GitLab

I move creation to RootWindowControllerCommon, but I think
RootWindowControllerCommon should go away. I'm going to do that next,
which will result in some additional cleanup.

BUG=624173
TEST=covered by tests
R=msw@chromium.org

Review-Url: https://codereview.chromium.org/2292183003
Cr-Commit-Position: refs/heads/master@{#415374}

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/14882d030b9fc5b4dbd97030681b80f46bb96f4e Add configs to set up WebView 32bit perf bots. (mikecase@chromium.org)

TBR=martiniss@chromium.org,phajdan.jr@chromium.org
BUG=

Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Review-Url: https://codereview.chromium.org/2292223002
Cr-Commit-Position: refs/heads/master@{#415373}

Currently when screen is off (or Chrome goes to background) for more than 1 min,
peerconnection will be stoppted by PowerMonitor to save battery power. This is
not necesary now since we'll suspend video call when the tab is background. And
users prefer that audio call can continue while they are doing something else.

BUG=513633

Review-Url: https://codereview.chromium.org/2263373003
Cr-Commit-Position: refs/heads/master@{#415372}

ViewMsg_ThemeChanged was never sent, and even if it was, the only code
path that did anything was

#if USE_AURA
// nothing
#elif OS_WIN
// do stuff
...

so it was never going to run. Noticed while trying to clean up
NativeThemeWin usage from Blink, which also makes little-to-no sense and
causes the linked bug.

R=jam@chromium.org
BUG=258201

Review-Url: https://codereview.chromium.org/2287813002
Cr-Commit-Position: refs/heads/master@{#415371}

Also, rename use of internal codename in the file.

BUG=none

Review-Url: https://codereview.chromium.org/2292203002
Cr-Commit-Position: refs/heads/master@{#415370}

When the sizes attribute changes, that has the potential of modifying the
image's intrinsic dimensions. However, the LayoutImage object was not
notified of that explicitly. While the intrinsic dimensions were changed
when a new resource was downloaded as a result of the sizes attribute
change, that wasn't the case when such the resource was not changed.

This CL fixes that, by explicitly notifying the LayoutImage of the change.

BUG=639845

Review-Url: https://codereview.chromium.org/2292013002
Cr-Commit-Position: refs/heads/master@{#415369}

The Blimp browser tests now cover every action of NavigationFeature.

GoForward is an additional step in the existing GoBack test, rather than a new test, because you can't test GoForward without GoBack.

Also adds no-op behavior on the engine for when the client requests GoBack or GoForward but those operations aren't valid, and tests for those cases.

Tested log output for the invalid GoBack and GoForward cases with the following:
blimp_browsertests --single_process --gtest_filter=EngineBrowserTest.InvalidGoBack | grep Ignoring
blimp_browsertests --single_process --gtest_filter=EngineBrowserTest.InvalidGoForward | grep Ignoring

BUG=

Review-Url: https://codereview.chromium.org/2295543002
Cr-Commit-Position: refs/heads/master@{#415368}

Use methods which returns id instead of deprecated methods which return
NSString. JS execution API returned NSString because that was the only
supported option from UIWebView. Now WKWebView returns id which should
be used everywhere.

BUG=None

Review-Url: https://codereview.chromium.org/2286773002
Cr-Commit-Position: refs/heads/master@{#415367}

The goal is to let Blimp reuse this logic on browser process, since
Blimp client doesn't have renderer process.

BUG=577985

Review-Url: https://codereview.chromium.org/2285083004
Cr-Commit-Position: refs/heads/master@{#415366}

On UI, the text for installing a WebAPK in the App Banner is:
1) "Add to Homescreen" -> "Adding" -> "Open" for a sucessful
   installation
2) "Add to Homescreen" -> "Adding" -> "Unable to save" toast
   for an unsucessful installation.

WebAPK-specific logic is added to AppBannerInfobarDelegateAndroid
to show the above text. Besides, checking whether a site is WebAPK
Compattive is based on 1) installable check + 2) url is over
 "https" + 3) WebAPK is enabled on command line.

Currently, WebAPK is installed as an untrusted source with Android
installation dialog pops up. User could choose cancel the installation
even after the WebAPK has been downloaded locally. To detect the
cancel behavior, AppBannerInfobarDelegateAndroid checks whether the
WebAPK has been installed or not after Chrome is resumed. This isn't an
ideal way, but it seems we can't find a better way to detect the cancel
event. This logic will be removed when WebAPK is installed by Play soon.

This CL disables creating WebAPKs by selecting add-to-homescreen from
the app menu. Instead, in a follow up CL, selecting from the app menu
will make the app banner info bar shown to install WebAPKs.

BUG=627250

Review-Url: https://codereview.chromium.org/2259553002
Cr-Commit-Position: refs/heads/master@{#415365}

BUG=641597

Review-Url: https://codereview.chromium.org/2294843002
Cr-Commit-Position: refs/heads/master@{#415364}

This patch adds a new (currently failing) filter repaint test to show
how we improperly repaint turbulence filters. In addition, the existing
test for 640264 has been unskipped and rebaselined with a failing
result. These two changes should help catch progressions in the future.

BUG=598051,640264

Review-Url: https://codereview.chromium.org/2294633002
Cr-Commit-Position: refs/heads/master@{#415363}

BUG=641559

Review-Url: https://codereview.chromium.org/2283163002
Cr-Commit-Position: refs/heads/master@{#415362}

This reverts commit 64fea5e2.

The CL broke the Windows build for some users.

NOPRESUBMIT=true
NOTRY=true
TBR=dpranke@chromium.org
BUG=634184

Review-Url: https://codereview.chromium.org/2294903002
Cr-Commit-Position: refs/heads/master@{#415361}

This directory has not been touched since 2008, appears to be from the very
early history of tracing, and contains only some very primitive trace examples.

The frontend tracing code lives in Catapult, so let's remove this dir.

Review-Url: https://codereview.chromium.org/2290393002
Cr-Commit-Position: refs/heads/master@{#415360}

BUG=636093
TBR=jam@chromium.org

Review-Url: https://codereview.chromium.org/2271653006
Cr-Commit-Position: refs/heads/master@{#415359}

A partial list of things done here:

  * Move tests that verified NTPSnippet::best_source() behavior into
    ntp_snippet_test.cc; they don't belong here.
  * Switch it to use the chromecontentsuggestions-pa. Testing new
    features like server-provided categories is not possible without
    this.
  * Replace MockContentSuggestionsProviderObserver with a fake. It
    shouldn't have been a mock, since we weren't using it to test
    behavior, only state.
  * In one case, test using what the observer sees, instead of peeking
    in through debugging hooks.
  * Get rid of special test base class and Mock::VerifyAndClear calls; I
    consider these anti-patterns.
  * Initialize the service better; give it an empty list of snippets for
    its first fetch, and wait for it to complete the fetch before
    returning (I think this is what was most strongly implicated by the
    failures with the other CL).
  * Create the service and expect the scheduling directly from each
    case, so they are more readable individually.

Review-Url: https://codereview.chromium.org/2285133004
Cr-Commit-Position: refs/heads/master@{#415358}

BUG=642078
COMPONENT=HaTS, Chrome OS

Review-Url: https://codereview.chromium.org/2294473003
Cr-Commit-Position: refs/heads/master@{#415357}

Inline sources into GN files.

Review-Url: https://codereview.chromium.org/2291223002
Cr-Commit-Position: refs/heads/master@{#415356}

Update toolkit_delegate_views to launch the menu as async

TEST=manually testing the various ways of calling the context menu. Normal views, guest_views
BUG=557136
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2255923002
Cr-Commit-Position: refs/heads/master@{#415355}

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/bb5de0678735599d572dad88f9a15e54e509f874 Add support for running WebView tests with new cts test runner. (mikecase@chromium.org)

TBR=martiniss@chromium.org,phajdan.jr@chromium.org
BUG=

Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Review-Url: https://codereview.chromium.org/2297773002
Cr-Commit-Position: refs/heads/master@{#415354}

I've noticed some problematic edge cases with the reporting of find-in-page results from FindRequestManager. While trying to write some new tests in find_request_manager_browsertest.cc, I noticed that the new tests sometimes fail because of these edge cases. The existing tests in there could definitely fail from this as well, though much less likely. I have seen them fail very rarely before though and this is probably why.

The details of the issue, edge cases, and my solution are explained in this design doc: https://docs.google.com/a/google.com/document/d/1BCvuSW9XSBH7GMx5VC0TgBh6XmJJDOzz_XZKItjF6oM/edit?usp=sharing

UPDATE: Anther find-in-page CL I landed recently (https://codereview.chromium.org/2186113002/) was reverted because of a test that began timing out. I believe that failure is caused by the problems fixed here, so that bugfix is now blocked on this CL.
Review-Url: https://codereview.chromium.org/2249133002
Cr-Commit-Position: refs/heads/master@{#415353}

In some cases the certificates_ array would be empty thus selecting the first
row (which is non-existing) would crash Chrome.

BUG=630319

Review-Url: https://codereview.chromium.org/2277133003
Cr-Commit-Position: refs/heads/master@{#415352}

This includes:
- Moving the mapping between autofill credit card types and types
  defined in the spec to components/autofill.
- Adding the basic credit card response type from the spec.

BUG=602666

Review-Url: https://codereview.chromium.org/2285523002
Cr-Commit-Position: refs/heads/master@{#415351}

BUG=635081
TEST=mac bots
NOTRY=true
R=kbr@chromium.org,erikchen@chromium.org

Review-Url: https://codereview.chromium.org/2251263002
Cr-Commit-Position: refs/heads/master@{#415350}

https://chromium.googlesource.com/skia.git/+log/fc6cb7366493..9a44565e9946

$ git log fc6cb7366..9a44565e9 --date=short --no-merges --format='%ad %ae %s'
2016-08-30 bsalomon Remove custom data from GrGpuResource

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
TBR=bsalomon@google.com

Review-Url: https://codereview.chromium.org/2290383002
Cr-Commit-Position: refs/heads/master@{#415349}

This is occurring because we try to get the RootScrollerController from the top
level Document by calling topDocument(). Since we're in a remote Frame though,
topDocument returns the current RootScrollerController's Document and so we
infinitely recurse into isViewportScrollCallback.

This is a quick fix to the problem while I come up with a better long term
solution for OOPIFs.

BUG=641775

Review-Url: https://codereview.chromium.org/2290353002
Cr-Commit-Position: refs/heads/master@{#415348}

This is done by making SPDY's stream info map own the StreamInfos, and by making sync's MetahandlesMap own the EntryKernels.

BUG=555865

Review-Url: https://codereview.chromium.org/2287783002
Cr-Commit-Position: refs/heads/master@{#415347}

m_useCurrentView being true implies that m_viewSpec is non-null. Perhaps
defensively, most code that checks it is written in a way such that it
expects that m_viewSpec can be null if m_useCurrentView is true - which
is not the case, and adds unnecessary code.
Instead change the code to adhere to the rule above, and assert that in
the various places instead.
Also add some checks of the invariant in methods that set the flag and
m_viewSpec. Refactor SVGSVGElement::setupInitialView to avoid the need
to sprinkle invariant checks "all over" it.

BUG=110195

Review-Url: https://codereview.chromium.org/2290293003
Cr-Commit-Position: refs/heads/master@{#415346}

I added this flag with the new plumbing for subresources with cert
errors in https://crbug.com/634171, but neglected to clear it on
navigation.

BUG=641180

Review-Url: https://codereview.chromium.org/2280753003
Cr-Commit-Position: refs/heads/master@{#415345}

HKDF was added to the spec trunk:
https://github.com/w3c/webcrypto/commit/36bc6686c4cc630a91df3d9e9f38bfc9b8f0ce47

BUG=642045

Review-Url: https://codereview.chromium.org/2295613002
Cr-Commit-Position: refs/heads/master@{#415344}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/89be526f6da4..bf2b35a6b699

$ git log 89be526f6..bf2b35a6b --date=short --no-merges --format='%ad %ae %s'
2016-08-30 benjhayden Migrate clockSyncLatencyMetric from ScalarNumeric to Histogram

TBR=catapult-sheriff@chromium.org

Review-Url: https://codereview.chromium.org/2296683003
Cr-Commit-Position: refs/heads/master@{#415343}

The icon and icon murmur2 hash are sent to the WebApk Minting Server
when creating/updating a WebAPK. For updating, these two values are
fetched and computed by ManifestUpdateDetector, so WebApkInstaller
doesn't need to download the icon and compute icon hash again.

BUG=639000

Review-Url: https://codereview.chromium.org/2263673003
Cr-Commit-Position: refs/heads/master@{#415342}

I think this is safer to merge than https://codereview.chromium.org/2291923002/. I'll remove the check once the bug 64223 is fixed.

BUG=642222

Review-Url: https://codereview.chromium.org/2291933002
Cr-Commit-Position: refs/heads/master@{#415341}

Pulling in a new v8 fuzzer.
Adding a download step for the wasm and wasm->asm.js module fuzzers.

BUG=None
TEST=None
R=kcc@chromium.org,ahaas@chromium.org,mmoroz@chromium.org

Review-Url: https://codereview.chromium.org/2284393003
Cr-Commit-Position: refs/heads/master@{#415340}

Breakage introduced by prior patch here:
https://codereview.chromium.org/2208463002

We had split off GetPluginContentSetting into a utils file, but
forgot to add the widevine header file include.

This means that WIDEVINE_CDM_AVAILABLE is never defined, and the
Widevine-treated-as-JavaScript for content settings clause is
never activated, breaking Widevine on Corp machines.

BUG=641706

Review-Url: https://codereview.chromium.org/2295563002
Cr-Commit-Position: refs/heads/master@{#415339}

Discussed with bettes@ to decide that the profile badge on RTL layouts
should be on the LHS instead.
Also fixed the position of the circular mask for profile icon on RTL
layouts (see first bug) - the circular mask needs to be bumped to the
right on RTL layouts so that it's right-aligned.

See comparative screenshots:
https://drive.google.com/drive/folders/0B7Fvv7JszRyGN1FKakw4RXE2Vkk?usp=sharing

BUG=640907
BUG=615893

Review-Url: https://codereview.chromium.org/2285413002
Cr-Commit-Position: refs/heads/master@{#415338}

When merging two ruby bases ensure that the new base has a complete float list.

BUG=619380,624028

Review-Url: https://codereview.chromium.org/2283413003
Cr-Commit-Position: refs/heads/master@{#415337}

The current name of these tests did not reflect the component actually
tested. It was also inconsistent with how the tests for the the other
implementations of KeyStorage were named.

Review-Url: https://codereview.chromium.org/2292933002
Cr-Commit-Position: refs/heads/master@{#415336}

Move and rename ash/desktop_background/*
Move and rename ash/wm/system_background_controller*
Rename identifiers; update comments.

Remove unused src_container for Reparent* functions.

BUG=629605
TEST=No functional changes; just renaming.
R=jamescook@chromium.org
TBR=sky@chromium.org,tsepez@chromium.org,jochen@chromium.org

Review-Url: https://codereview.chromium.org/2290473004
Cr-Commit-Position: refs/heads/master@{#415335}