GitLab

Revert of Avoids releasing capture in EndMoveLoop if capture was switched to another window while dragging (https://codereview.chromium.org/265843004/)

Reason for revert:
Appears to be responsible for ash_unittests failing on Linux ChromiumOS Tests(1): http://build.chromium.org/p/chromium.chromiumos/builders/Linux%20ChromiumOS%20Tests%20%281%29/builds/43114

Original issue's description:
> Avoids releasing capture in EndMoveLoop if capture was switched to another window while dragging.
> This variant relies on checking if a drag input window still has capture before releasing the grab.
> A better alternative is in CL https://codereview.chromium.org/262893002/.
> 
> BUG=363503
> TEST=Follow the exact steps in the bug description.
> TEST=interactive_ui_tests --gtest_filter=*TabDragging*
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=269593

TBR=sadrul@chromium.org,varkha@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=363503

Review URL: https://codereview.chromium.org/276163002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269629 0039d316-1c4b-4281-b951-d872f2087c98

Reason for revert:
Memory leak detected: http://build.chromium.org/p/chromium.memory/builders/Linux%20Chromium%20OS%20ASan%20LSan%20Tests%20%281%29/builds/1239/steps/aura_unittests/logs/stdio

Original issue's description:
> Unified Gesture Recognizer
>
> This adds the unified GR for Aura behind the --use-unified-gesture-detector flag.
>
> BUG=332418
> TEST=GestureRecognizer/GestureRecognizerTest.*
>   MotionEventUITest.*
>
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=269325

Review URL: https://codereview.chromium.org/278183002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269628 0039d316-1c4b-4281-b951-d872f2087c98

It emits NCN.NetworkOperatorMCCMNC_NewMetricsLog on new metric log creation with the network operator MCC_MNC code 

BUG=355604

Review URL: https://codereview.chromium.org/253203002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269627 0039d316-1c4b-4281-b951-d872f2087c98

Also add an explicit check for whitelisting pepper/nacl
permissions based on importing a shared module, so a shared
module can potentially limit allowing import of its resources
and permissions its allowed.

You can still use nonsfi mode for Linux or ChromeOS by
passing in --enable-nacl-nonsfi-mode, but it is on for a
whitelisted set of extensions for ChromeOS without flags.

BUG=355141
TEST=deploy_chrome to daisy, test app which includes whitelisted module

Review URL: https://codereview.chromium.org/264923011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269626 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=true

Review URL: https://codereview.chromium.org/275683002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269625 0039d316-1c4b-4281-b951-d872f2087c98

BUG=chromium:369679
TEST=Tested manually on daisy

Review URL: https://codereview.chromium.org/276023002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269624 0039d316-1c4b-4281-b951-d872f2087c98

This has been broken since https://src.chromium.org/viewvc/blink?revision=172846&view=revision landed.

Review URL: https://codereview.chromium.org/276983002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269623 0039d316-1c4b-4281-b951-d872f2087c98

Enable auto-refreshing of the gcm-internals page whenever a GCM activity is recorded, and fix a bug that previously causes this cl to crash the browser upon signin.

Basically GCMStatsRecorder::RecordRegistrationResponse() method did not check whether is_recording_ is true or not, so it always records. When the user signs in, a registration event is automatically triggered, but since the gcm-internals page has not been loaded yet, a callback object is still NULL. Therefore recorder calling this callback will crash the browser. I have added the check as well as test case for this scenario.

BUG=341256, 371259

Review URL: https://codereview.chromium.org/277743002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269622 0039d316-1c4b-4281-b951-d872f2087c98

If the process is unable to create a new thread we've got bigger issues
to deal with than reporting an error. A CHECK() will also give us
a sense for how often this happens in the wild.

BUG=110814

Review URL: https://codereview.chromium.org/276593007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269621 0039d316-1c4b-4281-b951-d872f2087c98

BUG=371897
R=rockot@chromium.org
TEST=ran UsbApiTest.TransferEvent under Dr. Memory

Review URL: https://codereview.chromium.org/275993002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269620 0039d316-1c4b-4281-b951-d872f2087c98

HttpServer: Handling of multiple header fields with the same name and multiple values of "Connection".

BUG=370437

Review URL: https://codereview.chromium.org/274813002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269619 0039d316-1c4b-4281-b951-d872f2087c98

Separate out context creation that does not need surface or service
into separate CreateOffscreenContext call.

Do not create GLSurfaceStubs on client side, and pass in NULL window
so the context creates GLSurfaceStubs on service side.

Clean up share context code since compositor does not need an offscreen
context anymore.

BUG=370566

Review URL: https://codereview.chromium.org/271763002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269618 0039d316-1c4b-4281-b951-d872f2087c98

BUG=368936

Review URL: https://codereview.chromium.org/270333003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269617 0039d316-1c4b-4281-b951-d872f2087c98

This is the last change to deliver video frames to Cast streaming API
on the IO thread. It will no longer be blocked by Javascript events.

This needs a change in the MediaStreamVideoSink interface. Also getting
rid of the interface of MediaStreamVideoTrack that delivers frames on
the render thread. All clients now handle the thread hopping directly.

BUG=335327, 371775

Review URL: https://codereview.chromium.org/264363005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269616 0039d316-1c4b-4281-b951-d872f2087c98

finished.

BUG=362679

Review URL: https://codereview.chromium.org/263273002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269615 0039d316-1c4b-4281-b951-d872f2087c98

When config changes, we currently use a round trip IPC to request the new config.
This increases the overall latency when processing config changes.
This CL appends the new config data to the DemuxerData which contains the config change access unit.

BUG=325528

Review URL: https://codereview.chromium.org/257323003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269614 0039d316-1c4b-4281-b951-d872f2087c98

This is just a code clean-up. As long as it compiles, no change to the functionality.

BUG=348523

Review URL: https://codereview.chromium.org/271173002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269613 0039d316-1c4b-4281-b951-d872f2087c98

only add control types in addtional to user selectable types.

BUG=362679

Review URL: https://codereview.chromium.org/266883007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269612 0039d316-1c4b-4281-b951-d872f2087c98

BUG=367677
R=danakj@chromium.org,tfarina@chromium.org

Review URL: https://codereview.chromium.org/253723006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269611 0039d316-1c4b-4281-b951-d872f2087c98

I accidentally got the base64 <-> base64url conversion stuff wrong
in https://codereview.chromium.org/278593005/.

BUG=369895
TBR=rockot@chromium.org

Review URL: https://codereview.chromium.org/273093002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269610 0039d316-1c4b-4281-b951-d872f2087c98

Add baked-in Domain Reliability configs for {drive,mail}.google.com.

BUG=356791

Review URL: https://codereview.chromium.org/278773003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269609 0039d316-1c4b-4281-b951-d872f2087c98

Trying to get rid of more globals in ui/compositor.

BUG=None

Review URL: https://codereview.chromium.org/277713002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269608 0039d316-1c4b-4281-b951-d872f2087c98

This test was writing to the user's downloads directory instead of a
temporary path.

BUG=372017
R=asargent@chromium.org

Review URL: https://codereview.chromium.org/274193008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269607 0039d316-1c4b-4281-b951-d872f2087c98

BUG=371655

Review URL: https://codereview.chromium.org/271573003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269606 0039d316-1c4b-4281-b951-d872f2087c98

Also includes a few small cleanup changes I noticed while I was in there.

BUG=108012

Review URL: https://codereview.chromium.org/261843004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269605 0039d316-1c4b-4281-b951-d872f2087c98

If wrong notification is received, NOTREACHED() is called.

BUG=None
R=benwells@chromium.org

Review URL: https://codereview.chromium.org/263833019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269604 0039d316-1c4b-4281-b951-d872f2087c98

During Chrome shutdown, DeviceMonitorMac deallocs
CrAVFoundationDeviceObserver, which is registered as an
observer of suspended and connected events on the
CrAVCaptureDevice. Every so seldom (see bugs), the
removal of these observers causes a crash. Observers
removal seems to be a long standing problem with
Cocoa and Objective-C++.

This CL speculates with a solution for this trouble
via testing if the protocol method -observationInfo [1]
is not |nil| before removing the observers. This is
based on the observation that:
a) There might be an AVFoundation-internal race
between observer removal and AVCaptureDevice destruction.
b) A previous, unrelated Chrome crash might have
left the AVFoundation into some unstable meta state.
(This is based on xians@ hitting this bug during
browser_tests).

Tested locally and seems to work fine.

BUG=288562, 371271


[1] https://developer.apple.com/library/ios/documentation/Cocoa/Reference/Foundation/Protocols/NSKeyValueObserving_Protocol/Reference/Reference.html#jumpTo_10

Review URL: https://codereview.chromium.org/274073002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269603 0039d316-1c4b-4281-b951-d872f2087c98

When StartupTimebomb goes off, collect the duration stats (difference
between when the timebomb went off and when it is started). We collect
this duration in terms of clock time, montonic clock time and thread CPU
time.

BUG=366699
R=jar@chromium.org

Review URL: https://codereview.chromium.org/273963002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269602 0039d316-1c4b-4281-b951-d872f2087c98

ChromeStringCanonicalizer does case and diacritic insensitive string
comparison through an ICU Collator. To get general Unicode character
rules, the root locale should be used. The system default locale is
different from system to system, and using that will cause the code to
break when executed on a system where it doesn't have the properties
that this code assumes.

This change also adds #include statemens for all ICU header files from
which the code in this file uses declarations.

R=rouslan@chromium.org
BUG=

Review URL: https://codereview.chromium.org/275913002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269601 0039d316-1c4b-4281-b951-d872f2087c98

Check for NULL |activation_client| in ExtensionPopup::OnWidgetDestroying.

Add a BrowserActionInteractiveTest that crashes without this fix.
Add a BrowserActionTestUtil::InspectPopup helper on Views.

BUG=327776
TEST=Inspecting an extension popup bubble and closing the underlying browser window does not crash.
R=sky@chromium.org,finnur@chromium.org

Review URL: https://codereview.chromium.org/275783002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269600 0039d316-1c4b-4281-b951-d872f2087c98

URLFixerUpper may potentially mishandle "about:blank".
The suspicion arose from crbug.com/371030 investigations.
(this unit test case doesn't seem to trigger that, though)

BUG=NONE
TEST=NONE
R=sky@chromium.org

Review URL: https://codereview.chromium.org/279593003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269599 0039d316-1c4b-4281-b951-d872f2087c98

BUG=318769

Review URL: https://codereview.chromium.org/207743004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269598 0039d316-1c4b-4281-b951-d872f2087c98

The task manager is a CustomFrameView. This had the logic for its button
placement hardcoded. The view now adds itself as a listener for button
configuration changes on linux. For other systems it falls back on a
default set that matches the current (minimize, maximize, close) order.

The layout of the buttons has changed to consult this order definition.

Loading the assets for the close button has been moved out of layout.
ResourceBundle only loads the asset once, so this can be done during
initialization.

The top padding for the buttons has been updated to provide more spacing. It now fits the rest of the padding in the class.

TEST=CustomFrameViewTest
BUG=351917

Review URL: https://codereview.chromium.org/240163006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269597 0039d316-1c4b-4281-b951-d872f2087c98

This matches the current ChromeOS implementation.

BUG=355788
R=rpaquay@chromium.org

Review URL: https://codereview.chromium.org/279503003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269596 0039d316-1c4b-4281-b951-d872f2087c98

using Java conversion functions.  This is part of making it possible
to build net/ without ICU, to get binary size down when building net/
as a library

BUG=362608

Review URL: https://codereview.chromium.org/266053002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269595 0039d316-1c4b-4281-b951-d872f2087c98

Main thread animations provide no guarantees that the initial fling timestamp
will be strictly positive.  Consequently, a (valid) timestamp of 0 applied to
Android's fling animation curve will yield a zero-sized scroll delta.  Feeding
this scroll delta through WebViewImpl reports that scrolling is impossible, in
which case the fling will early terminate.  Avoid this case by early returning
on non-positive fling timestamps, aligning with Aura's fling behavior.

BUG=371854

Review URL: https://codereview.chromium.org/275113002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269594 0039d316-1c4b-4281-b951-d872f2087c98

This variant relies on checking if a drag input window still has capture before releasing the grab.
A better alternative is in CL https://codereview.chromium.org/262893002/.

BUG=363503
TEST=Follow the exact steps in the bug description.
TEST=interactive_ui_tests --gtest_filter=*TabDragging*

Review URL: https://codereview.chromium.org/265843004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269593 0039d316-1c4b-4281-b951-d872f2087c98

Someone noticed that occasionally the uploader sent an empty upload. I
looked into the scheduler and found that it was clearing the "upload
pending" bit early, when the upload was scheduled. This meant that if a
beacon arrived between scheduling and starting the upload, the bit was
still set on its behalf, despite it being included in the upload. Once
the upload finished, the scheduler would see the bit still set and
schedule another.

This patch adds a couple of unittests for interesting interleavings
(beacons that arrive between scheduling and starting the upload, and
between starting and finishing the upload), and clears the upload
pending bit in the right place.

BUG=356791

Review URL: https://codereview.chromium.org/272773004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269592 0039d316-1c4b-4281-b951-d872f2087c98

> With ongoing work on fixing bugs and performance in D3D11, as well
> as blacklisting old drviers, we might have more stable results on
> Intel. If possible we can narrow down ranges of crashing configs
> and blacklist those individually.
>
> BUG=363721

BUG=369475

Review URL: https://codereview.chromium.org/273723004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269591 0039d316-1c4b-4281-b951-d872f2087c98

Remove API version 1 of the credentials passing API and rename version 2
to version 1. This renumbering is backwards-incompatible and would break
any existing users of API version 1. However, the API is still under
development and thee are no users yet. Thus, we should take the
opportunity to remove the cruft of an obsolete experimental API version
before we release anything to the public.

BUG=367847
TEST=Updated browser test
TBR=nkostylev

Review URL: https://codereview.chromium.org/276903002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@269590 0039d316-1c4b-4281-b951-d872f2087c98