GitLab

This includes PNaCl's merge of LLVM 3.4.

Due to Clang 3.4's new warnings, I've disabled some warnings in the
NaCl SDK's build of gtest, which gave:
gtest-filepath.cc:87:12: error: unused variable 'kPathSeparatorString' [-Werror,-Wunused-const-variable]

This pulls in the following Native Client changes:

r12811: (jfb) Update TOOL_REVISIONS for PNaCl r12786->r12809
r12812: (kschimpf) Test pnacl-compress fixpointing.
r12813: (jfb) Update TOOL_REVISIONS for PNaCl r12809->r12810
r12814: (dyen) Attempt to unify our platform/OS and architecture collapsing in python
r12815: (jvoung) Remove NACL_RODATA_FLAG to go between -Trodata-segment, and --section-start.
r12816: (mseaborn) PNaCl: Merge LLVM 3.4: update Git revs, driver scripts and test expectations
r12817: (jfb) Update TOOL_REVISIONS for PNaCl r12810->r12816 to bring in merge of LLVM 3.4
r12818: (mseaborn) Fix download_toolchains.py (breakage from r12814)
r12819: (mseaborn) PNaCl: LLVM 3.4 merge: Update test expectations for flaky tests on Cygwin
r12820: (mcgrathr) Update chrome_rev to current LKGR
r12822: (noelallen) Roll bionic
r12823: (mseaborn) IRT: Hide the "dev_getpid" IRT interface by default
r12824: (noelallen) Roll bionic toolchain into TOOL_REVs
r12825: (bradnelson) Adding timeout mechanism to command tester.
r12826: (jvoung) Apply -f*-sections and --gc-sections to IRT build.
r12827: (mseaborn) IRT: Disable various IRT interfaces for in-browser-translated PNaCl pexes
r12828: (jvoung) Remove -reduce-memory-footprint: just make it the default w/ streaming.
r12829: (sbc) Roll glibc version in tools/REVISIONS
r12830: (bradnelson) Fix multidomain_test flake on x86-32 windows.
r12831: (jvoung) Make pnacl-readelf a slightly better wrapper around readelf.
r12832: (mseaborn) PNaCl: Remove toolchain's dependency on x86-32 version of libtinfo on Linux
r12833: (bradnelson) Wait longer for debug_stub_tests, kill more effectively on windows.
r12834: (jvoung) PNaCl: Update LLVM revision in pnacl/COMPONENT_REVISIONS
r12835: (jfb) Update TOOL_REVISIONS for PNaCl r12816->r12832

BUG=none
TEST=browser_tests and nacl_integration

Review URL: https://codereview.chromium.org/187673006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255395 0039d316-1c4b-4281-b951-d872f2087c98

The playlist goes wrong when the 'expanded' status is changed in some condition.

BUG=342610, chrome-os-partner:26202
TEST=manually tested

Review URL: https://codereview.chromium.org/185653014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255394 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255392 0039d316-1c4b-4281-b951-d872f2087c98

http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog_blink.html?url=/trunk&range=168618:168640&mode=html
TBR=
BUG=

Review URL: https://codereview.chromium.org/186663007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255391 0039d316-1c4b-4281-b951-d872f2087c98

Moved all state from SiteIsolationPolicy out to the PendingRequestInfo struct in the ResourceScheduler

This removes a few redundant map lookups, simplifies the logic, and also paves the way for using this class from multiple threads.
R=creis@chromium.org
BUG=277124

Review URL: https://codereview.chromium.org/187063006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255389 0039d316-1c4b-4281-b951-d872f2087c98

This is a work around for a driver bug on Galaxy Nexus

BUG=349667

Review URL: https://codereview.chromium.org/184103037

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255386 0039d316-1c4b-4281-b951-d872f2087c98

UserActionClient is used to navigate back/forward when the back/forward on a
supported mouse-device is clicked on X11. This can instead be achieved by
installing an event-handler on the browser window.

This does change the behaviour a little on ChromeOS: clicking these navigation
buttons when the cursor is not on top of the browser window will not navigate
after this change. I have confirmed with UX that this is a desirable change.

BUG=319636
R=sky@chromium.org

Review URL: https://codereview.chromium.org/183853037

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255385 0039d316-1c4b-4281-b951-d872f2087c98

app_shell needs the chrome.storage.local extension API so the core storage implementation is moving to //extensions/browser/api. However, the "managed storage" API implementation needs to remain in Chrome.

* Introduce ExtensionsApiClient in extensions/browser/api
* Introduce ChromeExtensionsApiClient in chrome/browser/extensions/api
* Delegate out the storage ManagedValueStoreCache creation to Chrome
* Move settings_namespace.h/cc to e/b/api/storage so we can use an enum in the client interface definition

Most APIs will probably need to make only one or two client interface calls, so we're starting with a single ExtensionsApiClient interface. It we find individual APIs tend to need a lot of methods we can split it into one client interface per API, but that seems like overkill right now.

Reland of https://codereview.chromium.org/180213008/ with fixes for Android GCMProfileServiceSingleProfileTest.Unregister

BUG=348058
TEST=unit_tests *Settings* and browser_tests ExtensionSettings*
TBR=rdevlin.cronin@chromium.org, rockot@chromium.org, yoz@chromium.org

Review URL: https://codereview.chromium.org/184853012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255384 0039d316-1c4b-4281-b951-d872f2087c98

BUG=
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/187423003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255383 0039d316-1c4b-4281-b951-d872f2087c98

> Switch Quickoffice to new internal repo
> 
> BUG=338855
> TEST=manual
> NOTRY=true
> R=jhawkins@chromium.org, zork@chromium.org
> 
> Review URL: https://codereview.chromium.org/187983003

TBR=dpolukhin@chromium.org

Review URL: https://codereview.chromium.org/183883032

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255382 0039d316-1c4b-4281-b951-d872f2087c98

BUG=338855
TEST=manual
NOTRY=true
R=jhawkins@chromium.org, zork@chromium.org

Review URL: https://codereview.chromium.org/187983003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255380 0039d316-1c4b-4281-b951-d872f2087c98

In Surprise Me mode, changes are: 
- Top categories don't highlight anymore on Hover
- Online Wallpaper Attributes (author + url + thumbnail) are not displayed
- Selected Item Checkmark is not displayed
- "Add New" button is not highlighted on Hover and don't prompt user anymore.
- White Outline on thumbnails is not displayed.

Screenshot: http://i.imgur.com/UPhSQZS.jpg

Review URL: https://codereview.chromium.org/183853030

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255379 0039d316-1c4b-4281-b951-d872f2087c98

This causes problems as it means the aura::Window may be deleted out
from under the DesktopNativeWidgetAura, which causes all sorts of woe.

This code is no longer needed since DesktopWindowTreeHostX11 now deals
with owned windows.

BUG=344265
TEST=none
R=erg@chromium.org

Review URL: https://codereview.chromium.org/186483007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255377 0039d316-1c4b-4281-b951-d872f2087c98

BUG=349948
TBR=dpolukhin@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/184243004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255374 0039d316-1c4b-4281-b951-d872f2087c98

This is the third part of a three-sided MediaSource
Chromium/Blink+roll Blink/Chromium set of changes to enable
JavaScript web apps to continue to round-trip
SourceBuffer.timestampOffset set/get if SourceBuffer append
operations do not update the timestamp offset, and to
retrieve any updated timestamp offset as can occur during
upcoming "sequence" append mode compliant coded frame
processing.

This change removes the obsolete Chromium
WebSourceBufferImpl::append(,) now that Blink
WebSourceBuffer.h only has append(,,double*).

R=acolwell@chromium.org
BUG=249422
TEST=All media_unittests and http/tests/media/media-source layout tests pass locally on Linux

Review URL: https://codereview.chromium.org/183933002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255373 0039d316-1c4b-4281-b951-d872f2087c98

Reverted due to memory leaks.

> Migrate chrome.bluetooth API backend to use device::BluetoothDiscoverySession.
> 
> Modified chrome.bluetooth API code to use the new discovery session API for
> device discovery.
> 
> BUG=346982
> TEST=1. unit_tests, browser_tests
>      2. Use multiple running extension instances with the bluetooth permission and check for the following behavior:
>         - Only one session per extension. Calling chrome.bluetooth.stopDiscovery from one running app should fail if that app never called chrome.bluetooth.startDiscovery.
>         - An app's discovery sessions should get cleaned up if that app crashes or unloads.
>         - Use bt_console and run "power off". This should cause all discovery sessions that have been assigned to apps to get marked as inactive. Verify by requesting a new discovery session from apps, which should succeed.
> 
> Review URL: https://codereview.chromium.org/179123008

TBR=armansito@chromium.org
BUG=349942

Review URL: https://codereview.chromium.org/188663003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255372 0039d316-1c4b-4281-b951-d872f2087c98

When the browser actions container's animation ended, we should set 'suppress_chevron_' to false then re-layout it.

If we keep 'suppress_chevron_' being true and notify to re-layout the toolbar, then finish the animation, because there is no chance to layout toolbar again, some browser action views would not be showed, neither chevron. That makes people puzzled.

TBR=sky
R=finnur@chromium.org
BUG=328307

Review URL: https://codereview.chromium.org/188353002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255371 0039d316-1c4b-4281-b951-d872f2087c98

This patch provides the support needed for setting http referrers on URL
loads triggered from the context menu:

- adds Referrer.java holder corresponding to the native
  content::Referrer
- ContextMenuParams.java now holds a Referrer field set (along with
  sanitization) by ContextMenuHelper
- ContextMenuItemDelegate methods that trigger new navigations accept a
  Referrer parameter (save for OpenInNewIncognitoTab, as we drop the
  referrer in this case)
- LoadUrlParams.java also gets a Referrer field, so that embedders can
  set it in their ContextMenuItemDelegate implementations

BUG=340295

Review URL: https://codereview.chromium.org/178193033

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255370 0039d316-1c4b-4281-b951-d872f2087c98

This adds the Java hooks needed and a menu option to distill the current
web page in the test shell.

Original CL: https://codereview.chromium.org/185113002/

BUG=319881

Review URL: https://codereview.chromium.org/183893029

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255369 0039d316-1c4b-4281-b951-d872f2087c98

. Eliminate the implementation of LayerAnimationObserver. It seems no one expects WED to be a LAO anymore.
. Remove WindowTreeHostDelegate. This involves moving some of the functions to WTH, removing some completely, and moving the remainder to the WED public API. A little icky for now, but I plan to take a pass on cleaning up WED's public API once I get it pared down.
. window_tree_host_x11_unittest provided an implementation of WTHD that was not WED. This is not a valid situation now since various functions in WTH now call directly through to WED. Replaced this with a simple EventHandler to test that events are being dispatched.

R=sky@chromium.org
http://crbug.com/308843

Review URL: https://codereview.chromium.org/188223002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255368 0039d316-1c4b-4281-b951-d872f2087c98

BUG=341607
TEST=Make sure that you have the --new-profile-management flag turned on, at least two profiles, and at least one of them signed in. Go to chrome://settings. All the users should display an avatar image in the Users section, including the signed in users, which will have their Gaia photo.

Review URL: https://codereview.chromium.org/183893028

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255367 0039d316-1c4b-4281-b951-d872f2087c98

BUG=348629
TBR=padolph@netflix.com

Review URL: https://codereview.chromium.org/183893040

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255364 0039d316-1c4b-4281-b951-d872f2087c98

Reason for revert:
Experimentally reverting this CL because it may have broken AshTestHelperTest.AshTestHelper on Windows Vista . Will revert the revert if this CL is innocent

See http://build.chromium.org/p/chromium.win/builders/Vista%20Tests%20%282%29/builds/41640/steps/ash_unittests/logs/AshTestHelper

Original issue's description:
> Enable the volume slider in Ash for windows.
> 
> BUG=227247
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=255215

TBR=henrika@chromium.org,tommi@chromium.org,jennyz@chromium.org,dewittj@chromium.org,zturner@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=227247

Review URL: https://codereview.chromium.org/188693004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255363 0039d316-1c4b-4281-b951-d872f2087c98

Move to focusedElement() as focusedNode is deprecated and will be removed soon..

BUG=258763
NOTRY=true

Review URL: https://codereview.chromium.org/184933002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255362 0039d316-1c4b-4281-b951-d872f2087c98

The 1x resources landed here: https://codereview.chromium.org/176903008/

BUG=324052
R=oshima@chromium.org

Review URL: https://codereview.chromium.org/184273004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255361 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255359 0039d316-1c4b-4281-b951-d872f2087c98

In this way, more information can be added (currently added 'description' and 'owner' for each action) 

A few functions are moved from tools/metrics/histograms to tools/metrics/common to be shared by tools/metrics and tools/histograms.

BUG=340735
NOTRY=true

Review URL: https://codereview.chromium.org/149503005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255357 0039d316-1c4b-4281-b951-d872f2087c98

Revert of android: Make dex_action.gypi not call md5sum. (https://codereview.chromium.org/183883024/)

Reason for revert:
Experimentally reverting because it is possible that this CL broke the android build http://build.chromium.org/p/chromium/builders/Android/builds/21418/steps/compile/logs/stdio

Will undo the revert if this CL is innocent

Original issue's description:
> android: Make dex_action.gypi not call md5sum.
> 
> No intended behavior change.
> 
> dex_action.gypi didn't pass its input_paths variable to dex.py,
> but that was only set in a single place, and only to a single file there.
> It was only used in java_apk.gypi which does manual threading of
> stamp files to order actions (since it's a gypi, it can't easily use type
> none targets with dependencies). Since dex.py doesn't look at this
> stamp file at all, the it doesn't need to rerun when the stamp file disappears.
> 
> To make this a bit more obvious, remove dex_action.gypi's input_paths
> variable and set 'inputs' directly in the one place with the stamp file.
> (dex.py will still rerun if the name of the stamp file changes, due to
> regular timestamp handling.)
> 
> BUG=177552
> NOTRY=true
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=255325

TBR=cjhopman@chromium.org,thakis@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=177552

Review URL: https://codereview.chromium.org/185563012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255356 0039d316-1c4b-4281-b951-d872f2087c98

The tap and click_element page actions have a very similar purpose but use a very
different implementation. This patch renames click_element to javascript_click
to indicate that it is not using synthetic gestures as tap does. It also replaces
click_element with tap inside smoothness measurements, so we exercise the entire
input stack. In other places we keep click_element (now javascript_click) because
tap's functionality isn't always sufficient (click elements outside of the view;
clicking on very small elements causes Chrome to magnify the element first).
Further, tap now supports all ways to select an element that click_element
supported (by text, xpath etc).

This CL also removes the tap_element page action, which was only used inside
smoothness for the pica page set. This case is currently broken and adamk@ said
he'd re-implement it using synthetic gestures if needed.

BUG=339233

Review URL: https://codereview.chromium.org/181483002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255355 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Eliminate the dependence of tts extension API for tts_controller.cc (https://codereview.chromium.org/181123004/)

Reason for revert:
Breakes ChromeVox spoken feedback on OOBE.

Original issue's description:
> Eliminate the dependence of tts extension API for tts_controller.cc
> 
> This is the first patch to move TTS implementation from browser layer
> to content. To move TTS to conent, the direct dependence of browser
> compoent should be eliminated from TTS classes first.
> 
> R=tommi@chromium.org, dmazzoni@chromium.org, sky@chromium.org
> BUG=347045
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=254740

TBR=dmazzoni@chromium.org,tommi@chromium.org,sky@chromium.org,chun.gao@intel.com
NOTREECHECKS=true
NOTRY=true
BUG=347045

Review URL: https://codereview.chromium.org/188633003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255354 0039d316-1c4b-4281-b951-d872f2087c98

Allow wrapping a media::VideoFrame with a new view_rect as long as the new view_rect is within the original rect.

BUG= 349450
R=fischman@chromium.org, tommi@chromium.org

Review URL: https://codereview.chromium.org/187573006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255353 0039d316-1c4b-4281-b951-d872f2087c98

This CL introduce the core classes for handling classification of
data in a hierarchical fashion. It introduces two concepts:
- A rule tree: hierarchy of rules defined by the end-user.
- A result tree: the corresponding tree which aggregates the
memory counters.

BUG=340294
NOTRY=true

Review URL: https://codereview.chromium.org/183173003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255352 0039d316-1c4b-4281-b951-d872f2087c98

From this patch, libjingle is able to get the signal level and stats from the local audio track in chrome.

Note, that this patch works with and without kEnableAudioTrackProcessing. When kEnableAudioTrackProcessing is set, libjingle gets values from the chromium track, otherwise it will continue to use the values got from webrtc.

After https://codereview.chromium.org/178223013/ is landed, this CL will rebase and implement the GetSignalLevel() correctly.

BUG=334273
TEST=content_unittests, nothing breaks;

Review URL: https://codereview.chromium.org/185413009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255351 0039d316-1c4b-4281-b951-d872f2087c98

BUG=348850

Review URL: https://codereview.chromium.org/187453004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255350 0039d316-1c4b-4281-b951-d872f2087c98

This CL adds AnimationFrameCallback class to update the properties of progress
center items at the correct timing, and ensure it invokes the CSS animation.

Previously, we use setTimeout function to ensure changes of progress items
properties are applied to one by one, but if the number of milliseconds is less
than the frame inteval, it does not work.

BUG=348493
TEST=Make a repro function and call it multiple times.

Review URL: https://codereview.chromium.org/185193003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255349 0039d316-1c4b-4281-b951-d872f2087c98

This patch simplifies the code and removed the dependency on mount path, which was redundant after migrating to volume ids and file systems per volume.

TEST=Tested manually, mounted and unmounted a archive and a usb device.
BUG=341263

Review URL: https://codereview.chromium.org/181203002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255348 0039d316-1c4b-4281-b951-d872f2087c98

Without CoreAnimation, the bookmark bar buttons got redraw
messages whenever they were damaged (like, moved around,
made visible, etc). With CoreAnimation, the bookmark bar buttons
have their drawn contents cached and therefore need to be
explicitly told to re-draw.

The function BookmarkBarController::layoutSubviews should do
this, but it only iterates through the buttons in the buttons_ array,
which skips the "apps" and "other" bookmark buttons.

This patch adds those two back in. The general question of "where
else is the apps button assumed to be part of the buttons_ array" is
not answered.

BUG=340238
NOTRY=True

Review URL: https://codereview.chromium.org/184273010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255347 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=true

BUG=264611

Review URL: https://codereview.chromium.org/177113012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255346 0039d316-1c4b-4281-b951-d872f2087c98

BUG=314791

Review URL: https://codereview.chromium.org/187843005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255345 0039d316-1c4b-4281-b951-d872f2087c98

PageTestRunner.GetModernizedTestName was used to support old testnames to be specified, but it is not used anymore. The conversion map in tools/perf/run_measurement
 was already empty.

NOTRY=true
BUG=None

Review URL: https://codereview.chromium.org/183893036

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255344 0039d316-1c4b-4281-b951-d872f2087c98