GitLab

This CL changes list-button rows to be 'actionable'. That means that
a hand mouse cursor is shown rather than an I-beam mouse cursor when the
mouse is over the button.

BUG=635819
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2261803002
Cr-Commit-Position: refs/heads/master@{#413472}

Adds interfaces, typemaps, and tests for collecting
StackSamplingProfiler result types. The collection interface will be
used to receive stack profiles from non-browser processes.

This change was previously reverted in
https://codereview.chromium.org/2264613002/ for introducing a skia
dependency that broke cronet. The dependency has been removed and the
new code has been split into independent targets.

BUG=517958

Committed: https://crrev.com/a6e65bb0cea385fc9fa82a5a1af4a5c3616e43e4
Review-Url: https://codereview.chromium.org/2253013008
Cr-Original-Commit-Position: refs/heads/master@{#413285}
Cr-Commit-Position: refs/heads/master@{#413471}

This patch fixes a bug introduced by [1] where explicit test preference
overrides (e.g., testRunner.overridePreference(...)) would get re-used
in later tests.

BlinkTestController::PrepareForLayoutTest should completely reset
preferences before every test. On the first test run, default_prefs_
is initialized with default preferences plus compositor-test-specific
preferences. To ensure the compositor-test-specific preferences are
reset for subsequent runs, we call OverrideWebkitPrefs(&default_prefs_).
The bug in [1] was that default_prefs_ was read from the previous test
and would include explicitly overridden preferences. This patch removes
this line so default_prefs_ is no longer set to the previous test's
preferences.

Test that compositor-specific preferences are still reset:
run-webkit-tests compositing/animation/state-at-end-event-transform-layer.html fast/canvas/canvas-render-layer.html --additional-driver-flag=--enable-slimming-paint-v2 --num-retries=0 --child-processes=1

Test that explicit preference overrides are reset:
run-webkit-tests third_party/WebKit/LayoutTests/fast/harness/override-preferences.html third_party/WebKit/LayoutTests/fast/harness/results.html --num-retries=0 --child-processes=1

[1] https://crrev.com/a5191dca2b309396a7743816917f19d037ebc39a

R=carlosk@chromium.org,wangxianzhu@chromium.org
BUG=620126

Review-Url: https://codereview.chromium.org/2264883002
Cr-Commit-Position: refs/heads/master@{#413470}

This changes the pipeline that initializes prerender contents
with a size to use bounds instead. Note that not all addPrerenderX
calls are switched to use Rect since for everywhere except Android
this won't be needed. Android needs bounds instead of size to get
info about the Physical Backing Size, also the top container height
(to be used for toolbar hide on scroll). With a Rect, all these can
be initialized and we can avoid extra Resizes that happen when the
native View is properly tied to the Android View hierarcy on swap.

This change only updates the params and shouldn't be making any
functional changes. A follow up Android only patch will start setting
the value to the origin.

BUG=628302

Review-Url: https://codereview.chromium.org/2259533003
Cr-Commit-Position: refs/heads/master@{#413469}

R=stevenjb@chromium.org

Review-Url: https://codereview.chromium.org/2254173003
Cr-Commit-Position: refs/heads/master@{#413468}

First, use %16p in debugging prints in offline context so that the
context address lines up with other uses of the address.

Second, add numbers to the AudioNode:NodeType enum. I'm tired of
counting every time I need to look up a node number when
DEBUG_AUDIONODE_REFERENCES is enabled.

BUG=
TESTS=

Review-Url: https://codereview.chromium.org/2269543003
Cr-Commit-Position: refs/heads/master@{#413467}

...instead of the ugly global map.

R=rego@igalia.com,jfernandez@igalia.com

Review-Url: https://codereview.chromium.org/2262973002
Cr-Commit-Position: refs/heads/master@{#413466}

This CL:
* Fixes animation when using dom-if + iron-collapse.
* Uses CrScrollableBehavior to get keyboard focus and top/botom
  separators.

BUG=609156
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2260613003
Cr-Commit-Position: refs/heads/master@{#413465}

File platform_test.h will be included from code built with Objective C
Automatic Reference Counting that forbids explicit usage of
NSAutoreleasePools. The implementation file will not be built with ARC,
so the functionality is preserved. The implementation file cannot be
built with ARC accidentally because it mentions NSAutoreleasePool.

BUG=None
TEST=None

Review-Url: https://codereview.chromium.org/2121893003
Cr-Commit-Position: refs/heads/master@{#413464}

This patch refactors CachingWordShapeIterator::nextWordEndIndex so that:

1. Better readability.
2. Split 3 cases into 3 different loops to make it possible to identify
   the loop from the stack trace.

BUG=639085

Review-Url: https://codereview.chromium.org/2263083002
Cr-Commit-Position: refs/heads/master@{#413463}

An incoming service connection may outlive any ConnectionFilter
which added interfaces to it, so a ConnectionFilter impl must not
register any interface binders which hold unsafe references to
itself.

This fixes cases where that was being done.

BUG=639650
TBR=ben@chromium.org

Review-Url: https://codereview.chromium.org/2268603002
Cr-Commit-Position: refs/heads/master@{#413462}

It has been enabled by default since
https://codereview.chromium.org/1529323002/

BUG=156590

Review-Url: https://codereview.chromium.org/2265093002
Cr-Commit-Position: refs/heads/master@{#413461}

Currently, LoginDatabase::Init closes its sql::Connection on error. It also
uses an sql::Transaction to roll back changes made so far. Unfortunately, it
does so in the wrong order.

This CL fixes that and adds a test for it.

BUG=639803

Review-Url: https://codereview.chromium.org/2265103002
Cr-Commit-Position: refs/heads/master@{#413460}

This flag is now enabled by default, so we don't need to specify it.
Remove it from the build files.

R=thakis@chromium.org

Review-Url: https://codereview.chromium.org/2260913002
Cr-Commit-Position: refs/heads/master@{#413459}

Also remove it altogether as it isn't used anymore.

BUG=612908

Review-Url: https://codereview.chromium.org/2254813002
Cr-Commit-Position: refs/heads/master@{#413458}

Enable inline video playback on iOS10 or greater.

BUG=395206

Review-Url: https://codereview.chromium.org/2265653002
Cr-Commit-Position: refs/heads/master@{#413457}

When checking the recent data about bookmark opening stats, we found a
mystery that is hard to explain: the number of bookmark open is larger
than the number of bookmark manager open on phones. This CL adds some
extra loggings to the bookmark code to seek definite answer to this
mystery.

BUG=639512

Review-Url: https://codereview.chromium.org/2266633002
Cr-Commit-Position: refs/heads/master@{#413456}

Currently, there are three independent testing implementations of the
CredentialsFilter API, all providing similar no-op implementation. In a coming
CL there will be a fourth one.

This CL pulls the stub of the API into a separate file, simplifying the 3
(soon 4) places in code where it is needed.

BUG=621355

Review-Url: https://codereview.chromium.org/2260203004
Cr-Commit-Position: refs/heads/master@{#413455}

This option allows Cronet embedders to do end-to-end testing which
requires pointing requests to a particular test server.

BUG=594601

Review-Url: https://codereview.chromium.org/1817553002
Cr-Commit-Position: refs/heads/master@{#413454}

WindowManagerObserver can operate on the WMShell instance when the
WindowTreeClient is destroyed (e.g. AcceleratorRegistrarImpl). So
notify the observers before destroying the WMShell instance.

BUG=639039

Review-Url: https://codereview.chromium.org/2269473003
Cr-Commit-Position: refs/heads/master@{#413453}

This allows metrics to survive and be collected even if a
subprocess terminates before reporting.

Only the GPU process is monitored with this CL but it should
be trivial to add the remaining ones going forward.

BUG=546019

Review-Url: https://codereview.chromium.org/2224063002
Cr-Commit-Position: refs/heads/master@{#413452}

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/d2cc7436caf8acae3d3c8ac94858cc2afaf356bf chromium_trybot: extract chromium_tests.trybot_steps (phajdan.jr@chromium.org)

TBR=martiniss@chromium.org,phajdan.jr@chromium.org
BUG=584079

Recipe-Tryjob-Bypass-Reason: Autoroller
Review-Url: https://codereview.chromium.org/2267803002
Cr-Commit-Position: refs/heads/master@{#413451}

This CL adds isPaintable() to OffscreenCanvas which mimics HTMLCanvasElement::
isPaintable() with some adjustments. Certain common constants and utility
functions shared by OffscreenCanvas and HTMLCanvasElement have been moved to
platform/graphics/. This CL is to facilitate toBlob implementation.

BUG=563830
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2255683005
Cr-Commit-Position: refs/heads/master@{#413450}

The new result is wrong, more wrong than the old result, but
having it as a failure in test expectations means we do not
catch any regressions. We did not, for instance, catch the
regression that made the test result worse.

TBR=fmalita@chromium.org
BUG=502531,636475

Review-Url: https://codereview.chromium.org/2260193002
Cr-Commit-Position: refs/heads/master@{#413449}

This patch just moves some expectations around to enable
efficient triage.

TBR=chrishtr@chromium.org
BUG=636475

Review-Url: https://codereview.chromium.org/2263223002
Cr-Commit-Position: refs/heads/master@{#413448}

BUG=

Review-Url: https://codereview.chromium.org/2260073005
Cr-Commit-Position: refs/heads/master@{#413447}

BUG=623986
R=thakis@chromium.org

Review-Url: https://codereview.chromium.org/2263503002
Cr-Commit-Position: refs/heads/master@{#413446}

I am using it for the LayoutNG transition; there does not seem to be a need
to limit this to grid items as nothing else currently sets this.

This change also makes LayoutNG use this new function.

R=eae@chromium.org,ikilpatrick@chromium.org,glebl@chromium.org
BUG=635619

Review-Url: https://codereview.chromium.org/2257223002
Cr-Commit-Position: refs/heads/master@{#413445}

Revert of Delete the class WebRtcVideoCapturerAdapter::MediaVideoFrameFactory (patchset #5 id:80001 of https://codereview.chromium.org/2244213002/ )

Reason for revert:
Somehow breaks timestamping. Reverting for investigation. To reproduce, run

./browser_tests --gtest_filter='WebRtcVideoQualityBrowserTests/WebRtcVideoQualityBrowserTest*' --run-manual --ui-test-action-max-timeout=200000

This produces lots of warnings like

[1:14:0822/153830:WARNING:video_capture_input.cc(79)] Same/old NTP timestamp (3680861906110 <= 3680861906110) for incoming frame. Dropping.
[1:14:0822/153830:WARNING:video_capture_input.cc(79)] Same/old NTP timestamp (3680861906042 <= 3680861906042) for incoming frame. Dropping.

Original issue's description:
> Delete the class WebRtcVideoCapturerAdapter::MediaVideoFrameFactory
>
> Instead, use the AdaptFrame and OnFrame methods of the VideoCapturer
> base class.
>
> This cl also applies the new timestamp aligner logic, to translate
> Chrome's camera timestamps into the timescale of rtc::TimeMicros(). No
> other change in behavior is intended.
>
> BUG=webrtc:5682, webrtc:5740, 516700
>
> Committed: https://crrev.com/5b9a4b339447b033ca52244a02c0bef17067385f
> Cr-Commit-Position: refs/heads/master@{#413093}

TBR=perkj@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=webrtc:5682, webrtc:5740, 516700

Review-Url: https://codereview.chromium.org/2263183002
Cr-Commit-Position: refs/heads/master@{#413444}

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/57476cefd25ae2063bd373156fefab4f756ed90d Roll recipe dependencies (trivial). (recipe-roller@chromium.org)
depot_tools:
  https://crrev.com/e328cf9c78a02f44b24dfa809284a6e94bcec935 git cl issue: add --json switch for machine-readable output (phajdan.jr@chromium.org)

TBR=martiniss@chromium.org,phajdan.jr@chromium.org
BUG=none

Recipe-Tryjob-Bypass-Reason: Autoroller
Review-Url: https://codereview.chromium.org/2261193002
Cr-Commit-Position: refs/heads/master@{#413443}

Change the API around a little so that we can mock out the callback to
ensure that we are calling in the correct scenarios.

BUG=637640

Review-Url: https://codereview.chromium.org/2259823007
Cr-Commit-Position: refs/heads/master@{#413442}

UpdatePreferredLoginState needs to update PasswordStore with a slightly
modified version of some of the forms in PasswordFormManager's |best_matches_|.
Currently it does so by modifying |best_matches_| directly, and then updating
the PasswordStore.

That has the advantage of keeping an up-to-date copy in |best_matches_| without
reloading from PasswordStore (and also sparing PasswordForm copies inside
UpdatePreferredLoginState.

It also has a disadvantage: |best_matches_| cannot be const. That will become
a problem once the ownership of |best_matches_| will be transferred from
PasswordFormManager to a new class, FormFetcher, shared by multiple
PasswordFormManager instances. While it is possible to let all
PasswordFormManager instances modify the forms owned by a FormFetcher, that
will make the code hard to understand and prone to unexpected side-effects.

The advantage seems limited -- at the point UpdatePreferredLoginState is called,
the PasswordFormManager is saving a password form, and will be deleted soon.
It will not need the |best_matches_|. Once FormFetcher starts owning them,
it will need to wait for a refresh from PasswordStore to get them updated.
But that's not worse than today, when different PasswordFormManager instances do
not share their |best_matches_| at all.

BUG=621355

Review-Url: https://codereview.chromium.org/2261723002
Cr-Commit-Position: refs/heads/master@{#413441}

BUG=skia:5227

Review-Url: https://codereview.chromium.org/2264673003
Cr-Commit-Position: refs/heads/master@{#413440}

Since the RootScroller proposal allows arbitrary elements to become the outer
viewport scroll layer, the scroll chaining behavior need a bit of an
adjustment. Specifically, we shouldn't chain past the outer viewport scroll
layer. The current code assumes that once we hit the outer viewport scroll
layer, we can continue chaining since we'll get to the inner viewport scroll
layer.

In this CL, we'll add the inner viewport scroll layer to the scroll chain and
immediately break when we encounter the outer viewport scroll layer, relying on
the fact that the scrolling code will use cc::Viewport to scroll the inner
viewport scroll layer.

Also added a test to make sure TopControls movement works correctly when a non-
default outer viewport layer is used.

BUG=505516
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2243973002
Cr-Commit-Position: refs/heads/master@{#413439}

It's possible for frame_ to be deleted after this call, so add null
checks after call sites.

BUG=639688, 639689
TEST=No more OnStop or NavigateInternal crashes in RenderFrameImpl.

Review-Url: https://codereview.chromium.org/2260383002
Cr-Commit-Position: refs/heads/master@{#413438}

It turns out that the original patch
(https://codereview.chromium.org/2133093002) didn't actually enable
browser logging in the System Health smoke tests.

BUG=623058,639538

Review-Url: https://codereview.chromium.org/2267663003
Cr-Commit-Position: refs/heads/master@{#413437}

Inside PasswordFormManager and FormSaver, the |credentials_to_update| argument is used to pass copies of non-best matches with a modified password, in order to be saved to PasswordStore.

Currently, the PasswordForm in |not_best_matches_| is changed directly, and a pointer to it is passed in |credentials_to_update|. However, |not_best_matches_| will need to become constant as its ownership transfers to the coming FormFetcher class, to avoid potential for surprising side-effects when multiple PasswordFormManager change the shared object inside one FormFetcher.

This CL makes that possible by changing the element type of |credentials_to_update| from a pointer to a PasswordForm to a PasswordForm itself. That way it is possible to modify just the copy inside |credentials_to_update| and not the original in |not_best_matches_|. The fact that |not_best_matches_| fall out of sync with the PasswordStore at that point does not matter: the current PasswordFormManager is currently saving a PasswordForm, so it will be destroyed soon and will not need |not_best_matches_|.

BUG=621355

Review-Url: https://codereview.chromium.org/2263683002
Cr-Commit-Position: refs/heads/master@{#413436}

Google Cloud Messaging recently started including a "google.sent_time"
Long in the incoming message Bundle from Google Play Services. This
patch filters out non-String values to prevent warnings about
ClassCastExceptions.

BUG=639313

Review-Url: https://codereview.chromium.org/2265543002
Cr-Commit-Position: refs/heads/master@{#413435}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/e0e1556b..e0af8c62

Please follow these instructions for assigning/CC'ing issues:
https://github.com/v8/v8/wiki/Triaging%20issues

Please close rolling in case of a roll revert:
https://v8-roll.appspot.com/
This only works with a Google account.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

TBR=hablich@chromium.org,machenbach@chromium.org,littledan@chromium.org,vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/2264013003
Cr-Commit-Position: refs/heads/master@{#413434}

In mus+ash, toplevel windows don't share the same focus_client,
so we have to use exo::WMHelper for handling window focus.

BUG=637914

Review-Url: https://codereview.chromium.org/2259243002
Cr-Commit-Position: refs/heads/master@{#413433}