GitLab

The document mode migration code was changed in
4804efe6
to skip document mode migration if a shared preference was not set.
On older versions of Chrome, that shared preference wasn't set
even though the user was in document mode. Migration was skipped and
all tabs were lost.

Revert to the old logic of using another shared preference to help
determine if the user is in document mode.

BUG=646146

Review-Url: https://codereview.chromium.org/2340743004
Cr-Commit-Position: refs/heads/master@{#418722}

This depends on https://chromium-review.googlesource.com/#/c/384594/

BUG=b:31254747
TEST=manually confirmed that the method is actually called

Review-Url: https://codereview.chromium.org/2332153002
Cr-Commit-Position: refs/heads/master@{#418721}

This CL fixes the bug in which the media router dialog would sometimes fail to show on OSX.

We make MediaRouterDialogDelegate::GetDialogSize() set the dialog size the first time it's called by WebDialogConstrainedWindowSheet::updateSheetPosition(), so that the custom window is not initialized with size 0.

BUG=633403

Review-Url: https://codereview.chromium.org/2275413002
Cr-Commit-Position: refs/heads/master@{#418720}

R=asvitkine@chromium.org, rch@chromium.org
BUG=

Review-Url: https://codereview.chromium.org/2342663003
Cr-Commit-Position: refs/heads/master@{#418719}

This CL also adds more logging in update_w3c_test_expectations.py, and changes deps_updater.py to run this script with the --verbose flag.

Purpose of this CL:
 1. Make it easier to diagnose issues with w3c-test-autoroller in the future.
 2. Start to make update-w3c-test-expectations better to use directly.

BUG=643761

Review-Url: https://codereview.chromium.org/2337183002
Cr-Commit-Position: refs/heads/master@{#418718}

Introduces a new content::AssociatedInterfaceRegistry and
content::AssociatedInterfaceProvider to serve as Channel-associated
interface counterparts to their shell namesakes.

RenderFrameHost and RenderFrame each have both of these things on them,
and they're routed to the other side using a new RouteProvider interface
buried in RenderProcessHostImpl / ChildThreadImpl on either end.

Also adds a WebContentsInterfaceRegistry to every WebContents,
and a helper WebContentsFrameBindingSet<T> to allow consumers to
very easily manage the lifetime of associated interface bindings
for each frame in a WebContents.

See the follow-up patch for example usage.

Part a series of CLs to enable and demonstrate WebContents associated
interfaces:

  1. https://codereview.chromium.org/2309513002
  2. https://codereview.chromium.org/2316963005
  3. This CL
  4. https://codereview.chromium.org/2310583002

BUG=612500
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2310563002
Cr-Commit-Position: refs/heads/master@{#418717}

Add initial unit tests for chromeos::PowerManagerClient's
suspend-delay-handling code. More coming soon.

BUG=646912

Review-Url: https://codereview.chromium.org/2345593002
Cr-Commit-Position: refs/heads/master@{#418716}

This is to get feedback from canary users.  It will be reverted on 2016-09-16.

BUG=624074

Review-Url: https://codereview.chromium.org/2335363003
Cr-Commit-Position: refs/heads/master@{#418715}

Revert of BackgroundGradientView: -drawBackground: → -drawRect: (patchset #2 id:40001 of https://codereview.chromium.org/2336453002/ )

Reason for revert:
AnimatableView inherits from BackgroundGradientView, so a number of
other views which *don't* want backgrounds end up inheriting from
BackgroundGradientView. This widened the scope of the change beyond what
I intended.

Original issue's description:
> BackgroundGradientView: -drawBackground: → -drawRect:
>
> BackgroundGradientView had a -drawBackground: method that its clients
> were expected to call. This change moves background drawing to
> -drawRect: so that the it happens by default.
>
> Subclasses that did custom drawing already override -drawRect:, but this
> removes a couple of overrides that just forwarded to -drawBackground:.
>
> BUG=589943
>
> Committed: https://crrev.com/6b1939a26d2839136aeeba2aa0e359fce0c20281
> Cr-Commit-Position: refs/heads/master@{#418358}

TBR=thakis@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=647018

Review-Url: https://codereview.chromium.org/2339143003
Cr-Commit-Position: refs/heads/master@{#418714}

Add a bool argument in FilterData() to signal that upstream's eof has been
reached.

R=rdsmith@chromium.org

BUG=474859

Review-Url: https://codereview.chromium.org/2331193005
Cr-Commit-Position: refs/heads/master@{#418713}

BUG=631262

Review-Url: https://codereview.chromium.org/2346623002
Cr-Commit-Position: refs/heads/master@{#418712}

BUG=626321

Review-Url: https://codereview.chromium.org/2301323003
Cr-Commit-Position: refs/heads/master@{#418711}

BUG=

Review-Url: https://codereview.chromium.org/2332393005
Cr-Commit-Position: refs/heads/master@{#418710}

BUG=642380

Review-Url: https://codereview.chromium.org/2337063003
Cr-Commit-Position: refs/heads/master@{#418709}

This is a step towards removing SidebarTreeElement, and moving
all tree outlines into shadow dom.

BUG=none

Review-Url: https://codereview.chromium.org/2335983002
Cr-Commit-Position: refs/heads/master@{#418708}

In 0.5 client, it's done in BlimpClientSession, which broadcasts
connection events, assignment result to the Java layer.

1. This CL port these functionalities to native code, in
core/session/connection_status. ConnectionStatus will cache engine info
and broadcast network events to other components in Blimp.

2. Also use delegate pattern to decouple the settings JNI bridge and
BlimpClientContextImpl.

3. Some utilities functions prepared for error messages handling in UI.

BUG=624451, 646655

Review-Url: https://codereview.chromium.org/2322843002
Cr-Commit-Position: refs/heads/master@{#418707}

and have it in arc::ArcIntentHelperBridge.

BUG=b:29079617, b:31406784
TEST=try, ARC still loads activity icons

Review-Url: https://codereview.chromium.org/2337193005
Cr-Commit-Position: refs/heads/master@{#418706}

Implement WebM subsample support according to the specification at http://wiki.webmproject.org/encryption/webm-subsample-encryption.

BUG=630344

Review-Url: https://codereview.chromium.org/2174533002
Cr-Commit-Position: refs/heads/master@{#418705}

The CSS Transitions specification recently added a section specifying how timing
functions should be serialized:
https://hg.csswg.org/drafts/rev/12d32649a91f
https://drafts.csswg.org/css-transitions/#serializing-a-timing-function

This patch changes how step timing functions are serialized in order to bring it
in line with the spec. The corresponding parsing function devtools/front_end is
also updated.

The removal of "step-middle" and "steps(x, middle)" as valid values is deferred
until frames() is implemented (see crbug.com/646265).

BUG=631875,600248

Review-Url: https://codereview.chromium.org/2330283004
Cr-Commit-Position: refs/heads/master@{#418704}

This will otherwise cause assertion failures on many pages.

We could also compute this in ng code during margin collapsing and just set it
on the old object, but I'm not sure that would be worth it.

This and https://codereview.chromium.org/2337763005/ fixes most (but not all)
of the assertion failures during legacy/ng interop.

R=ikilpatrick@chromium.org,eae@chromium.org,glebl@chromium.org
BUG=635619

Review-Url: https://codereview.chromium.org/2340113003
Cr-Commit-Position: refs/heads/master@{#418703}

Revert of Enforce capability spec renderer <--> browser (patchset #15 id:280001 of https://codereview.chromium.org/2259903002/ )

Reason for revert:
This is still broken, as it happens.

Original issue's description:
> Enforce capability spec renderer <--> browser.
>
> This implements:
>
> - allows the content embedder to merge a manifest into the base ones provided by content, allowing it to add to the list of interfaces exposed to the renderer.
> - adds such an overlay to chrome/browser
>
> So.. from now on if you add an interface to either the browser/renderer, you'll need to list it in the manifest. I wonder how we should broadcast this. Also as a separate step we should require these manifests get security review.
>
> R=rockot@chromium.org,tsepez@chromium.org
>
> Committed: https://crrev.com/4ea17059bf78553528f3ffb4a9de84a447622fd3
> Committed: https://crrev.com/807a926c9ca02ac8ae1b52e82537834ef78fb9ba
> Committed: https://crrev.com/91c25a3cedc4fadcb8c84e91cc59a8f90411058a
> Committed: https://crrev.com/cdf40b4b434b43af6accf0c822ea964bc6bfd931
> Cr-Original-Original-Original-Commit-Position: refs/heads/master@{#418282}
> Cr-Original-Original-Commit-Position: refs/heads/master@{#418398}
> Cr-Original-Commit-Position: refs/heads/master@{#418582}
> Cr-Commit-Position: refs/heads/master@{#418691}

TBR=rockot@chromium.org,tsepez@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review-Url: https://codereview.chromium.org/2336043008
Cr-Commit-Position: refs/heads/master@{#418702}

We are updating this value based on the status of the lst animation of
the element which is wrong.
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2346583002
Cr-Commit-Position: refs/heads/master@{#418701}

BUG=646099

Review-Url: https://codereview.chromium.org/2334693003
Cr-Commit-Position: refs/heads/master@{#418700}

This file is used by V8's gyp. It will be moved out from platform/ soon.

R=brettw@chromium.org,dgozman@chromium.org

Review-Url: https://codereview.chromium.org/2333863002
Cr-Commit-Position: refs/heads/master@{#418699}

Original change: https://codereview.chromium.org/2323103002/

Fix: On Android, the delegate callback called in DestroySubtree was
accessing AXTree::root() after we had deleted it but before we set it
to null. We need to clear it first and then destroy it.

BUG=646777
TBR=aboxhall,mmoroz

Review-Url: https://codereview.chromium.org/2340893002
Cr-Commit-Position: refs/heads/master@{#418698}

BUG=646762

Review-Url: https://codereview.chromium.org/2339783003
Cr-Commit-Position: refs/heads/master@{#418697}

Introduce ExpectApproxEq, a helper function that allows making assertions that certain values are equal within the margin of specified relative error.

This is to avoid manually hacking around test failures like:
Expected equality of these values:
  expected_pacing_rate
    Which is: 10.78 Mbits/s (1.35 Mbytes/s)
  sender_->PacingRate(0)
    Which is: 10.78 Mbits/s (1.35 Mbytes/s)

Merge internal change: 132613274

R=rch@chromium.org
BUG=

Review-Url: https://codereview.chromium.org/2336913006
Cr-Commit-Position: refs/heads/master@{#418696}

The overflow size is the content size without being affected by width/height
and max/min properties, so it's exactly what we need here.

This is important because without this change, there are DCHECK failures
if we're a child of a flexbox as the size remains at -1.

R=eae@chromium.org,ikilpatrick@chromium.org,glebl@chromium.org
BUG=635619

Review-Url: https://codereview.chromium.org/2337763005
Cr-Commit-Position: refs/heads/master@{#418695}

https://chromium.googlesource.com/chromiumos/platform/system_api.git/+log/83212329..6d7653e3

$ git log 83212329..6d7653e3 --date=short --no-merges --format='%ad %ae %s'
2016-09-12 yusukes@google.com Add a dbus constant for EmitArcBooted()

BUG=b:31254747
TEST=gclient sync, try

Review-Url: https://codereview.chromium.org/2344643002
Cr-Commit-Position: refs/heads/master@{#418694}

https://chromium.googlesource.com/external/khronosgroup/webgl.git/+log/582169b..65c5e7b

BUG=563852

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/2339603004
Cr-Commit-Position: refs/heads/master@{#418693}

Add some performance metrics for the MD and non-MD extensions pages so
we can begin gathering data. To start, use the signals for
DocumentLoadedInFrame() for the main frame and
DocumentOnLoadCompletedInMainFrame().

BUG=529395

Review-Url: https://codereview.chromium.org/2341553002
Cr-Commit-Position: refs/heads/master@{#418692}

This implements:

- allows the content embedder to merge a manifest into the base ones provided by content, allowing it to add to the list of interfaces exposed to the renderer.
- adds such an overlay to chrome/browser

So.. from now on if you add an interface to either the browser/renderer, you'll need to list it in the manifest. I wonder how we should broadcast this. Also as a separate step we should require these manifests get security review.

R=rockot@chromium.org,tsepez@chromium.org

Committed: https://crrev.com/4ea17059bf78553528f3ffb4a9de84a447622fd3
Committed: https://crrev.com/807a926c9ca02ac8ae1b52e82537834ef78fb9ba
Committed: https://crrev.com/91c25a3cedc4fadcb8c84e91cc59a8f90411058a
Review-Url: https://codereview.chromium.org/2259903002
Cr-Original-Original-Original-Commit-Position: refs/heads/master@{#418282}
Cr-Original-Original-Commit-Position: refs/heads/master@{#418398}
Cr-Original-Commit-Position: refs/heads/master@{#418582}
Cr-Commit-Position: refs/heads/master@{#418691}

Also make sure that we remember the color space even if the compositor has not
been created yet.

BUG=622133

Review-Url: https://codereview.chromium.org/2307083003
Cr-Commit-Position: refs/heads/master@{#418690}

Instead of only listing the number of nearby Physical Web URLs, also include
the page title of the top URL. The title will be truncated if it is too long
to fit in the omnibox popup.

BUG=630769

Review-Url: https://codereview.chromium.org/2319033006
Cr-Commit-Position: refs/heads/master@{#418689}

BUG=638297
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2339583002
Cr-Commit-Position: refs/heads/master@{#418688}

*PERF SHERRIF: this CL intentionally regressions dummy_benchmark.stable_benchmark_1 for bisect FYI.
We are doing this to refresh the revision ranges used by bisect FYI jobs.

BUG=646628
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.perf:linux_perf_cq;master.tryserver.chromium.perf:mac_retina_perf_cq;master.tryserver.chromium.perf:winx64_10_perf_cq

Review-Url: https://codereview.chromium.org/2337323002
Cr-Commit-Position: refs/heads/master@{#418687}

Revert of Moving CRD Windows targets to subdirectories. (patchset #4 id:770001 of https://codereview.chromium.org/2308813002/ )

Reason for revert:
https://build.chromium.org/p/chromium/builders/Win/builds/47116/
FAILED: obj/remoting/host/win/remoting_core/chromoting_lib.res
C:/b/depot_tools/python276_bin/python.exe ...
../../remoting/host/win/chromoting_lib.rc(5) : error RC2135 : file not found: remoting/host/chromoting_lib.tlb

Original issue's description:
> Moving CRD Windows targets to subdirectories.
>
> This is the windows host focused GN rewrite for Remoting Host. This
> work attempts to simplify the build for Remoting so it can be more
> easily understood and maintained. This is done by pushing more of the target config checking deeper into the platform specific targets and
> producing build files that are more in-line with best-practice GN style.
>
> Committed: https://crrev.com/ac59f3fd88f185f04e1cd177fcc2d05a425308e6
> Cr-Commit-Position: refs/heads/master@{#418668}

TBR=zijiehe@chromium.org,joedow@chromium.org,brettw@chromium.org,nicholss@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review-Url: https://codereview.chromium.org/2339163004
Cr-Commit-Position: refs/heads/master@{#418686}

Do not collapse top/bottom margins if there is a padding or border between
first/last in-flow child and its parent.
W3C spec: https://www.w3.org/TR/CSS2/box.html#collapsing-margins

BUG=635619

Review-Url: https://codereview.chromium.org/2346473003
Cr-Commit-Position: refs/heads/master@{#418685}

CodeMirror.refresh is expensive, and calling it when CodeMirror is hidden breaks the gutters.

BUG=none

Review-Url: https://codereview.chromium.org/2342643002
Cr-Commit-Position: refs/heads/master@{#418684}

With a comment that if you're doing something beyond add/remove of
files you should get an owner.

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

Review-Url: https://codereview.chromium.org/2338403002
Cr-Commit-Position: refs/heads/master@{#418683}