GitLab

BUG=703113
TEST=create html notification, exit browser, wait until notification closes and/or close notification.

Review-Url: https://codereview.chromium.org/2803593003
Cr-Commit-Position: refs/heads/master@{#463377}

Cr-Commit-Position: refs/heads/master@{#463376}

This patch fixes a null deref crash during detachLayoutTree.

BUG=709916
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2809563003
Cr-Commit-Position: refs/heads/master@{#463375}

The method is unused. It was introduced in https://codereview.chromium.org/2445193005.

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2803163003
Cr-Commit-Position: refs/heads/master@{#463374}

BUG=707631
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2795763003
Cr-Commit-Position: refs/heads/master@{#463373}

It is necessary to notify the RenderWidgetHostViewChildFrame of the
surface associated with the child-frame, so that it can correctly notify
the embedder.

This fixes the login screen, opening pdf files in chrome, etc.

BUG=694846
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2804753002
Cr-Commit-Position: refs/heads/master@{#463372}

This should give better memory locality and fewer allocations for our
workloads.

Most dictionaries are small and previously sorted (JSON Chrome writes
like preferences is written in keyed-order). This makes the flat_map
a better storage option because it avoids separate allocation for each
node and has better memory locality.

Performance from base_perftests JSONPerfTest.StressTest:
  Old:                        84,807ms
  Just replacing the storage: 70,363ms
  Optimizing JSON parser:     69,286ms

Primarily the improvement is in read performance (in creating the
dictionary nodes) but write performance is also improved slightly,
presumably due to better memory locality.

I did some manual checking for the creation of large dictionary values
(where this change may perform poorly) and did not find any in normal
Chrome runs other than the JSON parser and the value copy constructor.

Bringing flat_map.h into values_unittests.cc seems to confuse MSVC's
ADL resolution of swap and it gives errors about calling
flat_map::swap with base::Values. I wasn't able to figure out the
source of this problem. The test is really checking for operator= that
is generated by the default std::swap implementation (base::Value
doesn't specialize a swap so the "using std/swap" pattern for ADL is a
no-op), so I qualified the call to swap with a std::.

Review-Url: https://codereview.chromium.org/2807953002
Cr-Commit-Position: refs/heads/master@{#463371}

Also updated base/chromoting_event.cc to use NameValueMap

Review-Url: https://codereview.chromium.org/2807803002
Cr-Commit-Position: refs/heads/master@{#463370}

This intergrates some work that has landed in remoting base in the last month,
remoting/client/ios now as access to common oauth classes in remoting/base. This
is the integration with those classes and it's use. The app is still in
development so there are still mocked methods and functions but this is a working
first pass.

R=joedow@chromium.org

BUG=652781

Review-Url: https://codereview.chromium.org/2794013005
Cr-Commit-Position: refs/heads/master@{#463369}

This change was motivated by the need to remove the
references to wheezy.  I noticed that the target_sysroot_dir
argument could be generalized which avoids the extra
conditional in sysroot.gni.

BUG=697494

Review-Url: https://codereview.chromium.org/2810633004
Cr-Commit-Position: refs/heads/master@{#463368}

Review-Url: https://codereview.chromium.org/2806643002
Cr-Commit-Position: refs/heads/master@{#463367}

https://chromium.googlesource.com/angle/angle.git/+log/bf67aa8..20e005b

BUG=chromium:708835

TBR=geofflang@chromium.org

TEST=bots

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

Review-Url: https://codereview.chromium.org/2807233002
Cr-Commit-Position: refs/heads/master@{#463366}

BUG=710024

Review-Url: https://codereview.chromium.org/2805343002
Cr-Commit-Position: refs/heads/master@{#463365}

BUG=676012

Review-Url: https://codereview.chromium.org/2801393002
Cr-Commit-Position: refs/heads/master@{#463364}

Similar to https://codereview.chromium.org/2804913002/
Tests should create, but not own the ChromeLauncherControllerImpl.
Instead, install the CLC as Shell's ShelfDelegate.

Also cleanup: expose command enum, remove friend tests.

Fixes the test from https://codereview.chromium.org/2798173002

BUG=557406,709729
TEST=Automated tests still pass; more closely match production.
R=jamescook@chromium.org

Review-Url: https://codereview.chromium.org/2812693002
Cr-Commit-Position: refs/heads/master@{#463363}

Removes --git-svn-go-deeper and -d/--default-lastchange.
No intended behavior change.

BUG=none

Review-Url: https://codereview.chromium.org/2808193002
Cr-Commit-Position: refs/heads/master@{#463362}

BUG=709939

Review-Url: https://codereview.chromium.org/2806423002
Cr-Commit-Position: refs/heads/master@{#463361}

After the content detectors were removed, the only remaining users of
these are in WebView (and we don't want to add any other users for this
primitive, legacy address parser) so move the files there and adjust
namespaces.

BUG=462596

Review-Url: https://codereview.chromium.org/2803163002
Cr-Commit-Position: refs/heads/master@{#463360}

Reporting is a spec for delivering out-of-band reports from various
other parts of the browser. See http://wicg.github.io/reporting/ for
the spec, or https://goo.gl/pygX5I for details of the planned
implementation in Chromium.

This implements the garbage collector, which cleans out reports that
haven't been delivered for too long or through too many attempts (and
also clears reports on network changes, if so configured).

BUG=704259

Review-Url: https://codereview.chromium.org/2740833004
Cr-Commit-Position: refs/heads/master@{#463359}

On Linux, older versions of UPower did not have the GetDisplayDevice
method. Trying to call it generates log spam.

Modernize the modified file with base::MakeUnique while we are here.

BUG=695383

Review-Url: https://codereview.chromium.org/2795713003
Cr-Commit-Position: refs/heads/master@{#463358}

Without this the bots try to use ash_mus_unittests, which doesn't
exist.

BUG=705715
TEST=this is test change
R=msw@chromium.org

Review-Url: https://codereview.chromium.org/2802163004
Cr-Commit-Position: refs/heads/master@{#463357}

This reverts commit 1204d239.

This introduced one bug and exposed another.  Yay!

BUG=709884,709886
TBR=bauerb,peconn

Review-Url: https://codereview.chromium.org/2806373003
Cr-Commit-Position: refs/heads/master@{#463356}

In the mode where CCT only shows the page title, we should never have an
empty title displayed. Force this by checking the title even if there
are no title update after Toolbar creation since those updates might
have been missing while prerendering. If there is no title set yet, post
a callback to check this and fallback to showing the domain if there is
not title set after a second.

BUG=708957

Review-Url: https://codereview.chromium.org/2806593003
Cr-Commit-Position: refs/heads/master@{#463355}

Try to simplify coordinate handling in NGBlockLayoutAlgorithm::Layout by
using NGLogicalSize instead of raw LayoutUnits for the block/inline dir.

R=glebl@chromium.org
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_layout_ng

Review-Url: https://codereview.chromium.org/2808863002
Cr-Commit-Position: refs/heads/master@{#463354}

Moved the timer task from per-renderer to per-frame in the following CL.
  https://codereview.chromium.org/2640293002/

At that time, the spec didn't mention it properly but it has become clear
recently. The spec says that we should use UserInteraction task source.

Related spec change:
  https://github.com/w3c/browser-payment-api/pull/405

BUG=none

Review-Url: https://codereview.chromium.org/2808573002
Cr-Commit-Position: refs/heads/master@{#463353}

When Crashpad self-monitoring is enabled, crashes in crashpad_handler
will be written to the crash report database and uploaded according to
the user's preference. Self-monitoring dedicates an additional
crashpad_handler process to serve as the handler for the original
instance. Having a dedicated process is somewhat expensive, so
self-monitoring is not enabled at all times.

An existing mechanism, the "fallback crash handler," is already in place
on Windows. The fallback crash handler does not create any additional
processes until a crashpad_handler crash occurs. The fallback handler
remains effective in cases where Crashpad's own self-monitoring is
disabled. The fallback handler and Crashpad self-monitoring have
different attributes, so there are no plans to move Windows entirely
onto Crashpad self-monitoring.

For branded Chrome, the dev channel gets Crashpad self-monitoring in 25%
of browser starts, and the beta channel gets it in 10% of browser
starts. On macOS, the stable channel gets self-monitoring in 1% of
browser starts, and the canary and Chromium builds get it all the time.
On Windows, the Chrome stable channel never gets self-monitoring because
the fallback crash handler is sufficient, and the canary and Chromium
builds get it 50% of the time, allowing for some use of the fallback
crash handler in those cases.

While reports will be collected for crashes that occur while Crashpad is
attempting to upload another report, it's unlikley that these reports
will be uploaded automatically under the current one-attempt-per-hour,
no-retry rate limiting strategy. See
https://crashpad.chromium.org/bug/23.

BUG=crashpad:143
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.win:win10_chromium_x64_rel_ng

Review-Url: https://codereview.chromium.org/2799013002
Cr-Commit-Position: refs/heads/master@{#463352}

BUG=709930
R=dcheng@chromium.org,thakis@chromium.org,dpranke@chromium.org

Review-Url: https://codereview.chromium.org/2808813002
Cr-Commit-Position: refs/heads/master@{#463351}

Add icon to the Data Saver footer, update the strings to the new spec,
and add a divider.

BUG=702799

Review-Url: https://codereview.chromium.org/2795653006
Cr-Commit-Position: refs/heads/master@{#463350}

https://skia.googlesource.com/skia.git/+log/b8ab7f7634fc..4482a14db9b1

$ git log b8ab7f763..4482a14db --date=short --no-merges --format='%ad %ae %s'
2017-04-10 mtklein fix -Fast bot
2017-04-10 bsalomon constexprify GrBlend.h and GrPorterDuffXferProcessor.cpp
2017-04-10 brianosman Remove Index_8 handling in cacherator heuristic code
2017-04-10 borenet Recipes: Remove depot_tools path hack

Created with:
  roll-dep src/third_party/skia


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.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=mtklein@chromium.org

Change-Id: I41dee01fac1b0d0270b6f667ef92d0a9d10aabff
Reviewed-on: https://chromium-review.googlesource.com/473406Reviewed-by: Skia Deps Roller <skia-deps-roller@chromium.org>
Commit-Queue: Skia Deps Roller <skia-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#463349}

This CL implements a TODO to use the original URL of the favicon
downloaded from the Google server as the key in the database.

BUG=644102

Review-Url: https://codereview.chromium.org/2810733002
Cr-Commit-Position: refs/heads/master@{#463348}

This change makes it safe to call HidConnection::Read from within the
callback passed to a previous call to this method. On macOS and Linux
this means moving protected input report filtering to before the report
is queued. On Windows this wasn't an issue but filtered input reports
would cause a callback to never be executed.

BUG=None

Review-Url: https://codereview.chromium.org/2804313005
Cr-Commit-Position: refs/heads/master@{#463347}

The method VideoCaptureDeviceFactory::EnumerateDeviceDescriptors()
is the asynchronous equivalent of method VideoCaptureDeviceFactory
::GetDeviceDescriptors(). Its current implementation delegates to
GetDeviceDescriptors() and invokes the callback synchronously.

This CL removes EnumerateDeviceDescriptors() and lets all clients
call GetDeviceDescriptors() directly.

BUG=708233

Review-Url: https://codereview.chromium.org/2805863002
Cr-Commit-Position: refs/heads/master@{#463346}

This CL is identical to https://codereview.chromium.org/2592143003,
except that RemoveObserver() works when called from any sequence
(existing code rely on that).

Previously, observers could only be added to an ObserverListThreadSafe
from single-threaded tasks. Observers were notified on the registration
thread.

With this CL, observers can also be added to an ObserverListThreadSafe
from sequenced tasks. They are notified on the registration sequence.

ObserverListThreadSafe behaves almost the same way as before when used
from single-threaded tasks. The following things changed:
- The order in which observers registered from the same thread
  are notified is no longer deterministic.
- One notification task is posted for each observer rather than one
  notification task per thread.
- If an observer is added to a NOTIFY_ALL ObserverListThreadSafe from
  a stack of notification dispatches, only the notification on top of
  the stack is sent to the newly added observer.

BUG=675631

Review-Url: https://codereview.chromium.org/2805933002
Cr-Commit-Position: refs/heads/master@{#463345}

Removes the promo_type_ field from notification_promo.h/.cc
and uses of it. It is unnecessary as the only promo type in use is the
MOBILE_NTP_WHATS_NEW_PROMO.

BUG=608525

Review-Url: https://codereview.chromium.org/2805443007
Cr-Commit-Position: refs/heads/master@{#463344}

Ensure the image has loaded before testing layout.

TBR=eae@chromium.org

BUG=707918

Review-Url: https://codereview.chromium.org/2807973004
Cr-Commit-Position: refs/heads/master@{#463343}

Payment manifest has been split in two:

   1. Payment method manifest. Example:

   {
    "default_applications": ["https://bobpay.com/app.json"],
    "supported_origins": ["https://alicepay.com"]
   }

   2. Web app manifest. Example:

   {
    "related_applications": [{
      "platform": "play",
      "id": "com.bobpay.app",
      "min_version": "1",
      "fingerprints": [{
        "type": "sha256_cert",
        "value": "59:5C:88:65:FF:C4:E8:20:CF:F7:3E:C8:64:D0:95:F0:06:19"
      }]
    }]
   }

Only "default_applications" is used in this patch. The
"supported_origins" field will be used in an upcoming patch.

Using these manifests is behind a flag:
chrome://flags/#android-payment-apps

Intent to implement:
https://groups.google.com/a/chromium.org/d/msg/blink-dev/YlMpaO-E2mE/-_GgiMiRCgAJ

Design:
https://docs.google.com/document/d/1izV4uC-tiRJG3JLooqY3YRLU22tYOsLTNq0P_InPJeE

Explainer:
https://github.com/zkoch/zkoch.github.io/blob/master/pmi-v2.md

BUG=683329

Review-Url: https://codereview.chromium.org/2802043002
Cr-Commit-Position: refs/heads/master@{#463342}

Cleaning up before I head off into the sunset.

BUG=none
R=nparker@chromium.org

Review-Url: https://codereview.chromium.org/2810823002
Cr-Commit-Position: refs/heads/master@{#463341}

It is not being used in any IPC messages.

Review-Url: https://codereview.chromium.org/2806083002
Cr-Commit-Position: refs/heads/master@{#463340}

The settings button was unintentionally broken for Windows and
Linux users in 2799653004, as was gathering of some UMA. This
CL offers a slight refactoring that meets all goals.

BUG=706658

Review-Url: https://codereview.chromium.org/2807623002
Cr-Commit-Position: refs/heads/master@{#463339}

The reference to TextInputManager instance (owned by WebContentsImpl) in RenderWidgetHostView is alive as long as
  1) WebContents is alive,
  2) RenderWidgetHostView is registered with TextInputManager.

On Mac, WebContents destruction might not lead to immediate destruction of the RenderWidgetHostViewMac. Therefore,
there can be a period of time where (after WebContentsImpl is deleted) |text_input_manager_| is null. Then it is only
safe to check |text_input_manager_| before dereferencing it.

BUG=709864

Review-Url: https://codereview.chromium.org/2813493003
Cr-Commit-Position: refs/heads/master@{#463338}