GitLab

I'm breaking up my local mega patch to land in several chunks. This is
the Blink portion that's needed to build blink_tests successfully with
WebLocalFrame subclassing WebFrame.

Overall, the changes are pretty straightforward. One method changed back
to taking a WebFrame*. It's not yet clear how we're going to coordinate
frame detach for remote frames. Once there's a clear story, I'll update
it appropriately.

The typedef remains for now, as I work on landing the Chrome bits that
depend on this.

BUG=346764
R=eseidel@chromium.org

Review URL: https://codereview.chromium.org/222203007

git-svn-id: svn://svn.chromium.org/blink/trunk@170717 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Since only floats can specify shape-outside, the RasterShapeIntervals
class only needs to track the first and last above threshold pixel column
(x1 and x2 in the implementation) for each row. Removed code for dealing with
multiple "runs" per row as well as shape-inside internals.

This is a merge of http://trac.webkit.org/changeset/166522 by Hans Muller <hmuller@adobe.com>.

BUG=359098

Review URL: https://codereview.chromium.org/212223006

git-svn-id: svn://svn.chromium.org/blink/trunk@170716 bbb929c8-8fbe-4397-9dbb-9b2b20218538

To make sure that the new MediaQueryParser doesn't cause compatibility issues in M35, it was disabled for main thread MQ evaluation in https://codereview.chromium.org/219183002/ .
This CL adds a runtime flag that would enable turning it back on.

BUG=358078

Review URL: https://codereview.chromium.org/221023002

git-svn-id: svn://svn.chromium.org/blink/trunk@170715 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Fix ASSERT failure in RenderGeomtryMap for SVGs (https://codereview.chromium.org/143363004/)

Reason for revert:
This broke rendering in Google Presentations (https://code.google.com/p/chromium/issues/detail?id=359142).

Original issue's description:
> Fix ASSERT failure in RenderGeomtryMap for SVGs
> 
> SVGRenderSupport::mapLocalToContainer() was trying to apply transforms
> in the incorrect order. Specifically, it would attempt to apply its
> localToParentTransform before its localToBorderBoxTransform. This
> was causing an ASSERT to fail when the computed transforms didn't
> match up to those computed by RenderGeometryMap.
> 
> BUG=302719
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=167941

TBR=fs@opera.com,schenney@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=302719

Review URL: https://codereview.chromium.org/223063002

git-svn-id: svn://svn.chromium.org/blink/trunk@170714 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=170660

BUG=320139,355463
TBR=dsinclair@chromium.org

Review URL: https://codereview.chromium.org/222203008

git-svn-id: svn://svn.chromium.org/blink/trunk@170711 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This test was flaky because it would sometimes print the url of the blocked
resource and sometimes not depending on when the navigation happened.

Instead we should wait for onload so that the url is always printed.

TBR=ojan@chromium.org

Review URL: https://codereview.chromium.org/221743006

git-svn-id: svn://svn.chromium.org/blink/trunk@170710 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=170660

BUG=320139
TBR=dsinclair@chromium.org

Review URL: https://codereview.chromium.org/221963006

git-svn-id: svn://svn.chromium.org/blink/trunk@170709 bbb929c8-8fbe-4397-9dbb-9b2b20218538

All concrete WebInputEvent derivatives currently accept an unsigned size
parameter in the constructor, used to zero-initialize the object.  This can
lead to unexpected runtime behavior, e.g., a WebInputEvent::Type value can
implicitly construct a WebInputEvent object.  Prevent this by making the
size-accepting constructor protected for all types, and preventing naked
construction of a base WebInputEvent instance.

Review URL: https://codereview.chromium.org/132183008

git-svn-id: svn://svn.chromium.org/blink/trunk@170708 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=shans
BUG=248938

Review URL: https://codereview.chromium.org/222903003

git-svn-id: svn://svn.chromium.org/blink/trunk@170707 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch replaces isCollapsibleWhitespace() with RenderStyle::isCollapsibleWhiteSpace
in places where renderer is available. The latter provides more accurate decision so we
should always use it when possible. Some test results require rebaseline but none of
them failed the test really and the new behavior is correct.

BUG=357226

Review URL: https://codereview.chromium.org/214523008

git-svn-id: svn://svn.chromium.org/blink/trunk@170706 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This shows up in the Shadow DOM benchmark where childrenChanged is 2% of
the profile because we're calling into inActiveDocument() which is not
inline and then doing markAncestorsWithChildNeedsStyleRecalc() which is
also not inline without checking if the ancestor marking is actually needed.

markAncestorsWithChildNeedsStyleRecalc() also blindly calls
document().scheduleRenderTreeUpdate() which is not inline and does a bunch
of work.

This patch fixes the checks in ::childrenChanged to check the childNeedsStyleRecalc()
bit before calling markAncestorsWithChildNeedsStyleRecalc. This is equivalent to
checking inActiveDocument() as well since all nodes not in the tree have this bit set,
and all nodes in an inactive document will also have this bit set, so we'd never call
into markAncestorsWithChildNeedsStyleRecalc() in those trees.

It also adds an early  out so we don't call into scheduleRenderTreeUpdate() so often
when invalidating  style.

R=eseidel@chromium.org,abarth@chromium.org
BUG=357087

Review URL: https://codereview.chromium.org/219993002

git-svn-id: svn://svn.chromium.org/blink/trunk@170705 bbb929c8-8fbe-4397-9dbb-9b2b20218538

MediaQueryTokenizer handling of escapes was faulty, causing media queries with escaped code points to be considered invalid.
This CL fixes that.

BUG=358833

Review URL: https://codereview.chromium.org/221383003

git-svn-id: svn://svn.chromium.org/blink/trunk@170704 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=170660

BUG=320139
TBR=dsinclair@chromium.org

Review URL: https://codereview.chromium.org/214133008

git-svn-id: svn://svn.chromium.org/blink/trunk@170703 bbb929c8-8fbe-4397-9dbb-9b2b20218538

It is possible that HTMLStyleElement::removedFrom() is called
before HTMLStyleElement::didNotifySubtreeInsertionsToDocument().

BUG=356653
TEST=append-child-style-crash.html,javascript-url-style-crash.html
R=esprehn@chromium.org, abath@chromium.org

Review URL: https://codereview.chromium.org/221673003

git-svn-id: svn://svn.chromium.org/blink/trunk@170702 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Only set this bit for non-fullscreen video, since toggling is ignored for fullscreen video.

Chromium side change is https://codereview.chromium.org/201573013.

BUG=179489

Review URL: https://codereview.chromium.org/200723010

git-svn-id: svn://svn.chromium.org/blink/trunk@170701 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Need to recalculate the repaint rects for children inside fixed objects when we
find the repaint rects for fixed objects after scrolling so we invalidate the fixed
objects with the correct scroll position.

BUG=358550

Review URL: https://codereview.chromium.org/222473004

git-svn-id: svn://svn.chromium.org/blink/trunk@170700 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=357672

Review URL: https://codereview.chromium.org/215973005

git-svn-id: svn://svn.chromium.org/blink/trunk@170699 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=337425
R=japhet@chromium.org

Review URL: https://codereview.chromium.org/221813005

git-svn-id: svn://svn.chromium.org/blink/trunk@170698 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This reverts r170690. Typo in filename breaks the build

TBR=shans
BUG=

Review URL: https://codereview.chromium.org/216603008

git-svn-id: svn://svn.chromium.org/blink/trunk@170697 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Review URL: https://codereview.chromium.org/219333009

git-svn-id: svn://svn.chromium.org/blink/trunk@170695 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Document::loader() might be null if re-entrancy is involved.

BUG=357570

Review URL: https://codereview.chromium.org/222433003

git-svn-id: svn://svn.chromium.org/blink/trunk@170693 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=rossberg
BUG=359194

Review URL: https://codereview.chromium.org/222773003

git-svn-id: svn://svn.chromium.org/blink/trunk@170692 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Note: This is a rework of issue 212023004

This patch refactors KeyframeEffectModel to support different types of
keyframe values. The common keyframe logic is retained in
KeyframeEffectModelBase to be inherited by the keyframe type specialisations.

KeyframeEffectModel is then templated on Keyframe type, with a hierarchy of Keyframes
including StringKeyframe and AnimatableValueKeyframe being introduced.

String based keyframes are being introduced to properly support the JS API
for Web Animations element.animate().

Review URL: https://codereview.chromium.org/215883005

git-svn-id: svn://svn.chromium.org/blink/trunk@170690 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is a speculative fix for crashes that are reported from wild.

BUG=358844
R=esprehn@chromium.org

Review URL: https://codereview.chromium.org/220723012

git-svn-id: svn://svn.chromium.org/blink/trunk@170689 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Update the test so that adjustment is needed and use a weak hash set
of mixin pointers.

I realized this didn't work when I needed to use it for the
Node hierarchy.

R=erik.corry@gmail.com, kouhei@chromium.org, oilpan-reviews@chromium.org, vegorov@chromium.org

Review URL: https://codereview.chromium.org/222653003

git-svn-id: svn://svn.chromium.org/blink/trunk@170688 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Since I'm responsible to many of the recent changes in media query parsing, I've added myself to the watchlist for that code.
I also added the 'parser/MediaQuery' path, to cover the new code.

BUG=

Review URL: https://codereview.chromium.org/221843002

git-svn-id: svn://svn.chromium.org/blink/trunk@170687 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=108012
R=cmumford,jsbell,abarth

Review URL: https://codereview.chromium.org/205413004

git-svn-id: svn://svn.chromium.org/blink/trunk@170686 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch keeps all ranges in CSSRules up-to-date after setPropertyText/setRuleSelector methods gets called.

It also updates links in stylesidebarpane sections titles.

Review URL: https://codereview.chromium.org/220403005

git-svn-id: svn://svn.chromium.org/blink/trunk@170684 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Consider the script resolved during font-fallback when determining the
minimum anti-aliasing font size. This matches the logic in
createFontPlatformData and ensures that CJK characters rendering using a
fallback font are legible.

R=dglazkov@chromium.org
BUG=357864

Review URL: https://codereview.chromium.org/222543002

git-svn-id: svn://svn.chromium.org/blink/trunk@170683 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Enables the MQ parser unit test to verify that both the Bison parser and MediaQueryParser are functioning as they should be.
Also fixes a couple of bugs that were found in the process, and adds testing for these cases.

BUG=358041, 358967

Review URL: https://codereview.chromium.org/221693004

git-svn-id: svn://svn.chromium.org/blink/trunk@170682 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Move GraphicsLayerTest, ImageLayerChromiumTest,
OpaqueRectTrackingContentLayerDelegateTest, FilterOperationsTest,
ImageFilterBuilderTest, and WEBPImageDecoderTest.

BUG=353585

Review URL: https://codereview.chromium.org/216423011

git-svn-id: svn://svn.chromium.org/blink/trunk@170681 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We haven't built Chrome without TextTrack support for a long
time so I'm removing the VideoTrack feature flag and removing
all conditional code that depended on it. It is part of the
HTML5 spec and should be on by default for any compliant
implementation that supports media.

Review URL: https://codereview.chromium.org/205583009

git-svn-id: svn://svn.chromium.org/blink/trunk@170680 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=ager@chromium.org, dstockwell@chromium.org, ericwilligers@chromium.org, jochen@chromium.org, mikhail.pozdnyakov@intel.com
BUG=

Review URL: https://codereview.chromium.org/219413002

git-svn-id: svn://svn.chromium.org/blink/trunk@170679 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates WebMediaSource to have an addSourceBuffer() version that
allows caller to request which of LegacyFrameProcessor or a (not yet
landed in Chromium) new frame processor to use to process frames parsed
during appends to the source buffer.

The choice of which frame processor to request is based on
--enable-experimental-web-platform-features, as exposed through
RuntimeEnabledFeatures::mediaSourceExperimentalEnabled(): if enabled,
the new frame processor is requested. Otherwise, LegacyFrameProcessor is
requested.

Until the updated Chromium side WebMediaSourceImpl lands and (later)
the new Chromium frame processor lands, LegacyFrameProcessor will be
used regardless of the requested processor.

R=acolwell@chromium.org,abarth@chromium.org
BUG=249422
TEST=All MediaSource layout tests pass both before and after Chromium
side WMSI change is applied locally, and MediaSourceBase logging shows
frameProcessorChoice reflects usage (or lack) of cmdline switch
--enable-experimental-web-platform-features

Review URL: https://codereview.chromium.org/220593010

git-svn-id: svn://svn.chromium.org/blink/trunk@170678 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Traverse (in preorder) the whole subtree under a render block with a first-letter selector when
looking for the text renderers which would need to get the apropriate style applied, instead of
only considering the first renderer only, which might not contain all the text that should be
considered as a "first letter" (leading and trailing punctuation should be included too).

The present patch also considers leading spaces, if any, as part of the first-letter element,
to match the behaviour of other renderers such as IE and Opera (Firefox does not do that).

R=eae@chromium.org, leviw@chromium.org
BUG=354403
TEST=Add new test to check first-letter elements composed of different renderers.

Review URL: https://codereview.chromium.org/207553007

git-svn-id: svn://svn.chromium.org/blink/trunk@170677 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This could help better understand the Blink resource-fetching overhead
seen, for example, in XMLHttpRequest.send().

R=japhet@chromium.org
BUG=348655

Review URL: https://codereview.chromium.org/213353003

git-svn-id: svn://svn.chromium.org/blink/trunk@170676 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Prevents a warning with GCC 4.8 on Android.

BUG=

Review URL: https://codereview.chromium.org/221873006

git-svn-id: svn://svn.chromium.org/blink/trunk@170675 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=357672

Review URL: https://codereview.chromium.org/219183004

git-svn-id: svn://svn.chromium.org/blink/trunk@170674 bbb929c8-8fbe-4397-9dbb-9b2b20218538

On pressing alty+down key, datalist is open and no stepDown is generated.

R=tkent@chromium.org, isherman@chromium.org, keishi1@chromium.org
BUG=229073
TEST= On down+alt key combination does not generate stepDown.

Review URL: https://codereview.chromium.org/218683010

git-svn-id: svn://svn.chromium.org/blink/trunk@170673 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The problem with AlgorithmRegistry is that it is not threadsafe.

BUG=357673,245025

Review URL: https://codereview.chromium.org/211423008

git-svn-id: svn://svn.chromium.org/blink/trunk@170672 bbb929c8-8fbe-4397-9dbb-9b2b20218538