GitLab

JS may invoke Image.decode without capturing the Image object. If GC
kicks in, ImageLoader will be destroyed as no one owns it, which leads
to unresolved/unrejected Promise. The callback of
ChromeClient::RequestDecode should take a persistent reference instead
of weak reference to avoid ImageLoader GC.

BUG=881060
TEST=LayoutTest image-decode-promise-when-gc.html

Change-Id: Ief9c7956ddd0df81ea1f1643eb6f6c7bac946a7c
Reviewed-on: https://chromium-review.googlesource.com/1205798Reviewed-by: Nate Chapin <japhet@chromium.org>
Reviewed-by: Hiroshige Hayashizaki <hiroshige@chromium.org>
Reviewed-by: vmpstr <vmpstr@chromium.org>
Commit-Queue: Yuchen Liu <yucliu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589687}

Use ServiceWorkerContext::RunTask() to make sure that
ServiceWorkerContext* is kept alive while transitioning from
UI thread to IO thread.

Bug: 875376
Change-Id: Ib961931260b453e8a99828b8598d77c7821cff06
Reviewed-on: https://chromium-review.googlesource.com/1211971
Commit-Queue: Istiaque Ahmed <lazyboy@chromium.org>
Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589686}

Update "Mac FYI dEQP Release AMD" on the GPU FYI waterfall to look for
machines running OS 10.13.6.

Bug: chromium:881952
No-Try: True
Change-Id: I12e0f82f4cc29fdbe36f7f76ad5d945228eaf4a7
Reviewed-on: https://chromium-review.googlesource.com/1213704
Commit-Queue: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589685}

This reverts commit f17b2cd3.

Reason for revert: <INSERT REASONING HERE>

Original change's description:
> Update ResourceLoader to also fetch cached code for inline scripts
> 
> V8 also stores code caches for inline scripts in the html resources.
> This cl updates ResourceLoader to fetch cached code for inline scripts
> in MainResources.
> 
> Bug: chromium:880212, chromium:812168
> Change-Id: I822453b52897aac292444df20e50937d8198064d
> Reviewed-on: https://chromium-review.googlesource.com/1208391
> Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
> Commit-Queue: Mythri Alle <mythria@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#589137}

TBR=kinuko@chromium.org,mythria@chromium.org

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

Bug: chromium:880212, chromium:812168
Change-Id: I4e24815c37378a86e334c8fe88129dff39ee87dd
Reviewed-on: https://chromium-review.googlesource.com/1213094Reviewed-by: Mythri Alle <mythria@chromium.org>
Commit-Queue: Mythri Alle <mythria@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589684}

They need large core machines.
This is fine since auto_builder_dimensions is on.

Bug: 868213
TBR: tandrii
Change-Id: I3aa153e09afbd7e9ddc885441a09183f73727232
Reviewed-on: https://chromium-review.googlesource.com/1214220Reviewed-by: Ryan Tseng <hinoka@chromium.org>
Commit-Queue: Ryan Tseng <hinoka@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589683}

This adds filter file support to all test types that previously
supported gtest-style test filtering, notably including instrumentation
and junit tests.

This was requested in crrev.com/c/1208442

Change-Id: I552843898735b52b773bdc397c39ecf64e4df275
Reviewed-on: https://chromium-review.googlesource.com/1214083
Commit-Queue: John Budorick <jbudorick@chromium.org>
Reviewed-by: agrieve <agrieve@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589682}

Change-Id: I65c3968959acf071019294acfa05ce5e0b4f23ba
Reviewed-on: https://chromium-review.googlesource.com/1214165
Commit-Queue: Tibor Goldschwendt <tiborg@chromium.org>
Reviewed-by: Eric Stevenson <estevenson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589681}

But splitview is still tablet only.

Test: manual
Bug: 879761
Change-Id: I8957eaab9c37b98a219caa2addcfab35bacffc1d
Reviewed-on: https://chromium-review.googlesource.com/1211109
Commit-Queue: Sammie Quon <sammiequon@chromium.org>
Reviewed-by: Xiaoqian Dai <xdai@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589680}

VariationsService was not initializing ResourceRequestsAllowedNotifier
on the OS_ANDROID portion of the #ifdef, which was preventing it from
being notified of network changes.

This should fix the regressions in Variations.SeedFreshness on Android:
https://uma.googleplex.com/variations?sid=aeb0078452c432f2d6b3aa76b79354cd

Bug: 868021
Change-Id: I40eaeea5dbd97e87c5ed16c7161781e3f7f67659
Reviewed-on: https://chromium-review.googlesource.com/1213316Reviewed-by: Alexei Svitkine <asvitkine@chromium.org>
Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Commit-Queue: Clark DuVall <cduvall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589679}

As explained in the bug, "tracking document" should be associated with
IntersectionObservation, not IntersectionObserver.

One side effect of this change is that for a given observer, each of its
observations runs its algorithm independently, so that notifications may
not be generated in the order in which the observations were created
(i.e., the order of calls to observer.observe()). To preserve the
idiomatic web platform behavior that notifications are delivered in
the order in which the targets were observed, this patch moves the
storage of pending notifications from IntersectionObserver to
IntersectionObservation.

BUG=879798
R=chrishtr@chromium.org

Change-Id: I9b6f6ad8a26387f0c072ccdff7f18cea9a88004c
Reviewed-on: https://chromium-review.googlesource.com/1200388Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Commit-Queue: Stefan Zager <szager@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589678}

To fix the issue of decoder reporting different sample rate
from the demuxer.

some AAC streams' (from NBC Sports/ESPN, etc) ADTS header info,
especially the sample rate, does not match the payload. The
demuxer reports wrong sample rate to mixer, causing incorrect
audio resampling. Currently the decodedbuffer does not include
the sample rate property, so add AudioConfig field to the callback
to return the info back to mixer and adjust the sample rate when
needed. It should apply to all audio paths that needs AudioConfig.

Bug: internal b/112898981
Test: local test NBC sports, audio sounds normal

Depends-On: eureka-internal/199064

Change-Id: Iebef3acf50e856c8fcb47c8d0b13c3598c6df20f
Reviewed-on: https://chromium-review.googlesource.com/1213320
Commit-Queue: Yi Fan <yfa@chromium.org>
Reviewed-by: Kenneth MacKay <kmackay@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589677}

Previously, the benchmark timing files & shard map files are manually named. This
CL simplifies all of them to be based on the builder names. This simplifies
the implementation of tools/perf/core/bot_platforms.py file.

This should be a pure refactoring & has no visible effect on how we schedule tests

This is a follow up of Juan's comment in https://chromium-review.googlesource.com/c/chromium/src/+/1188945/6/tools/perf/core/bot_platforms.py#85

Bug:863768

Cq-Include-Trybots: master.tryserver.chromium.perf:obbs_fyi
Change-Id: I7aa08e10978f54b7f2b15759d20d64a9c7bd8533
Reviewed-on: https://chromium-review.googlesource.com/1213317Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: Ned Nguyen <nednguyen@google.com>
Cr-Commit-Position: refs/heads/master@{#589676}

UI review requests to put the Continue Reading behind the
Open Tabs toggle.
Changes:
1. Revert previous setting for Continue Reading.
2. Use syncPrefs.tabsSynced to control the feature.

Bug: 880933, 843839
Test: manual.
Change-Id: Iac3464a7c7a08b3bb33e9e3a93d48e9dc365acec
Reviewed-on: https://chromium-review.googlesource.com/1208031
Commit-Queue: Tao Wu <wutao@chromium.org>
Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589675}

Introduces the "outer shell" API to for GLScaler, with upcoming patches
coming soon to fill-in the rest of the implementation. This is a major
refactoring of GLHelper+GLHelperScaler to: 1) fix dangerous memory
management issues, 2) provide a cleaner interface for client-side code,
3) unblock performance and color management efforts.

Bug: 870036,758057
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel
Change-Id: Ib8f0ed9b988bfe914bd8983940b089262543e8c7
Reviewed-on: https://chromium-review.googlesource.com/1194783
Commit-Queue: Yuri Wiitala <miu@chromium.org>
Reviewed-by: Xiangjun Zhang <xjz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589674}

Don't close the ZoomBubbleView on timeout if one of the buttons inside
it has focus. (The bubble doesn't close on timeout if the mouse is
inside the bubble; screenreaders should have the same experience).

Also add an accessible title (read with Insert+T on JAWS and NVDA).

Bug: 849486
Test: ZoomBubbleBrowserTest.FocusPreventsClose

Change-Id: Idbe2e46ab0c6d896691850ff6fee5a843abee2f9
Reviewed-on: https://chromium-review.googlesource.com/1189044Reviewed-by: Evan Stade <estade@chromium.org>
Commit-Queue: James Wallace-Lee <jamwalla@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589673}

Bug: 878178
Change-Id: I90a27bea589fafc048cb165ffb698c44cb7bb049
Reviewed-on: https://chromium-review.googlesource.com/1212762
Commit-Queue: Kevin Strohbehn <ginko@google.com>
Reviewed-by: Weidong Guo <weidongg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589672}

Create opaque SecurityOrigins for nonstandard schemes, unless
they are local. Allow creation of url::Origins for local,
nonstandard schemes.

The root problem here was inconsistency between url::Origin and
SecurityOrigin -- url::SchemeHostPort (and thus url::Origin)
rejects anything with a nonstandard protocol, but
SecurityOrigin::Create(KURL) would allow this creation to
proceed as a tuple origin ("protocol", "", 0).

The allowance of local+nonstandard schemes seems to be
necessary, at least temporarily, to avoid breaking the
Android-only "content:" scheme, and the ChromeOS-only
"externalfile:" scheme.

Other nonstandard schemes shouldn't generally commit, but this does
seem to be possible via LoadDataWithBaseUrl; for example in the test
"NavigationControllerBrowserTest.CrossDomainResourceRequestLoadDataWithBaseUrl"

BUG=862282

Change-Id: I351a067521cd846d2c77cca2498e58580613b4e2
Reviewed-on: https://chromium-review.googlesource.com/1208811
Commit-Queue: Nick Carter <nick@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Mike West <mkwst@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589671}

Fixes a NOTREACHED crash on eng builds whenever you touch the screen.

Bug: internal b/112437110
Test: touch screen on eng build
Change-Id: Iab827b9943df6869a919443b43bfd2eba43da3fc
Reviewed-on: https://chromium-review.googlesource.com/1211707
Commit-Queue: Luke Halliwell <halliwell@chromium.org>
Reviewed-by: Sergey Volk <servolk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589670}

When resources for a screen density are missing in a WebAPK (e.g. the
WebAPK only contains an icon for xxxhdpi),
Resources#getDrawableForDensity() returns a scaled bitmap.

On Android versions up to Android M, Resources#getDrawableForDensity()
would sometimes return an incorrectly scaled cached bitmap instead of
rerequesting the resource and scaling it for the passed in density. See
the bug for more details.

This CL moves the icons to the mipmap-nodpi folder in order to disable
the automatic scaling when requesting a resource. The scaling now occurs
inside of ImageView. The move fixes the caching bug.

It also increases how often we use the webapp_splash_screen_no_icon.xml
layout on high density devices. In particular, if the WebAPK only has
a 144x144px sized icon on xxxhdpi devices:
Prior to this CL: We used webapp_splash_screen_small.xml and
                  upscaled the icon.
With this CL: We will use webapp_splash_screen_no_icon.xml

BUG=879194

Change-Id: I3c87898ca39cc7a47b614d4b219aee5f1e41ce74
Reviewed-on: https://chromium-review.googlesource.com/1199768Reviewed-by: Xi Han <hanxi@chromium.org>
Reviewed-by: Yaron Friedman <yfriedman@chromium.org>
Commit-Queue: Peter Kotwicz <pkotwicz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589669}

CL was reverted here:
https://chromium-review.googlesource.com/c/chromium/src/+/1214105

The issue was another CL that renamed a function landed at the same
time.

It used to get the Primary Account's email but should get the active
account's one instead if in Butter state.

Bug: 881730
Change-Id: I9b5b489bc5237d044e7e36fde5eda7f7da7ada27
Reviewed-on: https://chromium-review.googlesource.com/1214063
Commit-Queue: Sebastien Seguin-Gagnon <sebsg@chromium.org>
Reviewed-by: Mathieu Perreault <mathp@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589668}

This change modifies the disk_cache API to support 64-bit cache sizes.
Individual entries are still limited to 32-bit sizes.  The blockfile
and memory backends will enforce a 32-bit cache size limit at runtime.
The simple backend is currently the only implementation that fully
supports 64-bit cache sizes.

R=jsbell@chromium.org, morlovich@chromium.org

Bug: 653630
Cq-Include-Trybots: luci.chromium.try:ios-simulator-cronet;luci.chromium.try:ios-simulator-full-configs;luci.chromium.try:linux_mojo
Change-Id: I9278b9e14e0a3253a66d20e57cf77026a462b4ac
Reviewed-on: https://chromium-review.googlesource.com/1194508
Commit-Queue: Ben Kelly <wanderview@chromium.org>
Reviewed-by: Eugene But <eugenebut@chromium.org>
Reviewed-by: Joshua Bell <jsbell@chromium.org>
Reviewed-by: Maks Orlovich <morlovich@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589667}

MDCAppBar will be deprecated soon, so eliminating any existing imports now will prevent errors later.

Bug: 874471
Change-Id: I67073fc90e935daee1c55d94db84eced20062999
Reviewed-on: https://chromium-review.googlesource.com/1213373Reviewed-by: Sergio Collazos <sczs@chromium.org>
Commit-Queue: Chris Lu <thegreenfrog@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589666}

Google-internal design doc:
https://docs.google.com/document/d/1X6Xoz9hl3qoAj3EhAKcvZ6hWLVcY_QpQgmUexpgDHKA/edit?usp=sharing

Move prefs creation from AwBrowserContext in PreMainMessageLoopRun to
AwBrowserMainParts in PreCreateThreads. This way, variations can use the
same pref service as the rest of WebView, while still being initialized
as early as possible. (See BrowserMainLoop::CreateStartupTasks() for an
overview of the startup sequence.)

Before, AwBrowserContext created and owned PrefService and
BrowserPolicyConnector. Now, AwBrowserMainParts creates them in
PreCreateThreads, and owns them until PreMainMessageLoopRun, at which
point ownership is transferred to AwBrowserContext.

BUG=866722

Change-Id: Ie8eb6df60bd4a0f872d8c6401cc14ff0bca38368
Reviewed-on: https://chromium-review.googlesource.com/1212268Reviewed-by: Richard Coles <torne@chromium.org>
Commit-Queue: Paul Miller <paulmiller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589665}

Previously we assumed that Chrome would always start-up
on regular browsing mode (never in incognito). The linked
bugs prove that this is not true - sometimes the app starts
in incognito. This CL sets the incognito state on initialize
with native.

Bug: 881837, 880178
Change-Id: I67cf33b4be10997c5f62a85d893febde0c278b87
Reviewed-on: https://chromium-review.googlesource.com/1213313
Commit-Queue: Pedro Amaral <amaralp@chromium.org>
Reviewed-by: Pedro Amaral <amaralp@chromium.org>
Reviewed-by: Matthew Jones <mdjones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589664}

Some failures are caused by changed but valid results instead of the bug.
Rebaseline so that we can track real flakiness.

TBR=yoichio@chromium.org

Bug: 774437
Change-Id: If7f5664902b67700bd18ab70a209cbc160e028d5
Reviewed-on: https://chromium-review.googlesource.com/1213269Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org>
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589663}

This makes us more consistent in how the ImmersiveFullscreenController
gets enabled. This patch also fixes the UMA for Chrome app windows.

Change-Id: Ib81d7401239cdcf42a415e81b01c29f3f378075c
Reviewed-on: https://chromium-review.googlesource.com/1207213Reviewed-by: Scott Violet <sky@chromium.org>
Commit-Queue: Evan Stade <estade@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589662}

https://chromium.googlesource.com/chromiumos/chromite.git/+log/990810e4ba69..6dd7eccb9a96


git log 990810e4ba69..6dd7eccb9a96 --date=short --no-merges --format='%ad %ae %s'
2018-09-07 bhthompson@google.com Add grunt-android-pi-pre-flight-branch builder for R70


Created with:
  gclient setdep -r src/third_party/chromite@6dd7eccb9a96

The AutoRoll server is located here: https://autoroll.skia.org/r/chromite-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=chrome-os-gardeners@chromium.org

Change-Id: Iff7f27d540739234f83832944ae556f2e12150d6
Reviewed-on: https://chromium-review.googlesource.com/1213943Reviewed-by: Chromite Chromium Autoroll <chromite-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Commit-Queue: Chromite Chromium Autoroll <chromite-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#589661}

https://chromium.googlesource.com/chromium/tools/depot_tools.git/+log/515e7fe03790..1cefd62ebe68


git log 515e7fe03790..1cefd62ebe68 --date=short --no-merges --format='%ad %ae %s'
2018-09-07 recipe-roller@chromium.org Roll recipe dependencies (trivial).


Created with:
  gclient setdep -r src/third_party/depot_tools@1cefd62ebe68

The AutoRoll server is located here: https://autoroll.skia.org/r/depot-tools-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=agable@chromium.org

Change-Id: Iad944e9977d12797b6bf434fd6c347ea864d9e7e
Reviewed-on: https://chromium-review.googlesource.com/1213947Reviewed-by: depot-tools-chromium-autoroll <depot-tools-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Commit-Queue: depot-tools-chromium-autoroll <depot-tools-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#589660}

This fix would actually apply to desktop views if the test created
DesktopNativeWidgetAura there as well.

This also removes a CHECK in event_generator_test that was added here:
https://codereview.chromium.org/147203004/diff/140001/ui/aura/test/event_generator.cc
and makes EventGenerator reset current_target_ if the dispatcher is destroyed.
This assumes either the target is reset, or the generator destroyed.

BUG=594260
TEST=test only changes

Change-Id: I860f7e254a0d9a5826955011b74da8e25d50f042
Reviewed-on: https://chromium-review.googlesource.com/1213374Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org>
Commit-Queue: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589659}

This simplifies conditions around the code.

Bug: 561800
Change-Id: Iaa24aeb45df5465e88de8f48b504b98737557dbd
Reviewed-on: https://chromium-review.googlesource.com/1180635
Commit-Queue: Nicolas Zea <zea@chromium.org>
Reviewed-by: Nicolas Zea <zea@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589658}

This also fixs a bug when using operator= with an event. Specifically we
were not correctly calling ReleaseCopiedNativeEvent() as well as
CopyNativeEvent(). I suspect not many places use operator=, so it wasn't
really an issue. I was tempted to make Event's operator= private, but KeyEvent
explicitly has operator=, which implies we need to support at Event as well.

BUG=881893
TEST=covered by tests

Change-Id: I16480937b04d881849ca7ac222f2df48d20e763d
Reviewed-on: https://chromium-review.googlesource.com/1213584Reviewed-by: Will Harris <wfh@chromium.org>
Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org>
Commit-Queue: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589657}

A number of changes have landed recently that may eliminate the
flakieness seen in this test. This CL re-enables the test, but feel
free to revert this if the flakes continue.

If this CL sticks, then all tests related to this bug have been
re-enabled.

Bug: 833380
Change-Id: I38147cf870ff11b51fa67dba1a209dfa40a7643c
Reviewed-on: https://chromium-review.googlesource.com/1213302Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Commit-Queue: James MacLean <wjmaclean@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589656}

The tests have been running for a while now and look stable.  Move them over
to the main waterfall so they start being run by the CQ.

Bug: 874090
Bug: 847585
Change-Id: I8b013c568d0abbfb2697d5166dd6edda31bbbf88
Reviewed-on: https://chromium-review.googlesource.com/1212093
Commit-Queue: Ryan Cui <rcui@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589655}

- added images/ directory
- placing .svg images in images/
- moved default_N_percent directory to images/
- replaced the help_outline PNGs with SVG

Bug: 852872
Change-Id: I7c3083526a64304f461ed964b040767ba6d8afe4
Reviewed-on: https://chromium-review.googlesource.com/1210962Reviewed-by: Edward Jung <edwardjung@chromium.org>
Reviewed-by: Asanka Herath <asanka@chromium.org>
Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Commit-Queue: Dan H <harringtond@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589654}

Chrome extensions do not have direct access to a user's Chrome profile information, information such as saved passwords and autofill profiles.
However, the Action Recorder Extension may derive a user's saved passwords and autofill profile by observing the password fields, shipping info fields and payment fields that Chrome autofills.

To land this change, I also had to change how the extension detects the Chrome Autofill action. Prior to this change, the extension detects Chrome Autofill using a combination of keydown and onchange event listeners. This strategy can distinguish user typing actions from Chrome autofill actions, but this stategy does not distinguish Chrome autofill actions from Page JavaScript changing input values. In this change, I switched the extension to using the method outlined in https://medium.com/@brunn/detecting-autofilled-fields-in-javascript-aed598d25da7 to detect Chrome autofill.

Bug: 855284
Change-Id: I22283d5684237084ca2a8b95d9f4723be46b7746
Reviewed-on: https://chromium-review.googlesource.com/1208446
Commit-Queue: Yiming Zhou <uwyiming@google.com>
Reviewed-by: Jared Saul <jsaul@google.com>
Cr-Commit-Position: refs/heads/master@{#589653}

Per https://drafts.csswg.org/cssom/#serialize-a-string double quotes
should be used when serializing strings.

Bug: 873527
Change-Id: I3b7a3c6425c6a5d0d6f96399663f5ae8b3929a59
Reviewed-on: https://chromium-review.googlesource.com/1213171
Commit-Queue: Eric Willigers <ericwilligers@chromium.org>
Reviewed-by: Eric Willigers <ericwilligers@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589652}

The patch to call Unalias caused a nullptr dereference, because of
another bug that previously only triggered a DCHECK. This patch ensures
to do a "safe unalias" where we first check for null.

This check can be reverted once the underlying issues that trigger the
DCHECK are fixed (see the two referenced bugs for the nullptr
dereference and for the DCHECK failure).

R=pdr@chromium.org, trchen@chromium.org

Bug: 881788, 853357
Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I412d60581466a40e4543b0ce78b18fb1a739f02b
Reviewed-on: https://chromium-review.googlesource.com/1213869Reviewed-by: Philip Rogers <pdr@chromium.org>
Commit-Queue: Philip Rogers <pdr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589651}

As per [1], CaptureRequest.get() may return nullptr. Crash reports show
that for some phone models some of these calls actually do. This leads
to nullptr exception when the return value is directly converted to a
primitive value.

This CL checks for null before assuming that the return value can be
converted to a primitive.

[1] https://developer.android.com/reference/android/hardware/camera2/CaptureRequest#get(android.hardware.camera2.CaptureRequest.Key%3CT%3E)

Bug: 880855
Change-Id: I4f83499f78208b97c181fa621a1ea621d5e08de1
Reviewed-on: https://chromium-review.googlesource.com/1214047Reviewed-by: Emircan Uysaler <emircan@chromium.org>
Commit-Queue: Christian Fremerey <chfremer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589650}

In preparation for enabling an ErrorProne check to ban them

Bug: 878853
Cq-Include-Trybots: master.tryserver.chromium.android:android_cronet_tester
Change-Id: I9cf9b07303fb3b00ce4d904f8fe56f47bb0b6268
Reviewed-on: https://chromium-review.googlesource.com/1213880Reviewed-by: Misha Efimov <mef@chromium.org>
Commit-Queue: Aiden Benner <abenner@google.com>
Cr-Commit-Position: refs/heads/master@{#589649}

If we haven't gone through prepaint since the LayoutObject was attached,
we will crash accessing the transform node.

Bug: 881921, 881735, 881643
Change-Id: I3c27c045a87319a3da2f1803707d42139d41cbfd
Reviewed-on: https://chromium-review.googlesource.com/1213805
Commit-Queue: Steve Kobes <skobes@chromium.org>
Reviewed-by: Timothy Dresser <tdresser@chromium.org>
Cr-Commit-Position: refs/heads/master@{#589648}