GitLab

Gardening.

NOTRY=true
TBR=joelo@chromium.org
BUG=521764,527130

Review URL: https://codereview.chromium.org/1285263010

git-svn-id: svn://svn.chromium.org/blink/trunk@200921 bbb929c8-8fbe-4397-9dbb-9b2b20218538

TBR=leviw
BUG=509025

Review URL: https://codereview.chromium.org/1307593002

git-svn-id: svn://svn.chromium.org/blink/trunk@200920 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=522799
NOTRY=true

Review URL: https://codereview.chromium.org/1288063010

git-svn-id: svn://svn.chromium.org/blink/trunk@200919 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The test is not really slow -- it just happens to be the first test run
by one of the workers when running the layout tests. And the first test
run takes much longer than usual. See the bug for details.

BUG=522646
TEST=test still passes locally

Review URL: https://codereview.chromium.org/1284413005

git-svn-id: svn://svn.chromium.org/blink/trunk@200918 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch introduces |isVisuallyEquivalentCandidate()| as replacement of
|PositionAlgorithm::isCandidate()| to make what this functions for
improving readability.

Following patch will move |isVisuallyEquivalentCandidate()| to
"VisibleUnits.cpp" to reduce usages of |InlineBox| from "Position.cpp" as
part of making |PositionAlgorithm| class to free from |InlineBox|.

BUG=518738
TEST=n/a; no behavior changes

Review URL: https://codereview.chromium.org/1300303002

git-svn-id: svn://svn.chromium.org/blink/trunk@200917 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Gardening.

NOTRY=true
TBR=drott@chromium.org
BUG=523021

Review URL: https://codereview.chromium.org/1300213008

git-svn-id: svn://svn.chromium.org/blink/trunk@200916 bbb929c8-8fbe-4397-9dbb-9b2b20218538

In https://codereview.chromium.org/1132103007 the problem with the trailing float
in fast/block/float/trailing-float.html is that it doesn't have a linebox, so 
when line layout is run on a selection of dirty lineboxes on the parent it is
not re-added to the floating objects list and so doesn't get a paint. So the correct
fix is to ensure any floats not added to the floating objects list during line
layout are added, rather than to attempt to dirty more lines in the hope that
line layout will hit the float, add it to the floats list 
and so ensure it gets a paint.

BUG=510703

Review URL: https://codereview.chromium.org/1298603004

git-svn-id: svn://svn.chromium.org/blink/trunk@200915 bbb929c8-8fbe-4397-9dbb-9b2b20218538

When recording its filterPicture, FEImage::createImageFilterForLayoutObject() transforms the
content such that it draws into dstRect - but it keeps the recording bounds at origin
(FloatRect(FloatPoint(), dstRect.size())).

This means that the picture content can draw outside its declared bounds, and is susceptible to
incorrect culling.

Since we're mapping the content into dstRect, we should use the same for picture culling.

R=senorblanco@chromium.org

Review URL: https://codereview.chromium.org/1298283003

git-svn-id: svn://svn.chromium.org/blink/trunk@200914 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=463547
TBR=pfeldman,paulirish
NOTRY=true

Review URL: https://codereview.chromium.org/1291983004

git-svn-id: svn://svn.chromium.org/blink/trunk@200913 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch updates USB.idl to add these two events, making it an
EventTarget. WebUSBClient still needs to be updated to support
delivering these events to Blink.

BUG=492204

Review URL: https://codereview.chromium.org/1291753004

git-svn-id: svn://svn.chromium.org/blink/trunk@200912 bbb929c8-8fbe-4397-9dbb-9b2b20218538

DeprecatedPaintLayerScrollableArea will not create an overlay scrollbar if there
is no overflow on that axis, even if overflow:scroll is specified.  This is
intentional behavior added in http://crbug.com/415031.

It's correct for intrinsicScrollbarLogicalWidth to return 0 in this case, as it
already does for all overlay scrollbars.

BUG=520662
TEST=WebViewTest.GeolocationAPIEmbedderHasAccessAllow passes on Mac

Review URL: https://codereview.chromium.org/1292003007

git-svn-id: svn://svn.chromium.org/blink/trunk@200911 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We can't detect the iframe load failure without timeout.
So this CL adds timeout_enabled flag to iframe_test() and set it only for failure tests.

BUG=522587

Review URL: https://codereview.chromium.org/1293023003

git-svn-id: svn://svn.chromium.org/blink/trunk@200910 bbb929c8-8fbe-4397-9dbb-9b2b20218538

For closed shadow roots, /deep/ combinator or
::shadow pseudo element should not pierce the
closed shadow root boundary.

BUG=459136
TEST=closed-shadow-and-deep-combinator.html, closed-mode-deep-combinators.html

Review URL: https://codereview.chromium.org/1270313002

git-svn-id: svn://svn.chromium.org/blink/trunk@200909 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Don't hide the mute button for anything with audio when using the new
media playback UI.  The old behavior is unchanged.

Note that on clank, we don't really know if an audio track is present,
so we always show the mute button.

BUG=521789

Review URL: https://codereview.chromium.org/1303553003

git-svn-id: svn://svn.chromium.org/blink/trunk@200908 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch moves |parentEditingBoundary()| to "VisibleUnits.{cpp,h}" from
"Position.{cpp,h}" with |nonShadowBoundaryParentNode()| which is used only by
|parentEditingBoundary()| to cleanup "Position.{cpp,h}" for improving code
health, since |parentEditingBoundary()| is used only in "VisibleUnits.cpp".

BUG=518738
TEST=n/a; no behavior changes

Review URL: https://codereview.chromium.org/1288063008

git-svn-id: svn://svn.chromium.org/blink/trunk@200907 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch moves |computeInlineBoxPosition()| out from |PositionAlgorithm|
template class to reduce |InlineBox| dependency in |PositionAlgorithm| class
and simplify |PositionAlgorithm| class for improving code health.

Following patch will move |computeInlineBoxPosition()| to "VisibleUnits.cpp".

BUG=518738
TEST=n/a; no behavior changes

Review URL: https://codereview.chromium.org/1295893006

git-svn-id: svn://svn.chromium.org/blink/trunk@200906 bbb929c8-8fbe-4397-9dbb-9b2b20218538

* Inlined clear() and disable() methods
* Inlined promiseTracker() getter
* Disable PromiseTracker when disabling debugger instead of simply clearing the former

BUG=None

Review URL: https://codereview.chromium.org/1298373005

git-svn-id: svn://svn.chromium.org/blink/trunk@200905 bbb929c8-8fbe-4397-9dbb-9b2b20218538

It is possible for an input event that will be forwarded to a remote
frame to also cause the iframe element in the local renderer to become
hidden, which destroys the layoutObject for that element. This was
causing a crash because the forwarding code requires a layoutObject for
mouse event conversion.

This patch causes event forwarding to abort in that situation.

No test because we don't yet have layout tests running with
--site-per-process.

BUG=520705
R=dcheng@chromium.org

Review URL: https://codereview.chromium.org/1298973002

git-svn-id: svn://svn.chromium.org/blink/trunk@200904 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is in line with the FIXMEs in these tests, and hopefully end up
curing some flakyness. Fingers crossed.

BUG=522649, 521856

Review URL: https://codereview.chromium.org/1286093008

git-svn-id: svn://svn.chromium.org/blink/trunk@200903 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The outer FormData in RequestInit::RequestInit is shadowed by the inner
one and completely useless.

The variable was introduced at [1] and the bug was found by brucedawson@.

1: https://codereview.chromium.org/1301523002

BUG=457484

Review URL: https://codereview.chromium.org/1293873003

git-svn-id: svn://svn.chromium.org/blink/trunk@200902 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This argument was a Dictionary (unconstrained).
Now it is a WebIdl "dictionary" with members "video" and "audio".

Tests still pass.

BUG=522612

Review URL: https://codereview.chromium.org/1302793002

git-svn-id: svn://svn.chromium.org/blink/trunk@200901 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Debug and slow bots can exceed the 1s timeout here.  Unfortunately
there's no way to tell if this has been displayed yet, so the best
we can do is add a longer timeout.

BUG=513398
TEST=bots

Review URL: https://codereview.chromium.org/1296343003

git-svn-id: svn://svn.chromium.org/blink/trunk@200900 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Move rootUserSelectAllForNode() to EditingStrategy template class from PositionAlgorithm template class

This patch moves |rootUserSelectAllForNode()| to |EditingStrategy| template
class from |PositionAlgorithm| template class to simplify |PositionAlgorithm|
class for improving code health, since an implementation of
|rootUserSelectAllForNode()| is independent from |PositionAlgorithm|.

BUG=518738
TEST=n/a; no behavior changes

Review URL: https://codereview.chromium.org/1287053007

git-svn-id: svn://svn.chromium.org/blink/trunk@200899 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is for fixing gcc builds caused by r200693.
https://src.chromium.org/viewvc/blink?view=rev&revision=200693

BUG=

Review URL: https://codereview.chromium.org/1307503002

git-svn-id: svn://svn.chromium.org/blink/trunk@200898 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This series of CLs fixes WebCallback parameter types used in CacheStorage.

 - WebServiceWorkerResponse* (ownership not transferred)
  => const WebServiceWorkerResponse&
 - WebVector<WebServiceWorkerResponse>* (ownership not transferred)
  => const WebVector<WebServiceWorkerResponse>&
 - WebVector<WebServiceWorkerRequest>* (ownership not transferred)
  => const WebVector<WebServiceWorkerRequest>&
 - WebServiceWorkerCache* (ownership transferred)
  => WebPassOwnPtr<WebServiceWorkerCache>
 - WebVector<WebString>* (ownership not transferred)
  => const WebVector<WebString>&
 - WebServiceWorkerCacheError* (ownership transferred)
  => WebServiceWorkerCacheError

1/3: https://codereview.chromium.org/1284173004/
2/3: https://codereview.chromium.org/1284283003/
3/3: https://codereview.chromium.org/1296723005/

BUG=493531

Review URL: https://codereview.chromium.org/1296723005

git-svn-id: svn://svn.chromium.org/blink/trunk@200897 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch moves |hasRenderedNonAnonymousDescendantsWithHeight()| form
"Position.cpp" to "VisibleUnits.cpp" to reduce |InlineBox| dependency in
"Position.cpp" for improving code health, since |hasRender...()| is independent
from an implementation of |PositionAlgorithm| template class.

BUG=518738
TEST=n/a; no behavior changes

Review URL: https://codereview.chromium.org/1288873009

git-svn-id: svn://svn.chromium.org/blink/trunk@200896 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Bring our CSS font style matching closer to the relevant section of the
spec. [1] We previously had issues on matching correctly on font-stretch and
on font-style. These are addressed in this patch by reimplementing the
search for the correct font candidate out of the list of available
@font-face's.

[1] https://drafts.csswg.org/css-fonts/#font-style-matching

BUG=514751,513670

Review URL: https://codereview.chromium.org/1270723002

git-svn-id: svn://svn.chromium.org/blink/trunk@200895 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Touch dragging should select the item under the finger only if the popup is not scrolling. This matches IE behavior.

When touch begins, we enter "touch select mode". And when a scroll event is fired we disengage "touch select mode".

BUG=493033

Review URL: https://codereview.chromium.org/1287003002

git-svn-id: svn://svn.chromium.org/blink/trunk@200894 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Index setters of HTMLSelectElement and HTMLOptionsCollection should do nothing if the requested index is too large.

They should be consistent with their |length| setters.

Also, The index setters and the |length| setters have two behavior
changes as follows:

  * Do not skip processing if an option list has enough size.
    We don't limit an option list length if the options were parsed by the
    HTML parser.  So an option list length can be larger than maxSelectItems.
    There's no reason to skip the processing or shrink the length if
    the list already has enough size.

  * Show console warnings when they do nothing due to the length limitation.

BUG=522802

Review URL: https://codereview.chromium.org/1291263003

git-svn-id: svn://svn.chromium.org/blink/trunk@200893 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Move static member functions next{Previous,Next}Offset() and uncheckedPreviousOffsetForBackwardDeletion() out from PositionAlgorithm<Strategy> template class

This patch moves a member function |next{Previous,Next}Offset()| and
|uncheckedPreviousOffsetForBackwardDeletion()| out from
|PositionAlgorithm<Strategy>| template class as non-member functions since
they are independent from |Strategy| template parameter to simplify class
declaration for improving code halth.

BUG=518738
TEST=n/a; no behavior changes

Review URL: https://codereview.chromium.org/1301023003

git-svn-id: svn://svn.chromium.org/blink/trunk@200892 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch moves |isRenderedCharacter()| to "EditorCommand.cpp" from
"Position.cpp" to narrow visibility of |isRenderedCharacter()| for
improving code health, since it is used only classes derived from
|EditorCommand|.

BUG=518738
TEST=n/a; no behavior changes

Review URL: https://codereview.chromium.org/1298823003

git-svn-id: svn://svn.chromium.org/blink/trunk@200891 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch moves |inRenderedText()| out from |PositionAlgorithm| template class
as a preparation of moving |inRenderedText()| to "VisibileUnits.cpp" to reduce
dependency of |LayoutObject| in "Position.cpp" for improving code health.

BUG=518738
TEST=n/a; no behavior changes

Review URL: https://codereview.chromium.org/1301643003

git-svn-id: svn://svn.chromium.org/blink/trunk@200890 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The necessity of CrossThreadWeakPersistent arose in
https://codereview.chromium.org/1296243004/.

BUG=420515
R=haraken@chromium.org, oilpan-reviews@chromium.org, sigbjornf@opera.com

Review URL: https://codereview.chromium.org/1304573002

git-svn-id: svn://svn.chromium.org/blink/trunk@200889 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=enne@chromium.org

Review URL: https://codereview.chromium.org/1288093005

git-svn-id: svn://svn.chromium.org/blink/trunk@200888 bbb929c8-8fbe-4397-9dbb-9b2b20218538

In https://codereview.chromium.org/1292863003
shadow-dom/events/event-dispatch/test-001.html
shadow-dom/events/event-retargeting/test-004.html
missed to be removed from failure expectation.

This removes these 2 entries from TestExpectation.

BUG=505364
TBR=kojii@chromium.org

Review URL: https://codereview.chromium.org/1303683003

git-svn-id: svn://svn.chromium.org/blink/trunk@200887 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=393155

Review URL: https://codereview.chromium.org/1305463002

git-svn-id: svn://svn.chromium.org/blink/trunk@200886 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is a CL to fix (a part of) the regressions observed in tough_animation_cases
benchmarks on oilpan builds. In short, this CL fixes the heavy allocation design
of CSSAnimationUpdate.

Currently, a lot of CSSAnimationUpdate objects are allocated on Oilpan's heap.
This is problematic because the CSSAnimationUpdate objects hold HeapVectors and HeapHashMaps,
which produce a lot of unnecessary garbage in Oilpan's heap.

This CL makes CSSAnimationUpdate stack-allocated and explicitly clears the HeapVectors and HeapHashMaps
when the CSSAnimationUpdate is destructed on stack. When we need to hold a pending CSSAnimationUpdate
on CSSAnimation, it is held as a part of object of CSSAnimation. That way we can avoid
allocating unecessary HeapVectors, HeapHashMaps etc every time a pending CSSAnimationUpdate is created.

BUG=325467

Review URL: https://codereview.chromium.org/1281493004

git-svn-id: svn://svn.chromium.org/blink/trunk@200885 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Moved Position::up/downstream implementation to VisibleUnit.cpp and
renamed mostForwardCaretPosition/mostBackwardCaretPosition.
This CL splits Position implementation from layout object.
Position::upstream just calls mostForwardCaretPosition but the callers will
  call it directly.

BUG=518738

Review URL: https://codereview.chromium.org/1299323006

git-svn-id: svn://svn.chromium.org/blink/trunk@200884 bbb929c8-8fbe-4397-9dbb-9b2b20218538

ServiceWorker: Show a warning when ServiceWorker(Client) attempts to send a transferable ArrayBuffer

In the current implementation, ServiceWorker and ServiceWorkerClient do not
support sending an ArrayBuffer as a transferable object yet and unexpectedly
lose the buffer without any warning.

BUG=511119

Review URL: https://codereview.chromium.org/1303823002

git-svn-id: svn://svn.chromium.org/blink/trunk@200883 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch moves |nodeIsUserSelectAll()| out from |PositionAlgorithm| template
class to "EditingUtilities.cpp" to simplify |PositionAlgorithm| class for
improving code health, since |nodeIsUserSelectAll()| doesn't depend on
|Strategy| template parameter.

BUG=518738
TEST=n/a; no behavior changes

Review URL: https://codereview.chromium.org/1307523002

git-svn-id: svn://svn.chromium.org/blink/trunk@200882 bbb929c8-8fbe-4397-9dbb-9b2b20218538