GitLab

R=sky@chromium.org
TEST=Repro steps in bug #168746.
BUG=166348

Review URL: https://chromiumcodereview.appspot.com/11784036

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175674 0039d316-1c4b-4281-b951-d872f2087c98

This will track Shill Device info, including Wifi access point data.

BUG=167987


Review URL: https://chromiumcodereview.appspot.com/11743006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175673 0039d316-1c4b-4281-b951-d872f2087c98

BUG=168652


Review URL: https://chromiumcodereview.appspot.com/11788020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175672 0039d316-1c4b-4281-b951-d872f2087c98

Change the method to no longer write to both App GUIDs, just the one that matches the distribution of the caller. This means that GCAPI will only write to the Google Chrome product's client state key, and not to the Chrome Binaries.

Replace SetOmahaExperimentLabel in the GCAPI code with a more specific version, and have SetOmahaExperimentLabel perform the general clobber-write to the registry.

This is in preparation for future work in using the experiment_labels to transmit Chrome Variations.

BUG=160251


Review URL: https://chromiumcodereview.appspot.com/11280067

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175671 0039d316-1c4b-4281-b951-d872f2087c98

Now it doesn't dismiss the item. When click happens in the toast, it will hide from the popup but remain in the center.

BUG=165576

Review URL: https://codereview.chromium.org/11778012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175669 0039d316-1c4b-4281-b951-d872f2087c98

> Omnibox: Make HistoryQuick provider not re-issue queries if |minimal_changes| is true
> 
> BUG=
> 
> 
> Review URL: https://chromiumcodereview.appspot.com/11722020

I'm reverting this because it causes problems with backspace, at least when
certain field trials are enabled.  See
https://code.google.com/p/chromium/issues/detail?id=168725

We'll probably chose a different strategy to fix this in a better way (in
terms of how the code should work), but at least this puts the behavior
back to what people are used to.

BUG=168725
TBR=mpearson@chromium.org
Review URL: https://codereview.chromium.org/11820015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175668 0039d316-1c4b-4281-b951-d872f2087c98

Context menu button isn't useful for editing item and it casuses
a problem when displaying context menu on editing item.

BUG=76189

Review URL: https://codereview.chromium.org/11777026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175667 0039d316-1c4b-4281-b951-d872f2087c98

Host and NaCl builds currently use different headers which can cause
build problems.  This change moves the headers out of the toolchains
and instead all toolchains share the same set as the host builds.

NOTRY=true
BUG=168718
R=binji@chromium.org

Review URL: https://chromiumcodereview.appspot.com/11825003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175666 0039d316-1c4b-4281-b951-d872f2087c98

This change allows you to compile all .cc files for nacl-mounts. Previously you
need to include compile kernel_wrap.cc, but only copy (not compile) the other
kernel_wrap_*.cc files.

This is convenient for someone making a visual studio project -- they can just
drop all the .cc files into a project and everything "just works".

BUG=none
R=noelallen@chromium.org
NOTRY=true

Review URL: https://chromiumcodereview.appspot.com/11744010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175665 0039d316-1c4b-4281-b951-d872f2087c98

This updates a few more classes to be observers of
PowerManagerClient in addition to RootPowerManagerClient.
After powerm is moved into powerd, all of the
RootPowerManagerClient code will be deleted.

BUG=chromium-os:36804


Review URL: https://chromiumcodereview.appspot.com/11824005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175664 0039d316-1c4b-4281-b951-d872f2087c98

TBR=atwilson@chromium.org
BUG=168882
TEST=none
NOTRY=true


Review URL: https://chromiumcodereview.appspot.com/11825008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175663 0039d316-1c4b-4281-b951-d872f2087c98

Also allow InstantExtended field trial groups to override this behavior and set espv values as they see fit, to provide for flexibility for future experiments.

BUG=168298

Review URL: https://codereview.chromium.org/11743035

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175660 0039d316-1c4b-4281-b951-d872f2087c98

There is a linux adb binary checked in to platform-tools. We use it if on linux
and the adb command isn't found.

BUG=None
TEST=tools/perf/run_multipage_benchmarks --browser=list -v

Review URL: https://codereview.chromium.org/11830011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175659 0039d316-1c4b-4281-b951-d872f2087c98

Over in related https://codereview.chromium.org/11649028/ , I had to add HISTOGRAM_BOOLEAN. It's not right that UMA_HISTOGRAM_* and HISTOGRAM_* are not the same. This CL doesn't fix that, but it does at least make things less bad.

R=isherman@chromium.org
BUG=None


Review URL: https://chromiumcodereview.appspot.com/11794020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175658 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://chromiumcodereview.appspot.com/11788022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175657 0039d316-1c4b-4281-b951-d872f2087c98

Adds a mechanism to TemplateURL to optionally allow search term extraction only when a given key is present (for Google, this is "espv"). Uses this to ensure ToolbarModel no longer relies on the hardcoded "espv" parameter.

Depends on:
  https://codereview.chromium.org/11552020/

BUG=161602


Review URL: https://chromiumcodereview.appspot.com/11576044

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175656 0039d316-1c4b-4281-b951-d872f2087c98

BUG=168244
Owner:
TBR=sky@chromium.org

Review URL: https://chromiumcodereview.appspot.com/11766002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175655 0039d316-1c4b-4281-b951-d872f2087c98

Possibly the change of 
http://src.chromium.org/viewvc/chrome?view=rev&revision=175642

Broke the mac plist string generation

BUG=168890
TEST=mac bot not red
TBR=thakis
Review URL: https://codereview.chromium.org/11829014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175654 0039d316-1c4b-4281-b951-d872f2087c98

> ThreadedTest that a non-null OutputSurface that fails initialization immediately doesn't crash
> 
> Unit test to catch things like bug 168691
> 
> BUG=
> 
> 
> Review URL: https://chromiumcodereview.appspot.com/11782026

The new tests this change added fail on Mac and linux 

TBR=jamesr@chromium.org
Review URL: https://codereview.chromium.org/11830013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175653 0039d316-1c4b-4281-b951-d872f2087c98

BUG=


Review URL: https://chromiumcodereview.appspot.com/11829005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175651 0039d316-1c4b-4281-b951-d872f2087c98

This doesn't change over the lifetime of the prcoess, so we can
easily keep ahold of one instance.

BUG=


Review URL: https://chromiumcodereview.appspot.com/11791010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175650 0039d316-1c4b-4281-b951-d872f2087c98

See the attachments of crbug.com/166786#c2 the bubble has border for focus.

BUG=166786
TEST=open the message center bubble when there're no notifications and confirmed no focus borders appear.

Review URL: https://codereview.chromium.org/11737029

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175649 0039d316-1c4b-4281-b951-d872f2087c98

This CL localizes usage of DesktopEnvironment to a single method that creates audio/video schedulers and hook up the event executor. This prepares further refactoring of DesktopEnvironment that should become a factory for classes interacting with the specific desktop environment.

Related changes in this CL:
  - MouseClampingFilter receives screen size on the proper thread.
  - ClientSessionTest.* tests now run asynchronously to support MouseClampingFilter change.
  - ClientSession does not expose DesktopEnvironment.
  - InputEventTracker can be re-attached to a different InputStub.

BUG=104544


Review URL: https://chromiumcodereview.appspot.com/11781003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175648 0039d316-1c4b-4281-b951-d872f2087c98

BUG=166786
TEST=login and see if the shelf has the notification icon even though there're no notifications.

Review URL: https://codereview.chromium.org/11761032

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175647 0039d316-1c4b-4281-b951-d872f2087c98

R=tonyg
BUG=168905
NOTRY=True

Review URL: https://codereview.chromium.org/11824014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175646 0039d316-1c4b-4281-b951-d872f2087c98

This fixes the debug crash here:
DCHECK(!wm::IsWindowMinimized(gained_active))

This is testing kShowStateKey that has not yet been changed to the expected value.

As far as I can tell, when called from OnWindowActivated is already from ash::window::Show
when window.cc calls into the delegate:
   delegate_->OnWindowTargetVisibilityChanged(visible);

Which triggers BrowserView::RestoreFocus which triggers a cascade of focus notifications
ending up in aura::FocusManager::FocusWindow which calls ActivationController::ActivateWindow
which calls BaseLayoutManager::OnWindowActivated

In other word we re-enter and as a side effect the kShowStateKey has not yet been set.

Full stack in the bug.

BUG=166952
TEST=see bug
Review URL: https://codereview.chromium.org/11633029

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175645 0039d316-1c4b-4281-b951-d872f2087c98

Solution: the popup keeps mouse capture, so that any mouse click closes it (via WebKit's WebPopupMenuImpl.cpp) not just clicks in it or in its browser window.

BUG=161129
TEST=navigate to
http://www.w3schools.com/tags/tryit.asp?filename=tryhtml_select
Click on the "car name" to bring up the pop-up.
Click outside the browser window to see it close (just as for clicking inside the browser window)


Review URL: https://chromiumcodereview.appspot.com/11675002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175644 0039d316-1c4b-4281-b951-d872f2087c98

SystemMenuModel. I've made it so SystemMenuModel doesn't need it.

This is a precursor to sharing system menu building code among the
frame types.

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


Review URL: https://chromiumcodereview.appspot.com/11791039

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175643 0039d316-1c4b-4281-b951-d872f2087c98

This forces explicit normalization of paths and make path escaping security bugs much harder to exploit. See for example bug 167122

BUG=168890
TEST=included tests
Review URL: https://codereview.chromium.org/11782005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175642 0039d316-1c4b-4281-b951-d872f2087c98

disappeared, so removing debugging code.

BUG=166355
TEST=none


Review URL: https://chromiumcodereview.appspot.com/11775021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175641 0039d316-1c4b-4281-b951-d872f2087c98

BUG=83714
TEST=ViewSourceTest.ViewSourceInMenuDisabledOnAMediaPage


Review URL: https://chromiumcodereview.appspot.com/11775002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175639 0039d316-1c4b-4281-b951-d872f2087c98

Keep only one tab and only one icon animation object. So when
we add more cases we don't expand the class state.

BUG=3541
TEST=none
Review URL: https://codereview.chromium.org/11753021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175638 0039d316-1c4b-4281-b951-d872f2087c98

TBR=sky@chromium.org
BUG=157273

Review URL: https://codereview.chromium.org/11783039

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175636 0039d316-1c4b-4281-b951-d872f2087c98

to reflect the way it now works in developer console.

Developers no longer need to email Chrome apps to request their client ID.

Review URL: https://codereview.chromium.org/11761029

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175635 0039d316-1c4b-4281-b951-d872f2087c98

This change should stop the HTTP server thread from outputting the
following backtrace, which clutters the logs (and is sometimes
interleaved with other output):

Traceback (most recent call last):
  File ".../tools/browser_tester/browser_tester.py", line 247, in Serve
    server.handle_request()
  File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/SocketServer.py", line 262, in handle_request
    fd_sets = select.select([self], [], [], timeout)
select.error: (9, 'Bad file descriptor')

BUG=none
TEST=checked log output on Mac OS X

Review URL: https://codereview.chromium.org/11819005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175634 0039d316-1c4b-4281-b951-d872f2087c98

The switch kUseGpuInTests is used to inform tests that they are running on a GPU bot (instead of a bot which may not have a real GPU in it).

The existing GPU tests which use this flag are chrome tests, so this flag had been happy living in chrome. There exist some content GPU tests that don't use this flag, but the GPU memory manager test is the first content test which uses this flag (it renders some very complicated scenes to stress GPU memory, and we shouldn't count on being able to run all of them on software raster).

BUG=135525

Review URL: https://codereview.chromium.org/11777019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175633 0039d316-1c4b-4281-b951-d872f2087c98

Unit test to catch things like bug 168691

BUG=


Review URL: https://chromiumcodereview.appspot.com/11782026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175632 0039d316-1c4b-4281-b951-d872f2087c98

Also, makes non-badged shortcuts (when deleting down to 1 profile) not have a command-line or custom icon set, to be consistent with (i.e. the same as) the shortcut initially created for the browser when it's installed.

BUG=167981
TEST=With two profiles A and B, with only B having a shortcut, delete profile B. There should be a non-profile desktop shortcut created.

Review URL: https://chromiumcodereview.appspot.com/11785004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175631 0039d316-1c4b-4281-b951-d872f2087c98

This works around flickering on Qualcomm devices when
using virtual contexts. Qualcomm has a bug where switching
to FBO targets of different sizes corrupts the scissor rect.

BUG=165493

Review URL: https://chromiumcodereview.appspot.com/11734037

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175630 0039d316-1c4b-4281-b951-d872f2087c98

Since it's only used there. That way we will be able to move
window_open_disposition.h out of webkit/

BUG=146251
R=jam@chromium.org
TBR=darin@chromium.org

Review URL: https://codereview.chromium.org/11798003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@175629 0039d316-1c4b-4281-b951-d872f2087c98