon (document url is not empty and not about:blank) and that favicon url is in HTTP family. We should do the
same checks now before starting to load icons.

  • loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::startIconLoading):

Source/WebKit:

Ignore non HTTP favicons in glib API.

  • UIProcess/API/glib/WebKitIconLoadingClient.cpp:
10:44 PM Changeset in webkit [219904] by beidson@apple.com
  • 13 edits
    2 adds in trunk

ResourceLoadStatistics grandfathering happens much too often.
<rdar://problem/32655834> and ​https://bugs.webkit.org/show_bug.cgi?id=174825

Reviewed by Chris Dumez.

Source/WebKit:

The ResourceLoadStatistics grandfathering procedure happens too often.

  • With an empty memory store, even though an empty memory store is a perfectly valid state.
  • On each launch, even if grandfathering happened on the last launch - because grandfathering data would not automatically be saved to disk.
  • After clearing all website data, at which point no grandfathering can possibly be relevant because there is no data to grandfather.

This patch fixes those cases and adds API tests to verify they remain fixed.

  • Shared/WebsiteData/WebsiteDataType.h:
  • UIProcess/API/Cocoa/WKWebsiteDataStore.mm:

(+[WKWebsiteDataStore _allWebsiteDataTypesIncludingPrivate]): allWebsiteDataTypes, but even with the private ones.
(-[WKWebsiteDataStore _resourceLoadStatisticsClearInMemoryAndPersistentStore]): If the types being cleared cover all of

the types that ResourceLoadStatistics care about, don't grandfather afterwards.

(-[WKWebsiteDataStore _resourceLoadStatisticsClearInMemoryAndPersistentStoreModifiedSinceHours:]): Ditto.
(-[WKWebsiteDataStore _setResourceLoadStatisticsTestingCallback:]):

  • UIProcess/API/Cocoa/WKWebsiteDataStorePrivate.h:
  • UIProcess/Storage/ResourceLoadStatisticsPersistentStorage.cpp:

(WebKit::ResourceLoadStatisticsPersistentStorage::populateMemoryStoreFromDisk): Don't grandfather if the store read from

disk is empty (as being empty is perfectly fine), and also log the event of "populated without grandfathering".

  • UIProcess/Storage/ResourceLoadStatisticsPersistentStorage.h:
  • UIProcess/WebProcessProxy.cpp:

(WebKit::WebProcessProxy::topPrivatelyControlledDomainsWithWebsiteData):

  • UIProcess/WebResourceLoadStatisticsStore.cpp:

(WebKit::WebResourceLoadStatisticsStore::monitoredDataTypes):
(WebKit::WebResourceLoadStatisticsStore::WebResourceLoadStatisticsStore):
(WebKit::WebResourceLoadStatisticsStore::~WebResourceLoadStatisticsStore):
(WebKit::WebResourceLoadStatisticsStore::removeDataRecords):
(WebKit::WebResourceLoadStatisticsStore::grandfatherExistingWebsiteData): Schedule a write right away so we don't re-grandfather

on next launch, and also log the grandfathering event.

(WebKit::WebResourceLoadStatisticsStore::scheduleClearInMemoryAndPersistent): Takes a ShouldGrandfather flag

to tell whether or not data should be re-grandfathered after the store is cleared.

(WebKit::WebResourceLoadStatisticsStore::logTestingEvent): Log the event to the testing client.
(WebKit::dataTypesToRemove): Deleted.

  • UIProcess/WebResourceLoadStatisticsStore.h:
  • UIProcess/WebsiteData/WebsiteDataStore.cpp:

(WebKit::WebsiteDataStore::removeData):
(WebKit::WebsiteDataStore::setResourceLoadStatisticsEnabled):
(WebKit::WebsiteDataStore::enableResourceLoadStatisticsAndSetTestingCallback): Handles enabling ResourceLoadStatistics both

with and without a testing callback.

  • UIProcess/WebsiteData/WebsiteDataStore.h:

Tools:

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKit2Cocoa/EmptyGrandfatheredResourceLoadStatistics.plist: Added.
  • TestWebKitAPI/Tests/WebKit2Cocoa/ResourceLoadStatistics.mm: Added.

(TEST):

9:42 PM Changeset in webkit [219903] by commit-queue@webkit.org
  • 4 edits
    9 moves
    3 adds
    1 delete in trunk/PerformanceTests

Speedometer 2.0: Document es2015-babel-webpack build process
​https://bugs.webkit.org/show_bug.cgi?id=174252

Patch by Mathias Bynens <​mathias@qiwi.be> on 2017-07-25
Reviewed by Ryosuke Niwa.

  • Speedometer/resources/todomvc/vanilla-examples/es2015-babel-webpack/README.md: Added build docs.
  • Speedometer/resources/todomvc/vanilla-examples/es2015-babel-webpack/dist/*: Update generated build files.
  • Speedometer/resources/todomvc/vanilla-examples/es2015-babel-webpack/package.json: Remove unused items.
  • Speedometer/resources/todomvc/vanilla-examples/es2015-babel-webpack/package-lock.json: Added lockfile.
  • Speedometer/resources/todomvc/vanilla-examples/es2015-babel-webpack/src/index.html: Make title consistent.
9:31 PM Changeset in webkit [219902] by weinig@apple.com
  • 7 edits
    1 delete in trunk/Source/WebCore

[WebIDL] Remove custom bindings for HTMLCanvasElement
​https://bugs.webkit.org/show_bug.cgi?id=174847

Reviewed by Darin Adler.

  • CMakeLists.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSBindingsAllInOne.cpp:
  • bindings/js/JSHTMLCanvasElementCustom.cpp: Removed.

Remove JSHTMLCanvasElementCustom.

  • html/HTMLCanvasElement.cpp:

(WebCore::HTMLCanvasElement::getContext):
(WebCore::HTMLCanvasElement::getContext2d):
(WebCore::HTMLCanvasElement::getContextWebGL):
(WebCore::HTMLCanvasElement::getContextWebGPU):

  • html/HTMLCanvasElement.h:
  • html/HTMLCanvasElement.idl:

Move context picking code to HTMLCanvasElement. Give more
exact types to remaining context getters.

7:34 PM Changeset in webkit [219901] by Chris Dumez
  • 20 edits in trunk/Source

Clean up ExceptionCode enumeration
​https://bugs.webkit.org/show_bug.cgi?id=174812

Reviewed by Darin Adler.

Clean up ExceptionCode enumeration:

  • Updated comments to reflect the latest WebIDL specification
  • Drop special values given to some exception codes. Those legacy codes are already in the table inside DOMException.cpp. This also allows us to drop empty rows in the DOMException table
  • Drop NoException value in the enumeration and use std::optional<ExceptionCode> when needed instead.
  • Modules/indexeddb/shared/IDBError.cpp:

(WebCore::IDBError::IDBError):
(WebCore::IDBError::name):
(WebCore::IDBError::message):

  • Modules/indexeddb/shared/IDBError.h:

(WebCore::IDBError::code):
(WebCore::IDBError::encode):
(WebCore::IDBError::decode):

  • dom/DOMException.cpp:
  • dom/ExceptionCode.h:
  • fileapi/FileReaderSync.cpp:

(WebCore::errorCodeToException):
(WebCore::FileReaderSync::startLoading):

  • xml/XMLHttpRequest.cpp:

(WebCore::XMLHttpRequest::createRequest):

  • xml/XMLHttpRequest.h:
7:27 PM Changeset in webkit [219900] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Don't output "No message" for multi-value logs like console.log(x, y)
​https://bugs.webkit.org/show_bug.cgi?id=174842

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2017-07-25
Reviewed by Matt Baker.

  • Localizations/en.lproj/localizedStrings.js:

Removed "No message".

  • UserInterface/Views/ConsoleMessageView.js:

(WebInspector.ConsoleMessageView.prototype._appendFormattedArguments):
Track whether or not outputting the next parameter will need a divider
in front of it or not.

7:23 PM Changeset in webkit [219899] by jfbastien@apple.com
  • 6 edits in trunk/Source/JavaScriptCore

WebAssembly: generate smaller binaries
​https://bugs.webkit.org/show_bug.cgi?id=174818

Reviewed by Filip Pizlo.

This patch reduces generated code size for WebAssembly in 2 ways:

  1. Use the ZR register when storing zero on ARM64.
  2. Synthesize wasm context lazily.

This leads to a modest size reduction on both x86-64 and ARM64 for
large WebAssembly games, without any performance loss on WasmBench
and TitzerBench.

The reason this works is that these games, using Emscripten,
generate 100k+ tiny functions, and our JIT allocation granule
rounds all allocations up to 32 bytes. There are plenty of other
simple gains to be had, I've filed a follow-up bug at
webkit.org/b/174819

We should further avoid the per-function cost of tiering, which
represents the bulk of code generated for small functions.

  • assembler/MacroAssemblerARM64.h:

(JSC::MacroAssemblerARM64::storeZero64):

  • assembler/MacroAssemblerX86_64.h:

(JSC::MacroAssemblerX86_64::storeZero64):

  • b3/B3LowerToAir.cpp:

(JSC::B3::Air::LowerToAir::createStore): this doesn't make sense
for x86 because it constrains register reuse and codegen in a way
that doesn't affect ARM64 because it has a dedicated zero
register.

  • b3/air/AirOpcode.opcodes: add the storeZero64 opcode.
  • wasm/WasmB3IRGenerator.cpp:

(JSC::Wasm::B3IRGenerator::instanceValue):
(JSC::Wasm::B3IRGenerator::restoreWasmContext):
(JSC::Wasm::B3IRGenerator::B3IRGenerator):
(JSC::Wasm::B3IRGenerator::materializeWasmContext): Deleted.

6:58 PM Changeset in webkit [219898] by fpizlo@apple.com
  • 17 edits
    8 adds
    1 delete in trunk/Source

B3 should do LICM
​https://bugs.webkit.org/show_bug.cgi?id=174750

Reviewed by Keith Miller and Saam Barati.

Source/JavaScriptCore:

Added a LICM phase to B3. This phase is called hoistLoopInvariantValues, to conform to the B3 naming
convention for phases (it has to be an imperative). The phase uses NaturalLoops and BackwardsDominators,
so this adds those analyses to B3. BackwardsDominators was already available in templatized form. This
change templatizes DFG::NaturalLoops so that we can just use it.

The LICM phase itself is really simple. We are decently precise with our handling of everything except
the relationship between control dependence and side exits.

Also added a bunch of tests.

This isn't super important. It's perf-neutral on JS benchmarks. FTL already does LICM on DFG SSA IR, and
probably all current WebAssembly content has had LICM done to it. That being said, this is a cheap phase
so it doesn't hurt to have it.

I wrote it because I thought I needed it for bug 174727. It turns out that there's a better way to
handle the problem I had, so I ended up not needed it - but by then I had already written it. I think
it's good to have it because LICM is one of those core compiler phases; every compiler has it
eventually.

  • CMakeLists.txt:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • b3/B3BackwardsCFG.h: Added.

(JSC::B3::BackwardsCFG::BackwardsCFG):

  • b3/B3BackwardsDominators.h: Added.

(JSC::B3::BackwardsDominators::BackwardsDominators):

  • b3/B3BasicBlock.cpp:

(JSC::B3::BasicBlock::appendNonTerminal):

  • b3/B3Effects.h:
  • b3/B3EnsureLoopPreHeaders.cpp: Added.

(JSC::B3::ensureLoopPreHeaders):

  • b3/B3EnsureLoopPreHeaders.h: Added.
  • b3/B3Generate.cpp:

(JSC::B3::generateToAir):

  • b3/B3HoistLoopInvariantValues.cpp: Added.

(JSC::B3::hoistLoopInvariantValues):

  • b3/B3HoistLoopInvariantValues.h: Added.
  • b3/B3NaturalLoops.h: Added.

(JSC::B3::NaturalLoops::NaturalLoops):

  • b3/B3Procedure.cpp:

(JSC::B3::Procedure::invalidateCFG):
(JSC::B3::Procedure::naturalLoops):
(JSC::B3::Procedure::backwardsCFG):
(JSC::B3::Procedure::backwardsDominators):

  • b3/B3Procedure.h:
  • b3/testb3.cpp:

(JSC::B3::generateLoop):
(JSC::B3::makeArrayForLoops):
(JSC::B3::generateLoopNotBackwardsDominant):
(JSC::B3::oneFunction):
(JSC::B3::noOpFunction):
(JSC::B3::testLICMPure):
(JSC::B3::testLICMPureSideExits):
(JSC::B3::testLICMPureWritesPinned):
(JSC::B3::testLICMPureWrites):
(JSC::B3::testLICMReadsLocalState):
(JSC::B3::testLICMReadsPinned):
(JSC::B3::testLICMReads):
(JSC::B3::testLICMPureNotBackwardsDominant):
(JSC::B3::testLICMPureFoiledByChild):
(JSC::B3::testLICMPureNotBackwardsDominantFoiledByChild):
(JSC::B3::testLICMExitsSideways):
(JSC::B3::testLICMWritesLocalState):
(JSC::B3::testLICMWrites):
(JSC::B3::testLICMFence):
(JSC::B3::testLICMWritesPinned):
(JSC::B3::testLICMControlDependent):
(JSC::B3::testLICMControlDependentNotBackwardsDominant):
(JSC::B3::testLICMControlDependentSideExits):
(JSC::B3::testLICMReadsPinnedWritesPinned):
(JSC::B3::testLICMReadsWritesDifferentHeaps):
(JSC::B3::testLICMReadsWritesOverlappingHeaps):
(JSC::B3::testLICMDefaultCall):
(JSC::B3::run):

  • dfg/DFGBasicBlock.h:
  • dfg/DFGCFG.h:
  • dfg/DFGNaturalLoops.cpp: Removed.
  • dfg/DFGNaturalLoops.h:

(JSC::DFG::NaturalLoops::NaturalLoops):
(JSC::DFG::NaturalLoop::NaturalLoop): Deleted.
(JSC::DFG::NaturalLoop::header): Deleted.
(JSC::DFG::NaturalLoop::size): Deleted.
(JSC::DFG::NaturalLoop::at): Deleted.
(JSC::DFG::NaturalLoop::operator[]): Deleted.
(JSC::DFG::NaturalLoop::contains): Deleted.
(JSC::DFG::NaturalLoop::index): Deleted.
(JSC::DFG::NaturalLoop::isOuterMostLoop): Deleted.
(JSC::DFG::NaturalLoop::addBlock): Deleted.
(JSC::DFG::NaturalLoops::numLoops): Deleted.
(JSC::DFG::NaturalLoops::loop): Deleted.
(JSC::DFG::NaturalLoops::headerOf): Deleted.
(JSC::DFG::NaturalLoops::innerMostLoopOf): Deleted.
(JSC::DFG::NaturalLoops::innerMostOuterLoop): Deleted.
(JSC::DFG::NaturalLoops::belongsTo): Deleted.
(JSC::DFG::NaturalLoops::loopDepth): Deleted.

Source/WTF:

Moved DFG::NaturalLoops to WTF. The new templatized NaturalLoops<> uses the same Graph abstraction as
Dominators<>. This allows us to add a B3::NaturalLoops for free.

Also made small tweaks to RangeSet, which the LICM uses.

  • WTF.xcodeproj/project.pbxproj:
  • wtf/CMakeLists.txt:
  • wtf/Dominators.h:
  • wtf/NaturalLoops.h: Added.

(WTF::NaturalLoop::NaturalLoop):
(WTF::NaturalLoop::graph):
(WTF::NaturalLoop::header):
(WTF::NaturalLoop::size):
(WTF::NaturalLoop::at):
(WTF::NaturalLoop::operator[]):
(WTF::NaturalLoop::contains):
(WTF::NaturalLoop::index):
(WTF::NaturalLoop::isOuterMostLoop):
(WTF::NaturalLoop::dump):
(WTF::NaturalLoop::addBlock):
(WTF::NaturalLoops::NaturalLoops):
(WTF::NaturalLoops::graph):
(WTF::NaturalLoops::numLoops):
(WTF::NaturalLoops::loop):
(WTF::NaturalLoops::headerOf):
(WTF::NaturalLoops::innerMostLoopOf):
(WTF::NaturalLoops::innerMostOuterLoop):
(WTF::NaturalLoops::belongsTo):
(WTF::NaturalLoops::loopDepth):
(WTF::NaturalLoops::loopsOf):
(WTF::NaturalLoops::dump):

  • wtf/RangeSet.h:

(WTF::RangeSet::begin):
(WTF::RangeSet::end):
(WTF::RangeSet::addAll):

6:19 PM Changeset in webkit [219897] by fpizlo@apple.com
  • 5 edits in trunk/Source/JavaScriptCore

GC should be fine with trading blocks between destructor and non-destructor blocks
​https://bugs.webkit.org/show_bug.cgi?id=174811

Reviewed by Mark Lam.

Our GC has the ability to trade blocks between MarkedAllocators. A MarkedAllocator is a
size-class-within-a-Subspace. The ability to trade helps reduce memory wastage due to
fragmentation. Prior to this change, this only worked between blocks that did not have destructors.
This was partly a policy decision. But mostly, it was fallout from the way we use the empty block
set.

Here's how empty used to work. If a block is empty, we don't run destructors. We say that a block
is empty if:

A) It has no live objects and its a non-destructor block, or
B) We just allocated it (so it has no destructors even if it's a destructor block), or
C) We just stole it from another allocator (so it also has no destructors), or
D) We just swept the block and ran all destructors.

Case (A) is for trading blocks. That's how a different MarkedAllocator would know that this is a
block that could be stolen.

Cases (B) and (C) need to be detected for correctness, since otherwise we might try to run
destructors in blocks that have garbage bits. In that case, the isZapped check won't detect that
cells don't need destruction, so without having the empty bit we would try to destruct garbage
and crash. Currently, we know that we have cases (B) and (C) when the block is empty.

Case (D) is necessary for detecting which blocks can be removed when we shrink the heap.

If we tried to enable trading of blocks between allocators without making any changes to how
empty works, then it just would not work. We have to set the empty bits of blocks that have no
live objects in order for those bits to be candidates for trading. But if we do that, then our
logic for cases (B-D) will think that the block has no destructible objects. That's bad, since then
our destructors won't run and we'll leak memory.

This change fixes this issue by decoupling the "do I have destructors" question from the "do I have
live objects" question by introducing a new destructible bitvector. The GC flags all live blocks
as being destructible at the end. We clear the destructible bit in cases (B-D). Cases (B-C) are
handled entirely by the new destrictible bit, while case (D) is detected by looking for blocks that
are (empty & ~destructible).

Then we can simply remove all destructor-oriented special-casing of the empty bit. And we can
remove destructor-oriented special-casing of block trading.

This is a perf-neutral change. We expect most free memory to be in non-destructor blocks anyway,
so this change is more about clean-up than perf. But, this could reduce memory usage in some
pathological cases.

  • heap/MarkedAllocator.cpp:

(JSC::MarkedAllocator::findEmptyBlockToSteal):
(JSC::MarkedAllocator::tryAllocateWithoutCollecting):
(JSC::MarkedAllocator::endMarking):
(JSC::MarkedAllocator::shrink):
(JSC::MarkedAllocator::shouldStealEmptyBlocksFromOtherAllocators): Deleted.

  • heap/MarkedAllocator.h:
  • heap/MarkedBlock.cpp:

(JSC::MarkedBlock::Handle::lastChanceToFinalize):
(JSC::MarkedBlock::Handle::sweep):

  • heap/MarkedBlockInlines.h:

(JSC::MarkedBlock::Handle::specializedSweep):
(JSC::MarkedBlock::Handle::finishSweepKnowingSubspace):
(JSC::MarkedBlock::Handle::emptyMode):

6:03 PM Changeset in webkit [219896] by aestes@apple.com
  • 8 edits in trunk

[Apple Pay] Add "carteBancaire" as a supported payment network
​https://bugs.webkit.org/show_bug.cgi?id=174841
<rdar://problem/31935596>

Reviewed by Alex Christensen.

Source/WebCore:

Tests: http/tests/ssl/applepay/ApplePaySession.html

http/tests/ssl/applepay/ApplePaySessionV3.html

  • Modules/applepay/PaymentRequest.cpp:

(WebCore::PaymentRequest::isValidSupportedNetwork):

Source/WebKit:

  • UIProcess/ApplePay/cocoa/WebPaymentCoordinatorProxyCocoa.mm:

(WebKit::toSupportedNetwork):

LayoutTests:

  • http/tests/ssl/applepay/ApplePaySession-expected.txt:
  • http/tests/ssl/applepay/ApplePaySession.html:
  • http/tests/ssl/applepay/ApplePaySessionV3.html:
5:45 PM Changeset in webkit [219895] by keith_miller@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Remove Broken CompareEq constant folding phase.
​https://bugs.webkit.org/show_bug.cgi?id=174846
<rdar://problem/32978808>

Reviewed by Saam Barati.

This bug happened when we would get code like the following:

a: JSConst(Undefined)
b: GetLocal(SomeObjectOrUndefined)
...
c: CompareEq(Check:ObjectOrOther:b, Check:ObjectOrOther:a)

constant folding will turn this into:

a: JSConst(Undefined)
b: GetLocal(SomeObjectOrUndefined)
...
c: CompareEq(Check:ObjectOrOther:b, Other:a)

But the SpeculativeJIT/FTL lowering will fail to check b
properly which leads to an assertion failure in the AI.

I'll follow up with a more robust fix later. For now, I'll remove the
case that generates the code. Removing the code appears to be perf
neutral.

  • dfg/DFGConstantFoldingPhase.cpp:

(JSC::DFG::ConstantFoldingPhase::foldConstants):

5:45 PM Changeset in webkit [219894] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Fix warnings about console.assert lines without semicolons
​https://bugs.webkit.org/show_bug.cgi?id=174840

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2017-07-25
Reviewed by Brian Burg.

  • UserInterface/Views/SettingsTabContentView.js:

(WebInspector.SettingsTabContentView.prototype.set selectedSettingsView):
(WebInspector.SettingsTabContentView.prototype.setSettingsViewVisible):
Add semicolon to assert lines so they can be stripped in optimized builds.

4:31 PM Changeset in webkit [219893] by commit-queue@webkit.org
  • 5 edits in trunk/Source/WebInspectorUI

Web Inspector: Inline multiple console log values if they are simple
​https://bugs.webkit.org/show_bug.cgi?id=174746
<rdar://problem/33469376>

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2017-07-25
Reviewed by Matt Baker.

  • UserInterface/Models/IssueMessage.js:
  • UserInterface/Protocol/RemoteObject.js:

(WebInspector.RemoteObject.type): Deleted.
Remove this as it isn't as useful as directly checking the type.

  • UserInterface/Views/ConsoleMessageView.js:

(WebInspector.ConsoleMessageView.prototype._appendExtraParameters):
(WebInspector.ConsoleMessageView.prototype._appendFormattedArguments):
(WebInspector.ConsoleMessageView.prototype._hasSimpleDisplay):
(WebInspector.ConsoleMessageView.prototype._isStackTrace):
For leading primitive/simple values, display them inline on the console message.

  • UserInterface/Views/FormattedValue.js:

(WebInspector.FormattedValue.hasSimpleDisplay):
Provide a helper to determine if a formatted value will be simple.

4:06 PM Changeset in webkit [219892] by dbates@webkit.org
  • 3 edits in trunk/Source/WebCore

Make StyleResolver::isValid{Cue, Region}StyleProperty() static, inline, non-member functions
​https://bugs.webkit.org/show_bug.cgi?id=174827

Reviewed by Simon Fraser.

StyleResolver::isValid{Cue, Region}StyleProperty() access neither instance nor
class data. Moreover, they are private member functions and hence cannot be
used outside of StyleResolver's implementation. So, make these static, inline,
non-member functions.

  • css/StyleResolver.cpp:

(WebCore::isValidRegionStyleProperty):
(WebCore::isValidCueStyleProperty):
(WebCore::StyleResolver::CascadedProperties::addMatch):
(WebCore::StyleResolver::isValidRegionStyleProperty): Deleted.
(WebCore::StyleResolver::isValidCueStyleProperty): Deleted.

  • css/StyleResolver.h:
3:48 PM Changeset in webkit [219891] by pvollan@apple.com
  • 2 edits in trunk/Source/WebCore

Dragged links are blurry on 1x displays.
​https://bugs.webkit.org/show_bug.cgi?id=174831
<rdar://problem/33519698>

Reviewed by Simon Fraser.

When the width and height of the drag image is a multiple of 2, the drag image is not blurry
on a 1x display. This is a workaround which should be removed when <rdar://problem/33059739>
is fixed.

No new tests, since this is not straightforward to test with a layout test.

  • platform/mac/DragImageMac.mm:

(WebCore::LinkImageLayout::LinkImageLayout):

2:49 PM Changeset in webkit [219890] by Matt Lewis
  • 2 edits in trunk/LayoutTests

Marked multiple fast/ tests as crashing on High Sierra WK1.
<rdar://problem/33522420>

Unreviewed test gardening.

  • platform/mac-highsierra-wk1/TestExpectations:
2:46 PM Changeset in webkit [219889] by Matt Baker
  • 9 edits in trunk/Source

Web Inspector: Refactoring: extract async stack trace logic from InspectorInstrumentation
​https://bugs.webkit.org/show_bug.cgi?id=174738

Reviewed by Brian Burg.

Source/JavaScriptCore:

Move AsyncCallType enum to InspectorDebuggerAgent, which manages async
stack traces. This preserves the call type in JSC, makes the range of
possible call types explicit, and is safer than passing ints.

  • inspector/agents/InspectorDebuggerAgent.cpp:

(Inspector::InspectorDebuggerAgent::asyncCallIdentifier):
(Inspector::InspectorDebuggerAgent::didScheduleAsyncCall):
(Inspector::InspectorDebuggerAgent::didCancelAsyncCall):
(Inspector::InspectorDebuggerAgent::willDispatchAsyncCall):

  • inspector/agents/InspectorDebuggerAgent.h:

Source/WebCore:

Relocate AsyncCallType from InspectorInstrumentation to the debugger
agent. Plumbing for requestAnimationFrame notifications has been added
to PageDebuggerAgent, which is responsible for managing async stack traces.

  • dom/ScriptedAnimationController.cpp:

(WebCore::ScriptedAnimationController::registerCallback):
(WebCore::ScriptedAnimationController::cancelCallback):
(WebCore::ScriptedAnimationController::serviceScriptedAnimations):

  • inspector/InspectorInstrumentation.cpp:

(WebCore::InspectorInstrumentation::didInstallTimerImpl):
(WebCore::InspectorInstrumentation::didRemoveTimerImpl):
(WebCore::InspectorInstrumentation::willFireTimerImpl):
(WebCore::InspectorInstrumentation::didRequestAnimationFrameImpl):
(WebCore::InspectorInstrumentation::didCancelAnimationFrameImpl):
(WebCore::InspectorInstrumentation::willFireAnimationFrameImpl):
(): Deleted.
(WebCore::didScheduleAsyncCall): Deleted.

  • inspector/InspectorInstrumentation.h:

(WebCore::InspectorInstrumentation::didRequestAnimationFrame):
(WebCore::InspectorInstrumentation::didCancelAnimationFrame):
(WebCore::InspectorInstrumentation::willFireAnimationFrame):
Replaced Frame and Document pointers with references and moved
pointer validation upstream.

  • inspector/PageDebuggerAgent.cpp:

(WebCore::PageDebuggerAgent::didRequestAnimationFrame):
(WebCore::PageDebuggerAgent::willFireAnimationFrame):
(WebCore::PageDebuggerAgent::didCancelAnimationFrame):

  • inspector/PageDebuggerAgent.h:
2:43 PM Changeset in webkit [219888] by Nikita Vasilyev
  • 7 edits
    1 copy in trunk/Source/WebInspectorUI

Web Inspector: Styles: Add a switch for Spreadsheet model style editor to experimental settings
​https://bugs.webkit.org/show_bug.cgi?id=174741
<rdar://problem/33467954>

Reviewed by Brian Burg.

Show a blank panel when Spreadsheet Style Editor is enabled.

  • UserInterface/Base/Setting.js:
  • UserInterface/Main.html:
  • UserInterface/Views/CSSStyleDetailsSidebarPanel.js:

(WebInspector.CSSStyleDetailsSidebarPanel):
(WebInspector.CSSStyleDetailsSidebarPanel.prototype.computedStyleDetailsPanelShowProperty):

  • UserInterface/Views/RulesStyleSpreadsheetDetailsPanel.js: Added.

(WebInspector.RulesStyleSpreadsheetDetailsPanel):
(WebInspector.RulesStyleSpreadsheetDetailsPanel.prototype.filterDidChange):
(WebInspector.RulesStyleSpreadsheetDetailsPanel.prototype.scrollToSectionAndHighlightProperty):

  • UserInterface/Views/SettingsTabContentView.css:

(.content-view.settings > .settings-view > .container.hidden):
(.content-view.settings > .settings-view > .container-centered):
(.content-view.settings > .settings-view > .container button):

  • UserInterface/Views/SettingsTabContentView.js:

(WebInspector.SettingsTabContentView.prototype._createDebugSettingsView):

  • UserInterface/Views/SettingsView.js:

(WebInspector.SettingsView.prototype.addCenteredContainer):
(WebInspector.SettingsView):
Show "Reload Web Inspector" button when Spreadsheet Style Editor setting is modified.

2:22 PM Changeset in webkit [219887] by BJ Burg
  • 5 edits in trunk/Source/WebKit

Use SandboxExtension::HandleArray when sending extensions for file uploads to the web process
​https://bugs.webkit.org/show_bug.cgi?id=174828

Reviewed by Tim Horton.

Send all of the sandbox extension handles in a single message, rather than
one message per extension. Update message name and parameter types to match.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::didChooseFilesForOpenPanelWithDisplayStringAndIcon):
(WebKit::WebPageProxy::didChooseFilesForOpenPanel):

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::extendSandboxForFilesFromOpenPanel):
(WebKit::WebPage::extendSandboxForFileFromOpenPanel): Deleted.

  • WebProcess/WebPage/WebPage.h:
  • WebProcess/WebPage/WebPage.messages.in:
2:12 PM Changeset in webkit [219886] by Matt Lewis
  • 2 edits in trunk/LayoutTests

Marked imported/w3c/web-platform-tests/XMLHttpRequest/send-blob-with-no-mime-type.html as failing.
<rdar://problem/33251388>

Unreviewed test gardening.

  • platform/mac-highsierra/TestExpectations:
1:56 PM Changeset in webkit [219885] by mark.lam@apple.com
  • 5 edits in trunk/Source/JavaScriptCore

Fix bugs in probe code to change sp on x86, x86_64 and 32-bit ARM.
​https://bugs.webkit.org/show_bug.cgi?id=174809
<rdar://problem/33504759>

Reviewed by Filip Pizlo.

  1. When the probe handler function changes the sp register to point to the region of stack in the middle of the ProbeContext on the stack, there is a bug where the ProbeContext's register values to be restored can be over-written before they can be restored. This is now fixed.
  1. Added more robust probe tests for changing the sp register.
  1. Made existing probe tests to ensure that probe handlers were actually called.
  1. Added some verification to testProbePreservesGPRS().
  1. Change all the probe tests to fail early on discovering an error instead of batching till the end of the test. This helps point a finger to the failing issue earlier.

This patch was tested on x86, x86_64, and ARMv7. ARM64 probe code will be fixed
next in ​https://bugs.webkit.org/show_bug.cgi?id=174697.

  • assembler/MacroAssemblerARM.cpp:
  • assembler/MacroAssemblerARMv7.cpp:
  • assembler/MacroAssemblerX86Common.cpp:
  • assembler/testmasm.cpp:

(JSC::testProbeReadsArgumentRegisters):
(JSC::testProbeWritesArgumentRegisters):
(JSC::testProbePreservesGPRS):
(JSC::testProbeModifiesStackPointer):
(JSC::testProbeModifiesStackPointerToInsideProbeContextOnStack):
(JSC::testProbeModifiesStackPointerToNBytesBelowSP):
(JSC::testProbeModifiesProgramCounter):
(JSC::run):

1:55 PM Changeset in webkit [219884] by Matt Lewis
  • 2 edits in trunk/LayoutTests

Rebaselinged the test http/tests/quicklook/top-navigation-blocked.html
​https://bugs.webkit.org/show_bug.cgi?id=174755

Unreviewed test gardening.

  • http/tests/quicklook/top-navigation-blocked-expected.txt:
1:54 PM Changeset in webkit [219883] by jmarcell@apple.com
  • 7 edits in branches/safari-604-branch/Source

Versioning.

1:50 PM Changeset in webkit [219882] by commit-queue@webkit.org
  • 2 edits in trunk/Tools

Fix autoinstaller failing on autoinstall_everything
​https://bugs.webkit.org/show_bug.cgi?id=174830

Patch by Matthew Stewart <​matthew_r_stewart@apple.com> on 2017-07-25
Reviewed by Joseph Pecoraro.

  • Scripts/webkitpy/thirdparty/init.py:

(AutoinstallImportHook.find_module):
(AutoinstallImportHook.install_chromedriver):
(AutoinstallImportHook.install_geckodriver):
(AutoinstallImportHook._install_chromedriver): Deleted.
(AutoinstallImportHook._install_geckodriver): Deleted.

1:47 PM Changeset in webkit [219881] by achristensen@apple.com
  • 2 edits in trunk/Source/WebKit

Fix API tests after r219871.
​https://bugs.webkit.org/show_bug.cgi?id=174807

WebKit2.OpenAndCloseWindowAsync and WebKit2.OpenAsyncWithNil API tests are fixed.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::createNewPage):
createNewPageAsync or createNewPage should be called, not both.

1:45 PM Changeset in webkit [219880] by jmarcell@apple.com
  • 1 copy in tags/Safari-604.1.33

Tag Safari-604.1.33.

1:20 PM Changeset in webkit [219879] by jmarcell@apple.com
  • 7 edits in branches/safari-604-branch/Source

Versioning.

1:16 PM Changeset in webkit [219878] by jmarcell@apple.com
  • 1 delete in tags/Safari-604.1.33

Delete Tag.

11:56 AM Changeset in webkit [219877] by Matt Lewis
  • 2 edits in trunk/LayoutTests

Rebaselined missed test.
​https://bugs.webkit.org/show_bug.cgi?id=174734

Unreviewed test gardening.

  • fast/events/touch/ios/touches-client-coords-after-zoom-expected.txt:
11:53 AM Changeset in webkit [219876] by commit-queue@webkit.org
  • 60 edits
    9 adds in trunk

Async image decoding for large images should be disabled after the first time a tile is painted
​https://bugs.webkit.org/show_bug.cgi?id=174451
<rdar://problem/31246421>

Patch by Said Abou-Hallawa <​sabouhallawa@apple.com> on 2017-07-25
Reviewed by Simon Fraser.

Source/WebCore:

Flashing because of DOM mutation can be fixed by disabling the asynchronous
image decoding after the first time a tile was painted.

We can detect this by consulting the tile repaintCount. If it is zero, then
it is safe to use asynchronous image decoded. If the tile repaintCount is
greater than zero, we are not sure if the renderer rectangle has an image
drawn in it already or not. In this case we have to use the synchronous
image decoding to avoid causing a flash.

Tests: fast/images/async-image-background-change.html

fast/images/async-image-src-change.html
http/tests/multipart/multipart-async-image.html

  • html/shadow/MediaControlElements.cpp:

(WebCore::MediaControlTextTrackContainerElement::createTextTrackRepresentationImage):

  • page/FrameView.cpp:

(WebCore::FrameView::willPaintContents):
(WebCore::FrameView::paintContentsForSnapshot):

  • page/PageOverlayController.cpp:

(WebCore::PageOverlayController::paintContents):

  • page/PageOverlayController.h:
  • page/linux/ResourceUsageOverlayLinux.cpp:
  • page/mac/ServicesOverlayController.h:
  • page/mac/ServicesOverlayController.mm:

(WebCore::ServicesOverlayController::Highlight::paintContents):

  • platform/graphics/BitmapImage.cpp:

(WebCore::BitmapImage::draw):

  • platform/graphics/BitmapImage.h:
  • platform/graphics/GraphicsLayer.cpp:

(WebCore::GraphicsLayer::paintGraphicsLayerContents):

  • platform/graphics/GraphicsLayer.h:
  • platform/graphics/GraphicsLayerClient.h:

(WebCore::GraphicsLayerClient::paintContents):

  • platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.cpp:

(WebCore::LayerClient::platformCALayerPaintContents):

  • platform/graphics/ca/GraphicsLayerCA.cpp:

(WebCore::GraphicsLayerCA::platformCALayerPaintContents):

  • platform/graphics/ca/GraphicsLayerCA.h:
  • platform/graphics/ca/PlatformCALayer.h:
  • platform/graphics/ca/PlatformCALayerClient.h:

(WebCore::PlatformCALayerClient::platformCALayerRepaintCount):

  • platform/graphics/ca/TileCoverageMap.cpp:

(WebCore::TileCoverageMap::platformCALayerPaintContents):

  • platform/graphics/ca/TileCoverageMap.h:
  • platform/graphics/ca/TileGrid.cpp:

(WebCore::TileGrid::platformCALayerPaintContents):
(WebCore::TileGrid::platformCALayerRepaintCount):

  • platform/graphics/ca/TileGrid.h:
  • platform/graphics/ca/cocoa/PlatformCALayerCocoa.mm:

(PlatformCALayer::drawLayerContents):

  • platform/graphics/ca/win/PlatformCALayerWin.cpp:

(PlatformCALayer::drawLayerContents):

  • platform/graphics/ca/win/PlatformCALayerWinInternal.cpp:

(PlatformCALayerWinInternal::displayCallback):

  • platform/graphics/ca/win/WebTiledBackingLayerWin.cpp:

(WebTiledBackingLayerWin::displayCallback):

  • platform/graphics/mac/WebLayer.mm:

(-[WebLayer drawInContext:]):
(-[WebSimpleLayer drawInContext:]):

  • rendering/PaintPhase.h:
  • rendering/RenderBoxModelObject.cpp:

(WebCore::RenderBoxModelObject::decodingModeForImageDraw):

  • rendering/RenderElement.h:
  • rendering/RenderLayer.cpp:

(WebCore::RenderLayer::paintLayerContents):
(WebCore::RenderLayer::paintForegroundForFragments):

  • rendering/RenderLayerBacking.cpp:

(WebCore::RenderLayerBacking::paintContents):

  • rendering/RenderLayerBacking.h:
  • rendering/RenderLayerCompositor.cpp:

(WebCore::RenderLayerCompositor::paintContents):

  • rendering/RenderLayerCompositor.h:
  • rendering/RenderWidget.cpp:

(WebCore::RenderWidget::paintContents):

  • testing/Internals.cpp:

(WebCore::imageFromImageElement):
(WebCore::bitmapImageFromImageElement):
(WebCore::Internals::imageFrameIndex):
(WebCore::Internals::setImageFrameDecodingDuration):
(WebCore::Internals::resetImageAnimation):
(WebCore::Internals::isImageAnimating):
(WebCore::Internals::setClearDecoderAfterAsyncFrameRequestForTesting):
(WebCore::Internals::imageDecodeCount):
(WebCore::Internals::setLargeImageAsyncDecodingEnabledForTesting):

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

Source/WebKit:

  • Shared/mac/RemoteLayerBackingStore.mm:

(WebKit::RemoteLayerBackingStore::drawInContext):

  • WebProcess/InjectedBundle/DOM/InjectedBundleNodeHandle.cpp:

(WebKit::imageForRect):

  • WebProcess/InjectedBundle/DOM/InjectedBundleRangeHandle.cpp:

(WebKit::InjectedBundleRangeHandle::renderedImage):

  • WebProcess/WebPage/CoordinatedGraphics/CompositingCoordinator.cpp:

(WebKit::CompositingCoordinator::paintContents):

  • WebProcess/WebPage/CoordinatedGraphics/CompositingCoordinator.h:

Source/WebKitLegacy/mac:

  • WebView/WebFrame.mm:

(-[WebFrame _paintBehaviorForDestinationContext:]):
(-[WebFrame _drawRect:contentsOnly:]):

  • WebView/WebHTMLView.mm:

(imageFromRect):

Source/WebKitLegacy/win:

  • FullscreenVideoController.cpp:

(FullscreenVideoController::LayerClient::platformCALayerPaintContents):

  • WebCoreSupport/AcceleratedCompositingContext.cpp:

(AcceleratedCompositingContext::paintContents):

  • WebCoreSupport/AcceleratedCompositingContext.h:

LayoutTests:

To test async image decoding for large images, we have to create the <img>
element dynamically so we can listen to the load and webkitImageFrameReady
events and know reliably when to end the test. But with this patch the async
image decoding for large images will be disabled after the first paint.
That means async image decoding for large images will be disabled always
unless we force the async image decoding till the image is painted for the
first time. We use Internals::setLargeImageAsyncDecodingEnabledForTesting()
to force the async image decoding. So painting an image in this case will
require multiple paints; in all of them the async image decoding will be
enabled. But this is okay because it resembles the case where the <img> is
created from a static <img> tag in the HTML file.

For new tests, where we want to make sure that mutating the DOM will not
cause a flash, async image decoding will be forced till the image is drawn
for the first time. After that the async image decoding is enabled but not
forced.

Disable new tests for WK1 because the async image decoding is always enabled
because tiling does not necessarily exist in WK1 . But eventually the async
image decoding for large images will be always disabled for WK1.

  • fast/images/async-image-background-change-expected.html: Added.
  • fast/images/async-image-background-change.html: Added.
  • fast/images/async-image-background-image-repeated.html:
  • fast/images/async-image-background-image.html:
  • fast/images/async-image-body-background-image.html:
  • fast/images/async-image-multiple-clients-repaint.html:
  • fast/images/async-image-src-change-expected.html: Added.
  • fast/images/async-image-src-change.html: Added.
  • fast/images/resources/green-400x400.png: Added.
  • fast/images/resources/red-100x100.png: Added.
  • fast/images/resources/red-400x400.png: Added.
  • fast/images/sprite-sheet-image-draw.html:
  • http/tests/multipart/multipart-async-image-expected.txt: Added.
  • http/tests/multipart/multipart-async-image.html: Added.
  • platform/ios-wk1/TestExpectations:
  • platform/mac-wk1/TestExpectations:
11:28 AM Changeset in webkit [219875] by Jonathan Bedard
  • 7 edits in trunk/Tools

Replace --runtime with something for both ios-simulator and ios-device
​https://bugs.webkit.org/show_bug.cgi?id=173775
<rdar://problem/32952164>

Reviewed by Aakash Jain.

Add --version flag for both iOS simulator and iOS device.

  • Scripts/webkitpy/layout_tests/run_webkit_tests.py:

(parse_args): Add --version flag and move --no-install.

  • Scripts/webkitpy/port/factory.py:

(platform_options): Move --no-intall to run_webkit_tests.py.

  • Scripts/webkitpy/port/ios.py:

(IOSPort):
(IOSPort._is_valid_ios_version): Check that the provided version string is legal.
(IOSPort.get_option): If accessing the iOS version, check that it is a valid version string.

  • Scripts/webkitpy/port/ios_device.py:

(IOSDevicePort.ios_version): Consult --version flag before checking connected
devices for iOS version.

  • Scripts/webkitpy/port/ios_device_unittest.py:

(IOSDeviceTest.make_port): Set --version option so that we can generate test
expectation paths without devices connected.
(IOSDeviceTest.test_additional_platform_directory): Deleted.
(IOSDeviceTest.test_baseline_searchpath): Deleted.
(IOSDeviceTest.test_expectations_ordering): Deleted.

  • Scripts/webkitpy/port/ios_simulator.py:

(IOSSimulatorPort.simulator_runtime): If no runtime is specified, use the --version
flag to specify a runtime.
(IOSSimulatorPort.ios_version): First check the --version flag, then the --runtime
flag and then use the default runtime.

11:03 AM Changeset in webkit [219874] by BJ Burg
  • 31 edits in trunk/Source

Web Automation: add support for uploading files
​https://bugs.webkit.org/show_bug.cgi?id=174797
<rdar://problem/28485063>

Reviewed by Joseph Pecoraro.

Source/JavaScriptCore:

  • inspector/scripts/generate-inspector-protocol-bindings.py:

(generate_from_specification):
Start generating frontend dispatcher code if the target framework is 'WebKit'.

  • inspector/scripts/codegen/generate_cpp_frontend_dispatcher_implementation.py:

(CppFrontendDispatcherImplementationGenerator.generate_output):
Use a framework include for InspectorFrontendRouter.h since this generated code
will be compiled outside of WebCore.framework.

  • inspector/scripts/tests/all/expected/definitions-with-mac-platform.json-result:
  • inspector/scripts/tests/generic/expected/commands-with-async-attribute.json-result:
  • inspector/scripts/tests/generic/expected/commands-with-optional-call-return-parameters.json-result:
  • inspector/scripts/tests/generic/expected/definitions-with-mac-platform.json-result:
  • inspector/scripts/tests/generic/expected/domain-availability.json-result:
  • inspector/scripts/tests/generic/expected/domains-with-varying-command-sizes.json-result:
  • inspector/scripts/tests/generic/expected/enum-values.json-result:
  • inspector/scripts/tests/generic/expected/events-with-optional-parameters.json-result:
  • inspector/scripts/tests/generic/expected/generate-domains-with-feature-guards.json-result:
  • inspector/scripts/tests/generic/expected/same-type-id-different-domain.json-result:
  • inspector/scripts/tests/generic/expected/shadowed-optional-type-setters.json-result:
  • inspector/scripts/tests/generic/expected/type-declaration-aliased-primitive-type.json-result:
  • inspector/scripts/tests/generic/expected/type-declaration-array-type.json-result:
  • inspector/scripts/tests/generic/expected/type-declaration-enum-type.json-result:
  • inspector/scripts/tests/generic/expected/type-declaration-object-type.json-result:
  • inspector/scripts/tests/generic/expected/type-requiring-runtime-casts.json-result:
  • inspector/scripts/tests/generic/expected/type-with-open-parameters.json-result:
  • inspector/scripts/tests/generic/expected/worker-supported-domains.json-result:
  • inspector/scripts/tests/ios/expected/definitions-with-mac-platform.json-result:
  • inspector/scripts/tests/mac/expected/definitions-with-mac-platform.json-result:

Rebaseline code generator tests.

Source/WebKit:

The general strategy is to have automation clients set the list of files they want
to select ahead of time. Then, the client simulates a click on the <input type="file"> element.
When that causes WebPageProxy to ask the UI delegate to runOpenPanel, it instead
gives WebAutomationSession a chance to select files if the page is under control of
automation. WebAutomationSession validates its file list and selects the files if valid.

  • CMakeLists.txt: Add frontend dispatcher files.
  • DerivedSources.make:

Add frontend dispatcher code generator scripts as dependencies of the generated code.
Add generated frontend dispatcher files to the list of output files.

  • UIProcess/Automation/Automation.json:

Add a command to set the canned list of files that should be selected when prompted.
If no files are selected, then the effect is as if the user had clicked "cancel" in
the file chooser dialog.

Add an event to signal to the WebDriver service that the file selection request has
been either fulfilled or cancelled. This event is necessary because otherwise remotes
cannot distinguish whether selecting the files failed for some reason, or if the
UIProcess has not yet selected the files and sent them to the web content process.
In either case, the input element's "files" attribute would return an empty FileList.

  • UIProcess/Automation/WebAutomationSession.h:
  • UIProcess/Automation/WebAutomationSession.cpp:

(WebKit::WebAutomationSession::WebAutomationSession):
Add the frontend dispatcher for Automation domain. This is named m_domainNotifier since
m_domainDispatcher is already being used for the backend command dispatcher.

(WebKit::WebAutomationSession::handleRunOpenPanel):
Validate that the files-to-select list has valid entries and select the files if so.
If anything goes wrong, pretend to cancel out of the file chooser. Either way, notify the
frontend with an event that the file chooser was dismissed due to selecting files or cancelling.
Automation protocol clients may implement additional strategies to further restrict
the conditions upon which local files can be uploaded.

(WebKit::WebAutomationSession::setFilesToSelectForFileUpload): Added.
Store the list of files as a vector of strings. The file list is always replaced.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::runOpenPanel):
If the page is under automation, delegating the runOpenPanel is likely to hang
because WebDriver cannot interact directly with the file chooser dialog. Instead,
give WebAutomationSession a chance to select files and don't tell the delegate.

  • WebKit.xcodeproj/project.pbxproj:

Add frontend dispatcher files to "Derived Sources" group.

11:02 AM Changeset in webkit [219873] by weinig@apple.com
  • 20 edits
    1 add in trunk/Source/WebCore

[WebIDL] Add support for generating timer bindings
​https://bugs.webkit.org/show_bug.cgi?id=174766

Reviewed by Darin Adler.

Adds a new non-standard type, ScheduledAction, which stands in for the
standard (DOMString or Function). It would be good to move to that in
future, but for now, this allows for forward momentum on removing custom
bindings.

  • WebCore.xcodeproj/project.pbxproj:

Add JSDOMConvertScheduledAction.h.

  • bindings/IDLTypes.h:

Add IDLScheduledAction.

  • bindings/js/JSDOMConvertScheduledAction.h: Added.

(WebCore::Converter<IDLScheduledAction>::convert):
Add conversion from JSValue -> ScheduledAction. This is moved from the old ScheduledAction
create function.

  • bindings/js/JSDOMConvertVariadic.h:

(WebCore::convertVariadicArguments):
(WebCore::Detail::VariadicConverterBase::convert): Deleted.
(WebCore::Detail::VariadicConverterBase<IDLInterface<T>>::convert): Deleted.

  • bindings/js/JSDOMConvertBase.h:
  • bindings/js/JSDOMConvertAny.h:

(WebCore::VariadicConverter<IDLAny>::convert):

  • bindings/js/JSDOMConvertInterface.h:

(WebCore::VariadicConverter<IDLInterface<T>>::convert):
Rename VariadicConverter to VariadicConverterDetails, and remove base class. Rename
VariadicConverterBase to VariadicConverter, and move specializations to the file
containing the base converter for that IDL type

  • bindings/js/JSDOMWindowCustom.cpp:

(WebCore::JSDOMWindow::setTimeout): Deleted.
(WebCore::JSDOMWindow::setInterval): Deleted.

  • bindings/js/JSWorkerGlobalScopeCustom.cpp:

(WebCore::JSWorkerGlobalScope::setTimeout): Deleted.
(WebCore::JSWorkerGlobalScope::setInterval): Deleted.
Remove custom implementations of setTimeout and setInterval.

  • bindings/js/ScheduledAction.cpp:

(WebCore::ScheduledAction::create):
(WebCore::ScheduledAction::ScheduledAction):
(WebCore::ScheduledAction::~ScheduledAction):
(WebCore::ScheduledAction::addArguments):
(WebCore::ScheduledAction::executeFunctionInContext):

  • bindings/js/ScheduledAction.h:

(WebCore::ScheduledAction::ScheduledAction): Deleted.
Rework ScheduledAction. Now has two create functions, one for the function
form, one for the string form. These are now called by the Converter. Also,
rather than extracting the arguments directly from the ExecState, allow the
bindings to work as designed, and have the arguments come in as variadic
arguments to setTimeout/setInterval and get added to the ScheduledAction if
needed. Also, move ContentSecurityPolicy check out of construction, and into
setTimeout/setInterval.

  • bindings/scripts/CodeGenerator.pm:

(IsBuiltinType):
Add ScheduledAction to the builtin list.

  • bindings/scripts/CodeGeneratorJS.pm:

(AddToIncludesForIDLType):
Add the correct include when ScheduledAction is used.

(GenerateParametersCheck):
Remove rule disallowing optional arguments before variadic arguments. That works
just fine.

(GetBaseIDLType):
Add mapping of ScheduledAction -> IDLScheduledAction.

(JSValueToNativeDOMConvertNeedsGlobalObject):
Add ScheduledAction to the list of types that need a global object
for conversion.

  • page/WindowOrWorkerGlobalScope.idl:

Update interface to match spec and add FIXMEs for moving to TimerHandler.

  • page/DOMWindow.cpp:

(WebCore::DOMWindow::setTimeout):
(WebCore::DOMWindow::setInterval):

  • page/DOMWindow.h:
  • workers/WorkerGlobalScope.cpp:

(WebCore::WorkerGlobalScope::setTimeout):
(WebCore::WorkerGlobalScope::setInterval):

  • workers/WorkerGlobalScope.h:

Update for new signatures. WorkerGlobalScope now has to return ExceptionOr<int>
even though it never throws, due to having one IDL file defining these. This is
unfortunate and something we should look at addressing the future.

10:42 AM Changeset in webkit [219872] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

[GTK] Unreviewed test gardening
​https://bugs.webkit.org/show_bug.cgi?id=174820

Unreviewed gardening.

Patch by Charlie Turner <​cturner@igalia.com> on 2017-07-25

  • platform/gtk/TestExpectations:
10:17 AM Changeset in webkit [219871] by achristensen@apple.com
  • 22 edits in trunk/Source/WebKit

Modernize NavigationAction code
​https://bugs.webkit.org/show_bug.cgi?id=174807

Reviewed by Darin Adler.

Mostly just use more move semantics.

  • Platform/IPC/HandleMessage.h:

(IPC::callMemberFunctionImpl):

  • Shared/API/APIURLRequest.h:
  • Shared/NavigationActionData.h:
  • UIProcess/API/APINavigation.cpp:

(API::Navigation::Navigation):

  • UIProcess/API/APINavigation.h:

(API::Navigation::create):

  • UIProcess/API/APINavigationAction.h:
  • UIProcess/API/APINavigationClient.h:

(API::NavigationClient::decidePolicyForNavigationAction):

  • UIProcess/API/APIPolicyClient.h:
  • UIProcess/API/APIUIClient.h:

(API::UIClient::createNewPage):
(API::UIClient::createNewPageAsync):
(API::UIClient::canCreateNewPageAsync):

  • UIProcess/API/C/WKPage.cpp:

(WKPageLoadURLRequest):
(WKPageLoadURLRequestWithUserData):
(WKPageSetPageUIClient):
(WKPageSetPageNavigationClient):

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

(WebKit::NavigationState::NavigationClient::decidePolicyForNavigationAction):

  • UIProcess/Cocoa/UIDelegate.h:
  • UIProcess/Cocoa/UIDelegate.mm:

(WebKit::UIDelegate::UIClient::createNewPageCommon):
(WebKit::UIDelegate::UIClient::createNewPage):
(WebKit::UIDelegate::UIClient::canCreateNewPageAsync):
(WebKit::UIDelegate::UIClient::createNewPageAsync):

  • UIProcess/WebInspectorProxy.cpp:

(WebKit::decidePolicyForNavigationAction):

  • UIProcess/WebNavigationState.cpp:

(WebKit::WebNavigationState::createLoadRequestNavigation):

  • UIProcess/WebNavigationState.h:
  • UIProcess/WebPageProxy.cpp:

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

  • UIProcess/WebPageProxy.h:
10:07 AM Changeset in webkit [219870] by Devin Rousso
  • 11 edits in trunk

Web Inspector: add context menu item for taking a screenshot of a node
​https://bugs.webkit.org/show_bug.cgi?id=174754

Reviewed by Brian Burg.

Source/WebInspectorUI:

  • Localizations/en.lproj/localizedStrings.js:
  • UserInterface/Views/ContextMenuUtilities.js:

(WebInspector.appendContextMenuItemsForDOMNode):
Add "Capture Element Screenshot " context menu item that opens a save dialog with a picture of the node.

  • UserInterface/Base/Main.js:

(WebInspector.saveDataToFile):
Allow the saveData to specify whether the contained content is base64Encoded, instead of
having to go through a Blob to save base64 data.

  • UserInterface/Base/Utilities.js:

Add Number.pad to add leading zeros up to a certain width.

Drive-by: fix failing test for Number.constrain.

  • UserInterface/Models/DOMNode.js:

(WebInspector.DOMNode.prototype.scrollIntoView.resolvedNode.scrollIntoView): Added.
(WebInspector.DOMNode.prototype.scrollIntoView.resolvedNode): Added.
(WebInspector.DOMNode.prototype.scrollIntoView): Added.

  • UserInterface/Views/DOMTreeElement.js:

(WebInspector.DOMTreeElement.prototype._populateNodeContextMenu):
(WebInspector.DOMTreeElement.prototype._scrollIntoView.resolvedNode.scrollIntoView): Deleted.
(WebInspector.DOMTreeElement.prototype._scrollIntoView.resolvedNode): Deleted.
(WebInspector.DOMTreeElement.prototype._scrollIntoView): Deleted.

  • UserInterface/Views/DOMTreeOutline.js:

(WebInspector.DOMTreeOutline.prototype.populateContextMenu):
Drive-by: move "Scroll Into View" context menu command to WebInspector.appendContextMenuItemsForDOMNode
so that it can be triggered from more places than just the DOM tree.

LayoutTests:

  • inspector/unit-tests/number-utilities-expected.txt:
  • inspector/unit-tests/number-utilities.html:

Add test for Number.zeroPad.

Drive-by: Change tests to use expectEqual for better output on failure.

9:49 AM Changeset in webkit [219869] by commit-queue@webkit.org
  • 16 edits in trunk/Source/WebCore

[WinCairo] Fix build with AllInOnes disabled
​https://bugs.webkit.org/show_bug.cgi?id=174784
<rdar://problem/33488914>

Patch by Yoshiaki Jitsukawa <​Yoshiaki.Jitsukawa@sony.com> on 2017-07-25
Reviewed by Alex Christensen.

  • PlatformWin.cmake:

Add RenderThemeWin.cpp to WebCore_SOURCES

  • rendering/RenderingAllInOne.cpp:

Remove #include "RenderThemeWin.cpp"

  • testing/MemoryInfo.h:

Remove #include "JSDomWindow.h"

  • bindings/js/JSDOMGlobalObject.cpp:
  • bindings/js/JSDOMPromiseDeferred.cpp:
  • bindings/js/JSDOMWrapper.cpp:
  • bindings/js/JSImageDataCustom.cpp:
  • bindings/scripts/CodeGeneratorJS.pm:

(GenerateCallWith):

  • dom/ScriptExecutionContext.cpp:
  • inspector/InspectorController.cpp:
  • inspector/InspectorTimelineAgent.cpp:
  • rendering/RenderMediaControls.cpp:
  • rendering/RenderMediaControls.h:
  • rendering/RenderThemeWin.cpp:
  • bindings/scripts/test/JS/JSTestObj.cpp:

Update test results.

6:22 AM Changeset in webkit [219868] by Claudio Saavedra
  • 2 edits
    1 add in trunk/LayoutTests

REGRESSION (r143068): fast/events/clientXY-in-zoom-and-scroll.html failing on GTK, Qt
​https://bugs.webkit.org/show_bug.cgi?id=110090

Unreviewed gardening.

This test only needed the expectation to be updated.

  • platform/gtk/TestExpectations: Remove test below.
  • platform/gtk/fast/events/clientXY-in-zoom-and-scroll-expected.txt: Added.
4:43 AM Changeset in webkit [219867] by Konstantin Tokarev
  • 2 edits in trunk/Source/WebCore

Unreviewed, fix Mac CMake build after r219567

  • PlatformMac.cmake:
4:38 AM Changeset in webkit [219866] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebKit

REGRESSION(r218629): [GTK] ASSERTION FAILED: m_eglDisplay == EGL_NO_DISPLAY in WebCore::PlatformDisplay::~PlatformDisplay
​https://bugs.webkit.org/show_bug.cgi?id=174789

Reviewed by Žan Doberšek.

Before r218629, the wayland nested compositor display was a singleton created on demand by
AcceleratedSurfaceWayland, but now it's created unconditionally on web process creation when a valid display
name is received from the UI process. The side effect of this is that now the shared display is created after
the nested compositor display that is not a singleton anymore. The atexit used by PlatformDisplay to shutdown
egl display assumes that it will be registered by the shared display. Since we switched to on demand AC mode,
the shared display is only used in the web process when entering AC mode, that's why it only crashes when we
have entered AC at least once. We just need to ensure that the shared display is created before the nested
compositor display.

  • WebProcess/gtk/WaylandCompositorDisplay.cpp:

(WebKit::WaylandCompositorDisplay::create): Check we are under wayland before creating the WaylandCompositorDisplay.

4:36 AM Changeset in webkit [219865] by Carlos Garcia Campos
  • 4 edits in trunk/Tools

[GTK] TestWebKitAPI tests are no longer built since the WebKit2 rename
​https://bugs.webkit.org/show_bug.cgi?id=174815

Reviewed by Carlos Alberto Lopez Perez.

Only WebKitGTK specific tests are built, but not WebKit2, WTF nor WebCore ones. We are still checking
ENABLE_WEBKIT2 in the makefile, that was renamed to ENABLE_WEBKIT. Also, all WebKit2 test paths were updated to
use WebKit dir, while the tests are still in WebKit2 directory. This patch also removes
TestWebKitAPIInjectedBundle_PATH that is unused and now makes the configure to fail (it used to be a warning, I
think), probably because of a newer CMake version.

  • CMakeLists.txt:
  • TestWebKitAPI/CMakeLists.txt:
  • TestWebKitAPI/PlatformGTK.cmake:
4:29 AM WebKitGTK/KeepingTheTreeGreen edited by clopez@igalia.com
(diff)
4:28 AM WebKitGTK/KeepingTheTreeGreen edited by clopez@igalia.com
(diff)
4:25 AM WebKitGTK/KeepingTheTreeGreen edited by clopez@igalia.com
(diff)
3:22 AM Changeset in webkit [219864] by Carlos Garcia Campos
  • 2 edits in trunk

Unreviewed. Fix GTK+ WebKit2 library soname after r219488.

The WebKit2 -> WebKit rename shouldn't affect the GTK+ library that is still called libwebkit2gtk, so keep using
WEBKIT2 as the library name in CALCULATE_LIBRARY_VERSIONS_FROM_LIBTOOL_TRIPLE, since that's also what the WebKit
makefile passes to WEBKIT_POPULATE_LIBRARY_VERSION.

  • Source/cmake/OptionsGTK.cmake:

Jul 24, 2017:

11:49 PM Changeset in webkit [219863] by Carlos Garcia Campos
  • 3 edits in trunk/Source/WebKit

[GTK] ASSERTION FAILED: client in WebKit::IconDatabase::setClient
​https://bugs.webkit.org/show_bug.cgi?id=174788

Patch by Michael Catanzaro <Michael Catanzaro> on 2017-07-24
Reviewed by Carlos Garcia Campos.

IconDatabase ensures that setClient is never called with nullptr, but WebKitFaviconDatabase
does exactly that. First, ensure it's safe to unset the client. Next, we can either remove
the assertions from setClient or else automatically unset the client in IconDatabase::close.
I opted to do both.

  • UIProcess/API/glib/IconDatabase.cpp:

(WebKit::IconDatabase::close):
(WebKit::IconDatabase::setIconDataForIconURL):
(WebKit::IconDatabase::setIconURLForPageURL):

  • UIProcess/API/glib/WebKitFaviconDatabase.cpp:

(_WebKitFaviconDatabasePrivate::~_WebKitFaviconDatabasePrivate): Deleted.
(webkitFaviconDatabaseDispose): Deleted.

11:45 PM Changeset in webkit [219862] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebKit

[GTK][WPE] Remove useless conditional in IconDatabase::setIconDataForIconURL
​https://bugs.webkit.org/show_bug.cgi?id=174792

Patch by Michael Catanzaro <Michael Catanzaro> on 2017-07-24
Reviewed by Carlos Garcia Campos.

We have in IconDatabase::setIconDataForIconURL some code in a conditional to check if it's
not being executed on the icon sync thread. But at the top of the function is an assertion
to ensure the code is never executed on the icon sync thread. The condition is therefore
useless and the code should be executed unconditionally.

  • UIProcess/API/glib/IconDatabase.cpp:

(WebKit::IconDatabase::setIconDataForIconURL):

11:39 PM Changeset in webkit [219861] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebKit

[GTK] Icon database error and crash
​https://bugs.webkit.org/show_bug.cgi?id=174760

Reviewed by Michael Catanzaro.

The crash is a debug ASSERT that happens when the IconRecord image is created in one thread and destroyed in
another one. IconDatabase creates and destroys IconRecord objects in both database and main thread. The
IconRecord is destroyed when the icon is no longer retained, and we only release icons when we fail to get the
image data (including pages that don't have a favicon). We can prevent this crash from happening if we ensure we
never create an Image for an IconRecord when the given image data is nullptr.

  • UIProcess/API/glib/IconDatabase.cpp:

(WebKit::IconDatabase::IconRecord::setImageData):

9:36 PM Changeset in webkit [219860] by commit-queue@webkit.org
  • 7 edits in trunk/Tools

Fix BenchmarkRunner to only import selenium when needed
​https://bugs.webkit.org/show_bug.cgi?id=174808

Patch by Matthew Stewart <​matthew_r_stewart@apple.com> on 2017-07-24
Reviewed by Stephanie Lewis.

  • Scripts/webkitpy/benchmark_runner/browser_driver/linux_chrome_driver.py:

(LinuxChromeDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/linux_firefox_driver.py:

(LinuxFirefoxDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_chrome_driver.py:

(OSXChromeDriver.launch_driver):
(OSXChromeCanaryDriver.launch_driver):
(create_chrome_options):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_firefox_driver.py:

(OSXFirefoxDriver.launch_driver):
(OSXFirefoxNightlyDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_safari_driver.py:

(OSXSafariDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/webdriver_benchmark_runner.py:

(WebDriverBenchmarkRunner._run_one_test):

9:35 PM Changeset in webkit [219859] by jmarcell@apple.com
  • 2 edits in branches/safari-604-branch/Source/WebKit

Cherry-pick r219852. rdar://problem/33503476

9:29 PM Changeset in webkit [219858] by Darin Adler
  • 51 edits in trunk

More NeverDestroyed and related cleanup
​https://bugs.webkit.org/show_bug.cgi?id=174745

Reviewed by Alex Christensen.

Source/WebCore:

  • bindings/js/JSCustomElementInterface.cpp:

(WebCore::JSCustomElementInterface::constructElementWithFallback):
Use isNull() rather than comparing against nullAtom() since it's
slightly more efficient.

  • css/MediaQueryEvaluator.cpp:

(WebCore::prefersReducedMotionEvaluate): Use a switch statement for the
three states of forcedPrefersReducedMotionAccessibilityValue. Also update
to use Theme::singleton() instead of platformTheme().

  • dom/QualifiedName.h: Removed some unneeded includes, minor style tweaks,

and use isNull() rather than comparing against nullAtom() since it's
slightly more efficient.

  • html/Autofill.cpp:

(WebCore::AutofillData::createFromHTMLFormControlElement): Use isNull()
rather than comparing against nullAtom() since it's slightly more efficient.

  • html/HTMLDocument.cpp:

(WebCore::addLocalNameToSet): Deleted.
(WebCore::createHtmlCaseInsensitiveAttributesSet): Deleted.
(WebCore::HTMLDocument::isCaseSensitiveAttribute): Moved the case-sensitive
attribute set in here and made it use makeNeverDestroyed.

  • html/HTMLNameCollection.cpp:

(WebCore::WindowNameCollection::elementMatchesIfNameAttributeMatch):
Formatted vertically and sorted so it's easier to see what it does.
(WebCore::WindowNameCollection::elementMatches): Use boolean logic instead
of multiple return statements to make this a little more straightforward.
(WebCore::isObjectElementForDocumentNameCollection): Added. Calls the new
isExposed function. The old code called this isDocNamedItem.
(WebCore::DocumentNameCollection::elementMatchesIfIdAttributeMatch):
Use isObjectElementForDocumentNameCollection to clean up logic. Also
sorted classes.
(WebCore::DocumentNameCollection::elementMatchesIfNameAttributeMatch):
Use isObjectElementForDocumentNameCollection to clean up logic. Also
sorted classes and formatted vertically.
(WebCore::DocumentNameCollection::elementMatches): Write in terms of
above functions for clarity. Efficiency is roughly the same.

  • html/HTMLObjectElement.cpp:

(WebCore::HTMLObjectElement::HTMLObjectElement): Initialize data members
in class definition instead of here.
(WebCore::HTMLObjectElement::childrenChanged): Update for name change;
function is now called updateExposedState rather than updateDocNamedItem.
Also use m_useFallbackContent directly instead of through a function.
(WebCore::HTMLObjectElement::renderFallbackContent): Ditto.
(WebCore::preventsParentObjectFromExposure): Added. Replaces the old
isRecognizedTagName function, and does almost the same thing, but the
argument type is different and the param tag is now one of the tags
exempted here. Also broke up into two overloads so we can handle the
Element, Text, and other Node cases.
(WebCore::shouldBeExposed): Added. Helper for the updateExposedState
function that handles the logic of when the element should be exposed,
separating it from what to do when the state changes.
(WebCore::HTMLObjectElement::updateExposedState): Renamed from
updateDocNamedItem.

  • html/HTMLObjectElement.h: Renamed isDocNamedItem -> isExposed,

m_docNamedItem -> m_isExposed, updateDocNamedItem -> updateExposedState,
stopped using bit fields for the two boolean data members, and
initialized those data members here in the class definition.

  • platform/Theme.cpp:

(WebCore::Theme::baselinePositionAdjustment): Moved here from header.
(WebCore::Theme::controlFont): Ditto.
(WebCore::Theme::controlSize): Ditto.
(WebCore::Theme::minimumControlSize): Ditto.
(WebCore::Theme::controlRequiresPreWhiteSpace): Ditto.
(WebCore::Theme::paint): Ditto.
(WebCore::Theme::inflateControlPaintRect): Ditto.
(WebCore::Theme::userPrefersReducedMotion): Ditto.
(WebCore::Theme::controlBorder): Fixed switch statement indentation.
(WebCore::Theme::controlPadding): Ditto.

  • platform/Theme.h: Removed the many unnecessary includes. The one

I was really after here was FontCascade.h. Replaced the platformTheme
function with a singleton function, following our normal style. Also
changed the return type to a reference rather than a pointer.
Removed many unused function and data members.

  • platform/cocoa/ThemeCocoa.h: Tightened up implementation, removing

unnecessary members and making overrides private.

  • platform/cocoa/ThemeCocoa.mm: Moved include in here.
  • platform/gamepad/glib/GamepadsGlib.cpp:

(WebCore::sampleGamepads): Use NeverDestroyed instead of
DEPRECATED_DEFINE_STATIC_LOCAL.

  • platform/graphics/FontCascade.cpp:

(WebCore::useBackslashAsYenSignForFamily): Build the set in a cleaner
way and use a NeverDestroyed set instead of a heap-allocated one.
(WebCore::FontCascade::FontCascade): Moved initialization of data members
to the class definition.

  • platform/graphics/FontCascade.h: Initialize GlyphOverflow and

FonstCascade data members in the class definition. Remove the constructor
for GlyphOverflow entirely. Added FIXME about the peculiar != operator we
defined for GlyphOverflow.

  • platform/graphics/NamedImageGeneratedImage.cpp:

(WebCore::NamedImageGeneratedImage::draw): Use Theme::singleton() instead
of platformTheme().
(WebCore::NamedImageGeneratedImage::drawPattern): Ditto.

  • platform/graphics/avfoundation/cf/MediaPlayerPrivateAVFoundationCF.cpp:

(WebCore::CMTimeRangeStartKey): Use CFSTR directly instead of using
DEPRECATED_DEFINE_STATIC_LOCAL.
(WebCore::CMTimeRangeDurationKey): Ditto.
(WebCore::CACFContextNeedsFlushNotification): Ditto.

  • platform/graphics/ca/win/LayerChangesFlusher.cpp:

(WebCore::LayerChangesFlusher::singleton): Use NeverDestroyed instead of
DEPRECATED_DEFINE_STATIC_LOCAL.

  • platform/graphics/ca/win/LayerChangesFlusher.h: Ditto.
  • platform/graphics/cg/ImageSourceCGWin.cpp:

(WebCore::preferredExtensionForImageSourceType): Use HashMap constructor
instead of writing out a lot of add calls.

  • platform/graphics/harfbuzz/HarfBuzzFace.cpp:

(WebCore::harfBuzzFaceCache): Use NeverDestroyed instead of
DEPRECATED_DEFINE_STATIC_LOCAL.

  • platform/graphics/win/FontCacheWin.cpp:

(WebCore::FontCache::lastResortFallbackFont): Ditto.

  • platform/ios/ThemeIOS.h: Tightened up implementation, removing

unnecessary members and making overrides private.

  • platform/ios/ThemeIOS.mm:

(WebCore::Theme::singleton): Renamed from platformTheme.

  • platform/mac/ThemeMac.h: Tightened up implementation, removing

unnecessary members and making overrides private. Removed unneeded
WTF_MAKE_FAST_ALLOCATED since this is never allocated on the heap.
Removed unused -[NSFont webCoreFamilyName] method.

  • platform/mac/ThemeMac.mm:

(-[NSFont webCoreFamilyName]): Deleted.
(WebCore::Theme::singleton): Renamed from platformTheme.

  • platform/mediastream/mac/MockRealtimeAudioSourceMac.h:

Removed unneeded include of FontCascade.h.

  • platform/mediastream/mac/MockRealtimeVideoSourceMac.h:

Ditto.

  • platform/mock/MockRealtimeAudioSource.h: Ditto.
  • platform/network/win/DownloadBundleWin.cpp:

(WebCore::DownloadBundle::fileExtension): Use NeverDestroyed
instead of DEPRECATED_DEFINE_STATIC_LOCAL.

  • platform/win/ClipboardUtilitiesWin.cpp:

(WebCore::replaceNewlinesWithWindowsStyleNewlines): Use
appendLiteral instead of DEPRECATED_DEFINE_STATIC_LOCAL.

  • platform/win/FileSystemWin.cpp:

(WebCore::bundleName): Use NeverDestroyed instead of
DEPRECATED_DEFINE_STATIC_LOCAL, and used a lambda to initialize
rather than a boolean to control one-time code.
(WebCore::storageDirectory): Removed unneeded use of
DEPRECATED_DEFINE_STATIC_LOCAL for a prefix we are appending to
a String. The + operator can handle a literal just as efficiently
without first storing it in a String.

  • platform/wpe/ThemeWPE.cpp:

(WebCore::Theme::singleton): Renamed from platformTheme.

  • platform/wpe/ThemeWPE.h: Tightened up implementation, removing

unnecessary members and making overrides private.

  • rendering/RenderTheme.cpp:

(WebCore::RenderTheme::RenderTheme): No need to initialize m_theme.
(WebCore::RenderTheme::adjustStyle): Use Theme::singleton() instead
of accessing it through m_theme.
(WebCore::RenderTheme::paint): Ditto.
(WebCore::RenderTheme::baselinePosition): Ditto.
(WebCore::RenderTheme::adjustRepaintRect): Ditto.
(WebCore::RenderTheme::setCustomFocusRingColor): Ditto.
(WebCore::RenderTheme::platformTapHighlightColor): Ditto.

  • rendering/RenderTheme.h: Removed unneded m_theme and unneeded

inclusion of Theme.h.

  • rendering/line/LineBreaker.h: Removed unneeded include of

FontCascade.h.

Source/WebKitLegacy/mac:

  • DOM/DOMUIKitExtensions.mm: Removed a lot of unneeded "using"

for names that aren't even used in this file.
(-[DOMNode boundingBoxes]): Use @[] syntax instead of
-[NSArray arrayWithObject:].
(-[DOMNode absoluteQuads]): Ditto.
(-[DOMNode findExplodedTextNodeAtPoint:]): Use auto* instead
of writing out various types.
(-[DOMHTMLAreaElement boundingBoxesWithOwner:]): Use @[]
syntax instea of -[NSArray arrayWithObject:].
(-[DOMHTMLAreaElement absoluteQuadsWithOwner:]): Ditto.
(-[DOMHTMLImageElement dataRepresentation:]): Use auto* instead
of writing out various types.
(-[DOMHTMLImageElement mimeType]): Ditto.

Source/WebKitLegacy/win:

  • Plugins/PluginDatabase.cpp:

(WebCore::persistentPluginMetadataCachePath): Use NeverDestroyed
instead of DEPRECATED_DEFINE_STATIC_LOCAL.

  • Plugins/PluginMainThreadScheduler.cpp:

(WebCore::PluginMainThreadScheduler::scheduler): Ditto.

  • Plugins/PluginMainThreadScheduler.h: Ditto.
  • WebCoreSupport/WebPlatformStrategies.cpp:

(WebPlatformStrategies::initialize): Ditto.

  • WebCoreSupport/WebPlatformStrategies.h: Ditto.
  • WebHistory.cpp:

(sharedHistoryStorage): Ditto.

  • WebLocalizableStrings.cpp:

(mainBundleLocStrings): Ditto.
(frameworkLocStringsMutex): Ditto.
(frameworkLocStrings): Ditto.

  • WebView.cpp:

(WebView::standardUserAgentWithApplicationName): Ditto.
Also changed to keep one global string with the prefix of the user agent string
instead of keeping two globals with pieces of that prefix.

Tools:

  • DumpRenderTree/JavaScriptThreading.cpp:

(javaScriptThreadsMutex): Use NeverDestroyed instead of
DEPRECATED_DEFINE_STATIC_LOCAL.
(javaScriptThreads): Ditto.

9:10 PM Changeset in webkit [219857] by commit-queue@webkit.org
  • 4 edits
    2 adds in trunk/Tools

Subclass Benchmark Runner script for WebDriver support
​https://bugs.webkit.org/show_bug.cgi?id=174390

Patch by Matthew Stewart <​matthew_r_stewart@apple.com> on 2017-07-24
Reviewed by Dewei Zhu, Stephanie Lewis.

Add WebServerBenchmarkRunner and WebDriverBenchmarkRunner subclasses.

  • Scripts/webkitpy/benchmark_runner/benchmark_builder.py:

(BenchmarkBuilder.init):

  • Scripts/webkitpy/benchmark_runner/benchmark_runner.py:

(BenchmarkRunner.init):
(BenchmarkRunner._find_plan_file):
(BenchmarkRunner._run_one_test.and):
(BenchmarkRunner.execute):
(BenchmarkRunner._get_result): Deleted.

  • Scripts/webkitpy/benchmark_runner/run_benchmark.py:

(parse_args):
(start):

  • Scripts/webkitpy/benchmark_runner/webdriver_benchmark_runner.py: Added.

(WebDriverBenchmarkRunner):
(WebDriverBenchmarkRunner._get_result):
(WebDriverBenchmarkRunner._run_one_test):

  • Scripts/webkitpy/benchmark_runner/webserver_benchmark_runner.py: Added.

(WebServerBenchmarkRunner):
(WebServerBenchmarkRunner.init):
(WebServerBenchmarkRunner._get_result):
(WebServerBenchmarkRunner._run_one_test):

8:59 PM Changeset in webkit [219856] by Chris Dumez
  • 184 edits in trunk/Source

ExceptionCode enumeration should use modern exception names
​https://bugs.webkit.org/show_bug.cgi?id=174794

Reviewed by Geoffrey Garen.

Source/WebCore:

ExceptionCode enumeration should use modern exception names consistently.
This makes the code nicer and closer to the text of the specification.

Source/WebKitLegacy/mac:

  • DOM/ExceptionHandlers.mm:

(raiseNotSupportedErrorException):

8:07 PM Changeset in webkit [219855] by Chris Dumez
  • 3 edits in trunk/Source/WebKit

[WK2][Cocoa] Allow overriding the ITP data removal internal using a default
​https://bugs.webkit.org/show_bug.cgi?id=174802
<rdar://problem/33497898>

Reviewed by Sam Weinig.

Allow overriding the ITP data removal internal using a default to facilitate testing.
Can be used like so:
defaults write -g ResourceLoadStatisticsMinimumTimeBetweenDataRecordsRemoval 60

Will cause us to write every 60 seconds instead of 3600.

  • Shared/WebPreferencesDefinitions.h:
  • UIProcess/Cocoa/WebResourceLoadStatisticsStoreCocoa.mm:

(WebKit::WebResourceLoadStatisticsStore::registerUserDefaultsIfNeeded):

6:58 PM Changeset in webkit [219854] by commit-queue@webkit.org
  • 14 edits
    10 copies
    11 moves
    2 adds in trunk/Tools

Add BenchmarkRunner patches with WebDriver support for each testing plan
​https://bugs.webkit.org/show_bug.cgi?id=174443

Patch by Matthew Stewart <​matthew_r_stewart@apple.com> on 2017-07-24
Reviewed by Dewei Zhu, Stephanie Lewis.

  • Scripts/webkitpy/benchmark_runner/benchmark_builder.py:

(BenchmarkBuilder.enter):

  • Scripts/webkitpy/benchmark_runner/data/patches/webdriver/ARES-6.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webdriver/ContentAnimation.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webdriver/Dromaeo.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webdriver/JSBench.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webdriver/JetStream.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webdriver/Kraken.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webdriver/MotionMark.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webdriver/Octane.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webdriver/Speedometer.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webdriver/SunSpider.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/ARES-6.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/Animometer.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/ContentAnimation.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/Dromaeo.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/JSBench.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/JetStream.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/Kraken.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/MotionMark.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/Octane.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/Speedometer.patch:
  • Scripts/webkitpy/benchmark_runner/data/patches/webserver/SunSpider.patch:
  • Scripts/webkitpy/benchmark_runner/data/plans/ares6.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/content-animation.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/dromaeo-cssquery.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/dromaeo-dom.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/dromaeo-jslib.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/jetstream.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/jsbench.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/kraken.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/motionmark.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/octane.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/speedometer.plan:
  • Scripts/webkitpy/benchmark_runner/data/plans/sunspider.plan:
6:42 PM Changeset in webkit [219853] by commit-queue@webkit.org
  • 10 edits in trunk/Tools

Add functions to autoinstaller needed for Benchmark Runner script
​https://bugs.webkit.org/show_bug.cgi?id=174331

Patch by Matthew Stewart <​matthew_r_stewart@apple.com> on 2017-07-24
Reviewed by Dean Johnson, Dewei Zhu, Stephanie Lewis.

Adds autoinstaller functions to install selenium and webdriver binaries
which will be used by the browser driver part of Benchmark Runner.

  • Scripts/webkitpy/benchmark_runner/browser_driver/browser_driver.py:

(BrowserDriver.restore_env):
(BrowserDriver):
(BrowserDriver.get_webdriver_binary_path):

  • Scripts/webkitpy/benchmark_runner/browser_driver/linux_chrome_driver.py:

(LinuxChromeDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/linux_firefox_driver.py:

(LinuxFirefoxDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_chrome_driver.py:

(OSXChromeDriver.launch_driver):
(OSXChromeCanaryDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_firefox_driver.py:

(OSXFirefoxDriver.launch_driver):
(OSXFirefoxNightlyDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_safari_driver.py:
  • Scripts/webkitpy/benchmark_runner/utils.py:

(get_driver_binary_path):

  • Scripts/webkitpy/common/system/autoinstall.py:

(AutoInstaller._unzip):

  • Scripts/webkitpy/thirdparty/init.py:

(AutoinstallImportHook.find_module):
(AutoinstallImportHook._install_selenium):
(AutoinstallImportHook):
(AutoinstallImportHook._install_chromedriver):
(AutoinstallImportHook._install_geckodriver):
(AutoinstallImportHook.get_latest_pypi_url):
(AutoinstallImportHook.install_binary):
(autoinstall_everything):
(get_driver_filename):
(get_os_info):

6:17 PM Changeset in webkit [219852] by Simon Fraser
  • 2 edits in trunk/Source/WebKit

[iOS WK2] Crash under PageOverlayController::uninstallPageOverlay when doing multiple finds
​https://bugs.webkit.org/show_bug.cgi?id=174806
rdar://problem/33501664

Reviewed by Tim Horton.

Calling TextIndicator::createWithSelectionInFrame can trigger layout via VisibleSelection code,
which can re-enter FindController::updateFindIndicator which has by now removed m_findIndicatorOverlay
from pageOverlayController's map, but not deleted it. The second call to uninstallPageOverlay()
thus crashes at m_overlayGraphicsLayers.take(&overlay)->removeFromParent().

Fix by nulling out m_findIndicatorOverlay as soon as we uninstall it.

  • WebProcess/WebPage/ios/FindControllerIOS.mm:

(WebKit::FindController::updateFindIndicator):

5:51 PM Changeset in webkit [219851] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Settings: Some controls overlay settings header
​https://bugs.webkit.org/show_bug.cgi?id=174804

Patch by Nikita Vasilyev <​me@elv1s.ru> on 2017-07-24
Reviewed by Matt Baker.

  • UserInterface/Views/SettingsTabContentView.css:

(.content-view.settings .navigation-bar):

5:40 PM Changeset in webkit [219850] by commit-queue@webkit.org
  • 11 edits in trunk/Tools

Add WebDriver support in browser driver part of BenchmarkRunner
​https://bugs.webkit.org/show_bug.cgi?id=174445

Patch by Matthew Stewart <​matthew_r_stewart@apple.com> on 2017-07-24
Reviewed by Stephanie Lewis.

Adds a launch_driver function to each BrowserDriver subclass. This
function sets up the arguments for the webdriver and launches the
webdriver specific to that browser.

  • Scripts/webkitpy/benchmark_runner/browser_driver/browser_driver.py:

(BrowserDriver.launch_webdriver):
(BrowserDriver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/linux_browser_driver.py:

(LinuxBrowserDriver.launch_webdriver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/linux_chrome_driver.py:

(LinuxChromeDriver.launch_url):
(LinuxChromeDriver):
(LinuxChromeDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/linux_epiphany_driver.py:

(EpiphanyBrowserDriver.launch_url):
(EpiphanyBrowserDriver):
(EpiphanyBrowserDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/linux_firefox_driver.py:

(LinuxFirefoxDriver):
(LinuxFirefoxDriver.launch_url):
(LinuxFirefoxDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/linux_minibrowsergtk_driver.py:

(GTKMiniBrowserDriver.launch_url):
(GTKMiniBrowserDriver):
(GTKMiniBrowserDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_browser_driver.py:

(OSXBrowserDriver._launch_webdriver):
(OSXBrowserDriver):
(OSXBrowserDriver._screen_size):
(OSXBrowserDriver._insert_url):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_chrome_driver.py:

(OSXChromeDriver):
(OSXChromeDriver.launch_url):
(OSXChromeDriver.launch_driver):
(OSXChromeCanaryDriver):
(OSXChromeCanaryDriver.launch_url):
(OSXChromeCanaryDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_firefox_driver.py:

(OSXFirefoxDriver):
(OSXFirefoxDriver.launch_url):
(OSXFirefoxDriver.launch_driver):
(OSXFirefoxNightlyDriver):
(OSXFirefoxNightlyDriver.launch_url):
(OSXFirefoxNightlyDriver.launch_driver):

  • Scripts/webkitpy/benchmark_runner/browser_driver/osx_safari_driver.py:

(OSXSafariDriver.launch_driver):

4:54 PM Changeset in webkit [219849] by pvollan@apple.com
  • 2 edits in trunk/LayoutTests

[Win] The tests fast/text/line-height-minimumFontSize*.html are failing.
​https://bugs.webkit.org/show_bug.cgi?id=174801

Unreviewed test gardening, update test expectations for failing tests.

  • platform/win/TestExpectations:
4:52 PM Changeset in webkit [219848] by aestes@apple.com
  • 2 edits in trunk/Source/WebCore

Consider implementing W3C Payment Request API
​https://bugs.webkit.org/show_bug.cgi?id=174796

Reviewed by Daniel Bates.

Add Payment Request to features.json as "Under Consideration".

  • features.json:
4:41 PM Changeset in webkit [219847] by Matt Lewis
  • 2 edits in trunk/LayoutTests

Marked tests as failures.
<rdar://problem/33437597>

Unreviewed test gardening.

  • platform/mac-highsierra/TestExpectations:
4:12 PM Changeset in webkit [219846] by Matt Lewis
  • 2 edits in trunk/LayoutTests

Fixed typo in test expectations.
​https://bugs.webkit.org/show_bug.cgi?id=173010

Unreviewed gardening.

  • platform/ios-simulator-wk2/TestExpectations:
3:50 PM Changeset in webkit [219845] by Chris Dumez
  • 3 edits in trunk/Source/WebKit

Fix lifetime management issue in ResourceLoadStatisticsPersistentStorage::scheduleOrWriteMemoryStore()
​https://bugs.webkit.org/show_bug.cgi?id=174790

Reviewed by Brady Eidson.

Fix lifetime management issue in ResourceLoadStatisticsPersistentStorage::scheduleOrWriteMemoryStore().
WorkQueue::dispatchAfter() keeps the WorkQueue alive because its implementation keeps a strong ref to
|this|. As a result, the lambda passed to dispatchAfter(), which calls writeMemoryStoreToDisk(), can
get executed after the store is gone.

To address the issue, we now use a RunLoop::Timer to schedule the write, instead of a
WorkQueue::dispatchAfter() call. This way, we are guaranteed that the callback will not get called
after the store has been destroyed.

  • UIProcess/Storage/ResourceLoadStatisticsPersistentStorage.cpp:

(WebKit::ResourceLoadStatisticsPersistentStorage::ResourceLoadStatisticsPersistentStorage):
(WebKit::ResourceLoadStatisticsPersistentStorage::asyncWriteTimerFired):
(WebKit::ResourceLoadStatisticsPersistentStorage::scheduleOrWriteMemoryStore):
(WebKit::ResourceLoadStatisticsPersistentStorage::finishAllPendingWorkSynchronously):
(WebKit::ResourceLoadStatisticsPersistentStorage::ref):
(WebKit::ResourceLoadStatisticsPersistentStorage::deref):

  • UIProcess/Storage/ResourceLoadStatisticsPersistentStorage.h:
3:42 PM Changeset in webkit [219844] by pvollan@apple.com
  • 2 edits in trunk/LayoutTests

[Win] Async image tests are failing.
​https://bugs.webkit.org/show_bug.cgi?id=174653

Unreviewed test gardening, update test expectations for async image tests.

  • platform/win/TestExpectations:
3:29 PM Changeset in webkit [219843] by Simon Fraser
  • 2 edits in trunk/LayoutTests

Rebaseline this test. The previous result was generated using an older iOS SDK.

  • fast/events/ios/no-touch-events-when-stopping-momentum-scroll-in-mainframe-expected.txt:
3:26 PM Changeset in webkit [219842] by pvollan@apple.com
  • 2 edits in trunk/LayoutTests

Mark Skia font test as image only failure on Windows.

Unreviewed test gardening.

  • platform/win/TestExpectations:
2:57 PM Changeset in webkit [219841] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

LayoutTest userscripts/user-script-audio-document.html failing on High Sierra
​https://bugs.webkit.org/show_bug.cgi?id=174726
rdar://problem/30059973

Patch by Jeremy Jones <​jeremyj@apple.com> on 2017-07-24
Reviewed by Alexey Proskuryakov.

Underlying issue has been resolved.
Re-enable test: userscripts/user-script-audio-document.html

  • platform/mac-highsierra/TestExpectations:
2:57 PM Changeset in webkit [219840] by Konstantin Tokarev
  • 2 edits in trunk/Source/WebCore

Unreviewed attempt to fix Mac cmake build

  • PlatformMac.cmake:
2:53 PM Changeset in webkit [219839] by Simon Fraser
  • 2 edits in trunk/LayoutTests

Fix typo (Timout -> Timeout).

  • platform/mac-wk2/TestExpectations:
2:42 PM Changeset in webkit [219838] by Chris Dumez
  • 3 edits in trunk/Source/WebKit

Unreviewed, rolling out r219828.

Causes debug assertions to be hit on iOS

Reverted changeset:

"Fix lifetime management issue in
ResourceLoadStatisticsPersistentStorage::scheduleOrWriteMemoryStore()"
​https://bugs.webkit.org/show_bug.cgi?id=174790
​http://trac.webkit.org/changeset/219828

2:21 PM Changeset in webkit [219837] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

5 wav media layoutTest timeouts on High Sierra
​https://bugs.webkit.org/show_bug.cgi?id=174725
rdar://problem/30182218

Patch by Jeremy Jones <​jeremyj@apple.com> on 2017-07-24
Reviewed by Andreas Kling.

Underlying issue is fixed. Re-enable tests on High Sierra.

media/audio-controls-timeline-in-media-document.html
media/audio-data-url.html
media/media-document-audio-controls-visible.html
media/media-document-audio-repaint.html
media/media-document-audio-size.html

  • platform/mac-highsierra/TestExpectations:
1:58 PM Changeset in webkit [219836] by Wenson Hsieh
  • 3 edits
    1 copy in trunk/Tools

[Mac WK2] Add an API test to cover r219765 (null dereference in [WKWebView dealloc])
​https://bugs.webkit.org/show_bug.cgi?id=174793

Reviewed by Tim Horton.

Adds a new unit test that invokes some asynchronous NSTextInputClient SPI methods as the web view is tearing
down. Without r219765, this test will dereference null and crash.

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/TestWebKitAPI/mac/NSTextInputClientSPI.h: Added.
  • TestWebKitAPI/Tests/mac/WKWebViewSelectionTests.mm:

(TEST):

1:20 PM Changeset in webkit [219835] by jmarcell@apple.com
  • 7 edits in branches/safari-604-branch/Source

Versioning.

1:15 PM Changeset in webkit [219834] by jmarcell@apple.com
  • 1 copy in tags/Safari-604.1.33

Tag Safari-604.1.33.

12:16 PM Changeset in webkit [219833] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebDriver

WebDriver: rename m_browsingContext as m_currentBrowsingContext in Session
​https://bugs.webkit.org/show_bug.cgi?id=174783

Patch by Carlos Garcia Campos <​cgarcia@igalia.com> on 2017-07-24
Reviewed by Brian Burg.

We have m_toplevelBrowsingContext and m_browsingContext, which is confusing. m_browsingContext is actually the
current browsing context, and the spec also refers to it as the current browsing context, so better use
m_currentBrowsingContext.

  • Session.cpp:

(WebDriver::Session::switchToTopLevelBrowsingContext):
(WebDriver::Session::switchToBrowsingContext):
(WebDriver::Session::switchToFrame):
(WebDriver::Session::switchToParentFrame):
(WebDriver::Session::computeElementLayout):
(WebDriver::Session::findElements):
(WebDriver::Session::isElementSelected):
(WebDriver::Session::getElementText):
(WebDriver::Session::getElementTagName):
(WebDriver::Session::isElementEnabled):
(WebDriver::Session::isElementDisplayed):
(WebDriver::Session::getElementAttribute):
(WebDriver::Session::waitForNavigationToComplete):
(WebDriver::Session::elementClear):
(WebDriver::Session::elementSendKeys):
(WebDriver::Session::elementSubmit):
(WebDriver::Session::executeScript):

  • Session.h:
11:55 AM Changeset in webkit [219832] by Michael Catanzaro
  • 2 edits in trunk/Source/WebKit

[GTK][WPE] Stop using AutodrainedPool in IconDatabase
​https://bugs.webkit.org/show_bug.cgi?id=174791

Reviewed by Alex Christensen.

AutodrainedPool seems to be some NS garbage collection thing that has no implementation on
other platforms. Now that IconDatabase is used only by GTK and WPE, we can stop using it.

  • UIProcess/API/glib/IconDatabase.cpp:

(WebKit::IconDatabase::setIconDataForIconURL):
(WebKit::IconDatabase::setIconURLForPageURL):
(WebKit::IconDatabase::iconDatabaseSyncThread):
(WebKit::IconDatabase::performURLImport):
(WebKit::IconDatabase::readFromDatabase):

11:42 AM Changeset in webkit [219831] by Chris Dumez
  • 200 edits in trunk/Source

Make ExceptionCode a proper enumeration
​https://bugs.webkit.org/show_bug.cgi?id=174771

Reviewed by Sam Weinig.

Make ExceptionCode a proper enumeration instead of a typedef to uint8_t.

Source/WebCore:

  • Modules/fetch/FetchBody.h:

(WebCore::FetchBody::formData):

  • Modules/indexeddb/shared/IDBError.h:

(WebCore::IDBError::encode):
(WebCore::IDBError::decode):

  • bindings/js/JSSubtleCryptoCustom.cpp:

(WebCore::rejectWithException):

  • dom/Exception.h:
  • dom/ExceptionCode.h:
  • fileapi/FileReaderSync.cpp:

(WebCore::errorCodeToExceptionCode):

  • html/canvas/WebGLRenderingContextBase.cpp:

(WebCore::WebGLRenderingContextBase::texSubImage2D):
(WebCore::WebGLRenderingContextBase::texImage2D):

  • xml/XMLHttpRequest.cpp:

(WebCore::XMLHttpRequest::createRequest):

  • xml/XMLHttpRequest.h:

Source/WebKitLegacy/win:

  • DOMCoreClasses.cpp:

(DOMDocument::createEvent):

  • WebView.cpp:

(WebView::prepareCandidateWindow):

11:34 AM Changeset in webkit [219830] by commit-queue@webkit.org
  • 5 edits in trunk/Tools

[Win] Implement Authentication dialog in MiniBrowser
​https://bugs.webkit.org/show_bug.cgi?id=174662

Patch by Basuke Suzuki <Basuke Suzuki> on 2017-07-24
Reviewed by Alex Christensen.

  • MiniBrowser/win/Common.cpp:

(CustomUserAgent):
(DisplayAuthDialog):
(AuthDialogProc):

  • MiniBrowser/win/MiniBrowserLib.rc:
  • MiniBrowser/win/MiniBrowserLibResource.h:
  • MiniBrowser/win/ResourceLoadDelegate.cpp:

(ResourceLoadDelegate::didReceiveAuthenticationChallenge):

11:07 AM Changeset in webkit [219829] by Simon Fraser
  • 12 edits
    2 adds in trunk

Change "client" coordinates back to match scrolling coordinates
​https://bugs.webkit.org/show_bug.cgi?id=174734
rdar://problem/33336930

Reviewed by Tim Horton.

Baidu.com (which uses jQuery), and various iOS apps make the assumption that getBoundingClientRect() returns a rectangle
that can be used to set or compare with the scroll position. With visual viewports, that assumption is no longer valid
when the page is zoomed, or when the keyboard has caused the visual viewport to detach from the layout viewport.

At this point the compatibility cost of shipping layout viewport-based client rects seems higher than the gain, so revert
to the shipping behavior. This reverts r216803, and will re-introduce bugs that occurred on zoomed pages on macOS,
many of which are noted in webkit.org/b/170981.

  • page/FrameView.cpp:

(WebCore::FrameView::documentToClientOffset):

LayoutTests:
Revert "client" coordinates to be in scrolling coordinates
​https://bugs.webkit.org/show_bug.cgi?id=174734
rdar://problem/33336930

Reviewed by Tim Horton.

Rebaseline.

  • fast/events/ios/no-touch-events-when-stopping-momentum-scroll-in-mainframe-expected.txt:
  • fast/visual-viewport/client-coordinates-relative-to-layout-viewport-expected.txt:
  • fast/visual-viewport/client-rects-relative-to-layout-viewport-expected.txt:
  • fast/visual-viewport/zoomed-fixed-expected.txt:
  • fast/visual-viewport/zoomed-fixed-header-and-footer-expected.txt:
  • fast/zooming/client-rect-in-fixed-zoomed-expected.txt:
  • platform/ios-wk2/fast/visual-viewport/client-rects-relative-to-layout-viewport-expected.txt:
  • platform/ios-wk2/imported/w3c/web-platform-tests/cssom-view/elementFromPoint-expected.txt: Added.
  • platform/ios/fast/visual-viewport/zoomed-fixed-expected.txt:
  • platform/ios/fast/visual-viewport/zoomed-fixed-header-and-footer-expected.txt:
10:43 AM Changeset in webkit [219828] by Chris Dumez
  • 3 edits in trunk/Source/WebKit

Fix lifetime management issue in ResourceLoadStatisticsPersistentStorage::scheduleOrWriteMemoryStore()
​https://bugs.webkit.org/show_bug.cgi?id=174790

Reviewed by Brady Eidson.

Fix lifetime management issue in ResourceLoadStatisticsPersistentStorage::scheduleOrWriteMemoryStore().
WorkQueue::dispatchAfter() keeps the WorkQueue alive because its implementation keeps a strong ref to
|this|. As a result, the lambda passed to dispatchAfter(), which calls writeMemoryStoreToDisk(), can
get executed after the store is gone.

  • UIProcess/Storage/ResourceLoadStatisticsPersistentStorage.cpp:

(WebKit::ResourceLoadStatisticsPersistentStorage::ResourceLoadStatisticsPersistentStorage):
(WebKit::ResourceLoadStatisticsPersistentStorage::scheduleOrWriteMemoryStore):

  • UIProcess/Storage/ResourceLoadStatisticsPersistentStorage.h:

(WebKit::ResourceLoadStatisticsPersistentStorage::createWeakPtr):

10:31 AM Changeset in webkit [219827] by Jonathan Bedard
  • 3 edits in trunk/Tools

Handle case where line_numbers is None instead of an array of line numbers
​https://bugs.webkit.org/show_bug.cgi?id=174775

Reviewed by Aakash Jain.

Added files set line_numbers to None in process_file.

  • Scripts/webkitpy/style/filereader.py:

(TextFileReader.process_file): Handle case where line_number is set to None.

  • Scripts/webkitpy/style/main_unittest.py:

(ExpectationLinterInStyleCheckerTest.test_linter_added_file_with_error): Test case of an added file.

10:13 AM Changeset in webkit [219826] by Matt Baker
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: should only show Canvases in resource type filter if debug setting enabled
​https://bugs.webkit.org/show_bug.cgi?id=174765
<rdar://problem/33479009>

Reviewed by Devin Rousso.

  • UserInterface/Views/ResourceSidebarPanel.js:

(WebInspector.ResourceSidebarPanel):

9:41 AM Changeset in webkit [219825] by Matt Lewis
  • 2 edits in trunk/LayoutTests

Marked tiled-drawing/scrolling/scroll-snap/scroll-snap-mandatory-mainframe-vertical.html as failure.
​https://bugs.webkit.org/show_bug.cgi?id=172148

Unreviewed test gardening.

  • platform/mac-wk2/TestExpectations:
9:06 AM Changeset in webkit [219824] by Darin Adler
  • 2 edits in trunk/LayoutTests

Remove race from one plug-in test
​https://bugs.webkit.org/show_bug.cgi?id=130653

Reviewed by Antti Koivisto.

  • plugins/plugin-remove-readystatechange.html: Updated test to use waitUntilDone/notifyDone

rather than racing. Added a computation of offsetLeft to trigger layout and adding a long
comment explaining why this is both needed and potentially insufficient in the long run.

9:01 AM Changeset in webkit [219823] by clopez@igalia.com
  • 3 edits in trunk/Source/WebCore/PAL

[CMake] libtasn1 should not be required when not enabling subtle_crypto
​https://bugs.webkit.org/show_bug.cgi?id=174785

Reviewed by Žan Doberšek.

  • pal/PlatformGTK.cmake:
  • pal/PlatformWPE.cmake:
8:43 AM Changeset in webkit [219822] by jmarcell@apple.com
  • 4 edits
    10 adds in branches/safari-604-branch

Cherry-pick r219665.

7:46 AM Changeset in webkit [219821] by mark.lam@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Gardening: fixed C Loop build after r219790.
​https://bugs.webkit.org/show_bug.cgi?id=174696

Not reviewed.

  • assembler/testmasm.cpp:
7:27 AM Changeset in webkit [219820] by zandobersek@gmail.com
  • 5 edits in trunk

[WPE] Add libepoxy to the Jhbuild moduleset
​https://bugs.webkit.org/show_bug.cgi?id=174711

Reviewed by Carlos Alberto Lopez Perez.

.:

libepoxy 1.3.1 that's widely available in current stable distributions (Debian stable,
Ubuntu LTS) does suffice our needs, but also contains a bug that incorrectly determines
the current EGL context and the API version of OpenGL that's bound to that context.
Focusing on experiences on Debian systems, while the current stable Debian release has
this bug patched, the package for the testing distribution of Debian removed that same
patch for whatever reason, despite using the 1.3.1 version as the package basis.

To avoid this, libepoxy 1.4.3 is now added to WPE's Jhbuild moduleset. The WPE port
also now demands at least the 1.4.0 version to be available when configuring.

We should revisit the minimum version for libepoxy in the future if the 1.3 release
branch gets updated with the proper fix.

  • Source/cmake/FindLibEpoxy.cmake:
  • Source/cmake/OptionsWPE.cmake:

Tools:

  • wpe/jhbuild.modules: Add libepoxy 1.4.3 to WPE's Jhbuild moduleset.
7:24 AM Changeset in webkit [219819] by zandobersek@gmail.com
  • 11 edits
    3 adds
    1 delete in trunk

[WPE] libepoxy headers can use EGL_CAST, which might not be defined by eglplatform.h
​https://bugs.webkit.org/show_bug.cgi?id=174780

Reviewed by Carlos Garcia Campos.

Source/WebCore:

Newer versions of libepoxy provide EGL definitions that already use the EGL_CAST macro.
We'd like to compile on systems where the eglplatform.h header doesn't yet provide
that macro that was added only recently (e.g. it's available in Mesa 17.1, but not
before).

To address that, the EpoxyEGL.h header is added that first includes the libepoxy's EGL
header and then provides the EGL_CAST definition in case the eglplatform.h header doesn't.
This header should be included in place of <epoxy/egl.h> (which is included in this new
header).

The EpoxyEGL.h and EpoxyShims.h headers are also moved under the platform/graphics/epoxy/
directory, with the new directory added to the list of inclusion paths for WPE.

  • PlatformWPE.cmake:
  • platform/graphics/PlatformDisplay.cpp:
  • platform/graphics/cairo/CairoUtilities.cpp:
  • platform/graphics/cairo/ImageBufferCairo.cpp:
  • platform/graphics/egl/GLContextEGL.cpp:
  • platform/graphics/egl/GLContextEGLWPE.cpp:
  • platform/graphics/epoxy/EpoxyEGL.h: Added.
  • platform/graphics/epoxy/EpoxyShims.h: Renamed from Source/WebCore/platform/graphics/EpoxyShims.h.
  • platform/graphics/opengl/Extensions3DOpenGLES.cpp:
  • platform/graphics/wpe/PlatformDisplayWPE.cpp:

Tools:

  • WebKitTestRunner/wpe/HeadlessViewBackend.cpp:

Manually define the EGL_CAST macro to the C++ version that uses static_cast<>
in case the eglplatform.h header doesn't provide that macro. This replicates
the work in WebCore's EpoxyEGL.h header, but we can't include that header here.

6:50 AM Changeset in webkit [219818] by zandobersek@gmail.com
  • 1 edit
    1 delete in trunk/Source/WebCore

[CoordinatedGraphics] Remove unused GLDefs.h header
​https://bugs.webkit.org/show_bug.cgi?id=174712

Reviewed by Michael Catanzaro.

  • platform/graphics/opengl/GLDefs.h: Remove this unused header.

As far as my recollection goes it was used in the IPC-based
CoordinatedGraphics subsystem, which has been removed already.

5:46 AM Changeset in webkit [219817] by Carlos Garcia Campos
  • 1 copy in releases/WebKitGTK/webkit-2.16.6

WebKitGTK+ 2.16.6

5:45 AM Changeset in webkit [219816] by Carlos Garcia Campos
  • 4 edits in releases/WebKitGTK/webkit-2.16

Unreviewed. Update OptionsGTK.cmake and NEWS for 2.16.6 release.

.:

  • Source/cmake/OptionsGTK.cmake:

Source/WebKit2:

  • gtk/NEWS: Add release notes for 2.16.6.
4:57 AM WebKitGTK/2.16.x edited by Carlos Garcia Campos
(diff)
4:15 AM WebKitGTK/2.16.x edited by Carlos Garcia Campos
(diff)
4:13 AM Changeset in webkit [219815] by Carlos Garcia Campos
  • 2 edits in releases/WebKitGTK/webkit-2.16/Source/JavaScriptCore

Unreviewed. Fix the build with GCC 4.9 after merge r217438.

  • bytecompiler/BytecodeGenerator.h:

(JSC::StructureForInContext::addGetInst):

4:09 AM Changeset in webkit [219814] by Carlos Garcia Campos
  • 3 edits in releases/WebKitGTK/webkit-2.16/Source/ThirdParty/ANGLE

Merge r219446 - eglplatform.h does not support Wayland
​https://bugs.webkit.org/show_bug.cgi?id=163482

This makes it possible to build WebKitGTK+ when the target system has only Wayland support,
but no X11 (and therefore the X11 headers are not present).

Reviewed by Alex Christensen.

  • include/EGL/eglplatform.h: Add Wayland typedefs when WL_EGL_PLATFORM is defined (for example

by including wayland-egl.h before including EGL/egl.h). Also, include the X11 headers only
when ANGLE_USE_X11 is defined and, for consistency with Mesa's version of the header, when
MESA_EGL_NO_X11_HEADERS is not defined.

4:07 AM Changeset in webkit [219813] by Carlos Garcia Campos
  • 3 edits in releases/WebKitGTK/webkit-2.16/Source/WebCore

Merge r219332, r219447 - [GTK] Spin buttons on input type number appear over the value itself for small widths
​https://bugs.webkit.org/show_bug.cgi?id=173572

Reviewed by Carlos Garcia Campos.

Source/WebCore:

When drawing the spin buttons, override the width of the input
element to increment it with the width of the spin button.
This ensures that we don't end up covering the input values with
the spin buttons.

Do this also for user controlled styles, because most web authors
won't test how their site renders on WebKitGTK+, and they will
assume spin buttons in the order of 13 pixels wide (that is what
most browsers use), but the GTK+ spin button is much wider (66 pixels).

  • rendering/RenderTheme.cpp:

(WebCore::RenderTheme::adjustStyle):

  • rendering/RenderThemeGtk.cpp:

(WebCore::RenderThemeGtk::adjustTextFieldStyle): Call the theme's adjustTextFieldStyle() also for user controlled styles.
(WebCore::RenderThemeGtk::adjustInnerSpinButtonStyle):

REGRESSION(r219332): [GTK] 9 new failures on fast/forms spinbutton related tests
​https://bugs.webkit.org/show_bug.cgi?id=174395

Reviewed by Carlos Garcia Campos.

Source/WebCore:

Before r219332 the height of the spin button widget was
calculated as the maximum value between the individual button
( the [+] or [-] ) width (33 pixels) and height (16 pixels).
And r219332 caused the height of the widget to be calculated as
the height of the button (16 pixels), which was incorrect as
each button should be first expanded vertically to fit the
preferred size of the widget.

Fix this by making the calculations about the spin button widget
on a new function spinButtonSize() that takes this into account,
and use this values both for adjusting the style of the input
field and the spin button widget itself.

  • rendering/RenderThemeGtk.cpp:

(WebCore::spinButtonSize):
(WebCore::RenderThemeGtk::adjustTextFieldStyle):
(WebCore::RenderThemeGtk::adjustInnerSpinButtonStyle):

3:57 AM Changeset in webkit [219812] by Carlos Garcia Campos
  • 3 edits
    1 add in releases/WebKitGTK/webkit-2.16

Merge r218819 - Crash in JSC::Lexer<unsigned char>::setCode
​https://bugs.webkit.org/show_bug.cgi?id=172754

Reviewed by Mark Lam.

JSTests:

  • stress/dont-reserve-huge-capacity-lexer.js: Added.

(catch):

Source/JavaScriptCore:

The lexer was asking one of its buffers to reserve initial space that
was O(text size in bytes). For large sources, this would end up causing
the vector to overflow and crash. This patch changes this code be like
the Lexer's other buffers and to only reserve a small starting buffer.

  • parser/Lexer.cpp:

(JSC::Lexer<T>::setCode):

3:53 AM Changeset in webkit [219811] by Carlos Garcia Campos
  • 3 edits
    2 adds in releases/WebKitGTK/webkit-2.16

Merge r216813 - Elements should be inserted into a template element as its content's last child
​https://bugs.webkit.org/show_bug.cgi?id=171373
<rdar://problem/31862949>

Reviewed by Ryosuke Niwa.

Source/WebCore:

Before this change, our HTML parser obeys the following premises:
1) A fostering child whose parent is a table should be inserted before its parent and under its grandparent.
2) When inserting into a template element, an element should be inserted into its content.

Let's walk through the example:
a) Before eventhandler takes place
template
table

svg <- parser

b) After eventhandler takes place
template

table

svg <- parser

c) after parsing svg
template

content

svg
(table)

table

Finally, in the example, the svg element will be inserted into the content of the template element while
having its next sibling point to the table element. However, the table element is actually under the
template element not its content.

This messy tree is constructed because the second premise is incompleted. It should be: When inserting into
a template element, an element should be inserted into its content as its last child.
Quoted from Step 3 of ​https://html.spec.whatwg.org/multipage/syntax.html#appropriate-place-for-inserting-a-node
A correct tree will then looks like:
template

content

svg

table

Tests: fast/dom/HTMLTemplateElement/insert-fostering-child-crash.html

fast/dom/HTMLTemplateElement/insert-fostering-child.html

  • html/parser/HTMLConstructionSite.cpp:

(WebCore::insert):
By nullifying task.nextChild, it will force the parser to append the element as task.parent's last child.

LayoutTests:

  • fast/dom/HTMLTemplateElement/insert-fostering-child-expected.txt: Added.
  • fast/dom/HTMLTemplateElement/insert-fostering-child.html: Added.
3:51 AM Changeset in webkit [219810] by Carlos Garcia Campos
  • 3 edits
    1 add in releases/WebKitGTK/webkit-2.16

Merge r215351 - Should use flushDirect() when flushing the scopeRegister due to needsScopeRegister().
​https://bugs.webkit.org/show_bug.cgi?id=170661
<rdar://problem/31579046>

Reviewed by Filip Pizlo.

JSTests:

  • stress/regress-170661.js: Added.

Source/JavaScriptCore:

Previously, we were using flush() to flush the outermost frame's scopeRegister.
This is incorrect because flush() expects the VirtualRegister value passed to
it to be that of the top most inlined frame. In the event that we reach a
terminal condition while inside an inlined frame, flush() will end up flushing
the wrong register. The fix is simply to use flushDirect() instead.

  • dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::flush):

3:49 AM Changeset in webkit [219809] by Carlos Garcia Campos
  • 3 edits
    1 add in releases/WebKitGTK/webkit-2.16

Merge r215596 - virtualThunkFor() needs to materialize its of tagMaskRegister for tail calls.
​https://bugs.webkit.org/show_bug.cgi?id=171079
<rdar://problem/31684756>

Reviewed by Saam Barati.

JSTests:

  • stress/regress-171079.js: Added.

Source/JavaScriptCore:

This is needed because tail calls would restore callee saved registers (and
therefore, potentially clobber the tag registers) before jumping to the thunk.

  • jit/ThunkGenerators.cpp:

(JSC::virtualThunkFor):

3:48 AM Changeset in webkit [219808] by Carlos Garcia Campos
  • 4 edits in releases/WebKitGTK/webkit-2.16/Source/JavaScriptCore

Merge r216309 - Put does not properly consult the prototype chain
​https://bugs.webkit.org/show_bug.cgi?id=171754

Reviewed by Saam Barati.

We should do a follow up that cleans up the rest of put. See:
​https://bugs.webkit.org/show_bug.cgi?id=171759

  • runtime/JSCJSValue.cpp:

(JSC::JSValue::putToPrimitive):

  • runtime/JSObject.cpp:

(JSC::JSObject::putInlineSlow):

  • runtime/JSObjectInlines.h:

(JSC::JSObject::canPerformFastPutInline):

3:31 AM Changeset in webkit [219807] by Carlos Garcia Campos
  • 2 edits in releases/WebKitGTK/webkit-2.16/Source/JavaScriptCore

Unreviewed. Fix merge r217438.

See bug #174781.

  • bytecompiler/BytecodeGenerator.cpp:

(JSC::StructureForInContext::finalize): Use operand instead of unsignedValue.

3:28 AM Changeset in webkit [219806] by Carlos Garcia Campos
  • 13 edits
    3 adds in releases/WebKitGTK/webkit-2.16

Merge r216279 - putDirectIndex does not properly do defineOwnProperty
​https://bugs.webkit.org/show_bug.cgi?id=171591
<rdar://problem/31735695>

Reviewed by Geoffrey Garen.

JSTests:

  • stress/array-prototype-splice-making-typed-array.js:

(test):

  • stress/array-species-config-array-constructor.js:

(shouldThrow):
(test):

  • stress/put-direct-index-broken-2.js: Added.

(assert):
(test):
(makeLengthWritable):
(set get restoreOldDesc):

  • stress/put-direct-index-broken.js: Added.

(whatToTest):
(tryRunning):
(tryItOut):

  • stress/put-indexed-getter-setter.js: Added.

(foo.X.prototype.set 7):
(foo.X.prototype.get 7):
(foo.X):
(foo):

Source/JavaScriptCore:

This patch fixes putDirectIndex and its JIT implementations to be
compatible with the ES6 spec. I think our code became out of date
when we implemented ArraySpeciesCreate since ArraySpeciesCreate may
return arbitrary objects. We perform putDirectIndex on that arbitrary
object. The behavior we want is as if we performed defineProperty({configurable:true, enumerable:true, writable:true}).
However, we weren't doing this. putDirectIndex assumed it could just splat
data into any descendent of JSObject's butterfly. For example, this means
we'd just splat into the butterfly of a typed array, even though a typed
array doesn't use its butterfly to store its indexed properties in the usual
way. Also, typed array properties are non-configurable, so this operation
should throw. This also means if we saw a ProxyObject, we'd just splat
into its butterfly, but this is obviously wrong because ProxyObject should
intercept the defineProperty operation.

This patch fixes this issue by adding a whitelist of cell types that can
go down putDirectIndex's fast path. Anything not in that whitelist will
simply call into defineOwnProperty.

  • bytecode/ByValInfo.h:

(JSC::jitArrayModePermitsPutDirect):

  • dfg/DFGArrayMode.cpp:

(JSC::DFG::ArrayMode::refine):

  • jit/JITOperations.cpp:
  • runtime/ArrayPrototype.cpp:

(JSC::arrayProtoFuncSplice):

  • runtime/ClonedArguments.cpp:

(JSC::ClonedArguments::createStructure):

  • runtime/JSGenericTypedArrayViewInlines.h:

(JSC::JSGenericTypedArrayView<Adaptor>::defineOwnProperty):

  • runtime/JSObject.cpp:

(JSC::canDoFastPutDirectIndex):
(JSC::JSObject::defineOwnIndexedProperty):
(JSC::JSObject::putDirectIndexSlowOrBeyondVectorLength):
(JSC::JSObject::putDirectIndexBeyondVectorLength): Deleted.

  • runtime/JSObject.h:

(JSC::JSObject::putDirectIndex):
(JSC::JSObject::canSetIndexQuicklyForPutDirect): Deleted.

  • runtime/JSType.h:
2:54 AM Changeset in webkit [219805] by Carlos Garcia Campos
  • 5 edits
    2 adds in releases/WebKitGTK/webkit-2.16

Merge r216992 - Replace CryptoOperationData with BufferSource for WebKitSubtleCrypto
​https://bugs.webkit.org/show_bug.cgi?id=172146
<rdar://problem/32122256>

Reviewed by Brent Fulgham.

Source/WebCore:

In this patch, we replaces CryptoOperationData with BufferSource for WebKitSubtleCrypto in
the custom binding codes.

Test: crypto/webkitSubtle/import-export-raw-key-leak.html

  • bindings/js/JSWebKitSubtleCryptoCustom.cpp:

(WebCore::JSWebKitSubtleCrypto::encrypt):
(WebCore::JSWebKitSubtleCrypto::decrypt):
(WebCore::JSWebKitSubtleCrypto::sign):
(WebCore::JSWebKitSubtleCrypto::verify):
(WebCore::JSWebKitSubtleCrypto::digest):
(WebCore::JSWebKitSubtleCrypto::importKey):
(WebCore::JSWebKitSubtleCrypto::unwrapKey):

  • crypto/WebKitSubtleCrypto.idl:

LayoutTests:

  • crypto/webkitSubtle/argument-conversion-expected.txt:
  • crypto/webkitSubtle/import-export-raw-key-leak-expected.txt: Added.
  • crypto/webkitSubtle/import-export-raw-key-leak.html: Added.
2:51 AM Changeset in webkit [219804] by Carlos Garcia Campos
  • 2 edits in releases/WebKitGTK/webkit-2.16/JSTests

Merge r217018 - Unreviewed, address mlam's review feedback.

  • stress/arguments-elimination-varargs-too-many-args-arg-count.js:
2:50 AM Changeset in webkit [219803] by Carlos Garcia Campos
  • 3 edits
    1 add in releases/WebKitGTK/webkit-2.16

Merge r217016 - JSC: Incorrect LoadVarargs handling in ArgumentsEliminationPhase::transform
​https://bugs.webkit.org/show_bug.cgi?id=172208

Reviewed by Saam Barati.

JSTests:

  • stress/arguments-elimination-varargs-too-many-args-arg-count.js: Added.

(foo):
(bar):
(baz):

Source/JavaScriptCore:

  • dfg/DFGArgumentsEliminationPhase.cpp:
2:45 AM Changeset in webkit [219802] by Carlos Garcia Campos
  • 3 edits
    1 add in releases/WebKitGTK/webkit-2.16

Merge r215451 - JSArray::appendMemcpy() needs to handle copying from Undecided indexing type too.
​https://bugs.webkit.org/show_bug.cgi?id=170896
<rdar://problem/31651319>

Reviewed by JF Bastien and Keith Miller.

JSTests:

  • stress/regress-170896.js: Added.

Source/JavaScriptCore:

  • runtime/JSArray.cpp:

(JSC::JSArray::appendMemcpy):

2:41 AM Changeset in webkit [219801] by Carlos Garcia Campos
  • 13 edits
    1 add in releases/WebKitGTK/webkit-2.16

Merge r217438 - Our for-in optimization in the bytecode generator does its static analysis incorrectly
​https://bugs.webkit.org/show_bug.cgi?id=172532
<rdar://problem/32369452>

Reviewed by Mark Lam.

JSTests:

  • stress/for-in-invalidation-for-any-write.js: Added.

(assert):
(test):
(test.i):

Source/JavaScriptCore:

Our static analysis for when a for-in induction variable
is written to tried to its analysis as we generate
bytecode. This has issues, since it does not account for
the dynamic execution path of the program. Let's consider
a program where our old analysis worked:

`
for (let p in o) {

o[p]; We can transform this into a fast get_direct_pname
p = 20;
o[p]; We cannot transform this since p has been changed.

}
`

However, our static analysis did not account for loops, which exist
in JavaScript. e.g, it would incorrectly compile this program as:
`
for (let p in o) {

for (let i = 0; i < 20; ++i) {

o[p]; It transforms this to use get_direct_pname even though p will be over-written if we get here from the inner loop back edge!
p = 20;
o[p]; We correctly do not transform this.

}

}
`

Because of this flaw, I've made the optimization more conservative.
We now optimistically emit code for the optimized access. However,
if a for-in context is *ever* invalidated, before we pop it off
the stack, we rewrite the program's optimized accesses to no longer
be optimized. To do this, each context keeps track of its optimized
accesses.

This patch also adds a new bytecode, op_nop, which is just a no-op.
It was helpful to add this because reverting get_direct_pname to get_by_val
will leave us with an extra instruction word because get_direct_pname is
has a length of 7 where get_by_val has a length of 6. This leaves us with
an extra slot that we fill with an op_nop.

  • bytecode/BytecodeDumper.cpp:

(JSC::BytecodeDumper<Block>::dumpBytecode):

  • bytecode/BytecodeList.json:
  • bytecode/BytecodeUseDef.h:

(JSC::computeUsesForBytecodeOffset):
(JSC::computeDefsForBytecodeOffset):

  • bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::emitGetByVal):
(JSC::BytecodeGenerator::popIndexedForInScope):
(JSC::BytecodeGenerator::popStructureForInScope):
(JSC::BytecodeGenerator::invalidateForInContextForLocal):
(JSC::StructureForInContext::pop):
(JSC::IndexedForInContext::pop):

  • bytecompiler/BytecodeGenerator.h:

(JSC::StructureForInContext::addGetInst):
(JSC::IndexedForInContext::addGetInst):

  • dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::parseBlock):

  • dfg/DFGCapabilities.cpp:

(JSC::DFG::capabilityLevel):

  • jit/JIT.cpp:

(JSC::JIT::privateCompileMainPass):

  • jit/JIT.h:
  • jit/JITOpcodes.cpp:

(JSC::JIT::emit_op_nop):

  • llint/LowLevelInterpreter.asm:
2:40 AM Changeset in webkit [219800] by Carlos Garcia Campos
  • 2 edits in releases/WebKitGTK/webkit-2.16/Source/WebCore

Merge r216825 - AccessibilityRenderObject::textUnderElement needs to assert on unclean tree.
​https://bugs.webkit.org/show_bug.cgi?id=172065

Reviewed by Simon Fraser.

r192103 changed the assert logic incorrectly. If the tree is dirty, regardless of the renderer's type,
TextIterator will end up forcing style update/layout on the render tree.
The original assert would have hit with bug 171546 prior to r216726.

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::textUnderElement):

2:40 AM Changeset in webkit [219799] by Carlos Garcia Campos
  • 7 edits
    2 adds in releases/WebKitGTK/webkit-2.16

Merge r216726 - AX: Defer text changes until after the tree is clean if needed.
​https://bugs.webkit.org/show_bug.cgi?id=171546
<rdar://problem/31934942>

Reviewed by Simon Fraser.

Source/WebCore:

While updating an accessibility object state, we might
trigger unintentional style updates. This style update could
end up destroying renderes that are still referenced by functions
on the callstack.
To avoid that, defer such changes and let AXObjectCache operate on a clean tree.

Test: accessibility/crash-when-render-tree-is-not-clean.html

  • accessibility/AXObjectCache.cpp:

(WebCore::AXObjectCache::remove):
(WebCore::AXObjectCache::handleAttributeChanged):
(WebCore::AXObjectCache::labelChanged):
(WebCore::AXObjectCache::performDeferredCacheUpdate):
(WebCore::AXObjectCache::deferRecomputeIsIgnored):
(WebCore::AXObjectCache::deferTextChangedIfNeeded):
(WebCore::AXObjectCache::recomputeDeferredIsIgnored): Deleted.
(WebCore::AXObjectCache::deferTextChanged): Deleted.

  • accessibility/AXObjectCache.h: Decouple different type of changes.

(WebCore::AXObjectCache::deferRecomputeIsIgnored):
(WebCore::AXObjectCache::deferTextChangedIfNeeded):
(WebCore::AXObjectCache::recomputeDeferredIsIgnored): Deleted.
(WebCore::AXObjectCache::deferTextChanged): Deleted.

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::deleteLines):

  • rendering/RenderBlockLineLayout.cpp:

(WebCore::RenderBlockFlow::createAndAppendRootInlineBox):

  • rendering/RenderText.cpp:

(WebCore::RenderText::setText):

LayoutTests:

  • accessibility/crash-when-render-tree-is-not-clean-expected.txt: Added.
  • accessibility/crash-when-render-tree-is-not-clean.html: Added.
1:58 AM Changeset in webkit [219798] by Carlos Garcia Campos
  • 6 edits
    2 adds in releases/WebKitGTK/webkit-2.16

Merge r216096 - Defer AX cache update when text content changes until after layout is finished.
​https://bugs.webkit.org/show_bug.cgi?id=171429
<rdar://problem/31885984>

Reviewed by Simon Fraser.

Source/WebCore:

When the content of the RenderText changes (even as the result of a text-transform change)
instead of updating the AX cache eagerly (and trigger layout on a half-backed render tree)
we should just defer it until after the subsequent layout is done.

Test: accessibility/crash-while-adding-text-child-with-transform.html

  • accessibility/AXObjectCache.cpp:

(WebCore::AXObjectCache::remove):
(WebCore::AXObjectCache::performDeferredCacheUpdate):
(WebCore::AXObjectCache::recomputeDeferredIsIgnored):
(WebCore::AXObjectCache::deferTextChanged):
(WebCore::AXObjectCache::performDeferredIsIgnoredChange): Deleted.

  • accessibility/AXObjectCache.h:

(WebCore::AXObjectCache::deferTextChanged):
(WebCore::AXObjectCache::performDeferredCacheUpdate):
(WebCore::AXObjectCache::performDeferredIsIgnoredChange): Deleted.

  • page/FrameView.cpp:

(WebCore::FrameView::performPostLayoutTasks):

  • rendering/RenderText.cpp:

(WebCore::RenderText::setText):

LayoutTests:

  • accessibility/crash-while-adding-text-child-with-transform-expected.txt: Added.
  • accessibility/crash-while-adding-text-child-with-transform.html: Added.
1:33 AM Changeset in webkit [219797] by fred.wang@free.fr
  • 8 edits
    7 adds in trunk

Add attribute allow-top-navigation-by-user-activation to iframe sandbox
​https://bugs.webkit.org/show_bug.cgi?id=171327

Patch by Frederic Wang <​fwang@igalia.com> on 2017-07-11
Reviewed by Chris Dumez.

LayoutTests/imported/w3c:

This commit updates the expectation for the test verifying that a sandboxed iframe without
the allow-top-navigation-by-user-activation flag set can not perform top navigation if it is
not triggered by a user gesture. The navigation is still prohibited but the parsing of the
allow-top-navigation-by-user-activation flag should not raised any error message.
A similar update is done for the test verifying that the combination of the flags
allow-top-navigation-by-user-activation and allow-top-navigation.

  • web-platform-tests/html/semantics/embedded-content/the-iframe-element/iframe_sandbox_allow_top_navigation_by_user_activation_without_user_gesture-expected.txt: Remove the error message.
  • web-platform-tests/html/semantics/embedded-content/the-iframe-element/iframe_sandbox_allow_top_navigation-3-expected.txt: Ditto.

Source/WebCore:

Tests: http/tests/security/frameNavigation/sandbox-ALLOWED-top-navigation-with-user-gesture-1.html

http/tests/security/frameNavigation/sandbox-ALLOWED-top-navigation-with-user-gesture-2.html
imported/w3c/web-platform-tests/html/semantics/embedded-content/the-iframe-element/iframe_sandbox_allow_top_navigation_by_user_activation_without_user_gesture-expected.txt
imported/w3c/web-platform-tests/html/semantics/embedded-content/the-iframe-element/iframe_sandbox_allow_top_navigation-3-expected.txt

  • dom/Document.cpp:

(WebCore::Document::canNavigate): Case 2 of conformance verification is refined to match the
current specification: Top navigation is allowed when it is triggered by user activation and
when allow-top-navigation-by-user-activation is set. Because we have additional "security
origin" rules with respect to the specification we also add some early return to ensure that
navigation is really allowed for this new case.

  • dom/SecurityContext.cpp:

(WebCore::SecurityContext::isSupportedSandboxPolicy): Add allow-top-navigation-by-user-activation flag.
(WebCore::SecurityContext::parseSandboxPolicy): Parse the new flag to allow top user
navigation by user activation. Also, make sure that allow-top-navigation allows such
navigation as defined by the specification.

  • dom/SecurityContext.h: Declare new sandboxing flag.

LayoutTests:

This patch adds tests to verify that a sandboxed iframe with the flag
'allow-top-navigation-by-user-activation' or 'allow-top-navigation' can navigate the top
level page, if navigation is triggered by a user gesture. This is based on a test from the
Chromium repository.
It also verifies that navigation fails when 'allow-top-navigation-by-user-activation' is
absent, even when it is triggered by a user gesture.

  • http/tests/security/frameNavigation/resources/iframe-that-performs-parent-navigation-with-user-activation.html: Added. This is similar to iframe-that-performs-parent-navigation.html but it

performs navigation of its parent using user activation.

  • http/tests/security/frameNavigation/sandbox-ALLOWED-top-navigation-with-user-gesture-1.html: Added.

Test top navigation for a sandboxed frame with 'allow-top-navigation-by-user-activation'

  • http/tests/security/frameNavigation/sandbox-ALLOWED-top-navigation-with-user-gesture-1-expected.txt: Added.

Add PASS expectation.

  • http/tests/security/frameNavigation/sandbox-ALLOWED-top-navigation-with-user-gesture-2.html: Added.

Test top navigation for a sandboxed frame with 'allow-top-navigation'.

  • http/tests/security/frameNavigation/sandbox-ALLOWED-top-navigation-with-user-gesture-2-expected.txt: Added.

Add PASS expectation.

  • http/tests/security/frameNavigation/sandbox-DENIED-top-navigation-with-user-gesture.html: Added.

Test user-triggered navigation for a sandboxed frame without 'allow-top-navigation-by-user-activation'.

  • http/tests/security/frameNavigation/sandbox-DENIED-top-navigation-with-user-gesture-expected.txt: Added.

Add reference with navigation failure.

Note: See TracTimeline for information about the timeline view.