GitLab

BUG=490942
R=skobes@chromium.org

Review-Url: https://codereview.chromium.org/2571703002
Cr-Commit-Position: refs/heads/master@{#438021}

This reverts commit a12e7a00.

BUG=646464

TEST=Visit http://output.jsbin.com/vucefi, confirm that both boxes have
box shadows and the first box has sub-pixel AA text

Review-Url: https://codereview.chromium.org/2561183002
Cr-Commit-Position: refs/heads/master@{#438020}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/19565fdb148a..c69690acc34b

$ git log 19565fdb1..c69690acc --date=short --no-merges --format='%ad %ae %s'
2016-12-12 charliea Upload the serial log file when we fail to retrieve the BattOr git hash
2016-12-12 jessimb Fixed mapreduce.yaml to name: entity_kind
2016-12-12 dtu [firefighter] Replace builder name regexes with more fuzzy matching.
2016-12-12 jessimb A MapReduce job to add units into the anomaly entity.
2016-12-12 benjhayden Fix mismatched curly braces in sinebow_color_generator_test.
2016-12-12 eakuefner Revert of [Telemetry] Add temporary logging to desktop_browser_finder (patchset #1 id:1 of https://codereview.chromium.org/2567783002/ )
2016-12-12 eakuefner [Telemetry] Introduce LegacyHtmlOutputFormatter
2016-12-12 rnephew [BattOr] Detect malformed git hashes.
2016-12-12 eakuefner [Telemetry] Pass --disable-gpu in run_tests if starting Xvfb
2016-12-12 nednguyen Move browser_test_runner.LoadAllTestsInModule to serially_executed_browser_test_case
2016-12-12 nednguyen [Telemetry] Fix flaky browser test due to duplicated system stubbing
2016-12-12 benjhayden Make ChartJsonConverter support UMA Histograms and single-story results.

BUG=673411,636153

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.chromium.android:android_optional_gpu_tests_rel
TBR=catapult-sheriff@chromium.org

Review-Url: https://codereview.chromium.org/2574623002
Cr-Commit-Position: refs/heads/master@{#438019}

The rename is needed to avoid a naming collision after changing from
Blink to Chromium naming style.  Right now we have a |ReadyState| type
and a |readyState| accessor method (differing by case of the first
character); after a naive rename by the rewrite_to_chrome_style tool we
would end up with |ReadyState| being the name of both the type and the
accessor method (with both living in the same namespace).

Prepending a "get" prefix to the name of the accessor method is the
workaround that fits into the guidance on the recommended
post-Blink-to-Chromium-rename style suggested by esprehn@ in
https://crbug.com/582312#c17:
- Getters favor not using "Get", ex. FirstChild()
- Unless the type name conflicts, in which case you can either rename
  the type if it's easy and makes sense, or add "Get", ex. GetContext().

FWIW, this rename brings the name of the enum closer to the name of the
enum in the WHATWG spec.

BUG=582312

Review-Url: https://codereview.chromium.org/2567943004
Cr-Commit-Position: refs/heads/master@{#438018}

This CL adds the actual CertificateReportingService code that's going to be
used to send reports. The service is still not wired to any other service, so is
off by default.

BUG=554323

Review-Url: https://codereview.chromium.org/2543523002
Cr-Commit-Position: refs/heads/master@{#438017}

Today, navigator renders green checkmarks as a part of the title element.
This results in the editing field spanning over the icon as well.

This patch starts using treeelement.setTrailingIcons() method to
avoid this problem.

BUG=673086
R=dgozman

Review-Url: https://codereview.chromium.org/2568623002
Cr-Commit-Position: refs/heads/master@{#438016}

This fixes a bug where the 2nd arg to std::string::substr() was not the
length, but the end position.

Review-Url: https://codereview.chromium.org/2567263002
Cr-Commit-Position: refs/heads/master@{#438015}

BUG=660445

Review-Url: https://codereview.chromium.org/2555623002
Cr-Commit-Position: refs/heads/master@{#438014}

This is to gather early feedback in canary, before we flip this to stable (for M57).

Link to Intent to ship.
https://groups.google.com/a/chromium.org/forum/#!topic/blink-
dev/AT7aqhPcNdw

BUG=672982

Review-Url: https://codereview.chromium.org/2560313002
Cr-Commit-Position: refs/heads/master@{#438013}

In some cases, an input event being targeted to the root RWHV was going
through coordinate transformation, while in other cases it was not.
This is normally fine because it is the identity transform anyway, but
it can result in rounding error when there is a non-integer device
scale factor. This was noted to cause a bug related to selections not
being cleared because of coordinates from a transformed (rounded)
event being compared against coordinates from a non-tranformed
(unrounded) input event.

This CL prevents transformation in all cases when the root is being
targeted and the event is already in the coordinate space of the root.

BUG=670253
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2567093003
Cr-Commit-Position: refs/heads/master@{#438012}

Record Data Reduction Proxy (DRP) histogram on how long back savings were
automatically cleared because the system clock moved back. If this happens
too frequently, then we will need to add a mechanism to get around this.
This would also help in debugging 0% savings feedback reports.

BUG=658354

Review-Url: https://codereview.chromium.org/2546273002
Cr-Commit-Position: refs/heads/master@{#438011}

When RLS is enabled, the LayoutView gets the size of the FrameView, and no clipping to the FrameView is done.

BUG=490942
R=skobes@chromium.org,chrishtr@chromium.org
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2566223003
Cr-Commit-Position: refs/heads/master@{#438010}

This change updates the mapping from RSSI values into signal strength
icon levels. It is made after reviewing a month of recorded UMA
data for RSSI values encountered by users.

A new design of how we map values to displayed levels is documented
in the change.

See these charts for measured data:
https://goo.gl/photos/pCoAkF7mPyza9B1k7

BUG=629689

Review-Url: https://codereview.chromium.org/2561883002
Cr-Commit-Position: refs/heads/master@{#438009}

Made list-style-position property independent. Also updated
independent-inheritance-fast-path.html to include listStylePosition.

Tested using independent-inheritance-fast-path.html.

BUG=628043

Review-Url: https://codereview.chromium.org/2566923002
Cr-Commit-Position: refs/heads/master@{#438008}

Starting with Mac OS 10.12, path building stops at the first weak digest (md2 or md5).

BUG=629712

Review-Url: https://codereview.chromium.org/2558983002
Cr-Commit-Position: refs/heads/master@{#438007}

A static class variable was new-ed with no opportunity for a delete call.
Instead, convert this to using CR_DEFINE_STATIC_LOCAL so it's properly
deleted on exit.

Removes ExperimentalScoringEnabled() field trial function and
changed the default value for the topicality threshold when
the field trial setting is erroneous or partially-specified.
I checked that removing this logic change does not affect any currently- or
recently-running experiments.

Other than changing the semantics of the field trial configs,
there should be no functional changes.

In the process, remove most references to HQP in variable names and
comments.
These references are unnecessary.
All scoring in these files are for the HQP.

BUG=

Review-Url: https://codereview.chromium.org/2548363010
Cr-Commit-Position: refs/heads/master@{#438006}

For some bots, the git_revision of cipd_package is pinned to a
very old version (summer 2016). This causes issues when trying to
communicate with other systems expecting a newer version, and leads
to issues like:
Failed to start logdog: Missing project [LOGDOG_STREAM_PROJECT]

This CL updates the pinned revision number to a newer version,
suggested in the linked bug.

A better fix would probably be to pin this in a shared location,
making it simple to update all these at the same time, instead
of specifiying it once per target.

This CL was generated by running the following script:
git grep -l -e \
  'git_revision:3ff24775a900b675866fbcacf2a8f98a18b2a16a' \
  -- "*.json" | \
  xargs sed -ie \
    's/git_revision:3ff24775a900b675866fbcacf2a8f98a18b2a16a/git_revision:25755a2c316937ee44a6432163dc5e2f9c85cf58/'

BUG=670817

Review-Url: https://codereview.chromium.org/2574563002
Cr-Commit-Position: refs/heads/master@{#438005}

This is required for proprietory codec when streaming URL type
content.

Review-Url: https://codereview.chromium.org/2568003003
Cr-Commit-Position: refs/heads/master@{#438004}

In a future CL, the Physical Web's auto-optin behavior will be modified to
require that Google is configured as the default search provider. This CL
adds a new overload for StartPhysicalWebDiscovery so downstream users can
be modified to use the new signature that provides access to the browser
state, from which we can determine the default search provider.

BUG=630769

Review-Url: https://codereview.chromium.org/2566753002
Cr-Commit-Position: refs/heads/master@{#438003}

components/ui_devtools:ui_devtools is a source_set target depended on
by chrome/browser/ui:ui and ash:ash, which just embed ui_devtools and do not
expect it to export any symbols out.

BUG=669352

Review-Url: https://codereview.chromium.org/2568003002
Cr-Commit-Position: refs/heads/master@{#438002}

This CL includes:
- Introduce GooglePlayWebApkInstallDelegate which can bind to Phonesky's
  unpublished PlayInstallService to install WebAPKs.
- WebApkInstaller uses the GooglePlayWebApkInstallDelegate to ask Play to
  install WebAPKs. The changes are behind the finch flag.

The internal CL is: https://chrome-internal-review.googlesource.com/#/c/305912/

BUG=662149

Review-Url: https://codereview.chromium.org/2515293004
Cr-Commit-Position: refs/heads/master@{#438001}

The crash happens because by the time we call Demuxer::GetStream again
the main thread might have proceeded with destroying SourceBuffer and
thus ChunkDemuxer no longer allows accessing demuxer streams. Luckily
demuxer streams themselves are still alive, just moved to
ChunkDemuxer::removed_streams_ (see ChunkDemuxer::RemoveId). So cached
pointers can still be used safely.

BUG=668604

Review-Url: https://codereview.chromium.org/2558213002
Cr-Commit-Position: refs/heads/master@{#438000}

This changes the command line switch "n" to "process-service-name" and
makes mash-runner look for it.

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

Review-Url: https://codereview.chromium.org/2571603002
Cr-Commit-Position: refs/heads/master@{#437999}

Document has two methods for getting whether the context is secure.
These duplicate functionality, but one also allocates an error string
for insecure contexts.

Many callers do not need the error string, so this patch updates them
to use the other isSecureContext method. The patch also consolidates
duplicate logic.

The error string allocation was seen in profiles in
RenderFrameImpl::willSendRequest

BUG=348655

Review-Url: https://codereview.chromium.org/2567173002
Cr-Commit-Position: refs/heads/master@{#437998}

Revert of android: Increase NUM_PRIVILEGED_SERVICES to 20 (patchset #3 id:40001 of https://codereview.chromium.org/2555303004/ )

Reason for revert:
Android keeps the intent for a long time, so
randomize doesn't help at all. More details:
crbug.com/664341#c91

Original issue's description:
> android: Increase NUM_PRIVILEGED_SERVICES to 20
>
> This works in conjunction with
> https://codereview.chromium.org/2557273004/ to workaround
> crbug.com/664341. That CL ramdomizes the order of the named services to
> avoid collision, which is not actually effective unless there is a large
> number of named services to choose from.
>
> Sandboxed service already has 20 named services, so up the privileged
> (gpu) count to 20 as well.
>
> BUG=664341
>
> Committed: https://crrev.com/9b853cfcf82d8c57487cb0dd8d1eeaa5ca40260f
> Cr-Commit-Position: refs/heads/master@{#437634}

TBR=tedchoc@chromium.org,halliwell@chromium.org,sanfin@chromium.org,tsunghung@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=664341

Review-Url: https://codereview.chromium.org/2574593002
Cr-Commit-Position: refs/heads/master@{#437997}

This is the version all the runtime libraries seem to
use. So, I'm matching it on the assumption the
compiler may do something smarter if appropriate.

TBR=scottmg
NOTRY=true

Review-Url: https://codereview.chromium.org/2574443002
Cr-Commit-Position: refs/heads/master@{#437996}

The rename is needed to avoid a naming collision after changing from
Blink to Chromium naming style.  Right now we have a |DocumentState|
type and a |documentState| accessor method (differing by case of the
first character); after a naive rename by the rewrite_to_chrome_style
tool we would end up with |DocumentState| being the name of both the
type and the accessor method (with both living in the same namespace).

Prepending a "get" prefix to the name of the accessor method is the
workaround that fits into the guidance on the recommended
post-Blink-to-Chromium-rename style suggested by esprehn@ in
https://crbug.com/582312#c17:
- Getters favor not using "Get", ex. FirstChild()
- Unless the type name conflicts, in which case you can either rename
  the type if it's easy and makes sense, or add "Get", ex. GetContext().

BUG=582312
TBR=creis@chromium.org

Review-Url: https://codereview.chromium.org/2569883002
Cr-Commit-Position: refs/heads/master@{#437995}

This CL adds a test helper header platform/testing/PaintPropertyTestHelpers.h
for common operations that are used in SPv2-related tests.

createOpacityOnlyEffect is added to create effect nodes with only opacity,
other optional effects are assigned a no-op value, and local transform space
and clip are inherited from parent.

As a side change, rootPaintChunkProperties is also moved to the helper header
and renamed to defaultPaintChunkProperties.
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2565903002
Cr-Commit-Position: refs/heads/master@{#437994}

This patch adds a cache of the SVG marker transform, similar to the pattern
used in LayoutSVGViewportContainer. This should be a small perf win because
we no longer re-calculate the transform when painting each marker instance.
With this new cache logic, invalidation logic has been added which includes
a fix for paint property under-invalidation (marker-viewBox-changes.svg).

BUG=645667,600618

Review-Url: https://codereview.chromium.org/2565943002
Cr-Commit-Position: refs/heads/master@{#437993}

WebTaskRunner::Task is replaced by WTF::Closure/CrossThreadClosure,
and no longer used in the production.

Review-Url: https://codereview.chromium.org/2568063003
Cr-Commit-Position: refs/heads/master@{#437992}

This file seems to be autogenerated and currently produces ~1k eslint errors.

BUG=none
R=chenwilliam

Review-Url: https://codereview.chromium.org/2566553002
Cr-Commit-Position: refs/heads/master@{#437991}

This is preperatory work which will allow for cheaper update of the
req'd for activation/draw status in the client side scheduler work.

BUG=673434
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2564433003
Cr-Commit-Position: refs/heads/master@{#437990}

Also:
* Quick refactor of jinja2 context creation into a builder.
* Switched Attribute/Method to dicts.
* Added a bit more meat more to header template.
* Added nicer parameter handling to run-bindings-tests
* Added stub helper for unit tests.

BUG=654129
R=bashi,yukishiino,haraken

Review-Url: https://codereview.chromium.org/2568693002
Cr-Commit-Position: refs/heads/master@{#437989}

https://chromium.googlesource.com/chromium/third_party/ffmpeg.git/+log/26be2ced9076..f309edd7828e

$ git log 26be2ced9..f309edd78 --date=short --no-merges --format='%ad %ae %s'
2016-12-09 wolenetz A couple cherry picks from upstream to fix issue 670190

TBR=dalecurtis@chromium.org
BUG=670190

Review-Url: https://codereview.chromium.org/2567533003
Cr-Commit-Position: refs/heads/master@{#437988}

Patch removes preventDefault(), stopPropagation(), consume() and .type property.
This is a preparation for typed events, which will not inherit from Common.Event.

BUG=none

Review-Url: https://codereview.chromium.org/2553043003
Cr-Commit-Position: refs/heads/master@{#437987}

It doesn't seem like it's being respected. See build:
https://build.chromium.org/p/chromium.fyi/builders/WebKit%20Win%20-%20RandomOrder/builds/382

It's getting close to Wednesday - I'd like to remove [ Win ] so we can get onto more important work and then circle back around when we have more time to find the correct prefix.

R=qyearsley@chromium.org
BUG=670843

Review-Url: https://codereview.chromium.org/2560423003
Cr-Commit-Position: refs/heads/master@{#437986}

BUG=649495

Review-Url: https://codereview.chromium.org/2507753003
Cr-Commit-Position: refs/heads/master@{#437985}

testCanPlayePlatformMediaCodecs was disabled due to flakiness. It seems
that the test is no longer flaky and can be re-enabled (since we are no
longer using WMPA, or because of some unknown changes in test
reliability).

NOTE: If ever the screen is off when the test runs, the test fails.

BUG=620890
TEST=100x run on Nexus 5 and Nexus 7

Review-Url: https://codereview.chromium.org/2570483005
Cr-Commit-Position: refs/heads/master@{#437984}

It used to be used for data/ dir removal, but it is now
integrated into ArcSessionManager.
Remove the unused class now.

BUG=657687
BUG=b/31079732
TEST=Ran bots.

Review-Url: https://codereview.chromium.org/2573463002
Cr-Commit-Position: refs/heads/master@{#437983}

BUG=673022

Review-Url: https://codereview.chromium.org/2565673003
Cr-Commit-Position: refs/heads/master@{#437982}