GitLab

Additionally, slightly refactor DetermineNewSelectedIndex() so that it feels less upside-down: handle the "closing inactive tab" case first. Judging by the tests and the way this is called in TabStripModel::DetachWebContentsImpl(), this is the intended order of behavior.

I wasn't able to manually reproduce a case where a grouped tab has an opener. I believe this is actually a different bug in groups: when a tab is opened by an opener, it doesn't get grouped. I'll log this separately (it's already in the Remaining Work doc for Tab Groups). For now, I kept the group logic after the opener logic, so that in most cases it will continue to behave as before.

From my experimentation, this works as expected when closing multiple tabs at once, including if they're non-consecutive or in different groups. The behavior just falls back to closing one tab at a time, so it will respect the group of the rightmost tab, which is the last to become active (see TabStripModel::DetachWebContentsImpl()).

Bug: 992512
Change-Id: I347fc52c0b0a2b5aca84a3ad2e4a6264bf8b11ee
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1834641
Commit-Queue: Connie Wan <connily@chromium.org>
Reviewed-by: Taylor Bergquist <tbergquist@chromium.org>
Reviewed-by: Bret Sepulveda <bsep@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702625}

Use content URI will show the download in files app
once it completes. Since the target file path is given,
there is no need to generate content URIs.

BUG=1010451

Change-Id: Icb61e403a25e837400893d114737155d64f35b1b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838698Reviewed-by: Shakti Sahu <shaktisahu@chromium.org>
Commit-Queue: Min Qin <qinmin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702624}

Before this CL, there were several related issues related to connecting
icons; sometimes, icons would appear smaller than they should, and other
times, icons would grow and/or shrink as they animated.

The issue was that network_icon.cc kept four caches for connecting
icons: light bars, dark bars, light arcs, and dark arcs. Each time an
icon was requested, we'd look to see if one existed in the cache and
return it if it existed; otherwise, we'd lazily create a new one.

The problem here was that different areas of the UI request different
icon sizes (e.g., Quick Settings requires a larger icon than the system
tray), but the caches described above were agnostic to size. This meant
that if a small icon was requested first, it would fill up the cache for
that icon type; then, if a larger icon size was requested later, the
same small one would be returned. This made the user-visible bug change
depending on what UIs were used in what order, making this bug difficult
to reproduce consistently.

This CL updates this code to include a cache for each icon size as well
as light vs. dark and bars vs. arcs.

Bug: 961843
Change-Id: I468542f29eec4e694a6bbfaa8623ebfe78700a04
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1836454
Commit-Queue: Kyle Horimoto <khorimoto@chromium.org>
Reviewed-by: Evan Stade <estade@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702623}

Restore support of file download with headless Chrome on Mac. The feature
was previously disabled due to test failures. The cause of the failures
has been identified as https://crbug.com/1011000, which only affects
Chrome build with unusual combination of build flags (including commit
queue, but not normal releases).

Bug: chromedriver:2991
Change-Id: I8960f379ba0b536714bad47f72be59841da2ab9e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838951
Commit-Queue: John Chen <johnchen@chromium.org>
Reviewed-by: Caleb Rouleau <crouleau@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702622}

CompositorFrameMetadata's destructor is empty. It has no special
logic. However, because it has a body defined, it is not trivial to
destruct.

Making a destructor trivial has surpriging benefits. And it is
easy to do in this case, since the destructor's body is empty.

This CL makes CompositorFrameMetadata's destructor = default.

R=ericrk@chromium.org

Change-Id: Ia80fc9fe60161f94a452b14f1e6ceb647fb96994
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838032Reviewed-by: Eric Karl <ericrk@chromium.org>
Commit-Queue: Chris Blume <cblume@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702621}

Tiny CL that just makes the highlight circular.

Bug: 1205, 929477
Change-Id: I92fbd6bc6fd89c087d7cf7755a69cd7f6633b0aa
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838676Reviewed-by: manuk hovanesian <manukh@chromium.org>
Commit-Queue: Tommy Li <tommycli@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702620}

Text fields used to display cookie data were eating gesture events. This
creates a local subclass of Textfield that correctly forwards gesture
events to the containing ScrollView to ensure that the cookie data will
scroll correctly.

Bug: 1008806
Change-Id: I08490549392a03f5ed745077f5dc483d548cd82b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838143
Commit-Queue: Dana Fried <dfried@chromium.org>
Reviewed-by: Charlene Yan <cyan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702619}

3 Utility classes being added:

DistilledDevInfo: a distilled smaller version of EventDeviceInfo with
the relevant fields for Neural Palm Rejection.

Sample: a single Sample based on an InProgressTouch

Stroke: A group of samples in multi-touch.

We also add thorough unit tests for all 3 classes.

Bug: 1009290
Change-Id: Idd7e3f9f6a09dda2bd0d619c7b576fc3d86461be
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1832951
Commit-Queue: Rob Schonberger <robsc@chromium.org>
Reviewed-by: Michael Spang <spang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702618}

Bug: 949269
Change-Id: Ie3365d3b25b7c488c46541eb2119afb05352b3df
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1833043Reviewed-by: Jared Saul <jsaul@google.com>
Commit-Queue: Manas Verma <manasverma@google.com>
Cr-Commit-Position: refs/heads/master@{#702617}

Bug: 982762
Change-Id: I9a1e30f452d48489c5a512905ab9ae79842415b4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1837082
Commit-Queue: Alex Moshchuk <alexmos@chromium.org>
Auto-Submit: Andrew Grieve <agrieve@chromium.org>
Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702616}

Make sure that public methods are optimized for APKs.

Related to:
https://crrev.com/c/1813520

Bug: 1010777, 1005802
Change-Id: I62fdac9b4acd12f5e09de01dc975cbdb5f116b99
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838141Reviewed-by: Tibor Goldschwendt <tiborg@chromium.org>
Reviewed-by: Peter Wen <wnwen@chromium.org>
Commit-Queue: Fred Mello <fredmello@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702615}

 - ArrowBubbleDrawable
 - ImageTextBubble
 - PulseDrawable
 - PulseInterpolator
 - RoundedIconGenerator
 - TextBubble
 - ViewHighlighter
 - ViewHighlighterTest
 - ViewHighlighterTestUtils

BUG=951455

Change-Id: I2dd5b9310e39964df51cf9448194decdc6942ed7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1833792Reviewed-by: Matthew Jones <mdjones@chromium.org>
Commit-Queue: Theresa  <twellington@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702614}

Bug: 1010568
Change-Id: I4b132eb3609c0fd4df1bf2148856f00774fbac07
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838161Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: Sven Zheng <svenzheng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702613}

Right now, we are dispatch touch events one touch point by one touch
point to Devtool input protocol, but Devtool does not group them
together as one touch event. Therefore, when we send touch events to
Devtool we need to group all the touch points into one touch event and
then send to Devtool.

Bug: 999982
Change-Id: I11a85625ede0651b920d732a4e1006292f8be0a8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838138Reviewed-by: John Chen <johnchen@chromium.org>
Commit-Queue: Lan Wei <lanwei@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702612}

No change to logic, this just changes explicit calls to the
std::unique_ptr constructor to instead use make_unique() or implicit
conversion from nullptr. This is to conform with a chromium presubmit
check.

This makes MetricSample's constructor public for compatibility with
make_unique(), and documents the existing static methods are still the
preferred way to instantiate the class.

This also fixes some indentation problems found by 'git-cl format'.

Bug: 1010369
Test: git cl presubmit --upload
Change-Id: I734a48629c435a8c31fe15b8a8d3e44352dcae08
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1834656
Commit-Queue: Nate Fischer <ntfschr@chromium.org>
Reviewed-by: Ilya Sherman <isherman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702611}

Change-Id: I8cfd7c201e847596d54b750aa24bb7606b2d8498
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1836019Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: François Doray <fdoray@chromium.org>
Commit-Queue: Sébastien Marchand <sebmarchand@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702610}

This cl enables it in a way so that it will be easy to revert this if
needed. A follow up will clean up the file a bit once this has been
enabled for a while.

Test: ash_unittests
Bug: 978112
Change-Id: Ie648f965106e6c021c77f039af0112364006e75a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1833046
Commit-Queue: Sammie Quon <sammiequon@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702609}

This reverts commit 987ed6b5.

Reason for revert: Suspect that this breaks android compile. Details at crbug.com/1011149

Original change's description:
> Android: Move UkmUtilsForTest.java.
> 
> Moved to a subdirectory of the build target using it. The old location
> was confusing because it's in the same directory as non test code.
> 
> Bug: 1010218
> Change-Id: I1de7cd4ec72c7967506ebd8451fcf2b9378a9fe3
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838522
> Commit-Queue: Andrew Grieve <agrieve@chromium.org>
> Auto-Submit: Eric Stevenson <estevenson@chromium.org>
> Reviewed-by: Andrew Grieve <agrieve@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#702581}

TBR=agrieve@chromium.org,estevenson@chromium.org

Change-Id: I1302fe15525dec5a281fb69afdd98aa2387d9646
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 1010218
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838973Reviewed-by: Ramya Nagarajan <ramyan@chromium.org>
Commit-Queue: Ramya Nagarajan <ramyan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702608}

Some baselines for mac-mac10.13 were mistakenly manually put under
platform/mac-mac10.13 directory after some automatic rebaseline-cl
failures. They are moved to platform/mac. The original baselines
under platform/mac are moved to platform/mac-mac10.10 and
platform/mac-mac10.11 if the baselines under platform/mac-mac10.12
and platform/mac-mac10.13 were the same (which become the new
mac baseline).

Remove the stale mac-mac10.13 directory.

Bug: 1010946
Change-Id: I050cae84bb3d535f9fd5869fedd8fb7778d7b473
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1837096Reviewed-by: Robert Ma <robertma@chromium.org>
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702607}

This CL changes button row layout tree so that play/pause
button is always centered regardless the visibility of previous/next
track button.

Bug: 1010645
Change-Id: Iba274ccd820346885768fa7243ff523313df3632
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1834899
Commit-Queue: Jazz Xu <jazzhsu@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702606}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/563be8e7..43e91961

Please follow these instructions for assigning/CC'ing issues:
https://v8.dev/docs/triage-issues

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

CQ_INCLUDE_TRYBOTS=luci.chromium.try:linux-blink-rel
CQ_INCLUDE_TRYBOTS=luci.chromium.try:linux_optional_gpu_tests_rel
CQ_INCLUDE_TRYBOTS=luci.chromium.try:mac_optional_gpu_tests_rel
CQ_INCLUDE_TRYBOTS=luci.chromium.try:win_optional_gpu_tests_rel
CQ_INCLUDE_TRYBOTS=luci.chromium.try:android_optional_gpu_tests_rel

TBR=hablich@chromium.org,v8-waterfall-sheriff@grotations.appspotmail.com

Change-Id: Iff0d56d9ca7b68a5e372e207739e2d014184145a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838595Reviewed-by: v8-ci-autoroll-builder <v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com>
Commit-Queue: v8-ci-autoroll-builder <v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#702605}

This CL modifies the function that determins the install source of an
ARC app. This handles the special case of the Play Store app, which is
classified as a default app on the Android side, but cannot be
uninstalled as it's a system image app.

Bug: 1009397
Change-Id: Iaf7e92d46596b85f09a46c724a17bc75f55991ad
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1837552Reviewed-by: Dominick Ng <dominickn@chromium.org>
Commit-Queue: Jeevan Shikaram <jshikaram@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702604}

Landing this will affect wpt importer on the next run, so if something breaks then
this CL should be reverted.

This deletes the old methods that used the `git cl try-results` command and
replaces them with the buildbucket rpc variants.

Also updated a lot of unit tests to specify try job results via MockWeb instead
of lambdas. This uncovered a couple minor bugs related to missing fields in
the response.

Bug: 1005801
Change-Id: Ia727635aeb1f2569707546454c52a0d1813683ec
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1837488
Commit-Queue: Luke Z <lpz@chromium.org>
Reviewed-by: Robert Ma <robertma@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702603}

Makes the suggestion chip container respect 1/16 (or 1/12 for smaller
screen width) horizontal margin. This ensures that at least 3 chips are
shown - if the minimum number of chips does not fit the margins, their
width is reduced by using box layout flexing.

BUG=1010236

Change-Id: I4ee0d85ce5ac263de1fc583ba7bdaa926b5344ac
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1836453
Commit-Queue: Toni Baržić <tbarzic@chromium.org>
Reviewed-by: Alex Newcomer <newcomer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702602}

Using wpt-import in Chromium 865f2173.
With Chromium commits locally applied on WPT:
03ccd4e9 "html: Move autofocus tests to html/interaction/focus/the-autofocus-attribute/"
c19433ef "Reland "[ChromeDriver] Stop using --ignore-certificate-errors""
afc509ca "[WPT/common/security-features] Fix typo"
df884b61 "[WPT/referrer-policy] Add worker subresource tests"
7e2a4f06 "[LayoutNG] Fix bidi reordering of lines with empty inline boxes"


Note to sheriffs: This CL imports external tests and adds
expectations for those tests; if this CL is large and causes
a few new failures, please fix the failures by adding new
lines to TestExpectations rather than reverting. See:
https://chromium.googlesource.com/chromium/src/+/master/docs/testing/web_platform_tests.md

NOAUTOREVERT=true
TBR=foolip

No-Export: true
Change-Id: I33b0349f1484f2fc6c0d5da7de5995d94e4502f6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838593Reviewed-by: WPT Autoroller <wpt-autoroller@chops-service-accounts.iam.gserviceaccount.com>
Commit-Queue: WPT Autoroller <wpt-autoroller@chops-service-accounts.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#702601}

Bug: 1009706, 1006965
Change-Id: Ie7a4314e593adb2a991ff611063380a1b572a34c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1835219Reviewed-by: Manu Cornet <manucornet@chromium.org>
Commit-Queue: Matthew Mourgos <mmourgos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702600}

https://chromium.googlesource.com/chromium/src/third_party/freetype2.git/+log/c912690d22ab..1f4e5bcb19ea

git log c912690d22ab..1f4e5bcb19ea --date=short --no-merges --format='%ad %ae %s'
2019-10-03 wl@gnu.org * src/sfnt/sfwoff2 (reconstruct_font): Fix reallocation.

Created with:
  gclient setdep -r src/third_party/freetype/src@1f4e5bcb19ea

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/freetype-chromium
Please CC bungeman@google.com,drott@google.com,thestig@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

CQ_INCLUDE_TRYBOTS=luci.chromium.try:linux_chromium_msan_rel_ng;luci.chromium.try:linux-blink-rel;luci.chromium.try:mac10.12-blink-rel;luci.chromium.try:mac10.13_retina-blink-rel;luci.chromium.try:win10-blink-rel;luci.chromium.try:win7-blink-rel
TBR=bungeman@google.com,drott@google.com,thestig@google.com

Bug: None
Change-Id: I93e04e377c5ef1bfb0f648cc04c6a394ef613f52
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838524Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Reviewed-by: Ben Wagner <bungeman@chromium.org>
Commit-Queue: Ben Wagner <bungeman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702599}

https://chromium.googlesource.com/chromiumos/chromite.git/+log/a5afaefee543..088b59cee131

git log a5afaefee543..088b59cee131 --date=short --no-merges --format='%ad %ae %s'
2019-10-03 manojgupta@google.com config: amd64-generic-asan: Use 16gb rootfs.

Created with:
  gclient setdep -r src/third_party/chromite@088b59cee131

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/chromite-chromium-autoroll
Please CC chrome-os-gardeners@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

CQ_INCLUDE_TRYBOTS=luci.chromium.try:chromeos-kevin-rel
TBR=chrome-os-gardeners@google.com

Bug: None
Change-Id: I5d47931a470fc4e9deb56d8e733bf63adc8b36a9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838148Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#702598}

Also does other misc. cleanup.

Bug: 1005568
Change-Id: I0f6203636ae60937ddf35b01fa985d0bb415853f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1837079
Commit-Queue: Peter Kasting <pkasting@chromium.org>
Auto-Submit: Peter Kasting <pkasting@chromium.org>
Reviewed-by: Peter Boström <pbos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702597}

In unittests for NetLogging of cookie inclusion status, verify the entire NetLog params and not just the value of "exclusion_reason".

This way the test catches regressions where the parameters contain more information than it should (say the cookie name).

Bug: 1005217
Change-Id: I7531e822bae67654fcac14b1b9900f8b3b0b98b5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838552Reviewed-by: Lily Chen <chlily@chromium.org>
Commit-Queue: Eric Roman <eroman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702596}

When we are notified on the IndexedDB sequence that a frame acquired
an IndexedDB lock, we will post a task to the PM sequence with the
process and frame routing ids. We will then invoke
GraphImpl::GetFrameNodeById() (added by this CL) to retrieve the
FrameNode and update its "holds IndexedDB lock" property.

Bug: 980533
Change-Id: Ic80a06c9177bb7abb6e28b6caf0ee3c344df0dd6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1834402
Commit-Queue: François Doray <fdoray@chromium.org>
Reviewed-by: Chris Hamilton <chrisha@chromium.org>
Reviewed-by: Sébastien Marchand <sebmarchand@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702595}

This makes it easier to copy the triple to the goma upload commandline,
and it puts pre- and post-roll git revs in the CL title.

Bug: none
Change-Id: Ided99e2bf1977e12fb4626422e07daf9e2b2eb26
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1837078Reviewed-by: Reid Kleckner <rnk@chromium.org>
Commit-Queue: Nico Weber <thakis@chromium.org>
Auto-Submit: Nico Weber <thakis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702594}

This CL is adding a global singleton to hold the FontConfig config
used by gfx API. The singleton is initialized on startup and the
reference counter is incremented to avoid the FcConfig to be deleted.

The problem we are fixing is described here:
  https://bugs.chromium.org/p/chromium/issues/detail?id=1004254#c28

The FontConfig may be deleted by an other thread while being used. This
is causing memory corruption and use-after-free. To avoid these races
between threads, the gfx component is keeping the same configuration
for the whole execution.

The global configuration needs to be manually passed to the font-config
API to avoid to load the default global configuration. New code that
are calling FcFontMatch and FcFontSort should use that configuration
too. Otherwise, these calls will work but will be flaky and may crash.

Bug: 1004254
Change-Id: I1c12ff88c39c5eb5e384a7bfb572b427632d2a76
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1835026
Commit-Queue: Etienne Bergeron <etienneb@chromium.org>
Reviewed-by: Robert Liao <robliao@chromium.org>
Reviewed-by: Alexei Svitkine <asvitkine@chromium.org>
Reviewed-by: Thomas Anderson <thomasanderson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702593}

On Linux the Widevine CDM is loaded into the zygote at startup. When the
component updater runs sometime later and finds a newer version of the
Widevine CDM, don't register it as the newer version can't be used.
Instead, similar to Flash, save the path to the new Widevine CDM in the
file latest-component-updated-widevine-cdm. Next time at startup this
file will be checked, and if it references a usable Widevine CDM, use
this version instead of the bundled CDM.

Tested this by running Chrome built with an older version of the
Widevine CDM, and then checking chrome://components for a newer (QA)
version. As there is one, it gets updated. When restarting Chrome, the
newer version is loaded. In both cases Widevine content was played using
Shaka.

Bug: 971433
Test: see description, plus new unit_tests pass
Change-Id: I2d188f5206ba45de8e87b7c51cb45a59a2875b19
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1826077Reviewed-by: Lei Zhang <thestig@chromium.org>
Reviewed-by: Sorin Jianu <sorin@chromium.org>
Reviewed-by: Xiaohan Wang <xhwang@chromium.org>
Commit-Queue: John Rummell <jrummell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702592}

Bug: 1010111,1010458
Change-Id: I5b873760e6cc0ffc0a957705f566ef99a3018864
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1837333Reviewed-by: Reid Kleckner <rnk@chromium.org>
Commit-Queue: Nico Weber <thakis@chromium.org>
Auto-Submit: Nico Weber <thakis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702591}

GpuMemoryBufferVideoFramePool uses RGBA/BGRA as the output format for
I420A video frames, and those are promoted to overlays on all platforms.
This causes SwapBuffers to fail in the GPU process on Windows since the
direct composition path can only handle YUV video.  This wasn't a
problem until we started promoting TextureDrawQuads to overlays in
crrev.com/c/1746796 assuming that the is_overlay_candidate flag would
only be set for low latency canvas that uses a swap chain.

A pixel test with yuva420p vp9 video is added.  The video was generated
using the following ffmpeg command:

ffmpeg -i four-colors.mp4 -filter_complex \
"color=c=#7F7F7F7F:s=960x540:rate=12[a];[0][a]alphamerge,format=yuva420p" \
-c:v libvpx-vp9 -auto-alt-ref 0 four-colors-vp9-i420a.webm

Bug: 1005028
Change-Id: Idc84d29c588cf3b9273fb605b2542be5441d61d7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1829556
Commit-Queue: Sunny Sachanandani <sunnyps@chromium.org>
Reviewed-by: Daniele Castagna <dcastagna@chromium.org>
Reviewed-by: Frank Liberato <liberato@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702590}

This reverts commit aaa2ec5c.

Reason for revert: Compile failure in WebLayerBrowserTestsActivity.java.

Original change's description:
> Add weblayer browser_tests APK so we can run the tests on Android.
> 
> The Shell C++ class isn't used on Android (unlike in content_shell) because we want to consume WebLayer there using the idiomatic Java API. At the same time, we want to write browser tests the same way for all platforms. So fake this by creating a Shell that reuses the existing BrowserController class that was created by Java.
> 
> Change-Id: I946957f39eb79cff2e84c8abda56eed98b28d0a0
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1815884
> Reviewed-by: Clark DuVall <cduvall@chromium.org>
> Commit-Queue: John Abd-El-Malek <jam@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#702563}

TBR=jam@chromium.org,cduvall@chromium.org

Change-Id: I3e1fbf1c3668aa4fe1799a102d9009f2ffcb3283
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838619Reviewed-by: Samuel Huang <huangs@chromium.org>
Commit-Queue: Samuel Huang <huangs@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702589}

To make the names make more sense and easier to understand.

Change-Id: I86cf70e26f6248495d28f77bf85978f50c4b64e9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1832715Reviewed-by: Navid Zolghadr <nzolghadr@chromium.org>
Commit-Queue: Ella Ge <eirage@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702588}

Bug: 1008734
Change-Id: If035ff53997e314e1c7ade677d99af6088f82e18
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838292Reviewed-by: Max Moroz <mmoroz@chromium.org>
Commit-Queue: John Budorick <jbudorick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702587}

Bug: 1010827
Change-Id: Ib03c22e708bca1d2fb2fe48dad15c1757c8d4db0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1838299
Commit-Queue: Manu Cornet <manucornet@chromium.org>
Commit-Queue: Andrew Xu <andrewxu@chromium.org>
Auto-Submit: Manu Cornet <manucornet@chromium.org>
Reviewed-by: Andrew Xu <andrewxu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#702586}