⚠ Archived content — this site is no longer maintained.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

Timeline



Oct 18, 2015:

11:49 PM Changeset in webkit [191278] by Csaba Osztrogonác
  • 2 edits in trunk/Source/WebKit2

Fix the ENABLE(SECCOMP_FILTERS) build after r187030
​https://bugs.webkit.org/show_bug.cgi?id=150172

Reviewed by Alexey Proskuryakov.

  • Shared/linux/SeccompFilters/Syscall.cpp:

(WebKit::reportUnexpectedSyscall):

7:26 PM Changeset in webkit [191277] by commit-queue@webkit.org
  • 2 edits in trunk/Tools

[Win] Change the DLL name in MiniBrowserLib.vcxproj to match webkit.org/b/150124.
​https://bugs.webkit.org/show_bug.cgi?id=150307

Patch by Sungmann Cho <​sungmann.cho@navercorp.com> on 2015-10-18
Reviewed by Brent Fulgham.

The DLL build rules have been changed to generate "{Name}Lib.DLL",
but these changes have not yet been applied to the Visual Studio builds.

  • MiniBrowser/MiniBrowser.vcxproj/MiniBrowserLib.vcxproj:
5:51 PM Changeset in webkit [191276] by Yusuke Suzuki
  • 5 edits
    2 moves
    2 adds in trunk

Promise constructor should throw when not called with "new"
​https://bugs.webkit.org/show_bug.cgi?id=149380

Reviewed by Darin Adler.

Source/JavaScriptCore:

Implement handling new.target in Promise constructor. And
prohibiting Promise constructor call without "new".

  • runtime/JSPromiseConstructor.cpp:

(JSC::constructPromise):
(JSC::callPromise):
(JSC::JSPromiseConstructor::getCallData):

  • tests/es6.yaml:
  • tests/stress/promise-cannot-be-called.js: Added.

(shouldBe):
(shouldThrow):
(Deferred):
(super):

LayoutTests:

Fix js/dom/Promise-types.html. Before this change, it calls the Promise constructor without new and
expects it succeeds. And we move it from js/dom to js since we can execute this without DOM support.

  • js/Promise-types-expected.txt: Renamed from LayoutTests/js/dom/Promise-types-expected.txt.
  • js/Promise-types.html: Added.
  • js/script-tests/Promise-types.js: Renamed from LayoutTests/js/dom/Promise-types.html.

(aPromise.new.Promise):
(debug.string_appeared_here.shouldThrow.Promise):

  • resources/standalone-pre.js:

(shouldBeDefined):

5:36 PM Changeset in webkit [191275] by Yusuke Suzuki
  • 10 edits in trunk/Source/JavaScriptCore

[ES6] Handle asynchronous tests in tests/es6
​https://bugs.webkit.org/show_bug.cgi?id=150293

Reviewed by Darin Adler.

Since JSC can handle microtasks, some of ES6 Promise tests can be executed under the JSC shell.
Some of them still fail because it uses setTimeout that invokes macrotasks with explicit delay.

  • tests/es6.yaml:
  • tests/es6/Promise_Promise.all.js:

(test.asyncTestPassed):
(test):

  • tests/es6/Promise_Promise.all_generic_iterables.js:

(test.asyncTestPassed):
(test):

  • tests/es6/Promise_Promise.race.js:

(test.asyncTestPassed):
(test):

  • tests/es6/Promise_Promise.race_generic_iterables.js:

(test.asyncTestPassed):
(test):

  • tests/es6/Promise_basic_functionality.js:

(test.asyncTestPassed):
(test):

  • tests/es6/Promise_is_subclassable_Promise.all.js:

(test.asyncTestPassed):
(test):

  • tests/es6/Promise_is_subclassable_Promise.race.js:

(test.asyncTestPassed):
(test):

  • tests/es6/Promise_is_subclassable_basic_functionality.js:

(test.asyncTestPassed):
(test):

5:29 PM Changeset in webkit [191274] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Console Search should update when console is cleared
​https://bugs.webkit.org/show_bug.cgi?id=150141

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2015-10-18
Reviewed by Brian Burg.

  • UserInterface/Views/LogContentView.js:

(WebInspector.LogContentView.prototype._logCleared):
Reperform the search if there is a search query.

(WebInspector.LogContentView.prototype._filterMessageElements):
Reperform the current search.

4:51 PM Changeset in webkit [191273] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

Marking fast/canvas/webgl/oes-texture-float-linear.html as flaky
​https://bugs.webkit.org/show_bug.cgi?id=149930

Patch by Ryan Haddad <Ryan Haddad> on 2015-10-18
Reviewed by Darin Adler.

  • platform/mac/TestExpectations:
4:47 PM Changeset in webkit [191272] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebKit2

[GTK] Remove WebKitContextMenu related missing annotation warnings
​https://bugs.webkit.org/show_bug.cgi?id=150299

Patch by Joonghun Park <​jh718.park@samsung.com> on 2015-10-18
Reviewed by Darin Adler.

The warnings occur currently because when executing g-ir-scanner,
it doesn't consider WebKitContextMenu related .cpp files,
but only .h files.

  • PlatformGTK.cmake: Add WebKitContextMenu related cpp files.
4:47 PM Changeset in webkit [191271] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

Marking webarchive/loading/object.html as flaky due to crashes
​https://bugs.webkit.org/show_bug.cgi?id=150241
<rdar://problem/21690723>

Patch by Ryan Haddad <Ryan Haddad> on 2015-10-18
Reviewed by Darin Adler.

  • platform/mac-wk2/TestExpectations:
4:43 PM Changeset in webkit [191270] by Chris Dumez
  • 4 edits in trunk

Script element with an empty for or event attributes should not execute
​https://bugs.webkit.org/show_bug.cgi?id=148855
<rdar://problem/22588156>

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Rebaseline existing W3C HTML test now that more checks are passing.

  • web-platform-tests/html/semantics/scripting-1/the-script-element/script-for-event-expected.txt:

Source/WebCore:

A script element with an empty for or event attributes should not execute
as per:
​https://html.spec.whatwg.org/multipage/scripting.html#prepare-a-script (step 12)

Our implementation had a bug where it would not correctly differentiate
an empty attribute from a missing one. This patch fixes this.

No new tests, already covered by existing test.

  • dom/ScriptElement.cpp:

(WebCore::ScriptElement::isScriptForEventSupported):

4:42 PM Changeset in webkit [191269] by commit-queue@webkit.org
  • 2 edits in trunk/LayoutTests

Marking media/event-queue-crash.html as flaky
​https://bugs.webkit.org/show_bug.cgi?id=114177

Patch by Ryan Haddad <Ryan Haddad> on 2015-10-18
Reviewed by Darin Adler.

  • platform/mac/TestExpectations:
4:41 PM Changeset in webkit [191268] by Chris Dumez
  • 7 edits in trunk

WebKit must support all JavaScript MIME types in HTML5 spec
​https://bugs.webkit.org/show_bug.cgi?id=148854
<rdar://problem/22588195>

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Rebaseline W3C HTML test now that more checks are passing.

  • web-platform-tests/html/semantics/scripting-1/the-script-element/script-languages-02-expected.txt:

Source/WebCore:

WebKit did not execute certain EcmaScript MIME types although the HTML
specification says all user agents should support those:
​https://html.spec.whatwg.org/multipage/scripting.html#javascript-mime-type

In particular, the following MIME types are now recognized as valid and
executed: "application/x-ecmascript", "text/javascript1.0",
"text/javascript1.4", "text/javascript1.5", "text/x-javascript",
"text/x-ecmascript".

The new behavior is consistent with Firefox.

No new tests, already covered by existing test.

  • platform/MIMETypeRegistry.cpp:

(WebCore::initializeSupportedJavaScriptMIMETypes):

LayoutTests:

Rebaseline several existing layout tests now that we execute
more JS MIME types.

  • fast/html/script-allowed-types-languages-expected.txt:
  • fast/tokenizer/004-expected.txt:
4:39 PM Changeset in webkit [191267] by commit-queue@webkit.org
  • 8 edits in trunk/Source

[Win] Fix the Windows builds.
​https://bugs.webkit.org/show_bug.cgi?id=150300

Patch by Sungmann Cho <​sungmann.cho@navercorp.com> on 2015-10-18
Reviewed by Darin Adler.

Source/JavaScriptCore:

Add missing files to JavaScriptCore.vcxproj.

Source/WebCore:

Add missing files to WebCore.vcxproj.
Add missing #includes to CSSAllInOne.cpp and HTMLElementsAllInOne.cpp.

No new tests, no behavior change.

  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.vcxproj/WebCore.vcxproj.filters:
  • css/CSSAllInOne.cpp:
  • html/HTMLElementsAllInOne.cpp:
4:36 PM Changeset in webkit [191266] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

Fix the builds with ENABLE_CONTENT_EXTENSIONS=OFF.
​https://bugs.webkit.org/show_bug.cgi?id=150302

Patch by Sungmann Cho <​sungmann.cho@navercorp.com> on 2015-10-18
Reviewed by Darin Adler.

Add missing #if ENABLE(CONTENT_EXTENSIONS).

No new tests, no behavior change.

  • page/UserContentController.h:
4:29 PM Changeset in webkit [191265] by commit-queue@webkit.org
  • 5 edits in trunk/Source/WebCore

[EFL] Fix debug build break since r191198
​https://bugs.webkit.org/show_bug.cgi?id=150277

Patch by Joonghun Park <​jh718.park@samsung.com> on 2015-10-18
Reviewed by Darin Adler.

No new tests, no new behaviours.

Use the correct %"PRIu64" for uint64_t,
instead of %llu.

  • Modules/indexeddb/client/IDBDatabaseImpl.cpp:

(WebCore::IDBClient::IDBDatabase::fireVersionChangeEvent):

  • Modules/indexeddb/client/IDBOpenDBRequestImpl.cpp:

(WebCore::IDBClient::IDBOpenDBRequest::onUpgradeNeeded):

  • Modules/indexeddb/legacy/IDBDatabaseBackend.cpp:

(WebCore::IDBDatabaseBackend::processPendingOpenCalls):
(WebCore::IDBDatabaseBackend::openConnectionInternal):

  • Modules/indexeddb/server/UniqueIDBDatabase.cpp:

(WebCore::IDBServer::UniqueIDBDatabase::notifyConnectionsOfVersionChange):

4:27 PM Changeset in webkit [191264] by commit-queue@webkit.org
  • 4 edits in trunk/Source

Fix the builds with ENABLE_INDEX_DATABASE=OFF.
​https://bugs.webkit.org/show_bug.cgi?id=150301

Patch by Sungmann Cho <​sungmann.cho@navercorp.com> on 2015-10-18
Reviewed by Darin Adler.

Add missing #if ENABLE(INDEXED_DATABASE).

Source/WebCore:

No new tests, no behavior change.

  • page/Page.cpp:

Source/WebKit:

  • Storage/WebDatabaseProvider.h:
4:24 PM Changeset in webkit [191263] by mmaxfield@apple.com
  • 3 edits in trunk/Source/WebCore

[Cocoa] [Win] Remove unused code from GraphicsContext
​https://bugs.webkit.org/show_bug.cgi?id=150304

Reviewed by Darin Adler.

No new tests because there is no behavior change.

  • platform/graphics/GraphicsContext.h:
  • platform/graphics/cg/GraphicsContextCG.cpp:

(WebCore::GraphicsContext::setAllowsFontSmoothing): Deleted.

3:15 PM Changeset in webkit [191262] by Antti Koivisto
  • 11 edits
    2 adds in trunk

Computed style should work correctly with slotted elements that have display:none
​https://bugs.webkit.org/show_bug.cgi?id=150237

Source/WebCore:

Reviewed by Andreas Kling..

If an element has display:none we don't normally retain or even compute its style (as it is not rendered).
If getComputedStyle is invoked for such element we resolve the style (along with any ancestors) and cache
it separately to rare data. This path needs to work with slotted elements in shadow trees.

This patch also make computedStyle() iterative rather than recursive.

Test: fast/shadow-dom/computed-style-display-none.html

  • dom/Document.cpp:

(WebCore::Document::styleForElementIgnoringPendingStylesheets):

Pass in the parent style instead of invoking computedStyle() recursively.

  • dom/Document.h:
  • dom/Element.cpp:

(WebCore::beforeOrAfterPseudoElement):
(WebCore::Element::existingComputedStyle):
(WebCore::Element::resolveComputedStyle):

Iterative resolve function that uses composed tree iterator.

(WebCore::Element::computedStyle):

Factor into helpers.

  • dom/Element.h:
  • dom/Node.cpp:

(WebCore::Node::computedStyle):

Use the composed tree iterator.

  • html/HTMLSelectElement.cpp:

(WebCore::HTMLSelectElement::selectOption):

Call updateValidity() before calling renderer->updateFromElement(). Calling updateFromElement()
may end up in Element::computedStyle() which can asserts if validity is not up to date.

LayoutTests:

Reviewed by Andreas Kling.

  • editing/style/apply-style-atomic-expected.txt:

Rebase.

  • fast/css/getComputedStyle/getComputedStyle-with-pseudo-element-expected.txt:
  • fast/css/getComputedStyle/getComputedStyle-with-pseudo-element.html:

We now also compute style of display:none pseudo elements correctly.
This is a progression and matches other browsers.

  • fast/shadow-dom/computed-style-display-none-expected.txt: Added.
  • fast/shadow-dom/computed-style-display-none.html: Added.
10:19 AM Changeset in webkit [191261] by mmaxfield@apple.com
  • 2 edits in trunk/Source/WebKit/mac

Build fix after r191260

Unreviewed.

  • WebView/WebView.mm:

(+[WebView initialize]):

9:36 AM Changeset in webkit [191260] by mmaxfield@apple.com
  • 10 edits in trunk/Source

Stop honoring the user default "WebKitKerningAndLigaturesEnabledByDefault"
​https://bugs.webkit.org/show_bug.cgi?id=150287

Reviewed by Simon Fraser.

Source/WebCore:

This user default is currently on by default. Therefore, by setting the user default,
users can only disable kerning / ligatures (rather than enable it).

There are a few reasons why we should stop honoring it:

  1. In the brave new world of font-feature-settings and font-variant-ligatures, there

are many different kinds of ligatures which may be enabled at will. The simplistic
statement of "turn on ligatures" no longer has any meaning.

  1. If a user wants to disable kerning / ligatures, he/she can do it with a user

stylesheet.

  1. The default isn't able to be tested with DumpRenderTree or WebKitTestRunner.
  1. I have never heard of anyone actually using this user default.
  • platform/graphics/FontCascade.cpp:

(WebCore::FontCascade::setDefaultKerning): Deleted.
(WebCore::FontCascade::setDefaultLigatures): Deleted.

  • platform/graphics/FontCascade.h:

(WebCore::FontCascade::advancedTextRenderingMode):

Source/WebKit/mac:

  • WebView/WebView.mm:

(+[WebView initialize]): Deleted.

Source/WebKit2:

  • Shared/WebProcessCreationParameters.cpp:

(WebKit::WebProcessCreationParameters::WebProcessCreationParameters): Deleted.
(WebKit::WebProcessCreationParameters::encode): Deleted.
(WebKit::WebProcessCreationParameters::decode): Deleted.

  • Shared/WebProcessCreationParameters.h:
  • UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::registerUserDefaultsIfNeeded): Deleted.
(WebKit::WebProcessPool::platformInitializeWebProcess): Deleted.

  • WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::platformInitializeWebProcess): Deleted.

6:30 AM Changeset in webkit [191259] by Gyuyoung Kim
  • 2 edits in trunk/LayoutTests

Unreviewed EFL gardening. Mark css variables tests to pass
with incorrect result. Because the pass is not verified yet.

  • platform/efl/TestExpectations:
12:58 AM Changeset in webkit [191258] by Yusuke Suzuki
  • 3 edits in trunk/Source/WebKit2

[GTK] Implement WebContextMenuProxyGtk::showContextMenu after r191194
​https://bugs.webkit.org/show_bug.cgi?id=150276

Reviewed by Carlos Garcia Campos.

After r191194, showContextMenu signature becomes changed, and the implementation
is moved from the generic file to the Mac port. It poses the compile error in GTK port.
This patch implements WebContextMenuProxyGtk::showContextMenu. The logic is the same
to the old one except for dropping SERVICE_CONTROLS support. This is because
SERVICE_CONTROLS is not related to GTK.

  • UIProcess/gtk/WebContextMenuProxyGtk.cpp:

(WebKit::WebContextMenuProxyGtk::showContextMenu):

  • UIProcess/gtk/WebContextMenuProxyGtk.h:
12:11 AM Changeset in webkit [191257] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Fix a copy/paste mistake in the previous commit.

  • platform/mac-wk2/TestExpectations:
12:08 AM Changeset in webkit [191256] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Add one more flaky Blink test to TestExpectations.

  • platform/mac-wk2/TestExpectations:
12:02 AM Changeset in webkit [191255] by commit-queue@webkit.org
  • 10 edits in trunk/Source

Unreviewed, rolling out r191250 and r191253.
​https://bugs.webkit.org/show_bug.cgi?id=150296

Broke all the tests on Windows (Requested by ap on #webkit).

Reverted changesets:

"Stop honoring the user default
"WebKitKerningAndLigaturesEnabledByDefault""
​https://bugs.webkit.org/show_bug.cgi?id=150287
​http://trac.webkit.org/changeset/191250

"Build fix after r191250"
​http://trac.webkit.org/changeset/191253

Oct 17, 2015:

5:25 PM Changeset in webkit [191254] by mark.lam@apple.com
  • 2 edits in trunk/Source/WTF

Add CPU(X86) to the supported MASM_PROBE CPUs. This was accidentally left out in r191197.

Not reviewed.

  • wtf/Platform.h:
5:00 PM Changeset in webkit [191253] by mmaxfield@apple.com
  • 2 edits in trunk/Source/WebKit/mac

Build fix after r191250

Unreviewed.

  • WebView/WebView.mm:

(+[WebView initialize]):

4:13 PM Changeset in webkit [191252] by hyatt@apple.com
  • 15 edits
    16 adds in trunk

Implement the CSS4 'revert' keyword.
​https://bugs.webkit.org/show_bug.cgi?id=149702

Reviewed by Simon Fraser.

Source/WebCore:

Added new tests in fast/css and fast/css/variables.

  • CMakeLists.txt:
  • WebCore.xcodeproj/project.pbxproj:

Add CSSRevertValue to the project and makefiles.

  • css/CSSParser.cpp:

(WebCore::parseKeywordValue):
Make sure to handle "revert" in the keyword parsing path (along with inherit/initial/unset).

(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::parseCustomPropertyDeclaration):
At the parser level, "revert" is just like inherit/initial/unset and gets its own special
singleton value, CSSRevertValue.

  • css/CSSRevertValue.cpp: Added.

(WebCore::CSSRevertValue::customCSSText):

  • css/CSSRevertValue.h: Added.

(WebCore::CSSRevertValue::create):
(WebCore::CSSRevertValue::equals):
(WebCore::CSSRevertValue::CSSRevertValue):
This value is identical to the inherit/initial/unset values, i.e., its own special value
that can be used to indicate a revert when doing style resolution.

  • css/CSSValue.cpp:

(WebCore::CSSValue::cssValueType):
(WebCore::CSSValue::equals):
(WebCore::CSSValue::cssText):
(WebCore::CSSValue::destroy):

  • css/CSSValue.h:

(WebCore::CSSValue::isInheritedValue):
(WebCore::CSSValue::isInitialValue):
(WebCore::CSSValue::isUnsetValue):
(WebCore::CSSValue::isRevertValue):
Add the RevertClass to CSSValue and make sure it is handled in all the appropriate methods.

  • css/CSSValueKeywords.in:

Add the "revert" keyword to the list of allowed CSS keywords.

  • css/CSSValuePool.cpp:

(WebCore::CSSValuePool::CSSValuePool):

  • css/CSSValuePool.h:

(WebCore::CSSValuePool::createRevertValue):
Add support for a CSSRevertValue singleton, just like inherit/unset/initial.

  • css/FontLoader.cpp:

(WebCore::FontLoader::resolveFontStyle):
Add "unset" and "revert" as special keywords to be ignored. This code seems to be turned off,
but patching it anyway.

  • css/SelectorChecker.h:

Add a MatchDefault value of 0 to the LinkMatchMask. This enables it to be used as an index
to the correct value in Property (in the style resolution code).

  • css/StyleResolver.cpp:

(WebCore::StyleResolver::State::initForStyleResolve):
Delete any lingering old CascadedProperty rollbacks for UA/user rules.

(WebCore::StyleResolver::styleForKeyframe):
(WebCore::StyleResolver::styleForPage):
(WebCore::StyleResolver::applyMatchedProperties):
Pass along the MatchResult as an additional parameter, since we need it to lazily compute
the cascade rollbacks if the "revert" keyword is encountered.

(WebCore::StyleResolver::cascadedPropertiesForRollback):
This method will lazily create and return a new CascadedProperties pointer that is cached
in the StyleResolver's state. This will contain only UA rules (for user reverts) and UA/user
rules (for author reverts). These will only be computed at most once for a given element
when doing a reversion, and they will be computed lazily, i.e., only if a revert is
requested.

(WebCore::StyleResolver::applyProperty):
Pass along the LinkMatchMask and the MatchResult to applyProperty. This way we know specifically
which link type we were computing if we have to revert (so that we roll back and look at the
same index in the reverted version). The MatchResult is passed along because it is needed
to build the CascadedProperties rollbacks.

The basic idea is that if a revert is encountered, the level that the rule came from is
checked. If it is UA level, just treat as "unset." If it is author or user level, get
the correct CascadedProperties rollback and repeat the applyProperty using the property
found in the rollback. If the property is not present in the cascade rollback, then the
revert becomes an unset.

(WebCore::StyleResolver::CascadedProperties::hasCustomProperty):
(WebCore::StyleResolver::CascadedProperties::customProperty):
Helpers used by applyProperty to check on custom properties, since they can revert too
just like a regular property can.

(WebCore::StyleResolver::CascadedProperties::setPropertyInternal):
(WebCore::StyleResolver::CascadedProperties::set):
(WebCore::StyleResolver::CascadedProperties::setDeferred):
Passing along the CascadeLevel (UA, User, Author) so that it can be stored in the Property.
This way when we do property application, we always know where the rule came from so
that the reversion can be handled properly.

(WebCore::StyleResolver::CascadedProperties::addStyleProperties):
(WebCore::cascadeLevelForIndex):
(WebCore::StyleResolver::CascadedProperties::addMatches):
When style properties are added, also figure out the CascadeLevel and pass it along to be
stored in the Property. We use the MatchResult's ranges to know where a property comes from.

(WebCore::StyleResolver::CascadedProperties::applyDeferredProperties):
(WebCore::StyleResolver::CascadedProperties::Property::apply):
(WebCore::StyleResolver::applyCascadedProperties):
Pass along the MatchResult so we know how to build the rollback.

  • css/StyleResolver.h:

(WebCore::StyleResolver::State::cascadeLevel):
(WebCore::StyleResolver::State::setCascadeLevel):
(WebCore::StyleResolver::State::authorRollback):
(WebCore::StyleResolver::State::userRollback):
(WebCore::StyleResolver::State::setAuthorRollback):
(WebCore::StyleResolver::State::setUserRollback):
(WebCore::StyleResolver::state):
(WebCore::StyleResolver::cascadeLevel):
(WebCore::StyleResolver::setCascadeLevel):
Move CascadedProperties into the header. Add CascadeLevel to Property. Add the level and
rollbacks to the resolver's state.

LayoutTests:

  • fast/css/all-keyword-revert-expected.html: Added.
  • fast/css/all-keyword-revert.html: Added.
  • fast/css/revert-color-expected.html: Added.
  • fast/css/revert-color.html: Added.
  • fast/css/revert-margins-expected.html: Added.
  • fast/css/revert-margins.html: Added.
  • fast/css/variables/all-keyword-revert-expected.html: Added.
  • fast/css/variables/all-keyword-revert.html: Added.
  • fast/css/variables/revert-inheritance-expected.html: Added.
  • fast/css/variables/revert-inheritance.html: Added.
  • fast/css/variables/revert-no-inheritance-expected.html: Added.
  • fast/css/variables/revert-no-inheritance.html: Added.
  • fast/css/variables/revert-variable-reference-expected.html: Added.
  • fast/css/variables/revert-variable-reference.html: Added.
3:40 PM Changeset in webkit [191251] by mmaxfield@apple.com
  • 5 edits
    2 adds in trunk

Delete FontPlatformData::allowsLigatures()
​https://bugs.webkit.org/show_bug.cgi?id=150286

Reviewed by Dan Bernstein.

Source/WebCore:

This function is only used to force ligatures on for complex fonts (where "complex"
means "does not support the letter 'a'"). However, ligatures are turned on for all
fonts by default, which means that this function is unnecessary.

Required ligatures, such as those which make these complex scripts legible, are always
enabled, no matter what.

Test: fast/text/required-ligatures.html

  • platform/graphics/FontPlatformData.h:
  • platform/graphics/cocoa/FontPlatformDataCocoa.mm:

(WebCore::FontPlatformData::allowsLigatures): Deleted.

  • platform/graphics/mac/SimpleFontDataCoreText.cpp:

(WebCore::Font::getCFStringAttributes):

LayoutTests:

Make sure it's impossible to turn off required ligatures with CSS.

  • fast/text/required-ligatures-expected.html: Added.
  • fast/text/required-ligatures.html: Added.
3:31 PM Changeset in webkit [191250] by mmaxfield@apple.com
  • 10 edits in trunk/Source

Stop honoring the user default "WebKitKerningAndLigaturesEnabledByDefault"
​https://bugs.webkit.org/show_bug.cgi?id=150287

Reviewed by Simon Fraser.

Source/WebCore:

This user default is currently on by default. Therefore, by setting the user default,
users can only disable kerning / ligatures (rather than enable it).

There are a few reasons why we should stop honoring it:

  1. In the brave new world of font-feature-settings and font-variant-ligatures, there

are many different kinds of ligatures which may be enabled at will. The simplistic
statement of "turn on ligatures" no longer has any meaning.

  1. If a user wants to disable kerning / ligatures, he/she can do it with a user

stylesheet.

  1. The default isn't able to be tested with DumpRenderTree or WebKitTestRunner.
  1. I have never heard of anyone actually using this user default.
  • platform/graphics/FontCascade.cpp:

(WebCore::FontCascade::setDefaultKerning): Deleted.
(WebCore::FontCascade::setDefaultLigatures): Deleted.

  • platform/graphics/FontCascade.h:

(WebCore::FontCascade::advancedTextRenderingMode):

Source/WebKit/mac:

  • WebView/WebView.mm:

(+[WebView initialize]): Deleted.

Source/WebKit2:

  • Shared/WebProcessCreationParameters.cpp:

(WebKit::WebProcessCreationParameters::WebProcessCreationParameters): Deleted.
(WebKit::WebProcessCreationParameters::encode): Deleted.
(WebKit::WebProcessCreationParameters::decode): Deleted.

  • Shared/WebProcessCreationParameters.h:
  • UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::registerUserDefaultsIfNeeded): Deleted.
(WebKit::WebProcessPool::platformInitializeWebProcess): Deleted.

  • WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::platformInitializeWebProcess): Deleted.

2:32 PM Changeset in webkit [191249] by mitz@apple.com
  • 5 edits
    2 adds in trunk/Source

[Cocoa] Stop using WKAXRegisterRemoteApp
​https://bugs.webkit.org/show_bug.cgi?id=150283

Reviewed by Alexey Proskuryakov.

Source/WebCore:

  • platform/spi/ios/GraphicsServicesSPI.h: Added declaration of GSSystemRootDirectory.

Source/WebKit2:

  • Platform/spi/mac: Added.
  • Platform/spi/mac/AppKitSPI.h: Added. Includes declaration of +[NSAccessibilityRemoteUIElement setRemoteUIApp:].
  • WebKit2.xcodeproj/project.pbxproj: Added reference to new header. Also changed the path of the Platform/spi/Cocoa group to the corresponding path in the source tree.
  • WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::registerWithAccessibility): Added.
(WebKit::WebProcess::platformInitializeProcess): Use new registerWithAccessibility instead

of WKAXRegisterRemoteApp.

1:27 PM Changeset in webkit [191248] by fpizlo@apple.com
  • 9 edits
    1 add in trunk/Source/JavaScriptCore

Fix some generational heap growth pathologies
​https://bugs.webkit.org/show_bug.cgi?id=150270

Reviewed by Andreas Kling.

When doing generational copying, we would pretend that the size of old space was increased
just by the amount of bytes we copied. In reality, it would be increased by the number of
bytes used by the copied blocks we created. This is a larger number, and in some simple
pathological programs, the difference can be huge.

Fixing this bug was relatively easy, and the only really meaningful change here is in
Heap::updateAllocationLimits(). But to convince myself that the change was valid, I had to
add some debugging code and I had to refactor some stuff so that it made more sense.

This change does obviate the need for m_totalBytesCopied, because we no longer use it in
release builds to decide how much heap we are using at the end of collection. But I added a
FIXME about how we could restore our use of m_totalBytesCopied. So, I kept the logic, for
now. The FIXME references ​https://bugs.webkit.org/show_bug.cgi?id=150268.

Relanding with build fix.

  • CMakeLists.txt:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • heap/CopiedBlock.cpp: Added.

(JSC::CopiedBlock::createNoZeroFill):
(JSC::CopiedBlock::destroy):
(JSC::CopiedBlock::create):
(JSC::CopiedBlock::zeroFillWilderness):
(JSC::CopiedBlock::CopiedBlock):

  • heap/CopiedBlock.h:

(JSC::CopiedBlock::didSurviveGC):
(JSC::CopiedBlock::createNoZeroFill): Deleted.
(JSC::CopiedBlock::destroy): Deleted.
(JSC::CopiedBlock::create): Deleted.
(JSC::CopiedBlock::zeroFillWilderness): Deleted.
(JSC::CopiedBlock::CopiedBlock): Deleted.

  • heap/CopiedSpaceInlines.h:

(JSC::CopiedSpace::startedCopying):

  • heap/Heap.cpp:

(JSC::Heap::updateObjectCounts):
(JSC::Heap::resetVisitors):
(JSC::Heap::capacity):
(JSC::Heap::protectedGlobalObjectCount):
(JSC::Heap::collectImpl):
(JSC::Heap::willStartCollection):
(JSC::Heap::updateAllocationLimits):
(JSC::Heap::didFinishCollection):
(JSC::Heap::sizeAfterCollect): Deleted.

  • heap/Heap.h:
  • heap/HeapInlines.h:

(JSC::Heap::shouldCollect):
(JSC::Heap::isBusy):
(JSC::Heap::collectIfNecessaryOrDefer):

  • heap/MarkedBlock.cpp:

(JSC::MarkedBlock::create):
(JSC::MarkedBlock::destroy):

12:38 PM Changeset in webkit [191247] by Chris Dumez
  • 25 edits
    4 adds in trunk

td and th should use HTMLTableDataCellElement and HTMLTableHeaderCellElement interfaces
​https://bugs.webkit.org/show_bug.cgi?id=148859
<rdar://problem/22588664>

Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

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

Source/WebCore:

td and th should use HTMLTableDataCellElement and HTMLTableHeaderCellElement interfaces
as per the latest HTML specification:
​https://html.spec.whatwg.org/multipage/tables.html#htmltabledatacellelement
​https://html.spec.whatwg.org/multipage/tables.html#htmltableheadercellelement

This patch aligns our behavior with the specification and IE. Firefox and Chrome do not
seem to expose HTMLTableDataCellElement / HTMLTableHeaderCellElement at this time.

The compatibility risk is low, given that the API stays the same and those new
interfaces inherit the pre-existing HTMLTableCellElement interface.

No new tests, already covered by existing tests.

  • CMakeLists.txt:
  • DerivedSources.cpp:
  • DerivedSources.make:
  • WebCore.vcxproj/WebCore.vcxproj:
  • WebCore.vcxproj/WebCore.vcxproj.filters:
  • WebCore.xcodeproj/project.pbxproj:
  • html/HTMLTableCellElement.cpp:

(WebCore::HTMLTableCellElement::HTMLTableCellElement):
(WebCore::HTMLTableCellElement::colSpan): Deleted.

  • html/HTMLTableCellElement.h:
  • html/HTMLTableCellElement.idl:
  • html/HTMLTableDataCellElement.h: Added.
  • html/HTMLTableDataCellElement.idl: Added.
  • html/HTMLTableHeaderCellElement.h: Added.
  • html/HTMLTableHeaderCellElement.idl: Added.
  • html/HTMLTableRowElement.cpp:

(WebCore::HTMLTableRowElement::insertCell):

  • html/HTMLTagNames.in:

LayoutTests:

Rebaseline existing tests now that new HTMLTableDataCellElement / HTMLTableHeaderCellElement
types are exposed.

  • fast/dom/wrapper-classes-expected.txt:
  • fast/dom/wrapper-classes.html:
  • js/dom/global-constructors-attributes-expected.txt:
  • platform/efl/js/dom/global-constructors-attributes-expected.txt:
  • platform/gtk/js/dom/global-constructors-attributes-expected.txt:
  • platform/mac-mavericks/js/dom/global-constructors-attributes-expected.txt:
  • platform/mac-yosemite/js/dom/global-constructors-attributes-expected.txt:
  • platform/mac/js/dom/global-constructors-attributes-expected.txt:
  • platform/win/js/dom/global-constructors-attributes-expected.txt:
12:14 PM Changeset in webkit [191246] by Alan Bujtas
  • 3 edits in trunk/Source/WebCore

RenderBlockFlow::xPositionForFloatIncludingMargin/yPositionForFloatIncludingMargin/flipFloatForWritingModeForChild
should all take FloatingObject reference.
​https://bugs.webkit.org/show_bug.cgi?id=150267

Reviewed by Simon Fraser.

No change in behaviour.

  • rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::addOverflowFromFloats):
(WebCore::RenderBlockFlow::paintFloats):
(WebCore::RenderBlockFlow::clipOutFloatingObjects):
(WebCore::RenderBlockFlow::addOverhangingFloats):
(WebCore::RenderBlockFlow::flipFloatForWritingModeForChild):
(WebCore::RenderBlockFlow::hitTestFloats):
(WebCore::RenderBlockFlow::adjustForBorderFit):

  • rendering/RenderBlockFlow.h:

(WebCore::RenderBlockFlow::xPositionForFloatIncludingMargin):
(WebCore::RenderBlockFlow::yPositionForFloatIncludingMargin):

11:49 AM Changeset in webkit [191245] by Simon Fraser
  • 2 edits in trunk/Source/WebCore

Sort the project file.

  • WebCore.xcodeproj/project.pbxproj:
11:42 AM Changeset in webkit [191244] by commit-queue@webkit.org
  • 9 edits
    1 delete in trunk/Source/JavaScriptCore

Unreviewed, rolling out r191240.
​https://bugs.webkit.org/show_bug.cgi?id=150281

Broke 32-bit builds (Requested by smfr on #webkit).

Reverted changeset:

"Fix some generational heap growth pathologies"
​https://bugs.webkit.org/show_bug.cgi?id=150270
​http://trac.webkit.org/changeset/191240

11:34 AM Changeset in webkit [191243] by Simon Fraser
  • 44 edits
    2 adds in trunk/Source

Enhance TextStream for logging, remove subclasses, log more things
​https://bugs.webkit.org/show_bug.cgi?id=150269

Reviewed by Zalan Bujtas.

Remove the various TextStream subclasses that only existed to support indenting,
and output additional types. Add output for more WebCore and WebKit2 types, and
just use TextStream everywhere.

TextStream is enhance to support grouping (open paren and intent), with a
stack-based class to open/end a group.

Remove some SVG-specific duplicate output functions.

Outdent namespace contents of GraphicsTypes.h.
Source/WebCore:

  • CMakeLists.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • dom/ViewportArguments.cpp:

(WebCore::operator<<):

  • dom/ViewportArguments.h:
  • page/ViewportConfiguration.cpp:

(WebCore::operator<<):
(WebCore::ViewportConfiguration::description):
(WebCore::ViewportConfigurationTextStream::ViewportConfigurationTextStream): Deleted.
(WebCore::ViewportConfigurationTextStream::increaseIndent): Deleted.
(WebCore::ViewportConfigurationTextStream::decreaseIndent): Deleted.
(WebCore::dumpProperty): Deleted.
(WebCore::ViewportConfigurationTextStream::writeIndent): Deleted.
(WebCore::ViewportConfigurationTextStream::operator<<): Deleted.

  • page/ViewportConfiguration.h:
  • page/scrolling/ScrollingConstraints.cpp:

(WebCore::operator<<):

  • page/scrolling/ScrollingConstraints.h:
  • page/scrolling/ScrollingCoordinator.cpp:

(WebCore::operator<<):

  • page/scrolling/ScrollingCoordinator.h:
  • platform/animation/TimingFunction.cpp: Added.

(WebCore::operator<<):

  • platform/animation/TimingFunction.h:
  • platform/graphics/Color.cpp:

(WebCore::operator<<):

  • platform/graphics/Color.h:
  • platform/graphics/FloatPoint3D.cpp:

(WebCore::operator<<):

  • platform/graphics/FloatPoint3D.h:
  • platform/graphics/FloatRoundedRect.cpp:

(WebCore::operator<<):

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

(WebCore::operator<<):

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

(WebCore::operator<<):

  • platform/graphics/GraphicsTypes.h:
  • platform/graphics/ca/PlatformCAAnimation.cpp: Added.

(WebCore::operator<<):

  • platform/graphics/ca/PlatformCAAnimation.h:
  • platform/graphics/ca/PlatformCALayer.cpp:

(WebCore::operator<<):

  • platform/graphics/ca/PlatformCALayer.h:
  • platform/graphics/filters/FilterOperation.cpp:

(WebCore::operator<<):

  • platform/graphics/filters/FilterOperation.h:
  • platform/graphics/filters/FilterOperations.cpp:

(WebCore::operator<<):

  • platform/graphics/filters/FilterOperations.h:
  • platform/graphics/filters/PointLightSource.cpp:

(WebCore::operator<<): Deleted.

  • platform/graphics/filters/SpotLightSource.cpp:

(WebCore::operator<<): Deleted.

  • platform/graphics/transforms/AffineTransform.cpp:

(WebCore::operator<<):

  • platform/graphics/transforms/AffineTransform.h:
  • platform/graphics/transforms/TransformationMatrix.cpp:

(WebCore::operator<<):

  • platform/graphics/transforms/TransformationMatrix.h:
  • platform/text/TextStream.cpp:

(WebCore::TextStream::startGroup):
(WebCore::TextStream::endGroup):
(WebCore::TextStream::nextLine):
(WebCore::TextStream::writeIndent):

  • platform/text/TextStream.h:

(WebCore::TextStream::operator<<):
(WebCore::TextStream::dumpProperty):
(WebCore::TextStream::increaseIndent):
(WebCore::TextStream::decreaseIndent):
(WebCore::TextStream::GroupScope::GroupScope):
(WebCore::TextStream::GroupScope::~GroupScope):

  • rendering/svg/SVGRenderTreeAsText.cpp:

(WebCore::operator<<): Deleted.

  • rendering/svg/SVGRenderTreeAsText.h:

Source/WebKit2:

  • Shared/Scrolling/RemoteScrollingCoordinatorTransaction.cpp:

(WebKit::dump):
(WebKit::recursiveDumpNodes):
(WebKit::RemoteScrollingCoordinatorTransaction::description):
(WebKit::RemoteScrollingTreeTextStream::RemoteScrollingTreeTextStream): Deleted.
(WebKit::RemoteScrollingTreeTextStream::increaseIndent): Deleted.
(WebKit::RemoteScrollingTreeTextStream::decreaseIndent): Deleted.
(WebKit::RemoteScrollingTreeTextStream::writeIndent): Deleted.
(WebKit::dumpProperty): Deleted.
(WebKit::RemoteScrollingTreeTextStream::operator<<): Deleted.
(WebKit::RemoteScrollingTreeTextStream::dump): Deleted.
(WebKit::RemoteScrollingTreeTextStream::recursiveDumpNodes): Deleted.

  • Shared/mac/RemoteLayerTreeTransaction.mm:

(WebKit::operator<<):
(WebKit::dumpChangedLayers):
(WebKit::RemoteLayerTreeTransaction::description):
(WebKit::RemoteLayerTreeTextStream::RemoteLayerTreeTextStream): Deleted.
(WebKit::RemoteLayerTreeTextStream::increaseIndent): Deleted.
(WebKit::RemoteLayerTreeTextStream::decreaseIndent): Deleted.
(WebKit::dumpProperty): Deleted.
(WebKit::RemoteLayerTreeTextStream::operator<<): Deleted.
(WebKit::RemoteLayerTreeTextStream::writeIndent): Deleted.

  • WebProcess/WebPage/mac/PlatformCAAnimationRemote.h:
  • WebProcess/WebPage/mac/PlatformCAAnimationRemote.mm:

(WebKit::operator<<):

11:04 AM Changeset in webkit [191242] by commit-queue@webkit.org
  • 3 edits in trunk/Source/JavaScriptCore

[Win] Fix the Windows build.
​https://bugs.webkit.org/show_bug.cgi?id=150278

Patch by Sungmann Cho <​sungmann.cho@navercorp.com> on 2015-10-17
Reviewed by Brent Fulgham.

10:52 AM Changeset in webkit [191241] by mark.lam@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Fixed typos from r191224.

Not reviewed.

  • jit/JITSubGenerator.h:

(JSC::JITSubGenerator::generateFastPath):

10:26 AM Changeset in webkit [191240] by fpizlo@apple.com
  • 9 edits
    1 add in trunk/Source/JavaScriptCore

Fix some generational heap growth pathologies
​https://bugs.webkit.org/show_bug.cgi?id=150270

Reviewed by Andreas Kling.

When doing generational copying, we would pretend that the size of old space was increased
just by the amount of bytes we copied. In reality, it would be increased by the number of
bytes used by the copied blocks we created. This is a larger number, and in some simple
pathological programs, the difference can be huge.

Fixing this bug was relatively easy, and the only really meaningful change here is in
Heap::updateAllocationLimits(). But to convince myself that the change was valid, I had to
add some debugging code and I had to refactor some stuff so that it made more sense.

This change does obviate the need for m_totalBytesCopied, because we no longer use it in
release builds to decide how much heap we are using at the end of collection. But I added a
FIXME about how we could restore our use of m_totalBytesCopied. So, I kept the logic, for
now. The FIXME references ​https://bugs.webkit.org/show_bug.cgi?id=150268.

  • CMakeLists.txt:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • heap/CopiedBlock.cpp: Added.

(JSC::CopiedBlock::createNoZeroFill):
(JSC::CopiedBlock::destroy):
(JSC::CopiedBlock::create):
(JSC::CopiedBlock::zeroFillWilderness):
(JSC::CopiedBlock::CopiedBlock):

  • heap/CopiedBlock.h:

(JSC::CopiedBlock::didSurviveGC):
(JSC::CopiedBlock::createNoZeroFill): Deleted.
(JSC::CopiedBlock::destroy): Deleted.
(JSC::CopiedBlock::create): Deleted.
(JSC::CopiedBlock::zeroFillWilderness): Deleted.
(JSC::CopiedBlock::CopiedBlock): Deleted.

  • heap/CopiedSpaceInlines.h:

(JSC::CopiedSpace::startedCopying):

  • heap/Heap.cpp:

(JSC::Heap::updateObjectCounts):
(JSC::Heap::resetVisitors):
(JSC::Heap::capacity):
(JSC::Heap::protectedGlobalObjectCount):
(JSC::Heap::collectImpl):
(JSC::Heap::willStartCollection):
(JSC::Heap::updateAllocationLimits):
(JSC::Heap::didFinishCollection):
(JSC::Heap::sizeAfterCollect): Deleted.

  • heap/Heap.h:
  • heap/HeapInlines.h:

(JSC::Heap::shouldCollect):
(JSC::Heap::isBusy):
(JSC::Heap::collectIfNecessaryOrDefer):

  • heap/MarkedBlock.cpp:

(JSC::MarkedBlock::create):
(JSC::MarkedBlock::destroy):

1:23 AM Changeset in webkit [191239] by Csaba Osztrogonác
  • 3 edits in trunk/Source/WebKit2

URTBF after r191194 to fix EFL build.
​https://bugs.webkit.org/show_bug.cgi?id=150272

Patch by Hunseop Jeong <Hunseop Jeong> on 2015-10-17
Reviewed by Csaba Osztrogonác.

  • UIProcess/efl/WebContextMenuProxyEfl.cpp:

(WebKit::WebContextMenuProxyEfl::showContextMenu):

  • UIProcess/efl/WebContextMenuProxyEfl.h:
12:10 AM Changeset in webkit [191238] by youenn.fablet@crf.canon.fr
  • 2 edits in trunk/Source/WebCore

Finalize bug 149952 patch
​https://bugs.webkit.org/show_bug.cgi?id=150238

Reviewed by Darin Adler.

No change in behavior.

  • bindings/js/JSDOMConstructor.h:

(WebCore::JSDOMConstructor<JSClass>::finishCreation): Marked as inline.
(WebCore::JSDOMConstructor<JSClass>::getConstructData): Marked as inline.

Note: See TracTimeline for information about the timeline view.