WebKit-https.git
2 years agoString#utf8() allocates new CString
utatane.tea@gmail.com [Mon, 4 Sep 2017 01:38:09 +0000 (01:38 +0000)]
String#utf8() allocates new CString
https://bugs.webkit.org/show_bug.cgi?id=176302

Reviewed by Sam Weinig.

Source/WebCore:

Several places uses String#utf8() like an accessor. In reality, it allocates new CString per call.
It is very costly. Furthermore, some places uses this incorrectly. For example,
`std::vector<char>(srtpAuth.utf8().data(), srtpAuth.utf8().data() + srtpAuth.utf8().length());`
is incorrect since each time strpAuth.utf8() allocates different CString.

This patch calls utf8() first and use it.

* Modules/websockets/WebSocketChannel.cpp:
(WebCore::WebSocketChannel::startClosingHandshake):
* platform/wpe/PlatformPasteboardWPE.cpp:
(WebCore::PlatformPasteboard::write):

Source/WebKit:

* NetworkProcess/webrtc/NetworkRTCSocket.cpp:
(WebKit::NetworkRTCSocket::sendTo):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221562 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[SOUP] Use fastMalloced SoupBuffer in ResourcehandleSoup
utatane.tea@gmail.com [Mon, 4 Sep 2017 01:37:05 +0000 (01:37 +0000)]
[SOUP] Use fastMalloced SoupBuffer in ResourcehandleSoup
https://bugs.webkit.org/show_bug.cgi?id=176311

Reviewed by Michael Catanzaro.

Use fastMalloced memory instead of g_malloc for SoupBuffer.

* platform/network/soup/ResourceHandleSoup.cpp:
(WebCore::ResourceHandle::ensureReadBuffer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221561 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[WTF] Add C++03 allocator interface for GCC < 6
utatane.tea@gmail.com [Mon, 4 Sep 2017 01:00:25 +0000 (01:00 +0000)]
[WTF] Add C++03 allocator interface for GCC < 6
https://bugs.webkit.org/show_bug.cgi?id=176301

Reviewed by Darin Adler.

Source/JavaScriptCore:

* dfg/DFGObjectAllocationSinkingPhase.cpp:

Source/WTF:

Unfortunately, std::list in GCC < 6 does not support C++11 allocator interface.
This patch adds C++03 allocator interface to FastAllocator to make it usable
for std::list. It also allows us to use FastAllocator for data structures that
only support C++03 interface.

* wtf/FastMalloc.h:
(WTF::FastAllocator::allocate):
(WTF::FastAllocator::construct):
(WTF::FastAllocator::destroy):
(WTF::FastAllocator::max_size const):
(WTF::FastAllocator::select_on_container_copy_construction const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221560 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Cocoa] Tidy a few things in legacy WebHTMLView
darin@apple.com [Sun, 3 Sep 2017 22:28:04 +0000 (22:28 +0000)]
[Cocoa] Tidy a few things in legacy WebHTMLView
https://bugs.webkit.org/show_bug.cgi?id=176280

Reviewed by Dan Bernstein.

* WebCoreSupport/WebChromeClient.mm:
(WebChromeClient::attachRootGraphicsLayer): Made this function Mac-only, because the
methods it calls on WebHTMLView were doing nothing on iOS.

* WebView/WebFrame.mm:
(-[WebFrame _paintBehaviorForDestinationContext:]): Changed iOS code so that it no longer
depends on -[WebHTMLView _web_isDrawingIntoLayer], which always returned YES on iOS.

* WebView/WebHTMLView.mm: Define WebEvent on both iOS and Mac so more event handling
code can be shared without #if below. Use #if PLATFORM(MAC) instead of #if !PLATFORM(IOS)
in a lot more places. Move many of the SPI declarations up to the top in alphabetical
order. Removed all sorts of unused things. Made more code and fields Mac-only where
what was compiled was doing nothing on iOS and it makes the resulting code simpler.
Use RetainPtr instead of writing retain/release wherever possible.
(+[WebMenuTarget sharedMenuTarget]): Moved global inside this method and used C++
initialization for it. Removed various unused methods and declarations of methods.
(promisedDataClient): Use NeverDestroyed instead of just doing a "new" on the heap.
(-[WebHTMLView _imageExistsAtPaths:]): Deleted.
(-[WebHTMLView _documentFragmentWithPaths:]): Use RetainPtr and modern for loop.
(+[WebHTMLView _excludedElementsForAttributedStringConversion]): Use C++ initialization,
and remove CFRetain no longer needed since we don't support garbage collection any more.
(uniqueURLWithRelativePart): Simplified using [NSUUID UUID].
(-[WebHTMLView _plainTextFromPasteboard:]): Simplified with RetainPtr.
(-[WebHTMLView _pasteWithPasteboard:allowPlainText:]): Ditto.
(-[WebHTMLView _pasteAsPlainTextWithPasteboard:]): Ditto.
(-[WebHTMLView _removeWindowObservers]): Removed code dealing with empty
methods windowWillOrderOnScreen: and windowWillOrderOffScreen:.
(-[WebHTMLView _shouldDeleteRange:]): Deleted. Was unused.
(-[WebHTMLView _writeSelectionWithPasteboardTypes:toPasteboard:cachedAttributedString:]):
Simplified with stringByReplacingOccurrencesOfString and modern string literal syntax.
(-[WebHTMLView _setMouseDownEvent:]): Ditto.
(-[WebHTMLView _web_setPrintingModeRecursive:adjustViewSize:]): Added. Helper method so
the next three methhods can be super-simple. Used RetainPtr and a modern for loop to make
this method simpler than the functions it replaced. Also got rid of the enumeratingSubviews
debug flag; we haven't used it for years and it's not clearly helpful.
(-[WebHTMLView _web_setPrintingModeRecursive]): Turned into a one-liner with the above.
(-[WebHTMLView _web_clearPrintingModeRecursive]): Ditto.
(-[WebHTMLView _web_setPrintingModeRecursiveAndAdjustViewSize]): Ditto.
(-[WebHTMLView _web_makePluginSubviewsPerformSelector:withObject:]): Moved from the
WebInternal category to the WebHTMLViewFileInternal category so it can be called by
two additional methods in the WebPrivate category and still be called by the two
methods in the WebInternal category that were already using it.
(-[WebHTMLView didAddSubview:]): Deleted.
(-[WebHTMLView _insideAnotherHTMLView]): Deleted.
(-[WebHTMLView _clearLastHitViewIfSelf]): Deleted. Inlined the implementation in the one
place this was called, the close method.
(-[WebHTMLView _setToolTip:]): Simplified with RetainPtr.
(-[WebHTMLView pasteboard:provideDataForType:]): Ditto.
(-[WebHTMLView _autoscroll]): Simplified by making this Mac-only.
(-[WebHTMLView _hasHTMLDocument]): Deleted. The one caller was removed a long time ago.
(-[WebHTMLView _web_documentFragmentFromPasteboard:pasteboardType:imageMIMEType:]):
Added helper method to make the one below simpler.
(-[WebHTMLView _documentFragmentFromPasteboard:forType:inContext:subresources:]):
Compared pasteboard type strings using isEqualToString: rather than using == since
we want to actually compare the string contents. Also updated to use modern for loop,
renamed a local variable from "s" to "localSubresources", and factored out the handling
of three image types to use the helper method above.
(-[WebHTMLView _pauseNullEventsForAllNetscapePlugins]): Rewrite using the
_web_makePluginSubviewsPerformSelector:withObject: method.
(-[WebHTMLView _resumeNullEventsForAllNetscapePlugins]): Ditto.
(-[NSView _web_addDescendentWebHTMLViewsToArray:]): Fixed spelling to use the adjective
"descendent" rather than the noun "descendant". Also removed Mac-only optimization to
not call the -[NSView subviews] method to make this identical on both platforms and used
a modern for loop.
(-[NSMutableDictionary _web_setObjectIfNotNil:forKey:]): Deleted.
(-[WebHTMLView acceptsFirstResponder]): Reduced amount of platform-specific code by
defining WebEventMouseDown as a synonym for NSEventTypeLeftMouseDown on Mac.
(-[WebHTMLView addWindowObservers]): Removed code dealing with empty
methods windowWillOrderOnScreen: and windowWillOrderOffScreen:.
(-[WebHTMLView willRemoveSubview:]): Removed the enumeratingSubviews check.
(-[WebHTMLView drawSingleRect:]): Rearranged this method to make it clearer that most
of it is Mac-only.
(-[WebHTMLView windowWillOrderOnScreen:]): Deleted. All the current version of this
method did was call itself on the main thread, and do nothing if called on main thread.
(-[WebHTMLView windowWillOrderOffScreen:]): Deleted. Ditto.
(-[WebHTMLView _isSelectionEvent:]): Merged macOS and iOS versions.
(-[WebHTMLView resignFirstResponder]): Removed iOS-specific code in the macOS-specific
half of this method.
(-[WebHTMLView setDataSource:]): Simplified using RetainPtr.
(-[WebHTMLView _setPrinting:minimumPageLogicalWidth:logicalHeight:originalPageWidth:originalPageHeight:maximumShrinkRatio:adjustViewSize:paginateScreenContent:]):
Simplified using a modern for loop and RetainPtr.
(-[WebHTMLView knowsPageRange:]): Simplified using RetainPtr.
(-[WebHTMLView keyDown:]): Ditto.
(-[WebHTMLView _selectionStartFontAttributesAsRTF]): Ditto.
(-[WebHTMLView performKeyEquivalent:]): Ditto.
(-[WebHTMLView _styleForAttributeChange:]): Ditto.
(-[WebHTMLView complete:]): Ditto.
(-[WebHTMLView _windowChangedKeyState]): Removed iOS-specific code since this method is
now macOS-only.
(-[WebHTMLView _selectionChanged]): Removed call to _updateSelectionForInputManager on
iOS since that is now macOS-only.
(-[WebHTMLView promisedDragTIFFDataSource]): Deleted.
(-[WebHTMLView _startAutoscrollTimer:]): Moved this into an internal category instead of
a private one because it is not used or needed outside WebKit. Also made it macOS-only
and reorganized to use early return.
(-[WebHTMLView _stopAutoscrollTimer]): Moved into internal category. It remains a no-op
on iOS, but kept there so we don't have to do #if at every call site.
(-[WebHTMLView _pluginController]): Moved into internal category.
(-[WebHTMLView validAttributesForMarkedText]): Simplified by removing CFRetain that was
there only for GC, using C++ initialization, removing obsolete comment.
(-[WebHTMLView _updateSelectionForInputManager]): Made this macOS-only and removed
peculiar iOS-only comment.
(+[NSURL _web_uniqueWebDataURL]): Simplified using [NSUUID UUID].

* WebView/WebHTMLViewInternal.h: Removed unused promisedDragTIFFDataSource method.
Made attachRootLayer:, detachRootLayer, _web_isDrawingIntoLayer,
_web_isDrawingIntoAcceleratedLayer all macOS-only. Moved _startAutoscrollTimer:,
_stopAutoscrollTimer, and _pluginController here from SPI header since they aren't
needed or used as SPI. And made _startAutoscrollTimer macOS-only.

* WebView/WebHTMLViewPrivate.h: Removed unneeded and unused SPI: _insideAnotherHTMLView,
_clearLastHitViewIfSelf, _pluginController, _startAutoscrollTimer:, and _stopAutoscrollTimer.
Made _transparentBackground and _setTransparentBackground: macOS-only. These probably
are not needed on macOS either, but I wasn't able to make absolutely sure I could remove
them at this time.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221559 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoHave getFileMetadata() return a std::optional<FileMetadata>
cdumez@apple.com [Sun, 3 Sep 2017 22:03:33 +0000 (22:03 +0000)]
Have getFileMetadata() return a std::optional<FileMetadata>
https://bugs.webkit.org/show_bug.cgi?id=176293

Reviewed by Darin Adler.

Source/WebCore:

Have getFileMetadata() return a std::optional<FileMetadata> instead of
using an out parameter for the metadata.

* Modules/entriesapi/DOMFileSystem.cpp:
(WebCore::listDirectoryWithMetadata):
(WebCore::validatePathIsExpectedType):
(WebCore::DOMFileSystem::getEntry):
* html/FileListCreator.cpp:
(WebCore::appendDirectoryFiles):
* platform/FileSystem.cpp:
(WebCore::fileIsDirectory):
* platform/FileSystem.h:
* platform/glib/FileSystemGlib.cpp:
(WebCore::getFileMetadata):
* platform/network/BlobDataFileReference.cpp:
(WebCore::BlobDataFileReference::startTrackingModifications):
* platform/network/mac/BlobDataFileReferenceMac.mm:
(WebCore::BlobDataFileReference::generateReplacementFile):
* platform/posix/FileSystemPOSIX.cpp:
(WebCore::getFileMetadata):
* platform/win/FileSystemWin.cpp:
(WebCore::getFileMetadata):

Tools:

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

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221558 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, rolling out r221555.
cdumez@apple.com [Sun, 3 Sep 2017 21:55:26 +0000 (21:55 +0000)]
Unreviewed, rolling out r221555.

Did not fix Windows build

Reverted changeset:

"Unreviewed attempt to fix Windows build."
http://trac.webkit.org/changeset/221555

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221557 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoConsolidate entries-api tests under entries-api/ folders
cdumez@apple.com [Sun, 3 Sep 2017 21:49:16 +0000 (21:49 +0000)]
Consolidate entries-api tests under entries-api/ folders
https://bugs.webkit.org/show_bug.cgi?id=176315

Reviewed by Darin Adler.

Consolidate entries-api tests under entries-api/ folders, for clarity.
Also drop an older test for directory upload which did not pass because
it was missing its resources. We now have good coverage for this feature
so this test is not needed.

* editing/pasteboard/enties-api/datatransfer-items-drop-directoryReader-error-expected.txt: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-directoryReader-error-expected.txt.
* editing/pasteboard/enties-api/datatransfer-items-drop-directoryReader-error.html: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-directoryReader-error.html.
* editing/pasteboard/enties-api/datatransfer-items-drop-directoryReader-expected.txt: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-directoryReader-expected.txt.
* editing/pasteboard/enties-api/datatransfer-items-drop-directoryReader-root-expected.txt: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-directoryReader-root-expected.txt.
* editing/pasteboard/enties-api/datatransfer-items-drop-directoryReader-root.html: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-directoryReader-root.html.
* editing/pasteboard/enties-api/datatransfer-items-drop-directoryReader.html: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-directoryReader.html.
* editing/pasteboard/enties-api/datatransfer-items-drop-fileEntry-file-expected.txt: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-fileEntry-file-expected.txt.
* editing/pasteboard/enties-api/datatransfer-items-drop-fileEntry-file.html: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-fileEntry-file.html.
* editing/pasteboard/enties-api/datatransfer-items-drop-getAsEntry-expected.txt: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getAsEntry-expected.txt.
* editing/pasteboard/enties-api/datatransfer-items-drop-getAsEntry.html: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getAsEntry.html.
* editing/pasteboard/enties-api/datatransfer-items-drop-getDirectory-expected.txt: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getDirectory-expected.txt.
* editing/pasteboard/enties-api/datatransfer-items-drop-getDirectory.html: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getDirectory.html.
* editing/pasteboard/enties-api/datatransfer-items-drop-getFile-expected.txt: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getFile-expected.txt.
* editing/pasteboard/enties-api/datatransfer-items-drop-getFile.html: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getFile.html.
* editing/pasteboard/enties-api/datatransfer-items-drop-getParent-expected.txt: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getParent-expected.txt.
* editing/pasteboard/enties-api/datatransfer-items-drop-getParent-root-expected.txt: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getParent-root-expected.txt.
* editing/pasteboard/enties-api/datatransfer-items-drop-getParent-root.html: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getParent-root.html.
* editing/pasteboard/enties-api/datatransfer-items-drop-getParent.html: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getParent.html.
* editing/pasteboard/enties-api/datatransfer-items-drop-getParent2-expected.txt: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getParent2-expected.txt.
* editing/pasteboard/enties-api/datatransfer-items-drop-getParent2.html: Renamed from LayoutTests/editing/pasteboard/datatransfer-items-drop-getParent2.html.
* fast/forms/file/entries-api/resources/testFiles/file1.txt: Renamed from LayoutTests/fast/forms/file/resources/testFiles/file1.txt.
* fast/forms/file/entries-api/resources/testFiles/file2.txt: Renamed from LayoutTests/fast/forms/file/resources/testFiles/file2.txt.
* fast/forms/file/entries-api/resources/testFiles/subfolder1/file3.txt: Renamed from LayoutTests/fast/forms/file/resources/testFiles/subfolder1/file3.txt.
* fast/forms/file/entries-api/resources/testFiles/subfolder2/file4.txt: Renamed from LayoutTests/fast/forms/file/resources/testFiles/subfolder2/file4.txt.
* fast/forms/file/entries-api/resources/testFiles/subfolder2/subfolder2a/file5.txt: Renamed from LayoutTests/fast/forms/file/resources/testFiles/subfolder2/subfolder2a/file5.txt.
* fast/forms/file/entries-api/webkitdirectory-drag-folder-expected.txt: Renamed from LayoutTests/fast/forms/file/webkitdirectory-drag-folder-expected.txt.
* fast/forms/file/entries-api/webkitdirectory-drag-folder.html: Renamed from LayoutTests/fast/forms/file/webkitdirectory-drag-folder.html.
* fast/forms/file/entries-api/webkitdirectory-open-panel-expected.txt: Renamed from LayoutTests/fast/forms/file/webkitdirectory-open-panel-expected.txt.
* fast/forms/file/entries-api/webkitdirectory-open-panel.html: Renamed from LayoutTests/fast/forms/file/webkitdirectory-open-panel.html.
* fast/forms/file/input-file-directory-upload-expected.txt: Removed.
* fast/forms/file/input-file-directory-upload.html: Removed.
* platform/ios/TestExpectations:
* platform/mac-wk1/fast/forms/file/entries-api/webkitdirectory-open-panel-expected.txt: Renamed from LayoutTests/platform/mac-wk1/fast/forms/file/webkitdirectory-open-panel-expected.txt.
* platform/mac/TestExpectations:
* platform/win/TestExpectations:
* platform/wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221556 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed attempt to fix Windows build.
cdumez@apple.com [Sun, 3 Sep 2017 20:38:52 +0000 (20:38 +0000)]
Unreviewed attempt to fix Windows build.

* runtime/JSGlobalObjectFunctions.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221555 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMove shared part of the Remote Layer Tree code out of a Mac-specific directory
timothy_horton@apple.com [Sun, 3 Sep 2017 19:08:06 +0000 (19:08 +0000)]
Move shared part of the Remote Layer Tree code out of a Mac-specific directory
https://bugs.webkit.org/show_bug.cgi?id=176297

Reviewed by Dan Bernstein.

UI-side compositing works on all Cocoa platforms, so the code should
not live inside Mac/. Also merge the Scrolling directory in here,
since those files are very related.

* PlatformMac.cmake:
* Shared/RemoteLayerTree/RemoteLayerBackingStore.h: Renamed from Shared/mac/RemoteLayerBackingStore.h.
* Shared/RemoteLayerTree/RemoteLayerBackingStore.mm: Renamed from Shared/mac/RemoteLayerBackingStore.mm.
* Shared/RemoteLayerTree/RemoteLayerBackingStoreCollection.h: Renamed from Shared/mac/RemoteLayerBackingStoreCollection.h.
* Shared/RemoteLayerTree/RemoteLayerBackingStoreCollection.mm: Renamed from Shared/mac/RemoteLayerBackingStoreCollection.mm.
* Shared/RemoteLayerTree/RemoteLayerTreePropertyApplier.h: Renamed from Shared/mac/RemoteLayerTreePropertyApplier.h.
* Shared/RemoteLayerTree/RemoteLayerTreePropertyApplier.mm: Renamed from Shared/mac/RemoteLayerTreePropertyApplier.mm.
* Shared/RemoteLayerTree/RemoteLayerTreeTransaction.h: Renamed from Shared/mac/RemoteLayerTreeTransaction.h.
* Shared/RemoteLayerTree/RemoteLayerTreeTransaction.mm: Renamed from Shared/mac/RemoteLayerTreeTransaction.mm.
* Shared/RemoteLayerTree/RemoteScrollingCoordinatorTransaction.cpp: Renamed from Shared/Scrolling/RemoteScrollingCoordinatorTransaction.cpp.
* Shared/RemoteLayerTree/RemoteScrollingCoordinatorTransaction.h: Renamed from Shared/Scrolling/RemoteScrollingCoordinatorTransaction.h.
* WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221554 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, rolling out r221552.
cdumez@apple.com [Sun, 3 Sep 2017 18:54:35 +0000 (18:54 +0000)]
Unreviewed, rolling out r221552.

Broke the build

Reverted changeset:

"[WTF] Add C++03 allocator interface for GCC < 6"
https://bugs.webkit.org/show_bug.cgi?id=176301
http://trac.webkit.org/changeset/221552

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221553 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[WTF] Add C++03 allocator interface for GCC < 6
utatane.tea@gmail.com [Sun, 3 Sep 2017 18:49:25 +0000 (18:49 +0000)]
[WTF] Add C++03 allocator interface for GCC < 6
https://bugs.webkit.org/show_bug.cgi?id=176301

Reviewed by Darin Adler.

Source/JavaScriptCore:

* dfg/DFGObjectAllocationSinkingPhase.cpp:

Source/WTF:

Unfortunately, std::list in GCC < 6 does not support C++11 allocator interface.
This patch adds C++03 allocator interface to FastAllocator to make it usable
for std::list. It also allows us to use FastAllocator for data structures that
only support C++03 interface.

* wtf/FastMalloc.h:
(WTF::FastAllocator::allocate):
(WTF::FastAllocator::construct):
(WTF::FastAllocator::destroy):
(WTF::FastAllocator::max_size const):
(WTF::FastAllocator::select_on_container_copy_construction const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221552 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[JSC] Clean up BytecodeLivenessAnalysis
utatane.tea@gmail.com [Sun, 3 Sep 2017 17:51:45 +0000 (17:51 +0000)]
[JSC] Clean up BytecodeLivenessAnalysis
https://bugs.webkit.org/show_bug.cgi?id=176295

Reviewed by Saam Barati.

Previously, computeDefsForBytecodeOffset was a bit customizable.
This is used for try-catch handler's liveness analysis. But after
careful generatorification implementation, it is now not necessary.
This patch drops this customizability.

* bytecode/BytecodeGeneratorification.cpp:
(JSC::GeneratorLivenessAnalysis::computeDefsForBytecodeOffset): Deleted.
(JSC::GeneratorLivenessAnalysis::computeUsesForBytecodeOffset): Deleted.
* bytecode/BytecodeLivenessAnalysis.cpp:
(JSC::BytecodeLivenessAnalysis::computeKills):
(JSC::BytecodeLivenessAnalysis::computeDefsForBytecodeOffset): Deleted.
(JSC::BytecodeLivenessAnalysis::computeUsesForBytecodeOffset): Deleted.
* bytecode/BytecodeLivenessAnalysis.h:
* bytecode/BytecodeLivenessAnalysisInlines.h:
(JSC::BytecodeLivenessPropagation::stepOverInstruction):
(JSC::BytecodeLivenessPropagation::computeLocalLivenessForBytecodeOffset):
(JSC::BytecodeLivenessPropagation::computeLocalLivenessForBlock):
(JSC::BytecodeLivenessPropagation::getLivenessInfoAtBytecodeOffset):
(JSC::BytecodeLivenessPropagation::runLivenessFixpoint):
(JSC::BytecodeLivenessPropagation<DerivedAnalysis>::stepOverInstruction): Deleted.
(JSC::BytecodeLivenessPropagation<DerivedAnalysis>::computeLocalLivenessForBytecodeOffset): Deleted.
(JSC::BytecodeLivenessPropagation<DerivedAnalysis>::computeLocalLivenessForBlock): Deleted.
(JSC::BytecodeLivenessPropagation<DerivedAnalysis>::getLivenessInfoAtBytecodeOffset): Deleted.
(JSC::BytecodeLivenessPropagation<DerivedAnalysis>::runLivenessFixpoint): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221551 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL should support structs
fpizlo@apple.com [Sun, 3 Sep 2017 17:46:48 +0000 (17:46 +0000)]
WSL should support structs
https://bugs.webkit.org/show_bug.cgi?id=176290

Reviewed by Mylex Maxfield.

This adds parsing of struct types and makes them work.

This required adding a struct layout phase, which in turn required being a bit more careful
about how we instantiate functions and types. We cannot have any instantiation TypeRefs left
once we're doing struct layout.

* WebGPUShadingLanguageRI/All.js:
* WebGPUShadingLanguageRI/CallFunction.js:
(callFunction):
* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitProtocolDecl.set throw):
* WebGPUShadingLanguageRI/DotExpression.js: Added.
(DotExpression):
(DotExpression.prototype.get struct):
(DotExpression.prototype.get fieldName):
(DotExpression.prototype.get isLValue):
(DotExpression.prototype.toString):
* WebGPUShadingLanguageRI/EBufferBuilder.js:
(EBufferBuilder.prototype._createEPtr):
* WebGPUShadingLanguageRI/EPtr.js:
(EPtr.prototype.copyFrom): Deleted.
(EPtr.prototype.toString): Deleted.
* WebGPUShadingLanguageRI/Evaluator.js:
(Evaluator.prototype.visitAssignment):
(Evaluator.prototype.visitDotExpression):
* WebGPUShadingLanguageRI/Field.js:
(Field):
* WebGPUShadingLanguageRI/FuncDef.js:
(FuncDef.prototype.rewrite):
* WebGPUShadingLanguageRI/FuncInstantiator.js:
(FuncInstantiator.prototype.getUnique):
(FuncInstantiator):
* WebGPUShadingLanguageRI/Inline.js:
(inline):
(_inlineFunction):
(resolveInlinedFunction):
* WebGPUShadingLanguageRI/Inliner.js:
(Inliner.prototype.visitDotExpression):
* WebGPUShadingLanguageRI/InstantiateImmediates.js:
(InstantiateImmediates.prototype.visitTypeRef):
(InstantiateImmediates.prototype.visitVariableDecl):
(InstantiateImmediates):
* WebGPUShadingLanguageRI/NameResolver.js:
(NameResolver.prototype._resolveTypeArguments):
(NameResolver.prototype.visitTypeRef):
* WebGPUShadingLanguageRI/Node.js:
(Node.prototype.visit):
* WebGPUShadingLanguageRI/OverloadResolutionFailure.js: Added.
(OverloadResolutionFailure):
(OverloadResolutionFailure.prototype.get func):
(OverloadResolutionFailure.prototype.get reason):
(OverloadResolutionFailure.prototype.toString):
* WebGPUShadingLanguageRI/Parse.js:
(parsePossibleSuffix):
(parseField):
(parseStructType):
(parse):
* WebGPUShadingLanguageRI/ResolveOverloadImpl.js:
(resolveOverloadImpl):
* WebGPUShadingLanguageRI/Rewriter.js:
(Rewriter.prototype.visitDotExpression):
(Rewriter.prototype.visitVariableRef):
* WebGPUShadingLanguageRI/StructLayoutBuilder.js: Added.
(StructLayoutBuilder):
(StructLayoutBuilder.prototype.visitStructType):
(StructLayoutBuilder.prototype.get offset):
(StructLayoutBuilder.prototype.visitField):
(StructLayoutBuilder.prototype.visitVariableRef):
(StructLayoutBuilder.prototype.visitVariableDecl):
* WebGPUShadingLanguageRI/StructType.js:
(StructType.prototype.add):
(StructType.prototype.instantiate):
(StructType.prototype.populateDefaultValue):
* WebGPUShadingLanguageRI/Substitution.js:
(Substitution.prototype.visitTypeRef):
(Substitution.prototype.visitVariableRef):
(Substitution):
* WebGPUShadingLanguageRI/Test.js:
(TEST_genericStructInstance):
* WebGPUShadingLanguageRI/Type.js:
(Type.prototype.inherits):
* WebGPUShadingLanguageRI/TypeRef.js:
(TypeRef.instantiate):
* WebGPUShadingLanguageRI/VariableRef.js:
(VariableRef.wrap):
* WebGPUShadingLanguageRI/Visitor.js:
(Visitor.prototype.visitTypeRef):
(Visitor.prototype.visitDotExpression):
(Visitor.prototype.visitVariableRef):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221550 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL NullType should be a special type variable so it can unify with any pointer type
fpizlo@apple.com [Sun, 3 Sep 2017 17:43:25 +0000 (17:43 +0000)]
WSL NullType should be a special type variable so it can unify with any pointer type
https://bugs.webkit.org/show_bug.cgi?id=176235

Reviewed by Myles Maxfield.

Null is just a type variable that plays by weird rules:

- Other type variables don't recognize it, so if it's unified with T, it fails.
- It behaves like a type variable for pointer and array reference types.
- It arises only in argument (RHS) position.
- Once Checker finds it to be equals to something, we need to remember what that was.

I refactored some of the type system code to allow for the possibility of type variables being
unified during equals() tests. It turns out that we were playing fast and loose here; I needed
to beef up UnificationContext.prototype.verify. I added tests for this.

I added an equalsWithCommit() that triggers commitUnification(), which allows NullType to
remember what type it got unified with.

NullLiteral then just instantiates its own NullType, which is what Checker uses as the type of
the NullLiteral. Everything else just works.

This adds a bunch of tests to make sure we generate type errors in various weird cases while
also accepting a number of safe cases.

I also added a way to run the tests in-browser.

* WebGPUShadingLanguageRI/All.js:
* WebGPUShadingLanguageRI/ArrayRefType.js:
(ArrayRefType.prototype.get isArrayRef):
* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitProtocolDecl.set throw):
* WebGPUShadingLanguageRI/ConstexprTypeParameter.js:
(ConstexprTypeParameter.prototype.typeVariableUnify):
* WebGPUShadingLanguageRI/Evaluator.js:
(Evaluator.prototype.visitDereferenceExpression):
(Evaluator.prototype.visitNullLiteral):
* WebGPUShadingLanguageRI/Func.js:
(Func.prototype.toString):
(Func):
* WebGPUShadingLanguageRI/Intrinsics.js:
(Intrinsics):
* WebGPUShadingLanguageRI/NativeFunc.js:
(NativeFunc.prototype.toDeclString):
(NativeFunc):
(NativeFunc.prototype.toString): Deleted.
* WebGPUShadingLanguageRI/NativeFuncInstance.js:
(NativeFuncInstance.prototype.toDeclString):
(NativeFuncInstance):
(NativeFuncInstance.prototype.toString): Deleted.
* WebGPUShadingLanguageRI/Node.js:
(Node.unify):
(Node.prototype._typeVariableUnifyImpl):
(Node.prototype.equals):
* WebGPUShadingLanguageRI/NullLiteral.js: Added.
(NullLiteral):
(NullLiteral.prototype.toString):
* WebGPUShadingLanguageRI/NullType.js:
(NullType):
(NullType.prototype.get isPrimitive):
(NullType.prototype.get isUnifiable):
(NullType.prototype.typeVariableUnify):
(NullType.prototype.unifyImpl):
(NullType.prototype.verifyAsArgument):
(NullType.prototype.verifyAsParameter):
(NullType.prototype.toString):
* WebGPUShadingLanguageRI/Parse.js:
(parseTerm):
(parseType.getAddressSpace):
(parseType):
* WebGPUShadingLanguageRI/Rewriter.js:
(Rewriter.prototype.visitNullLiteral):
* WebGPUShadingLanguageRI/Test.html: Added.
* WebGPUShadingLanguageRI/Test.js:
(this.window.this.print):
(TEST_doubleGenericCallsSingleGeneric):
(TEST_dereferenceDefaultNull):
(TEST_dereferenceDefaultNullArrayRef):
(doTest):
(let.s.in.this.s.startsWith.string_appeared_here.s.match): Deleted.
* WebGPUShadingLanguageRI/Type.js:
(Type.prototype.get isArrayRef):
* WebGPUShadingLanguageRI/TypeVariable.js:
(TypeVariable.prototype.typeVariableUnify):
* WebGPUShadingLanguageRI/UnificationContext.js:
(UnificationContext.prototype.verify):
(UnificationContext):
(UnificationContext.prototype.isUnunified): Deleted.
* WebGPUShadingLanguageRI/VariableRef.js:
(VariableRef.prototype.get unifyNode):
* WebGPUShadingLanguageRI/Visitor.js:
(Visitor.prototype.visitNullLiteral):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221549 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoLarge virtual memory region allocation requires MMAP_NORESERVE in Linux
utatane.tea@gmail.com [Sun, 3 Sep 2017 10:31:29 +0000 (10:31 +0000)]
Large virtual memory region allocation requires MMAP_NORESERVE in Linux
https://bugs.webkit.org/show_bug.cgi?id=176211

Reviewed by Geoffrey Garen.

In Linux, we cannot allocate very large memory region without MMAP_NORESERVE.
Linux kernel needs to reserve swap area for allocated memory region. If the
swap area is exhausted, kernel fails to allocate the memory region with ENOMEM.

This patch adds MMAP_NORESERVE to mmap flags in Linux. By adding this flag,
mmap does not need to reserve swap area for the reserved memory region.
This allows us to reserve very large memory region that is necessary for Gigacage.

* bmalloc/BPlatform.h:
* bmalloc/VMAllocate.h:
(bmalloc::tryVMAllocate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221548 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove CanvasProxy
commit-queue@webkit.org [Sun, 3 Sep 2017 08:42:11 +0000 (08:42 +0000)]
Remove CanvasProxy
https://bugs.webkit.org/show_bug.cgi?id=176288

Patch by Sam Weinig <sam@webkit.org> on 2017-09-03
Reviewed by Yusuke Suzuki.

CanvasProxy does not appear to be in any current HTML spec
and was disabled and unimplemented in our tree. Time to
get rid of it.

.:

* Source/cmake/OptionsMac.cmake:
* Source/cmake/OptionsWin.cmake:
* Source/cmake/WebKitFeatures.cmake:
* Source/cmake/tools/vsprops/FeatureDefines.props:
* Source/cmake/tools/vsprops/FeatureDefinesCairo.props:

Source/JavaScriptCore:

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

* CMakeLists.txt:
* Configurations/FeatureDefines.xcconfig:
* DerivedSources.cpp:
* DerivedSources.make:
* WebCore.xcodeproj/project.pbxproj:
* html/canvas/CanvasProxy.cpp: Removed.
* html/canvas/CanvasProxy.h: Removed.
* html/canvas/CanvasProxy.idl: Removed.

Source/WebCore/PAL:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit:

* Configurations/FeatureDefines.xcconfig:

Source/WebKitLegacy/mac:

* Configurations/FeatureDefines.xcconfig:

Source/WTF:

* wtf/FeatureDefines.h:

Tools:

* Scripts/webkitperl/FeatureList.pm:
* TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221547 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoNeed an API to get the global context from JSObjectRef
oliver@apple.com [Sun, 3 Sep 2017 04:42:51 +0000 (04:42 +0000)]
Need an API to get the global context from JSObjectRef
https://bugs.webkit.org/show_bug.cgi?id=176291

Reviewed by Saam Barati.

Very simple additional API, starting off as SPI on principle.

* API/JSObjectRef.cpp:
(JSObjectGetGlobalContext):
* API/JSObjectRefPrivate.h:
* API/tests/testapi.c:
(main):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221546 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoDOMMatrix/DOMMatrixReadOnly validation is incorrect for NaN values
commit-queue@webkit.org [Sun, 3 Sep 2017 04:05:00 +0000 (04:05 +0000)]
DOMMatrix/DOMMatrixReadOnly validation is incorrect for NaN values
https://bugs.webkit.org/show_bug.cgi?id=176287

Patch by Sam Weinig <sam@webkit.org> on 2017-09-02
Reviewed by Chris Dumez.

LayoutTests/imported/w3c:

* web-platform-tests/css/geometry-1/DOMMatrixInit-validate-fixup-expected.txt:
Update test results.

Source/WebCore:

* css/DOMMatrixReadOnly.cpp:
(WebCore::sameValueZero): Added.
(WebCore::DOMMatrixReadOnly::validateAndFixup):

    Switch from using standard != to added sameValueZero helper
    which implements https://tc39.github.io/ecma262/#sec-samevaluezero

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221545 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoImplement FileSystemFileEntry.file()
cdumez@apple.com [Sun, 3 Sep 2017 04:02:05 +0000 (04:02 +0000)]
Implement FileSystemFileEntry.file()
https://bugs.webkit.org/show_bug.cgi?id=176166
<rdar://problem/34187756>

Reviewed by Sam Weinig.

Source/WebCore:

Implement FileSystemFileEntry.file():
- https://wicg.github.io/entries-api/#dom-filesystemfileentry-file

Test: editing/pasteboard/datatransfer-items-drop-fileEntry-file.html

* Modules/entriesapi/DOMFileSystem.cpp:
(WebCore::validatePathIsExpectedType):
(WebCore::DOMFileSystem::listDirectory):
(WebCore::DOMFileSystem::getParent):
(WebCore::DOMFileSystem::getFile):
* Modules/entriesapi/DOMFileSystem.h:
* Modules/entriesapi/FileSystemFileEntry.cpp:
(WebCore::FileSystemFileEntry::file):

LayoutTests:

Add layout test coverage. I have verified that this test passes in Chrome as well.

* editing/pasteboard/datatransfer-items-drop-fileEntry-file-expected.txt: Added.
* editing/pasteboard/datatransfer-items-drop-fileEntry-file.html: Added.
* platform/wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221544 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[CA] Upstream QuartzCore-related WebKitSystemInterface functions
aestes@apple.com [Sun, 3 Sep 2017 03:43:08 +0000 (03:43 +0000)]
[CA] Upstream QuartzCore-related WebKitSystemInterface functions
https://bugs.webkit.org/show_bug.cgi?id=176270

Reviewed by Tim Horton.

Source/WebCore:

* platform/graphics/ca/cocoa/PlatformCALayerCocoa.mm:
(PlatformCALayerCocoa::enumerateRectsBeingDrawn):
* platform/graphics/cocoa/WebCoreCALayerExtras.h:
* platform/graphics/cocoa/WebCoreCALayerExtras.mm:
(+[CALayer _web_renderLayerWithContextID:]):
* platform/ios/WebCoreSystemInterfaceIOS.mm:
* platform/mac/WebCoreSystemInterface.h:
* platform/mac/WebCoreSystemInterface.mm:

Source/WebCore/PAL:

* pal/spi/cocoa/QuartzCoreSPI.h:

Source/WebKit:

* Platform/mac/LayerHostingContext.mm:
(WebKit::LayerHostingContext::createForPort):
(WebKit::LayerHostingContext::createForExternalHostingProcess):
* UIProcess/mac/PageClientImpl.mm:
(WebKit::PageClientImpl::enterAcceleratedCompositingMode):
(WebKit::PageClientImpl::updateAcceleratedCompositingMode):
* UIProcess/mac/RemoteLayerTreeHost.mm:
(WebKit::RemoteLayerTreeHost::createLayer):
* WebProcess/Plugins/Netscape/mac/PluginProxyMac.mm:
(WebKit::makeRenderLayer):
* WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:
(InitWebCoreSystemInterface):

Source/WebKitLegacy/mac:

* Plugins/Hosted/NetscapePluginHostManager.mm:
(WebKit::NetscapePluginHostManager::spawnPluginHost):
* Plugins/Hosted/WebHostedNetscapePluginView.h:
* Plugins/Hosted/WebHostedNetscapePluginView.mm:
(WebKit::SoftwareCARenderer::SoftwareCARenderer):
(WebKit::SoftwareCARenderer::~SoftwareCARenderer):
(WebKit::SoftwareCARenderer::context const):
(WebKit::SoftwareCARenderer::renderContext const):
(WebKit::SoftwareCARenderer::render):
(-[WebHostedNetscapePluginView createPlugin]):
(-[WebHostedNetscapePluginView createPluginLayer]):
(-[WebHostedNetscapePluginView destroyPlugin]):
(-[WebHostedNetscapePluginView drawRect:]):
* WebCoreSupport/WebSystemInterface.mm:
(InitWebCoreSystemInterface):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221543 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWrong getComputedStyle result for pseudo-elements in display: none subtrees.
commit-queue@webkit.org [Sun, 3 Sep 2017 02:31:13 +0000 (02:31 +0000)]
Wrong getComputedStyle result for pseudo-elements in display: none subtrees.
https://bugs.webkit.org/show_bug.cgi?id=176259

Patch by Emilio Cobos Álvarez <emilio@crisal.io> on 2017-09-02
Reviewed by Antti Koivisto.

LayoutTests/imported/w3c:

* web-platform-tests/cssom/getComputedStyle-pseudo-expected.txt:

Source/WebCore:

Before this patch, we were wrongly returning the parent style when there wasn't
a cached pseudo style.

With this, the only remaining failures in getComputedStyle-pseudo is a
serialisation bug for the "content" property, and the display: contents test,
which is because we don't support pseudo-elements of display: contents elements
yet.

Tests: imported/w3c/web-platform-tests/cssom/getComputedStyle-pseudo.html

* dom/Document.cpp:
(WebCore::Document::styleForElementIgnoringPendingStylesheets):
* dom/Document.h:
* dom/Element.cpp:
(WebCore::Element::resolvePseudoElementStyle):
(WebCore::Element::computedStyle):
* dom/Element.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221542 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMove UI Process part of the Remote Layer Tree code out of a Mac-specific directory
timothy_horton@apple.com [Sat, 2 Sep 2017 23:52:34 +0000 (23:52 +0000)]
Move UI Process part of the Remote Layer Tree code out of a Mac-specific directory
https://bugs.webkit.org/show_bug.cgi?id=176273

Reviewed by Dan Bernstein.

UI-side compositing works on all Cocoa platforms, so the code should
not live inside Mac/. Also merge the Scrolling directory in here,
since those files are very related.

* UIProcess/RemoteLayerTree/RemoteLayerTreeDrawingAreaProxy.h: Renamed from SourceebKit/UIProcess/mac/RemoteLayerTreeDrawingAreaProxy.h.
* UIProcess/RemoteLayerTree/RemoteLayerTreeDrawingAreaProxy.messages.in: Renamed from UIProcess/mac/RemoteLayerTreeDrawingAreaProxy.messages.in.
* UIProcess/RemoteLayerTree/RemoteLayerTreeDrawingAreaProxy.mm: Renamed from UIProcess/mac/RemoteLayerTreeDrawingAreaProxy.mm.
* UIProcess/RemoteLayerTree/RemoteLayerTreeHost.h: Renamed from UIProcess/mac/RemoteLayerTreeHost.h.
* UIProcess/RemoteLayerTree/RemoteLayerTreeHost.mm: Renamed from UIProcess/mac/RemoteLayerTreeHost.mm.
* UIProcess/RemoteLayerTree/RemoteLayerTreeScrollingPerformanceData.h: Renamed from UIProcess/Cocoa/RemoteLayerTreeScrollingPerformanceData.h.
* UIProcess/RemoteLayerTree/RemoteLayerTreeScrollingPerformanceData.mm: Renamed from UIProcess/Cocoa/RemoteLayerTreeScrollingPerformanceData.mm.
* UIProcess/RemoteLayerTree/RemoteScrollingCoordinatorProxy.cpp: Renamed from UIProcess/Scrolling/RemoteScrollingCoordinatorProxy.cpp.
* UIProcess/RemoteLayerTree/RemoteScrollingCoordinatorProxy.h: Renamed from UIProcess/Scrolling/RemoteScrollingCoordinatorProxy.h.
* UIProcess/RemoteLayerTree/RemoteScrollingTree.cpp: Renamed from UIProcess/Scrolling/RemoteScrollingTree.cpp.
* UIProcess/RemoteLayerTree/RemoteScrollingTree.h: Renamed from UIProcess/Scrolling/RemoteScrollingTree.h.
* UIProcess/RemoteLayerTree/ios/RemoteLayerTreeHostIOS.mm: Renamed from UIProcess/ios/RemoteLayerTreeHostIOS.mm.
* UIProcess/RemoteLayerTree/ios/RemoteScrollingCoordinatorProxyIOS.mm: Renamed from UIProcess/ios/RemoteScrollingCoordinatorProxyIOS.mm.
* UIProcess/RemoteLayerTree/ios/ScrollingTreeOverflowScrollingNodeIOS.h: Renamed from UIProcess/Scrolling/ios/ScrollingTreeOverflowScrollingNodeIOS.h.
* UIProcess/RemoteLayerTree/ios/ScrollingTreeOverflowScrollingNodeIOS.mm: Renamed from UIProcess/Scrolling/ios/ScrollingTreeOverflowScrollingNodeIOS.mm.
* WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221541 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoImplement FileSystemDirectoryEntry.getDirectory()
cdumez@apple.com [Sat, 2 Sep 2017 21:31:49 +0000 (21:31 +0000)]
Implement FileSystemDirectoryEntry.getDirectory()
https://bugs.webkit.org/show_bug.cgi?id=176168
<rdar://problem/34187787>

Reviewed by Darin Adler.

Source/WebCore:

Implement FileSystemDirectoryEntry.getDirectory() as per:
- https://wicg.github.io/entries-api/#dom-filesystemdirectoryentry-getdirectory

Test: editing/pasteboard/datatransfer-items-drop-getDirectory.html

* Modules/entriesapi/DOMFileSystem.cpp:
(WebCore::isValidVirtualPath):
(WebCore::resolveRelativeVirtualPath):
(WebCore::DOMFileSystem::getEntry):
* Modules/entriesapi/DOMFileSystem.h:
* Modules/entriesapi/FileSystemDirectoryEntry.cpp:
(WebCore::FileSystemDirectoryEntry::getEntry):
(WebCore::FileSystemDirectoryEntry::getFile):
(WebCore::FileSystemDirectoryEntry::getDirectory):
* Modules/entriesapi/FileSystemDirectoryEntry.h:

LayoutTests:

Add layout test coverage.

* editing/pasteboard/datatransfer-items-drop-getDirectory-expected.txt: Added.
* editing/pasteboard/datatransfer-items-drop-getDirectory.html: Added.
* platform/wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221540 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoPerformance / PerformanceObserver leaks seen on bots
commit-queue@webkit.org [Sat, 2 Sep 2017 18:31:00 +0000 (18:31 +0000)]
Performance / PerformanceObserver leaks seen on bots
https://bugs.webkit.org/show_bug.cgi?id=176269

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2017-09-02
Reviewed by Darin Adler.

* page/DOMWindow.cpp:
(WebCore::DOMWindow::removeAllEventListeners):
* workers/WorkerGlobalScope.cpp:
(WebCore::WorkerGlobalScope::removeAllEventListeners):
Also use this opportunity to break up the reference cycle
between PerformanceObservers and the Performance object.

* page/Performance.h:
* page/Performance.cpp:
(WebCore::Performance::removeAllObservers):
Disassociate and clear observers.

* page/PerformanceObserver.h:
* page/PerformanceObserver.cpp:
(WebCore::PerformanceObserver::disassociate):
Clear the back-reference to the Performance object to
break the cycle.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221539 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWeb Automation: -[_WKAutomationSession init] leaks
commit-queue@webkit.org [Sat, 2 Sep 2017 18:29:40 +0000 (18:29 +0000)]
Web Automation: -[_WKAutomationSession init] leaks
https://bugs.webkit.org/show_bug.cgi?id=176268

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2017-09-02
Reviewed by Darin Adler.

* UIProcess/API/Cocoa/_WKAutomationSession.mm:
(-[_WKAutomationSession init]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221538 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoFix some style issues in the Web Process part of the Remote Layer Tree code
timothy_horton@apple.com [Sat, 2 Sep 2017 10:48:34 +0000 (10:48 +0000)]
Fix some style issues in the Web Process part of the Remote Layer Tree code
https://bugs.webkit.org/show_bug.cgi?id=176272

Reviewed by Dan Bernstein.

In bug 176271, the style bot noticed some existing style issues in the UI-side
compositing code. Fix them, because there aren't many and they're straightforward.

* WebProcess/WebPage/Cocoa/RemoteLayerTree/GraphicsLayerCARemote.cpp:
* WebProcess/WebPage/Cocoa/RemoteLayerTree/PlatformCAAnimationRemote.h:
* WebProcess/WebPage/Cocoa/RemoteLayerTree/PlatformCAAnimationRemote.mm:
(WebKit::PlatformCAAnimationRemote::setValues):
* WebProcess/WebPage/Cocoa/RemoteLayerTree/PlatformCALayerRemote.cpp:
(WebKit::PlatformCALayerRemote::appendSublayer):
(WebKit::PlatformCALayerRemote::insertSublayer):
(WebKit::PlatformCALayerRemote::replaceSublayer):
* WebProcess/WebPage/Cocoa/RemoteLayerTree/PlatformCALayerRemote.h:
* WebProcess/WebPage/Cocoa/RemoteLayerTree/PlatformCALayerRemoteCustom.mm:
(WebKit::PlatformCALayerRemoteCustom::clone const):
* WebProcess/WebPage/Cocoa/RemoteLayerTree/PlatformCALayerRemoteTiledBacking.h:
* WebProcess/WebPage/Cocoa/RemoteLayerTree/RemoteLayerTreeContext.h:
* WebProcess/WebPage/Cocoa/RemoteLayerTree/RemoteLayerTreeDrawingArea.mm:
(WebKit::RemoteLayerTreeDrawingArea::flushLayers):
* WebProcess/WebPage/Cocoa/RemoteLayerTree/RemoteScrollingCoordinator.h:
* WebProcess/WebPage/Cocoa/RemoteLayerTree/RemoteScrollingCoordinator.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221537 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMove Web Process part of the Remote Layer Tree code out of a Mac-specific directory
timothy_horton@apple.com [Sat, 2 Sep 2017 09:38:13 +0000 (09:38 +0000)]
Move Web Process part of the Remote Layer Tree code out of a Mac-specific directory
https://bugs.webkit.org/show_bug.cgi?id=176271

Rubber-stamped by Dan Bernstein.

UI-side compositing works on all Cocoa platforms, so the code should
not live inside Mac/. Also merge the Scrolling directory in here,
since those files are very related.

* WebKit.xcodeproj/project.pbxproj:
* DerivedSources.make:
* PlatformMac.cmake:
* WebProcess/WebPage/RemoteLayerTree/GraphicsLayerCARemote.cpp: Renamed from WebProcess/WebPage/mac/GraphicsLayerCARemote.cpp.
* WebProcess/WebPage/RemoteLayerTree/GraphicsLayerCARemote.h: Renamed from WebProcess/WebPage/mac/GraphicsLayerCARemote.h.
* WebProcess/WebPage/RemoteLayerTree/PlatformCAAnimationRemote.h: Renamed from WebProcess/WebPage/mac/PlatformCAAnimationRemote.h.
* WebProcess/WebPage/RemoteLayerTree/PlatformCAAnimationRemote.mm: Renamed from WebProcess/WebPage/mac/PlatformCAAnimationRemote.mm.
* WebProcess/WebPage/RemoteLayerTree/PlatformCALayerRemote.cpp: Renamed from WebProcess/WebPage/mac/PlatformCALayerRemote.cpp.
* WebProcess/WebPage/RemoteLayerTree/PlatformCALayerRemote.h: Renamed from WebProcess/WebPage/mac/PlatformCALayerRemote.h.
* WebProcess/WebPage/RemoteLayerTree/PlatformCALayerRemoteCustom.h: Renamed from WebProcess/WebPage/mac/PlatformCALayerRemoteCustom.h.
* WebProcess/WebPage/RemoteLayerTree/PlatformCALayerRemoteCustom.mm: Renamed from WebProcess/WebPage/mac/PlatformCALayerRemoteCustom.mm.
* WebProcess/WebPage/RemoteLayerTree/PlatformCALayerRemoteTiledBacking.cpp: Renamed from WebProcess/WebPage/mac/PlatformCALayerRemoteTiledBacking.cpp.
* WebProcess/WebPage/RemoteLayerTree/PlatformCALayerRemoteTiledBacking.h: Renamed from WebProcess/WebPage/mac/PlatformCALayerRemoteTiledBacking.h.
* WebProcess/WebPage/RemoteLayerTree/RemoteLayerTreeContext.h: Renamed from WebProcess/WebPage/mac/RemoteLayerTreeContext.h.
* WebProcess/WebPage/RemoteLayerTree/RemoteLayerTreeContext.mm: Renamed from WebProcess/WebPage/mac/RemoteLayerTreeContext.mm.
* WebProcess/WebPage/RemoteLayerTree/RemoteLayerTreeDisplayRefreshMonitor.h: Renamed from WebProcess/WebPage/Cocoa/RemoteLayerTreeDisplayRefreshMonitor.h.
* WebProcess/WebPage/RemoteLayerTree/RemoteLayerTreeDisplayRefreshMonitor.mm: Renamed from WebProcess/WebPage/Cocoa/RemoteLayerTreeDisplayRefreshMonitor.mm.
* WebProcess/WebPage/RemoteLayerTree/RemoteLayerTreeDrawingArea.h: Renamed from WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.h.
* WebProcess/WebPage/RemoteLayerTree/RemoteLayerTreeDrawingArea.mm: Renamed from WebProcess/WebPage/mac/RemoteLayerTreeDrawingArea.mm.
* WebProcess/WebPage/RemoteLayerTree/RemoteScrollingCoordinator.h: Renamed from WebProcess/Scrolling/RemoteScrollingCoordinator.h.
* WebProcess/WebPage/RemoteLayerTree/RemoteScrollingCoordinator.messages.in: Renamed from WebProcess/Scrolling/RemoteScrollingCoordinator.messages.in.
* WebProcess/WebPage/RemoteLayerTree/RemoteScrollingCoordinator.mm: Renamed from WebProcess/Scrolling/RemoteScrollingCoordinator.mm.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221536 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[DFG] Relax arity requirement
utatane.tea@gmail.com [Sat, 2 Sep 2017 08:35:46 +0000 (08:35 +0000)]
[DFG] Relax arity requirement
https://bugs.webkit.org/show_bug.cgi?id=175523

Reviewed by Saam Barati.

JSTests:

* stress/arity-mismatch-arguments-length.js: Added.
(shouldBe):
(test1):
(test):
* stress/arity-mismatch-get-argument.js: Added.
(shouldBe):
(builtin.createBuiltin):
(test):
* stress/arity-mismatch-inlining-extra-slots.js: Added.
(shouldBe):
(inlineTarget):
(test):
* stress/arity-mismatch-inlining.js: Added.
(shouldBe):
(inlineTarget):
(test):
* stress/arity-mismatch-rest.js: Added.
(shouldBe):
(test2):
(test1):
(test):

Source/JavaScriptCore:

Our DFG pipeline gives up inlining when the arity of the target function is more than the number of the arguments.
It effectively prevents us from inlining and optimizing functions, which takes some optional arguments in the form
of the pre-ES6.

This patch removes the above restriction by performing the arity fixup in DFG.

SixSpeed shows improvement when we can inline arity-mismatched functions. (For example, calling generator.next()).

                               baseline                  patched

defaults.es5             1232.1226+-20.6775    ^    442.3326+-26.1883       ^ definitely 2.7855x faster
rest.es6                    5.3406+-0.8588     ^      3.5812+-0.5388        ^ definitely 1.4913x faster
spread-generator.es6      320.9107+-12.4808         310.4295+-12.0047         might be 1.0338x faster
generator.es6             318.3514+-9.6023     ^    286.4974+-12.6203       ^ definitely 1.1112x faster

* bytecode/InlineCallFrame.cpp:
(JSC::InlineCallFrame::dumpInContext const):
* bytecode/InlineCallFrame.h:
(JSC::InlineCallFrame::InlineCallFrame):
* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):
* dfg/DFGArgumentsEliminationPhase.cpp:
* dfg/DFGArgumentsUtilities.cpp:
(JSC::DFG::argumentsInvolveStackSlot):
(JSC::DFG::emitCodeToGetArgumentsArrayLength):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::setLocal):
(JSC::DFG::ByteCodeParser::setArgument):
(JSC::DFG::ByteCodeParser::findArgumentPositionForLocal):
(JSC::DFG::ByteCodeParser::flush):
(JSC::DFG::ByteCodeParser::getArgumentCount):
(JSC::DFG::ByteCodeParser::inliningCost):
(JSC::DFG::ByteCodeParser::inlineCall):
(JSC::DFG::ByteCodeParser::attemptToInlineCall):
(JSC::DFG::ByteCodeParser::parseBlock):
(JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry):
* dfg/DFGCommonData.cpp:
(JSC::DFG::CommonData::validateReferences):
* dfg/DFGConstantFoldingPhase.cpp:
(JSC::DFG::ConstantFoldingPhase::foldConstants):
* dfg/DFGGraph.cpp:
(JSC::DFG::Graph::isLiveInBytecode):
* dfg/DFGGraph.h:
(JSC::DFG::Graph::forAllLocalsLiveInBytecode):
* dfg/DFGOSRAvailabilityAnalysisPhase.cpp:
(JSC::DFG::LocalOSRAvailabilityCalculator::executeNode):
* dfg/DFGOSRExit.cpp:
(JSC::DFG::OSRExit::emitRestoreArguments):
* dfg/DFGOSRExitCompilerCommon.cpp:
(JSC::DFG::reifyInlinedCallFrames):
* dfg/DFGPreciseLocalClobberize.h:
(JSC::DFG::PreciseLocalClobberizeAdaptor::readTop):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::emitGetLength):
(JSC::DFG::SpeculativeJIT::compileCreateDirectArguments):
* dfg/DFGStackLayoutPhase.cpp:
(JSC::DFG::StackLayoutPhase::run):
* ftl/FTLCompile.cpp:
(JSC::FTL::compile):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileGetMyArgumentByVal):
(JSC::FTL::DFG::LowerDFGToB3::getArgumentsLength):
* ftl/FTLOperations.cpp:
(JSC::FTL::operationMaterializeObjectInOSR):
* interpreter/StackVisitor.cpp:
(JSC::StackVisitor::readInlinedFrame):
* jit/AssemblyHelpers.h:
(JSC::AssemblyHelpers::argumentsStart):
* jit/SetupVarargsFrame.cpp:
(JSC::emitSetupVarargsFrameFastCase):
* runtime/ClonedArguments.cpp:
(JSC::ClonedArguments::createWithInlineFrame):
* runtime/CommonSlowPaths.h:
(JSC::CommonSlowPaths::numberOfExtraSlots):
(JSC::CommonSlowPaths::numberOfStackPaddingSlots):
(JSC::CommonSlowPaths::numberOfStackPaddingSlotsWithExtraSlots):
(JSC::CommonSlowPaths::arityCheckFor):
* runtime/StackAlignment.h:
(JSC::stackAlignmentBytes):
(JSC::stackAlignmentRegisters):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221528 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GTK][Wayland] Use fast malloc to allocate backing store cairo surfaces data
carlosgc@webkit.org [Sat, 2 Sep 2017 08:10:53 +0000 (08:10 +0000)]
[GTK][Wayland] Use fast malloc to allocate backing store cairo surfaces data
https://bugs.webkit.org/show_bug.cgi?id=176226

Reviewed by Michael Catanzaro.

Source/WebCore:

It reduces the memory consumption and improves the performance. Also, since performance is better we don't need
to keep the scroll surface allocated all the time, we can create it on demand and keep it allocated only while
scrolling too fast.

* platform/graphics/cairo/BackingStoreBackendCairoImpl.cpp:
(WebCore::createCairoImageSurfaceWithFastMalloc): Helper to create a cairo image surface using fast malloc.
(WebCore::BackingStoreBackendCairoImpl::BackingStoreBackendCairoImpl): Create the main surface, but no the
scroll one. Initialize the hysteresis activity to discard the scroll surface after scrolling.
(WebCore::BackingStoreBackendCairoImpl::scroll): Create the scroll surface only if it's nullptr, and notify the
hysteresis activity.
* platform/graphics/cairo/BackingStoreBackendCairoImpl.h:

Source/WebKit:

Do not create the cairo surface for BackingStoreBackendCairoImpl. When not using X11 we are going to use an
image surface anyway, so we don't really need to use the create similar. This way we only need to pass the size
and device scale factor to BackingStoreBackendCairoImpl that will create the image surface using fast malloc to
allocate the data.

* UIProcess/cairo/BackingStoreCairo.cpp:
(WebKit::BackingStore::createBackend):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221521 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GTK][Wayland] Opening FedoraProject's pastebin chews CPU
carlosgc@webkit.org [Sat, 2 Sep 2017 07:05:46 +0000 (07:05 +0000)]
[GTK][Wayland] Opening FedoraProject's pastebin chews CPU
https://bugs.webkit.org/show_bug.cgi?id=175942

Reviewed by Žan Doberšek.

This regressed when we introduced the display refresh monitor. The monitor schedules updates immediately,
because we removed the option to not do frame sync in X11 to let swapBuffers do the throttling, but that's
not happening in Wayland because the nested compositor is dispatching frame callbacks on surface commit.
We need to ensure that frame callbacks are dispatched on every monitor refresh, because swapBuffers waits for
frame callbacks to be queued on display.

* UIProcess/gtk/WaylandCompositor.cpp:
(WebKit::WaylandCompositor::Surface::~Surface): Destroy pending frame callbacks too.
(WebKit::WaylandCompositor::Surface::setWebPage): Add a tick callback to the web view widget to flush all
committed frame callbacks on every frame update.
(WebKit::WaylandCompositor::Surface::requestFrame): Add the callbacks to m_pendingFrameCallbackList.
(WebKit::WaylandCompositor::Surface::flushFrameCallbacks): Dispatch all committed frame callabcks.
(WebKit::WaylandCompositor::Surface::flushPendingFrameCallbacks): Dispatch all pending frame callbacks.
(WebKit::WaylandCompositor::Surface::commit): Do not dispatch frame callbacks here, move them to the list of
committed frame callbacks that will be dispatched on the next frame clock update.
* UIProcess/gtk/WaylandCompositor.h:
(WebKit::WaylandCompositor::Surface::setWebPage): Moved to the cpp.
* WebProcess/WebPage/gtk/AcceleratedSurfaceWayland.cpp:
(WebKit::AcceleratedSurfaceWayland::AcceleratedSurfaceWayland): Move surface initialization and destruction to
the compositing thread.
(WebKit::AcceleratedSurfaceWayland::initialize):
(WebKit::AcceleratedSurfaceWayland::finalize):
(WebKit::AcceleratedSurfaceWayland::~AcceleratedSurfaceWayland): Deleted.
* WebProcess/WebPage/gtk/AcceleratedSurfaceWayland.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL should check unreachable code and no returns
fpizlo@apple.com [Sat, 2 Sep 2017 02:33:46 +0000 (02:33 +0000)]
WSL should check unreachable code and no returns
https://bugs.webkit.org/show_bug.cgi?id=176262

Reviewed by Myles Maxfield.

This adds phases to check whether non-void functions return and whether there is any code after
a return in any block. Added tests for simple versions of both of those things. We cannot add
more complex versions until we support control flow statements.

* WebGPUShadingLanguageRI/All.js:
* WebGPUShadingLanguageRI/CheckReturns.js: Added.
(checkReturns):
* WebGPUShadingLanguageRI/CheckUnreachableCode.js: Added.
(checkUnreachableCode):
* WebGPUShadingLanguageRI/Parse.js:
(parseReturn):
* WebGPUShadingLanguageRI/Prepare.js:
(prepare):
* WebGPUShadingLanguageRI/ReturnChecker.js: Added.
(ReturnChecker.prototype.visitFuncDef):
(ReturnChecker.prototype.visitBlock):
(ReturnChecker.prototype.visitReturn):
(ReturnChecker):
* WebGPUShadingLanguageRI/Test.js:
(TEST_simpleNoReturn):
* WebGPUShadingLanguageRI/UnreachableCodeChecker.js: Added.
(UnreachableCodeChecker.prototype.visitBlock):
(UnreachableCodeChecker):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221513 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoDOMMatrix and DOMMatrixReadOnly should be available in workers
weinig@apple.com [Sat, 2 Sep 2017 01:38:07 +0000 (01:38 +0000)]
DOMMatrix and DOMMatrixReadOnly should be available in workers
https://bugs.webkit.org/show_bug.cgi?id=176255

Reviewed by Simon Fraser.

LayoutTests/imported/w3c:

* web-platform-tests/css/geometry-1/DOMMatrix-css-string.worker-expected.txt:
* web-platform-tests/css/geometry-1/interfaces.worker-expected.txt:
Update test results.

Source/WebCore:

* WebCore.xcodeproj/project.pbxproj:

    Add GlobalPerformance.idl to the project. It was missing.

* bindings/scripts/CodeGeneratorJS.pm:
(IsAlwaysExposedOnInterface):
(NeedsRuntimeCheck):
(GenerateRuntimeEnableConditionalStringForExposed):
(GenerateRuntimeEnableConditionalString):

    Adds support for [Exposed] extended attributes on operations, attributes and
    constants. This allows an interface that is exposed to both windows and workers
    (with [Exposed=(Window,Worker)]), to indicate that a specific property should
    only be be exposed somewhere, e.g only the window. This is need by DOMMatrixReadOnly's
    stringifier, which is only exposed in the window environment.

* bindings/scripts/IDLParser.pm:
(parseExtendedAttributeRest2):

    Fix bug where array extended attributes, like those found in Exposed expressions
    where being converter to scalars.

* bindings/scripts/test/JS/JSTestGenerateIsReachable.cpp:
* bindings/scripts/test/JS/JSTestGlobalObject.cpp:
* bindings/scripts/test/JS/JSTestObj.cpp:
* bindings/scripts/test/TestObj.idl:

    Add tests and update results. Some results are just improved
    includes.

* css/DOMMatrix.cpp:
* css/DOMMatrix.h:
* css/DOMMatrix.idl:
* css/DOMMatrixReadOnly.cpp:
* css/DOMMatrixReadOnly.h:
* css/DOMMatrixReadOnly.idl:

    Expose DOMMatrix and DOMMatrixReadOnly to workers and update the constructors
    to match the current spec, throwing if the argument is a string, and the context
    is not the window / document context. Also did some refactoring to avoid allocating
    the DOMMatrix or DOMMatrixReadOnly until after all the exceptional cases were handled.

LayoutTests:

* js/dom/global-constructors-attributes-dedicated-worker-expected.txt:
* platform/mac-wk1/js/dom/global-constructors-attributes-dedicated-worker-expected.txt
* platform/wpe/js/dom/global-constructors-attributes-dedicated-worker-expected.txt

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221512 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL should use & for getting a ^ and @ for getting a []
fpizlo@apple.com [Sat, 2 Sep 2017 00:57:27 +0000 (00:57 +0000)]
WSL should use & for getting a ^ and @ for getting a []
https://bugs.webkit.org/show_bug.cgi?id=176245

Reviewed by Myles Maxfield.

Using \ as an operator was a bad decision since that's an escape character in so many contexts.

Henceforth, & is for making pointers, for example:

thread int^ foo() { int x; return &x; }

And @ is for making array references, for example:

thread int[] foo() { int[42] x; return @x; }

Having separate operators makes it easy to have pointers to fixed-size arrays:

thread int[42]^ foo() { int[42] x; return &x; }

* WebGPUShadingLanguageRI/Intrinsics.js:
(Intrinsics):
* WebGPUShadingLanguageRI/Lexer.js:
(Lexer.prototype.next):
(Lexer):
* WebGPUShadingLanguageRI/MakePtrExpression.js:
(MakePtrExpression.prototype.toString):
(MakePtrExpression):
* WebGPUShadingLanguageRI/Parse.js:
(parsePossibleSuffix):
(parsePossiblePrefix):
(parseFuncName):
* WebGPUShadingLanguageRI/StandardLibrary.js:
* WebGPUShadingLanguageRI/Test.js:
(TEST_simpleMakePtr):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221511 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoImplement FileSystemDirectoryEntry.getFile()
cdumez@apple.com [Sat, 2 Sep 2017 00:19:55 +0000 (00:19 +0000)]
Implement FileSystemDirectoryEntry.getFile()
https://bugs.webkit.org/show_bug.cgi?id=176167
<rdar://problem/34187775>

Reviewed by Andreas Kling.

Source/WebCore:

Implement FileSystemDirectoryEntry.getFile():
- https://wicg.github.io/entries-api/#dom-filesystemdirectoryentry-getfile

Test: editing/pasteboard/datatransfer-items-drop-getFile.html

* Modules/entriesapi/DOMFileSystem.cpp:
(WebCore::isValidPathNameCharacter):
(WebCore::isValidPathSegment):
(WebCore::isValidRelativeVirtualPath):
(WebCore::isValidAbsoluteVirtualPath):
(WebCore::isValidVirtualPath):
Implement various path validation functions as per:
- https://wicg.github.io/entries-api/#names-paths

(WebCore::resolveRelativeVirtualPath):
- Use StringView for second parameter for efficiency. Had to keep
  a String for the first parameter because I need the split to
  return a Vector.
- If the input path is absolute, call recursively with '/' as
  base path so that the virtual path gets sanitized (i.e. '..'
  and '.' in the absolute URL get resolved).

(WebCore::DOMFileSystem::listDirectory):
Use String instead of auto. It is not much longer and is clearer.

(WebCore::validatePathIsDirectory):
(WebCore::DOMFileSystem::getParent):
Move logic to validate that the path is a directory from getParent() to a
separate function. This only makes sure the ScriptExecutionContext is only
ref'd / deref'd on the main thread.

(WebCore::validatePathIsFile):
Logic for validating that a path is a file, used by getFile().

(WebCore::DOMFileSystem::getFile):
Implement getFile() as per:
- https://wicg.github.io/entries-api/#dom-filesystemdirectoryentry-getfile

* Modules/entriesapi/DOMFileSystem.h:
* Modules/entriesapi/FileSystemDirectoryEntry.cpp:
(WebCore::FileSystemDirectoryEntry::getFile):
Add implementation of FileSystemDirectoryEntry::getFile() which merely calls
DOMFileSystem::getFile().

LayoutTests:

Add layout test coverage.

* editing/pasteboard/datatransfer-items-drop-getFile-expected.txt: Added.
* editing/pasteboard/datatransfer-items-drop-getFile.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221510 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoASSERTION FAILED: taken.get() == &job in WebCore::ServiceWorkerContainer::jobDidFinis...
beidson@apple.com [Fri, 1 Sep 2017 23:59:21 +0000 (23:59 +0000)]
ASSERTION FAILED: taken.get() == &job in WebCore::ServiceWorkerContainer::jobDidFinish(WebCore::ServiceWorkerJob &).
https://bugs.webkit.org/show_bug.cgi?id=176234

Rubberstamped by Tim Horton, reluctantly.

* workers/service/ServiceWorkerContainer.cpp:
(WebCore::ServiceWorkerContainer::jobDidFinish): It's valid for the job to be missing after navigations, which is
  why this was hitting downstream tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL's lexer will never emit keyword tokens
mmaxfield@apple.com [Fri, 1 Sep 2017 23:54:23 +0000 (23:54 +0000)]
WSL's lexer will never emit keyword tokens
https://bugs.webkit.org/show_bug.cgi?id=176248

Reviewed by Filip Pizlo.

Because all tokens are also identified as idents, we need to handle them together.

* WebGPUShadingLanguageRI/Lexer.js:
(Lexer.prototype.next):
(Lexer):
* WebGPUShadingLanguageRI/Test.js:
(doLex):
(checkLexerToken):
(TEST_lexerKeyword):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221508 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSkipped multiple webgl tests.
jlewis3@apple.com [Fri, 1 Sep 2017 23:43:42 +0000 (23:43 +0000)]
Skipped multiple webgl tests.
<rdar://problem/34209038>

Unreviewed test gardening.

* platform/ios-11/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoDo not Reject CacheStorage promises when updating the persistent filesystem data...
commit-queue@webkit.org [Fri, 1 Sep 2017 23:28:09 +0000 (23:28 +0000)]
Do not Reject CacheStorage promises when updating the persistent filesystem data fails
https://bugs.webkit.org/show_bug.cgi?id=176241

Patch by Youenn Fablet <youenn@apple.com> on 2017-09-01
Reviewed by Alex Christensen.

Source/WebCore:

Open/Remove caches may succeed in updating the memory representation but the write-to-disk operation may fail.
In that case, the corresponding promise is now resolved since the web page can still proceeed.

There is no way to test that currently as this would require a write disk failure to be triggered.

* Modules/cache/CacheStorage.cpp:
(WebCore::logConsolePersistencyError):
(WebCore::CacheStorage::open):
(WebCore::CacheStorage::remove):
* Modules/cache/DOMCacheEngine.h:
(WebCore::DOMCacheEngine::CacheIdentifierOperationResult::encode const):
(WebCore::DOMCacheEngine::CacheIdentifierOperationResult::decode):

Source/WebKit:

Open/Remove caches may succeed in the memory representation but the write-to-disk operation may fail.
In that case, the callback does not return an error but a value containing the cache identifier.
The value will also contain a boolean flag set to true to notify the client that persistent storage failed this time.

* NetworkProcess/cache/CacheStorageEngine.cpp:
(WebKit::CacheStorage::Engine::open):
(WebKit::CacheStorage::Engine::remove):
* NetworkProcess/cache/CacheStorageEngineCaches.cpp:
(WebKit::CacheStorage::Caches::open):
(WebKit::CacheStorage::Caches::remove):
* NetworkProcess/cache/CacheStorageEngineCaches.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRename WebKit2 API tests after directory rename
achristensen@apple.com [Fri, 1 Sep 2017 22:56:15 +0000 (22:56 +0000)]
Rename WebKit2 API tests after directory rename
https://bugs.webkit.org/show_bug.cgi?id=176250

Reviewed by Tim Horton.

* TestWebKitAPI/Tests/Lots of files...
Renamed WebKit2 to WebKit and WebKit1 to WebKitLegacy.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221505 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Fetch API] Add support for consuming a Request ReadableStream body
commit-queue@webkit.org [Fri, 1 Sep 2017 22:53:23 +0000 (22:53 +0000)]
[Fetch API] Add support for consuming a Request ReadableStream body
https://bugs.webkit.org/show_bug.cgi?id=176182

Patch by Youenn Fablet <youenn@apple.com> on 2017-09-01
Reviewed by Alex Christensen.

Source/WebCore:

Tests: http/wpt/fetch/request-stream-empty.html
       http/wpt/fetch/request-stream-error.html

Adding a ReadableStreamSink which allows to pipe all ReadableStream data back to C++ code.
This sink only takes BufferSource as the only user is Fetch which has this restriction.
This sink is tied to the readableStreamPipeTo builtin internal function that reads all data from the ReadableStream
and send it to the sink.

Adding a ReadableStreamToSharedBufferSink specialization to get the data as a SharedBuffer.

Use that class in FetchBodyConsumer to get the data back from the ReadableStream and transform it according
the desired body type (text, JSON, ArrayBuffer).

* CMakeLists.txt:
* DerivedSources.make:
* Modules/fetch/FetchBody.cpp:
(WebCore::FetchBody::consume):
(WebCore::FetchBody::loadingSucceeded):
* Modules/fetch/FetchBodyConsumer.cpp:
(WebCore::resolveWithTypeAndData):
(WebCore::FetchBodyConsumer::clean):
(WebCore::FetchBodyConsumer::resolveWithData):
(WebCore::FetchBodyConsumer::resolve):
* Modules/fetch/FetchBodyConsumer.h:
* Modules/fetch/FetchResponse.cpp:
(WebCore::FetchResponse::finishConsumingStream):
* Modules/streams/ReadableStreamInternals.js:
(readableStreamPipeTo):
* Modules/streams/ReadableStreamSink.cpp: Added.
(WebCore::ReadableStreamToSharedBufferSink::ReadableStreamToSharedBufferSink):
(WebCore::ReadableStreamToSharedBufferSink::pipeFrom):
(WebCore::ReadableStreamToSharedBufferSink::enqueue):
(WebCore::ReadableStreamToSharedBufferSink::close):
(WebCore::ReadableStreamToSharedBufferSink::error):
* Modules/streams/ReadableStreamSink.h: Added.
* Modules/streams/ReadableStreamSink.idl: Added.
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/ReadableStream.cpp:
(WebCore::ReadableStream::pipeTo):
* bindings/js/ReadableStream.h:

LayoutTests:

* http/wpt/fetch/request-stream-disturbed-1-expected.txt:
* http/wpt/fetch/request-stream-disturbed-2-expected.txt:
* http/wpt/fetch/request-stream-disturbed-3-expected.txt:
* http/wpt/fetch/request-stream-empty-expected.txt: Added.
* http/wpt/fetch/request-stream-empty.html: Added.
* http/wpt/fetch/request-stream-error-expected.txt: Added.
* http/wpt/fetch/request-stream-error.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221504 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd WKUIDelegatePrivate equivalent of WKPageUIClient's toolbarsAreVisible
achristensen@apple.com [Fri, 1 Sep 2017 22:25:20 +0000 (22:25 +0000)]
Add WKUIDelegatePrivate equivalent of WKPageUIClient's toolbarsAreVisible
https://bugs.webkit.org/show_bug.cgi?id=176246
<rdar://problem/29270035>

Reviewed by Tim Horton.

Source/WebKit:

* UIProcess/API/APIUIClient.h:
(API::UIClient::toolbarsAreVisible):
* UIProcess/API/C/WKPage.cpp:
(WKPageSetPageUIClient):
* UIProcess/API/Cocoa/WKUIDelegatePrivate.h:
* UIProcess/Cocoa/UIDelegate.h:
* UIProcess/Cocoa/UIDelegate.mm:
(WebKit::UIDelegate::setDelegate):
(WebKit::UIDelegate::UIClient::toolbarsAreVisible):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::getToolbarsAreVisible):
* UIProcess/WebPageProxy.h:
* UIProcess/WebPageProxy.messages.in:

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/UIDelegate.mm:
(-[ToolbarDelegate _webView:getToolbarsAreVisibleWithCompletionHandler:]):
(-[ToolbarDelegate webView:runJavaScriptAlertPanelWithMessage:initiatedByFrame:completionHandler:]):
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221503 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMake PageClientImplCocoa a proper superclass of PageClientImpls for Mac and iOS
achristensen@apple.com [Fri, 1 Sep 2017 22:19:26 +0000 (22:19 +0000)]
Make PageClientImplCocoa a proper superclass of PageClientImpls for Mac and iOS
https://bugs.webkit.org/show_bug.cgi?id=176239

Reviewed by Tim Horton.

* UIProcess/Cocoa/PageClientImplCocoa.h: Added.
(WebKit::PageClientImplCocoa::PageClientImplCocoa):
* UIProcess/Cocoa/PageClientImplCocoa.mm:
(WebKit::PageClientImplCocoa::isPlayingAudioWillChange):
(WebKit::PageClientImplCocoa::isPlayingAudioDidChange):
(WebKit::PageClientImpl::isPlayingAudioWillChange): Deleted.
(WebKit::PageClientImpl::isPlayingAudioDidChange): Deleted.
* UIProcess/Cocoa/WebViewImpl.mm:
* UIProcess/ios/PageClientImplIOS.h:
* UIProcess/ios/PageClientImplIOS.mm:
(WebKit::PageClientImpl::PageClientImpl):
* UIProcess/mac/PageClientImpl.h: Removed.
* UIProcess/mac/PageClientImpl.mm: Removed.
* UIProcess/mac/PageClientImplMac.h: Copied from Source/WebKit/UIProcess/mac/PageClientImpl.h.
* UIProcess/mac/PageClientImplMac.mm: Copied from Source/WebKit/UIProcess/mac/PageClientImpl.mm.
(WebKit::PageClientImpl::PageClientImpl):
* UIProcess/mac/WebContextMenuProxyMac.mm:
* UIProcess/mac/WebPageProxyMac.mm:
* UIProcess/mac/WebPopupMenuProxyMac.mm:
* WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWrong getComputedStyle behavior for pseudo-elements for layout-dependent properties.
commit-queue@webkit.org [Fri, 1 Sep 2017 22:16:01 +0000 (22:16 +0000)]
Wrong getComputedStyle behavior for pseudo-elements for layout-dependent properties.
https://bugs.webkit.org/show_bug.cgi?id=175936

Patch by Emilio Cobos Álvarez <emilio@crisal.io> on 2017-09-01
Reviewed by Antti Koivisto.

LayoutTests/imported/w3c:

* web-platform-tests/cssom/getComputedStyle-pseudo-expected.txt:
* web-platform-tests/cssom/getComputedStyle-pseudo.html: Sync test with upstream.

Source/WebCore:

Before this patch we may wrongly end up using the wrong renderer to resolve
this.

Tests: imported/w3c/web-platform-tests/cssom/getComputedStyle-pseudo.html

* css/CSSComputedStyleDeclaration.cpp:
(WebCore::ComputedStyleExtractor::styledElement const):
(WebCore::ComputedStyleExtractor::styledRenderer const):
(WebCore::ComputedStyleExtractor::propertyValue):
* css/CSSComputedStyleDeclaration.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSwitch HTMLMediaElement to release logging
eric.carlson@apple.com [Fri, 1 Sep 2017 22:14:55 +0000 (22:14 +0000)]
Switch HTMLMediaElement to release logging
https://bugs.webkit.org/show_bug.cgi?id=176065

Unreviewed build fix.

* TestWebKitAPI/Tests/WebCore/Logging.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221500 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agohttp/tests/misc/slow-loading-animated-image.html is flaky on iOS simulator and Windows.
pvollan@apple.com [Fri, 1 Sep 2017 22:09:24 +0000 (22:09 +0000)]
http/tests/misc/slow-loading-animated-image.html is flaky on iOS simulator and Windows.
https://bugs.webkit.org/show_bug.cgi?id=157916

Unreviewed test gardening.

* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221499 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Win] Compile error, 'Cache' is not declared.
pvollan@apple.com [Fri, 1 Sep 2017 21:47:57 +0000 (21:47 +0000)]
[Win] Compile error, 'Cache' is not declared.
https://bugs.webkit.org/show_bug.cgi?id=176062

Reviewed by Youenn Fablet.

Rename Cache to DOMCache.

No new tests, covered by existing tests.

* DerivedSources.make:
* Modules/cache/Cache.cpp: Removed.
* Modules/cache/Cache.h: Removed.
* Modules/cache/Cache.idl: Removed.
* Modules/cache/CacheStorage.cpp:
(WebCore::doSequentialMatch):
(WebCore::startSequentialMatch):
(WebCore::copyCaches):
(WebCore::CacheStorage::match):
(WebCore::CacheStorage::retrieveCaches):
(WebCore::CacheStorage::open):
* Modules/cache/CacheStorage.h:
* Modules/cache/DOMCache.cpp: Copied from Source/WebCore/Modules/cache/Cache.cpp.
(WebCore::DOMCache::DOMCache):
(WebCore::DOMCache::~DOMCache):
(WebCore::DOMCache::match):
(WebCore::DOMCache::doMatch):
(WebCore::DOMCache::matchAll):
(WebCore::DOMCache::add):
(WebCore::DOMCache::requestFromInfo):
(WebCore::DOMCache::addAll):
(WebCore::DOMCache::put):
(WebCore::DOMCache::remove):
(WebCore::DOMCache::keys):
(WebCore::DOMCache::retrieveRecords):
(WebCore::DOMCache::queryCache):
(WebCore::DOMCache::queryCacheWithTargetStorage):
(WebCore::DOMCache::batchDeleteOperation):
(WebCore::DOMCache::batchPutOperation):
(WebCore::DOMCache::updateRecords):
(WebCore::DOMCache::stop):
(WebCore::DOMCache::activeDOMObjectName const):
(WebCore::DOMCache::canSuspendForDocumentSuspension const):
(WebCore::Cache::Cache): Deleted.
(WebCore::Cache::~Cache): Deleted.
(WebCore::Cache::match): Deleted.
(WebCore::Cache::doMatch): Deleted.
(WebCore::Cache::matchAll): Deleted.
(WebCore::Cache::add): Deleted.
(WebCore::Cache::requestFromInfo): Deleted.
(WebCore::Cache::addAll): Deleted.
(WebCore::Cache::put): Deleted.
(WebCore::Cache::remove): Deleted.
(WebCore::Cache::keys): Deleted.
(WebCore::Cache::retrieveRecords): Deleted.
(WebCore::Cache::queryCache): Deleted.
(WebCore::Cache::queryCacheWithTargetStorage): Deleted.
(WebCore::Cache::batchDeleteOperation): Deleted.
(WebCore::Cache::batchPutOperation): Deleted.
(WebCore::Cache::updateRecords): Deleted.
(WebCore::Cache::stop): Deleted.
(WebCore::Cache::activeDOMObjectName const): Deleted.
(WebCore::Cache::canSuspendForDocumentSuspension const): Deleted.
* Modules/cache/DOMCache.h: Copied from Source/WebCore/Modules/cache/Cache.h.
* Modules/cache/DOMCache.idl: Copied from Source/WebCore/Modules/cache/Cache.idl.
* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221498 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL should have more tests of type checking failures
fpizlo@apple.com [Fri, 1 Sep 2017 21:45:30 +0000 (21:45 +0000)]
WSL should have more tests of type checking failures
https://bugs.webkit.org/show_bug.cgi?id=176244

Reviewed by Myles Maxfield.

* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitProtocolDecl.set throw):
* WebGPUShadingLanguageRI/Test.js:
(TEST_typeMismatchReturn):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221497 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL should be be able to call a function declared in a protocol from a generic function
fpizlo@apple.com [Fri, 1 Sep 2017 21:35:30 +0000 (21:35 +0000)]
WSL should be be able to call a function declared in a protocol from a generic function
https://bugs.webkit.org/show_bug.cgi?id=176242

Reviewed by Myles Maxfield.

It turns out that we need to know when a function is resolved to a protocol signature, so we
need to have a type for this. This introduces ProtocolFuncDecl.

Also, this introduces parsing of protocol decls.

When instantiating a function, we need to rewrite its CallExpressions if they were previously
resolved to a PrtococolFuncDecl instead of a FuncDef or NativeFunc. In that case, we need to
rerun the resolution on the program. That resolution is guaranteed to succeed if the type
system works correctly.

* WebGPUShadingLanguageRI/All.js:
* WebGPUShadingLanguageRI/CallExpression.js:
(CallExpression.prototype.resolve):
* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitProtocolDecl.set throw):
* WebGPUShadingLanguageRI/EBufferBuilder.js:
(EBufferBuilder.prototype._createEPtr):
* WebGPUShadingLanguageRI/Func.js:
(Func):
(Func.prototype.get origin):
* WebGPUShadingLanguageRI/FuncDef.js:
(FuncDef):
(FuncDef.prototype.get origin): Deleted.
* WebGPUShadingLanguageRI/FuncInstantiator.js:
(FuncInstantiator):
(FuncInstantiator.prototype.getUnique.InstantiationSubstitution.prototype.visitCallExpression):
(FuncInstantiator.prototype.getUnique.InstantiationSubstitution):
* WebGPUShadingLanguageRI/NativeFunc.js:
(NativeFunc):
(NativeFunc.prototype.get origin): Deleted.
* WebGPUShadingLanguageRI/NativeFuncInstance.js:
(NativeFuncInstance):
* WebGPUShadingLanguageRI/Node.js:
(Node.prototype.substitute):
* WebGPUShadingLanguageRI/Parse.js:
(parseProtocolFuncDecl):
(parseProtocolDecl):
(parse):
* WebGPUShadingLanguageRI/Program.js:
(Program):
(Program.prototype.get protocols):
(Program.prototype.add):
* WebGPUShadingLanguageRI/ProtocolDecl.js:
(ProtocolDecl.prototype.add):
(ProtocolDecl.prototype.hasHeir):
(ProtocolDecl.prototype.addSignature): Deleted.
* WebGPUShadingLanguageRI/ProtocolFuncDecl.js: Added.
(ProtocolFuncDecl):
* WebGPUShadingLanguageRI/Rewriter.js:
(Rewriter.prototype.visitStructType):
(Rewriter.prototype.visitTypeVariable):
(Rewriter.prototype.visitProtocolFuncDecl):
(Rewriter.prototype.visitCallExpression):
* WebGPUShadingLanguageRI/StructType.js:
(StructType.prototype.instantiate):
* WebGPUShadingLanguageRI/Substitution.js:
(Substitution):
(Substitution.mapping): Deleted.
* WebGPUShadingLanguageRI/Test.js:
(TEST_simpleGeneric):
(TEST_simpleAssignment):
(TEST_simpleDefault):
(TEST_simpleDereference):
(TEST_dereferenceStore):
(TEST_simpleMakePtr):
(TEST_threadArrayLoad):
(TEST_deviceArrayLoad):
(TEST_threadArrayStore):
(TEST_deviceArrayStore):
(TEST_simpleProtocol):
* WebGPUShadingLanguageRI/Visitor.js:
(Visitor.prototype.visitProtocolFuncDecl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221496 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMove ServiceWorkerJob from FetchLoader to ThreadableLoader.
beidson@apple.com [Fri, 1 Sep 2017 21:33:02 +0000 (21:33 +0000)]
Move ServiceWorkerJob from FetchLoader to ThreadableLoader.
https://bugs.webkit.org/show_bug.cgi?id=176231

Reviewed by Youenn Fablet.

No new tests (No behavior change).

* WebCore.xcodeproj/project.pbxproj:

Re-indent these headers. Yuck!
* loader/ThreadableLoader.h:
(WebCore::ThreadableLoader::ref):
(WebCore::ThreadableLoader::deref):
(WebCore::ThreadableLoader::ThreadableLoader):
(WebCore::ThreadableLoader::~ThreadableLoader):
* loader/ThreadableLoaderClient.h:
(WebCore::ThreadableLoaderClient::didSendData):
(WebCore::ThreadableLoaderClient::didReceiveResponse):
(WebCore::ThreadableLoaderClient::didReceiveData):
(WebCore::ThreadableLoaderClient::didFinishLoading):
(WebCore::ThreadableLoaderClient::didFail):
(WebCore::ThreadableLoaderClient::didFinishTiming):
(WebCore::ThreadableLoaderClient::ThreadableLoaderClient):
(WebCore::ThreadableLoaderClient::~ThreadableLoaderClient):

Move from FetchLoaderClient to ThreadableLoaderClient:
* workers/service/ServiceWorkerJob.cpp:
(WebCore::ServiceWorkerJob::fetchScriptWithContext):
(WebCore::ServiceWorkerJob::didReceiveResponse):
(WebCore::ServiceWorkerJob::didReceiveData):
(WebCore::ServiceWorkerJob::didFinishLoading):
(WebCore::ServiceWorkerJob::didFail):
(WebCore::ServiceWorkerJob::didSucceed): Deleted.
* workers/service/ServiceWorkerJob.h:

* workers/service/server/SWClientConnection.cpp: Add a now-required include.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221495 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSwitch HTMLMediaElement to release logging
eric.carlson@apple.com [Fri, 1 Sep 2017 20:40:11 +0000 (20:40 +0000)]
Switch HTMLMediaElement to release logging
https://bugs.webkit.org/show_bug.cgi?id=176065

Reviewed by Jer Noble.

Source/WebCore:

* dom/Document.cpp:
(WebCore::Document::privateBrowsingStateDidChange): Disable the logger when private browsing
mode is enabled.
(WebCore::Document::logger const):
* dom/Document.h:

Convert debug-only logging to configurable release logging.
* html/HTMLMediaElement.cpp:
(PAL::LogArgument<WebCore::URL>::toString): Logger template for URL that returns the url as a
String when only when the LOG_DISABLED build flag is not defined. Returns "[url]" when it is.
* html/HTMLMediaElement.h:

Source/WTF:

* wtf/MediaTime.cpp:
(WTF::MediaTime::dump const): Use toString.
(WTF::MediaTime::toString const): New.
* wtf/MediaTime.h:
(PAL::LogArgument<WTF::MediaTime>::toString): Logger template.

Tools:

* TestWebKitAPI/Tests/WebCore/Logging.cpp:
(TestWebKitAPI::LogObserver::level const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221494 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoTry again to fix the Internal iOS Simulator build after r221485.
aestes@apple.com [Fri, 1 Sep 2017 19:47:53 +0000 (19:47 +0000)]
Try again to fix the Internal iOS Simulator build after r221485.

* pal/spi/cg/CoreGraphicsSPI.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoTry to fix the Internal iOS Simulator build after r221485.
aestes@apple.com [Fri, 1 Sep 2017 19:31:55 +0000 (19:31 +0000)]
Try to fix the Internal iOS Simulator build after r221485.

* pal/spi/cg/CoreGraphicsSPI.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221492 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agodownload-latest-github-release.py should have friendlier output for non-404 errors
commit-queue@webkit.org [Fri, 1 Sep 2017 19:13:53 +0000 (19:13 +0000)]
download-latest-github-release.py should have friendlier output for non-404 errors
https://bugs.webkit.org/show_bug.cgi?id=176201

Patch by Ross Kirsling <ross.kirsling@sony.com> on 2017-09-01
Reviewed by Brent Fulgham.

* Scripts/download-latest-github-release.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221491 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[WSL] Add tests for storing to arrays
fpizlo@apple.com [Fri, 1 Sep 2017 18:57:15 +0000 (18:57 +0000)]
[WSL] Add tests for storing to arrays
https://bugs.webkit.org/show_bug.cgi?id=176237

Reviewed by Myles Maxfield.

Storing to arrays works now.

* WebGPUShadingLanguageRI/ArrayType.js:
(ArrayType):
* WebGPUShadingLanguageRI/Test.js:
(TEST_threadArrayStore):
(TEST_deviceArrayStore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221490 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoReplace WKUIDelegatePrivate's isPlayingMediaDidChange with KVO _isPlayingAudio on...
achristensen@apple.com [Fri, 1 Sep 2017 18:54:50 +0000 (18:54 +0000)]
Replace WKUIDelegatePrivate's isPlayingMediaDidChange with KVO _isPlayingAudio on WKWebView
https://bugs.webkit.org/show_bug.cgi?id=176212

Reviewed by Tim Horton.

Source/WebKit:

Dan suggested this in bug 176203, I think it's a good idea,
and I'll need to add more KVO properties soon anyways so I'd better learn how.
Determining if audio playing changed is important for Safari, so WKPageUIClient's
isPlayingAudioDidChange was correctly named but incorrectly fired if either audio
or video playing changed and had you check if audio was playing, possibly too often.

* UIProcess/API/APIUIClient.h:
(API::UIClient::isPlayingMediaDidChange):
* UIProcess/API/C/WKPage.cpp:
(WKPageSetPageUIClient):
* UIProcess/API/Cocoa/WKUIDelegatePrivate.h:
* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _isPlayingAudio]):
* UIProcess/API/Cocoa/WKWebViewPrivate.h:
* UIProcess/API/glib/WebKitUIClient.cpp:
* UIProcess/Cocoa/UIDelegate.h:
* UIProcess/Cocoa/UIDelegate.mm:
(WebKit::UIDelegate::setDelegate):
(WebKit::UIDelegate::UIClient::isPlayingMediaDidChange): Deleted.
* UIProcess/Cocoa/WebViewImpl.h:
* UIProcess/Cocoa/WebViewImpl.mm:
(WebKit::WebViewImpl::isPlayingAudioWillChange):
(WebKit::WebViewImpl::isPlayingAudioDidChange):
* UIProcess/PageClient.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::isPlayingMediaDidChange):
* UIProcess/mac/PageClientImpl.h:
* UIProcess/mac/PageClientImpl.mm:
(WebKit::PageClientImpl::isPlayingAudioDidChange):
(WebKit::PageClientImpl::isPlayingAudioWillChange):

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/UIDelegate.mm:
(-[AudioObserver observeValueForKeyPath:ofObject:change:context:]):
(TEST):
(-[MediaDelegate _webView:isPlayingMediaDidChange:]): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221489 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agotransformCanLikelyUseFastPath() can read off the end of a string
simon.fraser@apple.com [Fri, 1 Sep 2017 18:49:00 +0000 (18:49 +0000)]
transformCanLikelyUseFastPath() can read off the end of a string
https://bugs.webkit.org/show_bug.cgi?id=176232
rdar://problem/33851237

Reviewed by Tim Horton.
Source/WebCore:

Code added in r220382 could read one byte past the end of the string when looking for the 'z'
of a rotateZ() function. The code was actually incorrect, testing for the 'z at i+6 after
already incrementing i by 6. This patch makes the code correctly detect rotateZ().

Also, rotate functions at the end of a string could be ignored because kShortestValidTransformStringLength
was too long, so set it to the length of "rotate(0)", the shortest transform function that we currently
fast-parse.

There's an implicit assumption in this code that chars is not indexed past i+kShortestValidTransformStringLength.
If the 'translate' path is taken, i is incremented by 9 (==kShortestValidTransformStringLength), but that's
OK because WTF::find() doesn't index into chars if i >= length.

Test: fast/css/transform-fast-paths.html

* css/parser/CSSParserFastPaths.cpp:
(WebCore::transformCanLikelyUseFastPath):

LayoutTests:

* fast/css/transform-fast-paths-expected.txt: Added.
* fast/css/transform-fast-paths.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221488 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL Rewriter should be an identity on things that aren't inside function bodies
fpizlo@apple.com [Fri, 1 Sep 2017 18:48:27 +0000 (18:48 +0000)]
WSL Rewriter should be an identity on things that aren't inside function bodies
https://bugs.webkit.org/show_bug.cgi?id=176208

Reviewed by Myles Maxfield.

Previously, if the Rewriter encountered a FunctionDef, StructType, NativeType, etc., then it
would either crash or try to rewrite them. That's unfortunate because we use the Rewriter to
rewrite struct and function bodies. If a function calls another function, then rewriting the
caller should not mean also rewriting the callee. Previously we "fixed" this by religiously
wrapping references to types with TypeDef and doing other such hacks. But that's subtly wrong.
It only worked because Rewriter wasn't rewriting TypeRef.type. I think that Rewriter has to
rewrite that in the long run because it may refer to another TypeRef, and it may be an
instantiation that is using types that themselves need to be rewritten.

* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitProtocolDecl.set throw):
* WebGPUShadingLanguageRI/NullType.js:
(NullType):
* WebGPUShadingLanguageRI/Rewriter.js:
(Rewriter.prototype.visitFuncDef):
(Rewriter.prototype.visitNativeFunc):
(Rewriter.prototype.visitNativeFuncInstance):
(Rewriter.prototype.visitNativeType):
(Rewriter.prototype.visitTypeDef):
(Rewriter.prototype.visitStructType):
(Rewriter.prototype.visitTypeVariable):
(Rewriter.prototype.visitConstexprTypeParameter):
(Rewriter.prototype.visitNativeTypeInstance):
(Rewriter.prototype.visitTypeRef):
* WebGPUShadingLanguageRI/Visitor.js:
(Visitor.prototype.visitNativeTypeInstance):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221487 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoDisable ObjC WebGL policy SPI on iOS
achristensen@apple.com [Fri, 1 Sep 2017 18:47:41 +0000 (18:47 +0000)]
Disable ObjC WebGL policy SPI on iOS
https://bugs.webkit.org/show_bug.cgi?id=176233

Reviewed by Tim Horton.
Source/WebKit:

In r221465 I uploaded new SPI that is only needed on Mac with a beautiful test that only works on Mac.
Since it fails on iOS and is not needed on iOS, let's just disable it on iOS.

* UIProcess/API/Cocoa/WKNavigationDelegatePrivate.h:
* UIProcess/Cocoa/NavigationState.h:
* UIProcess/Cocoa/NavigationState.mm:
(WebKit::NavigationState::setNavigationDelegate):

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/WebGLPolicy.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221486 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[CG] Upstream CoreGraphics-related WebKitSystemInterface functions
aestes@apple.com [Fri, 1 Sep 2017 18:37:28 +0000 (18:37 +0000)]
[CG] Upstream CoreGraphics-related WebKitSystemInterface functions
https://bugs.webkit.org/show_bug.cgi?id=176200

Reviewed by Tim Horton.

Source/WebCore:

* platform/graphics/GraphicsContext.h:
* platform/graphics/ImageSource.cpp:
(WebCore::ImageSource::subsamplingLevelForScaleFactor):
* platform/graphics/cg/GraphicsContextCG.cpp:
(WebCore::GraphicsContext::clipOut):
(WebCore::applyShadowOffsetWorkaroundIfNeeded):
* platform/graphics/cg/PatternCG.cpp:
(WebCore::Pattern::createPlatformPattern const):
* platform/graphics/cocoa/GraphicsContextCocoa.mm:
(WebCore::GraphicsContext::focusRingColor):
(WebCore::drawFocusRingAtTime):
(WebCore::drawFocusRing):
(WebCore::drawFocusRingToContext):
(WebCore::drawFocusRingToContextAtTime):
* platform/ios/WebCoreSystemInterfaceIOS.mm:
* platform/mac/PasteboardMac.mm:
(WebCore::flipImageSpec):
(WebCore::setDragImageImpl):
(WebCore::Pasteboard::setDragImage):
* platform/mac/PlatformEventFactoryMac.mm:
* platform/mac/ScrollbarThemeMac.mm:
(WebCore::linenBackgroundColor):
* platform/mac/ThemeMac.mm:
(WebCore::drawCellFocusRingWithFrameAtTime):
(WebCore::drawCellFocusRing):
* platform/mac/WebCoreSystemInterface.h:
* platform/mac/WebCoreSystemInterface.mm:
* platform/mac/WebWindowAnimation.mm:
(flipRect):
(mainWindowServerConnectionID):
(setScaledFrameForWindow):
(-[WebWindowScaleAnimation setCurrentProgress:]):
(-[WebWindowFadeAnimation setCurrentProgress:]):

Source/WebCore/PAL:

* PAL.xcodeproj/project.pbxproj:
* pal/spi/cg/CoreGraphicsSPI.h:
* pal/spi/mac/HIServicesSPI.h:
* pal/spi/mac/NSGraphicsSPI.h: Copied from Source/WebCore/PAL/pal/spi/mac/HIServicesSPI.h.

Source/WebKit:

* PluginProcess/mac/PluginProcessMac.mm:
(WebKit::PluginProcess::platformInitializeProcess):
* UIProcess/mac/WindowServerConnection.h:
* UIProcess/mac/WindowServerConnection.mm:
(WebKit::registerOcclusionNotificationHandler):
(WebKit::WindowServerConnection::WindowServerConnection):
(WebKit::WindowServerConnection::applicationWindowModificationsStarted): Deleted.
* WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:
(InitWebCoreSystemInterface):

Source/WebKitLegacy/mac:

* Carbon/HIWebView.mm:
(-[NSWindowGraphicsContext _web_setGraphicsPort:]):
(overrideCGContext):
(restoreCGContext):
(Draw):
* Plugins/Hosted/WebHostedNetscapePluginView.mm:
(+[WebHostedNetscapePluginView initialize]):
* Plugins/WebBaseNetscapePluginView.h:
* Plugins/WebBaseNetscapePluginView.mm:
(WebKit::sendUserChangeNotifications):
* Plugins/WebNetscapePluginView.mm:
(+[WebNetscapePluginView initialize]):
(-[WebNetscapePluginView saveAndSetNewPortStateForUpdate:]):
* WebCoreSupport/WebSystemInterface.mm:
(InitWebCoreSystemInterface):
* WebView/WebFrame.mm:
(-[WebFrame _paintBehaviorForDestinationContext:]):
* WebView/WebFullScreenController.mm:
(setClipRectForWindow):
(-[WebFullScreenController finishedEnterFullScreenAnimation:]):
(-[WebFullScreenController finishedExitFullScreenAnimation:]):
(-[WebFullScreenController _startEnterFullScreenAnimationWithDuration:]):
(-[WebFullScreenController _startExitFullScreenAnimationWithDuration:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221485 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMarked webrtc/datachannel/bufferedAmountLowThreshold.html as flaky on Mac WK1.
jlewis3@apple.com [Fri, 1 Sep 2017 18:33:09 +0000 (18:33 +0000)]
Marked webrtc/datachannel/bufferedAmountLowThreshold.html as flaky on Mac WK1.
https://bugs.webkit.org/show_bug.cgi?id=172334

Unreviewed test gardening.

* platform/mac-wk1/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221484 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoReject promises in case of internal CacheStorage engine errors.
commit-queue@webkit.org [Fri, 1 Sep 2017 18:16:16 +0000 (18:16 +0000)]
Reject promises in case of internal CacheStorage engine errors.
https://bugs.webkit.org/show_bug.cgi?id=176194

Patch by Youenn Fablet <youenn@apple.com> on 2017-09-01
Reviewed by Alex Christensen.

No new tests as this covers error cases that should not happen in regular cases.
Replacing lambdas that take no parameters to lambdas taking an optional error.
Reject promise if an error is found.

* Modules/cache/Cache.cpp:
(WebCore::Cache::doMatch):
(WebCore::Cache::matchAll):
(WebCore::Cache::keys):
(WebCore::Cache::retrieveRecords):
(WebCore::Cache::queryCache):
* Modules/cache/Cache.h:
* Modules/cache/CacheStorage.cpp:
(WebCore::CacheStorage::match):
(WebCore::CacheStorage::has):
(WebCore::CacheStorage::retrieveCaches):
(WebCore::CacheStorage::open):
(WebCore::CacheStorage::remove):
(WebCore::CacheStorage::keys):
* Modules/cache/CacheStorage.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221483 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed WPE build fix after r221439.
zandobersek@gmail.com [Fri, 1 Sep 2017 17:53:40 +0000 (17:53 +0000)]
Unreviewed WPE build fix after r221439.

* html/canvas/WebGL2RenderingContext.cpp: Add the HeapInlines.h
header include in order to properly inline a Heap::vm() call.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221482 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoImplement FileSystemEntry.getParent()
cdumez@apple.com [Fri, 1 Sep 2017 17:37:00 +0000 (17:37 +0000)]
Implement FileSystemEntry.getParent()
https://bugs.webkit.org/show_bug.cgi?id=176165
<rdar://problem/34187743>

Reviewed by Andreas Kling.

Source/WebCore:

Implement FileSystemEntry.getParent():
- https://wicg.github.io/entries-api/#dom-filesystementry-getparent

Tests: editing/pasteboard/datatransfer-items-drop-getParent-root.html
       editing/pasteboard/datatransfer-items-drop-getParent.html
       editing/pasteboard/datatransfer-items-drop-getParent2.html

* Modules/entriesapi/DOMFileSystem.cpp:
(WebCore::toFileSystemEntries):
Take a ScriptExecutionContext now that FileSystemEntry is an ActiveDOMObject.

(WebCore::isAbsoluteVirtualPath):
Add utility function to determine if a virtual path is absolute:
- https://wicg.github.io/entries-api/#absolute-path

(WebCore::DOMFileSystem::root):
Take a ScriptExecutionContext now that FileSystemEntry is an ActiveDOMObject.

(WebCore::DOMFileSystem::fileAsEntry):
Take a ScriptExecutionContext now that FileSystemEntry is an ActiveDOMObject.

(WebCore::resolveRelativePath):
Add implementation for:
- https://wicg.github.io/entries-api/#resolve-a-relative-path

(WebCore::DOMFileSystem::listDirectory):
Take a ScriptExecutionContext now that FileSystemEntry is an ActiveDOMObject.

(WebCore::DOMFileSystem::getParent):
Add implementation of getParent() as per:
- https://wicg.github.io/entries-api/#dom-filesystementry-getparent

* Modules/entriesapi/DOMFileSystem.h:
(WebCore::DOMFileSystem::createEntryForFile):
* Modules/entriesapi/DOMFileSystem.idl:
* Modules/entriesapi/FileSystemDirectoryEntry.cpp:
(WebCore::FileSystemDirectoryEntry::FileSystemDirectoryEntry):
* Modules/entriesapi/FileSystemDirectoryEntry.h:
* Modules/entriesapi/FileSystemDirectoryReader.cpp:
(WebCore::FileSystemDirectoryReader::readEntries):
Take a ScriptExecutionContext now that FileSystemEntry is an ActiveDOMObject.

* Modules/entriesapi/FileSystemEntry.cpp:
(WebCore::FileSystemEntry::FileSystemEntry):
(WebCore::FileSystemEntry::activeDOMObjectName const):
(WebCore::FileSystemEntry::canSuspendForDocumentSuspension const):
(WebCore::FileSystemEntry::getParent):
* Modules/entriesapi/FileSystemEntry.h:
* Modules/entriesapi/FileSystemEntry.idl:
- Add implementation of FileSystemEntry.getParent() which relies on
  DOMFileSystem::getParent()
- Make FileSystemEntry an ActiveDOMObject as getParent() is an asynchronous
  operation which causes delayed JS execution.

* Modules/entriesapi/FileSystemFileEntry.cpp:
(WebCore::FileSystemFileEntry::FileSystemFileEntry):
* Modules/entriesapi/FileSystemFileEntry.h:
* dom/DataTransferItem.cpp:
(WebCore::DataTransferItem::getAsEntry const):
* dom/DataTransferItem.h:
* dom/DataTransferItem.idl:
Take a ScriptExecutionContext now that FileSystemEntry is an ActiveDOMObject.

LayoutTests:

* editing/pasteboard/datatransfer-items-drop-getParent-expected.txt: Added.
* editing/pasteboard/datatransfer-items-drop-getParent-root-expected.txt: Added.
* editing/pasteboard/datatransfer-items-drop-getParent-root.html: Added.
* editing/pasteboard/datatransfer-items-drop-getParent.html: Added.
* editing/pasteboard/datatransfer-items-drop-getParent2-expected.txt: Added.
* editing/pasteboard/datatransfer-items-drop-getParent2.html: Added.
Add layout test coverage for FileSystemEntry.getParent().

* http/wpt/entries-api/interfaces-expected.txt:
Rebaseline test now that one more check is passing.

* platform/wk2/TestExpectations:
Skip new tests on WK2 since they rely on beginDragWithFiles.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221481 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoEnsure RenderStyle and SameSizeAsRenderStyle have the same size
commit-queue@webkit.org [Fri, 1 Sep 2017 17:17:33 +0000 (17:17 +0000)]
Ensure RenderStyle and SameSizeAsRenderStyle have the same size
https://bugs.webkit.org/show_bug.cgi?id=176210

Patch by Yoshiaki Jitsukawa <Yoshiaki.Jitsukawa@sony.com> on 2017-09-01
Reviewed by Daniel Bates.

SameSizeAsRenderStyle::m_nonInheritedFlags requires a 64 bit alignment,
which currently produces an extra 32 bit padding with 32 bit build.

* rendering/style/RenderStyle.cpp:
Make m_nonInheritedFlags 32 bit aligned and restore the size check.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221480 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[FTL] FTL allocation for async Function is incorrect
utatane.tea@gmail.com [Fri, 1 Sep 2017 16:05:42 +0000 (16:05 +0000)]
[FTL] FTL allocation for async Function is incorrect
https://bugs.webkit.org/show_bug.cgi?id=176214

Reviewed by Saam Barati.

In FTL, allocating async function / async generator function was incorrectly using
JSFunction logic. While it is not observable right now since sizeof(JSFunction) == sizeof(JSAsyncFunction),
but it is a bug.

* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileNewFunction):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221479 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Win] Compile error, 'Cache' is not declared.
pvollan@apple.com [Fri, 1 Sep 2017 15:08:01 +0000 (15:08 +0000)]
[Win] Compile error, 'Cache' is not declared.
https://bugs.webkit.org/show_bug.cgi?id=176062

Reviewed by Youenn Fablet.

Rename DOMCache to DOMCacheEngine.

Source/WebCore:

No new tests, covered by existing tests.

* Modules/cache/Cache.cpp:
(WebCore::Cache::addAll):
(WebCore::Cache::put):
(WebCore::queryCacheMatch):
(WebCore::Cache::batchDeleteOperation):
(WebCore::toConnectionRecord):
(WebCore::Cache::batchPutOperation):
* Modules/cache/Cache.h:
* Modules/cache/CacheStorage.cpp:
(WebCore::CacheStorage::open):
(WebCore::CacheStorage::remove):
* Modules/cache/CacheStorageConnection.cpp:
* Modules/cache/CacheStorageConnection.h:
(WebCore::CacheStorageConnection::clearMemoryRepresentation):
(WebCore::CacheStorageConnection::openCompleted):
(WebCore::CacheStorageConnection::removeCompleted):
(WebCore::CacheStorageConnection::doOpen):
(WebCore::CacheStorageConnection::doRemove):
(WebCore::CacheStorageConnection::doBatchDeleteOperation):
(WebCore::CacheStorageConnection::doBatchPutOperation):
* Modules/cache/DOMCache.cpp: Removed.
* Modules/cache/DOMCache.h: Removed.
* Modules/cache/DOMCacheEngine.cpp: Copied from Source/WebCore/Modules/cache/DOMCache.cpp.
(WebCore::DOMCacheEngine::isolatedResponseBody):
(WebCore::DOMCacheEngine::copyResponseBody):
(WebCore::DOMCache::errorToException): Deleted.
(WebCore::DOMCache::queryCacheMatch): Deleted.
(WebCore::DOMCache::isolatedResponseBody): Deleted.
(WebCore::DOMCache::copyResponseBody): Deleted.
(WebCore::DOMCache::Record::copy const): Deleted.
* Modules/cache/DOMCacheEngine.h: Copied from Source/WebCore/Modules/cache/DOMCache.h.
* Modules/cache/WorkerCacheStorageConnection.cpp:
* Modules/cache/WorkerCacheStorageConnection.h:
* WebCore.xcodeproj/project.pbxproj:
* testing/Internals.cpp:
(WebCore::Internals::clearCacheStorageMemoryRepresentation):

Source/WebKit:

* NetworkProcess/cache/CacheStorageEngine.cpp:
(WebKit::CacheStorage::Engine::queryCache):
(WebKit::CacheStorage::Engine::writeFile):
* NetworkProcess/cache/CacheStorageEngine.h:
* NetworkProcess/cache/CacheStorageEngineCache.h:
* NetworkProcess/cache/CacheStorageEngineCaches.cpp:
(WebKit::CacheStorage::Caches::initialize):
* NetworkProcess/cache/CacheStorageEngineCaches.h:
* NetworkProcess/cache/CacheStorageEngineConnection.cpp:
* NetworkProcess/cache/CacheStorageEngineConnection.h:
* NetworkProcess/cache/CacheStorageEngineConnection.messages.in:
* Shared/WebCoreArgumentCoders.cpp:
(IPC::ArgumentCoder<DOMCacheEngine::CacheInfo>::encode):
(IPC::ArgumentCoder<DOMCacheEngine::CacheInfo>::decode):
(IPC::ArgumentCoder<DOMCacheEngine::Record>::encode):
(IPC::ArgumentCoder<DOMCacheEngine::Record>::decode):
(IPC::ArgumentCoder<DOMCache::CacheInfo>::encode): Deleted.
(IPC::ArgumentCoder<DOMCache::CacheInfo>::decode): Deleted.
(IPC::ArgumentCoder<DOMCache::Record>::encode): Deleted.
(IPC::ArgumentCoder<DOMCache::Record>::decode): Deleted.
* Shared/WebCoreArgumentCoders.h:
* WebProcess/Cache/WebCacheStorageConnection.cpp:
* WebProcess/Cache/WebCacheStorageConnection.h:
* WebProcess/Cache/WebCacheStorageConnection.messages.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221478 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GTK] Mark audio-mpeg-supported.html as passing.
commit-queue@webkit.org [Fri, 1 Sep 2017 12:44:06 +0000 (12:44 +0000)]
[GTK] Mark audio-mpeg-supported.html as passing.
https://bugs.webkit.org/show_bug.cgi?id=131535

Unreviewed test gardening.

The last time it intermittently timed out was on r194436.
(It also timed out between r200951 and r201016, but that was consistent
and unrelated.)

Patch by Ms2ger <Ms2ger@igalia.com> on 2017-09-01

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221477 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMark media-ended.html as passing.
commit-queue@webkit.org [Fri, 1 Sep 2017 11:47:45 +0000 (11:47 +0000)]
Mark media-ended.html as passing.
https://bugs.webkit.org/show_bug.cgi?id=131534

Unreviewed test gardening.

The test needs to run through the entire audio file twice, so it will
take several seconds by design. GTK used to have a shorter timeout,
which was extended in r218270. The test hasn't timed out since.

Patch by Ms2ger <Ms2ger@igalia.com> on 2017-09-01

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221476 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[JSC] Fix "name" and "length" of Proxy revoke function
utatane.tea@gmail.com [Fri, 1 Sep 2017 06:40:55 +0000 (06:40 +0000)]
[JSC] Fix "name" and "length" of Proxy revoke function
https://bugs.webkit.org/show_bug.cgi?id=176155

Reviewed by Mark Lam.

JSTests:

* test262.yaml:

Source/JavaScriptCore:

ProxyRevoke's length should be configurable. And it does not have
its own name. We add NameVisibility enum to InternalFunction to
control visibility of the name.

* runtime/InternalFunction.cpp:
(JSC::InternalFunction::finishCreation):
* runtime/InternalFunction.h:
* runtime/ProxyRevoke.cpp:
(JSC::ProxyRevoke::finishCreation):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221475 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[GTK] Improve the way unit test are run and the results reported
carlosgc@webkit.org [Fri, 1 Sep 2017 06:36:32 +0000 (06:36 +0000)]
[GTK] Improve the way unit test are run and the results reported
https://bugs.webkit.org/show_bug.cgi?id=176104

Reviewed by Carlos Alberto Lopez Perez.

There are several issues with the way unit tests are run by run-gtk-tests and also with the way results are
reported:

 - The results summary only mentions the test binaries, not the actual test cases, so you always have to scroll
   up to find the actual test cases failing.
 - The number of reported failures is the number of test binaries that failed, so if a new test case fails for
   the same binary in a new revision, we won't notice it just looking at the number of failures.
 - We show detailed information about skipped test in the results summary, which is just noise.
 - In the case of glib tests, when a test case times out, we finish the test suite, instead of continuing with the
   rest of the test cases like we do for normal failures or crashes. If a new test case fails after a test case that
   timed out we will not notice it until we fix or skip the test cases timing out.
 - In the case of glib tests, the timeout is applied to the whole suite, instead of per test case, we have a hack
   to make it longer only for that. It has worked so far, but it doesn't scale, and it's an ugly hack.
 - It's not currently possible to detect flaky tests, because again, we know the binaries/suites that failed but
   not the actual test cases.

This patch fixes all these issues and makes it possible to add support for flaky tests in a follow up patch.

* BuildSlaveSupport/build.webkit.org-config/master.cfg:
(RunGtkAPITests.commandComplete): Update the RunGtkAPITests step to parse the new output.
* Scripts/run-gtk-tests:
(TestRunner._start_timeout): Helper to start the timeout if needed.
(TestRunner._start_timeout._alarm_handler): Raise timeout exception.
(TestRunner._stop_timeout): Helper to stop the timeout if needed.
(TestRunner._waitpid): Merged waitpid and return_code_from_exit_status.
(TestRunner._run_test_glib): Do not double the timeout anymore, we now start/stop the timeout for every test
case. Return a dictionary where keys are test cases and values the results only in case of failures.
(TestRunner._run_test_glib.parse_line): Update the parser to also detect test cases and the results.
(TestRunner._run_test_glib.parse_line.set_test_result): Helper to set the result of a test case.
(TestRunner._run_google_test): Return a dictionary where key is the test case and value is the result only in
case of failure.
(TestRunner._run_google_test_suite): Updated now that _run_google_test returns a dictionary.
(TestRunner.run_tests): Handle the results dictionary and show the results with information about test cases
failing and grouped by test binaries.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[Mac] Upstream AppKit-related WebKitSystemInterface functions
aestes@apple.com [Fri, 1 Sep 2017 06:00:33 +0000 (06:00 +0000)]
[Mac] Upstream AppKit-related WebKitSystemInterface functions
https://bugs.webkit.org/show_bug.cgi?id=176175

Reviewed by Brady Eidson.

Source/WebCore:

* Configurations/WebCore.xcconfig: Added -ObjC to OTHER_LDFLAGS so that Objective-C classes
in PAL are exported by WebCore even if WebCore doesn't use them.
* platform/mac/PlatformEventFactoryMac.h:
* platform/mac/PlatformEventFactoryMac.mm:
(WebCore::getWheelEventDeltas):
(WebCore::keyCharForEvent):
(WebCore::PlatformWheelEventBuilder::PlatformWheelEventBuilder):
* platform/mac/WebCoreSystemInterface.h:
* platform/mac/WebCoreSystemInterface.mm:

Source/WebCore/PAL:

* PAL.xcodeproj/project.pbxproj:
* pal/spi/mac/HIToolboxSPI.h:
* pal/spi/mac/NSResponderSPI.h: Copied from Source/WebCore/PAL/pal/spi/mac/NSWindowSPI.h.
* pal/spi/mac/NSWindowSPI.h:
* pal/system/mac/WebPanel.h: Copied from Source/WebCore/PAL/pal/spi/mac/NSWindowSPI.h.
* pal/system/mac/WebPanel.mm: Copied from Source/WebCore/PAL/pal/spi/mac/HIToolboxSPI.h.
(-[WebPanel init]):

Source/WebKit:

* PluginProcess/mac/PluginProcessMac.mm:
(WebKit::initializeCocoaOverrides):
* Shared/mac/WebEventFactory.mm:
(WebKit::WebEventFactory::createWebWheelEvent):
(WebKit::WebEventFactory::createWebKeyboardEvent):
* UIProcess/mac/WKTextInputWindowController.mm:
(-[WKTextInputPanel init]):
* WebProcess/WebCoreSupport/mac/WebSystemInterface.mm:
(InitWebCoreSystemInterface):

Source/WebKitLegacy/mac:

* Plugins/Hosted/NetscapePluginInstanceProxy.mm:
(WebKit::NetscapePluginInstanceProxy::keyEvent):
* Plugins/Hosted/WebTextInputWindowController.m:
(-[WebTextInputPanel init]):
* WebCoreSupport/WebSystemInterface.mm:
(InitWebCoreSystemInterface):
* WebView/WebDynamicScrollBarsView.mm:
(-[WebDynamicScrollBarsView scrollWheel:]):
* WebView/WebHTMLView.mm:
(-[WebHTMLView _removeWindowObservers]):
(-[WebHTMLView addWindowObservers]):
* WebView/WebView.mm:
(-[WebView addWindowObserversForWindow:]):
(-[WebView removeWindowObservers]):
(-[WebView viewWillMoveToWindow:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221473 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoThrowing an exception in the DFG/FTL should not cause a jettison
sbarati@apple.com [Fri, 1 Sep 2017 05:41:08 +0000 (05:41 +0000)]
Throwing an exception in the DFG/FTL should not cause a jettison
https://bugs.webkit.org/show_bug.cgi?id=176060
<rdar://problem/34143348>

Reviewed by Keith Miller.

Throwing an exception is not something that should be a jettison-able
OSR exit. We used to count Throw/ThrowStaticError towards our OSR exit
counts which could cause a CodeBlock to jettison and recompile. This
was dumb. Throwing an exception is not a reason to jettison and
recompile in the way that a speculation failure is. This patch
treats Throw/ThrowStaticError as true terminals in DFG IR.

* bytecode/BytecodeUseDef.h:
(JSC::computeUsesForBytecodeOffset):
* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
* dfg/DFGInPlaceAbstractState.cpp:
(JSC::DFG::InPlaceAbstractState::mergeToSuccessors):
* dfg/DFGNode.h:
(JSC::DFG::Node::isTerminal):
(JSC::DFG::Node::isPseudoTerminal):
(JSC::DFG::Node::errorType):
* dfg/DFGNodeType.h:
* dfg/DFGOperations.cpp:
* dfg/DFGOperations.h:
* dfg/DFGPredictionPropagationPhase.cpp:
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileThrow):
(JSC::DFG::SpeculativeJIT::compileThrowStaticError):
* dfg/DFGSpeculativeJIT.h:
(JSC::DFG::SpeculativeJIT::callOperation):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileNode):
(JSC::FTL::DFG::LowerDFGToB3::compileThrow):
(JSC::FTL::DFG::LowerDFGToB3::compileThrowStaticError):
* jit/JITOperations.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221472 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoFix Windows build after r221422
achristensen@apple.com [Fri, 1 Sep 2017 04:10:20 +0000 (04:10 +0000)]
Fix Windows build after r221422
https://bugs.webkit.org/show_bug.cgi?id=174924

* platform/text/win/LocaleWin.cpp:
(WebCore::LocaleWin::getLocaleInfoString):
StringVector is now the type that can be adopted by String::adopt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221471 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoGraph::methodOfGettingAValueProfileFor compares NodeOrigin instead of the semantic...
sbarati@apple.com [Fri, 1 Sep 2017 03:05:39 +0000 (03:05 +0000)]
Graph::methodOfGettingAValueProfileFor compares NodeOrigin instead of the semantic CodeOrigin
https://bugs.webkit.org/show_bug.cgi?id=176206

Reviewed by Keith Miller.

JSTests:

* stress/compare-semantic-origin-op-negate-method-of-getting-a-value-profile.js: Added.
(a):
(b):
(foo):

Source/JavaScriptCore:

Mark fixed the main issue in Graph::methodOfGettingAValueProfileFor in r208560
when he fixed it from overwriting invalid parts of the ArithProfile when the
currentNode and the operandNode are from the same bytecode. However, the
mechanism used to determine same bytecode was comparing NodeOrigin. That's
slightly wrong. We need to compare semantic origin, since two NodeOrigins can
have the same semantic origin, but differ only in exitOK. For example,
in the below IR, the DoubleRep and the Phi have the same semantic
origin, but different NodeOrigins.

43 Phi(JS|PureInt, NonBoolInt32|NonIntAsdouble, W:SideState, bc#63, ExitInvalid)
58 ExitOK(MustGen, W:SideState, bc#63)
51 DoubleRep(Check:Number:Kill:@43, Double|PureInt, BytecodeDouble, Exits, bc#63)
54 ArithNegate(DoubleRep:Kill:@51<Double>, Double|UseAsOther|MayHaveDoubleResult, AnyIntAsDouble|NonIntAsdouble, NotSet, Exits, bc#63)

* dfg/DFGGraph.cpp:
(JSC::DFG::Graph::methodOfGettingAValueProfileFor):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221470 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAnother attempt to fix the Apple Windows build following <https://trac.webkit.org...
dbates@webkit.org [Fri, 1 Sep 2017 02:59:19 +0000 (02:59 +0000)]
Another attempt to fix the Apple Windows build following <https://trac.webkit.org/changeset/221459>
(https://bugs.webkit.org/show_bug.cgi?id=176171)

Temporarily weaken the size check for RenderStyle from a strict equality to <=.

* rendering/style/RenderStyle.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221469 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL should support loading from arrays
fpizlo@apple.com [Fri, 1 Sep 2017 02:52:14 +0000 (02:52 +0000)]
WSL should support loading from arrays
https://bugs.webkit.org/show_bug.cgi?id=176207

Reviewed by Saam Barati.

Getting a test that loads from arrays to work required doing a lot of things:

- a[i] now parses to ^(operator\[](a, i)).
- added operator\[] for four array reference types.
- to make thread|threadgroup|device T[] work, you need T:primitive.
- so this adds the magical primitive protocol.

This required a little bit of rejuggling in the name resolver and type checker. The main thing
is that to prevent the rewriter from copying the types referenced by some function, you need to
make sure that when you add a reference to a type, you wrap it in a TypeRef. This doesn't
completely feel right, but I'm sure we'll figure out the balance eventually. See bug 176208.

* WebGPUShadingLanguageRI/All.js:
* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitProtocolDecl.set throw):
* WebGPUShadingLanguageRI/EArrayRef.js: Added.
(EArrayRef):
(EArrayRef.prototype.get ptr):
(EArrayRef.prototype.get length):
(EArrayRef.prototype.toString):
* WebGPUShadingLanguageRI/EPtr.js:
(EPtr.prototype.plus):
(EPtr.prototype.toString):
(EPtr):
* WebGPUShadingLanguageRI/Evaluator.js:
(Evaluator.prototype.visitUintLiteral):
* WebGPUShadingLanguageRI/FuncInstantiator.js:
(FuncInstantiator.prototype.getUnique.Instantiate.prototype.visitNativeFunc):
(FuncInstantiator.prototype.getUnique.Instantiate):
(FuncInstantiator.prototype.getUnique):
(FuncInstantiator):
* WebGPUShadingLanguageRI/Intrinsics.js:
(Intrinsics):
* WebGPUShadingLanguageRI/Lexer.js:
(Lexer.prototype.next):
(Lexer):
* WebGPUShadingLanguageRI/NameContext.js:
(NameContext.prototype.recognizeIntrinsics):
* WebGPUShadingLanguageRI/NameResolver.js:
(NameResolver.prototype.visitProtocolRef):
* WebGPUShadingLanguageRI/NativeFuncInstance.js:
(NativeFuncInstance.prototype.get isNative):
* WebGPUShadingLanguageRI/Parse.js:
(parseTerm):
(parseTypeDef):
(parseNative):
(parsePossibleSuffix):
(parse):
* WebGPUShadingLanguageRI/ProtocolDecl.js:
(ProtocolDecl):
* WebGPUShadingLanguageRI/ProtocolRef.js:
(ProtocolRef.prototype.get isPrimitive):
(ProtocolRef):
* WebGPUShadingLanguageRI/PtrType.js:
(PtrType.prototype.populateDefaultValue): Deleted.
* WebGPUShadingLanguageRI/ReferenceType.js:
(ReferenceType.prototype.populateDefaultValue):
(ReferenceType):
* WebGPUShadingLanguageRI/Rewriter.js:
(Rewriter.prototype.visitUintLiteral):
(Rewriter.prototype.visitFunc): Deleted.
(Rewriter.prototype.visitTypeVariable): Deleted.
(Rewriter.prototype.visitConstexprTypeParameter): Deleted.
* WebGPUShadingLanguageRI/StandardLibrary.js:
* WebGPUShadingLanguageRI/Test.js:
(TEST_threadArrayLoad):
(TEST_deviceArrayLoad):
* WebGPUShadingLanguageRI/TypeRef.js:
(TypeRef.prototype.get instantiatedType):
* WebGPUShadingLanguageRI/TypeVariable.js:
(TypeVariable.prototype.toString):
(TypeVariable):
* WebGPUShadingLanguageRI/UintLiteral.js: Added.
(UintLiteral):
(UintLiteral.prototype.get value):
(UintLiteral.prototype.get isConstexpr):
(UintLiteral.prototype.toString):
* WebGPUShadingLanguageRI/Visitor.js:
(Visitor.prototype.visitNativeFuncInstance):
(Visitor.prototype.visitTypeVariable):
(Visitor.prototype.visitPtrType):
(Visitor.prototype.visitArrayRefType):
(Visitor.prototype.visitUintLiteral):
* WebGPUShadingLanguageRI/WTrapError.js: Added.
(WTrapError):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAttempt to fix the Apple Windows and WinCairo build following <https://trac.webkit...
dbates@webkit.org [Fri, 1 Sep 2017 02:01:43 +0000 (02:01 +0000)]
Attempt to fix the Apple Windows and WinCairo build following <https://trac.webkit.org/changeset/221459>
(https://bugs.webkit.org/show_bug.cgi?id=176171)

Explicitly cast PseudoIds to unsigned before computing a difference between two PseudoIds.

* rendering/style/RenderStyle.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd WKUIDelegatePrivate equivalent of WKPageUIClient's isPlayingAudioDidChange
achristensen@apple.com [Fri, 1 Sep 2017 01:47:33 +0000 (01:47 +0000)]
Add WKUIDelegatePrivate equivalent of WKPageUIClient's isPlayingAudioDidChange
https://bugs.webkit.org/show_bug.cgi?id=176203
<rdar://problem/29270035>

Reviewed by Tim Horton.

Source/WebKit:

* UIProcess/API/APIUIClient.h:
(API::UIClient::isPlayingMediaDidChange):
(API::UIClient::isPlayingAudioDidChange): Deleted.
* UIProcess/API/C/WKPage.cpp:
(WKPageSetPageUIClient):
* UIProcess/API/Cocoa/WKUIDelegatePrivate.h:
* UIProcess/Cocoa/UIDelegate.h:
* UIProcess/Cocoa/UIDelegate.mm:
(WebKit::UIDelegate::setDelegate):
(WebKit::UIDelegate::UIClient::isPlayingMediaDidChange):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::isPlayingMediaDidChange):

Tools:

* TestWebKitAPI/Tests/WebKitCocoa/UIDelegate.mm:
(-[MediaDelegate _webView:isPlayingMediaDidChange:]):
(-[MediaDelegate webView:didFinishNavigation:]):
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd ObjC SPI corresponding to WKPageLoaderClient's webGLLoadPolicy and resolveWebGLLo...
achristensen@apple.com [Fri, 1 Sep 2017 01:46:26 +0000 (01:46 +0000)]
Add ObjC SPI corresponding to WKPageLoaderClient's webGLLoadPolicy and resolveWebGLLoadPolicy
https://bugs.webkit.org/show_bug.cgi?id=175779
<rdar://problem/22367975>

Reviewed by Tim Horton.

Source/WebCore:

Covered by new API tests.

* loader/FrameLoaderClient.h:

Source/WebKit:

* UIProcess/API/APINavigationClient.h:
(API::NavigationClient::webGLLoadPolicy const):
(API::NavigationClient::resolveWebGLLoadPolicy const):
* UIProcess/API/Cocoa/WKNavigationDelegatePrivate.h:
* UIProcess/Cocoa/NavigationState.h:
* UIProcess/Cocoa/NavigationState.mm:
(WebKit::NavigationState::setNavigationDelegate):
(WebKit::toWebCoreWebGLLoadPolicy):
(WebKit::NavigationState::NavigationClient::webGLLoadPolicy const):
(WebKit::NavigationState::NavigationClient::resolveWebGLLoadPolicy const):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::webGLPolicyForURL):
(WebKit::WebPageProxy::resolveWebGLPolicyForURL):
* UIProcess/WebPageProxy.h:
* UIProcess/WebPageProxy.messages.in:
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::webGLPolicyForURL const):
(WebKit::WebFrameLoaderClient::resolveWebGLPolicyForURL const):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::webGLPolicyForURL):
(WebKit::WebPage::resolveWebGLPolicyForURL):
* WebProcess/WebPage/WebPage.h:
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::webGLPolicyForURL):
(WebKit::WebPage::resolveWebGLPolicyForURL):
* WebProcess/WebPage/mac/WebPageMac.mm:
(WebKit::WebPage::webGLPolicyForURL):
(WebKit::WebPage::resolveWebGLPolicyForURL):

Source/WebKitLegacy/mac:

* WebCoreSupport/WebFrameLoaderClient.h:
* WebCoreSupport/WebFrameLoaderClient.mm:
(WebFrameLoaderClient::webGLPolicyForURL const):
(WebFrameLoaderClient::resolveWebGLPolicyForURL const):

Tools:

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKitCocoa/WebGLPolicy.mm: Added.
(-[WebGLTestDelegate webView:startURLSchemeTask:]):
(-[WebGLTestDelegate webView:stopURLSchemeTask:]):
(-[WebGLTestDelegate _webView:webGLLoadPolicyForURL:decisionHandler:]):
(-[WebGLTestDelegate _webView:resolveWebGLLoadPolicyForURL:decisionHandler:]):
(-[WebGLTestDelegate webView:runJavaScriptAlertPanelWithMessage:initiatedByFrame:completionHandler:]):
(urlsEqual):
(runTest):
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoElement focus appearance update should be either immediate or a post-layout task
akling@apple.com [Fri, 1 Sep 2017 01:17:16 +0000 (01:17 +0000)]
Element focus appearance update should be either immediate or a post-layout task
https://bugs.webkit.org/show_bug.cgi?id=176193

Reviewed by Antti Koivisto.

This change removes an old mechanism for avoiding element focus appearance updates
while we might be in the middle of attaching a renderer.

Focus appearance updates depend on a clean layout, since they want to be able to
scroll the element into the visible viewport if needed.
We now simply do the updates either immediately after layout in Element::focus(),
or as a post-layout callback when needed for HTMLInputElement::didAttachRenderers().

* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::updateFocusAppearanceSoon): Deleted.
(WebCore::Document::cancelFocusAppearanceUpdate): Deleted.
(WebCore::Document::updateFocusAppearanceTimerFired): Deleted.
* dom/Document.h:
* dom/Element.cpp:
(WebCore::Element::focus):
(WebCore::Element::blur):
(WebCore::Element::clearStyleDerivedDataBeforeDetachingRenderer):
(WebCore::Element::updateFocusAppearanceAfterAttachIfNeeded): Deleted.
(WebCore::Element::cancelFocusAppearanceUpdate): Deleted.
* dom/Element.h:
* dom/ElementRareData.h:
(WebCore::ElementRareData::ElementRareData):
(WebCore::ElementRareData::needsFocusAppearanceUpdateSoonAfterAttach const): Deleted.
(WebCore::ElementRareData::setNeedsFocusAppearanceUpdateSoonAfterAttach): Deleted.
* html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::didAttachRenderers):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221464 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years ago[CMake] Make USE_CF conditional within Windows
don.olmstead@sony.com [Fri, 1 Sep 2017 00:36:53 +0000 (00:36 +0000)]
[CMake] Make USE_CF conditional within Windows
https://bugs.webkit.org/show_bug.cgi?id=176173

Reviewed by Alex Christensen.

.:

* Source/cmake/OptionsAppleWin.cmake:
* Source/cmake/OptionsWinCairo.cmake:

Source/JavaScriptCore:

* PlatformWin.cmake:

Source/WebCore:

No new tests. No change in behavior.

* PlatformWin.cmake:
* WebCorePrefix.h:
* testing/js/WebCoreTestSupportPrefix.h:

Source/WebKitLegacy:

* PlatformWin.cmake:

Source/WTF:

* wtf/Platform.h:
* wtf/PlatformWin.cmake:

Tools:

* DumpRenderTree/config.h:
* MiniBrowser/win/CMakeLists.txt:
* MiniBrowser/win/stdafx.h:
* TestWebKitAPI/PlatformWin.cmake:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221463 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoImplement DOMMatrix2DInit for setTransform()/addPath()
commit-queue@webkit.org [Fri, 1 Sep 2017 00:21:59 +0000 (00:21 +0000)]
Implement DOMMatrix2DInit for setTransform()/addPath()
https://bugs.webkit.org/show_bug.cgi?id=176048

Patch by Sam Weinig <sam@webkit.org> on 2017-08-31
Reviewed by Simon Fraser.

LayoutTests/imported/w3c:

* web-platform-tests/css/geometry-1/DOMMatrix-001.html:
* web-platform-tests/css/geometry-1/DOMMatrix-css-string.worker.js:
(string_appeared_here.forEach.constr.test):
* web-platform-tests/css/geometry-1/DOMMatrixInit-validate-fixup-expected.txt: Added.
* web-platform-tests/css/geometry-1/DOMMatrixInit-validate-fixup.html: Added.
* web-platform-tests/css/geometry-1/historical.html:
* web-platform-tests/css/geometry-1/interfaces-expected.txt: Added.
* web-platform-tests/css/geometry-1/interfaces.html: Added.
* web-platform-tests/css/geometry-1/interfaces.worker-expected.txt: Added.
* web-platform-tests/css/geometry-1/interfaces.worker.html: Added.
* web-platform-tests/css/geometry-1/interfaces.worker.js: Added.
* web-platform-tests/css/geometry-1/structured-serialization.html:
* web-platform-tests/css/geometry-1/w3c-import.log:

    Pull latest web-platform-tests/css/geometry-1/ and rebase the results.

Source/WebCore:

* DerivedSources.make:
* WebCore.xcodeproj/project.pbxproj:
* CMakeLists.txt:

    Add new files.

* bindings/js/CallTracerTypes.h:
* inspector/InspectorCanvas.cpp:

    Swap out DOMMatrixInit for DOMMatrix2DInit in canvas tracing infrastructure.

* css/DOMMatrix2DInit.h: Added.
* css/DOMMatrix2DInit.idl: Added.
* css/DOMMatrixInit.h:
* css/DOMMatrixInit.idl:
Split DOMMatrixInit in two, with the 2D bits going into DOMMatrix2DInit.

* css/DOMMatrixReadOnly.cpp:
* css/DOMMatrixReadOnly.h:

    Add a validateAndFixup overload for DOMMatrix2DInit.

* html/canvas/CanvasRenderingContext2D.cpp:
* html/canvas/CanvasRenderingContext2D.h:

    Swap out DOMMatrixInit for DOMMatrix2DInit in setTransform. This
    should have no observable difference, since we ignored the 3D prior.

* html/canvas/DOMPath.cpp:
* html/canvas/DOMPath.h:
* html/canvas/DOMPath.idl:

    Update DOMPath (actually called Path2D) to take an optional DOMMatrix2DInit
    rather than an SVGMatrix. Passing an SVGMatrix is still valid, as it conforms
    as a DOMMatrix2DInit dictionary (having the necessary properties), so the
    risk of compatibility issues is minimal.

LayoutTests:

* fast/canvas/canvas-path-addPath-expected.txt:
* fast/canvas/canvas-path-addPath.html:
Update test / results now that the transform is for the path is an optional dictionary.
Mostly this means we throw a little less.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221462 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoAdd (entirely incorrect) fetching of ServiceWorker scripts.
beidson@apple.com [Fri, 1 Sep 2017 00:12:52 +0000 (00:12 +0000)]
Add (entirely incorrect) fetching of ServiceWorker scripts.
https://bugs.webkit.org/show_bug.cgi?id=176179

Reviewed by Andy Estes.

Source/WebCore:

No new tests (Covered by changes to existing tests).

When the Storage process is running the "Update" algorithm and a ServiceWorker script file needs to be fetched, this patch:
  - Messages back to the WebContent process that started the register/update job
  - Executes a FetchLoad in that script context for the script
  - Sends the results back to the Storage process

We don't do anything with the results yet.

Soon.

* WebCore.xcodeproj/project.pbxproj:

* workers/service/ServiceWorkerContainer.cpp:
(WebCore::ServiceWorkerContainer::startScriptFetchForJob):
(WebCore::ServiceWorkerContainer::jobFinishedLoadingScript):
(WebCore::ServiceWorkerContainer::jobFailedLoadingScript):
* workers/service/ServiceWorkerContainer.h:

* workers/service/ServiceWorkerFetchResult.h: Copied from Source/WebCore/workers/service/server/SWClientConnection.h.
(WebCore::ServiceWorkerFetchResult::encode const):
(WebCore::ServiceWorkerFetchResult::decode):

* workers/service/ServiceWorkerJob.cpp:
(WebCore::ServiceWorkerJob::startScriptFetch):
(WebCore::ServiceWorkerJob::fetchScriptWithContext):
(WebCore::ServiceWorkerJob::didReceiveResponse):
(WebCore::ServiceWorkerJob::didReceiveData):
(WebCore::ServiceWorkerJob::didSucceed):
(WebCore::ServiceWorkerJob::didFail):
* workers/service/ServiceWorkerJob.h:

* workers/service/ServiceWorkerJobClient.h:

* workers/service/server/SWClientConnection.cpp:
(WebCore::SWClientConnection::finishedFetchingScript):
(WebCore::SWClientConnection::failedFetchingScript):
(WebCore::SWClientConnection::startScriptFetchForServer):
* workers/service/server/SWClientConnection.h:

* workers/service/server/SWServer.cpp:
(WebCore::SWServer::Connection::finishFetchingScriptInServer):
(WebCore::SWServer::startScriptFetch):
(WebCore::SWServer::scriptFetchFinished):
* workers/service/server/SWServer.h:

* workers/service/server/SWServerRegistration.cpp:
(WebCore::SWServerRegistration::scriptFetchFinished):
(WebCore::SWServerRegistration::runUpdateJob):
(WebCore::SWServerRegistration::startScriptFetchFromMainThread):
(WebCore::SWServerRegistration::startScriptFetchForCurrentJob):
* workers/service/server/SWServerRegistration.h:

Source/WebKit:

* StorageProcess/ServiceWorker/WebSWServerConnection.cpp:
(WebKit::WebSWServerConnection::startScriptFetchInClient):
* StorageProcess/ServiceWorker/WebSWServerConnection.h:
* StorageProcess/ServiceWorker/WebSWServerConnection.messages.in:

* WebProcess/Storage/WebSWClientConnection.cpp:
(WebKit::WebSWClientConnection::finishFetchingScriptInServer):
* WebProcess/Storage/WebSWClientConnection.h:
* WebProcess/Storage/WebSWClientConnection.messages.in:

LayoutTests:

* http/tests/workers/service/basic-register-expected.txt:
* http/tests/workers/service/resources/basic-register.js:
* http/tests/workers/service/resources/registration-task-queue-scheduling-1.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL should support makeptr (\)
fpizlo@apple.com [Fri, 1 Sep 2017 00:09:14 +0000 (00:09 +0000)]
WSL should support makeptr (\)
https://bugs.webkit.org/show_bug.cgi?id=176197

Reviewed by Saam Barati.

This required introducing some sanity about how pointers are carried around in the evaluator.
It turns out that we need to be careful about when a pointer to an rvalue is dereferenced. It
should be dereferenced basically immediately. If it cannot be, then we need to snapshot the
value. To do that, we need to know the types of things in a few more places. That's what this
change accomplishes.

* WebGPUShadingLanguageRI/All.js:
* WebGPUShadingLanguageRI/CallFunction.js:
(callFunction):
* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitProtocolDecl.set throw):
* WebGPUShadingLanguageRI/DereferenceExpression.js:
(DereferenceExpression.prototype.get isLValue):
* WebGPUShadingLanguageRI/Evaluator.js:
(Evaluator.prototype._snapshot):
(Evaluator.prototype.runBody):
(Evaluator.prototype.visitFunctionLikeBlock):
(Evaluator.prototype.visitDereferenceExpression):
(Evaluator.prototype.visitMakePtrExpression):
(Evaluator.prototype.visitCommaExpression):
(Evaluator.prototype.visitCallExpression):
(Evaluator):
(Evaluator.prototype._dereference): Deleted.
* WebGPUShadingLanguageRI/FuncInstantiator.js:
(FuncInstantiator.prototype.getUnique.Instantiate.prototype.visitFuncDef):
(FuncInstantiator.prototype.getUnique.Instantiate.prototype.visitNativeFunc):
(FuncInstantiator.prototype.getUnique.Instantiate):
(FuncInstantiator.prototype.getUnique):
(FuncInstantiator):
* WebGPUShadingLanguageRI/FunctionLikeBlock.js:
(FunctionLikeBlock):
(FunctionLikeBlock.prototype.get returnType):
(FunctionLikeBlock.prototype.toString):
* WebGPUShadingLanguageRI/Inliner.js:
(Inliner.prototype.visitCallExpression):
(Inliner):
* WebGPUShadingLanguageRI/MakePtrExpression.js: Added.
(MakePtrExpression):
(MakePtrExpression.prototype.get lValue):
(MakePtrExpression.prototype.toString):
* WebGPUShadingLanguageRI/NativeFuncInstance.js: Added.
(NativeFuncInstance):
(NativeFuncInstance.prototype.get func):
(NativeFuncInstance.prototype.toString):
* WebGPUShadingLanguageRI/Rewriter.js:
(Rewriter.prototype.visitMakePtrExpression):
(Rewriter.prototype.visitCallExpression):
* WebGPUShadingLanguageRI/Test.js:
(TEST_dereferenceStore):
(TEST_simpleMakePtr):
* WebGPUShadingLanguageRI/Value.js:
(Value.prototype.get isLValue):
(Value):
* WebGPUShadingLanguageRI/VariableRef.js:
(VariableRef.prototype.get isLValue):
(VariableRef.prototype.get addressSpace):
* WebGPUShadingLanguageRI/Visitor.js:
(Visitor.prototype.visitMakePtrExpression):
(Visitor.prototype.visitCallExpression):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221460 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoMake RenderStyle's non-inherited flags more human friendly
dbates@webkit.org [Thu, 31 Aug 2017 23:55:12 +0000 (23:55 +0000)]
Make RenderStyle's non-inherited flags more human friendly
https://bugs.webkit.org/show_bug.cgi?id=176171

Reviewed by Antti Koivisto.

Revert r166465 as we never followed through and made the CSS JIT update
RenderStyle's non-inherited flags directly.

* rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::RenderStyle):
(WebCore::RenderStyle::hashForTextAutosizing const):
(WebCore::RenderStyle::equalForTextAutosizing const):
(WebCore::RenderStyle::changeRequiresLayout const):
* rendering/style/RenderStyle.h:
(WebCore::RenderStyle::styleType const):
(WebCore::RenderStyle::setStyleType):
(WebCore::RenderStyle::setHasViewportUnits):
(WebCore::RenderStyle::hasViewportUnits const):
(WebCore::RenderStyle::affectedByHover const):
(WebCore::RenderStyle::affectedByActive const):
(WebCore::RenderStyle::affectedByDrag const):
(WebCore::RenderStyle::setAffectedByHover):
(WebCore::RenderStyle::setAffectedByActive):
(WebCore::RenderStyle::setAffectedByDrag):
(WebCore::RenderStyle::isFloating const):
(WebCore::RenderStyle::display const):
(WebCore::RenderStyle::position const):
(WebCore::RenderStyle::floating const):
(WebCore::RenderStyle::overflowX const):
(WebCore::RenderStyle::overflowY const):
(WebCore::RenderStyle::verticalAlign const):
(WebCore::RenderStyle::unicodeBidi const):
(WebCore::RenderStyle::clear const):
(WebCore::RenderStyle::tableLayout const):
(WebCore::RenderStyle::hasExplicitlySetDirection const):
(WebCore::RenderStyle::isLink const):
(WebCore::RenderStyle::setDisplay):
(WebCore::RenderStyle::setOriginalDisplay):
(WebCore::RenderStyle::setPosition):
(WebCore::RenderStyle::setFloating):
(WebCore::RenderStyle::setOverflowX):
(WebCore::RenderStyle::setOverflowY):
(WebCore::RenderStyle::setVerticalAlign):
(WebCore::RenderStyle::setUnicodeBidi):
(WebCore::RenderStyle::setClear):
(WebCore::RenderStyle::setTableLayout):
(WebCore::RenderStyle::setHasExplicitlySetDirection):
(WebCore::RenderStyle::setIsLink):
(WebCore::RenderStyle::hasExplicitlySetWritingMode const):
(WebCore::RenderStyle::setHasExplicitlySetWritingMode):
(WebCore::RenderStyle::hasExplicitlySetTextAlign const):
(WebCore::RenderStyle::setHasExplicitlySetTextAlign):
(WebCore::RenderStyle::unique const):
(WebCore::RenderStyle::setUnique):
(WebCore::RenderStyle::emptyState const):
(WebCore::RenderStyle::setEmptyState):
(WebCore::RenderStyle::firstChildState const):
(WebCore::RenderStyle::setFirstChildState):
(WebCore::RenderStyle::lastChildState const):
(WebCore::RenderStyle::setLastChildState):
(WebCore::RenderStyle::setHasExplicitlyInheritedProperties):
(WebCore::RenderStyle::hasExplicitlyInheritedProperties const):
(WebCore::RenderStyle::inheritUnicodeBidiFrom):
(WebCore::RenderStyle::NonInheritedFlags::operator!= const):
(WebCore::RenderStyle::NonInheritedFlags::hasAnyPublicPseudoStyles const):
(WebCore::RenderStyle::originalDisplay const):
(WebCore::RenderStyle::NonInheritedFlags::operator== const):
(WebCore::RenderStyle::NonInheritedFlags::copyNonInheritedFrom):
(WebCore::RenderStyle::NonInheritedFlags::hasPseudoStyle const):
(WebCore::RenderStyle::NonInheritedFlags::setHasPseudoStyle):
(WebCore::RenderStyle::NonInheritedFlags::overflowX const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setOverflowX): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::overflowY const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setOverflowY): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::clear const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setClear): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::effectiveDisplay const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setEffectiveDisplay): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::position const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setPosition): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::originalDisplay const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setOriginalDisplay): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::unicodeBidi const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setUnicodeBidi): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::hasViewportUnits const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setHasViewportUnits): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::verticalAlign const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setVerticalAlign): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::hasExplicitlyInheritedProperties const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setHasExplicitlyInheritedProperties): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::isFloating const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::floating const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setFloating): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::tableLayout const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setTableLayout): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::styleType const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setStyleType): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::isUnique const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setIsUnique): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::emptyState const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setEmptyState): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::firstChildState const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setFirstChildState): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::lastChildState const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setLastChildState): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::affectedByHover const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setAffectedByHover): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::affectedByActive const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setAffectedByActive): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::affectedByDrag const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setAffectedByDrag): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::isLink const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setIsLink): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::hasExplicitlySetDirection const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setHasExplicitlySetDirection): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::hasExplicitlySetWritingMode const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setHasExplicitlySetWritingMode): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::hasExplicitlySetTextAlign const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setHasExplicitlySetTextAlign): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::flagsMemoryOffset): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::flagIsaffectedByActive): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::flagIsaffectedByHover): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::flagPseudoStyle): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setFirstChildStateFlags): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setLastChildStateFlags): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::flagIsUnique): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::flagFirstChildState): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::flagLastChildState): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::NonInheritedFlags): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::setHasPseudoStyles): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::updateBoolean): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::getBoolean const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::updateValue): Deleted.
(WebCore::RenderStyle::InheritedFlags::operator== const): Deleted.
(WebCore::RenderStyle::NonInheritedFlags::getValue const): Deleted.
(WebCore::adjustForAbsoluteZoom): Deleted.
(WebCore::adjustFloatForAbsoluteZoom): Deleted.
(WebCore::adjustLayoutUnitForAbsoluteZoom): Deleted.
(WebCore::collapsedBorderStyle): Deleted.
(WebCore::RenderStyle::hasBackground const): Deleted.
(WebCore::RenderStyle::autoWrap): Deleted.
(WebCore::RenderStyle::preserveNewline): Deleted.
(WebCore::RenderStyle::collapseWhiteSpace): Deleted.
(WebCore::RenderStyle::isCollapsibleWhiteSpace const): Deleted.
(WebCore::RenderStyle::breakOnlyAfterWhiteSpace const): Deleted.
(WebCore::RenderStyle::breakWords const): Deleted.
(WebCore::RenderStyle::hasInlineColumnAxis const): Deleted.
(WebCore::RenderStyle::imageOrientation const): Deleted.
(WebCore::RenderStyle::setLogicalWidth): Deleted.
(WebCore::RenderStyle::setLogicalHeight): Deleted.
(WebCore::RenderStyle::setBorderRadius): Deleted.
(WebCore::RenderStyle::setZoom): Deleted.
(WebCore::RenderStyle::setEffectiveZoom): Deleted.
(WebCore::RenderStyle::setTextOrientation): Deleted.
(WebCore::RenderStyle::adjustBackgroundLayers): Deleted.
(WebCore::RenderStyle::adjustMaskLayers): Deleted.
(WebCore::RenderStyle::clearAnimations): Deleted.
(WebCore::RenderStyle::clearTransitions): Deleted.
(WebCore::RenderStyle::setShapeOutside): Deleted.
(WebCore::RenderStyle::setShapeImageThreshold): Deleted.
(WebCore::RenderStyle::setClipPath): Deleted.
(WebCore::RenderStyle::willChangeCreatesStackingContext const): Deleted.
(WebCore::RenderStyle::isDisplayRegionType const): Deleted.
(WebCore::RenderStyle::setWritingMode): Deleted.
(WebCore::RenderStyle::getShadowInlineDirectionExtent const): Deleted.
(WebCore::RenderStyle::getShadowBlockDirectionExtent const): Deleted.
(WebCore::RenderStyle::isDisplayReplacedType): Deleted.
(WebCore::RenderStyle::isDisplayInlineType): Deleted.
(WebCore::RenderStyle::isDisplayFlexibleBox): Deleted.
(WebCore::RenderStyle::isDisplayGridBox): Deleted.
(WebCore::RenderStyle::isDisplayFlexibleOrGridBox): Deleted.
(WebCore::RenderStyle::hasAnyPublicPseudoStyles const): Deleted.
(WebCore::RenderStyle::hasPseudoStyle const): Deleted.
(WebCore::RenderStyle::setHasPseudoStyle): Deleted.
(WebCore::RenderStyle::setHasPseudoStyles): Deleted.
(WebCore::RenderStyle::setBoxReflect): Deleted.
(WebCore::pseudoElementRendererIsNeeded): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSkip two slow JSC tests after r221422.
ryanhaddad@apple.com [Thu, 31 Aug 2017 23:44:51 +0000 (23:44 +0000)]
Skip two slow JSC tests after r221422.

Unreviewed test gardening.

* stress/regexp-prototype-match-on-too-long-rope.js:
* stress/regexp-prototype-test-on-too-long-rope.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221458 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoFix error in protocol revving
megan_gardner@apple.com [Thu, 31 Aug 2017 23:31:55 +0000 (23:31 +0000)]
Fix error in protocol revving
https://bugs.webkit.org/show_bug.cgi?id=176191

Protocol definition was mastered out of later iOS submissions, causing them to not build.

Reviewed by Tim Horton.

* UIProcess/ios/WKContentViewInteraction.mm:
(-[WKContentView changeSelectionWithGestureAt:withGesture:withState:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove newBlockInsideInlineModel and anonymous inline block
antti@apple.com [Thu, 31 Aug 2017 23:27:02 +0000 (23:27 +0000)]
Remove newBlockInsideInlineModel and anonymous inline block
https://bugs.webkit.org/show_bug.cgi?id=176181

Reviewed by Zalan Bujtas.

Source/WebCore:

Anonymous inline block exists to support newBlockInsideInlineModel which was planned as the replacement for continuations.
It has never been enabled and it is time to remove it.

* page/Settings.in:
* rendering/InlineElementBox.cpp:
(WebCore::InlineElementBox::paint):
(WebCore::InlineElementBox::nodeAtPoint):
* rendering/InlineFlowBox.cpp:
(WebCore::InlineFlowBox::addToLine):
(WebCore::InlineFlowBox::placeBoxesInBlockDirection):
(WebCore::InlineFlowBox::nodeAtPoint):
(WebCore::InlineFlowBox::paint):
(WebCore::InlineFlowBox::anonymousInlineBlock const): Deleted.
* rendering/InlineFlowBox.h:
(WebCore::InlineFlowBox::InlineFlowBox):
(WebCore::InlineFlowBox::hasAnonymousInlineBlock const): Deleted.
(WebCore::InlineFlowBox::setHasAnonymousInlineBlock): Deleted.
* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::lineHeight const):
(WebCore::RenderBlock::baselinePosition const):
(WebCore::RenderBlock::renderName const):
* rendering/RenderBlockFlow.cpp:
(WebCore::RenderBlockFlow::rebuildFloatingObjectSetFromIntrudingFloats):
(WebCore::RenderBlockFlow::childrenPreventSelfCollapsing const):
(WebCore::RenderBlockFlow::adjustLinePositionForPagination):
(WebCore::RenderBlockFlow::computeInlinePreferredLogicalWidths const):
* rendering/RenderBlockFlow.h:
* rendering/RenderBlockLineLayout.cpp:
(WebCore::updateLogicalInlinePositions):
(WebCore::RenderBlockFlow::computeInlineDirectionPositionsForLine):
(WebCore::RenderBlockFlow::layoutRunsAndFloats):
(WebCore::RenderBlockFlow::layoutRunsAndFloatsInRange):
(WebCore::RenderBlockFlow::layoutLineBoxes):
(WebCore::RenderBlockFlow::determineStartPosition):
(WebCore::RenderBlockFlow::matchedEndLine):
(WebCore::RenderBlockFlow::marginCollapseLinesFromStart): Deleted.
* rendering/RenderBox.cpp:
(WebCore::RenderBox::sizesLogicalWidthToFitContent const):
(WebCore::RenderBox::createsNewFormattingContext const):
(WebCore::RenderBox::avoidsFloats const):
* rendering/RenderInline.cpp:
(WebCore::RenderInline::addChildIgnoringContinuation):
* rendering/RenderObject.cpp:
(WebCore::RenderObject::isAnonymousInlineBlock const): Deleted.
* rendering/RenderObject.h:
(WebCore::RenderObject::isAnonymousBlock const):
* rendering/RootInlineBox.cpp:
(WebCore::RootInlineBox::alignBoxesInBlockDirection):
(WebCore::RootInlineBox::ascentAndDescentForBox const):
* rendering/line/BreakingContext.h:
(WebCore::BreakingContext::BreakingContext):
(WebCore::BreakingContext::handleReplaced):
* rendering/line/LineBreaker.cpp:
(WebCore::LineBreaker::nextLineBreak):
* rendering/line/LineBreaker.h:
* rendering/line/LineLayoutState.h:
(WebCore::LineLayoutState::marginInfo):
(WebCore::LineLayoutState::prevFloatBottomFromAnonymousInlineBlock): Deleted.
(WebCore::LineLayoutState::maxFloatBottomFromAnonymousInlineBlock): Deleted.

LayoutTests:

* TestExpectations:
* fast/block/inside-inlines/crash-on-first-line-change.html:

    Don't set newBlockInsideInlineModel.

* fast/block/inside-inlines/new-model: Removed.
* fast/block/inside-inlines/new-model/basic-float-intrusion-expected.html: Removed.
* fast/block/inside-inlines/new-model/basic-float-intrusion.html: Removed.
* fast/block/inside-inlines/new-model/block-width-check-expected.html: Removed.
* fast/block/inside-inlines/new-model/block-width-check.html: Removed.
* fast/block/inside-inlines/new-model/breaking-behavior-expected.html: Removed.
* fast/block/inside-inlines/new-model/breaking-behavior.html: Removed.
* fast/block/inside-inlines/new-model/empty-block-expected.html: Removed.
* fast/block/inside-inlines/new-model/empty-block.html: Removed.
* fast/block/inside-inlines/new-model/margin-collapse: Removed.
* fast/block/inside-inlines/new-model/margin-collapse/001-expected.html: Removed.
* fast/block/inside-inlines/new-model/margin-collapse/001.html: Removed.
* fast/block/inside-inlines/new-model/margin-collapse/002-expected.html: Removed.
* fast/block/inside-inlines/new-model/margin-collapse/002.html: Removed.
* fast/block/inside-inlines/new-model/margin-collapse/003-expected.html: Removed.
* fast/block/inside-inlines/new-model/margin-collapse/003.html: Removed.
* fast/block/inside-inlines/new-model/margin-collapse/004-expected.html: Removed.
* fast/block/inside-inlines/new-model/margin-collapse/004.html: Removed.
* fast/block/inside-inlines/new-model/margins-pushing-below-float-expected.html: Removed.
* fast/block/inside-inlines/new-model/margins-pushing-below-float.html: Removed.
* fast/block/inside-inlines/new-model/paint-order-expected.html: Removed.
* fast/block/inside-inlines/new-model/paint-order.html: Removed.
* fast/block/inside-inlines/new-model/self-collapsing-test-expected.html: Removed.
* fast/block/inside-inlines/new-model/self-collapsing-test.html: Removed.

    These are reftests comparing newBlockInsideInlineModel and the current model and don't have value anymore.

* fast/block/inside-inlines/opacity-on-inline.html:

    Don't set newBlockInsideInlineModel.
    Marked this as failing since it only worked with newBlockInsideInlineModel enabled.
    The test passes in Firefox and fails in Chrome.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUnreviewed, rolling out r221445.
ryanhaddad@apple.com [Thu, 31 Aug 2017 23:25:45 +0000 (23:25 +0000)]
Unreviewed, rolling out r221445.

This change broke Sierra Release builds.

Reverted changeset:

"Switch HTMLMediaElement to release logging"
https://bugs.webkit.org/show_bug.cgi?id=176065
http://trac.webkit.org/changeset/221445

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221455 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoUpdate CacheStorage caches only if it is updated
commit-queue@webkit.org [Thu, 31 Aug 2017 23:20:03 +0000 (23:20 +0000)]
Update CacheStorage caches only if it is updated
https://bugs.webkit.org/show_bug.cgi?id=176135

Patch by Youenn Fablet <youenn@apple.com> on 2017-08-31
Reviewed by Alex Christensen.

Source/WebCore:

Covered by existing tests.

When retrieving caches from the engine, CacheStorage gets an update counter.
When this counter is different from the previous received one, CacheStorage will update its cache list.
Otherwise it will skip this update step.

* Modules/cache/CacheStorage.cpp:
(WebCore::CacheStorage::retrieveCaches):
* Modules/cache/CacheStorage.h:
* Modules/cache/CacheStorageConnection.cpp:
(WebCore::CacheStorageConnection::retrieveCaches):
* Modules/cache/CacheStorageConnection.h:
(WebCore::CacheStorageConnection::doRetrieveCaches):
* Modules/cache/DOMCache.cpp:
(WebCore::DOMCache::CacheInfos::isolatedCopy):
* Modules/cache/DOMCache.h:
(WebCore::DOMCache::CacheInfos::encode const):
(WebCore::DOMCache::CacheInfos::decode):
* Modules/cache/WorkerCacheStorageConnection.cpp:
(WebCore::WorkerCacheStorageConnection::doRetrieveCaches):
* Modules/cache/WorkerCacheStorageConnection.h:

Source/WebKit:

For each change to CacheStorageEngineCaches list of cache, CacheStorageEngineCaches increments a counter.
When sending the list of caches, CacheStorageEngineCaches will compare its counter with the one provided.
If they sare the same, CacheStorageEngineCaches will return nothing.
Otherwise it returns the whole list of caches.

* NetworkProcess/cache/CacheStorageEngine.cpp:
(WebKit::CacheStorage::Engine::retrieveCaches):
* NetworkProcess/cache/CacheStorageEngine.h:
* NetworkProcess/cache/CacheStorageEngineCaches.cpp:
(WebKit::CacheStorage::Caches::initialize):
(WebKit::CacheStorage::Caches::open):
(WebKit::CacheStorage::Caches::remove):
(WebKit::CacheStorage::Caches::clearMemoryRepresentation):
(WebKit::CacheStorage::Caches::cacheInfos const):
* NetworkProcess/cache/CacheStorageEngineCaches.h:
(WebKit::CacheStorage::Caches::updateCounter const):
* NetworkProcess/cache/CacheStorageEngineConnection.cpp:
(WebKit::CacheStorageEngineConnection::caches):
* NetworkProcess/cache/CacheStorageEngineConnection.h:
* NetworkProcess/cache/CacheStorageEngineConnection.messages.in:
* WebProcess/Cache/WebCacheStorageConnection.cpp:
(WebKit::WebCacheStorageConnection::doRetrieveCaches):
* WebProcess/Cache/WebCacheStorageConnection.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agouseSeparatedWXHeap should never be true when not on iOS
sbarati@apple.com [Thu, 31 Aug 2017 23:19:31 +0000 (23:19 +0000)]
useSeparatedWXHeap should never be true when not on iOS
https://bugs.webkit.org/show_bug.cgi?id=176190

Reviewed by JF Bastien.

If you set useSeparatedWXHeap to true on X86_64, and launch the jsc shell,
the process insta-crashes. Let's silently ignore that option and set it
to false when not on iOS.

* runtime/Options.cpp:
(JSC::recomputeDependentOptions):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221453 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL should support dereference (^)
fpizlo@apple.com [Thu, 31 Aug 2017 23:00:22 +0000 (23:00 +0000)]
WSL should support dereference (^)
https://bugs.webkit.org/show_bug.cgi?id=176192

Reviewed by Myles Maxfield.

This implements DereferenceExpression.

It also renames Evaluator.prototype.visitFunctionBody, because that name confusingly implies
that there is a class named FunctionBody. It's now called runBody.

I made the check to see if a type is a pointer type be the isPtr method, so that we can have
other pointer types that are implemented via something other than PtrType. We might want it if
for example we wanted a pointer type that is generic over address space.

* WebGPUShadingLanguageRI/All.js:
* WebGPUShadingLanguageRI/CallFunction.js:
(callFunction):
* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitProtocolDecl.set throw):
* WebGPUShadingLanguageRI/DereferenceExpression.js: Added.
(DereferenceExpression):
(DereferenceExpression.prototype.get ptr):
(DereferenceExpression.prototype.toString):
* WebGPUShadingLanguageRI/Evaluator.js:
(Evaluator.prototype.visitFunctionLikeBlock):
(Evaluator.prototype._dereference):
(Evaluator.prototype.visitDereferenceExpression):
(Evaluator.prototype.visitFunctionBody): Deleted.
* WebGPUShadingLanguageRI/PtrType.js:
(PtrType.prototype.get isPtr):
(PtrType.prototype.populateDefaultValue):
* WebGPUShadingLanguageRI/Rewriter.js:
(Rewriter.prototype.visitAssignment):
(Rewriter.prototype.visitDereferenceExpression):
* WebGPUShadingLanguageRI/Test.js:
(TEST_simpleDereference):
* WebGPUShadingLanguageRI/Type.js:
(Type.prototype.get isPtr):
* WebGPUShadingLanguageRI/Visitor.js:
(Visitor.prototype.visitDereferenceExpression):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoThere should only be one callFunction API in WSL
fpizlo@apple.com [Thu, 31 Aug 2017 22:22:59 +0000 (22:22 +0000)]
There should only be one callFunction API in WSL
https://bugs.webkit.org/show_bug.cgi?id=176189

Reviewed by Saam Barati.

This removes the need for callFunctionByRef, which was weird. It's now the case the TypedValue
is always a tuple of type and ePtr, and TypedValue has a super simple API for boxing and
unboxing single-cell values like ints.

* WebGPUShadingLanguageRI/CallFunction.js:
(callFunction):
(callFunctionByRef): Deleted.
* WebGPUShadingLanguageRI/Test.js:
(makeInt):
* WebGPUShadingLanguageRI/TypedValue.js:
(TypedValue):
(TypedValue.prototype.get type):
(TypedValue.prototype.get ePtr):
(TypedValue.box):
(TypedValue.prototype.get value):
(TypedValue.prototype.toString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221449 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL EPtr does not need to carry around the type
fpizlo@apple.com [Thu, 31 Aug 2017 22:15:18 +0000 (22:15 +0000)]
WSL EPtr does not need to carry around the type
https://bugs.webkit.org/show_bug.cgi?id=176188

Reviewed by Saam Barati.

To validate that the interpreter is modeling a statically typed execution environment, it's
really best if values don't know their types at all. A pointer is just a pointer (i.e a buffer
and offset in our world), and it's the job of the AST node that uses it to figure out what its
type should be.

I think this makes the code simpler overall.

* WebGPUShadingLanguageRI/All.js:
* WebGPUShadingLanguageRI/CallFunction.js:
(callFunctionByRef):
(callFunction):
* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitProtocolDecl.set throw):
* WebGPUShadingLanguageRI/EBufferBuilder.js:
(EBufferBuilder.prototype._createEPtr):
* WebGPUShadingLanguageRI/EPtr.js:
(EPtr):
(EPtr.box):
(EPtr.prototype.copyFrom):
(EPtr.prototype.toString):
* WebGPUShadingLanguageRI/EValue.js: Removed.
* WebGPUShadingLanguageRI/Evaluator.js:
(Evaluator.prototype.visitFunctionLikeBlock):
(Evaluator.prototype.visitVariableDecl):
(Evaluator.prototype.visitAssignment):
(Evaluator.prototype.visitIntLiteral):
* WebGPUShadingLanguageRI/Intrinsics.js:
(Intrinsics):
* WebGPUShadingLanguageRI/Rewriter.js:
(Rewriter.prototype.visitAssignment):
* WebGPUShadingLanguageRI/StructType.js:
(StructType.prototype.instantiate):
* WebGPUShadingLanguageRI/Test.js:
(checkInt):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221448 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoIntroduce CacheStorageEngineCache to handle cache records
commit-queue@webkit.org [Thu, 31 Aug 2017 21:55:17 +0000 (21:55 +0000)]
Introduce CacheStorageEngineCache to handle cache records
https://bugs.webkit.org/show_bug.cgi?id=176137

Patch by Youenn Fablet <youenn@apple.com> on 2017-08-31
Reviewed by Alex Christensen.

CacheStorageEngineCache is the equivalent of WebCore Cache.
It contains a list of records that it needs to manage (add, update, remove).
Moving the logic of this management from CacheStorageEngine to this new class so as to ease future development.

* CMakeLists.txt:
* NetworkProcess/cache/CacheStorageEngine.cpp:
(WebKit::CacheStorage::Engine::open):
(WebKit::CacheStorage::Engine::remove):
(WebKit::CacheStorage::Engine::retrieveRecords):
(WebKit::CacheStorage::Engine::putRecords):
(WebKit::CacheStorage::Engine::deleteMatchingRecords):
(WebKit::CacheStorage::Engine::cache):
(WebKit::CacheStorage::Engine::writeCacheRecords): Deleted.
(WebKit::CacheStorage::Engine::removeCacheRecords): Deleted.
(WebKit::CacheStorage::Engine::queryCache): Deleted.
* NetworkProcess/cache/CacheStorageEngine.h:
* NetworkProcess/cache/CacheStorageEngineCache.cpp: Added.
* NetworkProcess/cache/CacheStorageEngineCache.h:
(WebKit::CacheStorage::Cache::identifier const):
(WebKit::CacheStorage::Cache::name const):
(WebKit::CacheStorage::Cache::info const):
* NetworkProcess/cache/CacheStorageEngineCaches.cpp:
(WebKit::CacheStorage::Caches::find):
(WebKit::CacheStorage::Caches::open):
(WebKit::CacheStorage::Caches::remove):
(WebKit::CacheStorage::encodeCacheNames):
(WebKit::CacheStorage::Caches::readCachesFromDisk):
(WebKit::CacheStorage::Caches::cacheInfos const):
* WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221447 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL doesn't need to wrap primitives like ints and floats in specialized classes like...
fpizlo@apple.com [Thu, 31 Aug 2017 21:50:57 +0000 (21:50 +0000)]
WSL doesn't need to wrap primitives like ints and floats in specialized classes like EInt and EFloat
https://bugs.webkit.org/show_bug.cgi?id=176184

Reviewed by Saam Barati.

The only use for EInt and EFloat was that users of the interpreter want to be able to reason
about values that know their type. But for the actual interpreter, it makes most sense for
EBuffer to hold values directly - for example, ints in the case of values of type int. Then,
all of the logic of doing math on those things can be implemented in Intrinsics.js.

So, this removes EInt and EFloat but introduces a new TypedValue class that is used only on the
boundary of the interpreter. You can either use the interpreter by speaking its internal
language (EPtr to an EBuffer that has the values) or by using the simple API (which uses
TypedValue as a trade-off between power and convenience).

* WebGPUShadingLanguageRI/All.js:
* WebGPUShadingLanguageRI/CallFunction.js:
(callFunction):
* WebGPUShadingLanguageRI/EFloat.js: Removed.
* WebGPUShadingLanguageRI/EInt.js: Removed.
* WebGPUShadingLanguageRI/EPtr.js:
(EPtr.box):
* WebGPUShadingLanguageRI/Evaluator.js:
(Evaluator.prototype.visitIntLiteral):
* WebGPUShadingLanguageRI/Intrinsics.js:
(Intrinsics):
* WebGPUShadingLanguageRI/Test.js:
(makeInt):
(checkInt):
(TEST_add1):
(TEST_simpleGeneric):
(TEST_simpleAssignment):
(TEST_simpleDefault):
* WebGPUShadingLanguageRI/TypedValue.js: Added.
(TypedValue):
(TypedValue.prototype.toString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221446 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoSwitch HTMLMediaElement to release logging
eric.carlson@apple.com [Thu, 31 Aug 2017 21:41:01 +0000 (21:41 +0000)]
Switch HTMLMediaElement to release logging
https://bugs.webkit.org/show_bug.cgi?id=176065

Reviewed by Jer Noble.

Source/WebCore:

* dom/Document.cpp:
(WebCore::Document::privateBrowsingStateDidChange): Disable the logger when private browsing
mode is enabled.
(WebCore::Document::logger const):
* dom/Document.h:

Convert debug-only logging to configurable release logging.
* html/HTMLMediaElement.cpp:
(PAL::LogArgument<WebCore::URL>::toString): Logger template for URL that returns the url as a
String when only when the LOG_DISABLED build flag is not defined. Returns "[url]" when it is.
* html/HTMLMediaElement.h:

Source/WTF:

* wtf/MediaTime.cpp:
(WTF::MediaTime::dump const): Use toString.
(WTF::MediaTime::toString const): New.
* wtf/MediaTime.h:
(PAL::LogArgument<WTF::MediaTime>::toString): Logger template.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221445 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWKNavigationDelegatePrivate client redirect SPI needs to be able to detect redirects...
commit-queue@webkit.org [Thu, 31 Aug 2017 21:26:24 +0000 (21:26 +0000)]
WKNavigationDelegatePrivate client redirect SPI needs to be able to detect redirects scheduled before the document finishes loading
https://bugs.webkit.org/show_bug.cgi?id=176128
rdar://problem/34068476

Patch by David Quesada <david_quesada@apple.com> on 2017-08-31
Reviewed by Brady Eidson.

Source/WebCore:

Removed FrameLoaderClient::dispatchDidPerformClientRedirect() since no client cares about this event anymore.
Also removed FrameLoader::performClientRedirect() since it wouldn't do anything but call changeLocation().

No new tests - no change in functionality.

* loader/FrameLoader.cpp:
* loader/FrameLoader.h:
* loader/FrameLoaderClient.h:
* loader/NavigationScheduler.cpp:

Source/WebKit:

_webView:didPerformClientRedirect: isn't useful for delegates that want to know about client redirects
started before the document is finished loading. This is because the method would be called after the
navigation scheduler's timer fires and the navigation for the redirect has begun. Since this happens in
a later iteration of the run loop, the document has already finished loading. Address this by replacing
the method with two that give the navigation delegate more information about when client redirects are
scheduled and canceled.

* UIProcess/API/APINavigationClient.h:
(API::NavigationClient::willPerformClientRedirect):
(API::NavigationClient::didCancelClientRedirect):
* UIProcess/API/Cocoa/WKNavigationDelegatePrivate.h:
* UIProcess/Cocoa/NavigationState.h:
* UIProcess/Cocoa/NavigationState.mm:
(WebKit::NavigationState::setNavigationDelegate):
(WebKit::NavigationState::NavigationClient::willPerformClientRedirect):
(WebKit::NavigationState::NavigationClient::didCancelClientRedirect):
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::willPerformClientRedirectForFrame):
(WebKit::WebPageProxy::didCancelClientRedirectForFrame):
* UIProcess/WebPageProxy.h:
* UIProcess/WebPageProxy.messages.in:
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDidCancelClientRedirect):
(WebKit::WebFrameLoaderClient::dispatchWillPerformClientRedirect):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:

Tools:

Removed API test for the deleted WKNavigationDelegatePrivate method,
and added two new tests for the two new methods.

* TestWebKitAPI/Tests/WebKitCocoa/Navigation.mm:
(-[ClientRedirectNavigationDelegate _webView:willPerformClientRedirectToURL:delay:]):
(-[ClientRedirectNavigationDelegate _webViewDidCancelClientRedirect:]):
(-[ClientRedirectNavigationDelegate webView:didFinishNavigation:]):
(TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221444 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoRemove IsBlockSelection flag
megan_gardner@apple.com [Thu, 31 Aug 2017 21:13:38 +0000 (21:13 +0000)]
Remove IsBlockSelection flag
https://bugs.webkit.org/show_bug.cgi?id=176141

Reviewed by Dean Jackson.

Block selection is disabled. Removed code associated with this flag to allow removal of the
flag in UIKit.

No tests for a removed feature.

* Platform/spi/ios/UIKitSPI.h:
* Shared/ios/GestureTypes.h:
* UIProcess/ios/WKContentViewInteraction.mm:
(toUIWKSelectionFlags):
* WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::rangeForWebSelectionAtPosition):
(WebKit::WebPage::selectWithGesture):
(WebKit::WebPage::contractedRangeFromHandle):
(WebKit::WebPage::computeExpandAndShrinkThresholdsForHandle):
(WebKit::WebPage::updateBlockSelectionWithTouch):
(WebKit::WebPage::updateSelectionWithTouches):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221443 268f45cc-cd09-0410-ab3c-d52691b4dbfc

2 years agoWSL should handle variable assignments and variable declarations
fpizlo@apple.com [Thu, 31 Aug 2017 20:57:45 +0000 (20:57 +0000)]
WSL should handle variable assignments and variable declarations
https://bugs.webkit.org/show_bug.cgi?id=176180

Reviewed by Saam Barati.

Just fills in VariableDecl and Assignment logic so that we can test variables.

* WebGPUShadingLanguageRI/Checker.js:
(Checker.prototype.visitProtocolDecl.set throw):
* WebGPUShadingLanguageRI/EBufferBuilder.js:
(EBufferBuilder.prototype._createEPtrForNode):
(EBufferBuilder.prototype.visitFuncParameter):
(EBufferBuilder.prototype.visitVariableDecl):
(EBufferBuilder):
* WebGPUShadingLanguageRI/Evaluator.js:
(Evaluator.prototype.visitVariableDecl):
(Evaluator.prototype.visitAssignment):
* WebGPUShadingLanguageRI/NameResolver.js:
(NameResolver.prototype.visitVariableDecl):
* WebGPUShadingLanguageRI/Rewriter.js:
(Rewriter.prototype.visitFuncParameter):
(Rewriter.prototype.visitVariableDecl):
* WebGPUShadingLanguageRI/Test.js:
(makeInt):
(TEST_add1):
(TEST_simpleGeneric):
(TEST_simpleAssignment):
(TEST_simpleDefault):
* WebGPUShadingLanguageRI/Visitor.js:
(Visitor.prototype.visitVariableDecl):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@221442 268f45cc-cd09-0410-ab3c-d52691b4dbfc