GitLab

http://src.chromium.org/viewvc/blink?view=revision&revision=190743

BUG=461128
TBR=bashi@chromium.org

Review URL: https://codereview.chromium.org/953873002

git-svn-id: svn://svn.chromium.org/blink/trunk@190747 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=460701

Review URL: https://codereview.chromium.org/951313002

git-svn-id: svn://svn.chromium.org/blink/trunk@190746 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Otherwise sync notifications are not going through.

BUG=460697

Review URL: https://codereview.chromium.org/945303002

git-svn-id: svn://svn.chromium.org/blink/trunk@190745 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Simplify form validation handling (patchset #1 id:20001 of https://codereview.chromium.org/943303002/)

Reason for revert:
This CL breaks browser test:
AllForms_FormStructureBrowserTest.DataDrivenHeuristics_108

http://build.chromium.org/p/chromium.webkit/builders/Mac10.8%20Tests/builds/12230
http://build.chromium.org/p/chromium.webkit/builders/Linux%20Tests/builds/42755
http://build.chromium.org/p/chromium.webkit/builders/Linux%20ChromiumOS%20Tests%20%281%29/builds/10647
http://build.chromium.org/p/chromium.webkit/builders/Linux%20ChromiumOS%20Tests%20%28dbg%29%281%29/builds/6767

Bisected locally on Linux:
./out/Release/browser_tests  --gtest_filter=AllForms/FormStructureBrowserTest.DataDrivenHeuristics/108

Blink 190722 / Chromium 317759: OK
Blink 190723 / Chromium 317759: NG

So Blink 190723 is reverted.

Original issue's description:
> Simplify form validation handling
> 
> This CL contains some part of https://codereview.chromium.org/460343004/ by
> sudarshan.p, which was reverted by a performance regression.
> 
> This CL simplifies form validation state handling code by
>  - lazy update of m_isValid
>    setNeedsValidityCheck just makes m_isValid dirty.
>  - m_isValid takes care of willValidate state too.
>    m_isValid = !willValidate() || valid()
> 
> We switched to invalidation sets for :valid and :invalid by Blink
> r184392. Calling pseudoStateChanged() for unchanged validity state won't have so
> bad performance. So, we can evaluate m_isValid lazily.
> 
> We need to change the meaning of m_isValid slightly so that it takes care of
> willValidate(). This is necessary to clean m_validityIsDirty of associated
> controls in HTMLFormElement::checkInvalidControlsAndCollectUnhandled.
> Without this change, there would be a case where willValidate state
> change won't invalidate :valid :invalid style of <form>.
> 
> This CL will have a few percent regression on blink_perf.dom:textarea-edit.
> A fix for the performance regression will be landed soon.
> 
> BUG=
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=190723

TBR=keishi@chromium.org,tkent@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=

Review URL: https://codereview.chromium.org/948373002

git-svn-id: svn://svn.chromium.org/blink/trunk@190744 bbb929c8-8fbe-4397-9dbb-9b2b20218538

OTS rejects gw432047.ttf because the font is broken. We should not use
it. Instead, use a variant of the Ahem-font which has a variation
selector for 'a'. When a variation selector U+E0100 follows 'a', the
glyph for 'a' should be the same glyph as 'p' (which is a 0.2em height
rectangle) on platforms which support IVS (Note that we don't support
IVS yet on some platforms).

I use a script[1] to create this font.

[1] https://gist.github.com/bashi/1048e2e84394723fdc65

BUG=461128

Review URL: https://codereview.chromium.org/954603003

git-svn-id: svn://svn.chromium.org/blink/trunk@190743 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=461238
TBR=leviw@chromium.org

Review URL: https://codereview.chromium.org/948333003

git-svn-id: svn://svn.chromium.org/blink/trunk@190742 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=haraken@chromium.org, jl@opera.com

Review URL: https://codereview.chromium.org/947123002

git-svn-id: svn://svn.chromium.org/blink/trunk@190741 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/952813002

git-svn-id: svn://svn.chromium.org/blink/trunk@190740 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/926963003

git-svn-id: svn://svn.chromium.org/blink/trunk@190739 bbb929c8-8fbe-4397-9dbb-9b2b20218538

To check if the stack memory has enough room for recursive
calls, we tried to allocate a large object and assumed the size.
This CL estimate the real thread stack limit using APIs.
On some platforms where such APIs do not work correctly,
we set the limit using the old algorithm.

On ASAN build, we disable it because sanitizer creates and uses "fake stacks".

---
This CL is re-land of https://codereview.chromium.org/910663002/, https://codereview.chromium.org/927213002/, and https://codereview.chromium.org/943293002/.

Test was changed in another CL.


BUG=420515, 457982

Review URL: https://codereview.chromium.org/955563002

git-svn-id: svn://svn.chromium.org/blink/trunk@190738 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=461187
TBR=joshualitt@chromium.org

Review URL: https://codereview.chromium.org/952823002

git-svn-id: svn://svn.chromium.org/blink/trunk@190737 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Mark virtual/threaded/fast/scroll-behavior/overflow-scroll-animates.html as flaky (timeout) on Win Debug/Linux

BUG=458420
TBR=skobes@chromium.org

Review URL: https://codereview.chromium.org/948333002

git-svn-id: svn://svn.chromium.org/blink/trunk@190736 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=435867
TBR=keonho07.kim@samsung.com

Review URL: https://codereview.chromium.org/949153002

git-svn-id: svn://svn.chromium.org/blink/trunk@190735 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/949103002

git-svn-id: svn://svn.chromium.org/blink/trunk@190734 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=448795

Review URL: https://codereview.chromium.org/951253002

git-svn-id: svn://svn.chromium.org/blink/trunk@190733 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/954683002

git-svn-id: svn://svn.chromium.org/blink/trunk@190732 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=450612
R=dsinclair@chromium.org,pdr@chromium.org,tkent@chromium.org

Review URL: https://codereview.chromium.org/931423003

git-svn-id: svn://svn.chromium.org/blink/trunk@190731 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/953763002

git-svn-id: svn://svn.chromium.org/blink/trunk@190730 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/947393002

git-svn-id: svn://svn.chromium.org/blink/trunk@190729 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Strict mode blocks insecure websocket connections, but doesn't do so in
a way that prevents them from signaling to the embedder. We wish to
suppress this signal, as it triggers visible UI in Chrome's omnibox.
This CL cleans up the implementation to make sure that happens.

BUG=450458
R=jww@chromium.org

Review URL: https://codereview.chromium.org/948853002

git-svn-id: svn://svn.chromium.org/blink/trunk@190728 bbb929c8-8fbe-4397-9dbb-9b2b20218538

- input[type=datetime-local] should have larger default width.
- If the content box has no enough space for the value, the value should be cut
  off.

BUG=414412

Review URL: https://codereview.chromium.org/938793005

git-svn-id: svn://svn.chromium.org/blink/trunk@190727 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/953693002

git-svn-id: svn://svn.chromium.org/blink/trunk@190726 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/949093002

git-svn-id: svn://svn.chromium.org/blink/trunk@190725 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/955583002

git-svn-id: svn://svn.chromium.org/blink/trunk@190724 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL contains some part of https://codereview.chromium.org/460343004/ by
sudarshan.p, which was reverted by a performance regression.

This CL simplifies form validation state handling code by
 - lazy update of m_isValid
   setNeedsValidityCheck just makes m_isValid dirty.
 - m_isValid takes care of willValidate state too.
   m_isValid = !willValidate() || valid()

We switched to invalidation sets for :valid and :invalid by Blink
r184392. Calling pseudoStateChanged() for unchanged validity state won't have so
bad performance. So, we can evaluate m_isValid lazily.

We need to change the meaning of m_isValid slightly so that it takes care of
willValidate(). This is necessary to clean m_validityIsDirty of associated
controls in HTMLFormElement::checkInvalidControlsAndCollectUnhandled.
Without this change, there would be a case where willValidate state
change won't invalidate :valid :invalid style of <form>.

This CL will have a few percent regression on blink_perf.dom:textarea-edit.
A fix for the performance regression will be landed soon.

BUG=

Review URL: https://codereview.chromium.org/943303002

git-svn-id: svn://svn.chromium.org/blink/trunk@190723 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/941363003

git-svn-id: svn://svn.chromium.org/blink/trunk@190722 bbb929c8-8fbe-4397-9dbb-9b2b20218538

It was added in 2006 <https://trac.webkit.org/changeset/12568> and was
accurate at the time.

Review URL: https://codereview.chromium.org/939303004

git-svn-id: svn://svn.chromium.org/blink/trunk@190721 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is a speculative attempt to fix Issue 456636.

CustomElementSyncMicrotaskRunQueue::dispatchIfAlive checks that the
object is alive when it starts running callbacks, but doesn't ensure
the object lives for the duration of the callbacks. Since the
callbacks can run author JavaScript, the assumption that whatever kept
the object alive will continue to do so might be too optimistic.

Additionally, strengthen the assertion that queue processing is
non-reentrant.

BUG=456636

Review URL: https://codereview.chromium.org/916403003

git-svn-id: svn://svn.chromium.org/blink/trunk@190720 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

Note to reviewers:
This CL is created manually, as DEFINE_TRACE macros doesn't support template methods.

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/940293002

git-svn-id: svn://svn.chromium.org/blink/trunk@190719 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The container's is already kept alive by a stack reference with Oilpan.

R=haraken
BUG=458776
NOTRY=true

Review URL: https://codereview.chromium.org/955573002

git-svn-id: svn://svn.chromium.org/blink/trunk@190718 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=455253
TEST=fast/dom/HTMLAreaElement/area-cursor.html

Review URL: https://codereview.chromium.org/934563002

git-svn-id: svn://svn.chromium.org/blink/trunk@190717 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The flakiness dashboard results after [1] shows the flakiness is not caused
by the flaky test itself, but caused by another test. The test logs say
xmlhttprequest-responseText-exception.html is the culprit.
This CL fixes the flakiness caused by repeated readystatechange events
on loading.
This CL makes the test use js-test.js, as recommended in [2].

1: https://src.chromium.org/viewvc/blink?view=rev&revision=190652
2: https://docs.google.com/document/d/1Yl4SnTLBWmY1O99_BTtQvuoffP8YM9HZx2YPkEsaduQ/edit#heading=h.ui2te0d6ongo

BUG=415872

Review URL: https://codereview.chromium.org/950063004

git-svn-id: svn://svn.chromium.org/blink/trunk@190716 bbb929c8-8fbe-4397-9dbb-9b2b20218538

All local frames being swapped in must be provisional local frames. The
other code path is completely unexercised outside of some unit tests.
Once the PlzNavigate work is closer to completion, the non-provisional
local frame path can be reintroduced and properly tested.

BUG=346764

Review URL: https://codereview.chromium.org/948143002

git-svn-id: svn://svn.chromium.org/blink/trunk@190715 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Revert of Added code to fix visibility of select all option in context menu. (patchset #3 id:40001 of https://codereview.chromium.org/402403003/)

Reason for revert:
This CL is causing Issue 438135 and Issue 450540. So reverting this.

Original issue's description:
> Added code to fix visibility of select all option in context menu.
> 
> If context menu is appearing on some input field. Select all option
> should appear only if some text is present inside input field.
> 
> BUG=395994
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=179636

TBR=darin@chromium.org,aelias@chromium.org,abarth@chromium.org,esprehn@chromium.org,ankit2.kumar@samsung.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=395994

Review URL: https://codereview.chromium.org/942183005

git-svn-id: svn://svn.chromium.org/blink/trunk@190714 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The class was added to share RenderLayer with SVG (see
https://bugs.webkit.org/show_bug.cgi?id=86022) with the
goal of having composited SVG but no progress was ever
made in this direction so it's better to revert back to
a more simple strategy. Also slimming paint should
change the way we composite and thus remove some deep
assumption about how we composite.

Review URL: https://codereview.chromium.org/945803004

git-svn-id: svn://svn.chromium.org/blink/trunk@190713 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/940843003

git-svn-id: svn://svn.chromium.org/blink/trunk@190712 bbb929c8-8fbe-4397-9dbb-9b2b20218538

It is consistently used as a null check for mediaControls() and is no
more efficient, so keep only mediaControls().

toMediaControls() includes an equivalent null check and
ASSERT_WITH_SECURITY_IMPLICATION.

BUG=448795

Review URL: https://codereview.chromium.org/943333003

git-svn-id: svn://svn.chromium.org/blink/trunk@190711 bbb929c8-8fbe-4397-9dbb-9b2b20218538

No change to generated code, as there are no checks to verify that a
non-nullable return type is never null.

BUG=460722

Review URL: https://codereview.chromium.org/950893002

git-svn-id: svn://svn.chromium.org/blink/trunk@190710 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Updates trace methods as of https://groups.google.com/a/chromium.org/d/msg/blink-dev/oi4Snt9jNx8/XcES55c7QaQJ

Note to reviewers:
This is manually created patchpatches, as the automated migration script didn't support inline trace methods definitions made out of class.


R=oilpan-reviews
BUG=455524

Review URL: https://codereview.chromium.org/938283002

git-svn-id: svn://svn.chromium.org/blink/trunk@190709 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Review URL: https://codereview.chromium.org/947303002

git-svn-id: svn://svn.chromium.org/blink/trunk@190707 bbb929c8-8fbe-4397-9dbb-9b2b20218538