GitLab

Reason for revert:
This has broken android_aosp compilation.

http://build.chromium.org/p/tryserver.chromium/builders/android_aosp/builds/58228

Original issue's description:
> Fix dependencies for EventInterfaces.in
> 
> EventInterfaces.in should depend on event IDL files.
> It currently doesn't, which means it fails to rebuild correctly.
> This fixes it.
> 
> TBR=haraken
> BUG=341748
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=169998

TBR=haraken@chromium.org,acolwell@chromium.org,nbarth@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=341748

Review URL: https://codereview.chromium.org/212713002

git-svn-id: svn://svn.chromium.org/blink/trunk@170044 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Only needed by Apple's setPagination API, which doesn't exist in Blink anymore.

Link to "intent to remove" post:
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/7n-oLhmTkuM

Review URL: https://codereview.chromium.org/201573009

git-svn-id: svn://svn.chromium.org/blink/trunk@170043 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This complicates things significantly and also leads to repeated computations of the data object for each cell.

BUG=None

Review URL: https://codereview.chromium.org/211273005

git-svn-id: svn://svn.chromium.org/blink/trunk@170042 bbb929c8-8fbe-4397-9dbb-9b2b20218538

@jochen or tkent: Could you take a look at the wtf change. I needed to change the trait's Pass[In|Out]Type and Peek[In|Out]Type to make it compile on both gcc and clang for both oilpan and non-oilpan.

Also change the StyleResolver's ViewportStyleResolver to an OwnPtrWillBeMember and removed the call to the clearDocument method.
Added DISALLOW_ALLOCATION to TreeBoundaryCrossingRules since it is only allocated as a part object and traced it from StyleResolver.
Had to change CanvasRenderingContext2D's vector of State to a Vector of pointers to State since we don't yet support vectors of objects with vtables.

R=ager@chromium.org, erik.corry@gmail.com, haraken@chromium.org, oilpan-reviews@chromium.org, tkent@chromium.org, vegorov@chromium.org, zerny@chromium.org
BUG=341815

Review URL: https://codereview.chromium.org/211073002

git-svn-id: svn://svn.chromium.org/blink/trunk@170040 bbb929c8-8fbe-4397-9dbb-9b2b20218538

git-svn-id: svn://svn.chromium.org/blink/trunk@170039 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Since StyleColor stores the color value (32-bit RGBA) plus a single flag,
it is a bit inefficient.  In particular when multiple StyleColor objects
are members of another object, there will be a lot of unnecessary padding.

This padding can be avoided by storing the color value and the flag as
separate members of the containing structure, which allows the flags to
be packed into a bitfield.

The actual saving from this change, measured in a 64-bit Linux build, is
20 bytes per StyleRareInheritedData object.

BUG=356124

Review URL: https://codereview.chromium.org/196103013

git-svn-id: svn://svn.chromium.org/blink/trunk@170038 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Introduce an ExceptionMessage helper for emitting the desired error
message when calling a constructor as a function & have the IDL compiler
generate calls to it.

At the same time, make 'detail' a required argument to all 'failedToXX'
methods on ExceptionMessage -- better if the caller is required to
supply meaningful info.

R=nbarth@chromium.org,haraken@chromium.org
BUG=

Review URL: https://codereview.chromium.org/211333004

git-svn-id: svn://svn.chromium.org/blink/trunk@170037 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Reason for revert:
Breaks many browser tests: http://build.chromium.org/p/chromium.webkit/builders/Linux%20Tests/builds/35489

Original issue's description:
> Remove prefixed Shadow DOM APIs in Element.
> 
> It's time to remove. We've shipped unprefixed APIs for these APIs.
> 
> BUG=336121
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=170008

TBR=dglazkov@chromium.org,esprehn@chromium.org,hayato@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=336121

Review URL: https://codereview.chromium.org/212303005

git-svn-id: svn://svn.chromium.org/blink/trunk@170036 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=vsevik@chromium.org, vsevik

Review URL: https://codereview.chromium.org/212223003

git-svn-id: svn://svn.chromium.org/blink/trunk@170035 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=pfeldman
BUG=337138
NOTRY=true

Review URL: https://codereview.chromium.org/212503002

git-svn-id: svn://svn.chromium.org/blink/trunk@170034 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=wibling@chromium.org
BUG=

Review URL: https://codereview.chromium.org/208893005

git-svn-id: svn://svn.chromium.org/blink/trunk@170033 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=331301

Review URL: https://codereview.chromium.org/212333002

git-svn-id: svn://svn.chromium.org/blink/trunk@170032 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=340522

Review URL: https://codereview.chromium.org/212143002

git-svn-id: svn://svn.chromium.org/blink/trunk@170031 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=354373

Review URL: https://codereview.chromium.org/210763005

git-svn-id: svn://svn.chromium.org/blink/trunk@170030 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Misuse of this method could easily cause
memory leaks and UAFs. It wasn't actually doing anything
in the one place it was called, so I removed it.

I also fixed another FIXME about how unsigned/signed
constructors are confusing.  At least now we'll ASSERT
if you got the wrong one.

Review URL: https://codereview.chromium.org/209353004

git-svn-id: svn://svn.chromium.org/blink/trunk@170029 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This was helpful to allow a number of tests to run under other browsers, but
as we're starting to unprefix some properties we will add additional tests
instead.

Review URL: https://codereview.chromium.org/212073002

git-svn-id: svn://svn.chromium.org/blink/trunk@170028 bbb929c8-8fbe-4397-9dbb-9b2b20218538

DatabaseManager holds DatabaseContext references instead.

- We need to move DatabaseManager::registerDatabaseContext call from the
  constructor to the factory because registerDatabaseContext now increments the
  reference counter of a DatabaseContext and we can't update the reference
  counter in the constructor.

- We move DatabaseManager::unregisterDatabaseContext call from stopDatabases to
  contextDestroyed in order to match to deref() in the old code.

- We don't need m_isRegistered any longer because contextDestroyed is called
  just once. stopDatabases can be called twice.

- We need to move ~DatabaseManager implementation to DatabaseManager.cpp because
  we add RefPtr<DatabaseContext> in a member of DatabaseManager and
  ~DatabaseManager needs a complete definition of DatabaseContext.

BUG=347902

Review URL: https://codereview.chromium.org/212133002

git-svn-id: svn://svn.chromium.org/blink/trunk@170027 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=ager

Review URL: https://codereview.chromium.org/212243002

git-svn-id: svn://svn.chromium.org/blink/trunk@170026 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch is in continuation of <https://codereview.chromium.org/210363006>.
While the last patch added an overloaded TextIterator constructor, this patch
adds similar overloads to findPlainText(), which will be used in the find-in-
page implementation in WebFrameImpl.

The added function is not used yet. Hence this patch should have no impact on
the product behavior.

BUG=353430

Review URL: https://codereview.chromium.org/203973005

git-svn-id: svn://svn.chromium.org/blink/trunk@170025 bbb929c8-8fbe-4397-9dbb-9b2b20218538

(still needs a lot of work)

Review URL: https://codereview.chromium.org/207553010

git-svn-id: svn://svn.chromium.org/blink/trunk@170024 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Omitting "int" when using "unsigned" modifier as per Blink Coding Style Guidelines.

BUG=NONE

Review URL: https://codereview.chromium.org/207783002

git-svn-id: svn://svn.chromium.org/blink/trunk@170023 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Persistent handles in on-heap objects are not allowed (because it's likely to cause memory leaks). This CL removes them from the Event hierarchy, which was moved to oilpan's heap in r169941.

BUG=340522

Review URL: https://codereview.chromium.org/208743011

git-svn-id: svn://svn.chromium.org/blink/trunk@170022 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Implements backface-visibility as defined in CSS Transforms[1]. Behavior is
identical to -webkit-backface-visibility.

Note, this property is guarded by the CSSTransformsUnprefixed runtime feature
(enabled only for testing).

[1] http://www.w3.org/TR/css3-transforms/

BUG=154772

Review URL: https://codereview.chromium.org/211213002

git-svn-id: svn://svn.chromium.org/blink/trunk@170021 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is a preparation to move DatabaseContext to Oilpan heap.

We don't need to use RefPtr/PassRefPtr of DatabaseContext for function arguments
and function return value. They don't move ownership of DatabaseContext. They
might be meaningful if we didn't have self->ref() in DatabaseContext::create.

Also, this CL simplifies DatabaseContext::backend, and makes
DatabaseManager::databaseContextFor private.

BUG=347902
TEST=none; no behavior changes.

Review URL: https://codereview.chromium.org/209043008

git-svn-id: svn://svn.chromium.org/blink/trunk@170020 bbb929c8-8fbe-4397-9dbb-9b2b20218538

 * importKey-badParameters.html
 * importKey-normalize-usages.html

The reason for deleting the successful key import tests, is that there are much more exhaustive key import tests as part of the clone-* layout tests (which import a large combination of keys in order to serialize them).

Lastly this change fixes a typo in an error message "HmacImporParams" --> "HmacImportParams".

BUG=245025

Review URL: https://codereview.chromium.org/211933003

git-svn-id: svn://svn.chromium.org/blink/trunk@170019 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=307614,288072
NOTRY=true

Review URL: https://codereview.chromium.org/211933008

git-svn-id: svn://svn.chromium.org/blink/trunk@170018 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This was added in WebKit r87692 in order to fix "Scrubbing a Vimeo movie
when in fullscreen stops playback; no way to make it start again":
https://bugs.webkit.org/show_bug.cgi?id=61717

The commit message and bug comments makes it clear that this special
case is no longer needed:

 1. The media controls fullscreen button now uses largely the same code
    path as Element::webkitRequestFullscreen. "Video-only full screen is
    a mode where we use the shadow DOM as an implementation detail that
    should not be detectable by the web content" is no longer true at
    all, since Web content can observe the events fired, etc.

 2. Scrubbing no longer uses HTMLMediaElement::setPausedInternal(),
    which looks like the cause of "no way to make it start again".

BUG=none
TEST=None originally added with the code, but existing tests pass.

Review URL: https://codereview.chromium.org/210763004

git-svn-id: svn://svn.chromium.org/blink/trunk@170017 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The :host-context() functional pseudo-class tests whether there is an ancestor, outside the shadow tree,
which matches a particular selector. Its syntax is:

   :host-context( <compound-selector> )

Spec link: http://drafts.csswg.org/css-scoping/#host-selector

BUG=354287
TEST=no new tests because just renaming.

Review URL: https://codereview.chromium.org/210813003

git-svn-id: svn://svn.chromium.org/blink/trunk@170016 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL removes use of {,un}setPendingActivity from AudioContext. They were used to make the wrappers alive until Document is destroyed. This CL replaces the hack with a |hasPendingActivity()| impl.

Also, this removes ASSERT(m_isStopScheduled) from d-tor. This is never set when OfflineAudioContext::create() fails after creating an instance.

TEST=media/
BUG=354845

Review URL: https://codereview.chromium.org/208493002

git-svn-id: svn://svn.chromium.org/blink/trunk@170015 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=340522
NOTRY=true

Review URL: https://codereview.chromium.org/211623006

git-svn-id: svn://svn.chromium.org/blink/trunk@170014 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Implements transform-style as defined in CSS Transforms[1]. Behavior is
identical to -webkit-transform-style.

Note, this property is guarded by the CSSTransformsUnprefixed runtime feature
(enabled only for testing).

[1] http://www.w3.org/TR/css3-transforms/

BUG=154772

Review URL: https://codereview.chromium.org/211123002

git-svn-id: svn://svn.chromium.org/blink/trunk@170013 bbb929c8-8fbe-4397-9dbb-9b2b20218538

updateStyle was doing too many things, this patch moves the style updating
logic out into a new function called updateStyle and renames the original
one to updateRenderTree. It also renames all the other updateStyle related
things to be named with "updateRenderTree".

Review URL: https://codereview.chromium.org/211773002

git-svn-id: svn://svn.chromium.org/blink/trunk@170012 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Often in traces it's hard to know what events are being hanled and how
much time is spent where in the EventDispatcher and EventHandler. This
patch adds a bunch of new TRACE_EVENT's to make it easier.

Review URL: https://codereview.chromium.org/210823007

git-svn-id: svn://svn.chromium.org/blink/trunk@170011 bbb929c8-8fbe-4397-9dbb-9b2b20218538

After r169941, the SVG leak tests reports failures, but delta is <0, so number of live nodes is not increasing. This CL loosen the check.

Also removes FIXME comments in the tests.

BUG=340522
NOTRY=true

Review URL: https://codereview.chromium.org/211933007

git-svn-id: svn://svn.chromium.org/blink/trunk@170010 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch also cleans up CSSSelector so that it doesn't use ShadowDOM flag.
Runtime ShadowDOM flag was already set to true at r169201. This change shouldn't have any effects.

BUG=336121

Review URL: https://codereview.chromium.org/210763002

git-svn-id: svn://svn.chromium.org/blink/trunk@170009 bbb929c8-8fbe-4397-9dbb-9b2b20218538

It's time to remove. We've shipped unprefixed APIs for these APIs.

BUG=336121

Review URL: https://codereview.chromium.org/208503004

git-svn-id: svn://svn.chromium.org/blink/trunk@170008 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This indirection doesn't help and just introduces unnecessary complexity.

BUG=356257
TEST=none
R=dominicc@chromium.org,dglazkov@chromium.org

Review URL: https://codereview.chromium.org/211663002

git-svn-id: svn://svn.chromium.org/blink/trunk@170006 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Adds a runtime flag (enabled only for testing) and defines property names
for each property defined in CSS Transforms[1]. Future patches will make
these properties functional.

[1] http://www.w3.org/TR/css3-transforms/

NOTRY=true
BUG=154772

Review URL: https://codereview.chromium.org/211033002

git-svn-id: svn://svn.chromium.org/blink/trunk@170005 bbb929c8-8fbe-4397-9dbb-9b2b20218538

It broke a bunch of tests in oilpan bots:
http://build.chromium.org/p/chromium.webkit/builders/WebKit%20Mac%20Oilpan/builds/2500

> Oilpan: Move CSSAnimations and CSSAnimationUpdate to oilpan's heap
> 
> BUG=341032
> 
> Review URL: https://codereview.chromium.org/211113002

TBR=haraken@chromium.org

Review URL: https://codereview.chromium.org/199663007

git-svn-id: svn://svn.chromium.org/blink/trunk@170004 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch fixes a crash caused by stale LoadMediaResource flag in
m_pendingActionFlags when load() is explicitly called on a media
element. The insertion of a <track> element triggers the crash by
triggering the scheduling of the m_loadTimer, which ends up using the
stale flag data when the timer fires. The fix is to clear the
LoadMediaResource flag from m_pendingActionFlags when a new load is
initiated.

BUG=356352
TEST=LayoutTests/media/track/track-insert-after-load-crash.html

Review URL: https://codereview.chromium.org/211373009

git-svn-id: svn://svn.chromium.org/blink/trunk@170003 bbb929c8-8fbe-4397-9dbb-9b2b20218538