GitLab

This CL adds a timer to WebApkIconHasher. Consequently, classes that
invoke WebApkIconHasher::DownloadAndComputeMurmur2Hash no longer need
to maintain a timer dedicated to WebApkIconHasher.

This CL also simplifies the error checking and handling logic.
WebApkIconHasher now checks for URL validity itself, and would return
an empty hash string for any errors encountered.

BUG=649771

Review-Url: https://codereview.chromium.org/2771793003
Cr-Commit-Position: refs/heads/master@{#460508}

even for zero byte files.

URLRequestExtensionJob depends on this behavior, so add a zero-byte
test to url_request_file_job_unittest. Also update other relevant tests
in that file to check for DoneReading.

This is merely a test addition/modification for crbug.com/703892

BUG=703892
Test=None, internal change only.

Review-Url: https://codereview.chromium.org/2776873002
Cr-Commit-Position: refs/heads/master@{#460507}

Added a concrete example in comments on how to call Chrome OS Component API.

BUG=690521
TEST=None

Review-Url: https://codereview.chromium.org/2762393002
Cr-Commit-Position: refs/heads/master@{#460506}

When the default browser is not set, the 'Open in browser' option in the
context menu for CCT does not work for image and video link.

This is because for image and video, the linkUrl of the
ContextMenuParams is null. It is a same problem as the bug here:
https://bugs.chromium.org/p/chromium/issues/detail?id=696417.

And to fix it, follow this CL:
https://codereview.chromium.org/2739723004/ that checks the type of link
first then passes different urls for different types of link.

BUG=706209

Review-Url: https://codereview.chromium.org/2779113006
Cr-Commit-Position: refs/heads/master@{#460505}

compositor_frame_sink_id won't exist once we switch to mojo, so it
should not be exposed to non-IPC methods. Instead of passing
compositor_frame_sink_id to the views, we just keep track of it in
RenderWidgetHostImpl and call DidCreateNewRendererCompositorFrameSink
on the view when its value changes. When
DidCreateNewRendererCompositorFrameSink is called, the views ensure
that they never return resources of the old RendererCompositorFrameSink
by recreating CompositorFrameSinkSupport.

bug=697864
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2780713004
Cr-Commit-Position: refs/heads/master@{#460504}

If user opens recent/history menu and at the same moment syncronization
is updating session from other devices then user will can see
non-relevant a list of opened tabs from the other devices.

It's easy to reproduce.
1. Open a browser with enabled syncronization.
2. Open chrome://sync-internals page.
3. Open recent/history menu and wait for updating session on
   chrome://sync-internals page.
4. Open new tab on other synced device.
5. When chrome://sync-internals shows that session was updated the
   opened menu will show previous state of list list of opened tabs
   from the other devices.

R=pkasting@chromium.org, pavely@chromium.org

Review-Url: https://codereview.chromium.org/2768633003
Cr-Commit-Position: refs/heads/master@{#460503}

This way it shrinks if the text is bigger than expected, or expands if
more space is provided.

BUG=none
TEST=none
R=bsep@chromium.org

Review-Url: https://codereview.chromium.org/2782743003
Cr-Commit-Position: refs/heads/master@{#460502}

When a result of |string.c_str()| expression is passed to a function that takes
a StringPiece, it will measure the length of the string using strlen() function.
This is useless operation because we know the length already.

No behaviour changes.

BUG=679479

Review-Url: https://codereview.chromium.org/2779893005
Cr-Commit-Position: refs/heads/master@{#460501}

BUG=693525

Review-Url: https://codereview.chromium.org/2774293002
Cr-Commit-Position: refs/heads/master@{#460500}

BUG=676164
TBR=estade

Review-Url: https://codereview.chromium.org/2779523005
Cr-Commit-Position: refs/heads/master@{#460499}

Previously, params that were queried via their associated features
would not get correctly picked by perfbots using the field trial
testing config.

This was caused by the generated command-line not using the
--enable-features=FeatureName<TrialName syntax. This syntax is
needed in order to estable the link between a feature and its
associated trial, so that when a field-trial param is looked up
by feature, it's able to find it from the associated trial.

Adds quotations to these params since "<" characters would
otherwise get interpreted by some shells.

PERF SHERIFFS: This CL may cause perf bots to move - if so, it
means that one of the entries in the field trial testing config
that uses both params and features is causing the change but was
previously not being correctly tested. If so, we will need to
narrow down which one.

BUG=706414

Review-Url: https://codereview.chromium.org/2778313003
Cr-Commit-Position: refs/heads/master@{#460498}

BUG=646928

Review-Url: https://codereview.chromium.org/2785533002
Cr-Commit-Position: refs/heads/master@{#460497}

The BindingSet for mojom.GpuService is bound on the IO thread. So make
sure it gets destroyed on the IO thread too.

BUG=706357

Review-Url: https://codereview.chromium.org/2779903005
Cr-Commit-Position: refs/heads/master@{#460496}

BUG=697333

Review-Url: https://codereview.chromium.org/2782193002
Cr-Commit-Position: refs/heads/master@{#460495}

BUG=695613

Review-Url: https://codereview.chromium.org/2784633002
Cr-Commit-Position: refs/heads/master@{#460494}

This CL makes the paper icon button focus ripples (disc) consistent.
Before this CL, the menu icon and clear search icon had a different
focus color than the search icon, for example. Alan wants them to all
have a white focus highlight.
This is a step toward fixing the bug (not a complete fix).

BUG=693383
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2778463003
Cr-Commit-Position: refs/heads/master@{#460493}

This CL fixes some desktop integration issues on Ubuntu 17.04, where
XDG_CURRENT_DESKTOP has changed from 'Unity' to 'Unity:Unity7' (or
'Unity:Unity8' if you're of the Mir persuasion).

BUG=706317
R=thestig@chromium.org

Review-Url: https://codereview.chromium.org/2780863003
Cr-Commit-Position: refs/heads/master@{#460492}

There is a current refcounting known issue where a blob can be sent to
another process but while that IPC is scheduled the source process
dies. The blob will be non-existent. We want to gracefully handle this
error in IndexedDB instead of crashing.

(also fixes missing include in the connection class)

BUG=677780

Review-Url: https://codereview.chromium.org/2779433002
Cr-Commit-Position: refs/heads/master@{#460491}

Revert of Make VR screen tap tests more stable (patchset #6 id:100001 of https://codereview.chromium.org/2768583002/ )

Reason for revert:
Breaking compile on several bots including Android x64 builder:
  symbol:   class OnDispatchTouchEventCallback
  location: class WebVrTest
../../chrome/android/javatests/src/org/chromium/chrome/browser/vr_shell/WebVrTest.java:269: error: cannot find symbol
        ((VrShellImpl) VrShellDelegate.getVrShellForTesting()

Original issue's description:
> Make VR screen tap tests more stable
>
> The single tap sent during the VR screen tap tests wasn't always
> getting registered since device response can be a bit slow during
> the VR transition. For the Cardboard case, wait on Javascript to say
> it received another click event. For the Daydream case, wait on VrShell
> saying its parent has consumed the event before checking.
> BUG=
>
> Review-Url: https://codereview.chromium.org/2768583002
> Cr-Commit-Position: refs/heads/master@{#460452}
> Committed: https://chromium.googlesource.com/chromium/src/+/3e2ce51e88f19f0cc5196284f580583ddd4dbd9a

TBR=dtrainor@chromium.org,tedchoc@chromium.org,mthiesse@chromium.org,bsheedy@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=

Review-Url: https://codereview.chromium.org/2782963002
Cr-Commit-Position: refs/heads/master@{#460490}

This requires changing its signature. This is a preliminary stage to making it
private.

BUG=633030

Review-Url: https://codereview.chromium.org/2698673003
Cr-Commit-Position: refs/heads/master@{#460489}

The previous CL, https://codereview.chromium.org/2607373002/, that fixed
 a resizing problem in accelerated 2d canvas, has introduced this error
 message "GL_INVALID_ENUM : glTexParameteri: param was GL_FALSE" because
the resource.filter in OffscreenCanvasFrameDispatcherImpl is not set.
This causes ResourceProvider::DeleteAndReturnUnusedResourcesToChild to run
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, 0) as it discovers
a difference between original_filter and filter in resource when closing
down browser.

We then find out that whilst quad.nearest_neighbor indicates the desired
filtering effect on the rendered quad, the resource.filter actually
indicates the filtering algorithm on the resource inherently. To remove
this error msg whilst maintaining the correctness of filtering algorithm
applied on accelerated 2d canvas, we need to specify the resource filter to
GL_NEAREST. Then the overriding mechanism about resource->original_filter
and resource->filter in resource_provider.cc will be able to function
correctly.

In addition, I added GL_TEXTURE_MAG_FILTER as there is no reason why
we only apply linear filtering on MIN_FILTER and not MAG_FILTER.

BUG=692599

Review-Url: https://codereview.chromium.org/2758863003
Cr-Commit-Position: refs/heads/master@{#460488}

This is a follow-up to https://crrev.com/2774673002.

This CL lands server change 151348843 by bnc.

BUG=488484

Review-Url: https://codereview.chromium.org/2778043004
Cr-Commit-Position: refs/heads/master@{#460487}

WebData is already a RefPtr<SharedBuffer>. We don't need to copy the data.

Review-Url: https://codereview.chromium.org/2787573002
Cr-Commit-Position: refs/heads/master@{#460486}

BUG=704361
R=dpranke@chromium.org

Review-Url: https://codereview.chromium.org/2765423003
Cr-Commit-Position: refs/heads/master@{#460485}

BUG=706416
TBR=primiano@chromium.org

Review-Url: https://codereview.chromium.org/2786713002
Cr-Commit-Position: refs/heads/master@{#460484}

https://boringssl.googlesource.com/boringssl/+log/d04ca95356c5b79e9ad80f3e8d1bd9784213dc93..a0ba400c33c5919fb0c25f165133446cb6dec5c0

BUG=none

Review-Url: https://codereview.chromium.org/2783923002
Cr-Commit-Position: refs/heads/master@{#460483}

This patch fixes and reverts the changes caused here:
https://codereview.chromium.org/2692653003. When we moved the checks
into network it no longer was able to receive messages when changed from
some other panels like timeline or emulation if network was not loaded
yet (ie: lazy loaded).

R=dgozman,pfeldman@chromium.org
BUG=700797

Review-Url: https://codereview.chromium.org/2765443005
Cr-Commit-Position: refs/heads/master@{#460482}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/b7dd32be..04f54bf6

Please follow these instructions for assigning/CC'ing issues:
https://github.com/v8/v8/wiki/Triaging%20issues

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

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel

TBR=hablich@chromium.org,machenbach@chromium.org,kozyatinskiy@chromium.org,vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/2777363007
Cr-Commit-Position: refs/heads/master@{#460481}

This CL updates PhotoCapabilities.idl and associated logic to the extent
possible before touching the mojo interface, which essentially means:

- |redEyeReduction| changes to be an enum of supported modes
- |fillLightMode| changes to be an array of supported modes

Referring to the following Spec PRs :

https://github.com/w3c/mediacapture-image/pull/164
 (which addressed https://github.com/w3c/mediacapture-image/issues/163)
and
https://github.com/w3c/mediacapture-image/pull/165
 (which addressed https://github.com/w3c/mediacapture-image/issues/156)

BUG=706171

Review-Url: https://codereview.chromium.org/2785503003
Cr-Commit-Position: refs/heads/master@{#460480}

CommandExecutor can take a PauseExectionCallback to pause command
execution. This is where the PreemptionFlag is hooked into and where
the scheduler will too.

R=piman@chromium.org
BUG=514813
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2763753002
Cr-Commit-Position: refs/heads/master@{#460479}

BUG=

Review-Url: https://codereview.chromium.org/2777233004
Cr-Commit-Position: refs/heads/master@{#460478}

BUG=686284,686286

Review-Url: https://codereview.chromium.org/2781793002
Cr-Commit-Position: refs/heads/master@{#460477}

Implement the SubresourceFilterActivationNavigationThrottle for accessing the Subresource Filter list.

In this Cl NavigationThrottle acts only on WillProcessResonce signal.
This is temporal behavior and will changed in followup CLs.

Part 1, https://codereview.chromium.org/2690413002
Part 2, https://codereview.chromium.org/2645283007 (this one)
Part 3, https://codereview.chromium.org/2770153004

BUG=671962

Review-Url: https://codereview.chromium.org/2645283007
Cr-Commit-Position: refs/heads/master@{#460476}

A SchedulerWorker may outlive the test environment when it's released
as a result of SingleThreadTaskRunner unregistration in the
SchedulerSingleThreadTaskRunnerManager. As a result, SchedulerWorker
should prefer checking its own fields first to validate that
it should continue to run before consulting the environment provided
TaskTracker.

BUG=705500

Review-Url: https://codereview.chromium.org/2781933003
Cr-Commit-Position: refs/heads/master@{#460475}

Other minor changes:
1. Separate out QUIC tests to quic.py.
2. Separate out QUIC test in two different tests.
3. Add chrome arg --data-reduction-proxy-enable-quic-on-non-core-proxies
to the tests so that QUIC usage would be enabled even when Chrome is
using a non-core proxy.

BUG=663584

Review-Url: https://codereview.chromium.org/2775173003
Cr-Commit-Position: refs/heads/master@{#460474}

Review-Url: https://codereview.chromium.org/2752253003
Cr-Commit-Position: refs/heads/master@{#460473}

They were attempting to disable the beforeunload timeout, but the switch provided no longer worked.

BUG=705306, 418266, 700271, 700641
TEST=Linux MSan Tests goes green
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation;master.tryserver.chromium.linux:linux_chromium_msan_rel_ng

Review-Url: https://codereview.chromium.org/2777013004
Cr-Commit-Position: refs/heads/master@{#460472}

This CL adds two dialogs that are shown by ConfirmSyncDataStateMachine.
Dialogs implementation is in ConfirmSyncDataStateMachineDelegate.
The first dialog is a progress dialog that is shown while Chrome requests
server for account management policy.
The second dialog is a timeout dialog that is shown if Chrome doesn't receive
account management policy within 30 seconds.

BUG=650121

Review-Url: https://codereview.chromium.org/2772203004
Cr-Commit-Position: refs/heads/master@{#460471}

This should improve performance slightly when used in certain cases like a
std::vector.

Review-Url: https://codereview.chromium.org/2778183003
Cr-Commit-Position: refs/heads/master@{#460470}

This mainly to fix bugs where line numbers appear outside the gutter.
This patch also removes the SimpleMode since it was never used.

BUG=none
R=lushnikov@chromium.org

Review-Url: https://codereview.chromium.org/2772343006 .
Cr-Commit-Position: refs/heads/master@{#460469}