GitLab

This is a refactoring and does not change behavior.

BUG=None

Review URL: https://codereview.chromium.org/728453002

git-svn-id: svn://svn.chromium.org/blink/trunk@185413 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=back-send-referrer.html%2C%20canvas-composite-image.html%C2%A0

BUG=433710, 433711
TBR=dongseong.hwang@intel.com
NOTRY=true

Review URL: https://codereview.chromium.org/729313002

git-svn-id: svn://svn.chromium.org/blink/trunk@185412 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Disable AcceptEx on Win32 Apache (patchset #1 id:1 of https://codereview.chromium.org/695193006/)

Reason for revert:
It was not the culprit of the serviceworker test flakiness.

Original issue's description:
> Disable AcceptEx on Win32 Apache
> 
> Some layout tests time out since the switch to apache server, with
> AcceptEx failures logged on error log. This may fix the issue.
> 
> BUG=30536, 431172
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=185174

TBR=dpranke@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=30536, 431172

Review URL: https://codereview.chromium.org/734553002

git-svn-id: svn://svn.chromium.org/blink/trunk@185411 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The problem is that Selection is not getting cleared 
after tapping of gesture. This was affected by 
https://codereview.chromium.org/507533002, 
because initialization for m_selectionInitiationState was 
skipped in case of gesture. 
This patch makes m_selectionInitiationState be initialized
after dispatching mousedown event for gesture tapping.

BUG=429967
R=rbyers@chromium.org
TEST=LayoutTests/fast/events/touch/gesture/gesture-tap-reset-selection-range.html

Review URL: https://codereview.chromium.org/705733002

git-svn-id: svn://svn.chromium.org/blink/trunk@185410 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=haraken
BUG=420515

Review URL: https://codereview.chromium.org/725293002

git-svn-id: svn://svn.chromium.org/blink/trunk@185409 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Previously RenderTreeBuilder::createRendererForTextIfNeeded()
and RenderTreeBuilder::createRendererForElementIfNeeded() are called
from Element and Text, respectively.

This refactoring separates logics needed for creating renderer
for Element and for Text, reducing isElement and isText checks.

More importantly, this refactoring exposes 
RenderTreeBuilderForText::createRenderer()
so that we can test Text::textRendererIsNeeded() only once when
we reattach whitespace siblings
(for https://codereview.chromium.org/684633006/).

BUG=482997

Review URL: https://codereview.chromium.org/699213003

git-svn-id: svn://svn.chromium.org/blink/trunk@185407 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Review URL: https://codereview.chromium.org/727093002

git-svn-id: svn://svn.chromium.org/blink/trunk@185406 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is a reland of first letter as a psuedo element after the revert
in r184113.

Currently, first letter renderers are created and destroyed through the
updateFirstLetters() call in RenderBlock. This has, historically, been
problematic as we may miss places where the call was required, leading to
accessing invalid memory.

This CL converts the first letter code to use a PseudoElement for the
first-letter instead of creating classes during layout.

With the PseudoElement implementation the two tests which are currently in
LayoutTests/fast/css/first-letter-removed-added.html work as expected.

BUG=391288

Review URL: https://codereview.chromium.org/672953002

git-svn-id: svn://svn.chromium.org/blink/trunk@185405 bbb929c8-8fbe-4397-9dbb-9b2b20218538

These forward references are not needed anymore. After
this patch, Source/core has no unneeded forward references
any more.

Review URL: https://codereview.chromium.org/732803003

git-svn-id: svn://svn.chromium.org/blink/trunk@185404 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This has been moved into the infra repo at:
https://chromium.googlesource.com/infra/infra/+/master/appengine/sheriff_o_matic/

NOTRY=true
R=dpranke@chromium.org

Review URL: https://codereview.chromium.org/728023004

git-svn-id: svn://svn.chromium.org/blink/trunk@185403 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This allows us to get rid of a related state bit on RenderLayer as well.

Review URL: https://codereview.chromium.org/727633002

git-svn-id: svn://svn.chromium.org/blink/trunk@185402 bbb929c8-8fbe-4397-9dbb-9b2b20218538

(Have not yet checked how many tests pass with slimming paint on, that will be next.)

BUG=423071

Review URL: https://codereview.chromium.org/721773005

git-svn-id: svn://svn.chromium.org/blink/trunk@185401 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=412088

Review URL: https://codereview.chromium.org/729823002

git-svn-id: svn://svn.chromium.org/blink/trunk@185400 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=chrishtr@chromium.org

Review URL: https://codereview.chromium.org/721363002

git-svn-id: svn://svn.chromium.org/blink/trunk@185399 bbb929c8-8fbe-4397-9dbb-9b2b20218538

These are flaky after https://codereview.chromium.org/467153003, which
made a 0ms setTimeout() no longer be a minimum 1ms delay.

These tests all depend on doing work during or after the load event, but
attempt to do so in ways that aren't strictly guaranteed to meet that
requirement.

BUG=

Review URL: https://codereview.chromium.org/733633003

git-svn-id: svn://svn.chromium.org/blink/trunk@185398 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=406422
TEST=Updated results to fast/loader/create-frame-in-DOMContentLoaded.html

Review URL: https://codereview.chromium.org/725593004

git-svn-id: svn://svn.chromium.org/blink/trunk@185397 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Part of my ultimate goal of de-coupling invalidation and painting of selection
gaps from one function so I can move the painting into BlockFlowPainter.

We don't actually paint selection gaps for non-blockFlow RenderBlocks. Make the
code actually represent that!

Review URL: https://codereview.chromium.org/720313002

git-svn-id: svn://svn.chromium.org/blink/trunk@185396 bbb929c8-8fbe-4397-9dbb-9b2b20218538

RLSA RenderBlock needs layouts excluding the CustomScrollbar thickness.
Set children of RenderBlock width is changed upon CustomScrollbar 
thickness change. This ensures to re-layout RLSA RenderBlock 
considering the CustomScrollbar thickness.

BUG= 431528

Review URL: https://codereview.chromium.org/708283002

git-svn-id: svn://svn.chromium.org/blink/trunk@185395 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Skobes has 60 patches in core/rendering[1] including work deep in layout
for text autosizing, and scrollbar code.

[1] git shortlog --author=skobes Source/core/rendering/ | wc -l

R=dsinclair,pdr

Review URL: https://codereview.chromium.org/725273002

git-svn-id: svn://svn.chromium.org/blink/trunk@185394 bbb929c8-8fbe-4397-9dbb-9b2b20218538

These belong on the KeyboardEvent interface, so add them to
KeyboardEvent.idl and count anything that still falls back to UIEvent.

MouseEvent.which does not appear to be in any spec, but it is supported
by most browsers. Measure its usage to help determine its future.

Review URL: https://codereview.chromium.org/725073003

git-svn-id: svn://svn.chromium.org/blink/trunk@185393 bbb929c8-8fbe-4397-9dbb-9b2b20218538

- implement synthetic FirstPaint markers on front-end (they used to be generated
    on the back-end with old Timeline instrumentation which is an ex-parrot)
- remove brown marks near the frame markers (regressed at r177959)

BUG=

Review URL: https://codereview.chromium.org/732683002

git-svn-id: svn://svn.chromium.org/blink/trunk@185392 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The width variant is actually just a font feature request.
Now that HarfBuzzShaper::setFontFeatures() takes care of setting these
font features based on the FontDescription, there is no longer a need
to plumb these values though FontPlatformData.

Review URL: https://codereview.chromium.org/715363005

git-svn-id: svn://svn.chromium.org/blink/trunk@185391 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Checking the existence of _WIN32_WINNT_WINBLUE is incorrect, rather
we should check that the value of _WIN32_WINNT is the desired value
as that's the level that the headers are exposing.

Also, move #include of windows.h up above check for include of
versionhelpers.h, otherwise the defines are changed mid-file which
causes problems.

Also, fix typo in spelling of "condition".

(Follow up to https://codereview.chromium.org/39693002)

R=yosin@chromium.org, tkent@chromium.org
BUG=311488, 432748

Review URL: https://codereview.chromium.org/732673002

git-svn-id: svn://svn.chromium.org/blink/trunk@185390 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This may have been used for AppKit integration at some point in the
distant past. Two checks for the left mouse button have been moved
up a bit higher and combined, since drags cannot be triggered by
any other button.

BUG=none

Review URL: https://codereview.chromium.org/726703002

git-svn-id: svn://svn.chromium.org/blink/trunk@185389 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Unnecessary GC::scale() alias.

R=schenney@chromium.org,senorblanco@chromium.org
BUG=424655

Review URL: https://codereview.chromium.org/725043002

git-svn-id: svn://svn.chromium.org/blink/trunk@185388 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=409580
TBR=pfeldman

Review URL: https://codereview.chromium.org/720613003

git-svn-id: svn://svn.chromium.org/blink/trunk@185387 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Redundant include left behind when moving code to paint/.
Also remove some other (unused) painting related includes.

BUG=412088

Review URL: https://codereview.chromium.org/721413004

git-svn-id: svn://svn.chromium.org/blink/trunk@185386 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Add a flag to indicate if the compositor thread is performing the
Mac overscroll rubber-band bounce effect.

This is false by default and will be set to true by
--enable-threaded-event-handling-mac until developement is finished,
whereupon it will be true by default and removed.

BUG=133097
TBR=dglazkov

Review URL: https://codereview.chromium.org/722353002

git-svn-id: svn://svn.chromium.org/blink/trunk@185385 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=185306

Review URL: https://codereview.chromium.org/722683003

git-svn-id: svn://svn.chromium.org/blink/trunk@185384 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of HTMLPlugInElement: Use custom focus logic only when there is a plugin. (patchset #1 id:1 of https://codereview.chromium.org/717193003/)

Reason for revert:
probable cause of use-after-free:
https://code.google.com/p/chromium/issues/detail?id=433357

Original issue's description:
> HTMLPlugInElement: Use custom focus logic only when there is a plugin.
> 
> This allows focus to enter and leave plugin placeholders, so that constituent
> elements (such as a close button) can be keyboard-focused.
> 
> TEST=fast/plugins/plugin/placeholder-focus.html
> BUG=364716
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=185290

TBR=mkwst@chromium.org,esprehn@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=364716

Review URL: https://codereview.chromium.org/729573002

git-svn-id: svn://svn.chromium.org/blink/trunk@185383 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=399094

Review URL: https://codereview.chromium.org/707753002

git-svn-id: svn://svn.chromium.org/blink/trunk@185382 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The Chromium IDL parser was updated to support Promise<T> syntax.
Update the IDLs to include types from the spec. Note that the type
isn't actually used for anything so this is a no-op change, for
documentation only.

Review URL: https://codereview.chromium.org/723643002

git-svn-id: svn://svn.chromium.org/blink/trunk@185381 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is a prereqisite for smooth file split.

BUG=

Review URL: https://codereview.chromium.org/724413002

git-svn-id: svn://svn.chromium.org/blink/trunk@185380 bbb929c8-8fbe-4397-9dbb-9b2b20218538

We've now agreed to document areas of ownership in core/OWNERS
(https://groups.google.com/a/chromium.org/forum/#!msg/blink-dev/StkI3k4Pm5U/i5PdztzMgNMJ).
This adds a few descriptions of areas people gave me on a blink-core-owners-discuss thread
back in May.

NOTRY=True
BUG=None

Review URL: https://codereview.chromium.org/723063005

git-svn-id: svn://svn.chromium.org/blink/trunk@185379 bbb929c8-8fbe-4397-9dbb-9b2b20218538

NOTRY=true

Review URL: https://codereview.chromium.org/732603002

git-svn-id: svn://svn.chromium.org/blink/trunk@185378 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=427569
TBR=dpranke@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/726943002

git-svn-id: svn://svn.chromium.org/blink/trunk@185377 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=431697

Review URL: https://codereview.chromium.org/733473002

git-svn-id: svn://svn.chromium.org/blink/trunk@185376 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=pfeldman

Review URL: https://codereview.chromium.org/724333004

git-svn-id: svn://svn.chromium.org/blink/trunk@185375 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This makes it possible for developers to call waitUntil() on the
event, in case its lifetime should be extended.

Specification link (Push API):
  https://w3c.github.io/push-api/#the-push-event


BUG=430888

Review URL: https://codereview.chromium.org/716203003

git-svn-id: svn://svn.chromium.org/blink/trunk@185374 bbb929c8-8fbe-4397-9dbb-9b2b20218538

ATTN Sheriffs: There is good probability that that this patch may have unintended side effects. For example there may be some JS tests that use setTimeout(...0) and hope that layout will occur.  With this patch that's much less likely to be the case and is actually a bug in the test.  I tried to fix some of those here https://codereview.chromium.org/644093003/

Another possibility is that something using setInterval(... 0) may suddenly show a perf regression due to running much more frequently.

BUG=402694

Review URL: https://codereview.chromium.org/467153003

git-svn-id: svn://svn.chromium.org/blink/trunk@185372 bbb929c8-8fbe-4397-9dbb-9b2b20218538