GitLab

Spec: http://www.whatwg.org/specs/web-apps/current-work/multipage/timers.html#dom-navigator-languages

BUG=365123

Review URL: https://codereview.chromium.org/288393002

git-svn-id: svn://svn.chromium.org/blink/trunk@175161 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Create the compositor in WebPopupMenuImpl::initialize because it's always going to be used.

BUG=378029

Review URL: https://codereview.chromium.org/304273002

git-svn-id: svn://svn.chromium.org/blink/trunk@175160 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Avoid redundant initialization of GraphicsContextState. (https://codereview.chromium.org/303613002/)

Reason for revert:
This change is the cause of a
leak in PhishingDOMFeatureExtractorTest.ScriptAndImageFeatures
under LSAN. See
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%281%29/builds/2642/steps/browser_tests/logs/ScriptAndImageFeatures

Original issue's description:
> Avoid redundant initialization of GraphicsContextState.
> 
> When creating a new GraphicsContextState the constructor initializes the state
> with various default values. In GraphicsContext::realizePaintState we immediately
> overwrite these values after creation. This patch adds a copy constructor to
> GraphicsContextState to avoid that.
> 
> Even though this seems like a clear win, the benefits are not visible when
> profiling. Looks like the overhead of the redundant initialization is
> not very large. Still seems worth fixing though.
> 
> BUG=377687
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=174985

TBR=schenney@chromium.org,fmalita@chromium.org,dominikg@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=377687

Review URL: https://codereview.chromium.org/303403006

git-svn-id: svn://svn.chromium.org/blink/trunk@175159 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Revert of Prevent <canvas> destructor from causing nested garbage collection (https://codereview.chromium.org/307013006/)

Reason for revert:
This isn't the right CL, was off-by-one.

Original issue's description:
> Revert of Prevent <canvas> destructor from causing nested garbage collection (https://codereview.chromium.org/303803002/)
> 
> Reason for revert:
> This change is the cause of a
> leak in PhishingDOMFeatureExtractorTest.ScriptAndImageFeatures
> under LSAN. See
> http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%281%29/builds/2642/steps/browser_tests/logs/ScriptAndImageFeatures
> 
> Original issue's description:
> > Prevent <canvas> destructor from causing nested garbage collection
> > 
> > In the destruction of HTMLCanvasElement, we call
> > HTMLCanvasElement::existingDrawingContext(), whose behavior was changed
> > incorrectly in r162330 and was triggering the allocation of an ImageBuffer
> > during tear-down.  This is not only wasteful, but it can trigger a
> > garbage collection because the creation of an ImageBuffer will increase
> > the amount of externallay allocated memory that is declared to V8.
> > Triggering a GC during the finalization of a garbage collected object
> > is bad.
> > 
> > BUG=378289
> > 
> > Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=174984
> 
> TBR=senorblanco@chromium.org,junov@chromium.org
> NOTREECHECKS=true
> NOTRY=true
> BUG=378289
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175157

TBR=senorblanco@chromium.org,junov@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=378289

Review URL: https://codereview.chromium.org/301243009

git-svn-id: svn://svn.chromium.org/blink/trunk@175158 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Prevent <canvas> destructor from causing nested garbage collection (https://codereview.chromium.org/303803002/)

Reason for revert:
This change is the cause of a
leak in PhishingDOMFeatureExtractorTest.ScriptAndImageFeatures
under LSAN. See
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%281%29/builds/2642/steps/browser_tests/logs/ScriptAndImageFeatures

Original issue's description:
> Prevent <canvas> destructor from causing nested garbage collection
>
> In the destruction of HTMLCanvasElement, we call
> HTMLCanvasElement::existingDrawingContext(), whose behavior was changed
> incorrectly in r162330 and was triggering the allocation of an ImageBuffer
> during tear-down.  This is not only wasteful, but it can trigger a
> garbage collection because the creation of an ImageBuffer will increase
> the amount of externallay allocated memory that is declared to V8.
> Triggering a GC during the finalization of a garbage collected object
> is bad.
>
> BUG=378289
>
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=174984

TBR=senorblanco@chromium.org,junov@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=378289

Review URL: https://codereview.chromium.org/307013006

git-svn-id: svn://svn.chromium.org/blink/trunk@175157 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=174930

BUG=373240
TBR=dominik.rottsches@intel.com

Review URL: https://codereview.chromium.org/308643005

git-svn-id: svn://svn.chromium.org/blink/trunk@175155 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The browser side code of the new WebSocket stack needs to
know the ID of the render frame where the corresponding
WebSocketBridge is created in order to handle certificate errors.

This CL just ports the code used by the old WebSocket
implementation to new one.

This CL also contains clean up on the old stack.

Chromium side change: https://codereview.chromium.org/308513004/

BUG=364361

Review URL: https://codereview.chromium.org/304113002

git-svn-id: svn://svn.chromium.org/blink/trunk@175154 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch simplifies parsing CSS animations and transitions slightly by
making us only return CSSValueLists from the parser, including when we
only have a single value in the list. This was probably done differently
in the past (with single values not being wrapped in a CSSValueList) as
premature optimisation, which adds needless complexity to the code. Note
this changes the foo.style.getPropertyCSSValue output slightly.

I've also made BisonCSSParser::parseAnimationTimingFunctionValue return
a single (non-valuelist-wrapped) CSSValue which isn't initial or inherit
to better match the intention of the callers.

R=alancutter,dstockwell

Review URL: https://codereview.chromium.org/293113007

git-svn-id: svn://svn.chromium.org/blink/trunk@175153 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Follow-up to r175144

TBR=jchaffraix@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/306053006

git-svn-id: svn://svn.chromium.org/blink/trunk@175152 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Patch 1 of 4 to specify the deviceSource of a desired fling animation
request via enum throughout Blink and Chromium.

BUG=343327

Review URL: https://codereview.chromium.org/307733002

git-svn-id: svn://svn.chromium.org/blink/trunk@175151 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch updates DUMP_NODE_STATISTICS as it has gotten stale.
Attributes are no longer lazily created so those counts have been
removed. Clang does not allow exit time destructors so the static
liveNodeSet() function has been added. Lastly, fix a few type errors.

Review URL: https://codereview.chromium.org/310433002

git-svn-id: svn://svn.chromium.org/blink/trunk@175150 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=375171

Review URL: https://codereview.chromium.org/289323004

git-svn-id: svn://svn.chromium.org/blink/trunk@175149 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=340522,375671

Review URL: https://codereview.chromium.org/303133005

git-svn-id: svn://svn.chromium.org/blink/trunk@175148 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Compute and validate the shape parameters before deciding which code-path
should be used (path vs. primitive). Don't use the fallback when the
bounding box is empty.

Based on WebKit patch:
  https://bugs.webkit.org/attachment.cgi?id=232191&action=prettypatch
by:
  Nikos Andronikos <nikos.andronikos-webkit@cisra.canon.com.au>

BUG=377610

Review URL: https://codereview.chromium.org/307643003

git-svn-id: svn://svn.chromium.org/blink/trunk@175147 bbb929c8-8fbe-4397-9dbb-9b2b20218538

ScriptState::current should be called in the binding layer.

BUG=357144
NOTRY=true

Review URL: https://codereview.chromium.org/303103002

git-svn-id: svn://svn.chromium.org/blink/trunk@175146 bbb929c8-8fbe-4397-9dbb-9b2b20218538

RenderGeometryMap::dumpSteps doesn't have any side effects other then writing
information to stderr. Marking it const so we can use it from other const
methods.

NOTRY=true

Review URL: https://codereview.chromium.org/301243006

git-svn-id: svn://svn.chromium.org/blink/trunk@175145 bbb929c8-8fbe-4397-9dbb-9b2b20218538

These stopped crashing after r273650.

TBR=jbauman@chromium.org
NOTRY=true
BUG=None

Review URL: https://codereview.chromium.org/310463004

git-svn-id: svn://svn.chromium.org/blink/trunk@175144 bbb929c8-8fbe-4397-9dbb-9b2b20218538

AnimationPlayer should stop indicating activity when active dom object is
stopped.

BUG=379032

Review URL: https://codereview.chromium.org/301363002

git-svn-id: svn://svn.chromium.org/blink/trunk@175143 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Windows XP specific rebaselines for r175038 as the XP > Win7 fallback is
no longer valid for text tests.

TBR=dglazkov@chromium.org

Review URL: https://codereview.chromium.org/306953007

git-svn-id: svn://svn.chromium.org/blink/trunk@175142 bbb929c8-8fbe-4397-9dbb-9b2b20218538

For example: console.log(new Date)

BUG=327109
R=pfeldman

Review URL: https://codereview.chromium.org/297163014

git-svn-id: svn://svn.chromium.org/blink/trunk@175141 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Windows XP specific rebaselines for r175038 as the XP > Win7 fallback is
no longer valid for text tests.

TBR=dglazkov@chromium.org

Review URL: https://codereview.chromium.org/304233009

git-svn-id: svn://svn.chromium.org/blink/trunk@175140 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=378681

Review URL: https://codereview.chromium.org/304113003

git-svn-id: svn://svn.chromium.org/blink/trunk@175138 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=267592
R=pfeldman, yurys

Review URL: https://codereview.chromium.org/304563002

git-svn-id: svn://svn.chromium.org/blink/trunk@175136 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Spec: http://www.whatwg.org/specs/web-apps/current-work/multipage/section-index.html#event-languagechange

BUG=365123

Review URL: https://codereview.chromium.org/284793003

git-svn-id: svn://svn.chromium.org/blink/trunk@175135 bbb929c8-8fbe-4397-9dbb-9b2b20218538

To avoid the need to call ReThrow() on the v8::TryCatch that may or
may not be declared, and the risk of forgetting to do so in some case,
use an auto-rethrowing v8::TryCatch variant instead.

BUG=362388

Review URL: https://codereview.chromium.org/304223007

git-svn-id: svn://svn.chromium.org/blink/trunk@175134 bbb929c8-8fbe-4397-9dbb-9b2b20218538

According to Geolocation API specification getCurrentPosition and
watchPosition does not raise exception, and current RaisesException
for them in the Geolocation.idl are meaningless Because it never
affect to generating bindings codes at all. Therefore we do not
need to keep them in that idl.

BUG=368184

Review URL: https://codereview.chromium.org/297353002

git-svn-id: svn://svn.chromium.org/blink/trunk@175133 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=361045
TBR=alph@chromium.org

Review URL: https://codereview.chromium.org/307093002

git-svn-id: svn://svn.chromium.org/blink/trunk@175132 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Geolocation spec defines default values for both timeout and 
maximumAge but current implementation is working with
hasTimeout and hasMaximumAge functions for emulating that.

Setting initial values of timeout and maximumAge is more proper than using these "has" functions.
In addition, hasZeroTimeout can be removed because it is
enough to check timeout is zero or not and clearMaximumAge
can also be removed, as it is enough to set 0 to maximumAge.

BUG=368184

Review URL: https://codereview.chromium.org/297143003

git-svn-id: svn://svn.chromium.org/blink/trunk@175131 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The lazy initialization required the way to check wheather audio thread is finished or not to avoid lazy initialization by creation of other node after context uninitialzation.

A m_audioThreadFinished is not needed anymore since initialization() will be called only once in construct.

BUG=none

Review URL: https://codereview.chromium.org/304103002

git-svn-id: svn://svn.chromium.org/blink/trunk@175130 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The SVGResourceCycleSolver checks for cycles from within the resources
a RenderObject references to the resources it references, or any resources
in the ancestor-chain of the starting RenderObject (including itself if
it is a resource container).
This suffers from a few issues. One is that a resource that is referenced
both by the starting RenderObject and another resource referenced from the
starting RenderObject will be flagged as a cycle because all the resources
are added to the resources set.
Another issue is that longer cycles may not be detected because only the
resources referenced by the starting RenderObject is checked, and
depending on how the render tree is built, some cycles could be missed.
Additionally, a resource defined within another resource can cause a
"false cycle" to be flagged when the inner resource references a resource
in the resource set - eventhough the inner resource itself is not reachable
from the starting RenderObject.

To fix the above, make the SVGResourcesCycleSolver traverse the entire
resource "graph", and verify that it is acyclic. The cycle-checking is
achieved by keeping track of the current "path" through the graph, and
checking any discovered references against that set. When traversing
a subtree looking for references, any resources discovered will be skipped
to avoid inject false dependencies that way.
Also keep a simple "cache" of sub-graphs that has been checked already
(and verified cycle-free) to avoid having to revisit them.

BUG=351713

Review URL: https://codereview.chromium.org/303693009

git-svn-id: svn://svn.chromium.org/blink/trunk@175129 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Required to keep chromium working until it handles the callback method.

BUG=162827

Review URL: https://codereview.chromium.org/302703003

git-svn-id: svn://svn.chromium.org/blink/trunk@175128 bbb929c8-8fbe-4397-9dbb-9b2b20218538

To complete the move to transition types for Element in html/ and dom/,
convert PendingScript and the script runner objects to trace their
embedded script element references. The script runner objects
(ScriptRunner and HTMLScriptRunner) are now garbage collected objects,
with PendingScript being kept in-line or as a part objects of those
runner objects.

To enable detaching of HTMLScriptRunner without having to rely on its
host (HTMLDocumentParser), the HTMLScriptRunner is instead made the
client of the resources it loads. This simplifies dependencies, but
requiring some follow-on changes to the HTMLScriptRunnerHost interface.

Besides script element references, convert a few other remaining Element
{Pass}RefPtr uses to transition types.

R=ager@chromium.org,haraken@chromium.org,eseidel@chromium.org
BUG=357163

Review URL: https://codereview.chromium.org/298863010

git-svn-id: svn://svn.chromium.org/blink/trunk@175127 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=357163

Review URL: https://codereview.chromium.org/308073005

git-svn-id: svn://svn.chromium.org/blink/trunk@175126 bbb929c8-8fbe-4397-9dbb-9b2b20218538

|invokeMethodOnAllTargetProperties| is now only used from |endAnimation|.
The code is more readable when inlined.

BUG=378670

Review URL: https://codereview.chromium.org/305033004

git-svn-id: svn://svn.chromium.org/blink/trunk@175125 bbb929c8-8fbe-4397-9dbb-9b2b20218538

same while shaping Complex text. More than one langauge can point to
single Script and single font can have support for different languages.

BUG=372502

Review URL: https://codereview.chromium.org/276573010

git-svn-id: svn://svn.chromium.org/blink/trunk@175124 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Webkit has css properties through which custom scroll bars can be added.
In case the element's visibility property is hidden and custom scrollbar 
properties are applied the scrollbar width is not considered when querying
the element.clientWidth. In case of non-custom scrollbars whether the
element's visibility property is hidden or not correct scrollbar width is
considered.

When a custom scrollbar is created there is a check whether the renderer to
which scrollbar is added is visible or not. This check is not required since
for non-custom scrollbars the same check is not present. Also whether the
element's visibility property is hidden or not a placeholder is set for the
element.

Both behavior i.e custom and non-custom scrollbars should be same.

I already landed this patch on Webkit. Just modified the layout test file.
http://trac.webkit.org/changeset/155323

R=esprehn@chromium.org,eseidel@chromium.org
BUG=343521

Review URL: https://codereview.chromium.org/163383002

git-svn-id: svn://svn.chromium.org/blink/trunk@175123 bbb929c8-8fbe-4397-9dbb-9b2b20218538

For open-ended intervals such as [3) - i.e. begin=3s and end/duration/etc
unspecified - every call to resolveFirstInterval() would think that a new
interval had been created, and notify/require a reschedule of the
animations.
Add a new type SMILInterval and a suitable operator for that, and then
use it to compare the raw values (i.e. if both endpoints are exacly the same.)

BUG=377329

Review URL: https://codereview.chromium.org/298043005

git-svn-id: svn://svn.chromium.org/blink/trunk@175122 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Long ago this was added in hopes of generating a FlakyTests file.
This patch finally finishes that work and adds the FlakyTests
file as well as renames the command to update-flaky-tests
and writes to the file instead of stdout.

TestExpectationLine changed syntax a little since
flaky-tests was written.  I've also added support
for specifying path and specifiers values so that
the resulting test output looks right.

I explored having it update TestExpectations directly
(which we're definitely very close to being able to do)
but the consensus was FlakyTests as a separate file would
work better, so lets start with that.

BUG=242366

Review URL: https://codereview.chromium.org/301853003

git-svn-id: svn://svn.chromium.org/blink/trunk@175121 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=369451

Review URL: https://codereview.chromium.org/303363002

git-svn-id: svn://svn.chromium.org/blink/trunk@175120 bbb929c8-8fbe-4397-9dbb-9b2b20218538

nodeUnderMouse function is unimplemented in the complete code, so removing it.

Review URL: https://codereview.chromium.org/297293007

git-svn-id: svn://svn.chromium.org/blink/trunk@175119 bbb929c8-8fbe-4397-9dbb-9b2b20218538