⚠ Archived content — this site is no longer maintained.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

Timeline



Apr 2, 2016:

6:55 PM Changeset in webkit [198979] by commit-queue@webkit.org
  • 2 edits in trunk/Source/JavaScriptCore

Unreviewed, rolling out r198976.
​https://bugs.webkit.org/show_bug.cgi?id=156140

"Causes js/regress/array-nonarray-polymorhpic-access.html to
crash." (Requested by ddkilzer on #webkit).

Reverted changeset:

"[JSC] Initialize SSA's live values at tail lazily"
​https://bugs.webkit.org/show_bug.cgi?id=156126
​http://trac.webkit.org/changeset/198976

6:22 PM Changeset in webkit [198978] by gskachkov@gmail.com
  • 3 edits in trunk/LayoutTests

[ES6] Arrow function syntax. Add missed test for using 'with' and 'this' in arrow function.
​https://bugs.webkit.org/show_bug.cgi?id=156059

Reviewed by Saam Barati.

Added test case that cover using 'with' with lexically bound 'this' in arrow function, this case was not
covered by tests for arrow function.

  • js/arrowfunction-lexical-bind-this-expected.txt:
  • js/script-tests/arrowfunction-lexical-bind-this.js:
3:04 PM Changeset in webkit [198977] by dbates@webkit.org
  • 2 edits in trunk/Source/WebKit2

REGRESSION (r198933): Web Inspector and inspected page hang in local builds
​https://bugs.webkit.org/show_bug.cgi?id=156138
<rdar://problem/25507724>

Reviewed by Alexey Proskuryakov.

Similar to the sandbox policy for WebProcess and NetworkProcess, grant a Database process
read-only filesystem access and mach lookup capability to various Security framework files
and SecurityServer, respectively.

  • DatabaseProcess/mac/com.apple.WebKit.Databases.sb.in:
2:16 PM Changeset in webkit [198976] by commit-queue@webkit.org
  • 2 edits in trunk/Source/JavaScriptCore

[JSC] Initialize SSA's live values at tail lazily
​https://bugs.webkit.org/show_bug.cgi?id=156126

Patch by Benjamin Poulain <​bpoulain@apple.com> on 2016-04-02
Reviewed by Mark Lam.

Setting up the clean state early looks harmless but it is
actually quite expensive.

The problem is AbstractValue is gigantic, you really want
to minimize how much you touch that memory.

By removing the initialization, most blocks only
get 2 or 3 accesses. Once to setup the value, and a few
queries for merging the current block with the successors.

  • dfg/DFGInPlaceAbstractState.cpp:

(JSC::DFG::InPlaceAbstractState::endBasicBlock):
(JSC::DFG::setLiveValues): Deleted.
(JSC::DFG::InPlaceAbstractState::initialize): Deleted.

2:14 PM Changeset in webkit [198975] by commit-queue@webkit.org
  • 3 edits in trunk/Source/JavaScriptCore

[JSC] Add an option to avoid disassembling baseline code for the JSC Profiler
​https://bugs.webkit.org/show_bug.cgi?id=156127

Patch by Benjamin Poulain <​bpoulain@apple.com> on 2016-04-02
Reviewed by Mark Lam.

The profiler run out of memory on big programs if you dump
the baseline disassembly.

  • jit/JIT.cpp:

(JSC::JIT::privateCompile):

  • runtime/Options.h:
1:01 PM Changeset in webkit [198974] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebCore

Cleanup DataDetection.mm a little bit
​https://bugs.webkit.org/show_bug.cgi?id=156128

Patch by Sam Weinig <​sam@webkit.org> on 2016-04-02
Reviewed by Dan Bernstein.

  • html/HTMLAttributeNames.in:

Add x-apple-data-detectors, x-apple-data-detectors-type, x-apple-data-detectors-result

  • editing/cocoa/DataDetection.mm:

(WebCore::DataDetection::requiresExtendedContext):
(WebCore::DataDetection::dataDetectorIdentifier):
(WebCore::DataDetection::shouldCancelDefaultAction):
(WebCore::removeResultLinksFromAnchor):
(WebCore::searchForLinkRemovingExistingDDLinks):
Use new generated qualified name attributes, switch to using fastGetAttribute and
equalIgnoringASCIICase, and fix the position of some *s.

(WebCore::dataDetectorStringForPath):
Fix some *s and switch to using StringBuilder.

(WebCore::buildQuery):
Fix some *s.

(WebCore::DataDetection::detectContentInRange):
Use new generated qualified name attributes.

12:34 PM Changeset in webkit [198973] by dbates@webkit.org
  • 2 edits in trunk/Source/WebKit2

Another attempt to fix the Apple Internal Yosemite and El Capitan builds following
<​http://trac.webkit.org/changeset/198933> (​https://bugs.webkit.org/show_bug.cgi?id=155455)

Similar to the sandbox policy for WebProcess and NetworkProcess, grant a Database process
access to the passed user's cache directory and passed user's temporary directory so that
the Security framework can verify the code signature of the app that launched it when the
Database process is itself code signed.

  • DatabaseProcess/mac/com.apple.WebKit.Databases.sb.in:
9:34 AM Changeset in webkit [198972] by mitz@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

jsc binary embedded in relocatable JavaScriptCore.framework links against system JavaScriptCore.framework
​https://bugs.webkit.org/show_bug.cgi?id=156134
<rdar://problem/25443824>

Reviewed by Mark Lam.

  • Configurations/JSC.xcconfig: Define WK_RELOCATABLE_FRAMEWORKS_LDFLAGS when building relocatable frameworks to include a -dyld_env option setting DYLD_FRAMEWORK_PATH to point to the directory containing JavaScript.framework, and add WK_RELOCATABLE_FRAMEWORKS_LDFLAGS to OTHER_LDFLAGS.
1:59 AM Changeset in webkit [198971] by ddkilzer@apple.com
  • 2 edits in trunk/LayoutTests

REGRESSION (r198963): css3/filters/backdrop/dynamic-backdrop-filter-change.html frequently fails
<​http://webkit.org/b/149319>

Unreviewed attempt to fix layout test.

  • css3/filters/backdrop/dynamic-backdrop-filter-change.html:

Use testRunner.notifyDone() after testRunner.waitUntilDone().

12:01 AM Changeset in webkit [198970] by mmaxfield@apple.com
  • 9 edits in trunk/Source/WebCore

Migrate BidiRunList and BidiRun to automatic memory management
​https://bugs.webkit.org/show_bug.cgi?id=156123

Reviewed by Simon Fraser.

BidiRunList, BidiRun, and BidiCharacterRun have all been doing manual
"new"s and "delete"s for years. This patch migrates those classes to
using std::unique_ptr.

No new tests because there is no behavior change.

  • platform/graphics/GraphicsContext.cpp:

(WebCore::GraphicsContext::drawBidiText):

  • platform/text/BidiResolver.h:

(WebCore::BidiCharacterRun::BidiCharacterRun):
(WebCore::BidiCharacterRun::next):
(WebCore::BidiCharacterRun::takeNext):
(WebCore::BidiCharacterRun::setNext):
(WebCore::Subclass>::appendRunInternal):

  • platform/text/BidiRunList.h:

(WebCore::BidiRunList::BidiRunList):
(WebCore::BidiRunList::firstRun):
(WebCore::BidiRunList<Run>::appendRun):
(WebCore::BidiRunList<Run>::prependRun):
(WebCore::BidiRunList<Run>::moveRunToEnd):
(WebCore::BidiRunList<Run>::moveRunToBeginning):
(WebCore::BidiRunList<Run>::replaceRunWithRuns):
(WebCore::BidiRunList<Run>::clear):
(WebCore::BidiRunList<Run>::reverseRuns):
(WebCore::BidiRunList<Run>::clearWithoutDestroyingRuns): Deleted.
(WebCore::BidiRunList<Run>::deleteRuns): Deleted.

  • rendering/BidiRun.cpp:

(WebCore::BidiRun::takeNext):

  • rendering/BidiRun.h:

(WebCore::BidiRun::next):
(WebCore::BidiRun::takeNext):

  • rendering/InlineIterator.h:

(WebCore::addPlaceholderRunForIsolatedInline):

  • rendering/RenderBlockLineLayout.cpp:

(WebCore::createRun):
(WebCore::RenderBlockFlow::handleTrailingSpaces):
(WebCore::RenderBlockFlow::layoutRunsAndFloatsInRange):

  • rendering/line/LineBreaker.cpp:

(WebCore::LineBreaker::skipLeadingWhitespace):

Apr 1, 2016:

11:38 PM MathML/Early_2016_Refactoring edited by fred.wang@free.fr
(diff)
10:41 PM Changeset in webkit [198969] by Joseph Pecoraro
  • 2 edits in trunk/Source/WebInspectorUI

Unreviewed follow-up to r198968. Remove debug statements.

  • Scripts/remove-console-asserts.pl:

(removeConsoleAssertsInFile):

10:23 PM Changeset in webkit [198968] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebInspectorUI

Web Inspector: Remove console.asserts from Worker code in optimized builds
​https://bugs.webkit.org/show_bug.cgi?id=156124
<rdar://problem/25505031>

Patch by Joseph Pecoraro <Joseph Pecoraro> on 2016-04-01
Reviewed by Timothy Hatcher.

  • Scripts/copy-user-interface-resources.pl:

Remove console asserts on the Worker results directory.

  • Scripts/remove-console-asserts.pl:

(removeConsoleAssertsInFile):
(removeConsoleAssertsInDirectory):
Refactor into functions for a file and a file for
recursing through directories.

8:21 PM Changeset in webkit [198967] by dbates@webkit.org
  • 2 edits in trunk/Source/WebKit2

Attempt to fix the Apple Internal Yosemite and El Capitan builds following
<​http://trac.webkit.org/changeset/198933> (​https://bugs.webkit.org/show_bug.cgi?id=155455)
<rdar://problem/25502181>

Only check the code signature of the client app if the "user-directory-suffix" sandbox
initialization parameter will not be used. This parameter is only used/passed as an
initialization parameter if the client app is not sandboxed, which seems sufficient
criterion to determine whether it is safe to honor it.

  • Shared/mac/ChildProcessMac.mm:

(WebKit::ChildProcess::initializeSandbox):

7:30 PM Changeset in webkit [198966] by bshafiei@apple.com
  • 5 edits in tags/Safari-602.1.26.0.1/Source

Versioning.

7:28 PM Changeset in webkit [198965] by bshafiei@apple.com
  • 1 copy in tags/Safari-602.1.26.0.1

New tag.

6:48 PM Changeset in webkit [198964] by jiewen_tan@apple.com
  • 3 edits
    1 move in trunk/Tools

Build fix for r198956.

Unreviewed.

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKit2Cocoa/ProvisionalURLNotChange.mm: Renamed from Tools/TestWebKitAPI/Tests/WebKit2Cocoa/ProvisionalURLChange.mm.

(-[ProvisionalURLNotChangeController webView:didFinishNavigation:]):
(-[ProvisionalURLNotChangeController webView:didFailProvisionalNavigation:withError:]):
(TEST):

  • TestWebKitAPI/Tests/WebKit2ObjC/CustomProtocolsInvalidScheme.mm:

(-[LoadInvalidSchemeDelegate browsingContextController:didFailProvisionalLoadWithError:]):
(TestWebKitAPI::TEST):

5:19 PM Changeset in webkit [198963] by Simon Fraser
  • 3 edits
    2 adds in trunk

Backdrop Filter should repaint when changed via script
​https://bugs.webkit.org/show_bug.cgi?id=149319
rdar://problem/22749892

Reviewed by Tim Horton.

Source/WebCore:

When backdrop filters change, we need to trigger a recomposite, just as we do
for filters. It's ok to re-use ContextSensitivePropertyFilter because
adjustStyleDifference() does the right thing.

Test: css3/filters/backdrop/dynamic-backdrop-filter-change.html

  • rendering/style/RenderStyle.cpp:

(WebCore::RenderStyle::changeRequiresLayerRepaint):

LayoutTests:

  • css3/filters/backdrop/dynamic-backdrop-filter-change-expected.html: Added.
  • css3/filters/backdrop/dynamic-backdrop-filter-change.html: Added.
5:09 PM Changeset in webkit [198962] by bshafiei@apple.com
  • 5 edits in branches/safari-601-branch/Source

Versioning.

5:08 PM Changeset in webkit [198961] by bshafiei@apple.com
  • 5 edits in branches/safari-601.1.46-branch/Source

Versioning.

4:52 PM Changeset in webkit [198960] by bshafiei@apple.com
  • 1 copy in tags/Safari-601.1.46.125

New tag.

4:51 PM Changeset in webkit [198959] by bshafiei@apple.com
  • 1 copy in tags/Safari-601.6.14

New tag.

4:25 PM Changeset in webkit [198958] by Alan Bujtas
  • 3 edits
    2 adds in trunk

putImageData leaves visible artifacts on retina display
​https://bugs.webkit.org/show_bug.cgi?id=156039
<rdar://problem/25482243>

Reviewed by Simon Fraser.

Inflate the repaint rect to cover anti-aliasing bits.

Source/WebCore:

Test: fast/canvas/hidpi-repaint-on-retina-leaves-bits-behind.html

  • html/HTMLCanvasElement.cpp:

(WebCore::HTMLCanvasElement::didDraw):

LayoutTests:

  • fast/canvas/hidpi-repaint-on-retina-leaves-bits-behind-expected.html: Added.
  • fast/canvas/hidpi-repaint-on-retina-leaves-bits-behind.html: Added.
2:44 PM Changeset in webkit [198957] by matthew_hanson@apple.com
  • 5 edits in branches/safari-601.1.46-branch/Source/WebCore

Roll out r198035 via r198903. rdar://problem/25467558

2:40 PM Changeset in webkit [198956] by jiewen_tan@apple.com
  • 9 edits
    3 adds in trunk

WebKit should dispatchDidFailProvisionalLoad while loading invalid URLs
​https://bugs.webkit.org/show_bug.cgi?id=155995
<rdar://problem/14967004>

Reviewed by Andy Estes.

Source/WebCore:

Added API Tests.

If a loading request contains an invalid URL, DocumentLoader will now dispatch
cannotShowURLError to the clients.

  • loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::startLoadingMainResource):

Source/WebKit2:

Ensure that alternative HTML string will not be loaded back to back for
failing provisional loads.

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::loadAlternateHTMLString):
(WebKit::WebPageProxy::didFinishLoadForFrame):

  • UIProcess/WebPageProxy.h:

Tools:

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKit2Cocoa/LoadAlternateHTMLString.mm:

(-[LoadAlternateHTMLStringFromProvisionalLoadErrorController webView:didFailProvisionalNavigation:withError:]):
(-[LoadAlternateHTMLStringFromProvisionalLoadErrorController webView:didStartProvisionalNavigation:]):
(TEST):

  • TestWebKitAPI/Tests/WebKit2Cocoa/LoadInvalidURLRequest.mm: Added.

(literalURL):
(-[LoadInvalidURLNavigationActionDelegate webView:didCommitNavigation:]):
(-[LoadInvalidURLNavigationActionDelegate webView:didFailProvisionalNavigation:withError:]):
(TestWebKitAPI::TEST):

  • TestWebKitAPI/Tests/mac/LoadInvalidURLRequest.html: Added.
  • TestWebKitAPI/Tests/mac/LoadInvalidURLRequest.mm: Added.

(-[LoadInvalidURLWebFrameLoadDelegate webView:didCommitLoadForFrame:]):
(-[LoadInvalidURLWebFrameLoadDelegate webView:didFailProvisionalLoadWithError:forFrame:]):
(TestWebKitAPI::TEST):

2:10 PM Changeset in webkit [198955] by Brent Fulgham
  • 23 edits
    9 adds in trunk

Confirm <a download> satisfies specification criteria
​https://bugs.webkit.org/show_bug.cgi?id=156057

Reviewed by Andy Estes.

Source/WebKit2:

  • WebProcess/InjectedBundle/API/c/WKBundleNavigationAction.cpp:

(WKBundleNavigationActionGetHasDownloadAttribute): Added.

  • WebProcess/InjectedBundle/API/c/WKBundleNavigationActionPrivate.h:
  • WebProcess/InjectedBundle/InjectedBundleNavigationAction.cpp:

(WebKit::InjectedBundleNavigationAction::InjectedBundleNavigationAction):

  • WebProcess/InjectedBundle/InjectedBundleNavigationAction.h:

(WebKit::InjectedBundleNavigationAction::hasDownloadAttribute): Added.

Tools:

Revise WebKitTestRunner:
(1) Provide a download client delegate.
(2) Teach WKTR to wait to finish a test until a download finishes.
(3) Allow WKTR to notify the current TestRunner that it can finish.

  • WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
  • WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:

(WTR::InjectedBundle::didReceiveMessageToPage): Update to recognize a new
message "NotifyDownloadDone".

  • WebKitTestRunner/InjectedBundle/InjectedBundlePage.cpp:

(WTR::InjectedBundlePage::decidePolicyForNavigationAction): Handle 'download'
attribute case.

  • WebKitTestRunner/InjectedBundle/TestRunner.cpp:

(WTR::TestRunner::waitUntilDownload): Added.

  • WebKitTestRunner/InjectedBundle/TestRunner.h:

(WTR::TestRunner::shouldFinishAfterDownload): Added.

  • WebKitTestRunner/TestController.cpp:

(WTR::TestController::createWebViewWithOptions): Set up the download client.
(WTR::TestController::downloadDidStart): Added.
(WTR::TestController::decideDestinationWithSuggestedFilename): Added.
(WTR::TestController::downloadDidFinish): Added.
(WTR::TestController::downloadDidFail): Added.
(WTR::TestController::downloadDidCancel): Added.

  • WebKitTestRunner/TestController.h:
  • WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::notifyDownloadDone): Added.

  • WebKitTestRunner/TestInvocation.h:

LayoutTests:

Import <a download> tests from the Blink project.

Based on a Blink change (patch by <​asanka@chromium.org>):
<​https://src.chromium.org/viewvc/blink?revision=172767&view=revision>

  • TestExpectations: Mark two expected failures with the new tests.
  • fast/dom/HTMLAnchorElement/anchor-download-expected.txt: Updated for

expected output (blob URL currently not working).

  • fast/dom/HTMLAnchorElement/anchor-nodownload-set-expected.txt: Ditto.
  • http/tests/security/anchor-download-allow-blob-expected.txt: Added.
  • http/tests/security/anchor-download-allow-blob.html: Added.
  • http/tests/security/anchor-download-allow-data-expected.txt: Added.
  • http/tests/security/anchor-download-allow-data.html: Added.
  • http/tests/security/anchor-download-allow-sameorigin-expected.txt: Added.
  • http/tests/security/anchor-download-allow-sameorigin.html: Added.
  • http/tests/security/anchor-download-block-crossorigin-expected.txt: Added.
  • http/tests/security/anchor-download-block-crossorigin.html: Added.
  • http/tests/security/resources/attachment.php: Added.
  • platform/ios-simulator-wk1/TestExpectations: Disable download tests until WK1 support

is implemented.

  • platform/ios-simulator-wk2/TestExpectations: Disable download tests until

NETWORK_SESSION support exists.

  • platform/mac/TestExpectations: Reactivate the HTMLAnchorElement tests.
  • platform/mac-wk1/TestExpectations: Disable download tests until WK1 support

is implemented.

1:58 PM Changeset in webkit [198954] by Ryan Haddad
  • 4 edits in trunk/LayoutTests

Rebaseline tests added with r198951
​https://bugs.webkit.org/show_bug.cgi?id=156092

Unreviewed test gardening.

  • http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-child-src-expected.txt:
  • http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-default-src-expected.txt:
  • http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-frame-src-expected.txt:
12:20 PM Changeset in webkit [198953] by commit-queue@webkit.org
  • 4 edits in trunk/Source/JavaScriptCore

[JSC][x86] Add the 3 operands form of floating point substraction
​https://bugs.webkit.org/show_bug.cgi?id=156095

Patch by Benjamin Poulain <​bpoulain@apple.com> on 2016-04-01
Reviewed by Geoffrey Garen.

Same old, same old. Add the AVX form of subsd and subss.

Unfortunately, we cannot benefit from the 3 register form
in B3 yet because the Air script does not support CPU flags yet.
That can be fixed later.

  • assembler/MacroAssemblerX86Common.h:

(JSC::MacroAssemblerX86Common::subDouble):
(JSC::MacroAssemblerX86Common::subFloat):

  • assembler/X86Assembler.h:

(JSC::X86Assembler::vsubsd_rr):
(JSC::X86Assembler::subsd_mr):
(JSC::X86Assembler::vsubsd_mr):
(JSC::X86Assembler::vsubss_rr):
(JSC::X86Assembler::subss_mr):
(JSC::X86Assembler::vsubss_mr):
(JSC::X86Assembler::X86InstructionFormatter::SingleInstructionBufferWriter::memoryModRM):

  • b3/air/AirOpcode.opcodes:
12:15 PM Changeset in webkit [198952] by commit-queue@webkit.org
  • 3 edits
    2 adds in trunk

Reset CSS spacing rules on the <math> element.
​https://bugs.webkit.org/show_bug.cgi?id=155433

Patch by Frederic Wang <​fwang@igalia.com> on 2016-04-01
Reviewed by Brent Fulgham.

Source/WebCore:

Test: mathml/presentation/css-spacing.html

When the authors has changed some CSS spacing rules on a page, they are currently inherited on MathML too, potentially causing broken math rendering.
For example, this caused a bug in Firefox on Wikipedia in the past. Hence we align on Gecko and just reset the CSS spacing rules by default. We also import Mozilla's test.

  • css/mathml.css:

(math): Set line-height, word-spacing and letter-spacing to normal.

LayoutTests:

We import a test from Mozilla to ensure that CSS spacing rules line-height, word-spacing and letter-spacing are by default reset to normal on the <math> element.

  • mathml/presentation/css-spacing-expected.html: Added.
  • mathml/presentation/css-spacing.html: Added.
11:40 AM Changeset in webkit [198951] by dbates@webkit.org
  • 6 edits
    6 adds in trunk

CSP: child-src violations reported as frame-src violation
​https://bugs.webkit.org/show_bug.cgi?id=156092
<rdar://problem/25478509>

Reviewed by Andy Estes.

Source/WebCore:

Tests: http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-child-src.html

http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-default-src.html
http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-frame-src.html

  • page/csp/ContentSecurityPolicy.cpp:

(WebCore::ContentSecurityPolicy::allowChildFrameFromSource): Determine the name of the effective violation
directive from the name of the violated directive. If the name of the violated directive is "frame-src"
then use that name for the name of the effective violated directive. Otherwise, use "child-src" for the
name of the effective violated directive. A byproduct of this decision is that we report child-src as the
effective violated directive when a frame load was blocked by the default-src directive. This seems reasonable
because directive frame-src is deprecated in Content Security Policy Level 2. The child-src directive is
its replacement.

LayoutTests:

  • TestExpectations: Mark the newly added tests as PASS so that we run them.
  • http/tests/security/contentSecurityPolicy/1.1/child-src/frame-blocked-expected.txt: Substitute "child-src" for "frame-src" in the

expected console warning.

  • http/tests/security/contentSecurityPolicy/1.1/child-src/frame-redirect-blocked-expected.txt: Ditto.
  • http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-child-src-expected.txt: Added.
  • http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-child-src.html: Added.
  • http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-default-src-expected.txt: Added.
  • http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-default-src.html: Added.
  • http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-frame-src-expected.txt: Added.
  • http/tests/security/contentSecurityPolicy/1.1/securityviolationpolicy-block-frame-using-frame-src.html: Added.
11:22 AM Changeset in webkit [198950] by dbates@webkit.org
  • 2 edits in trunk

Fix up ChangeLog entries for r198936 (​https://bugs.webkit.org/show_bug.cgi?id=156079) to reference
the update bug title that reflects that r198936 fixes the regression caused by r196012.

11:15 AM Changeset in webkit [198949] by adachan@apple.com
  • 2 edits in trunk/LayoutTests

Add a way to reset the eventTrigger in ControlsTest in LayoutTests/media/controls/controls-test-helpers.js
​https://bugs.webkit.org/show_bug.cgi?id=154616

Reviewed by Dean Jackson.

  • media/controls/controls-test-helpers.js:

(ControlsTest.prototype.resetEventTrigger):

10:55 AM Changeset in webkit [198948] by achristensen@apple.com
  • 2 edits in trunk/Source/WebKit2

CMake build fix.

  • PlatformMac.cmake:
9:45 AM Changeset in webkit [198947] by commit-queue@webkit.org
  • 19 edits
    1 copy
    1 add in trunk

Compile DumpRenderTree with CMake on Mac
​https://bugs.webkit.org/show_bug.cgi?id=156088

Patch by Alex Christensen <​achristensen@webkit.org> on 2016-04-01
Reviewed by Brent Fulgham.

Source/WebCore:

  • PlatformMac.cmake:
  • platform/mock/mediasource/MockBox.cpp:

Source/WebKit:

  • PlatformMac.cmake:

Source/WebKit/mac:

  • Misc/WebKit.h:

Alphabetize headers.

Source/WebKit2:

  • CMakeLists.txt:
  • PlatformMac.cmake:
  • UIProcess/API/Cocoa/WebKitLegacy.h: Added.

Tools:

  • CMakeLists.txt:
  • DumpRenderTree/CMakeLists.txt:
  • DumpRenderTree/DefaultPolicyDelegate.m:

(-[DefaultPolicyDelegate webView:decidePolicyForNavigationAction:request:frame:decisionListener:]):

  • DumpRenderTree/DumpRenderTreeFileDraggingSource.m:
  • DumpRenderTree/PlatformMac.cmake: Added.
  • DumpRenderTree/PlatformWin.cmake:
  • DumpRenderTree/mac/AppleScriptController.m:

(convertAEDescToObject):

  • DumpRenderTree/mac/DumpRenderTreeMain.mm:
  • DumpRenderTree/mac/LayoutTestHelper.m:
9:18 AM Changeset in webkit [198946] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

Marking three media controls tests as flaky on El Capitan
​https://bugs.webkit.org/show_bug.cgi?id=156112

Unreviewed test gardening.

  • platform/mac/TestExpectations:
3:49 AM WebKitGTK/2.12.x edited by berto@igalia.com
(diff)
3:36 AM Changeset in webkit [198945] by berto@igalia.com
  • 2 edits in trunk/Source/JavaScriptCore

[JSC] Missing PATH_MAX definition
​https://bugs.webkit.org/show_bug.cgi?id=156102

Reviewed by Yusuke Suzuki.

Not all systems define PATH_MAX, so add a fallback value that is
long enough.

  • jsc.cpp:
3:21 AM Changeset in webkit [198944] by jh718.park@samsung.com
  • 2 edits
    1 add in trunk/LayoutTests

Unreviewed EFL gardening about new RTL scrollbar testcase
This testcase has been added in r198859.

  • platform/efl/TestExpectations:
  • platform/efl/fast/scrolling/rtl-scrollbars-animation-property-expected.txt: Added.
2:54 AM Changeset in webkit [198943] by Antti Koivisto
  • 41 edits
    5 adds in trunk

Separate render tree updating from style resolve
​https://bugs.webkit.org/show_bug.cgi?id=155298

Reviewed by Andreas Kling.

Source/WebCore:

This patch splits computing document style and applying the results into two distinct steps:

Style::TreeResolver::resolve()

|
| Style::Update
V

RenderTreeUpdater::commit()

Style::TreeResolver::resolve() returns a Style::Update object that contains all the changes to be made
for the whole composed tree. RenderTreeUpdater then applies the changes updating, building or tearing
down portions of the render tree as needed.

Style::Update consists of a map that contains new style for each newly resolved element along with some
metadata. A separate map contains text nodes that require reconstruction. It also tracks change roots so
RenderTreeUpdater needs to traverse the changed subtrees only.

The patch eliminates the recursive render tree build code path replacing it with iterative functions.

This will enable future optimizations. For example we won't need to commit to immediate rendering
changes simply because some script or internal function requires up-to-date style.

  • CMakeLists.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • css/StyleResolver.cpp:

(WebCore::StyleResolver::State::State):
(WebCore::StyleResolver::styleForElement):

  • css/StyleResolver.h:

(WebCore::StyleResolver::setOverrideDocumentElementStyle):
(WebCore::StyleResolver::State::State):

Root element style is needed for resolving other elements. Add a way to provide it without looking
into active document style.

  • dom/Document.cpp:

(WebCore::Document::recalcStyle):

Resolve the document style and commit it immediately (for now).

(WebCore::Document::styleForElementIgnoringPendingStylesheets):

  • dom/Document.h:

(WebCore::Document::setNeedsNotifyRemoveAllPendingStylesheet):
(WebCore::Document::inStyleRecalc):
(WebCore::Document::inRenderTreeUpdate):

  • dom/Element.cpp:

(WebCore::Element::setChildIndex):

Setting the unique bit is now done by style relations update code.

  • dom/Node.cpp:

(WebCore::Node::setNeedsStyleRecalc):

Prevent spurious style invalidation during render tree updating.

  • rendering/RenderBox.cpp:

(WebCore::RenderBox::styleDidChange):

Capturing body element color for color:-webkit-text is now done by TreeResolver.

  • rendering/RenderElement.h:

(WebCore::RenderElement::setAnimatableStyle): Deleted.

No longer used.

  • style/RenderTreePosition.cpp:

(WebCore::RenderTreePosition::nextSiblingRenderer):

Skip over non-rendered slot elements.

  • style/RenderTreeUpdater.cpp: Added.

(WebCore::RenderTreeUpdater::Parent::Parent):
(WebCore::RenderTreeUpdater::RenderTreeUpdater):
(WebCore::hasDisplayContents):
(WebCore::findRenderingRoot):
(WebCore::RenderTreeUpdater::commit):

Call updateRenderTree for each change root.

(WebCore::shouldCreateRenderer):
(WebCore::RenderTreeUpdater::updateRenderTree):

Iteratively traverse the composed tree starting for a change root.
Apply the changes calling updateElementRenderer and updateTextRenderer as needed.
Enter subtrees that haves changes to apply.

(WebCore::RenderTreeUpdater::renderTreePosition):

We may not create renderers for all elements (<slot> or more generally display:contents) that
have rendered descendants. Search the parent stack to find the valid position.

(WebCore::RenderTreeUpdater::pushParent):
(WebCore::RenderTreeUpdater::popParent):
(WebCore::RenderTreeUpdater::popParentsToDepth):

Maintain parent stack.

(WebCore::pseudoStyleCacheIsInvalid):
(WebCore::RenderTreeUpdater::updateElementRenderer):

Create, delete or update the renderer.

(WebCore::moveToFlowThreadIfNeeded):
(WebCore::RenderTreeUpdater::createRenderer):
(WebCore::textRendererIsNeeded):
(WebCore::createTextRenderer):
(WebCore::RenderTreeUpdater::updateTextRenderer):
(WebCore::RenderTreeUpdater::invalidateWhitespaceOnlyTextSiblingsAfterAttachIfNeeded):

This is moved from TreeResolver.

(WebCore::needsPseudoElement):
(WebCore::RenderTreeUpdater::updateBeforeOrAfterPseudoElement):

Pseudo elements are handled entirely during render tree construction. Compute their style and
create or delete them as needed.

  • style/RenderTreeUpdater.h: Added.

(WebCore::RenderTreeUpdater::parent):

  • style/StyleRelations.cpp:

(WebCore::Style::commitRelationsToRenderStyle):
(WebCore::Style::commitRelations):

Commit to Style::Update instead of the document if needed.

(WebCore::Style::commitRelationsToDocument): Deleted.

  • style/StyleRelations.h:
  • style/StyleSharingResolver.cpp:

(WebCore::Style::elementHasDirectionAuto):
(WebCore::Style::SharingResolver::resolve):

Fetch the shareable style from Style::Update instead of the active document style.

(WebCore::Style::SharingResolver::findSibling):
(WebCore::Style::SharingResolver::canShareStyleWithElement):

  • style/StyleSharingResolver.h:
  • style/StyleTreeResolver.cpp:

(WebCore::Style::TreeResolver::Parent::Parent):

No need for render tree position anymore.

(WebCore::Style::TreeResolver::popScope):
(WebCore::Style::TreeResolver::styleForElement):
(WebCore::Style::invalidateWhitespaceOnlyTextSiblingsAfterAttachIfNeeded):
(WebCore::Style::createTextRendererIfNeeded):
(WebCore::Style::updateTextRendererAfterContentChange):
(WebCore::Style::resetStyleForNonRenderedDescendants):
(WebCore::Style::detachChildren):
(WebCore::Style::detachSlotAssignees):
(WebCore::Style::detachRenderTree):
(WebCore::Style::TreeResolver::resolveElement):

Just resolve the style and return it, no more applying or entering render tree construction code paths.

(WebCore::Style::resolveTextNode):
(WebCore::Style::elementImplicitVisibility):
(WebCore::Style::TreeResolver::pushParent):
(WebCore::Style::TreeResolver::popParent):
(WebCore::Style::TreeResolver::popParentsToDepth):
(WebCore::Style::shouldResolvePseudoElement):
(WebCore::Style::TreeResolver::resolveComposedTree):

Add style changes to Style::Update.

(WebCore::Style::TreeResolver::resolve):

Return Style::Update object if non-empty.

(WebCore::Style::postResolutionCallbackQueue):
(WebCore::Style::shouldCreateRenderer): Deleted.
(WebCore::Style::moveToFlowThreadIfNeeded): Deleted.
(WebCore::Style::TreeResolver::createRenderer): Deleted.
(WebCore::Style::TreeResolver::createRenderTreeForChildren): Deleted.
(WebCore::Style::TreeResolver::createRenderTreeForShadowRoot): Deleted.
(WebCore::Style::beforeOrAfterPseudoElement): Deleted.
(WebCore::Style::setBeforeOrAfterPseudoElement): Deleted.
(WebCore::Style::clearBeforeOrAfterPseudoElement): Deleted.
(WebCore::Style::needsPseudoElement): Deleted.
(WebCore::Style::TreeResolver::createRenderTreeForBeforeOrAfterPseudoElement): Deleted.
(WebCore::Style::TreeResolver::createRenderTreeForSlotAssignees): Deleted.
(WebCore::Style::TreeResolver::createRenderTreeRecursively): Deleted.
(WebCore::Style::pseudoStyleCacheIsInvalid): Deleted.
(WebCore::Style::TreeResolver::resolveBeforeOrAfterPseudoElement): Deleted.

Remove the recursive render tree building code path.

  • style/StyleTreeResolver.h:

(WebCore::Style::TreeResolver::scope):

  • style/StyleUpdate.cpp: Added.

(WebCore::Style::Update::Update):
(WebCore::Style::Update::elementUpdate):
(WebCore::Style::Update::textUpdate):
(WebCore::Style::Update::elementStyle):
(WebCore::Style::Update::addElement):
(WebCore::Style::Update::addText):
(WebCore::Style::Update::addPossibleRoot):

  • style/StyleUpdate.h: Added.

(WebCore::Style::Update::roots):
(WebCore::Style::Update::document):

  • svg/SVGElement.h:

(WebCore::SVGElement::updateRelativeLengthsInformation):

  • svg/SVGUseElement.cpp:

(WebCore::SVGUseElement::svgAttributeChanged):
(WebCore::SVGUseElement::willRecalcStyle):
(WebCore::SVGUseElement::willAttachRenderers): Deleted.

Switvh willAttachRenderers to willRecalcStyle as the former is now called too late.

  • svg/SVGUseElement.h:

LayoutTests:

Skip mathml/presentation/menclose-notation-attribute-change-value.html. It will be fixed by upcoming MathML refactoring.

  • css3/blending/repaint/blend-mode-isolate-stacking-context-expected.txt:
  • css3/viewport-percentage-lengths/viewport-percentage-lengths-resize-expected.txt:

This is a progression.

  • editing/mac/spelling/autocorrection-contraction-expected.txt:
  • editing/mac/spelling/autocorrection-removing-underline-after-paste-expected.txt:
  • editing/mac/spelling/autocorrection-removing-underline-expected.txt:
  • editing/mac/spelling/autocorrection-simple-expected.txt:
  • editing/style/remove-underline-from-stylesheet-expected.txt:
  • editing/style/typing-style-003-expected.txt:

Non-rendered whitespace related changes.

  • platform/ios-simulator/TestExpectations:

Skip fast/regions/position-writing-modes-in-variable-width-regions.html on iOS. Similar tests are mostly already skipped.

  • platform/ios-simulator/editing/style/typing-style-003-expected.txt: Added.
  • platform/mac-wk2/editing/mac/spelling/autocorrection-contraction-expected.txt:
  • platform/mac/editing/inserting/editable-html-element-expected.txt:
  • platform/mac/editing/inserting/editing-empty-divs-expected.txt:
  • platform/mac/editing/inserting/insert-at-end-02-expected.txt:
  • platform/mac/editing/pasteboard/4989774-expected.txt:
  • platform/mac/editing/selection/4983858-expected.txt:

Non-rendered whitespace related changes.

1:06 AM WebKitNightlyElCapWorkaround edited by Manuel Rego Casasnovas
run-safari was not working after the workaround, explained how to run … (diff)
12:23 AM WebKitGTK/2.12.x edited by berto@igalia.com
(diff)
Note: See TracTimeline for information about the timeline view.