GitLab

Also don't bother compiling it on ChromeOS at all.

Review-Url: https://codereview.chromium.org/2285843002
Cr-Commit-Position: refs/heads/master@{#414781}

R=iannucci@chromium.org
BUG=640251

Review-Url: https://codereview.chromium.org/2282933002
Cr-Commit-Position: refs/heads/master@{#414780}

Unless the user manually tabs in the message center bubble, we shouldn't request
the focus by default to prevent the ugly blue border from showing automatically
[See screenshot: https://drive.google.com/a/google.com/file/d/0B6G_-uQnf1_LelZhbUN2aWtMSzg/view?usp=sharing]

BUG=620801

Review-Url: https://codereview.chromium.org/2281983002
Cr-Commit-Position: refs/heads/master@{#414779}

Need to alloc bitmap before creating canvas or it will create
an empty bitmap instead of one in WHITE.

BUG=636630

Review-Url: https://codereview.chromium.org/2238173003
Cr-Commit-Position: refs/heads/master@{#414778}

BUG=622445

Review-Url: https://codereview.chromium.org/2275513003
Cr-Commit-Position: refs/heads/master@{#414777}

We can no longer listen after events on something that is an instance in
a <use>. Move the executeTest() call to the <use> element instead.

Since the mentioned file is almost a stub, the actual change is in
svg/animations/script-tests/use-animate-width-and-height.js.

BUG=641398

Review-Url: https://codereview.chromium.org/2283893002
Cr-Commit-Position: refs/heads/master@{#414776}

Interface control messages were added before union support was ready so
the structs used for messages manually imitate unions. Now that union
support is ready, this workaround is no longer necessary and the structs
used for interface control messages can contain unions of the possible
message types.

Review-Url: https://codereview.chromium.org/2277853003
Cr-Commit-Position: refs/heads/master@{#414775}

BUG=none

Review-Url: https://codereview.chromium.org/2272043002
Cr-Commit-Position: refs/heads/master@{#414774}

The context menu code for gesture long-press used to fire mousedown &
contextmenu events with wrong button/buttons attributes (w.r.t. both
the spec & other browsers). This CL fixes those attribute values.

Moreover, the existing code achieved the desired outcome of a
long-press (setting focus before context menu) through calling the raw
PlatformMouseEvent entrypoint. This CL cherry-picks the needed steps,
thus provides a cleaner alternative to double-pointer-event firing
(minimally fixed through part of crrev.com/2141993003). We don't expect
any change in behavior for this part of the CL.

BUG=579564,629876

Review-Url: https://codereview.chromium.org/2249663002
Cr-Commit-Position: refs/heads/master@{#414773}

As the old bitmap DesktopView is removed, there is no need to allow the
DesktopView to directly access the render data. This CL moves mRenderData into
TouchInputHandler to simplify the logic of the DesktopView a little bit.

BUG=641123

Review-Url: https://codereview.chromium.org/2272253004
Cr-Commit-Position: refs/heads/master@{#414772}

BUG=639416

Review-Url: https://codereview.chromium.org/2260933002
Cr-Commit-Position: refs/heads/master@{#414771}

affects MAC OSX for channels other than canary and Linux for
all the channels. Other platforms don't go through the same
first run flow.

BUG=600391

Review-Url: https://codereview.chromium.org/2285483002
Cr-Commit-Position: refs/heads/master@{#414770}

ProcessManifest can return null in a few failure modes.
This result gets passed into Reader::OnReadManifest which
was doing no null checking. This fixes that.

BUG=632916
TBR=ben@chromium.org

Review-Url: https://codereview.chromium.org/2284843002
Cr-Commit-Position: refs/heads/master@{#414769}

DevTools: remove DevToolsTargetDescriptor and its implementations, we are now based on devtools agent hosts.

BUG=627517
TBR=this is a mechanical change

Review-Url: https://codereview.chromium.org/2273063002
Cr-Commit-Position: refs/heads/master@{#414768}

This change results in chunks from the input getting removed if they
contain repetitive strings or mostly spaces.

BUG=

Review-Url: https://codereview.chromium.org/2283683002
Cr-Commit-Position: refs/heads/master@{#414767}

This CL allows collecting v8 runtime call stats data as part of deep reports.

R=oysteine@chromium.org
BUG=641402

Review-Url: https://codereview.chromium.org/2282013003
Cr-Commit-Position: refs/heads/master@{#414766}

https://codereview.chromium.org/1821863002 changed the display scheduler
on Android to use the CompositorImpl ExternalBeginFrameSource directly
instead of a synthetic BeginFrameSource that had its vsync parameters
set from WindowAndroidCompositor::OnVSync.

The frame time given to the display scheduler is always in the past,
which made the past logic always give an immediate deadline, causing
the display scheduler to run immediately.  Fix this by providing a
correct deadline in the future.

This fixes bad jank where the display scheduler would never give any
other compositors time to give frames.  So, if the browser compositor
was producing frames (such as a loading spinner) then the renderer
frames would be extremely janky as they would often miss the immediate
(incorrect) deadline.

BUG=613193
R=sievers@chromium.org

Review-Url: https://codereview.chromium.org/2275253003
Cr-Commit-Position: refs/heads/master@{#414765}

BUG=

Review-Url: https://codereview.chromium.org/2277103003
Cr-Commit-Position: refs/heads/master@{#414764}

SSL status needs to be reset if navigation fails with error. But
back-forward navigation calls navigation failure callback with
"cancelled" error. In this case SSL status should not be changed.

The other case where navigation failure callback is called with
cancelled error is load cancellation which should not reset SSL status
as well.

BUG=637832

Review-Url: https://codereview.chromium.org/2286803002
Cr-Commit-Position: refs/heads/master@{#414763}

This CL fixes the bug where UnindexedRulesetWriter doesn't trim the
CodedOutputStream on Finish, resulting in the output having some extra
uninitialized bytes at the end.

BUG=609747

Review-Url: https://codereview.chromium.org/2279803002
Cr-Commit-Position: refs/heads/master@{#414762}

BUG=

Review-Url: https://codereview.chromium.org/2282793002
Cr-Commit-Position: refs/heads/master@{#414761}

When download snackbar shows up, clicking the open button will open
download home for offline page.
This CL corrects that behavior so that the corresponding offline page
will be opened.

BUG=641262

Review-Url: https://codereview.chromium.org/2274573008
Cr-Commit-Position: refs/heads/master@{#414760}

When a fullscreen quad is promoted to overlay, it should be removed
from quad_list to be GL composited.

This CL fixes a bug where OverlayStrategyFullscreen assumed that the
first quad would always be the one to remove.

BUG=b/29430506
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2280823002
Cr-Commit-Position: refs/heads/master@{#414759}

BluetoothChooserController, which is in chrome/, was including BluetoothMetrics
which is in content/. To avoid this we create a new histogram specifically
for interactions with the chooser.

BUG=641105

Review-Url: https://codereview.chromium.org/2278003003
Cr-Commit-Position: refs/heads/master@{#414758}

https://pdfium.googlesource.com/pdfium.git/+log/9b57ec9ff50a..3e98158a6c47

$ git log 9b57ec9ff..3e98158a6 --date=short --no-merges --format='%ad %ae %s'
2016-08-26 tonikitoo Extend pdfium_test capability so that more Javascript can be executed

TBR=dsinclair@chromium.org

Review-Url: https://codereview.chromium.org/2281053002
Cr-Commit-Position: refs/heads/master@{#414757}

This change adds a StyleSheetResourceClient on all CSS
preloads. The client holds a CSSPreloadScanner which receives
the stream of CSS as it comes in.

This is guarded by two Blink settings,
cssExternalScannerPreload and cssExternalScannerNoPreload.
The former allows for scanning and preloading and the latter just
allows for scanning, as a baseline for the overhead of the feature.

This is partially a reland of
https://codereview.chromium.org/1819593002, which was
reverted manually here:
https://codereview.chromium.org/1947053002/

The revert was due to a crash and some faulty logic with
regard to all css preload requests being considered
"referenced". This CL attempts to fix that by adding an enum
to mark a ResourceClient as "passive".

BUG=610437,596676

Review-Url: https://codereview.chromium.org/1976463003
Cr-Commit-Position: refs/heads/master@{#414756}

This is an initial Blimp integration test that exercises navigation history. The purpose is to test the integration with WebContentsImpl and NavigationControllerImpl, as well as the proper flow of Blimp's NavigationStateChangeMessage protobuf across the client-engine network connection.

The test visits two different web pages and then goes back to the first one.

Additional supporting changes:
* Moving some setup code from the initial LoadUrl test into SetUpOnMainThread.
* Helper method ExpectPageLoad, which sets mock expectations for a page load, and can be used with the helper method RunAndVerify to wait for Blimp to load a page.
* Always storing the latest page title when it changes on the client side, making it easier to assert about what page was most recently loaded.

Review-Url: https://codereview.chromium.org/2272513003
Cr-Commit-Position: refs/heads/master@{#414755}

Notification PendingIntents having the same name will conflict
with each other. This CL fixes the conflict for MediaNotification
stop button and swipe away, which now fire different actions but
we treat both as ACTION_STOP.

BUG=640301

Review-Url: https://codereview.chromium.org/2279263002
Cr-Commit-Position: refs/heads/master@{#414754}

Remove the mostly useless CountBits abstraction and better align the
generic feature and CSS property usage tracking.  Also a couple other
minor cleanups and comments warning of outstanding issues.

BUG=597963

Review-Url: https://codereview.chromium.org/2284503002
Cr-Commit-Position: refs/heads/master@{#414753}

The main use of these by end-users is to disable cursor blink by setting them
to extremely large values.

BUG=500168

Review-Url: https://codereview.chromium.org/2282533002
Cr-Commit-Position: refs/heads/master@{#414752}

Also tighten up the matcher to also require that |value| field:
- is public,
- is inside a struct-or-class that doesn't have any methods.

BUG=640749

Review-Url: https://codereview.chromium.org/2283553002
Cr-Commit-Position: refs/heads/master@{#414751}

This was mostly tracked because while lost the main thread Renderer
capabilities would not be valid. However a few things have changed now:
1) No more renderer capabilities. So
2) We don't care if we main frame without an output surface.
3) We don't defer commits on the main thread so we wouldn't try to
start a frame on the main thread while the output surface is lost
anyways.

The only exception is SingleThreadProxy needs to know to request
and output surface when doing synchronous composite things. So it does
that itself now.

R=enne
BUG=606056
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2270573002
Cr-Commit-Position: refs/heads/master@{#414750}

BUG=635644

Review-Url: https://codereview.chromium.org/2273373003
Cr-Commit-Position: refs/heads/master@{#414749}

This CL:
* Uses Mus' IMEServer to do the IME logic in InputMethodMus.
* Adds unittests to test InputMethodMus.
* Starts test_ime_driver on mus+ash session startup.
* Modifies test_ime_driver to not-handle non character events.

BUG=548407

Review-Url: https://codereview.chromium.org/2230393002
Cr-Commit-Position: refs/heads/master@{#414748}

(This is my first CL so please let me know if I'm doing something wrong)

BUG=614963

Review-Url: https://codereview.chromium.org/2273363004
Cr-Commit-Position: refs/heads/master@{#414747}

The copy constructor was trying to call DeepCopy() against a nullptr. This adds
a check if the type is NULLTYPE and return base::Value() if so.

BUG=b:30963813
TEST=call the copy constructor with a NULLTYPE instance to see if it crashes

Review-Url: https://codereview.chromium.org/2284713002
Cr-Commit-Position: refs/heads/master@{#414746}

NGConstraintSpace now has a backing NGPhysicalConstaintSpace and
provides abstract coordinate system accessors for everything.

Makes NG*ConstraintSpace GarbagedCollected as well as we'll need
this once we move to a state machine / for caching fragment results.

BUG=635619

Review-Url: https://codereview.chromium.org/2267383003
Cr-Commit-Position: refs/heads/master@{#414745}

BUG=None

Review-Url: https://codereview.chromium.org/2277903004
Cr-Commit-Position: refs/heads/master@{#414744}

BUG=641400
TBR=michaelpg@chromium.org
NOTRY=true
NOTREECHECKS=true

Review-Url: https://codereview.chromium.org/2284823002
Cr-Commit-Position: refs/heads/master@{#414743}

This CL removes calls to these methods from gpu/ files:
- MessageLoop::PostTask
- MessageLoop::PostDelayedTask
- MessageLoop::ReleaseSoon
- MessageLoop::DeleteSoon
- MessageLoop::Run
- MessageLoop::RunUntilIdle

BUG=616447
CQ_INCLUDE_TRYBOTS=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/2274213004
Cr-Commit-Position: refs/heads/master@{#414742}