GitLab

Oilpan: Prepare to move FormController and DocumentState to Oilpan heap, and remove RefPtrs to HTMLFormControlElement and HTMLFormControlElementWithState.

Note: ValidationMessage has a raw pointer member to HTMLFormControlElement. This
CL doesn't make it traceable because a CL to remove ValidationMessage class is
in-progress. A ValidationMessage object is owned by the HTMLFormControlElement
object and the raw pointer is safe.

BUG=357163

Review URL: https://codereview.chromium.org/277753003

git-svn-id: svn://svn.chromium.org/blink/trunk@173710 bbb929c8-8fbe-4397-9dbb-9b2b20218538

If you squash on to your compositing container, you'll draw in the wrong order
because you'll draw after other RenderLayers that come after you in paint order
into your compositing container.

R=chrishtr@chromium.org
BUG=370862

Review URL: https://codereview.chromium.org/277763002

git-svn-id: svn://svn.chromium.org/blink/trunk@173709 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173673

BUG=361729
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/278763002

git-svn-id: svn://svn.chromium.org/blink/trunk@173708 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=enne@chromium.org
BUG=371651

Review URL: https://codereview.chromium.org/272893002

git-svn-id: svn://svn.chromium.org/blink/trunk@173707 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This code is dead, we always pass a null parentLayer which causes us to
early return. I think we were supposed to call
childLayer->compositedLayerMapping()->childForSuperlayers()->removeFromParent()
in the original design, but since r44876 [1] we always early return when the
parentLayer is null, and that patch was back in 2009 so clearly the code
is not needed.

[1] http://src.chromium.org/viewvc/blink?revision=44876&view=revision

R=abarth@chromium.org

Review URL: https://codereview.chromium.org/279643002

git-svn-id: svn://svn.chromium.org/blink/trunk@173705 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173673

BUG=361729
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/278743002

git-svn-id: svn://svn.chromium.org/blink/trunk@173704 bbb929c8-8fbe-4397-9dbb-9b2b20218538

SelectorChecker::parentElement should walk up to ShadowHost only if an element is in a shadow tree hosted by the ShadowHost and scope is the ShadowHost.

:host can only match shadow host. :host depends on context.scope with ScopeIsShadowHost behavior.
So only if context.scope is shadow host and context.element's shadow host is context.scope, we need to check the shadow host.

BUG=370303
TEST=fast/dom/shadow/style-with-cat.html, fast/dom/shadow/style-with-hat.html

Review URL: https://codereview.chromium.org/273523008

git-svn-id: svn://svn.chromium.org/blink/trunk@173703 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173673

BUG=361729
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/278733002

git-svn-id: svn://svn.chromium.org/blink/trunk@173702 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert "Build fix after r173654"

Reverts r173654 and r173655. Broke inspector/tracing-session-id.html.

TBR=yurys@chromium.org

Review URL: https://codereview.chromium.org/276793002

git-svn-id: svn://svn.chromium.org/blink/trunk@173701 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173673

BUG=361729
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/274983002

git-svn-id: svn://svn.chromium.org/blink/trunk@173700 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The original patch was reverted in r173234, because the chagne exposed
a hidden bug of SVG fonts (crbug.com/369633). That bug is being addressed
in https://codereview.chromium.org/271633002/ so I'm relanding this patch.

BUG=369633

> Make CSSFontFace::willUseFontData() load fonts with unicode-range
>
> Before this patch CSSFontFace::willUseFontData() loads font faces that
> have no unicode-range. Since font faces with no unicode-range tends to
> be used as fallback font of segmented font family, this behavior leads
> to unnecessary font downloads.
>
> This patch makes willUseFontData() loads the first unloaded font face
> whose unicode-range intersects with given text. That check does not
> need to be 100% precise (false negative is ok), so it only checks the
> first character of the text, for speed.
>
> TEST=fast/css/font-face-unicode-range-overlap-load.html
> BUG=247920, 246492
>
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=172943

Review URL: https://codereview.chromium.org/270813003

git-svn-id: svn://svn.chromium.org/blink/trunk@173699 bbb929c8-8fbe-4397-9dbb-9b2b20218538

-Add flaky tests from https://codereview.chromium.org/262963003.
A revert would be tricky.
-Rebaseline fast/transforms/shadows.html as a followup to
https://codereview.chromium.org/279513002.
-Add a flaky imports test.

TBR=loislo@chromium.org,enne@chromium.org,morrita@chromium.org

Review URL: https://codereview.chromium.org/279663002

git-svn-id: svn://svn.chromium.org/blink/trunk@173698 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173666

BUG=369609
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/278623003

git-svn-id: svn://svn.chromium.org/blink/trunk@173697 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We've fixed all the know bugs. Hopefully this time it will stick. If you find
that this CL caused a regression, please mark the bug as blocking
http://crbug.com/365701

BUG=365701

Review URL: https://codereview.chromium.org/267343005

git-svn-id: svn://svn.chromium.org/blink/trunk@173696 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173666

BUG=369609
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/272063002

git-svn-id: svn://svn.chromium.org/blink/trunk@173695 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173666

BUG=369609
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/276803002

git-svn-id: svn://svn.chromium.org/blink/trunk@173694 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This test was added in http://src.chromium.org/viewvc/blink?revision=173653&view=revision.
Adam tells me it's fine to rebaseline.

TBR=abarth@chromium.org

Review URL: https://codereview.chromium.org/274963002

git-svn-id: svn://svn.chromium.org/blink/trunk@173693 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173666

BUG=369609
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/275663002

git-svn-id: svn://svn.chromium.org/blink/trunk@173692 bbb929c8-8fbe-4397-9dbb-9b2b20218538

In particular, allow a layer L to squash if their clipping
ancestor differs from the squashing owner, but the clipping
ancestor for L is a descendant of a layer that does squash
into the squashing owner.

BUG=368410

Review URL: https://codereview.chromium.org/264013002

git-svn-id: svn://svn.chromium.org/blink/trunk@173690 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Also moved the testing for invalid values from
AudioContext::createBuffer to AudioBuffer::create, which knows what
the valid ranges are instead of having the tests in both
AudioBuffer::create and AudioContext::createBuffer.

Test case added for createBuffer(0,...).

BUG=362657

Review URL: https://codereview.chromium.org/235653002

git-svn-id: svn://svn.chromium.org/blink/trunk@173689 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173661

BUG=371111,361729
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/273923002

git-svn-id: svn://svn.chromium.org/blink/trunk@173688 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173653

BUG=370667
TBR=abarth@chromium.org

Review URL: https://codereview.chromium.org/271993002

git-svn-id: svn://svn.chromium.org/blink/trunk@173687 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The frame / iframe src attribute value was not correctly updated upon
serializing because HTMLFrameElementBase::hasLegalLinkAttribute() was returning
true if the attribute name was 'href' instead of 'src'. The attribute was thus
not detected as a link attribute and its value was not correctly updated with
the path to the serialized frame.

R=tkent@chromium.org, adamk@chromium.org
BUG=370772
TEST=WebPageNewSerializeTest.SubFrameSerialization

Review URL: https://codereview.chromium.org/275433011

git-svn-id: svn://svn.chromium.org/blink/trunk@173686 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert "Web Animations: Fix error in clock update logic and re-enable timeline-time test"

This reverts r173583 and r173609. r173583 causes 
ImageResourceTest.MultipartImage to fail on Android debug, e.g.,
http://build.chromium.org/p/tryserver.chromium/builders/android_dbg_triggered_tests/builds/152167/steps/webkit_unit_tests/logs/stdio

TBR=dstockwell@chromium.org

Review URL: https://codereview.chromium.org/272543005

git-svn-id: svn://svn.chromium.org/blink/trunk@173685 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173660

BUG=368078
TBR=reveman@chromium.org

Review URL: https://codereview.chromium.org/277593006

git-svn-id: svn://svn.chromium.org/blink/trunk@173683 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173614

TBR=esprehn@chromium.org

Review URL: https://codereview.chromium.org/270783005

git-svn-id: svn://svn.chromium.org/blink/trunk@173682 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173641

BUG=361729
TBR=loislo@chromium.org

Review URL: https://codereview.chromium.org/272723010

git-svn-id: svn://svn.chromium.org/blink/trunk@173681 bbb929c8-8fbe-4397-9dbb-9b2b20218538

r173248 added it but there is a performance concern.
This change repalces it with a modest alternative that does same as what didRemoveAllPendingStylesheet() has been doing.

This change also removed 500ms sleep from a test that was added at r173248.

TEST=none
R=ojan@chromium.org,esprehn@chromium.org
BUG=367386

Review URL: https://codereview.chromium.org/279463002

git-svn-id: svn://svn.chromium.org/blink/trunk@173680 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173633

BUG=361729
TBR=loislo@chromium.org

Review URL: https://codereview.chromium.org/272793003

git-svn-id: svn://svn.chromium.org/blink/trunk@173679 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173616

BUG=361729
TBR=loislo@chromium.org

Review URL: https://codereview.chromium.org/275643002

git-svn-id: svn://svn.chromium.org/blink/trunk@173678 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL adds document lifecycle assertions in for RenderObject::createObject,
RnederObject::addChild and RenderObject::removeChild.

In all cases we assert that we are currently in the InRecalcStyle state.

BUG=368052

Review URL: https://codereview.chromium.org/267053002

git-svn-id: svn://svn.chromium.org/blink/trunk@173677 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173616

BUG=361729
TBR=loislo@chromium.org

Review URL: https://codereview.chromium.org/267303007

git-svn-id: svn://svn.chromium.org/blink/trunk@173674 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=361729

Review URL: https://codereview.chromium.org/275603002

git-svn-id: svn://svn.chromium.org/blink/trunk@173673 bbb929c8-8fbe-4397-9dbb-9b2b20218538

refreshPageScaleFactor only gets called if the page scale constraints are dirty.
This happens when the page changes width but not when the height changes. This
cuased a bug where changing the window height only wouldn't resize the main
frame/outer viewport.

BUG=370218

Review URL: https://codereview.chromium.org/271593005

git-svn-id: svn://svn.chromium.org/blink/trunk@173672 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=173616

BUG=361729
TBR=loislo@chromium.org

Review URL: https://codereview.chromium.org/271873006

git-svn-id: svn://svn.chromium.org/blink/trunk@173671 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=

Review URL: https://codereview.chromium.org/270283006

git-svn-id: svn://svn.chromium.org/blink/trunk@173670 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=370490
R=dglazkov@chromium.org

Review URL: https://codereview.chromium.org/278643002

git-svn-id: svn://svn.chromium.org/blink/trunk@173668 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is intended to make the MSE implementation more spec compliant.
More such CLs to follow after this initial round is checked in.

BUG=
TEST=new test pass

Review URL: https://codereview.chromium.org/267333006

git-svn-id: svn://svn.chromium.org/blink/trunk@173667 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Previously all of these tests were converted to not output repaint rects
and just output pixels.  This ended up being a loss of information that
was useful to detecting missing repaints, and so they should be turned
back on where it makes sense.

BUG=369609

Review URL: https://codereview.chromium.org/262963003

git-svn-id: svn://svn.chromium.org/blink/trunk@173666 bbb929c8-8fbe-4397-9dbb-9b2b20218538

In particular, don't do it  when attributes changed if there
is no style resolver or the style change type is subtree or
greater. In both of these cases, it will either have no
additional effect or the code is incorrect.

It could be incorrect if the element was not yet attached.
It could be not needed because if there is no style resolver,
re-making it will recalc the whole document's style. Also,
style invalidation cannot trigger a style that is greater than
subtree.

Also clear style invalidation bits unconditionally in Node::detach.

BUG=366788

Review URL: https://codereview.chromium.org/273783003

git-svn-id: svn://svn.chromium.org/blink/trunk@173665 bbb929c8-8fbe-4397-9dbb-9b2b20218538