GitLab

TBR=yzshen@chromium.org
NOTRY=true

Change-Id: I5e5362aa6c5547b3a93fba48020b42c21f0ab6a1
Reviewed-on: https://chromium-review.googlesource.com/575349Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Commit-Queue: John Abd-El-Malek <jam@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487225}

This CL migrates ArcTtsService.

BUG=672829
TEST=Ran try.

Change-Id: Ia635f234c7bd03ce19a2ffd2f78bc20bbe9ecc19
Reviewed-on: https://chromium-review.googlesource.com/572473
Commit-Queue: Hidehiko Abe <hidehiko@chromium.org>
Reviewed-by: Luis Hector Chavez <lhchavez@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487224}

Bug: none
Change-Id: Ie2b12dd21b061282111dec4f0fd1654e7ff695c8
Reviewed-on: https://chromium-review.googlesource.com/572824Reviewed-by: Scott Violet <sky@chromium.org>
Commit-Queue: Evan Stade <estade@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487223}

The old AppListView did not respond to scroll events from PEEKING mode.
Now the AppListView transitions from PEEKING to FULLSCREEN_ALL_APPS when
a scroll event (mousewheel or touchpad gesture) is received via
AppListView::HandleScroll.

I had to make a fix to ensure that events are properly propogating, and
I had to explicitly pass MouseWheel events from the SearchBoxView
widget.

New Tests:
MouseWheel Scroll events from the SearchBoxView from PEEKING.
MouseWheel scroll events from the AppListView shield from PEEKING.
Ditto, from other states.

BUG=739408

Review-Url: https://codereview.chromium.org/2982453002
Cr-Commit-Position: refs/heads/master@{#487222}

Bug: 729183
Change-Id: I4e5b3251048d57d592bb82e933d23adab146a31d
Reviewed-on: https://chromium-review.googlesource.com/568662
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Ilya Sherman <isherman@chromium.org>
Reviewed-by: Eric Karl <ericrk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487221}

Anticipating ~140 layout test diffs:
  https://skia-review.googlesource.com/c/20968/13

Change-Id: Ia5de315a6b04249ad5042add8f193da7c4db7269
Reviewed-on: https://chromium-review.googlesource.com/574871Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487220}

chromium.android.fyi: Memory Infra Tester
chromium.fyi: Site Isolation Android

Bug: 670879
Change-Id: Ia7baa91881a626dcd0a743d3f93510c27772d3bb
Reviewed-on: https://chromium-review.googlesource.com/572544Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: Benjamin Pastene <bpastene@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487219}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/6e3bfad0debc..64ca2b5ad4d4

$ git log 6e3bfad0d..64ca2b5ad --date=short --no-merges --format='%ad %ae %s'
2017-07-17 sullivan Add a 'benchmark' parametr to alerts API, to allow for limiting to just one benchmark. This makes it possible to get a much longer timeframe of data.

Created with:
  roll-dep src/third_party/catapult


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

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls


CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel
TBR=sullivan@chromium.org

Change-Id: I3f9cf9ac78883357558206c02b1bde2f3d9d7fde
Reviewed-on: https://chromium-review.googlesource.com/574764
Reviewed-by: <catapult-deps-roller@chromium.org>
Commit-Queue: <catapult-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487218}

This CL changes the clear (X button) in the upper right of a cr_dialog
from a paper-icon-button to a paper-icon-button-light. It also moves
some of the paper-icon-button-light styling from settings to cr CSS so
that it can apply in cr_*.

Bug: 736891
Change-Id: Idc211b5dc42d18261d5e0906fea6f5d385421f0e
Reviewed-on: https://chromium-review.googlesource.com/567728Reviewed-by: Scott Chen <scottchen@chromium.org>
Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Commit-Queue: Dave Schuyler <dschuyler@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487217}

This updates the change picture screen to use the new layout.

It does not update the profile pictures, that will be done behind
a flag in a followup.

Bug: 730031
Change-Id: I0645eb1cd034188f438062cc7cba8662e81f6e91
Reviewed-on: https://chromium-review.googlesource.com/572302
Commit-Queue: Steven Bennetts <stevenjb@chromium.org>
Reviewed-by: Alexander Alekseev <alemate@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487216}

This removes the GET_ACCOUNTS permission from the manifest which is no
longer needed. Also removes some permissions-related code.

Change-Id: I111a97b433ef0c9129a5486b1a22662725ea5c8e
Reviewed-on: https://chromium-review.googlesource.com/572600Reviewed-by: Joe Downing <joedow@chromium.org>
Commit-Queue: Lambros Lambrou <lambroslambrou@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487215}

Surface synchronization unit tests use CompositorFrameSinkSupport and
FrameSinkManager which makes them part of the frame_sinks layer.
This CL moves the file to components/viz/service/frame_sinks,
updating includes, namespaces and forward declarations as appropriate.

Bug: 722935
TBR: piman@chromium.org
Change-Id: Ic8395dd78dfdcbe10caaef8377998843d01bad0f
Reviewed-on: https://chromium-review.googlesource.com/572385Reviewed-by: Antoine Labour <piman@chromium.org>
Reviewed-by: Robert Kroeger <rjkroege@chromium.org>
Commit-Queue: Fady Samuel <fsamuel@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487214}

Release with 12 shards and debug with 20.

A release shard:
https://chromium-swarm.appspot.com/task?id=374bad1c5c50b910

A debug shard:
https://chromium-swarm.appspot.com/task?id=374bc4c487af1810

Bug: 670879
Change-Id: I3a9a0e2936269df5a8c72212cc2233803656ef3b
Reviewed-on: https://chromium-review.googlesource.com/567370Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: Benjamin Pastene <bpastene@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487213}

Make TabletPowerButtonController disable the touchscreen
when powerd reports that the display was turned off
automatically (typically due to user inactivity).

This is intended to remove an inconsistency from the user's
perspective where the touchscreen turns the screen on when
the device is awake but in a screen-off state due to
inactivity, but doesn't wake the device when it's in S3
(i.e. suspended, asleep).

BUG=743291
TEST=manual: run set_short_powerd_timeouts, then:
     a) tap power button and verify touchscreen is disabled.
        after tapping power button again, screen turns on
        and touchscreen works.
     b) wait 20 seconds for screen to turn off due to
        inactivity and verify touchscreen is disabled. after
        hitting a key, screen turns on and touchscreen works
        again.
     c) manually decrease brightness to 0, wait five
        seconds, and then verify that touchscreen increases
        backlight brightness to min-visible level

Change-Id: I16c9769f587fe91bf2e6c3e68d06ccb83b299294
Reviewed-on: https://chromium-review.googlesource.com/572766Reviewed-by: Qiang(Joe) Xu <warx@chromium.org>
Commit-Queue: Dan Erat <derat@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487212}

PasswordStoreWin has a complex way of retrieving stored logins, due to an
implicit import from IE7 stored logins in case of Chrome own password store
being empty for a given site. This CL adds histograms to see whether this
importing from IE7 is used sufficiently often to help the user to justify the
code complexity.

Bug: 456119
Change-Id: I46e8ee4977cbf1f5bdc99638596e28ae4ef860bc
Reviewed-on: https://chromium-review.googlesource.com/571228Reviewed-by: Robert Kaplow <rkaplow@chromium.org>
Reviewed-by: Balazs Engedy <engedy@chromium.org>
Commit-Queue: Vaclav Brozek <vabr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487211}

Specifically if an ancestor of the focused window moves to a different
display, then the FocusController is left in a base state. This is
because the FocusController is per display.

BUG=none
TEST=covered by test

Change-Id: I3fbcaf91f5991bd5bb3a8aa7113503937e4c97a2
Reviewed-on: https://chromium-review.googlesource.com/574927Reviewed-by: Elliot Glaysher <erg@chromium.org>
Commit-Queue: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487210}

Checks browser history after navigating to two pages in the VR browser. Both
the VR browser and regular "2D" Chrome are supposed to share history, which
up until now was only tested manually.

Bug: 728779
Change-Id: I48b3c075e9a043c9a2fadb67e71e9080cb9f7e9b
Reviewed-on: https://chromium-review.googlesource.com/570120
Commit-Queue: Brian Sheedy <bsheedy@chromium.org>
Reviewed-by: Tommy Nyquist <nyquist@chromium.org>
Reviewed-by: Michael Thiessen <mthiesse@chromium.org>
Reviewed-by: Theresa <twellington@chromium.org>
Reviewed-by: David Trainor <dtrainor@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487209}

BUG=706512

Change-Id: I51f6817dd9267ce842d77e8a85c5c2e15dacba22
Reviewed-on: https://chromium-review.googlesource.com/574436Reviewed-by: Nico Weber <thakis@chromium.org>
Commit-Queue: Robert Liao <robliao@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487208}

Previously, this was only done for /var/log/messages.

BUG=715263
R=afakhry@chromium.org

Change-Id: I1ece80c22d9a74f2ab28eab6c63fd3b6d5772ba7
Reviewed-on: https://chromium-review.googlesource.com/572582Reviewed-by: Ahmed Fakhry <afakhry@chromium.org>
Commit-Queue: Simon Que <sque@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487207}

* Removes protocol compatibility tests since all the baselines haven't
been maintained for a while and this work won't be done for a while.
* Removes a couple of console tests that have been migrated to the new
test infrastructure (they were copied in an earlier CL)

Bug: 667560
Change-Id: Ie73e4aeb0e238aa63cda0a9a96165480dfa6a03d
Reviewed-on: https://chromium-review.googlesource.com/572331Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Commit-Queue: Will Chen <chenwilliam@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487206}

The comment in .mojom file is fixed.

BUG=731758

Change-Id: I110c4220e6fa2f7bd49ed9786d8f21015c955e95
Reviewed-on: https://chromium-review.googlesource.com/571031Reviewed-by: Amirhossein Simjour <asimjour@chromium.org>
Reviewed-by: Ken Rockot <rockot@chromium.org>
Reviewed-by: Emily Stark <estark@chromium.org>
Commit-Queue: Amirhossein Simjour <asimjour@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487205}

Just moving the switch to a common location.

TBR=bradnelson@chromium.org

Bug: 742646
Change-Id: Ia537de1f26cf3a065d97f4fe2dca0256ab3ce0c5
Reviewed-on: https://chromium-review.googlesource.com/570888
Commit-Queue: Michael Giuffrida <michaelpg@chromium.org>
Reviewed-by: Michael Giuffrida <michaelpg@chromium.org>
Reviewed-by: Ben Wells <benwells@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487204}

This CL removes the usage of MessageLoop and TestBrowserThread in
ExtensionPrefsTest, both of which should not be used directly in tests, and
replaces them with a TestBrowserThreadBundle instance. This also (implicitly) 
gives ExtensionPrefsTest a TaskScheduler instance.

BUG=272091

Change-Id: I6d1c3594cd2f28488e8d5d44a31967d9a3a507f6
Reviewed-on: https://chromium-review.googlesource.com/570922Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Karan Bhatia <karandeepb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487203}

Bug: None
Change-Id: I87fcd72f0c29e7b0dedeb8749cc9ad6d1b990866
Reviewed-on: https://chromium-review.googlesource.com/574947Reviewed-by: Tibor Goldschwendt <tiborg@chromium.org>
Commit-Queue: Biao She <bshe@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487202}

Bug: 742727
Change-Id: Ia6bacffad903f7fd9d5a9e7d460dde293214b671
Reviewed-on: https://chromium-review.googlesource.com/572205Reviewed-by: Scott Chen <scottchen@chromium.org>
Commit-Queue: Demetrios Papadopoulos <dpapad@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487201}

This change is part of the work to replace BrowserThread::FILE instances
with TaskScheduler friendly constructs.

In our case, we will now provide a single threaded task runner which
allows blocking I/O to the ChromotingHostContext passed to the
Native Message Host.  Many of the classes in Chromoting expect thread
affinity so I did not want to use a sequenced task runner for this
change to reduce its potential impact.

Bug: 
Change-Id: I93aa30b2fb35165200aed0f6e926583ec077e635
Reviewed-on: https://chromium-review.googlesource.com/572340Reviewed-by: Sergey Ulanov <sergeyu@chromium.org>
Commit-Queue: Joe Downing <joedow@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487200}

1) Classify an intent as a VR intent only if the platform supports VR mode.
2) Return to Daydream home screen if VrShell is disabled. The splash screen
depends on VrShell features so the easiest thing to do is bail. This can only
happen if VrShell is manually disabled from chrome://flags and should be rare.

Bug: 
Change-Id: Ie14c164d6ac1c74de041e24073349c498df5f6a7
Reviewed-on: https://chromium-review.googlesource.com/572620Reviewed-by: Michael Thiessen <mthiesse@chromium.org>
Commit-Queue: Yash Malik <ymalik@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487199}

TBR=zmo@chromium.org
NOTRY=true

BUG=668223

Change-Id: Ibf09100c74fa4ec2c8e7738f7f977153f1816c69
Reviewed-on: https://chromium-review.googlesource.com/574929Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487198}

If shared SchedulerWorkers are released before joining, they are
effectively detached and may outlive the
SchedulerSingleThreadTaskRunnerManager.

BUG=735701

Change-Id: Ie83b50b99698a4059041454f579f4a2c1fbaabc0
Reviewed-on: https://chromium-review.googlesource.com/572233Reviewed-by: Gabriel Charette <gab@chromium.org>
Commit-Queue: Robert Liao <robliao@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487197}

Bug: 672263
Change-Id: I9c9ad8b4925f109718bef273b489bb9bbcacc8ca
Reviewed-on: https://chromium-review.googlesource.com/572212
Commit-Queue: Kyle Horimoto <khorimoto@chromium.org>
Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487196}

There doesn't appear to be a reason all BlobReaders should use the same
task runners for their file operations (and furthermore it doesn't seem
to matter if the task runner even for a single blob reader is
sequenced), so this change BlobReader to create its own TaskRunner.

This assumes that blobs aren't being read for no reason, so uses
USER_VISIBLE as priority for all blob reads.

Ideally the creation of the TaskRunner would be pushed even further down
into the stack (maybe all the way down to net::FileStream), but at that
point it doesn't seem safe to assume anymore that all users of
FileStream want the same priority and shutdown behavior.

Shutdown behavior for these tasks should not matter, as BlobReader is only
used on the IO thread, which gets shutdown before the task scheduler is
shut down, making it impossible for tasks to get posted after shutdown.

Change-Id: If44f32aad5d566d04fc845113f05ef6044592aa4
Reviewed-on: https://chromium-review.googlesource.com/570945Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Reviewed-by: Daniel Murphy <dmurph@chromium.org>
Commit-Queue: Marijn Kruisselbrink <mek@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487195}

BUG=703369

Review-Url: https://codereview.chromium.org/2975413002
Cr-Commit-Position: refs/heads/master@{#487194}

TBR=twellington

Bug: 744710
Change-Id: Ib01c1b081b5c34641aeae8dae8c4234d38472837
Reviewed-on: https://chromium-review.googlesource.com/574760Reviewed-by: Yoland Yan <yolandyan@chromium.org>
Commit-Queue: Yoland Yan <yolandyan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487193}

[CrOS Tether] Separate network stack code from MasterHostScanCache and convert MasterHostScanCache into a manager of both the networking and persistent caches.

This is the third step of persisting host scan results so that we can recover from crashes. Steps listed below:
(1) Create a HostScanCache interface and move the concrete implementation to MasterHostScanCache.
(2) Create a HostScanCache implementation which persists cache entries to user prefs.
(3 - this CL) Move code which adds entries in the network stack to another HostScanCache implementation, and have MasterHostScanCache manage the other two HostScanCache implementations.

After this CL, host scan results will be restored after a crash, but the active host will not yet be restored. That will occur in a follow-up CL.

Bug: 672263, 737273
Change-Id: I594ad829c5c5b010accc6089f9575aef9afb0cae
Reviewed-on: https://chromium-review.googlesource.com/570925Reviewed-by: Ryan Hansberry <hansberry@chromium.org>
Commit-Queue: Ryan Hansberry <hansberry@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487192}

This reverts commit 1b600061.

Reason for revert:
Causes a startup crash with chrome --mash, possibly because mash does not have a DeviceDataManager. You might want to check with sadrul@ or rjkroege@ about whether there's a good way to do this under mash.

Failing build (chromeos-side waterfall):
https://luci-milo.appspot.com/buildbot/chromeos.chrome/tricky-tot-chrome-pfq-informational/5205

Stack (from running manually with chrome --mash on Linux):

[60089:60089:0717/114019.863351:FATAL:device_data_manager.cc(84)] Check failed: instance_. DeviceDataManager was not created.
#0 0x7f06a03468bc base::debug::StackTrace::StackTrace()
#1 0x7f06a036a121 logging::LogMessage::~LogMessage()
#2 0x7f069aeb08ff ui::DeviceDataManager::GetInstance()
#3 0x55f183402a41 chromeos::LoginDisplayHostImpl::LoginDisplayHostImpl()
#4 0x55f1834066ad chromeos::ShowLoginWizard()
#5 0x55f1833e4bf7 chromeos::ChromeSessionManager::Initialize()
#6 0x55f1832f9cb4 chromeos::ChromeBrowserMainPartsChromeos::PostProfileInit()
#7 0x55f1837e7aa2 ChromeBrowserMainParts::PreMainMessageLoopRunImpl()
#8 0x55f1837e709d ChromeBrowserMainParts::PreMainMessageLoopRun()
#9 0x55f1832f8f4a chromeos::ChromeBrowserMainPartsChromeos::PreMainMessageLoopRun()
#10 0x7f069d9e5081 content::BrowserMainLoop::PreMainMessageLoopRun()
#11 0x7f069dde6407 content::StartupTaskRunner::RunAllTasksNow()
#12 0x7f069d9e353b content::BrowserMainLoop::CreateStartupTasks()
#13 0x7f069d9e7a72 content::BrowserMainRunnerImpl::Initialize()
#14 0x7f069d9e0b57 content::BrowserMain()
#15 0x7f069e1e8462 content::ContentMainRunnerImpl::Run()
#16 0x7f06a0890d39 service_manager::Main()
#17 0x7f069e1e7384 content::ContentMain()
#18 0x55f182e8132f ChromeMain
#19 0x7f069454af45 __libc_start_main
#20 0x55f182e81194 <unknown>


Original change's description:
> Listen to changes to touch input devices
> 
> In https://codereview.chromium.org/2964823002 the OobeDisplayChooser
> started using the DeviceDataManager to look for touchscreen devices when
> searching for a good primary display to use during OOBE.
> 
> On device cold boot the DeviceDataManager has not yet found any
> touchscreen devices at the time OobeUi::ShowOobeUI() is called (likely
> due to lower level systems not being fully initialized).
> 
> This CL make LoginDisplayHostImpl an observer of changes to connected
> touchscreen devices, re-triggering the OobeDisplayChooser when the
> DeviceDataManager is notified of the connected touchscreens. This
> overcomes the timing issues on cold boot.
> 
> Bug: 738885
> Change-Id: Iae488ddc9428b7c5e74d36cf18e35ba3d1235bbd
> Reviewed-on: https://chromium-review.googlesource.com/569958
> Reviewed-by: Jacob Dufault <jdufault@chromium.org>
> Commit-Queue: Felix Ekblom <felixe@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#487007}

TBR=alemate@chromium.org,jdufault@chromium.org,felixe@chromium.org

# Not skipping CQ checks because original CL landed > 1 day ago.

Bug: 738885
Change-Id: If31322734e679bbb1f4eef0a9aa802d34263cba4
Reviewed-on: https://chromium-review.googlesource.com/574731Reviewed-by: James Cook <jamescook@chromium.org>
Commit-Queue: James Cook <jamescook@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487191}

Previously, vr_shell_resources.grdp contained a condition that it be
included if enable_vr, but in browser_resources.grd we only included
this part if is_android. This allows vr_common_unittests to build on
linux when enable_vr is true.

We will eventually want to, for tests, use generic assets for the
controller (see crbug.com/743687), but this lets us test on linux
in the interim.

This change also relocates our resources to a more appropriate
location.

Bug: None
Change-Id: Ic17545487555595a867913fa58b74e559b26391d
Reviewed-on: https://chromium-review.googlesource.com/572943
Commit-Queue: Ian Vollick <vollick@chromium.org>
Reviewed-by: Tibor Goldschwendt <tiborg@chromium.org>
Reviewed-by: Michael Thiessen <mthiesse@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487190}

This CL migrates ArcStorageManager.

BUG=672829
TEST=Ran try.

Change-Id: If91bab2be503825fa3c10ee71053acf52c504057
Reviewed-on: https://chromium-review.googlesource.com/572886
Commit-Queue: Hidehiko Abe <hidehiko@chromium.org>
Reviewed-by: Luis Hector Chavez <lhchavez@chromium.org>
Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487189}

Merges server-side changes 161104822, 161260482 by dahollings, birenroy.

Bug: 488484
Change-Id: I7888fe8b6901132a83086c471c7f3e3935999679
Reviewed-on: https://chromium-review.googlesource.com/568857
Commit-Queue: Bence Béky <bnc@chromium.org>
Reviewed-by: Bence Béky <bnc@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487188}

- treat the Docs content editable as a root (note that the role=textbox has a parent that is also richly editable; the only way to distinguish is based on focus. Could have used role as well, but focused state seems more relevant). This means that jumping to the next link while ChromeVox range is inside of a content editable wraps inside the content editable. Also, control duplicate/overlapping output by suppressing the usual output in navigation when the node is richly editable.
- fix an issue where we were clearing text edit handler state on focus. We actually set focus after every jump command, so this resulted in us speaking "deleted" after every jump command.
- handle another state transition in AutomationRichEditableText. This one is non-overlapping selections from prev to cur. These occur as a result of jump commands among other possibilities.
TEST=jump around in a complex document.

Bug: 
Change-Id: I1f5b555e143a2998576b746671761eeb0705ede1
Reviewed-on: https://chromium-review.googlesource.com/574787Reviewed-by: Dominic Mazzoni <dmazzoni@chromium.org>
Commit-Queue: David Tseng <dtseng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487187}

Style nit.

R=piman@chromium.org
TBR=piman@chromium.org

Change-Id: I96477350fae2688d31143086797dc3e35d1c771c
Reviewed-on: https://chromium-review.googlesource.com/572330Reviewed-by: Antoine Labour <piman@chromium.org>
Reviewed-by: Justin Schuh <jschuh@chromium.org>
Commit-Queue: Justin Schuh <jschuh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487186}