GitLab

https://skia.googlesource.com/skia.git/+log/0c984a0af309..a6ae1f7cda07

$ git log 0c984a0af..a6ae1f7cd --date=short --no-merges --format='%ad %ae %s'
2017-03-16 ethannicholas wired up SPIR-V validator
2017-03-15 mtklein Support all gradient types in SkColorSpaceXformCanvas.
2017-03-16 bsalomon Outset dst texture bounds by a pixel to fix corruption in dstreadshuffle.
2017-03-16 robertphillips Remove budgeted parameter from SkSurface::makeImageSnapshot
2017-03-15 ccameron Add SK_API to SkCreateColorSpaceXformCanvas
2017-03-16 jvanverth Rename GrAADistanceFieldPathRenderer to GrSmallPathRenderer
2017-03-15 bsalomon Make SkTArray not allocate unless reserve or initial count > 0

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
TBR=borenet@chromium.org

Change-Id: I6171bf89ffb54c161088e835e7dcf011cdc9ef01
Reviewed-on: https://chromium-review.googlesource.com/455354Reviewed-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@{#457445}

BUG=699774

Review-Url: https://codereview.chromium.org/2753753006
Cr-Commit-Position: refs/heads/master@{#457444}

Previously, animations in WAITING_FOR_DELETION state would only be purged via
AnimationPlayer::PushPropertiesTo, which synchronizes impl- and main-thread
players.  But the player in ScrollOffsetAnimationsImpl has no main-thread copy,
so its animations never got deleted after they finished.

BUG=701025
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2757453003
Cr-Commit-Position: refs/heads/master@{#457443}

Popups are ephemeral views used to send commands.
When tab is deselected, they should be dismissed, specially popups relative to
the tab (page info).

BUG=697506, 697470

Review-Url: https://codereview.chromium.org/2729563003
Cr-Commit-Position: refs/heads/master@{#457442}

Also remove stray print.

BUG=702201
TBR=nednguyen@google.com

Review-Url: https://codereview.chromium.org/2752043003
Cr-Commit-Position: refs/heads/master@{#457441}

Trivial one-pixel rasterization diffs between Release and Debug builds,
Win7-only.

Cannot be solved by rebasline.

See discussion on http://crbug.com/484370 also.

BUG=702176
R=schenney,guidou

Review-Url: https://codereview.chromium.org/2751383002
Cr-Commit-Position: refs/heads/master@{#457440}

This removes all cases of virtual from DisplayItem, making them
into just data types. Makes the fields public like DrawQuads, but
also const.

R=enne@chromium.org
BUG=671440
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2751833007
Cr-Commit-Position: refs/heads/master@{#457439}

BUG=699587

Review-Url: https://codereview.chromium.org/2752643004
Cr-Commit-Position: refs/heads/master@{#457438}

Replacing ASSERT and ASSERT_NOT_REACHED in core/style/
with DCHECK_op, NOTREACHED respectively.

BUG=697688

Review-Url: https://codereview.chromium.org/2751653003
Cr-Commit-Position: refs/heads/master@{#457437}

Network traffic annotation is added to network request of
chrome/browser/safe_search_api/safe_search_url_checker.cc.

BUG=656607

Review-Url: https://codereview.chromium.org/2743923002
Cr-Commit-Position: refs/heads/master@{#457436}

BUG=702194
TBR=nednguyen@google.com

Review-Url: https://codereview.chromium.org/2754053002
Cr-Commit-Position: refs/heads/master@{#457435}

The baseline-position syntax has changed recently, so we need to update
the CSS properties using the old syntax. This patch address only the
content-alignment (align-content and justify-content).

I used this patch to adapt the implementation of the parsing logic for
these properties to the new Blink's CSS Parsing Design.

The new baseline syntax is "[first | last ]? baseline" which implies
modifying the parsing and computed value logic.

There are several layout tests affected by this change, so I'll update
them accordingly.

BUG=565883

Review-Url: https://codereview.chromium.org/2750843002
Cr-Commit-Position: refs/heads/master@{#457434}

BUG=angleproject:1874
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/2725873007
Cr-Commit-Position: refs/heads/master@{#457433}

BUG=702194

Review-Url: https://codereview.chromium.org/2758533003
Cr-Commit-Position: refs/heads/master@{#457432}

When using ozone_auto_platforms=true change the default Ozone platform
to be X11 for Chrome OS and Linux Desktop. This does not change what
Ozone platforms are built. The CQ trybots already runs tests using Ozone
X11 by default, this is just to make developers are also using the same
configuration.

The Ozone platform can always be picked at runtime with
--ozone-platform=X command line flag.

BUG=671355

Review-Url: https://codereview.chromium.org/2730733002
Cr-Commit-Position: refs/heads/master@{#457431}

BUG=701999
TBR=skau@chromium.org

Review-Url: https://codereview.chromium.org/2755923002
Cr-Commit-Position: refs/heads/master@{#457430}

Currently, when SVGImages need to be painted via shaders (e.g. rounded
borders) we build an SkImage backed by an SkPictureImageGenerator.  The
problem is SkPictureImageGenerator has an intrinsic resolution, so it
locks the rasterization scale prematurely.  If we later scale this
SkImage (pinch zoom, or device scale factor on certain platforms), we're
scaling a bitmap - not a vector recording.

To avoid this, implement an applyShader() override and use a
SkPictureShader (which does not lock the resolution) instead of
SkImage/SkPictureImageGenerator.

BUG=696575
R=fs@opera.com,schenney@chromium.org

Review-Url: https://codereview.chromium.org/2753633003
Cr-Commit-Position: refs/heads/master@{#457429}

Use HTTP test load-and-stall script to partially load the initial
frame of a animated GIF image.

Render the image over CSS background color green and confirm that
the CSS color appears in the undecoded regions of the image.

TBR=scroggo@chromium.org
BUG=702148

Review-Url: https://codereview.chromium.org/2756443005
Cr-Commit-Position: refs/heads/master@{#457428}

BufferIdAllocator was unnescessarily getting created in
ResourceProvider's ctor as it was not being used anywhere.
This commit removes IdAllocator/BufferIdAllocator class
and its allocation.

BUG=
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2751953003
Cr-Commit-Position: refs/heads/master@{#457427}

Network traffic annotation is added to network request of
content_hash_fetcher.cc.

BUG=656607

Review-Url: https://codereview.chromium.org/2728313002
Cr-Commit-Position: refs/heads/master@{#457426}

There is no longer any need to use a template for MonochromeApplication.
This CL adds a proper MonochromeApplication.java and changes
monochrome_public_apk to use it.

The template code will be removed in a follow-up CL once downstream
targets no longer depend on it.

BUG=698004

Review-Url: https://codereview.chromium.org/2739503002
Cr-Commit-Position: refs/heads/master@{#457425}

Network traffic annotation is added to network request of:
chrome/browser/supervised_user/child_accounts/family_info_fetcher.cc
c/b/s/child_accounts/permission_request_creator_apiary.cc
c/b/s/experimental/safe_search_url_reporter.cc
c/b/s/legacy/supervised_user_refresh_token_fetcher.cc

BUG=656607

Review-Url: https://codereview.chromium.org/2742743004
Cr-Commit-Position: refs/heads/master@{#457424}

BUG=402805
TBR=enne@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2758543002
Cr-Commit-Position: refs/heads/master@{#457423}

down in //content's browser_main_loop.cc. If we let the Device Sensor be hosted
by DeviceService, we cannot find a right time to do the DeviceSensorService
shutdown operation even as late as in destructor of DeviceService.

So, in this patch we let DeviceSensorService observe the thread it is
running on. For Non-Android platform, DeviceSensorService is run in IO thread.
It shuts itself down when IO thread is going to close. For Android platform, the
DeviceSensorService shutdown is only called when running browsertests, with this
modification the shutdown is later (for browsertest) but still safe.

BUG=698664

Review-Url: https://codereview.chromium.org/2730333002
Cr-Commit-Position: refs/heads/master@{#457422}

https://skia.googlesource.com/skia.git/+log/f49b1e0ad955..0c984a0af309

$ git log f49b1e0ad..0c984a0af --date=short --no-merges --format='%ad %ae %s'
2017-03-16 robertphillips More pre-emptive proxification

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
TBR=borenet@chromium.org

Change-Id: I392ecb9cf308ea9686ff2e881554881b369f6e52
Reviewed-on: https://chromium-review.googlesource.com/455993Reviewed-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@{#457421}

The pattern for *.typemap is missing in OWNERS. It causes warnings for CLs that change autofill_types.typemap

Review-Url: https://codereview.chromium.org/2755913002
Cr-Commit-Position: refs/heads/master@{#457420}

I believe this captures the purpose of the function more cleanly.

BUG=690006

Review-Url: https://codereview.chromium.org/2757603002
Cr-Commit-Position: refs/heads/master@{#457419}

The cursor would flash when the context menu is sent if the event is
preventDefaulted. With a change https://codereview.chromium.org/2637403012
this occurred more frequently because the move event would get suppressed.

Ultimately I found when this code was activated in
https://codereview.chromium.org/7270 and it seems it was mis-guided in
being added. I've check on Windows removing this code makes everything
work fine.

BUG=699232

Review-Url: https://codereview.chromium.org/2755783002
Cr-Commit-Position: refs/heads/master@{#457418}

On Windows' taskbar's jumplist, both "New window" and "New incognito window" links have main Chrome icon.
An Incognito icon is added to the repository and used instead of Chrome icon for new incognito window link.

BUG=680904
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.win:win10_chromium_x64_rel_ng

Review-Url: https://codereview.chromium.org/2660153002
Cr-Commit-Position: refs/heads/master@{#457032}
Committed: https://chromium.googlesource.com/chromium/src/+/596cbc2cc875076c361b1b77e840919a2697b67d

patch from issue 2660153002 at patchset 120001 (http://crrev.com/2660153002#ps120001)

In the previous attempt to land the CL, I had not updated the DEPS file of src-internal with the new icon and it broke the build.
Now the icon is added to Google Chrome by this CL: https://chromereviews.googleplex.com/573787013/
And DEPS file is updated in this CL: https://chromereviews.googleplex.com/569457014/

TBR=sky@chromium.org

Review-Url: https://codereview.chromium.org/2750023002
Cr-Commit-Position: refs/heads/master@{#457417}

BUG=702171
TBR=pfeldman@chromium.org

Review-Url: https://codereview.chromium.org/2747293006
Cr-Commit-Position: refs/heads/master@{#457416}

MockFormFetcherImpl currently appears in FormFetcherImpl unittest. Its purpose
is to ensure that ProcessMigratedForms is called on a FormFetcher if an
associated HttpPasswordMigrator receives results from PasswordStore.

This is not necessary. It is sufficient to check that (A) HttpPasswordMigrator
receiving results from PasswordStore calls ProcessMigratedForms on its
consumer, and (B), that forms which the FormFetcher ends up with are the
migrated ones. (A) is tested in HttpPasswordMigratorTest::TestFullStore, (B) is
tested in FormFetcherImplTest.TryToMigrateHTTPPasswordsOnHTTPSSites through
EXPECT_CALL(consumer_, ProcessMatches(...https_form...)).

Moreover, having a test double for the tested class is a bad practice, which
could result in the test bypassing the production code when it should not.

Therefore, this CL removes MockFormFetcherImpl.

BUG=571580

Review-Url: https://codereview.chromium.org/2756543004
Cr-Commit-Position: refs/heads/master@{#457415}

https://codereview.chromium.org/2743733002 already disabled several of
them, due to rounding errors on Android. There are two additional tests
that are failing on some of the Android bots. This CL disables them.

BUG=700007
TBR=kenrb@chromium.org, creis@chromium.org

Review-Url: https://codereview.chromium.org/2750343002
Cr-Commit-Position: refs/heads/master@{#457414}

Network traffic annotation is added to network request of
chrome/browser/extensions/api/dial/device_description_fetcher.cc.

BUG=656607

Review-Url: https://codereview.chromium.org/2738243003
Cr-Commit-Position: refs/heads/master@{#457413}

This CL addes flag for enabling/disabling video fullscreen detection. So
we can reduce the overhead of MediaCustomControlsFullscreenDetector if
the client side doesn't need it.

The downstream change will be in the subclass of AppHooks.java.

BUG=699565

Review-Url: https://codereview.chromium.org/2738893002
Cr-Commit-Position: refs/heads/master@{#457412}

This restores the old logic we had for blink to deal with these as mouse
move events but keeping the mouseleave events when we leave the window.

NotifyInferior messages are already dropped at the X11 desktop tree
host but I didn't want to risk breaking any views with addressing this
release block stable issue.

BUG=701637,450631,240300,352106,386896,569998,575208

Review-Url: https://codereview.chromium.org/2751833006
Cr-Commit-Position: refs/heads/master@{#457411}

https://skia.googlesource.com/skia.git/+log/b66fa526b882..f49b1e0ad955

$ git log b66fa526b..f49b1e0ad --date=short --no-merges --format='%ad %ae %s'
2017-03-15 halcanary FuzzCanvas: only valid ChannelSelectors
2017-03-15 iannucci Update recipes.py to f605e5c0bc2e748170ca851383cb552c7225316b

Created with:
  roll-dep src/third_party/skia
BUG=699379


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
TBR=borenet@chromium.org

Change-Id: I75b38ce9077100e93ad1c685382099face7c3f75
Reviewed-on: https://chromium-review.googlesource.com/455992Reviewed-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@{#457410}

Both of LocalFrame and RemoteFrame have a WindowProxyManager.
We can move it into a base class Frame.

windowProxy->initializedIfNeeded() is moved into WindowProxyManager
from ScriptController and RemoteFrame.
WindowProxyManager::windowProxy(world) now always returns
an initialized global proxy.  It's safer than before, I think.
(But no behavioral change in this CL at all.)

Also the following renaming is done in order to emphasize that
the returned object may not be initialized.

  WindowProxyManagerBase
  => WindowProxyManager
  (Same naming as Frame, DOMWindow and WindowProxy)

  WindowProxyManager::mainWorldProxy
  => mainWorldProxyMaybeUninitialized

  WindowProxyManager::windowProxy
  => windowProxyMaybeUninitialized

By the way, we don't much use RemoteWindowProxyManager.
We can simply remove the class in a follow-up CL.

BUG=

Review-Url: https://codereview.chromium.org/2702273004
Cr-Commit-Position: refs/heads/master@{#457409}

Revert of [Cronet] Write effective experimental options to NetLog (patchset #5 id:80001 of https://codereview.chromium.org/2738813004/ )

Reason for revert:
The bot

Android Cronet Builder Asan

is reporting

Failed steps failed cronet_unittests failed cronet_test_instrumentation_apk

I suspect this patch might be why.

https://uberchromegw.corp.google.com/i/chromium.android/builders/Android%20Cronet%20Builder%20Asan/builds/1791

Original issue's description:
> [Cronet] Write effective experimental options to NetLog
>
> This CL does the following:
> (1) Have a warning in Experimental Options parsing code if there is an
>     unrecognized option.
> (2) Make cronet_url_request_context_adapter use the new FileNetLogObserver which
>     allows writing polled data.
> (3) Put the effective experimental options in the polled data.
>
> BUG=584806,699567
> CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_cronet_tester
>
> Review-Url: https://codereview.chromium.org/2738813004
> Cr-Commit-Position: refs/heads/master@{#457158}
> Committed: https://chromium.googlesource.com/chromium/src/+/3c275e0a2333511ec831b3c396dc9f38a161a8c6

TBR=kapishnikov@chromium.org,mgersh@chromium.org,xunjieli@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=584806,699567

Review-Url: https://codereview.chromium.org/2751113004
Cr-Commit-Position: refs/heads/master@{#457408}

Network traffic annotation is added to network request of
components/autofill/core/browser/autofill_download_manager.cc.

BUG=656607

Review-Url: https://codereview.chromium.org/2740213003
Cr-Commit-Position: refs/heads/master@{#457407}

BUG=631966

Review-Url: https://codereview.chromium.org/2749583003
Cr-Commit-Position: refs/heads/master@{#457406}