GitLab

This CL adds |device_heartbeat_settings| to the device policy proto.
We accidentally added |device_heartbeat_settings| with index 34 on
the server and |reboot_on_shutdown| with the same index on the client.
To resolve this conflict, one of the two settings needs to change
index. Since |device_heartbeat_settings| is not being read or written
by anyone yet, it is the one to move.

We must avoid such conflicts in the future by adding all policies on
the client-side first and syncing the resulting proto to the server.

In general, new policies should be added to the device policy proto
and policy_templates.json at the same time. As an emergency fix, this
CL gets a free pass for that rule as it is an emergency fix.

BUG=430908
TEST=None

Review URL: https://codereview.chromium.org/887423002

Cr-Commit-Position: refs/heads/master@{#314346}

The value of the option is simply passed down to WaitUntilFullyBooted
when block=True.

BUG=

Review URL: https://codereview.chromium.org/895783005

Cr-Commit-Position: refs/heads/master@{#314345}

Contains fixes for error message creation.

https://chromium.googlesource.com/angle/angle/+log/bc393d..592ab9

BUG=454520

Review URL: https://codereview.chromium.org/892413003

Cr-Commit-Position: refs/heads/master@{#314344}

BUG=170587, 170592
NOTRY=true

Review URL: https://codereview.chromium.org/869933005

Cr-Commit-Position: refs/heads/master@{#314343}

Summary of changes:
https://chromium.googlesource.com/android_tools/+log/aaeda3d..f6e2370

BUG=427718

Review URL: https://codereview.chromium.org/899663002

Cr-Commit-Position: refs/heads/master@{#314342}

Now that we have access to the IO thread, pass the IO task runner to
the main thread and use that to register and listen for page flip
events.

BUG=none

Review URL: https://codereview.chromium.org/891703006

Cr-Commit-Position: refs/heads/master@{#314341}

Changes in wizard_controller.cc - chromeos namespace usage cleanup.

BUG=454801
TEST=existing browser tests + manual test with --test-auto-update-ui

Review URL: https://codereview.chromium.org/863263006

Cr-Commit-Position: refs/heads/master@{#314340}

BUG=451075

Review URL: https://codereview.chromium.org/893543003

Cr-Commit-Position: refs/heads/master@{#314339}

Move the chrome.diagnostics extension API from chrome/ to
extensions/.

BUG=395240

Review URL: https://codereview.chromium.org/883703004

Cr-Commit-Position: refs/heads/master@{#314338}

R=estade@chromium.org
BUG=421816

Review URL: https://codereview.chromium.org/657493003

Cr-Commit-Position: refs/heads/master@{#314337}

Make it possible to use cr for building on mac. Mostly this is a clone of what
we are doing on Linux, except that I had to work around "readlink -e" not being
available.

Review URL: https://codereview.chromium.org/898683003

Cr-Commit-Position: refs/heads/master@{#314336}

BUG=454562
TEST=Manual, see bug

Review URL: https://codereview.chromium.org/864123004

Cr-Commit-Position: refs/heads/master@{#314335}

The locale parameter (hl=) has been broken by the componentization of
web_resource, and is now always empty.
This CL restores the right behavior.

BUG=454485

Review URL: https://codereview.chromium.org/900633002

Cr-Commit-Position: refs/heads/master@{#314334}

This patch adds an extra dependency on content.gyp:content_browser
to browser_app_shim in order to generate permission_status.mojom.h
before compile.

BUG=454447

Review URL: https://codereview.chromium.org/894103002

Cr-Commit-Position: refs/heads/master@{#314333}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/fef4c32..532faa9

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=stephana@google.com

Review URL: https://codereview.chromium.org/875113003

Cr-Commit-Position: refs/heads/master@{#314332}

R=asvitkine@chromium.org

Review URL: https://codereview.chromium.org/875983007

Cr-Commit-Position: refs/heads/master@{#314331}

If a device is disabled, it should show the disabled screen and not
start any sessions. Regular sessions, auto-launch for public sessions
and auto-launch for consumer kiosk mode respected this. Auto-launch
for enterprise kiosk mode takes a different code path that had
slipped through the cracks. The effect was that single-app kiosk mode
would auto-launch even if a device was disabled.

BUG=454005
TEST=New browser test

Review URL: https://codereview.chromium.org/882043006

Cr-Commit-Position: refs/heads/master@{#314330}

Cr-Commit-Position: refs/heads/master@{#314329}

This will simplify use of the API for consumers who don't know a
Profile* at the time of getting the callback.

BUG=none
R=robertshield@chromium.org

Review URL: https://codereview.chromium.org/891793002

Cr-Commit-Position: refs/heads/master@{#314328}

Previously we tried to refcount the tracking id to slot id mapping.
This broke in some circumstances where the number of press events was
not equal to the number of release events.

This patch switches to marking some touch events such that they don't
modify the mapping, simplifying logic, and fixing a nasty bug.

Initially landed here:
https://codereview.chromium.org/785753002/

Reverted here:
https://codereview.chromium.org/869433007/
due to memory failures.

BUG=439051
TEST=EventsXTest.TouchEventNotRemovingFromNativeMapping

Review URL: https://codereview.chromium.org/886593004

Cr-Commit-Position: refs/heads/master@{#314327}

Review URL: https://codereview.chromium.org/887043002

Cr-Commit-Position: refs/heads/master@{#314326}

Revert of [android] Remove libaddressinput Java widget. (patchset #1 id:1 of https://codereview.chromium.org/890103002/)

Reason for revert:
This CL broke the gn android build.

To reproduce locally on a chromium checkout src:

> gn gen --args='os="android" cpu_arch="arm" use_goma=true' out_android_gn/Debug
> ninja -C out_android_gn/Debug

eventually fails with:

[6011/25521] CXX obj/third_party/libaddressinput/chromium/libaddressinput.chrome_address_validator.o
<snip>
../../third_party/libaddressinput/chromium/chrome_address_validator.cc:156:68: error: call of overloaded 'pow(int, int)' is ambiguous
       GetBaseRetryPeriod() * pow(2, attempts_number_[region_code]++));

Original issue's description:
> [android] Remove libaddressinput Java widget.
>
> Android uses a custom UI with the cross-platform strings for address
> input.
>
> BUG=454120
>
> Committed: https://crrev.com/e7e8b5f5a53c3577b84413bd165731ca66d6a837
> Cr-Commit-Position: refs/heads/master@{#314217}

TBR=estade@chromium.org,cpu@chromium.org,rouslan@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=454120

Review URL: https://codereview.chromium.org/892403002

Cr-Commit-Position: refs/heads/master@{#314325}

The output directory is the same as the build directory, so let's be
consistent to not confuse new users.

BUG=441955
TEST=gn help args, gn help gen, gn help ls, gn help refs
R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/895433002

Cr-Commit-Position: refs/heads/master@{#314324}

BUG=452071

Review URL: https://codereview.chromium.org/894363002

Cr-Commit-Position: refs/heads/master@{#314323}

Revert of Roll opus to opus-HEAD-66611f1. (patchset #3 id:40001 of https://codereview.chromium.org/865853005/)

Reason for revert:
This CL breaks WebRTC in Chrome since the new Opus version is not bit-exact with the previous version.

Original issue's description:
> Roll opus to opus-HEAD-66611f1.
>
> The third_party/opus/src CL is https://codereview.chromium.org/882843002.
>
> Support MIPS optimizations in opus.
>
> R=henrika@chromium.org,sergeyu@chromium.org
> BUG=452984
>
> Committed: https://crrev.com/ebbf54dafa1be134a4d45fc30c103388dfaa8ff0
> Cr-Commit-Position: refs/heads/master@{#313646}

TBR=sergeyu@chromium.org,tlegrand@chromium.org,wtc@chromium.org
BUG=452984

Review URL: https://codereview.chromium.org/899623002

Cr-Commit-Position: refs/heads/master@{#314322}

TBR=machenbach@chromium.org

Review URL: https://codereview.chromium.org/900553004

Cr-Commit-Position: refs/heads/master@{#314321}

This CL implements AffiliationBackend::GetAffiliation(), including support for caching, but without support for some error scenarios.

The missing parts, viz. Prefetch() functionality and error handling will be added later to reduce CL size. Given that the code is not yet exercised outside of tests, this should not create any problems.

BUG=437865
TBR=gcasto@chromium.org

Review URL: https://codereview.chromium.org/868253005

Cr-Commit-Position: refs/heads/master@{#314320}

BUG=453773
DIFFBASE=880223002

Review URL: https://codereview.chromium.org/889893002

Cr-Commit-Position: refs/heads/master@{#314319}

Make sure that at most one profile corruption error dialog is shown at
any given time. Currently I ocasionally get six dialogs with an
identical error message when I start Chrome while the profile is locked.
Although, the ShowProfileErrorDialog method is called with two different
messages (corrupt profile vs incompatible profile), I have only seen one
of them showing up at any time, so the boolean guard is applied to any
dialog generated by ShowProfileErrorDialog.

R=pkasting@chromium.org
BUG=446732,447791
TEST=Manually:
1. Start Chrome to initialize the profile directory.
2. Close Chrome.
3. Mark database as read-only: chmod a-r "/tmp/profile/Default/Web Data"
4. Start Chrome again.
5. Verify that only one dialog is shown (instead of six).
6. Close Chrome.
7. Start Chrome, with the --noerrdialogs flag.
8. Verify that no error dialogs are displayed at all (instead of six).

Review URL: https://codereview.chromium.org/873253002

Cr-Commit-Position: refs/heads/master@{#314318}

Review URL: https://codereview.chromium.org/895903002

Cr-Commit-Position: refs/heads/master@{#314317}

Per bug description, the "HtmlType" attribute wasn't carried over when processing a cached form for quality assessment.

BUG=454151
TEST=AutofillMetricsTest

Review URL: https://codereview.chromium.org/891983002

Cr-Commit-Position: refs/heads/master@{#314316}

On Android, WebTouchPoint.state of all touches is set to "StateMoved" thus
included in changedTouches when any touch moves. This patch compares current
touch points position to previous position in TouchMove event and updates state.
This comparision is done after all filtering and potential coalescing
in TouchEventQueue.

BUG=416236

Review URL: https://codereview.chromium.org/788923002

Cr-Commit-Position: refs/heads/master@{#314315}

Before this change, when there were a lot of font fallbacks happening on
the web site, the fallback font was mmaped multiple times, reaching up to
200 times in the TC url: http://jsfiddle.net/p5pe81vs/, leading to OOMs
and crashes for renderer process.

This happens after the change introduced in
https://codereview.chromium.org/307243002

This CL keeps track of the mmaps for each font ID to avoid any further
unnecessary IPC requests and mmaps for future requests of the same font
ID that would result in new FD.

BUG=430021

NOPRESUBMIT=true
brettw indicates this use of ScopedAllowIO is acceptable.
Previous code eluded IO checks, this change makes the IO use find-able.

Committed: https://crrev.com/78db5e535ef48c596223fe272572e5679fbb44fd
Cr-Commit-Position: refs/heads/master@{#313102}

Review URL: https://codereview.chromium.org/697383002

Cr-Commit-Position: refs/heads/master@{#314314}

When the Ctrl+Enter are pressed, the click is simulated.
The test checks if the new tab is opened on the background.

BUG=450069
R=jochen@chromium.org

Review URL: https://codereview.chromium.org/877513002

Cr-Commit-Position: refs/heads/master@{#314313}

When announcing the availability of the robot account token to its
observers, DeviceOAuth2TokenService needs to know both the token and
the robot account ID. During enrollment, the token becomes available
before the account ID. Thus, DeviceOAuth2TokenService needs to wait
for the account ID to be available before announcing the token.

BUG=453828
TEST=New unit test

Review URL: https://codereview.chromium.org/892633003

Cr-Commit-Position: refs/heads/master@{#314312}

For the login screen, if there is some web content displayed, and the user clicks a link that is specified to open in a new window, a new browser window indeed opens, which is highly undesirable at the login screen.
To avoid this, a new delegate method was introduced, which should decide if we want to allow displaying any additional contents.

BUG=443096
TEST=manual

Review URL: https://codereview.chromium.org/812763003

Cr-Commit-Position: refs/heads/master@{#314311}

 - fix missing files check for the case when the script runs outside of
   the repository root;

 - exclude not shipped dirs that present on the bot from the checks.

BUG=450048

Review URL: https://codereview.chromium.org/877583003

Cr-Commit-Position: refs/heads/master@{#314310}

Rather than lazily creatng the schedueler (which could happen on any
thread), create it in the constructor if there is a message loop.
Frustratingly the problem this solves only shows up if you
run all the webkit_tests, not just one of them.

BUG=391005

Review URL: https://codereview.chromium.org/890333003

Cr-Commit-Position: refs/heads/master@{#314309}

Revert of Add build targets for Opus tests. (patchset #2 id:20001 of https://codereview.chromium.org/894473002/)

Reason for revert:
This CL breaks WebRTC in Chrome and must be reverted. Contact tlegrand@ for more details.

Original issue's description:
> Add build targets for Opus tests.
>
> test_opus_api, test_opus_encode, test_opus_decode, and test_opus_padding
> can be run directly, with no command-line options, and they print out
> "OK" during the test and an overall status message such as
> "All API tests passed." or "Tests completed successfully." at the end.
>
> The celt component has several unit tests. I didn't add build targets
> for them because some of them don't print any output, so you'd need a
> test harness script to examine their exit status.
>
> R=henrika@chromium.org,sergeyu@chromium.org
> BUG=452984
>
> Committed: https://crrev.com/8b76b75f78ffdb9760ee7eea4f149c5d2b64bc89
> Cr-Commit-Position: refs/heads/master@{#314235}

TBR=sergeyu@chromium.org,brettw@chromium.org,wtc@chromium.org, tlegrand@chromium.org
BUG=452984

Review URL: https://codereview.chromium.org/884483005

Cr-Commit-Position: refs/heads/master@{#314308}

Some printing related //android_webview were copied over from
//chrome. Now that the ones in //chrome can move to a component,
webview should also use these files.

Depends on https://codereview.chromium.org/811563008/

BUG=444883, 311308

Review URL: https://codereview.chromium.org/822133003

Cr-Commit-Position: refs/heads/master@{#314307}