GitLab

Changes:
1) Run clang-format through source and header files.
2) Rename data member variables to use unix_hacker_ style.
3) Rename methods to use CamelCase style.
4) Rename file name to mock_spell_check.

BUG=331299

Review URL: https://codereview.chromium.org/410283003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285566 0039d316-1c4b-4281-b951-d872f2087c98

Instead, app_list::OmniboxProvider is responsible to set true to SearchTermsArgs::from_app_list.

BUG=388515

Review URL: https://codereview.chromium.org/399063002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285565 0039d316-1c4b-4281-b951-d872f2087c98

Revert of [WebSocket] Send a close frame when the renderer process is gone. (https://codereview.chromium.org/390773002/)

Reason for revert:
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%282%29/builds/5350

Original issue's description:
> [WebSocket] Send a close frame when the renderer process is gone.
> 
> Renderer processes may die without clearing its |WebSocketBridge|s.
> It is helpful to send a close frame with code = 1001 in such a case.
> 
> BUG=392534
> R=ricea@chromium.org
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=285557

TBR=ricea@chromium.org,jgraettinger@chromium.org,eroman@chromium.org,mmenke@chromium.org,yhirano@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=392534

Review URL: https://codereview.chromium.org/416333004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285564 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TBR=jianli@chromium.org (for notifications_api.cc)

Review URL: https://codereview.chromium.org/415143002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285562 0039d316-1c4b-4281-b951-d872f2087c98

During StartAnimation first it does a search for all animations which are
running and starting. After that it again searches the entire animation
list for animations waiting for target availability.

This patch reduces this search (for loops) by keeping track of the
animations waiting for target availability in the first search itself.


BUG=396562

Review URL: https://codereview.chromium.org/408833002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285561 0039d316-1c4b-4281-b951-d872f2087c98

Having a RequestAnimationFrame API on CommandBuffer makes no sense - the
context itself has nothing to do with scheduling and may not even be
swapping onscreen.  The spinning cube app was using this API as a way to
get delayed tasks without depending on base - basically
posttask-via-mojo-message. This adds basic delayed task support to the
public RunLoop and uses that instead. This only supports posting tasks
from the thread that owns the runloop. We can expand as needed. The mojo
RunLoop attempts to interleave delayed tasks with handle handlers, like
base's MessageLoop.

Review URL: https://codereview.chromium.org/384513003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285560 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285559 0039d316-1c4b-4281-b951-d872f2087c98

http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog_blink.html?url=/trunk&range=178842:178911&mode=html

TBR=jchaffraix@chromium.org

Review URL: https://codereview.chromium.org/418223006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285558 0039d316-1c4b-4281-b951-d872f2087c98

Renderer processes may die without clearing its |WebSocketBridge|s.
It is helpful to send a close frame with code = 1001 in such a case.

BUG=392534
R=ricea@chromium.org

Review URL: https://codereview.chromium.org/390773002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285557 0039d316-1c4b-4281-b951-d872f2087c98

BUG=336558
TEST=NO_TVCM=1 tools/telemetry/run_tests browser finder path

Review URL: https://codereview.chromium.org/412553006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285556 0039d316-1c4b-4281-b951-d872f2087c98

BUG=384861
TBR=jochen@chromium.org

Review URL: https://codereview.chromium.org/410993002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285555 0039d316-1c4b-4281-b951-d872f2087c98

Revert of skia/ext: Early out from analysis when we have more than 1 draw op. (https://codereview.chromium.org/418093003/)

Reason for revert:
This change broke some tests in Blink
Additionally, it renders DevTools shortcuts screen unusable (Open devTools -> F1 -> Shortcuts tab)

http://test-results.appspot.com/dashboards/flakiness_dashboard.html#showExpectations=true&tests=virtual%2Fandroid%2Ffullscreen%2Ffull-screen-fixed-pos-parent.html%2Cvirtual%2Fandroid%2Ffullscreen%2Ffull-screen-iframe-zIndex.html%2Cvirtual%2Fandroid%2Ffullscreen%2Ffull-screen-zIndex.html

Original issue's description:
> skia/ext: Early out from analysis when we have more than 1 draw op.
> 
> This patch puts in a heuristic that earlies out from analysis when
> we have more than one draw op. The intent here is that in most cases,
> if we see more than one draw operation, then analysis is likely not
> to be solid.
> 
> It also removes HasText from analysis canvas, since the only
> remaining use for this was a previous early out heuristic.
> 
> BUG=397198
> R=enne
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=285499

TBR=enne@chromium.org,senorblanco@google.com,junov@chromium.org,vmpstr@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=397198

Review URL: https://codereview.chromium.org/416333003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285554 0039d316-1c4b-4281-b951-d872f2087c98

Windows: Fix a corner case where we still think Adobe Reader is up to date after it has been uninstalled.

BUG=370746
TEST=Manual, see https://code.google.com/p/chromium/issues/detail?id=370746#c43

Review URL: https://codereview.chromium.org/419163003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285553 0039d316-1c4b-4281-b951-d872f2087c98

We'll need this for the new and improved WaitMany.

R=darin@chromium.org

Review URL: https://codereview.chromium.org/419613003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285552 0039d316-1c4b-4281-b951-d872f2087c98

The NSSCertDatabase singleton for Linux is now maintained by nss_context_linux.cc .

BUG=329735

Review URL: https://codereview.chromium.org/405973003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285551 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
Test=None, internal cleanup.

Review URL: https://codereview.chromium.org/411253009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285550 0039d316-1c4b-4281-b951-d872f2087c98

One more clean up before getting 397391 done.

HasGDocFileExtension is a very thin wrapper of the latter
function, and what all the callers need is the former interface.
This CL merges those two functions into one to make it more
easily searchable where in our code is relying on the extensions.

BUG=397391

Review URL: https://codereview.chromium.org/416223003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285549 0039d316-1c4b-4281-b951-d872f2087c98

BUG=
NOTRY=True

Review URL: https://codereview.chromium.org/408323002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285548 0039d316-1c4b-4281-b951-d872f2087c98

Clean up code a bit in MockPeerConnectionImpl. This addresses an issue in non-clang builds on ubuntu 12.04 using gcc 4.6.4 which has been in there for a while now:

The error is as follows:

../../content/renderer/media/mock_peer_connection_impl.cc: In member function
'virtual bool content::MockPeerConnectionImpl::GetStats(webrtc::StatsObserver*,
webrtc::MediaStreamTrackInterface*,
webrtc::PeerConnectionInterface::StatsOutputLevel)':
../../content/renderer/media/mock_peer_connection_impl.cc:280:3: error: in C++98
'value' must be initialized by constructor, not by '{...}'
../../content/renderer/media/mock_peer_connection_impl.cc:293:5: error: in C++98
'value2' must be initialized by constructor, not by '{...}'

R=xians@chromium.org

Review URL: https://codereview.chromium.org/415403002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285547 0039d316-1c4b-4281-b951-d872f2087c98

To make it easier to refer to a specific property in the documentation.

Review URL: https://codereview.chromium.org/421443002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285546 0039d316-1c4b-4281-b951-d872f2087c98

BUG=396466

Review URL: https://codereview.chromium.org/411253002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285545 0039d316-1c4b-4281-b951-d872f2087c98

BUG=
NOTRY=True

Review URL: https://codereview.chromium.org/410753002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285544 0039d316-1c4b-4281-b951-d872f2087c98

* drive::util::GetHostedDocumentMimeType is useless, the mime type
is directly obtained from ResourceEntry, especially after we've
moved to Drive API (see Bug 308836 for subtlety).
* kWapiRootDirectoryResourceId is just not used anywhere.

BUG=397391

Review URL: https://codereview.chromium.org/414203003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285543 0039d316-1c4b-4281-b951-d872f2087c98

This patch ensures that we only create tiling data iterators when
we visit the corresponding region.

Relevant perf data from a macbook:

Before:
tiling_raster_tile_iterator_construct:
0_0_100x100= 5828515 runs/s
50_0_100x100= 5530272 runs/s
100_0_100x100= 2730390 runs/s
150_0_100x100= 3735413.25 runs/s

tiling_raster_tile_iterator_construct_and_iterate:
32_100x100= 576050 runs/s
32_500x500= 601867.5625 runs/s
64_100x100= 353348.21875 runs/s
64_500x500= 362777.46875 runs/s

After:
tiling_raster_tile_iterator_construct:
0_0_100x100= 7913765 runs/s (+36%)
50_0_100x100= 7875735 runs/s (+42%)
100_0_100x100= 3382876.5 runs/s (+23%)
150_0_100x100= 5191805 runs/s (+39%)

tiling_raster_tile_iterator_construct_and_iterate:
32_100x100= 599836.6875 runs/s (+4%)
32_500x500= 629638.75 runs/s (+5%)
64_100x100= 357503.9375 runs/s (+<1%)
64_500x500= 372805.71875 runs/s (+3%)

R=reveman

Review URL: https://codereview.chromium.org/411183002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285542 0039d316-1c4b-4281-b951-d872f2087c98

Originally, ExtensionUninstallDialog took a Browser argument (which
determined which browser window it should be modal to), and a special
case was added so the dialog would be modal to the app list if the
Browser argument was NULL. Updated this constructor to take the parent
NativeWindow directly instead of a Browser, so the dialog can be modal
to any parent window. For example, this allows it to be parent to the
App Info dialog, and be a proper standalone dialog in
ExtensionStorageMonitor. Further refactors can be made as a result of
this CL since some of the callsites could be getting a Browser just to
be able to launch the dialog, whereas that is not needed anymore if the
intention is to launch it as a non-modal dialog.

Updated uninstall dialogs launched from the following callsites:
- The management Chrome API
- The extensions context menu
- The 'app disabled' message for when it has requested new permissions
- The chrome://extensions page
- The context menu of apps in the app list
- The 'remove' button in the App Info dialog
- The context menu on the New Tab page

BUG=388746,362308

Review URL: https://codereview.chromium.org/382133003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285541 0039d316-1c4b-4281-b951-d872f2087c98

Same what https://chromiumcodereview.appspot.com/11830002 did for ui_resources.

BUG=396209

Review URL: https://codereview.chromium.org/417493002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285540 0039d316-1c4b-4281-b951-d872f2087c98

Ephemeral apps should not trigger Chrome background mode and the
"New background app added" notification bubble should not be
displayed when they are launched as they do not have background
activity after they are closed.

BUG=339001
TEST=unit_tests

Review URL: https://codereview.chromium.org/383113002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285539 0039d316-1c4b-4281-b951-d872f2087c98

Add support for seccomp bpf sandboxing on MIPS architecture.
Enable testing of seccomp bpf sandbox.

Support for seccomp bpf for MIPS was added in Linux kernel version 3.15.

This patchset was reviewed in https://chromiumcodereview.appspot.com/260793003/ and re-created
to workaround a commit-queue issue.

BUG=369594
TEST=sandbox_linux_unittests

Review URL: https://codereview.chromium.org/409403003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285538 0039d316-1c4b-4281-b951-d872f2087c98

In search result, the entries have same names and it causes unstability of the
sort. The CL let the compare function to see URL when the names are same.

The CL also moves the sort functions from the UI class to the model class.

BUG=397386
TEST=manually

Review URL: https://codereview.chromium.org/419173002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285537 0039d316-1c4b-4281-b951-d872f2087c98

BUG=392981
TEST=manually

Review URL: https://codereview.chromium.org/419063004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285536 0039d316-1c4b-4281-b951-d872f2087c98

To show the focus ring in the correct size, the CL enlarges the size of range
inputs in Gallery. Instead, the CL adds pseudo element to draw the range track
(a thin horizontal bar).

BUG=397062
TEST=manually

Review URL: https://codereview.chromium.org/419153002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285535 0039d316-1c4b-4281-b951-d872f2087c98

Leverage OnFocus event to let component IME extension know about the current screen type (e.g. login, lock, etc.) so that on-screen keyboard won't popup new window/page at login/lock screens.

BUG=395621
TEST=None

Review URL: https://codereview.chromium.org/418273002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285534 0039d316-1c4b-4281-b951-d872f2087c98

- store the properties into DistilledPageInfo
- add test using OpenGraphProtocol and Schema.Org markup.

BUG=396124
TBR=nyquist

Review URL: https://codereview.chromium.org/411253008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285533 0039d316-1c4b-4281-b951-d872f2087c98

This refactor extracts common functionality to simplify adding the equivalent OpenSSL implementation:

 * Extracts the conversion of public exponent from a big integer to an unsigned integer
 * Restricts key sizes to multiples of 8, between 256 and 16K. This is the more restrictive range supported by NSS.

BUG=389317

Review URL: https://codereview.chromium.org/401233004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285532 0039d316-1c4b-4281-b951-d872f2087c98

Add a /update resource which accepts a new software package (assuming
the hash is correct) and performs an in-place upgrade. To avoid leaving
the device in an unusable state the old version remains running until
the new version starts to accept HTTP connections. At this point the
Tornado I/O loop is running correctly and even if USB functionality
is broken the device should accept another software update to fix it.

BUG=396682
R=rockot@chromium.org,rpaquay@chromium.org,kalman@chromium.org

Review URL: https://codereview.chromium.org/412243002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285531 0039d316-1c4b-4281-b951-d872f2087c98

'WebURLRequest::TargetType' has been dropped in Blink in favor of
'RequestContext' and 'FrameType', matching the Fetch specification's
changes, and allowing both more granularity in various Blink-side checks
and more clarity around the thing that caused a request, and the frame
that the request affects.

This CL should have zero practical effect, as it's simply hoisting the
translation work that 'WebURLRequest::setTargetType' is already doing
up into the content layer to bring us one step closer to removing the
concept entirely.

BUG=390497

Review URL: https://codereview.chromium.org/392333004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285530 0039d316-1c4b-4281-b951-d872f2087c98

If chrome::VersionInfo::GetVersionStringModifier() returns an empty
string (which is the case for the Stable channel and any unofficial
builds), don't append a space.

R=rtenneti@chromium.org,thestig@chromium.org
BUG=none

Review URL: https://codereview.chromium.org/420613004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285529 0039d316-1c4b-4281-b951-d872f2087c98

Fixed the crash when adding a chromium sink to remote audio track or adding a track to peerconnection more than once.

NOTRY=true

BUG=396447
TEST=open peerConnection demo, http://googlechrome.github.io/webrtc/samples/web/content/peerconnection/
2. start a call
3. Open JS console and input below codes:
# adding a chromium sink to remote audio track:
var v = new webkitMediaStream();
v.addTrack(pc2.getRemoteStreams()[0].getAudioTracks()[0]);
var a = document.createElement('audio');
a.src = URL.createObjectURL(v);

# add a track to peer connection more than once:
var v = new webkitMediaStream();
v.addTrack(pc2.getRemoteStreams()[0].getAudioTracks()[0]);
pc1.addStream(v);

Review URL: https://codereview.chromium.org/411903002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285528 0039d316-1c4b-4281-b951-d872f2087c98

There are a 8 basic test cases of varying types including text only html files, some html files with forms, an image, and a color-gradient.
BUG=388517

Review URL: https://codereview.chromium.org/416433002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285527 0039d316-1c4b-4281-b951-d872f2087c98

Removes an erroneous DCHECK added in this patch: https://codereview.chromium.org/385013002/

This DCHECK is erroneous, because the Install() function in DefaultComponentInstaller is actually called on a different task runner than the member task runner.

It's called from ComponentUnpacker::Install. ComponentUnpacker has a different task runner with a different sequence token, so it's erroneous to DCHECK it against the member task runner of DefaultComponentInstaller.

BUG=397264

Review URL: https://codereview.chromium.org/419103002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@285526 0039d316-1c4b-4281-b951-d872f2087c98