GitLab

When the change is rolled in, I'll rebaseline and enable the tests.

BUG=329407
TBR=dominicc,esprehn
NOTRY=true

Review URL: https://codereview.chromium.org/138003010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245116 0039d316-1c4b-4281-b951-d872f2087c98

The OVERRIDE needs to be right after the method name, and before other
attributes (e.g. WARN_UNUSED_RESULT) or gcc 4.8 gives a build error.
Clang does not seem to mind it though.

Review URL: https://codereview.chromium.org/139703005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245115 0039d316-1c4b-4281-b951-d872f2087c98

BUG=330082

Review URL: https://codereview.chromium.org/138933002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245114 0039d316-1c4b-4281-b951-d872f2087c98

BUG=334049

Review URL: https://codereview.chromium.org/138913002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245113 0039d316-1c4b-4281-b951-d872f2087c98

There are some unnecessary includes in generated thunks. We should remove these
for clarity, and (hopefully) to reduce build time. A small test showed that
ppb_view_thunk.cc built 18% faster after the change, but that's well within the
noise on my local system.

BUG=334727

Review URL: https://codereview.chromium.org/137353003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245111 0039d316-1c4b-4281-b951-d872f2087c98

Send multiple queries to the network in short intervals to account for potential
packet loss on noisy networks.

BUG=332629

Review URL: https://codereview.chromium.org/137343005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245110 0039d316-1c4b-4281-b951-d872f2087c98

that have changed from the previous context.

The most CPU consuming operations on the GPU thread for
Android are virtual context switches. We attempt to reduce the CPU usage by
context switches with this patch.

BUG=244701

Review URL: https://codereview.chromium.org/118203002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245109 0039d316-1c4b-4281-b951-d872f2087c98

Android mediaplayer won't be able to handle blob urls.
After xhr requests creates a temporary file with blob url, we need to pass the file path to the media player so it can play the file.

TBR=tedchoc
BUG=253465

Review URL: https://codereview.chromium.org/98823003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245108 0039d316-1c4b-4281-b951-d872f2087c98

There is too many abstractions in our surface classes to
determine 'real' SwapBuffers calls in traces. This adds
traces in all the relevant places where we do 'real' swaps.

BUG=334063

Review URL: https://codereview.chromium.org/137823002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245107 0039d316-1c4b-4281-b951-d872f2087c98

BUG=194304
NOTRY=true

Review URL: https://codereview.chromium.org/137803002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245106 0039d316-1c4b-4281-b951-d872f2087c98

Before, we regarded any HTML form on the landing page after a login attempt as failure.

After this CL, it's only a failure, if the new form has the same action as the old one.

BUG=329860

Review URL: https://codereview.chromium.org/137783003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245105 0039d316-1c4b-4281-b951-d872f2087c98

The change was needed because when there was a lot of information to be
displayed, and a if the query was not pasted, then the incremental search
took too long and the partial results rendering was unnecesarily refreshing
the results pane when what was actually needed was just the full query.

BUG=333414

Review URL: https://codereview.chromium.org/137753002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245104 0039d316-1c4b-4281-b951-d872f2087c98

Tab media indicator data model updates indirectly cause TabGtk::UpdateTooltipState() to be invoked several times per second.  This in turned resulted in calling gtk_widget_set_tooltip_text() redundantly.  GTK responds to any tooltip text setter calls by hiding *all* tooltips throughout the application.

The fix is to check whether the tooltip text actually is different, and not call the setter when the text is unchanged.

In the bigger picture, there are other times when normal browser functionality would cause UpdateTooltipState() to be called several times per second.  In general, the current implementation for handling tab data model updates is expensive and in bad need of rewriting.  However, efforts would be better spent on moving Linux Chrome over to Aura.

BUG=333002
TEST=Confirmed fixed using bug repro steps.
R=erg@chromium.org

Review URL: https://codereview.chromium.org/138413006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245103 0039d316-1c4b-4281-b951-d872f2087c98

BUG=334098
TEST=none

Review URL: https://codereview.chromium.org/137323005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245102 0039d316-1c4b-4281-b951-d872f2087c98

This refactors some repeated code, as we're about to add a few more extension calls.

BUG=161119

Review URL: https://codereview.chromium.org/137313003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245101 0039d316-1c4b-4281-b951-d872f2087c98

Remove most of the places where we content::RunMessageLoop() and then elsewhere
base::MessageLoop::current()->Quit(). Too many parts of the tests involve
spinning the message loop and waiting for one of three or four events to
happen, some complex conditions on some of the events when we want AND instead
of OR.

The primary change is a TestPrerender object vended out by the
TestPrerenderFactory. This object is a PrerenderObserver and holds a non-owning
reference to the TestPrerenderContents. Its lifetime is not tied to the
TestPrerenderContents and is managed by the caller. TestPrerender exposes
methods to wait for various events which replaces the content::RunMessageLoop
calls.

The hope is that this will make it easier to write tests which create multiple
prerenders and allow us to unbreak and reenable tests disabled because the old
wait conditions clobbered each other.

BUG=307592, 249179
TEST=Prerender*.*

Review URL: https://codereview.chromium.org/137263002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245100 0039d316-1c4b-4281-b951-d872f2087c98

BUG=334690

Review URL: https://codereview.chromium.org/137253006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245099 0039d316-1c4b-4281-b951-d872f2087c98

Move the DisplayConfigurationObserver to shell delegate's PreInit which is the right place to do initialization that happens before Shell::Init.

BUG=332852

Review URL: https://codereview.chromium.org/137193004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245098 0039d316-1c4b-4281-b951-d872f2087c98

When video gets interrupted by audio focus change, we sent an DidMediaPlayerPause IPC to WebMediaPlayer, and release the mediaplayer at the same time.
However, when we want to resume the playback, WebMediaPlayer does not know that the mediaplayer needs to reestablish the surface texture.
Instead of sending DidMediaPlayerPause, we should send MediaPlayerReleased IPC.
See internal b/12362468

BUG=

Review URL: https://codereview.chromium.org/137103004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245097 0039d316-1c4b-4281-b951-d872f2087c98

Android's InputEventConsistencyVerifier expects that all MotionEvents in a given
sequence have the same source type. Fix an issue where the source type for synthetic
cancellation MotionEvents went unspecified in ContentViewGestureHandler.  Also
fix an issue where MotionEvents were processed out-of-order with respect to
double-tap drag zoom and snap scrolling.

Review URL: https://codereview.chromium.org/137103002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245096 0039d316-1c4b-4281-b951-d872f2087c98

BUG=154400

Review URL: https://codereview.chromium.org/134463008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245095 0039d316-1c4b-4281-b951-d872f2087c98

BUG=330664
TEST=ash_unittests --gtest_filter=*DockedWindowResizerTest.ResizingKeepsDockedState*

Review URL: https://codereview.chromium.org/136983002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245094 0039d316-1c4b-4281-b951-d872f2087c98

BUG=333914
R=xiyuan@chromium.org
TEST=manually

Review URL: https://codereview.chromium.org/139293007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245093 0039d316-1c4b-4281-b951-d872f2087c98

This will help diagnosing bugs caused by the timeout. This change also
changes the UMA sampling of the times involved in the fetching process,
since UMA_HISTOGRAM_TIMES is clamped to a maximum sample of 10 seconds.

BUG=334107

Review URL: https://codereview.chromium.org/139293002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245092 0039d316-1c4b-4281-b951-d872f2087c98

BUG=324216
R=glider@chromium.org

Review URL: https://codereview.chromium.org/139263003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245091 0039d316-1c4b-4281-b951-d872f2087c98

views::corewm::CursorManager was no longer owned by anything, nor did it
register with the base::AtExitManager. This refcounts the number of DNWAs
and frees the shared CursorManager when the last one is destroyed. (We
can't just use a base::LazyInstance<> since that doesn't have any useful
guarantees on destruction ordering; we need to destroy CursorManager
before we tear down the X11 connection.)

BUG=334333
R=tdanderson
TBR=sadrul

Review URL: https://codereview.chromium.org/134103006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245090 0039d316-1c4b-4281-b951-d872f2087c98

Previously in RootWindowTest.TouchMovesHeld, sometimes a SHOW_PRESS
gesture would be fired, and sometimes it wouldn't, depending on
whether or not the 5ms timer used in tests for the SHOW_PRESS
gesture was fired. Now we explicitly wait for the gesture to be
fired before continuing.

Marked flaky here: https://codereview.chromium.org/136013002/

TEST=RootWindowTest.TouchMovesHeld
BUG=333644

Review URL: https://codereview.chromium.org/136843003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245087 0039d316-1c4b-4281-b951-d872f2087c98

There was only one implementation of the ClientPlugin interface, and we
don't have plans to add other implementations, so it doesn't make sense
to have it as interface. Moved ClientPluginAsync to ClientPlugin.

R=jamiewalch@chromium.org

Review URL: https://codereview.chromium.org/138503007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245086 0039d316-1c4b-4281-b951-d872f2087c98

ash_strings where included in Android build and none of them are actually used.
Restrict the include to only ash builds.

BUG=none

Review URL: https://codereview.chromium.org/136793004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245085 0039d316-1c4b-4281-b951-d872f2087c98

In the case where we use CookieManager before initializing the rest of
Chromium, we need to have initialized url_util at some reliable point
because its lazy init is not threadsafe. Do this in the entry point so
it's guaranteed we've always called it before any Chromium code runs.

BUG=334490

Review URL: https://codereview.chromium.org/136723010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245084 0039d316-1c4b-4281-b951-d872f2087c98

Reason for revert:
Causes blink layout test (media/encrypted-media/encrypted-media-needkey.html) to crash:

http://test-results.appspot.com/dashboards/flakiness_dashboard.html#group=%40ToT%20Blink&showExpectations=true&tests=media%2Fencrypted-media%2Fencrypted-media-needkey.html

Original issue's description:
> Add Stop() to AudioDecoder.
> 
> BUG=329379
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=244994

TBR=dalecurtis@chromium.org,scherkus@chromium.org,xhwang@chromium.org,rileya@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=329379

Review URL: https://codereview.chromium.org/139513007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245083 0039d316-1c4b-4281-b951-d872f2087c98

BUG=126823

Review URL: https://codereview.chromium.org/136603003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245082 0039d316-1c4b-4281-b951-d872f2087c98

When using SAML, the e-mail address typed into the GAIA login form and
the e-mail address of the user who actually ends up authenticating may
not match. This CL ensures that the correct e-mail address gets used by
querying GAIA for the authenticated user's e-mail instead of using the
e-mail obtained from the GAIA login form.

BUG=332132
TEST=New browser test

Review URL: https://codereview.chromium.org/136573002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245081 0039d316-1c4b-4281-b951-d872f2087c98

The 'DBus.Connection' property of a shill cellular device object is
deprecated. This CL fixes NetworkSMSHandler to use the DBus service
name, obtained from the 'DBus.Service' property of a shill cellular
device object, to create a DBus object proxy that interacts with a modem
object managed by cromo or ModemManager.

BUG=334337
TEST=Tested the following:
1. Run chromeos_unittests.
2. Send a SMS to a E936 modem, which is managed by cromo, and verify
   that Chrome pops up a notification of the received SMS.
3. Repeat (2) using a MU736 modem, which is managed by ModemManager.
4. Repeat (2) using a pseudomodem.

Review URL: https://codereview.chromium.org/136463004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245080 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none

Review URL: https://codereview.chromium.org/136193002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245079 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=true

Review URL: https://codereview.chromium.org/135993003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245078 0039d316-1c4b-4281-b951-d872f2087c98

BUG=334319
NOTRY=True

Review URL: https://codereview.chromium.org/138503003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245077 0039d316-1c4b-4281-b951-d872f2087c98

This causes problems for try bots as the app shims cannot connect to
the test chrome processes. Also are created in user-local (rather than
test-local) locations.

BUG=334079

TEST=Tests that install apps (e.g.PlatformAppBrowserTest) should
not create app shims (in ~/Applications/Chromium Apps/).

Review URL: https://codereview.chromium.org/135853008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245076 0039d316-1c4b-4281-b951-d872f2087c98

BUG=334473
TEST=None
R=jamescook
TBR=benwells (For trivial refactor in chrome/browser/ui/views/apps/native_app_window_views.cc)

Review URL: https://codereview.chromium.org/135053006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245075 0039d316-1c4b-4281-b951-d872f2087c98

Some window can not be maximized. We should not make
'Maximize' the default menu item for them.

BUG=333146
TEST=Open a window inside Chrome that can not be maximized,
for example, the "Report an issue" window in "About Chrome".
The menu in the title bar should not have "Maximize" in bold.

Review URL: https://codereview.chromium.org/137543006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245072 0039d316-1c4b-4281-b951-d872f2087c98