GitLab

The test was originally disabled due to https://crbug.com/101984. The
flakiness occurred for Vista builds only, and the bug has not had a
significant update since 2012.

Bug: 101984
Change-Id: Ifadf55ae2d867e816db2795c9a7eaed0377cc276
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1497128
Auto-Submit: Aran Gilman <gilmanmh@google.com>
Reviewed-by: David Tseng <dtseng@chromium.org>
Commit-Queue: Aran Gilman <gilmanmh@google.com>
Cr-Commit-Position: refs/heads/master@{#638651}

The go/ link doesn't have an 's' at the end.

Change-Id: Iae22758ebcc448ac8b86c02c735ba9f5a4ca5719
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506761
Auto-Submit: David Benjamin <davidben@chromium.org>
Reviewed-by: Sorin Jianu <sorin@chromium.org>
Commit-Queue: David Benjamin <davidben@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638650}

Currently we need to notify ImageElementTiming whenever an element will
be destroyed, so that it can be removed from the HashSet of images that
have been seen by ImageElementTiming. This CL adds unittests for this
behavior.

Bug: 879270
Change-Id: I7719b9c2bfc594cbf4070b1fe66c9e5325a3a552
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506660Reviewed-by: Steve Kobes <skobes@chromium.org>
Commit-Queue: Nicolás Peña Moreno <npm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638649}

ScopedTaskEnvironment::NextScheduledRunTime was returning Now()
(aka now_ticks_) because sequence_manager is idle, but then
ScopedTaskEnvironment::MaybeFastForwardToNextTask DCHECKS that now_ticks_
is strictly < run_time, which was causing test failures. Now that
ScopedTaskEnviornment is fixed we can re-enable these tests.

Bug: 935651
Change-Id: I5a0cb2967001995a54f2245bac4e5f9022514001
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1500018Reviewed-by: David Tseng <dtseng@chromium.org>
Commit-Queue: Katie Dektar <katie@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638648}

We use View.getGlobalVisibleRect, and the transform and viewport from
functor to compute info for the viewport:
* the tile memory budget, which ideally is computed from the size of
  the visible portion of webview
* the viewport and transform used to raster the right tiles at the
  right scale

We usually use getGlobalVisibleRect and transform from functor
together if is_layer is false, making the assumption that they both
transform between the window space and view space. This turns out not
to be true.

Instead, use getGlobalVisibleRect to compute the tile budget. And
always use the viewport and transform from functor together for
tile raster, since they are guaranteed to work together.

Change-Id: I4d1346501dab14dbb7b7ca160e0c88fee64aa3ca
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506759
Auto-Submit: Bo <boliu@chromium.org>
Reviewed-by: Eric Karl <ericrk@chromium.org>
Reviewed-by: Tobias Sargeant <tobiasjs@chromium.org>
Commit-Queue: Bo <boliu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638647}

This CL adds TCP and HTTP timing UMA on every successful request to the
client config service. This UMA will be used server side to compare with
other metrics for evaluating network quality.

Bug: 939121
Change-Id: I84d7867d03f16a57e4131e0043fbc922d50ff187
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506442Reviewed-by: Robert Kaplow <rkaplow@chromium.org>
Reviewed-by: Tarun Bansal <tbansal@chromium.org>
Commit-Queue: Robert Ogden <robertogden@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638646}

https://android.googlesource.com/platform/external/perfetto.git/+log/2d4d2a3cf7b9..356474278798


git log 2d4d2a3cf7b9..356474278798 --date=short --no-merges --format='%ad %ae %s'
2019-03-07 fmayer@google.com Improve error message if perfetto invocation fails.
2019-03-07 fmayer@google.com Read from socket to prevent clear notification to service.


Created with:
  gclient setdep -r src/third_party/perfetto@356474278798

The AutoRoll server is located here: https://autoroll.skia.org/r/perfetto-chromium-autoroll

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

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


TBR=perfetto-bugs@google.com

Change-Id: I496e166cba1724b1d47675e842f891db54d5cb71
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1508914Reviewed-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@{#638645}

VDA implementations require config.container_color_space and
config.target_color_space information to be set to provide correct output. For
WebRTC frames that carry this information for VP9.2, we can reinitialize VDA
with this info for the correct colors.

Bug: 939387
Change-Id: Ib2ef4c7fdfbeef5e6f25a887e32a5c9c4c2e170e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1476065
Commit-Queue: Emircan Uysaler <emircan@chromium.org>
Reviewed-by: Dan Sanders <sandersd@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638644}

SecurityExploitBrowserTest.AttemptDuplicateRenderWidgetHost is flaky
on all platorms.

TBR=boliu@chromium.org

Bug: 939338
Change-Id: I21f13d8651b52647f61a0bb8f92d943ff96e7def
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1508894Reviewed-by: Etienne Pierre-Doray <etiennep@chromium.org>
Commit-Queue: Etienne Pierre-Doray <etiennep@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638643}

Bug: 859172
Change-Id: I94d6400a9b0ef983ace670c3adfad68b71c75819
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506752
Auto-Submit: Elly Fong-Jones <ellyjones@chromium.org>
Commit-Queue: Alex Newcomer <newcomer@chromium.org>
Reviewed-by: Alex Newcomer <newcomer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638642}

If a navigation entry was marked as skippable because it added an entry
using pushState, then a user gesture on the page should reset the skippable
bit. If the pushState was done after the user gesture, then the entry would
not have been marked as skippable to begin with. This change will make the
behavior independent of the ordering between user gesture and pushState.

content_browsertests --gtest_filter=
NavigationControllerHistoryInterventionBrowserTest.
OnUserGestureResetSameDocumentEntriesSkipFlag
content_browsertests --gtest_filter=
NavigationControllerHistoryInterventionBrowserTest.
OnUserGestureDoNotResetDifferentDocumentEntrySkipFlag

Test: 
Bug: 937423
Change-Id: Ie1e55815163ca84950ed7e04fbfa55c06f86c2cf
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1499834
Commit-Queue: Shivani Sharma <shivanisha@chromium.org>
Reviewed-by: Charlie Reis <creis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638641}

This adds the original_url to the information returned when a page is
removed from the offline page model.

See bug for motivation.

Bug: 884291
Change-Id: I25d11e58347379e06c4a0de8a1cb5b1830cdf279
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1473830
Commit-Queue: Dan H <harringtond@google.com>
Reviewed-by: Carlos Knippschild <carlosk@chromium.org>
Reviewed-by: Filip Gorski <fgorski@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638640}

ThemeProperties::COLOR_BOOKMARK_BAR_INSTRUCTIONS_TEXT is no longer in use,
because BookmarkBarInstructionView is removed. However, it still remains,
so it is cleaned up.

Bug: NONE
Change-Id: I7ce14646fafaf07fc426aa96d1b4dab073b51ca4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506679Reviewed-by: Peter Kasting <pkasting@chromium.org>
Commit-Queue: Peter Kasting <pkasting@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638639}

This reverts commit ed15ce53.

Previous change prevented promos and search suggestions from appearing properly.

Required manual resolution for merge conflicts.

Bug: 770640
Change-Id: I37add613a23567d57912623c02441cdb68d261e0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1507249
Commit-Queue: Kyle Milka <kmilka@chromium.org>
Reviewed-by: Kristi Park <kristipark@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638638}

Before this CL, GetPendingItemIndex() used to return last committed item
index for the new navigations (none-reload, none-back-forward).
Presumably this was done before CRWSessionController implemented
pendingItemIndex support, when CRWSessionController immediately changed committed item
during back-forward navigations. pendingItemIndex support design doc: https://docs.google.com/document/d/1GvsaoNtMO7d87FzoysOW5GqH6XErI5g2cinKhkPk7bQ/edit

This CL returns -1 for new navigations, which matches //content.

There are 3 places where GetPendingItemIndex is used outside of ios/web:
 - IOSLiveTab::GetPendingEntryIndex (new implementation matches content,
   which is the right approach)
 - GetPossiblyPendingItemAtIndex (new implementation should have no
   effect because index should be in [0;size] bounds, but this is just
   an assumption and it's actually unclear if GetPossiblyPendingItemAtIndex
   relied on undocumented and incorrect behavior).
 - ReadingListWebStateObserver::LoadOfflineReadingListEntry (new
   implementation should make the whole block of code obsolete, but
   it should be fine to leave it as it is because
   ReadingListWebStateObserver is going to be removed soon).

Bug: 665189, 899827
Change-Id: I9538b693c924790620140b37d6c4fea6cce26fde
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1504231
Commit-Queue: Eugene But <eugenebut@chromium.org>
Reviewed-by: Gauthier Ambard <gambard@chromium.org>
Reviewed-by: Olivier Robin <olivierrobin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638637}

Bug: 906453
Change-Id: Ice07136126ce35d128054276e86ac641aba2c882
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1507235Reviewed-by: enne <enne@chromium.org>
Commit-Queue: ccameron <ccameron@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638636}

Bug: None
Change-Id: I37c84f9367743b3d2f758eaf9bf0d1094b65aad6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1507201Reviewed-by: Yusuf Ozuysal <yusufo@chromium.org>
Commit-Queue: Wei-Yin Chen (陳威尹) <wychen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638635}

Gets a new parameter (supported_card_bin_ranges) from the
GetDetailsForSaveCardResponse proto, which contains a list of pair of
numbers representing ranges of supported card bins. If a card doesn't
fall under any of the ranges, the card won't be offered upload save and
will be saved locally instead.

Change-Id: I17a8742f0d20788131ffb2af29d8a6bcb21c09f5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1496251Reviewed-by: Sebastien Seguin-Gagnon <sebsg@chromium.org>
Reviewed-by: Elly Fong-Jones <ellyjones@chromium.org>
Commit-Queue: Anne Lim <annelim@google.com>
Cr-Commit-Position: refs/heads/master@{#638634}

Bug: 766077
Change-Id: Ie52fc2eea6e96580bb064ac3b9c877a0ef96ff2a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1508466Reviewed-by: Tim Volodine <timvolodine@chromium.org>
Commit-Queue: Mugdha Lakhani <nator@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638633}

ConnectJobs no longer care about group names, so having them store the
name is a bit of a layering violation.

ConnectJob group names were used for a NetLog event and by the socket
pools to locate the group a ConnectJob belonged to. To resolve the first
issue, group names are now logged by a NetLog event by the socket pools
before starting each ConnectJob (which will require an update in the
net log viewer code). To resolve the second issue, Groups are now
ConnectJob::Delegates for all the ConnectJobs associated with the group.

Beyond fixing the layering violation, this CL serves two purposes: It
will make it easier to change group_name from a string to a struct, and
it will make it easier to make CommonConnectJobParameters a per-pool
object.

Bug: 927100, 533571
Change-Id: Ife570e1277ef18b4d916eb6d3ab91a9fccf2bed8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506567Reviewed-by: David Benjamin <davidben@chromium.org>
Commit-Queue: Matt Menke <mmenke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638632}

Record for each window.open(), href link activation, and location=xxx. Those
calls should be mutually exclusive.

For this metric to be accurate, it has to assume that "the navigation was
initiated in a frame with transient gesture" == "the navigation was caused
by this gesture", which can be wrong. But it may just work in practice and
it seems to be a general assumption we have been making throughout Chrome.

Bug: 864147
Change-Id: If7f75bd8d17e1d2ea572607e6f8f4149c78222fd
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1488111
Commit-Queue: Yao Xiao <yaoxia@chromium.org>
Reviewed-by: Bryan McQuade <bmcquade@chromium.org>
Reviewed-by: Camille Lamy <clamy@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Charlie Harrison <csharrison@chromium.org>
Reviewed-by: Mustaq Ahmed <mustaq@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638631}

The sync datatype is the last leftover that hasn't been wrapped in the
USS architecture, and could end up in the critical path to future
dismantling efforts.

The new controller has been simplified by instead introducing the
notion of model-readiness in SyncableService, which the USS codepath
honors. This makes the new controller a lot simpler, and is aligned
with our plans to avoid the notion of model readiness in controllers,
and narrow down the scope of DataTypeController::ReadyForStart() to
non-transient business logic (i.e. is a datatype allowed to start).

Sync integration tests are now parameterized to exercise both codepaths
in case the feature toggle must be used in the future as kill switch.

Bug: 870624,939329
Change-Id: I9d415eba7bd53f0fd9cf7bbfa00db98b7e916e1c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1507657Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Reviewed-by: Marc Treib <treib@chromium.org>
Commit-Queue: Mikel Astiz <mastiz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638630}

UpdateScoreDict is causing a crash when we are getting
keys and this could be caused by old data that does
not have all the keys. This hardens the checks when
loading the data.

BUG=934165

Change-Id: I35eafc9b2a232ff8cc79fd48b48de9cc09cfc65d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1508596Reviewed-by: Tommy Steimel <steimel@chromium.org>
Commit-Queue: Becca Hughes <beccahughes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638629}

This is the platform independent version of the notification trigger
which simply starts a timer. This requires Chrome to be running at the
time the notification should be displayed. If Chrome is not running, the
notification will be shown at the next startup.

A follow-up CL will override this behaviour for Android and use the
BackgroundTaskScheduler to wake up Chrome.

Bug: 891339
Change-Id: I679df47d725371aa481c30bcaa742f4760106e9c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1477840
Commit-Queue: Richard Knoll <knollr@chromium.org>
Reviewed-by: Peter Beverloo <peter@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638628}

Change client code to work as if IdentityManager::GetPrimaryAccount
where returning a CoreAccountInfo instead of an AccountInfo. Method
will be updated in a followup CL.

This CL was uploaded by git cl split.

Bug: 926204
Change-Id: Ic7e29b114fd39ee0c55cafb1a0e526eb1aa78244
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1499559Reviewed-by: Sergio Collazos <sczs@chromium.org>
Commit-Queue: Sylvain Defresne <sdefresne@chromium.org>
Auto-Submit: Sylvain Defresne <sdefresne@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638627}

When the input caret position is "AfterCharacter", the function should
look backwards for the result position in DOM. Hence, it should call
GetFirstPosition() instead of GetLastPosition().

Bug: 894651
Change-Id: If2c9e8346c38859d0b8674dd56f49fbc71aa3ccc
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506768Reviewed-by: Koji Ishii <kojii@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Commit-Queue: Xiaocheng Hu <xiaochengh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638626}

This patch is largely removing the use of 'suggestion' from a
generic MVC class.

Change-Id: I20e5932d0638da0489ed002ffa8555874bd10eba
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506206Reviewed-by: Theresa <twellington@chromium.org>
Commit-Queue: Matthew Jones <mdjones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638625}

This change adds accessibility content description for refine button

Bug: 939014
Change-Id: Ief2f5d1e5aa14498d03eda7827788dfd0d752ea6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506557
Commit-Queue: Ender <ender@google.com>
Auto-Submit: Ender <ender@google.com>
Reviewed-by: Theresa <twellington@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638624}

Needs https://android-review.googlesource.com/c/platform/external/perfetto/+/910574
to land and roll into Chrome before this can land.

R=ssid@chromium.org

Bug: 839084
Change-Id: I945eca9387c562a144af30dfee9f7759a23e3f49
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1500134
Commit-Queue: oysteine <oysteine@chromium.org>
Reviewed-by: ssid <ssid@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638623}

There are two icon properties, kAppIcon and kWindowIcon. For the
purposes of Ash, these are used interchangeably (kAppIcon is preferred
and kWindowIcon is a fallback). Thus we can skip sending kWindowIcon.
This should cut down on memory usage by half.

Further, the only place that the icon is used for single process
mash is in overview mode, which displays the icon at 24dip. We can
downscale the icon before sending it over. For tabbed browser windows,
that's 24dip*24dip instead of 192px * 192px.

Since the icons are set on the content window as well as the root
window, they are each copied to the window service twice. We do need
to transport the icons that are set on the root window, but not on
the content window. PropertyConverter doesn't distinguish on the window
type, so new keys are introduced for the root window: kAppIconSmallKey
and kAppIconLargeKey. These are the only ones that are sent to the WS.
They are only used in Mash. kAppIconKey and kWindowIconKey are retained
for use in the browser, e.g. by AppWindowLauncherItemController.
kAppIconLargeKey is only set by the task manager and used by
ShelfWindowWatcher for the shelf icon.

The net effect is that every browser window goes from allocating
192x192px four times, to 24x24dip once. According to chrome://tracing
this lops about 4MiB off malloc'd memory in the browser process when
there are 10 browser windows open.

Bug: 938105

Change-Id: I7bc04784ff3fb1d0857219eae2184542b1f14fd4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1504164Reviewed-by: Michael Wasserman <msw@chromium.org>
Reviewed-by: Tom Sepez <tsepez@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Auto-Submit: Evan Stade <estade@chromium.org>
Commit-Queue: Evan Stade <estade@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638622}

- New profile avatar icons are added to Chrome.
- The avatar selector is refreshed:
  - Only the new avatars are available for selection.
  - The icons are made round and bigger.
- Existing profiles will keep their current avatars, i.e.
  they don't need to be migrated.

Screenshots:
https://drive.google.com/file/d/1m5tDHR9k-2Z9ZkQfza-l7D3oZIsTgftu/view?usp=sharing
https://drive.google.com/file/d/11lzTzZJR_qBT4qziY_Z5lbgPgyrBHVBa/view?usp=sharing

Bug: 937834
Change-Id: Iffe141de2aa9d176690cc1caaed05afd76adccff
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1505942Reviewed-by: David Roger <droger@chromium.org>
Reviewed-by: Hector Carmona <hcarmona@chromium.org>
Commit-Queue: Thomas Tangl <tangltom@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638621}

Bug: 937007
Change-Id: Ibe088bf8b76446098520284a50ac84ec5140ce0a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1495801
Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
Reviewed-by: Carlos Knippschild <carlosk@chromium.org>
Reviewed-by: Xi Han <hanxi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638620}

Before this change, we do not capture mouse events to widget when
set/release pointer capture. It causes a bug that when mouse down on
main frame and set pointer capture, mouse still not capture to the main
frame widget because we neither set widget capture on main frame mouse
down nor set widget capture on pointer capture set.

This CL makes PEM::SetPointerCapture and ReleasePointerCapture returns
a boolean to shows whether pointer capture is changed, and we change
widget capture accordingly.

Change-Id: I1d571659237c81f8575d4556d725c22c8c2b862f
Bug: 936190, 919908
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1446500
Commit-Queue: David Bokan <bokan@chromium.org>
Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: David Bokan <bokan@chromium.org>
Reviewed-by: Navid Zolghadr <nzolghadr@chromium.org>
Reviewed-by: Mustaq Ahmed <mustaq@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638619}

We recently introduced a new AXMode that turns on image
labeling for a WebContents, and it's integrated into
BrowserAccessibilityState, the mechanism used to
set the global AXMode on most platforms.

On Chrome OS the mechanism is slightly different, using
the automation API, and we missed one place. Fix it
by merging the AXMode set for the automation API with
the image labeling flag if needed, and add a test for it.

Bug: 938637, 905419

Change-Id: I273b35e3ef2457207f8a5ff4482cfeae4178027d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1504695
Commit-Queue: Avi Drissman <avi@chromium.org>
Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: David Tseng <dtseng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638618}

Before this patch, if PaymentRequest.show() was called in a CCT, then
user opened the CCT page in Chrome, and a skip-UI payment app was slow
to respond to the `canmakepayment` event, then the activity to which
Chrome attached the payment UI would be gone by the time the payment UI
needs to be shown, which would cause a crash. The skip-UI bit is
important, because in this flow Chrome blocks UI interactions only after
launching the payment app (after the response to `canmakepayment`
event).

This patch waits to attach the payment UI to the current Chrome activity
via buildUI(chromeActivity) until immediately before blocking UI
interactions via mUI.show() or mUI.dimBackground().

After this patch, Chrome does not crash when PaymentRequest.show() was
called in a CCT for skip-UI flow and the user opened the CCT page in
Chrome before the payment app responded to the `canmakepayment` event.

Bug: 817073
Change-Id: I5ce50d000c2f64a60c62955d9e3e8f080498238d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506494
Commit-Queue: Rouslan Solomakhin <rouslan@chromium.org>
Reviewed-by: Danyao Wang <danyao@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638617}

This patch fixes disposition/misplacement of menu windows and let the
Wayland compositor decide how to place the windows if their pixels
cannot be shown on one display or if they are partly going to be hidden
because of being shown too close to some edges of a display.

First of all, Chromium positions windows using screen coordinates.
As long as Wayland does not provide global coordinates, the client
does not know the position of windows in screen coordinates and
always assumes that they are located on 0,0. What's left for us to do
is to translate bounds from screen coordinates to relative to parent
coordinates, which is easy to do.

Secondly, WaylandWindow re-positions nested menus back to the right side
of parent menus (if the size of a top level window is close or equal to the
work area of a display) and let Wayland compositor decide how to position
them based on calculated anchor rect, anchor and gravity. The exception
for this is based on the maximized state of a top level window.
If it's maximized, then the nested windows are not re positioned
and left on the left side of the parent menu windows.

Also, WaylandWindow must be able to recalculate origin back from
relative to parent to be relative to top level window aka on screen
location.

The unittests exercises all those mentioned cases and make sure
WaylandWindow is able to calculated bounds correctly (the values
are taken by manual usage of the Chromium browser and logs).

PS XdgVersionV5Test/WaylandWindowTest.AdjustPopupBounds/0 is skipped
as long as it does not support xdg positioner.

Test: XdgVersionV6Test/WaylandWindowTest.AdjustPopupBounds/0
Bug: 927280
Change-Id: I30573620562e7bc5dfd3a69487326bcda49fc6ed
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1474654Reviewed-by: Robert Kroeger <rjkroege@chromium.org>
Commit-Queue: Maksim Sisov <msisov@igalia.com>
Cr-Commit-Position: refs/heads/master@{#638616}

Before sending over to be logged by DevTools, filter out DOM nodes that
have a type attribute equal to "password", and that are not empty.

Bug: 934609
Change-Id: I147ad0c2bad13cc50394f4b5ff2f4bfb7293114b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1506498
Commit-Queue: Sebastien Lalancette <seblalancette@chromium.org>
Reviewed-by: Vadym Doroshenko <dvadym@chromium.org>
Reviewed-by: Mathieu Perreault <mathp@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638615}

Bug: 939160
Change-Id: I9371553975c56bb19450aed1ac12296708c0e412
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1507233Reviewed-by: Jacob Dufault <jdufault@chromium.org>
Commit-Queue: Evan Stade <estade@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638614}

Change-Id: Ide05b93e62e29bcf34e813022b1214f37af1c0be
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1498783Reviewed-by: Brian White <bcwhite@chromium.org>
Commit-Queue: Sebastien Lalancette <seblalancette@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638613}

Bug: 939148
Change-Id: If82c99bfdf0834589a818591502928212ad35e34
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1508694
Commit-Queue: Sam Maier <smaier@chromium.org>
Commit-Queue: Andrew Grieve <agrieve@chromium.org>
Auto-Submit: Sam Maier <smaier@chromium.org>
Reviewed-by: Andrew Grieve <agrieve@chromium.org>
Cr-Commit-Position: refs/heads/master@{#638612}