GitLab

The last public use of it is part of the pre-MD profile switcher.

BUG=633986

Review-Url: https://codereview.chromium.org/2556833002
Cr-Commit-Position: refs/heads/master@{#437967}

Review-Url: https://codereview.chromium.org/2566983004
Cr-Commit-Position: refs/heads/master@{#437966}

The test has been flakily failing on linux today

TBR=wangxianzhu@chromium.org

BUG=673539

Review-Url: https://codereview.chromium.org/2571553004 .
Cr-Commit-Position: refs/heads/master@{#437965}

See https://codereview.chromium.org/2512473004/ for additional context.

BUG=669638
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2555453003
Cr-Commit-Position: refs/heads/master@{#437964}

I messed up the mapping. As penance, add tests for everything and make
the Android tests less of a mess.

The failure mode here is P-521 client certificates won't quite work
right in TLS 1.3. I believe TLS 1.2 and below would not have been
affected.

BUG=672972

Review-Url: https://codereview.chromium.org/2567523003
Cr-Commit-Position: refs/heads/master@{#437963}

If VrCore exists on non-Daydream devices, WebVR Cardboard support should be enabled.

BUG=671370

Review-Url: https://codereview.chromium.org/2556093002
Cr-Commit-Position: refs/heads/master@{#437962}

The child process launcher now switches to the host executable's
path before running a new child process. That breaks mash_runner
code which launches the host executable if the host executable was
invoked by relative path.

This changes mash_runner to always reference the host executable
by absolute path when rewriting child command lines.

BUG=None
R=sky@chromium.org

Review-Url: https://codereview.chromium.org/2573593002
Cr-Commit-Position: refs/heads/master@{#437961}

Assets are yet to be added to this change.
BUG=664367

Review-Url: https://codereview.chromium.org/2531223003
Cr-Commit-Position: refs/heads/master@{#437960}

Since external experiment ids are not based on Chrome's low
entropy source, they do not have the same guarantees about
not identifying a user as Chrome's variations. As such, we
should only transmit them for signed in users, whose identity
is already known by Google so there's no risk of identifying
them through these headers.

Note: The signed-in state checking in this CL is only done for
web content area requests and not other internal requests,
like to the suggestion service, where it treats the state as
"not signed in". This is fine to do because variations service
ids are still sent, which is what the other call sites are
interested in.

BUG=672532
TBR=mpearson@chromium.org,mattm@chromium.org,donnd@chromium.org,afakhry@chromium.org

Review-Url: https://codereview.chromium.org/2558913003
Cr-Commit-Position: refs/heads/master@{#437959}

The browser tests for ResourcePrefetchPredictor check that the actual
request priority is equal to the expected value. But for the images the
priority cannot be precisely defined because there is a race between the
preload scanner issuing fetch requests and frame view performing layout
and promoting priorities of all visible images.

BUG=650253,673028

Review-Url: https://codereview.chromium.org/2569723002
Cr-Commit-Position: refs/heads/master@{#437958}

This is useful when the user creates a bound interface ptr and passes it without
actually using it to make any calls, which is pretty common.

BUG=579646

Review-Url: https://codereview.chromium.org/2563593005
Cr-Commit-Position: refs/heads/master@{#437957}

When changing from one shadow type to another, cross fade the two.

BUG=608852

Review-Url: https://codereview.chromium.org/2562143002
Cr-Commit-Position: refs/heads/master@{#437956}

The goal of this CL is to modify content_browser's IMEDriver so we have access
to ui::InputMethodChromeOS's functionality in mus+ash.

This CL:
- Modifies ui::InputMethodChromeOS so it can call an ack callback after
  processing events.
- Introduces InputMethodBridge to acts as a bridge between
  ui::mojom::InputMethod and ui::InputMethodChromeOS.
- Introduces RemoteTextInputClient to send commands received from
  ui::InputMethodChromeOS to a remote client over mojo IPC.
- Modifies content_browser's IMEDriver implementation to use
  ui::InputMethodChromeOS via InputMethodBridge.
- Some unittests to verify that InputMethodBridge works correctly.

BUG=665981

Review-Url: https://codereview.chromium.org/2557493002
Cr-Commit-Position: refs/heads/master@{#437955}

BookmarkBarController has a weak pointer in a |pulsingButton_| data
member. This is an NSView which can be destroyed in
-[BookmarkBarController redistributeButtonsOnBarAsNeeded] when it is
removed from its superview.

Since bubbles close asynchronously, and fade out, the bookmark bubble
can cause a UAF to |pulsingButton_| via stopPulsingBookmarkNode. This
can occur if a resize triggers redistributeButtonsOnBarAsNeeded while a
bookmark bubble is still alive.

There's no need for this to be a weak pointer. Make it scoped_nsobject
instead.

This affects both Cocoa and MacViews.

BUG=616051
TEST=Added unit_test:
BookmarkBarControllerTest.RedistributeButtonsOnBarAsNeeded

Review-Url: https://codereview.chromium.org/2567973002
Cr-Commit-Position: refs/heads/master@{#437954}

wayland-protocols: Add device_configuration_exists error to zcr_keyboard_configuration_v1 interface.

This error is produced if get_keyboard_device_configuration method is
called with a keyboard object that is already associated with
keyboard device configuration object.

BUG=670247

Review-Url: https://codereview.chromium.org/2564353002
Cr-Commit-Position: refs/heads/master@{#437953}

This patch adds a check to ensure that all of the required tiles are
in the now bin.

R=ericrk@chromium.org, sunnyps@chromium.org
BUG=622080
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2566603004
Cr-Commit-Position: refs/heads/master@{#437952}

Re-land https://codereview.chromium.org/2531953002 and
https://codereview.chromium.org/2547933003.

When a negative margin top pushes a block back up into its previous siblings
we need to check for any floats in those siblings it now needs to avoid or clear.

Previously we were just looking at its neighbour, we need to keep looking until
we reach a sibling that we don't overlap.

Fix the performance regression on mossiella.com in page_cycler_v2.tough_layout_cases
by only looking for floats if margin collapsing has actually moved the child up.

BUG=671645, 670325, 666487

Review-Url: https://codereview.chromium.org/2568643002
Cr-Commit-Position: refs/heads/master@{#437951}

This lets you drag files into the url bar to open them.

BUG=none

Review-Url: https://codereview.chromium.org/2566753003
Cr-Commit-Position: refs/heads/master@{#437950}

IntersectionGeometry is an ephemeral object, so make it stack
allocated, let it use raw pointers to LayoutObject, and move it to
core/layout.

Also, be explicit about the implicit root by introducing
rootIsImplicit.

R=ojan@chromium.org,xjz@chromium.org
BUG=

Review-Url: https://codereview.chromium.org/2553103004
Cr-Commit-Position: refs/heads/master@{#437949}

When a child of an overflow: hidden object is composited (and the ancestor is not) we were failing to apply the border-radius clip. This patch fixes the problem by adding a new masking layer for this situation, and modifying the layers used for painting to set the clip appropriately.

New tests are added to verify the behavior for border styles, and existing layout tests cover the bug (and fail before this patch). Unit testing is used for verifying update behavior.

There are remaining issues tracked in TODOs with corresponding bugs.

BUG=157218
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2194273002
Cr-Commit-Position: refs/heads/master@{#437948}

- The avatar button should only be set as active if the a bubble is
opened
- Removed the code for the fast chooser in the cocoa files

BUG=670613

Review-Url: https://codereview.chromium.org/2552423002
Cr-Commit-Position: refs/heads/master@{#437947}

BUG=663187

Review-Url: https://codereview.chromium.org/2511313002
Cr-Commit-Position: refs/heads/master@{#437946}

It was crashing due to StopIteration exception

BUG=671803

Review-Url: https://codereview.chromium.org/2565303003
Cr-Commit-Position: refs/heads/master@{#437945}

BUG=607968,660216

Review-Url: https://codereview.chromium.org/2568873003
Cr-Commit-Position: refs/heads/master@{#437944}

Revert of [ObjC ARC] Converts ios/chrome/browser/ui:ui to ARC. (patchset #1 id:1 of https://codereview.chromium.org/2569713002/ )

Reason for revert:
Speculatively reverting because downstream device tests are failing.

Original issue's description:
> [ObjC ARC] Converts ios/chrome/browser/ui:ui to ARC.
>
> Automatically generated ARCMigrate commit
> Notable issues:None
> BUG=624363
> TEST=None
>
> Committed: https://crrev.com/54842ee261813f25854ace5fd23854238a3f544a
> Cr-Commit-Position: refs/heads/master@{#437891}

TBR=marq@chromium.org,stkhapugin@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=624363

Review-Url: https://codereview.chromium.org/2567203002
Cr-Commit-Position: refs/heads/master@{#437943}

This CL also includes a few fixes for loading a WebUILoginView instance by itself without associated helper classes also being initialized.

There are three child CLs associated with this one:
- [1] contains fixes on the HTML side.
- [2] makes the lock screen show up correctly in the task manager.
- [3] preloads and reuses the WebUIScreenLocker instance (derived from WebUILoginView).

1: https://codereview.chromium.org/2555453003/
2: https://codereview.chromium.org/2550263002/
3: https://codereview.chromium.org/2553863002/

There is additional context at go/cros-lock-screen-optimization.

BUG=669638
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2512473004
Cr-Commit-Position: refs/heads/master@{#437942}

Most code that calls the UserGestureIndicator knows which thread it is
executing in. Calling isMainThread involves doing a thread id syscall which
can be quite slow, and there are places in the loading code that do this
in a tight loop (e.g. preloading resources).

This patch adds an optional argument to the indicator's static methods which
optionally allow callers to get the gesture state in a thread-safe way.

BUG=348655

Review-Url: https://codereview.chromium.org/2555013004
Cr-Commit-Position: refs/heads/master@{#437941}

This is the case for CronetSample.apk

BUG=none

Review-Url: https://codereview.chromium.org/2566293002
Cr-Commit-Position: refs/heads/master@{#437940}

AutoplayUmaHelperTest currently uses GMock for mock
AutoplayUmaHelper in order to check if events are correctly
received. However this will make the test flaky since
MockAutoplyUmaHelper is garbage-collected but GMock expect it to
be destroyed when the test finishes.

This CL marks MockAutoplayUmaHelper as AllowLeak which fixes the
issue.

BUG=672831

Review-Url: https://codereview.chromium.org/2562683004
Cr-Commit-Position: refs/heads/master@{#437939}

With no cursor in readonly editor, highlight disappears very fast and
it's impossible to say where it was.

BUG=670417

Review-Url: https://codereview.chromium.org/2574443003
Cr-Commit-Position: refs/heads/master@{#437938}

Split out from:
https://codereview.chromium.org/2559573002/

There is a downstream caller for one method, so will
land this first to allow a quicker patch time.

TBR=yusufo@chromium.org
BUG=671149

Review-Url: https://codereview.chromium.org/2565183005
Cr-Commit-Position: refs/heads/master@{#437937}

Stability and performance numbers from the current 50% GPU process enable rate
look good. Move on to the third experiment configuration in the associated bug,
to test performance impact of enabling the profiler for both GPU and browser
processes vs. enabling for neither process.

This CL deviates from the roll out percentages in the bug by performing the
experiment on 10% of launches rather than 50%. The experiment will be enabled
over the holiday period, so the lower percentage will provide enough data while
minimizing data volume differences for the browser process.

BUG=650869

Review-Url: https://codereview.chromium.org/2571483003
Cr-Commit-Position: refs/heads/master@{#437936}

The CNNIC whitelist was temporary, and only contained EE certs.

Remove the whitelist, fully distrusting CNNIC.

BUG=673083

Review-Url: https://codereview.chromium.org/2565743004
Cr-Commit-Position: refs/heads/master@{#437935}

BUG=none

Review-Url: https://codereview.chromium.org/2568993002
Cr-Commit-Position: refs/heads/master@{#437934}

As it happens, renderers actually do this, and we don't want to
unnecessarily crash them. Instead, use the previous behaviour of
returning OUTPUT_DEVICE_STATUS_ERROR_NOT_FOUND instead.

One way to trigger a nonsense id being sent to the browser is by using
the setSinkId method of an HTMLMediaElement.

BUG=671167

Review-Url: https://codereview.chromium.org/2561343002
Cr-Commit-Position: refs/heads/master@{#437933}

BUG=672852
TEST=No more content_unittests, events_unittests, and extensions_unittests failures on linux_chromium_tsan_rel_ng
Confirmed @ https://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_tsan_rel_ng/builds/377

Review-Url: https://codereview.chromium.org/2565893002
Cr-Commit-Position: refs/heads/master@{#437932}

Adds an optional header to DateDividedAdapter and adds a new layout
with a clear browsing data button and privacy disclaimers as a header
in the history ui.

BUG=654071

Review-Url: https://codereview.chromium.org/2562123003
Cr-Commit-Position: refs/heads/master@{#437931}

BUG=671459,671653

Review-Url: https://codereview.chromium.org/2569753002
Cr-Commit-Position: refs/heads/master@{#437930}

The intent can be saved by android and re-used for launching a child
service in edge cases. However the command line arguments are different
for each launch of the child process. The command line args are already
passed through an additional bundle through the aidl interface. So just
fallback to the aidl bundle and remove command line args from the intent
bundle.

Note this may and probably does decrease parallelism in child service
start up, thus causing a perf regression.

BUG=664341

Review-Url: https://codereview.chromium.org/2560403002
Cr-Commit-Position: refs/heads/master@{#437929}

We have no way of telling what changes beyond the first column break, so if we
cannot guarantee that the column height *is* and *was* non-uniform, we need to
re-lay out children that may stretch into the unknown.

Review-Url: https://codereview.chromium.org/2562273003
Cr-Commit-Position: refs/heads/master@{#437928}