GitLab

Make CachedMetrics during initialization have 50% chance to have all
histogram samples bypass it's cache and immediately forward metrics to
RecordHistogram (which now uses CachingUmaRecorder).

The state of the bypass is reported with a synthetic field trial:
 -  AndroidCachedMetrics.Control
 -  AndroidCachedMetrics.HistogramsBypassCache

Bug: 1046181
Change-Id: I7fe1d9d4edf6ccf8a6b2dde0348324aaf797f032
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2031443
Commit-Queue: bttk <bttk@chromium.org>
Reviewed-by: Andrew Grieve <agrieve@chromium.org>
Reviewed-by: Alexei Svitkine <asvitkine@chromium.org>
Reviewed-by: Sky Malice <skym@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738384}

This replaces the indefinite blocking during pulse::InitPulse() with a
timed wait of 5 seconds. We will now fall back to ALSA if we fail to
connect to pulse within that time.

I suspect this has the same root cause as issue 986021, but the fix
for those hangs disabled the audio process, which makes the browser
process now hang on startup.

R=jrummell

Bug: 986021, 1047655
Test: Local tests with manual shutdown of Pulse fall back to ALSA.
Change-Id: I337ef88bd953f56ef0bd9969a42f190200520a79
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037817
Commit-Queue: Dale Curtis <dalecurtis@chromium.org>
Reviewed-by: John Rummell <jrummell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738383}

What: This relands http://crrev/c/1993140 that was reverted in
http://crrev/c/2035374. The diff:
in OverlayInterstitialAdDetector::MaybeFireDetection, switch to
main_frame->ContentLayoutObject()->HitTestNoLifecycleUpdate for the hit
test.

Why: We don't need LocalFrameView::UpdateAllLifecyclePhasesExceptPaint
as we don't need to recurse into child document. Forcing the update
would change the document lifecycle state, that would break the test
expectations of FrameThrottlingTest.ThrottledLifecycleUpdate and would
introduce flakes for AccessibilityActionBrowserTest.ImgElementGetImage.


Bug: 1047921
Change-Id: Ic99420e125788d446bb11202848634ddc6e07841
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2034054
Commit-Queue: Yao Xiao <yaoxia@chromium.org>
Reviewed-by: Stefan Zager <szager@chromium.org>
Reviewed-by: Bryan McQuade <bmcquade@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738382}

This change makes UpgradeDetector an observer of the process's
BuildState, removing its own detection. It instead creates an instance
of either an InstalledVersionPoller (for desktop Chrome) or an
InstalledVersionUpdater (for Chrome OS) to do the actual detection work.

BUG=1043624

Change-Id: Ib04b0f588a3d035f2ce0300561ed86d9024c7b3e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2017352
Commit-Queue: Greg Thompson <grt@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Reviewed-by: David Roger <droger@chromium.org>
Reviewed-by: Julian Pastarmov <pastarmovj@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738381}

FeatureUtilities is not accurate anymore, as this shouldn't contain
code for specific features anymore, and it is (and has been) a
stateful class.

The responsibility of this class is to cache feature flags for use
without native being loaded.

This is CL 3/5 for renaming FeatureUtilities.

Bug: 1012975
Change-Id: I8b4c72a65361379ec34272bdff55edf416d35b55
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2028111
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: David Trainor <dtrainor@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738380}

This patch introduces variants of:
Resolve{Main,Min,Max}{Inline,Block}Length

With a callback function to provide the MinMaxSize, or IntrinsicBlockSize
as needed.

This in instead of logic used by other algorithms via. NeedMinMaxSize
et. al.

There are pros/cons to this approach:
Pro - Much more precise (and easier to maintain) calls to
      NGBlockNode::ComputeMinMaxSize and NGBlockNode::Layout.
Con - Duplicate logic for Resolve*Length.

I tried moving Resolve*InlineLength to all being callback based but some
other refactoring needs to occur first (mainly an additional ng cache).

An alternate version of this patch would be either:
1) A NeedsMinMaxSize / NeedsIntrinsicBlockSize similar to other callers
   of these functions.
2) Adding the branch before the Resolve*Length calls to check if we
   need to provide the appropriate argument.

Bug: 845235
Change-Id: I8e26e0171adc60a7e3d244af3cb21e119b2653f1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2027075Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: David Grogan <dgrogan@chromium.org>
Commit-Queue: Ian Kilpatrick <ikilpatrick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738379}

https://chrome-internal.googlesource.com/chrome/src-internal.git/+log/9a3ade44d9f4..cecc4d627631


Created with:
  gclient setdep -r src-internal@cecc4d627631

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://skia-autoroll.corp.goog/r/src-internal-chromium-autoroll
Please CC jbudorick@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.chrome.try:linux-chromeos-chrome
Bug: None
Tbr: jbudorick@google.com
Change-Id: I1d0910de3a5a395180e1a6b1953c5aa838b415a1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037696Reviewed-by: chromium-internal-autoroll <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com>
Commit-Queue: chromium-internal-autoroll <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#738378}

This reverts commit 919ba484.

Reason for revert: New test PaymentMethodViewControllerTest.DoNotShowAddCardWhenBasicCardIsNotSupported is consistently failing on linux-chromeos-chrome.  First failing build: https://ci.chromium.org/p/chrome/builders/ci/linux-chromeos-chrome/2687.

Test output (CRASH):
[ RUN      ] PaymentMethodViewControllerTest.DoNotShowAddCardWhenBasicCardIsNotSupported
[1:1:0204/141234.652201:WARNING:hugepage_text.cc(182)] The ordering seems incorrect, fall back to small page
[7821:7821:0204/141234.715769:WARNING:audio_manager_linux.cc(52)] Falling back to ALSA for audio output. PulseAudio is not available or could not be initialized.
[7821:7884:0204/141234.746168:ERROR:arc_property_util.cc(224)] arc-build-properties is not found
[7821:7884:0204/141234.746216:ERROR:arc_property_util.cc(276)] Failed to read /usr/share/arc/properties/default.prop: No such file or directory (2)
[7821:7884:0204/141234.746237:ERROR:arc_property_util.cc(295)] Failed to expand /usr/share/arc/properties/default.prop
[7821:7821:0204/141234.760051:INFO:profile_network_context_service.cc(246)] Using built-in cert verifier.
[7821:7821:0204/141234.761643:ERROR:proximity_auth_profile_pref_manager.cc(189)] Failed to find local state prefs for current user.
[7821:7821:0204/141234.816422:INFO:profile_network_context_service.cc(246)] Using built-in cert verifier.
[7821:7821:0204/141234.824119:INFO:profile_network_context_service.cc(246)] Using built-in cert verifier.
[7821:7821:0204/141234.827132:WARNING:remote_commands_service.cc(40)] Client is not registered.
[7821:7821:0204/141234.831227:WARNING:wallpaper_controller_client.cc(370)] Cannot get wallpaper files id in RemovePolicyWallpaper. This should never happen under normal circumstances.
[7821:7821:0204/141234.872150:ERROR:account_manager_migrator.cc(256)] Could not find a refresh token for the Device Account.
[7821:7821:0204/141234.876165:WARNING:drivefs_session.cc(62)] DriveFs mount failed with error: 3
[7821:7908:0204/141234.954453:WARNING:google_brand_chromeos.cc(39)] Brand code file missing: /opt/oem/etc/BRAND_CODE
BrowserTestBase received signal: Segmentation fault. Backtrace:
#0 0x55a6a7352149 base::debug::CollectStackTrace()
#1 0x55a6a72c17b3 base::debug::StackTrace::StackTrace()
#2 0x55a6a782e765 content::(anonymous namespace)::DumpStackTraceSignalHandler()
#3 0x7f81d61164b0 (/lib/x86_64-linux-gnu/libc-2.23.so+0x354af)
#4 0x55a6a7a68a22 net::test_server::EmbeddedTestServer::DoAcceptLoop()
#5 0x55a6a730574f base::TaskAnnotator::RunTask()
#6 0x55a6a731511a base::sequence_manager::internal::ThreadControllerWithMessagePumpImpl::DoWorkImpl()
#7 0x55a6a7314eb7 base::sequence_manager::internal::ThreadControllerWithMessagePumpImpl::DoSomeWork()
#8 0x55a6a736d6a9 base::MessagePumpLibevent::Run()
#9 0x55a6a73159c8 base::sequence_manager::internal::ThreadControllerWithMessagePumpImpl::Run()
#10 0x55a6a72efb57 base::RunLoop::Run()
#11 0x55a6a7328939 base::Thread::ThreadMain()
#12 0x55a6a735f3ce base::(anonymous namespace)::ThreadFunc()
#13 0x7f81d9fbf6ba start_thread
#14 0x7f81d61e841d clone


Original change's description:
> Reland "[Payments]Don't show "add card" button when basic card is not requested."
> 
> This is a reland of 08e5b730. The cl
> got reverted because of the added test failing on ChromeOS_Chrome_Linux
> build.
> 
> Original change's description:
> > [Payments]Don't show "add card" button when basic card is not requested.
> >
> > This is a Desktop only change to ensure that the "add card" button is
> > shown in payment method window only when the "basic card" method is
> > supported.
> >
> > Change-Id: Ibfc31141cf45595c0db216d017b5c8c484094424
> > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2024923
> > Commit-Queue: Sahel Sharify <sahel@chromium.org>
> > Commit-Queue: Rouslan Solomakhin <rouslan@chromium.org>
> > Reviewed-by: Rouslan Solomakhin <rouslan@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#735978}
> 
> TBR=rouslan@chromium.org
> 
> Change-Id: I90d4febb944d48cfe009c0c0e1511f9d493d1762
> Bug: 1046514
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037639
> Reviewed-by: Sahel Sharify <sahel@chromium.org>
> Commit-Queue: Sahel Sharify <sahel@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#738334}

TBR=rouslan@chromium.org,sahel@chromium.org

Change-Id: Idfe37889ed1bc2effd9bec6ae075559c293a0fcd
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 1046514
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037828Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Commit-Queue: Alex Moshchuk <alexmos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738377}

AppServiceAppItem is used to replace InternalAppItem. Update the unit
test to use AppServiceAppItem.

BUG=1016159

Change-Id: I659cd43fea543578ab4b6d7df6c1e7f57941cbdd
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037175Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Reviewed-by: Jeevan Shikaram <jshikaram@chromium.org>
Commit-Queue: Nancy Wang <nancylingwang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738376}

A similar DCHECK was removed in
http://crrev.com/0a0c3a85f4eeb668acc87dbc575405c6026971a1

The check is unnecessary, and is a no-op in release builds. The host
is meant to run as root for the login screen when curtain-mode is not
enabled.

Bug: 1020739
Change-Id: I5f764611b745ce58c47353b63678fbbadb3a68f3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037466
Commit-Queue: Lambros Lambrou <lambroslambrou@chromium.org>
Commit-Queue: Jamie Walch <jamiewalch@chromium.org>
Auto-Submit: Lambros Lambrou <lambroslambrou@chromium.org>
Reviewed-by: Jamie Walch <jamiewalch@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738375}

The affected targets are:
* components/send_tab_to_self/proto:send_tab_to_self_proto
* chrome/browser/web_applications/proto

Various sync .proto file imports were updated by running:
find components/sync/protocol/ | grep "\.proto$" | xargs sed -i 's/import \"/import \"components\/sync\/protocol\//g'

Bug: 1039009
Change-Id: I56bfae1ae47608bc3c4c4d27ea14f8cd13e8f43c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1988862
Commit-Queue: Sky Malice <skym@chromium.org>
Reviewed-by: Alexey Baskakov <loyso@chromium.org>
Reviewed-by: Marc Treib <treib@chromium.org>
Reviewed-by: vitaliii <vitaliii@chromium.org>
Reviewed-by: sebsg <sebsg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738374}

Using wpt-import in Chromium 56f7e59b.

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

Directory owners for changes in this CL:
foolip@chromium.org, lpz@chromium.org, robertma@chromium.org:
  external/wpt/tools

NOAUTOREVERT=true
TBR=foolip

No-Export: true
Change-Id: Ib3ad10069334748afcea611f447e090827436963
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037507Reviewed-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@{#738373}

The assistant service runs in the browser process on the UI thread.
There aren't any plans to move it out of process. Instead of using
the identity service, inject a pointer to the C++ IdentityManager
and use it to look up access tokens.

This reduces the number of customers of the identity service from
2 to 1.

Bug: none
Change-Id: I6a5155306f8ad433c4805fa6acd83f7606ca7412
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2033572
Commit-Queue: James Cook <jamescook@chromium.org>
Reviewed-by: David Roger <droger@chromium.org>
Reviewed-by: Xiaohui Chen <xiaohuic@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738372}

No change to logic. This exposes chrome_public_manifest_package as a GN
arg to support WebView developers who wish to compile
monochrome_public_apk. This updates WebView docs to mention the new GN
arg.

Bug: 1048327
Test: gn args out/Default --list=chrome_public_manifest_package
Test: out/Default/bin/monochrome_public_apk package-info
Test: tools/md_browser/md_browser.py android_webview/docs/build-instructions.md
Test: Upload to gerrit > open file > click "gitiles"
Change-Id: Iba14d6eae3bdde7cde8f7be1c0df29cd8674d8d9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036381Reviewed-by: Andrew Grieve <agrieve@chromium.org>
Commit-Queue: Nate Fischer <ntfschr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738371}

Reland of
https://chromium-review.googlesource.com/c/chromium/src/+/2032049
with fixes so avoid failing ChromeCustomTabsConnectionTest
(crbug.com/1048438).

Instead of runNowOrAfterNativeInitialization(), the method should
enqueue tasks to run when full browser starts, and not when native is
started. There is a distinction - reduced mode means native is started but full browser, which includes Profiles, isn't.

This fixes a crash when transitioning from reduced mode to full browser if a task that uses Profile was scheduled with
runNowOrAfterNativeInitialization(). Now, it will only run after the
transition, rather than immediately.

Also change hasNativeInitializationCompleted() to
isFullBrowserInitialized() and update clients. All but DownloadMetrics are meant to run in full browser, so that should actually revert their behavior back to what it was before reduced mode was introduced.

DownloadMetrics still checks if native was loaded, which includes in
reduced mode.

Bug: 1045945, 1014098, 1048438
Change-Id: If99179a5720d7e550ee5653a8625021f58fc262e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037818Reviewed-by: Mohamed Heikal <mheikal@chromium.org>
Reviewed-by: Michael Thiessen <mthiesse@chromium.org>
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738370}

Also to be consistent with dm_api.proto which is used on DMServer:
cl/293161272.

The field is not yet used anywhere, so it's safe to rename.

Change-Id: Ib348ae4c0504a71cfebf0df516b26afecfb971e6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2035988Reviewed-by: Pavol Marko <pmarko@chromium.org>
Commit-Queue: Askar Aitzhan <askaraitzhan@google.com>
Cr-Commit-Position: refs/heads/master@{#738369}

Test: manual
Change-Id: Id3142bbe89caa2aa002f0b41f846bb885f6b56a2
Fixed: 1048315
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037715Reviewed-by: Ahmed Fakhry <afakhry@chromium.org>
Commit-Queue: Avery Musbach <amusbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738368}

The |kAutofillAssistantDirectActionsName| constant is only declared for
OS_ANDROID in flag_descriptions.h, but is defined for all platforms in
flag_descriptions.cc. This did not trigger a compiler warning because
the definition has an erroneous extern keyword. Remove the extern
keyword and move the constant to the right spot. Do the same for
|kAutofillAssistantDirectActionsDescription|.

Change-Id: I0a8b174a2a27a360a97d0ffa7f5e69c110d2f970
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2025248Reviewed-by: Stephane Zermatten <szermatt@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738367}

This CL fixes the bug where extensions remain disabled with reason
custodianApprovalRequired even after supervision is removed. This CL
accomplishes this by removing supervised user related disable reasons
on startup if the logged in profile is not a child account.

Bug: 1045625
Change-Id: I35491e240610334aa9665cddc0c9f4fa0ea00037
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2026172Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
Reviewed-by: Aga Wronska <agawronska@chromium.org>
Reviewed-by: Toni Baržić <tbarzic@chromium.org>
Commit-Queue: Toby Huang <tobyhuang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738366}

- Changed the path to the simulator shared resources directory, where
the app also write rights, and won't be deleted after app is killed.
- Added "%m" to the file name, so that on-line profile merging is
enabled.
- Added a generated UUID to the file name, so that there won't be a
conflict when multiple apps are launched in one test suite.

Please see ios code coverage design doc for rationales of those changes:
go/bling-code-coverage-doc.

Bug: 1043273
Change-Id: I21b2c8f4018036fdd13eecbe7e1362c926886eca
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2031673
Commit-Queue: Zhaoyang Li <zhaoyangli@chromium.org>
Reviewed-by: John Budorick <jbudorick@chromium.org>
Reviewed-by: Eugene But <eugenebut@chromium.org>
Reviewed-by: Yuke Liao <liaoyuke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738365}

This code is complicating changing the feature, so time to remove it!

Bug: 1048736
Change-Id: Ia410d07f5ae16db540159a2d4b46aaec1731e15e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037814Reviewed-by: Tarun Bansal <tbansal@chromium.org>
Commit-Queue: Robert Ogden <robertogden@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738364}

Updating the Pixel Wrangler docs section on extending the rotation. Making them
more general for modifying the rotation. With a portion about swapping.

Change-Id: Ief4c3ffc86f87fe924e94900c16b017f38e9ed80
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037327Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Jonathan Ross <jonross@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738363}

This CL makes sure that a correct request_initiator_site_lock is used
when a URLLoaderFactory is pushed to a renderer process
during a ReadyToCommit->DidCommit transition window.  In particular,
during that window it is incorrect to use |last_committed_origin_|
(see the bug for details).

Bug: 1047436
Change-Id: If696a326f162585f42ab0a0ecbe8f8f42f07f5a2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2026168
Commit-Queue: Łukasz Anforowicz <lukasza@chromium.org>
Reviewed-by: Charlie Reis <creis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738362}

The CL https://crrev.com/737514 fixes an issue with the PDF viewer
extension not loading if another extension injects an additional iframe
into the top-level HTML document for the PDF.

Not caught in the bug description is the fact that PostMessaging support
is also broken; that was discovered (and fixed) while writing the
accompanying test.

Bug: 1046795
Change-Id: I650518162e24e5d961e45b471bfb1518e9b7531c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037343
Commit-Queue: James MacLean <wjmaclean@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738361}

For GAIA password pings, it is currently gated by SB
for >=M79 versions. However, since we are not showing warnings
to the users, these pings should instead be gated by SBER.

Bug: 1048440
Change-Id: I248b78388cb2aff093c67e774e31ee0c711c3499
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036286
Commit-Queue: Xinghui Lu <xinghuilu@chromium.org>
Reviewed-by: Varun Khaneja <vakh@chromium.org>
Reviewed-by: Bettina Dea <bdea@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738360}

Need this to ensure the quick settings widget is always activated
when focus is transferred from the message center widget.

Bug: 1048127
Change-Id: I6d142ea37bf270712d609024a0d0e09bd5a5d295
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036274
Commit-Queue: Ahmed Mehfooz <amehfooz@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Reviewed-by: Tim Song <tengs@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738359}

The previous a11y change:

https://chromium-review.googlesource.com/c/chromium/src/+/1955161

didn't work well on Windows. I'm just reverting it since I don't
have time to look into a proper fix.

This had minor conflicts, which may cause an issue merging to M81.

Bug: 1045192 1045199
Change-Id: Idbf00425dcfc8a3a516cd60a8c5338c1faea3e74
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036806Reviewed-by: Orin Jaworski <orinj@chromium.org>
Commit-Queue: Kevin Bailey <krb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738358}

This reverts commit 6fbca674.

Reason for revert: This was a short-term bandaid pending a riskier
but architecturally better fix that has now landed (crrev.com/c/2023047
by Lukasz).

Original change's description:
> Revert URLPattern wildcard to again match some invalid URLs
>
> This is a partial revert of r715320, which changed
> URLPattern::MatchesURL to decline to match all invalid
> (!GURL::is_valid) URLs. This was an inadvertent behavior change;
> it wasn't clear what the intended behavior was in this case.
>
> Reverting this seems to resolve crbug.com/1041880, a bug involving
> some PDF loads from blob: URLs getting stuck as gray boxes. While
> r715320 was originally made to resolve a crash---namely, MatchesURL
> would crash on URLs with empty paths---the follow-up commit r718423
> adds logic that will still suffice to prevent this crash (this
> is tested by ExtensionURLPatternTest.WildcardMatchesPathlessUrl and
> ExtensionURLPatternTest.NonwildcardDoesntMatchPathlessUrl).
>
> Changes:
> - URLPattern: revert the behavior change and update documentation
> to reflect that intended behavior is actually to allow the wildcard
> '<all_urls>' pattern to match invalid URLs
> - URLPattern tests: update to reflect the intended new behavior
>
> Bug: 1041880
>
> R=​devlin
>
> Change-Id: I4f4571c70d0f0dbd7b52cdadbf96a7ea6854cfe0
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2022251
> Reviewed-by: Charlie Reis <creis@chromium.org>
> Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
> Commit-Queue: David Van Cleve <davidvc@chromium.org>
> Commit-Queue: Charlie Reis <creis@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#735510}

R=rdevlin.cronin@chromium.org
CC=​creis@chromium.org,lukasza@chromium.org

Bug: 1046397
Change-Id: I3fd02ac96e3120eb15e1e8bf43ce7dad8185aa03
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036792Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
Commit-Queue: David Van Cleve <davidvc@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738357}

Use WeakPtr in PresentationCallback to avoid use-after-scope

Bug: 1043465
Test: run 100 times and no flaky
Change-Id: Iae88d5449c7ae22a05c1b968b938ee0cbdce4c2b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036873
Commit-Queue: Mitsuru Oshima <oshima@chromium.org>
Reviewed-by: Jun Mukai <mukai@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738356}

Moving libassistant shutdown a little bit earlier in the
shutdown process to help it finish shutdown.

Bug: 950146
Test: locally run and logout/shutdown
Change-Id: Ic0d3b2ca0bf0cc06702c71fb02659854540a0863
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036256
Commit-Queue: Xiaohui Chen <xiaohuic@chromium.org>
Reviewed-by: Sam McNally <sammc@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738355}

Previously code was sending CopyOutputRequest from ChildFrame only if
the root renderer submitted frame, which is incorrect. This CL
addresses the issue.

Bug: 805739
Change-Id: Ic1f1243fe0d9b660773a9dbd753a09218880f72b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037638Reviewed-by: Bo <boliu@chromium.org>
Commit-Queue: Vasiliy Telezhnikov <vasilyt@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738354}

This is the main part of the patch series migrating PlatformEvent to
ui::Event in Aura/X11. Design doc [1] describes in details such effort.

In this CL, Aura/X11 code which deals with ui::PlatformEvent is factored
to deal with ui::Event instead, which is currently a fundamental
difference between Aura/X11 and Ozone/X11. To accomplish it, this patch
does:

1. Refactor X11 PlatformWindow impl's dispatching logic to handle both
ui::Event (auto translated from XEvent and routed through
PlatformEventDispatcher API) and raw XEvents (X-specific events routed
through XEventDispatcher API). This makes it more similar to
X11WindowOzone, which also made it possible to;

2. Cleanup ui::XWindow code guarded by USE_X11, which consisted of
key/mouse/touch/scroll events handling, which is now handled at
X11EventSource level, i.e: XEvents are translated and routed through
{X,Platform}EventDispatcher infra.

3. Convert former PlatformEventDispatcher impls which deals with
X11-specific events into ui::XEventDispatchers;

4. Refactor some testing support code to make it work with this new
event flow scheme (i.e: X11-specific events routed through
XEventDispatcher instead of PlatformEventDispatcher, etc).

Even though this touches many files, the delta for each file is usually
small and trivial (e.g: PlatformEventDispatcher => XEventDispatcher).
The core changes are in X11Window and in some testing code. This should
not cause behavioral changes, but a single browser/interactive test
became flaky and was disabled for now (actually most tests from that
test suite are flaky and conditionally disabled for ozone configs, what
possibly points out to issues in the event processing pipeline when
ui::Event are used), the proposal for this is to further investigate and
address it in a separate CL.

[1] https://docs.google.com/document/d/1Neh2p6f8pew9SPuEAtnJMafozAY7uhAIfhwd43CLETc

Bug: 965991
Change-Id: I10539b8f94a5f0d4857735087ddaf5efd73f9b31
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2022280Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org>
Reviewed-by: Peng Huang <penghuang@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Thomas Anderson <thomasanderson@chromium.org>
Commit-Queue: Nick Yamane <nickdiego@igalia.com>
Cr-Commit-Position: refs/heads/master@{#738353}

Bug: 958893
Change-Id: Ie23f403e063cacdfd11cbc9a43e25aefdfc52a2e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037713Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Reviewed-by: Owen Min <zmin@chromium.org>
Commit-Queue: Yann Dago <ydago@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738352}

https://skia.googlesource.com/skia.git/+log/a2406f9218a3..25cf469a112f

git log a2406f9218a3..25cf469a112f --date=short --first-parent --format='%ad %ae %s'
2020-02-04 nifong@google.com Move Make functions for corner and discrete path effects into namespaces.
2020-02-04 mtklein@google.com remove misleading Uniforms apis
2020-02-04 mtklein@google.com thread alloc through to color filters
2020-02-04 jvanverth@google.com Set NSError on compilation/link timeouts.
2020-02-04 mtklein@google.com run .skps on PDF bots
2020-02-04 kjlubick@google.com [canvaskit] Fix flaky test

Created with:
  gclient setdep -r src/third_party/skia@25cf469a112f

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/skia-autoroll
Please CC scroggo@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:android_optional_gpu_tests_rel;luci.chromium.try:linux-blink-rel;luci.chromium.try:linux-chromeos-compile-dbg;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
Bug: None
Tbr: scroggo@google.com
Change-Id: I238ec7558815abe8201ccc04e5516e205f5848a9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2037619Reviewed-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@{#738351}

The Android framework does this some time after installation, and we
need to run this by default to closely emulate the real behavior.

Bug: 1048411
Change-Id: I80e10f72a8901a42ca0d397649c6a8fbd29198dd
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036526Reviewed-by: Oksana Zhuravlova <oksamyt@chromium.org>
Commit-Queue: Changwan Ryu <changwan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738350}

This change refreshses the expiration for DiscoverySession.Stop.Outcome.
This histogram will be analyzed to determine if its still worth keeping
or not.

Bug: 1043698
Change-Id: Ib441ed452900ec9e6ad1498b13352b66db555508
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036310
Commit-Queue: Brian White <bcwhite@chromium.org>
Reviewed-by: Brian White <bcwhite@chromium.org>
Auto-Submit: Ovidio de Jesús Ruiz-Henríquez <odejesush@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738349}

This CL breaks AddPeopleStrings in browser string provider into smaller
string providing functions that provide for specific People settings
polymer elements.  Shared components will have their providing function
inside the shared string provider, and OS exclusive components will have
their provided function inside the os string provider.
IDS_SETTINGS_SYNC_SIGN_IN_PROMPT_WITH_ACCOUNT in both chromium settings
and chrome os settings strings are deleted since they are no longer
used, and so is IDS_SETTINGS_USE_DEFAULT_SETTINGS. Also, people
strings that were located incorrectly in AddPeopleStrings are
moved to the correct function.

Bug: 967888
Change-Id: I2f8b94e6903a0ed413786195f467d8d49c7942ee
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2026111
Commit-Queue: Regan Hsu <hsuregan@chromium.org>
Reviewed-by: Demetrios Papadopoulos <dpapad@chromium.org>
Reviewed-by: Kyle Horimoto <khorimoto@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738348}

I previously thought that the cursor was always part of the captured
screen on Mac, but it turns out that's only true if curtain mode is
active. This CL makes creating of a composing capturer conditional
on the curtain mode setting, which should give the correct beheviour
for both remote support and uncurtained remote access.

Bug: 1048774
Change-Id: I28cc01b0b615794e06cd31308c10a40af550e0e1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2036546
Commit-Queue: Jamie Walch <jamiewalch@chromium.org>
Reviewed-by: Joe Downing <joedow@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738347}

If the language model doesn't return a target language, fall back to UI
language and then accept language list. In the very worst case default
to a target of English. Record uma on the target language origin.

Bug: 1041387
Change-Id: Ifd42f6a30fa0ba01cd04f714d595661bcedd2a47
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2032513Reviewed-by: Alexandre Frechette <frechette@chromium.org>
Reviewed-by: Alexei Svitkine <asvitkine@chromium.org>
Reviewed-by: anthonyvd <anthonyvd@chromium.org>
Commit-Queue: Megan Jablonski <megjablon@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738346}

By default, on POSIX, FileEnumerator keeps track of previously
visited directories by inode. On Fuchsia, this was not working
properly because all files under the /pkg directory have inode 1.

To work around this issue, the SHOW_SYM_LINKS flag was used. This flag
makes FileEnumerator always parse through SymLinks on POSIX platforms.

Since Fuchsia does not support SymLinks, this CL makes the
SHOW_SYM_LINKS flag have no effect on Fuchsia and tweaks the
FileEnumerator behavior to never keep track of inodes on Fuchsia.

Bug: 1033230
Change-Id: I2bb7a25b3e49f394323bfb14e1e5132dda92ca12
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967970
Commit-Queue: Fabrice de Gans-Riberi <fdegans@chromium.org>
Reviewed-by: Wez <wez@chromium.org>
Reviewed-by: Ryan Hamilton <rch@chromium.org>
Cr-Commit-Position: refs/heads/master@{#738345}