GitLab

Add iterator object to iterate efficiently over an Element's attributes instead
of using ElementData::length() / ElementData::attributeItem(index).

This is a bit more efficient as it avoids implicitly calling
ElementData::isUnique() for each loop iteration.

This is based on the following WebKit revision by benjamin@webkit.org:
http://trac.webkit.org/changeset/162394

R=esprehn@chromium.org, adamk@chromium.org

Review URL: https://codereview.chromium.org/298253009

git-svn-id: svn://svn.chromium.org/blink/trunk@175583 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This test was already marked as flaky on release retina bots, but is flake
on other Mac bots as well.

TBR=dcheng@chromium.org
BUG=380167

Review URL: https://codereview.chromium.org/317783011

git-svn-id: svn://svn.chromium.org/blink/trunk@175582 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=175562

BUG=374119
TBR=davve@opera.com

Review URL: https://codereview.chromium.org/316193002

git-svn-id: svn://svn.chromium.org/blink/trunk@175581 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=381035

Review URL: https://codereview.chromium.org/317843002

git-svn-id: svn://svn.chromium.org/blink/trunk@175580 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=vsevik,sergeyv
NOTRY=true

Review URL: https://codereview.chromium.org/315213003

git-svn-id: svn://svn.chromium.org/blink/trunk@175579 bbb929c8-8fbe-4397-9dbb-9b2b20218538

With new touch hit testing in compositor, which looks through all
the layers hitting the point, we don't need touch regions on overlay
layer anymore.

Note, that we still need main-thread scrolling to stay in sync with
page contents.

BUG=304271

Review URL: https://codereview.chromium.org/305383002

git-svn-id: svn://svn.chromium.org/blink/trunk@175578 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We will have two descendants of TimelineModel - one based on trace events and another one based on the timeline agent events. This change extracts the latter one.

I left WebInspector.TimelineModel.RecordType to keep the patch short. Will move it to WebInspector.TimelineModelImpl.RecordType later.

BUG=361045
R=caseq@chromium.org

Review URL: https://codereview.chromium.org/313333002

git-svn-id: svn://svn.chromium.org/blink/trunk@175577 bbb929c8-8fbe-4397-9dbb-9b2b20218538

I'm about to use this useful script in Source/modules, so a common
denominator location seemed like a good idea. As well, I added a
namespace option as the upcoming use in modules/serviceworkers will be
in WebCore not blink.

R=jochen@chromium.org
BUG=none

Review URL: https://codereview.chromium.org/311343002

git-svn-id: svn://svn.chromium.org/blink/trunk@175575 bbb929c8-8fbe-4397-9dbb-9b2b20218538

<integer>, <angle>, <number>, <length> and <percentage>.

BUG=377503,339899

Review URL: https://codereview.chromium.org/302643004

git-svn-id: svn://svn.chromium.org/blink/trunk@175574 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=vsevik

Review URL: https://codereview.chromium.org/315223002

git-svn-id: svn://svn.chromium.org/blink/trunk@175572 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Toolbox page is the same frontend launched with "toolbox=true" parameter.
It has its own app, which creates a view hierarchy and reports to the main app.

Currently, toolbox is only created for responsive design.

BUG=327641

Review URL: https://codereview.chromium.org/298913004

git-svn-id: svn://svn.chromium.org/blink/trunk@175571 bbb929c8-8fbe-4397-9dbb-9b2b20218538

CSS content is considered as active content, and therefore "runs" instead
of "displays". Change this in the test expectations.

TBR=dcheng@chromium.org, bashi@chromium.org
NOTRY=true
BUG=377864

Review URL: https://codereview.chromium.org/317173002

git-svn-id: svn://svn.chromium.org/blink/trunk@175570 bbb929c8-8fbe-4397-9dbb-9b2b20218538

As of Skia 9b14f26d (Alter SkCanvas::drawPicture (devirtualize, take const SkPicture, take pointer) - https://codereview.chromium.org/313613004) SkCanvas::drawPicture has an alternate signature. The old entry point is deprecated.

Note: this patch cannot be landed until the corresponding Skia change is rolled into Chromium & Blink.

Review URL: https://codereview.chromium.org/316863003

git-svn-id: svn://svn.chromium.org/blink/trunk@175569 bbb929c8-8fbe-4397-9dbb-9b2b20218538

scheduleStyleInvalidationForClassChange skipped comparison when oldClasses
was empty, but forgot to return afterwards, causing invalidation sets to be
scheduled twice.

This fix increases runs/s by 10%+ on the added performance test.

R=esprehn@chromium.org,chrishtr@chromium.org
BUG=380874

Review URL: https://codereview.chromium.org/313963006

git-svn-id: svn://svn.chromium.org/blink/trunk@175568 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The DragEvent was ignoring the trasnform for the drag point, causing
the dragimage to be drawn at an incorrect location on the canvas. The
patch corrects this.

BUG=322201

Review URL: https://codereview.chromium.org/286903011

git-svn-id: svn://svn.chromium.org/blink/trunk@175567 bbb929c8-8fbe-4397-9dbb-9b2b20218538

WorkerRunLoop::run shouldn't call SharedTimer::fire when processing debugger tasks.

BUG=377926

Review URL: https://codereview.chromium.org/298383002

git-svn-id: svn://svn.chromium.org/blink/trunk@175566 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Value updated inside change event, get overwritten by setTextAsOfLastFormControlChangeEvent.

New value if similar to previous value was not triggering change event.

R=tkent, keishi
BUG=378871
TEST=OnChange event should be fired, when value is changed inside change
handler and same text is re-entered.

Review URL: https://codereview.chromium.org/313013005

git-svn-id: svn://svn.chromium.org/blink/trunk@175565 bbb929c8-8fbe-4397-9dbb-9b2b20218538

During style invalidation we need to continue to traverse a subtree that is
marked with SubtreeStyleChange to clear any remaining invalidation bits.
But, when the childNeedsInvalidation bit is not set, there should be no
bits left to clear in the subtree. Hence, we can skip traversing that
subtree.

R=esprehn@chromium.org,chrishtr@chromium.org
BUG=381062

Review URL: https://codereview.chromium.org/317133002

git-svn-id: svn://svn.chromium.org/blink/trunk@175564 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch changes DummyPageHolder to use fillWithEmptyClients() instead of making specific empty clients for writing tests depending on client objects not created in current DumpPageHolder class, e.g. spell checker client for editing test.
 
BUG=n/a
TEST=n/a

Review URL: https://codereview.chromium.org/311263007

git-svn-id: svn://svn.chromium.org/blink/trunk@175563 bbb929c8-8fbe-4397-9dbb-9b2b20218538

When crossing the svg -> html boundary through a foreignObject, the
zoom factor is reset to its initial value. This is because SVG has a
special zoom model where the SVG fragment root puts the scale factor
on the graphics context, in contrast to non-SVG content where each
value is scaled individually.

This means the FontBuilder can't just inherit the font from parent
when crossing a foreignObject boundary. It may have to recompute the
font size because the effective zoom, included in the computed font
size, may change.

Example of behavior after this patch:

<html> <!-- computed font size: 16px. -->
  <body style="zoom: 2"> <!-- computed font size: 32px. -->
    <svg> <!-- scale(2) is applied to the graphics context; computed font size: 32px. -->
      <text>Sample svg text.</text> <!-- svg ignores computed font size to be able to support minimal font size. -->
      <foreignObject> <!-- html rendering again but scale(2) still applied to GC. computed font size must be reset to 16px. -->
        Sample html text. <!-- Text drawn with computed font size 16px but scaled to 32px by the GC -->

In this annotated test, 'Sample html text.' and 'Sample svg text.'
should both be displayed with a 32px font size.

Note: One difference is that the 'sample html text' is not sensitive
to the correct minimal font size since that happens before the scale
on the GC is applied. In order to get proper minimal font size working
on foreignObject, we would probably have to put foreignObject in a
layer separated from the graphical context of the SVG.

BUG=374119

Review URL: https://codereview.chromium.org/308123010

git-svn-id: svn://svn.chromium.org/blink/trunk@175562 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Reason for revert:
Broke every incremental compile bot on the waterfall:

InternalRuntimeFlags.h: In member function 'bool WebCore::InternalRuntimeFlags::styleScopedEnabled()':
InternalRuntimeFlags.h:110:40: error: 'styleScopedEnabled' is not a member of 'WebCore::RuntimeEnabledFeatures'
     bool styleScopedEnabled() { return RuntimeEnabledFeatures::styleScopedEnabled(); }

While a clobber build indeed fixes the problem, it also breaks local
builds, which increases the cost of this CL significantly. Let's try
to find a proper solution in the bug.

BUG=381111

Original issue's description:
> Remove scoped styles.
> 
> Removes <style scoped>.
> 
> This was discussed and approved with LGTMs in blink-dev:
> https://groups.google.com/a/chromium.org/d/msg/blink-dev/R1x18ZLS5qQ/Bjuh_cENhlQJ
> 
> Chrome status dashboard entry:
> http://www.chromestatus.com/features/5374137958662144
> 
> Note:
> this is based on esprehn's original cl:
> https://codereview.chromium.org/214693002/
> 
> BUG=379096
> TEST=pass all layout tests
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175555

TBR=esprehn@chromium.org,tasak@google.com,kochi@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=379096

Review URL: https://codereview.chromium.org/315173002

git-svn-id: svn://svn.chromium.org/blink/trunk@175561 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Before this patch, GC_TRACING dumped its result to stdout.
This has caused all LayoutTests to fail.

This patch changes dump output to stderr, so LayoutTest run is not affected.

Unfortunately there is no trybot with ENABLE(GC_TRACING).
NOTRY=true
BUG=381038

Review URL: https://codereview.chromium.org/314943004

git-svn-id: svn://svn.chromium.org/blink/trunk@175560 bbb929c8-8fbe-4397-9dbb-9b2b20218538

According to WebIDL, the Function.length property should be the length of
shortest argument list in the effective overload set for argument count 0.
This means for instance that variadic arguments don't contribute to the
length.

BUG=377718

Review URL: https://codereview.chromium.org/296403007

git-svn-id: svn://svn.chromium.org/blink/trunk@175559 bbb929c8-8fbe-4397-9dbb-9b2b20218538

For RefCountedGarbageCollected objects wrapped in a WebPrivatePtr, the
PtrStorageImpl::assign(PassRefPtr<T>) overload potentially created a
Persistent on assigning its underlying raw pointer, even if a RawPtr<>
was passed along (due to the implicit RawPtr<T> conversion that
PassRefPtr<T> provides.)

This assign() overload serves no real purpose with Oilpan enabled;
provide a more appropriate RawPtr<T> version there.

R=haraken@chromium.org,tkent@chromium.org
BUG=357163

Review URL: https://codereview.chromium.org/313173009

git-svn-id: svn://svn.chromium.org/blink/trunk@175558 bbb929c8-8fbe-4397-9dbb-9b2b20218538

In order to evaluate when we can start dropping support for passive
mixed content, this patch adds counters to ResourceFetcher for the four
resource types we currently don't block: images, media, tracks, and raw.
This patch also drops the 'MixedContentFont' enum, as we now consider
that active content.

Review URL: https://codereview.chromium.org/312253003

git-svn-id: svn://svn.chromium.org/blink/trunk@175557 bbb929c8-8fbe-4397-9dbb-9b2b20218538

If element.setAttributeNode(attr) is performed on an "element" with
an attribute that matches "attr.name", that attribute's value is
updated and attr is attached to "element".

Later accesses of "attr.value" must continue to resolve to "element"'s
underlying attribute, matching (local) names following the case
sensitivity of the element.

A normalized local name was previously assumed for the attribute,
leading to crashing conditions on both setting and getting "attr.value".
Address by having Attr record the local name of the element attribute
it is attached to and use that when looking up the attribute.

R=adamk@chromium.org
BUG=376718

Review URL: https://codereview.chromium.org/316583002

git-svn-id: svn://svn.chromium.org/blink/trunk@175556 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Removes <style scoped>.

This was discussed and approved with LGTMs in blink-dev:
https://groups.google.com/a/chromium.org/d/msg/blink-dev/R1x18ZLS5qQ/Bjuh_cENhlQJ

Chrome status dashboard entry:
http://www.chromestatus.com/features/5374137958662144

Note:
this is based on esprehn's original cl:
https://codereview.chromium.org/214693002/

BUG=379096
TEST=pass all layout tests

Review URL: https://codereview.chromium.org/310443002

git-svn-id: svn://svn.chromium.org/blink/trunk@175555 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This completes the split!

Note that this deletes the top-level:
* bindings/generated.gypi
* bindings/idl.gypi
...and eliminates the variable bindings_output_dir, b/c we do not output
anything to there any longer!

One layering violation is:
constructors for interfaces in core need to know about
global objects defined in modules,
namely ServiceWorker, because some of these are exposed
to ServiceWorker.

This can be fixed by moving
ServiceWorkerGlobalScopeCoreConstructors.idl
to modules, which I'll do in a followup.

R=haraken
BUG=358074

Review URL: https://codereview.chromium.org/315003004

git-svn-id: svn://svn.chromium.org/blink/trunk@175554 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch adds test based on the regression issue with styles sidebar
pane not showing styles for :host selectors. This test was originally
added in r174963, but the commit got reverted.

BUG=363468
TBR=apavlov

Review URL: https://codereview.chromium.org/317703002

git-svn-id: svn://svn.chromium.org/blink/trunk@175553 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Reason for revert:
Speculative revert: looks like this change broke compilation
http://build.chromium.org/p/chromium.webkit/builders/GPU%20Mac%20Builder%20%28dbg%29/builds/14856/steps/compile/logs/stdio

Original issue's description:
> Remove scoped styles.
> 
> Removes <style scoped>.
> 
> This was discussed and approved with LGTMs in blink-dev:
> https://groups.google.com/a/chromium.org/d/msg/blink-dev/R1x18ZLS5qQ/Bjuh_cENhlQJ
> 
> Chrome status dashboard entry:
> http://www.chromestatus.com/features/5374137958662144
> 
> Note:
> this is based on esprehn's original cl:
> https://codereview.chromium.org/214693002/
> 
> BUG=379096
> TEST=pass all layout tests
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175550

TBR=esprehn@chromium.org,tasak@google.com,kochi@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=379096

Review URL: https://codereview.chromium.org/317143002

git-svn-id: svn://svn.chromium.org/blink/trunk@175552 bbb929c8-8fbe-4397-9dbb-9b2b20218538

* Explicitly include v8.h where used.
* Do *not* forward declare v8 symbols in headers, per:
http://www.chromium.org/developers/coding-style/cpp-dos-and-donts#TOC-Exceptions

There are 27 files in core (9 .h + 18 .cpp) that use v8.
This corrects usage.

Review URL: https://codereview.chromium.org/318843003

git-svn-id: svn://svn.chromium.org/blink/trunk@175551 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Removes <style scoped>.

This was discussed and approved with LGTMs in blink-dev:
https://groups.google.com/a/chromium.org/d/msg/blink-dev/R1x18ZLS5qQ/Bjuh_cENhlQJ

Chrome status dashboard entry:
http://www.chromestatus.com/features/5374137958662144

Note:
this is based on esprehn's original cl:
https://codereview.chromium.org/214693002/

BUG=379096
TEST=pass all layout tests

Review URL: https://codereview.chromium.org/310443002

git-svn-id: svn://svn.chromium.org/blink/trunk@175550 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Followup to:
Cleanup v8 usage in core
https://codereview.chromium.org/318843003/

Review URL: https://codereview.chromium.org/319653002

git-svn-id: svn://svn.chromium.org/blink/trunk@175549 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=none

Review URL: https://codereview.chromium.org/315113002

git-svn-id: svn://svn.chromium.org/blink/trunk@175548 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Bug=350571

Review URL: https://codereview.chromium.org/306123003

git-svn-id: svn://svn.chromium.org/blink/trunk@175546 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=369451

Review URL: https://codereview.chromium.org/313263002

git-svn-id: svn://svn.chromium.org/blink/trunk@175545 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Upgrade pywebsocket to r804.

BUG=NONE
R=tyoshino

Review URL: https://codereview.chromium.org/314043002

git-svn-id: svn://svn.chromium.org/blink/trunk@175544 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Review URL: https://codereview.chromium.org/312193002

Review URL: https://codereview.chromium.org/312193002

git-svn-id: svn://svn.chromium.org/blink/trunk@175543 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Now that frames always get a RenderLayer, we don't need to
actually do a layer update anymore.

Not sure how to test this. The original patch that added it
https://codereview.chromium.org/104433003 didn't have any
new tests.

Review URL: https://codereview.chromium.org/318863002

git-svn-id: svn://svn.chromium.org/blink/trunk@175542 bbb929c8-8fbe-4397-9dbb-9b2b20218538

These two public/web/ types hadn't been updated to use transition types
over their underlying Oilpan-converted objects.

R=tkent@chromium.org,haraken@chromium.org
BUG=357163

Review URL: https://codereview.chromium.org/310393004

git-svn-id: svn://svn.chromium.org/blink/trunk@175540 bbb929c8-8fbe-4397-9dbb-9b2b20218538