GitLab

Fix various outdated version checks in the Android code:

1) Remove comparisons to Build.VERSION.CODENAME that refer to released
   SDKs and replace with correct SDK_INT comparisons instead.

2) Compare SDK_INT to constants instead of integer literals.

3) Remove old ICS-only code from NFC BeamCallback since we don't support
   pre-JB any more.

BUG=724622
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/2895293002
Cr-Commit-Position: refs/heads/master@{#473998}

Web developers should use 'basic-card' instead. Deprecated:

  new PaymentRequest(
    [{
      supportedMethods: ['visa', 'amex'],
    }],
    cart);

Should be used:

  new PaymentRequest(
    [{
      supportedMethods: ['basic-card'],
      data: {supportedNetworks: ['visa', 'amex']},
    }],
    cart);

Console warning message:
  [Deprecation] Card issuer network ("amex", "diners", "discover",
  "jcb", "mastercard", "mir", "unionpay", "visa") as payment method is
  deprecated and will be removed in M64, around January 2018. Please use
  payment method name "basic-card" with issuer network in the
  "supportedNetworks" field instead. See
  https://www.chromestatus.com/features/5725727580225536 for more
  details.

Intent to deprecate thread:
https://groups.google.com/a/chromium.org/d/msg/blink-dev/1udpnrlQK4Q/ot_E-htQAQAJ

BUG=719526

Review-Url: https://codereview.chromium.org/2865023003
Cr-Commit-Position: refs/heads/master@{#473997}

Skia is starting to use more advanced runtime-detected
CPU features, and this is problematic when running layout tests
(results are now machine-dependent).

Add a flag to disable Skia's runtime optimizations, and always
apply it when running layout tests.

BUG=skia:6657

Change-Id: I98e371ea60a5b0a290ba869c9134afe328ccae90
Reviewed-on: https://chromium-review.googlesource.com/512942Reviewed-by: Mike Klein <mtklein@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Cr-Commit-Position: refs/heads/master@{#473996}

 * Announce when the panel is opened and closed.
 * Prevent focus from going to the tab content when the bottom
   sheet is open.
 * Change the in-product help bubble when the expand button is
   enabled and use different strings for non-accessibility mode
   and accessibility mode.
 * Announce when a different bottom sheet tab is selected.
 * When the expand button isn't visible, make the entire toolbar
   focusable for accessibility and add a content description.

BUG=724694,724696,724693,724695

Review-Url: https://codereview.chromium.org/2895223003
Cr-Commit-Position: refs/heads/master@{#473995}

This will allow us to track when requests finish.  Adds a hook so that
internal code can have different behavior.  See bug for details.

BUG=720782

Review-Url: https://codereview.chromium.org/2888273003
Cr-Commit-Position: refs/heads/master@{#473994}

This CL adds base::Optional<int32_t> return type for H264POC::ComputePicOrderCnt(...).

BUG=None
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/2902693002
Cr-Commit-Position: refs/heads/master@{#473993}

BUG=722553
TEST=covered by tests
R=msw@chromium.org

Review-Url: https://codereview.chromium.org/2901663003
Cr-Commit-Position: refs/heads/master@{#473992}

This renames swap() to Swap() to harmonize with ComPtr::Swap().

BUG=638327
TBR=thestig@chromium.org
Mechanical rename.

Review-Url: https://codereview.chromium.org/2895313002
Cr-Commit-Position: refs/heads/master@{#473991}

BUG=723730

Review-Url: https://codereview.chromium.org/2901043002
Cr-Commit-Position: refs/heads/master@{#473990}

TalkBack and other accessibility services on Android now pay attention to
LocaleSpans to determine what language to use when speaking text. Wrap
text in AccessibilityNodeInfo in a LocaleSpan indicating the correct
language, but for efficiency, only do this when it's different than the
system language.

BUG=725134

Review-Url: https://codereview.chromium.org/2902533002
Cr-Commit-Position: refs/heads/master@{#473989}

fixes link errors for NEON init functions in certain configurations
(android + arm + arm_use_neon=false). in cases where NEON is disabled
stubs are produced, in the rest the NEON is compiled and in the android
case support is detected at runtime.

BUG=725225

Change-Id: I3b5db31bdd8f5f0be7778efa7e6df0bdc98fd674
Reviewed-on: https://chromium-review.googlesource.com/511382Reviewed-by: Paul Miller <paulmiller@chromium.org>
Reviewed-by: Urvang Joshi <urvang@chromium.org>
Commit-Queue: Urvang Joshi <urvang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#473988}

Special case for Android list items no longer needed to remove name, because list items are no longer name from contents.

BUG=None

Review-Url: https://codereview.chromium.org/2901003002
Cr-Commit-Position: refs/heads/master@{#473987}

This change exposes a gamepad with a single button, which is driven by the touch screen.

BUG=718665

Review-Url: https://codereview.chromium.org/2879973002
Cr-Commit-Position: refs/heads/master@{#473986}

The motivations for using Skia encoders are:
***Support for embedding ICC profiles
***Support for linear unpremultiplication and blending
***Support for multiple color types (including F16)
***Support for multiple alpha types

As a long term refactoring goal, we should be able to delete
the WebKit image encoders and the ui/gfx image encoders, thus
leaving one set of image encoders to maintain.

BUG=713862

Review-Url: https://codereview.chromium.org/2891373002
Cr-Commit-Position: refs/heads/master@{#473985}

On Chrome OS, IME candidate window is anchored to either bottom-left
of the whole composition text (when its WindowPosition settings is
"composition") or bottom-left of the cursor (when it's "cursor".)
https://developer.chrome.com/extensions/input_ime#type-WindowPosition

The former setting doesn't have any issues regarding IME candidate
window positioning, but the latter (anchor to "cursor") does. The
cursor position may change every single time when you press the
convert key because for views::TextField, or more precisely for its
gfx::RenderText, the cursor position is defined as the bottom-right
of the selected range (if any) but the width of the selected range
may change on each conversion. For example, when pressing the
conversion key, a character in the selected range may change from a
full-width Hiragana Japanese character to a half-width Katakana
character.

The CL swaps |start| and |end| of the selected range on Chrome OS
so that the cursor position becomes bottom *left* of the range which
never changes during conversion. Note that Blink does the same
(i.e. placing the cursor at the beginning of the selection range)
to prevent the window flicker.

This CL is for Chrome OS only and doesn't change other platforms'
behavior. This does not affect Chrome OS IMEs that use "composition"
WindowPosition.

Demo:
 Omnibox without this fix: https://goo.gl/WVpjhE
 Omnibox with this fix: https://goo.gl/FH1fs9

BUG=311206
TEST=Follow the repro steps in crbug.com/311206 to confirm Omnibox
 no longer has the issue. Do the same with the find bar (Ctrl+F).

Review-Url: https://codereview.chromium.org/2886313002
Cr-Commit-Position: refs/heads/master@{#473984}

Moves the tests in mus_gpu_unittests and resource_coordinator_unittest
into services_unittests and remove references to the old targets.

BUG=722546

Review-Url: https://codereview.chromium.org/2901623002
Cr-Commit-Position: refs/heads/master@{#473983}

Revert of [WebPayments] Adding FilterProfilesForShipping to profile comparator (patchset #5 id:80001 of https://codereview.chromium.org/2884393002/ )

Reason for revert:
Appears to be causing test failures in https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Fchromium.win%2FWin10_Tests_x64%2F11929%2F%2B%2Frecipes%2Fsteps%2Fbrowser_side_navigation_browser_tests_on_Windows-10-10586%2F0%2Flogs%2FPaymentRequestProfileListTest.PrioritizeCompleteness%2F0

 RUN      ] PaymentRequestProfileListTest.PrioritizeCompleteness
[640:3644:0523/102806.488:WARNING:embedded_test_server.cc(219)] Request not handled. Returning 404: /favicon.ico
c:\b\c\b\win\src\chrome\browser\ui\views\payments\profile_list_view_controller_browsertest.cc(46): error:       Expected: partial
      Which is: b5e544d1-6a66-49bf-bd4c-f2eb38671fd3 https://www.example.com/ Jane A. Smith jsmith@example.com ACME      48838  US  13105557889
To be equal to: *profiles[0]
      Which is: 63b73e74-aacb-4a0e-bde5-a3260918586c http://www.example.com/ John H. Doe johndoe@hades.com Underworld 666 Erebus St. Apt 8  Elysium CA 91111  US  16502111111
c:\b\c\b\win\src\chrome\browser\ui\views\payments\profile_list_view_controller_browsertest.cc(47): error:       Expected: complete
      Which is: 63b73e74-aacb-4a0e-bde5-a3260918586c http://www.example.com/ John H. Doe johndoe@hades.com Underworld 666 Erebus St. Apt 8  Elysium CA 91111  US  16502111111
To be equal to: *profiles[1]
      Which is: b5e544d1-6a66-49bf-bd4c-f2eb38671fd3 https://www.example.com/ Jane A. Smith jsmith@example.com ACME      48838  US  13105557889
[  FAILED  ] PaymentRequestProfileListTest.PrioritizeCompleteness, where TypeParam =  and GetParam() =  (1292 ms)

Original issue's description:
> [WebPayments] Adding FilterProfilesForShipping to profile comparator
>
> BUG=722949
>
> Review-Url: https://codereview.chromium.org/2884393002
> Cr-Commit-Position: refs/heads/master@{#473929}
> Committed: https://chromium.googlesource.com/chromium/src/+/eb8d574b98adfcbd3b5ce2eb6e521b83efbb428a

TBR=mahmadi@chromium.org,mathp@chromium.org,tmartino@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=722949

Review-Url: https://codereview.chromium.org/2897133002
Cr-Commit-Position: refs/heads/master@{#473982}

With the new convertible and detachable form factors for
laptops it is important to make sure the interaction media
features are updated whenever the keyboard/trackpad combo is
flipped (so inactive) or detached. This will allow content
author to react to media query changes to adapt the user
interface to better suit the new interaction method (often
touch vs trackpad).

The patch make sure that InputDevices are enabled by default. This
will make sure that the MQ can be set correctly. There is no reason
to keep track of the input device status on x11 as whenever the input
list is changed the list of devices is updated entirely (at least in
DeviceDataManager). However due to previous patch with CL
https://codereview.chromium.org/2827803002/ InputDevices were disabled
by default (due to ChromeOS and refactoring in the CL). In order to
keep the same behavior on ChromeOS we just mark the input device
as disabled in EventConverterEvDev which is the one carrying the
InputDevice instance.

BUG=442418

Change-Id: I95cc0f6c6ee8533046afe605ebcbc04e963b98b3
Reviewed-on: https://chromium-review.googlesource.com/508212Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org>
Commit-Queue: Alexis Menard <alexis.menard@intel.com>
Cr-Commit-Position: refs/heads/master@{#473981}

When a URL focus is in progress, we reset toolbar button visibility.
The expand button visibility needs to be guarded on whether the button
is enabled.

BUG=725392

Review-Url: https://codereview.chromium.org/2896303002
Cr-Commit-Position: refs/heads/master@{#473980}

SequencedWorkerPool is being deprecated in favor of TaskScheduler.

BUG=667892
R=jochen@chromium.org

Review-Url: https://codereview.chromium.org/2876213002
Cr-Commit-Position: refs/heads/master@{#473979}

Crashes are being observed in the Origin Trials token validation flow.
The crashes seem occur during service worker startup, trying to validate
stored tokens. This path uses ChromeOriginTrialPolicy, which is lazily
created. That creation path does not look thread-safe, so the theory is
there is a race during initialization causing the crash.

This CL implements a lock on creation/access to the policy object, to
prevent a race.

Bug: 714992,721144
Change-Id: I7e842f68af54d58857d8b43b592a7125c0daecab
Reviewed-on: https://chromium-review.googlesource.com/506509Reviewed-by: Nico Weber <thakis@chromium.org>
Reviewed-by: Ian Clelland <iclelland@chromium.org>
Commit-Queue: Jason Chase <chasej@chromium.org>
Cr-Commit-Position: refs/heads/master@{#473978}

The chrome.runtime API has a method called `getPackageDirectoryEntry`
which has custom bindings to retrieve the app's directory entry (a la
HTML5 filesystem) and return it. The custom bindings for this are
quite strange, since it needs to be done from the app's background
page.

Move this to the file_entry_binding_util.js file, where the similar
getFileBindingsForApi() function is implemented. In addition to grouping
similar logic, this solves a native bindings issue where with trying to
require an API directly (which is disallowed).

BUG=653596
TBR=michaelpg@chromium.org (moving a resource)

Review-Url: https://codereview.chromium.org/2895493004
Cr-Commit-Position: refs/heads/master@{#473977}

The IPC message BrowserPluginHostMsg_ImeFinishComposingText does not have a
parameter for the instance ID of BrowserPlugin and therefore, is dropped on
the browser side inside BrowserPluginMessageFilter. This causes a regression
in (Japanese) IME where clicking during an ongoing composition does not commit.

BUG=720023

Review-Url: https://codereview.chromium.org/2887973002
Cr-Commit-Position: refs/heads/master@{#473976}

Pass it directly to RecordPaintCanvas and ToSkPicture, and other skia
methods which is where it is used.

This allows us to more easily get rid of cc::DisplayItem and its
subclasses, replacing them with a PaintOpBuffer in DisplayItemList
directly instead. The difficulty I faced with that was that if
DisplayItemList has a single PaintOpBuffer, then it has a single
cull rect. However when painting, each "batch" of PaintOps can
have a different cull rect (corresponding to the PaintOps that
would have been in a single DisplayItem before). So, instead the
cull rect should be a property of recording at the
RecordPaintCanvas level, which is a temporary object. As such,
creators of cc::RecordPaintCanvas (mostly thru cc::PaintRecorder)
need to manage the cull rect themselves to pass to things that
want to use it with the cc::PaintOpBuffer (aka cc::PaintRecord at
this time).

Original code review was done on gerrit:
https://chromium-review.googlesource.com/c/503472

R=chrishtr@chromium.org, enne@chromium.org, pdr@chromium.org
BUG=671433, 646010, 724367
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2889653002 .
Cr-Original-Commit-Position: refs/heads/master@{#472917}
Committed: https://chromium.googlesource.com/chromium/src/+/c5f1b6126a7657234b9abc0c4359cbab45850b69
Review-Url: https://codereview.chromium.org/2889653002
Cr-Commit-Position: refs/heads/master@{#473975}

The previously generated skiafied 2x icon of the ON status of NightLight
was misaligned with the OFF status icon.

BUG=721952
TEST=Test on kevin, or on the emulator with --force-device-scale-factor=2.
Make sure the ON and OFF icons are aligned.

Review-Url: https://codereview.chromium.org/2900083002
Cr-Commit-Position: refs/heads/master@{#473974}

conformance2/textures/image_data/'tex-3d-rgb565-rgb-unsigned_short_5_6_5.html fails on Linux AMD R7 240

BUG=725556
TEST=WebGL2 Conformance / Linux AMD R7 240 bot turns green
TBR=kbr@chromium.org
NOTRY=true
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/2900103002
Cr-Commit-Position: refs/heads/master@{#473973}

Should fix the stack overflow issue for selectors with an excessive
amount of compound selectors.

R=meade@chromium.org
BUG=719374

Change-Id: I2a1cfb8cb2d00d96f8d46a6e7317c5871020c6cd
Reviewed-on: https://chromium-review.googlesource.com/506020Reviewed-by: Eddy Mead <meade@chromium.org>
Commit-Queue: Rune Lillesveen <rune@opera.com>
Cr-Commit-Position: refs/heads/master@{#473972}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/4eb9e4f0..03172a0f

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/2904553002
Cr-Commit-Position: refs/heads/master@{#473971}

BUG=679905

Review-Url: https://codereview.chromium.org/2900053002
Cr-Commit-Position: refs/heads/master@{#473970}

https://chromium.googlesource.com/external/github.com/google/googletest/+log/8c7f93fed..42bc671f4

Almost all no-ops. Pulling this in for
https://chromium.googlesource.com/external/github.com/google/googletest/+/b2cbbec04c141801dc9f648dc92b26bac5bad0da

BUG=550065

Review-Url: https://codereview.chromium.org/2901773003
Cr-Commit-Position: refs/heads/master@{#473969}

There was an off by one error that was making it 49 chars.  The
previous solution wouldn't work if the base ID was less than 25
chars either.  This solution should work for any non-empty ID.

TBR=yusufo@chromium.org
TEST=DCHECK doesn't fail
BUG=

Review-Url: https://codereview.chromium.org/2900963004
Cr-Commit-Position: refs/heads/master@{#473968}

Hi bokan@, what do you think of this possible rename?

Review-Url: https://codereview.chromium.org/2895863003
Cr-Commit-Position: refs/heads/master@{#473967}

Gtk2 did not have a global window scaling setting (so the only way to
scale up widgets was with a custom theme).  However, it did allow font
scaling with gdk-xft-dpi (backed by XSetting Xft/DPI).

Gtk3 adds (non-fractional) global window scaling with
gdk-window-scaling-factor (XSetting Gdk/WindowScalingFactor).  To
ensure that fonts were not scaled up twice (once from
gdk-window-scaling-factor and once from gdk-xft-dpi), a new setting
was added in [1] that overrides gdk-xft-dpi: gdk-unscaled-dpi
(XSetting Gdk/UnscaledDPI).  gdk-xft-dpi was kept around for
compatibility with apps like Chromium that still need it.

When modifying these settings, an invariant should be maintained:
gdk-xft-dpi = gtk-window-scaling-factor * gdk-unscaled-dpi.  Chromium
should have been able to keep using gdk-xft-dpi, but this invariant is
violated when changing the settings using gnome-tweak-tool, where I
have gdk-window-scaling-factor = 2, gdk-unscaled-dpi = 98304, and
gdk-xft-dpi = 98304 (gdk-xft-dpi should be 196608).

[2] changed Gtk builds to use the window scaling factor, which is
incorrect because it did not consider font scaling, making fractional
scaling impossible.

This CL takes gdk-unscaled-dpi into the calculation as well, and
continues to fallback on using gdk-xft-dpi if the other variables are
unavailable, which can happen on Gtk2.

[1] https://git.gnome.org/browse/gtk+/commit/?id=4b9c08f48d6f5be43b0795d3eee462d60b5f9e1f
[2] https://codereview.chromium.org/2869763004

BUG=723931
R=erg@chromium.org,chris.coulson@canonical.com
CC=oshima@chromium.org

Review-Url: https://codereview.chromium.org/2899943002
Cr-Commit-Position: refs/heads/master@{#473966}

Fixes hintWidth calculation when we reverse animation.

BUG=724624

Review-Url: https://codereview.chromium.org/2897753002
Cr-Commit-Position: refs/heads/master@{#473965}

Syntax error corrected in network traffic annotations of:
chrome/browser/extensions/install_signer.cc
extensions/browser/updater/extension_downloader.cc

BUG=656607

Review-Url: https://codereview.chromium.org/2900953002
Cr-Commit-Position: refs/heads/master@{#473964}

When a secondary UI dialog is opened, the omnibox icon it's anchored to
need to set its state as active. When the dialog is dismissed, the icon is
set to inactive.

BubbleAnchorHelper provides functions to retrieve the omnibox icons
and sets the active state.

BUG=715863

Review-Url: https://codereview.chromium.org/2882533003
Cr-Commit-Position: refs/heads/master@{#473963}

SupportsWeakPtr's weak pointers are invalidated only after derived class
members are destroyed which may lead to use-after-destroy issues.

This CL removes SupportsWeakPtr from GLES2Decoder and adds a WeakPtrFactory
instance to each of its subclasses.

BUG=None
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/2898953002
Cr-Commit-Position: refs/heads/master@{#473962}

The rectangle highlight was mostly faded out, with an occasional pulse
to fade in.  The spec calls for the opposite.  Switching the animation.

BUG=724996

Review-Url: https://codereview.chromium.org/2902573002
Cr-Commit-Position: refs/heads/master@{#473961}

The automation API specifies the hit test coordinates should be
integers which is fine since there's no reason for sub-pixel
precision.

However, it crashes if you pass it a non-integer, which is not good!
It should just convert to an integer.

BUG=725159
TBR=aboxhall

Review-Url: https://codereview.chromium.org/2900803003
Cr-Commit-Position: refs/heads/master@{#473960}

Requiring quoted URLs was cargo-culted from the HPKP implementation. The HPKP
spec does not actually say that report-uris must be quoted -- it's simply that
all the examples quote them. So it's possibly a bug that Chrome's HPKP implementation
requires quoted report-uris. The Expect-CT spec doesn't say anything about quoting
report-uris nor do I see a reason that it should, so Chrome's implementation shouldn't
require them.

BUG=679012

Review-Url: https://codereview.chromium.org/2895373002
Cr-Commit-Position: refs/heads/master@{#473959}