GitLab

ChannelIDKey does the signing.

Right now ChannelIDSource::GetChannelIDKey is still a synchronous function.
It will become asynchronous in a future CL. The assumption is that channel
ID keys may be stored on a disk, so the lookup may require disk access, but
once a channel ID key is loaded into memory, it can sign data without
blocking.

Corresponds to internal CL 68051215.

R=wtc@chromium.org
BUG=

Review URL: https://codereview.chromium.org/300223007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273748 0039d316-1c4b-4281-b951-d872f2087c98

This adds three new bindings to the automation API:
chrome.automation.Event
chrome.automation.Role
chrome.automation.State

Each is an object mapping an enum name to its value. Currently, this is an identity mapping. However, with the way the cl constructs the bindings, we could in the future substitute with integer values (and eliminate the stringification on either end of our extension/browser IPC).

BUG=308003003

Review URL: https://codereview.chromium.org/304293002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273747 0039d316-1c4b-4281-b951-d872f2087c98

Since chromium and chromium_trybot recipes expect different test spec file
format, chromium_arm.json cannot be used by chromium isolate, so duplicating
the test spec in chromium.fyi.json. Also we might want different set of tests
in these recipes.

Test spec may have compile targets. Had to change chromium_arm.json and
chromium_trybot.json format to include them. For backward compatibility,
chromium_trybot recipe will continue accepting an array as a test spec, until
this is committed. After that the old format support will be removed.

Removed cmd line args from chromium_arm.json because at the moment they
cannot be transferred to a tester with the current infrastructure.

Removed descriptions (comments) from test specs per phajdan.jr's
request.

R=phajdan.jr@chromium.org, vadimsh@chromium.org
BUG=359338

Review URL: https://codereview.chromium.org/297303012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273746 0039d316-1c4b-4281-b951-d872f2087c98

WebRtcRtpDumpHandler implements RTP header dump creation. It's owned by WebRtcLoggingHandlerHost and receives RTP packet callbacks through P2PSocketHost.

WebRtcRtpDumpWriter is owned by WebRtcDumpHandler and writes the RTP header into a memory buffer. When the in-memory buffer is full, it compresses the data and writes it to the disk.

After the dumping is stopped, WebRtcLoggingHandlerHost calls ReleaseDump to get the completed dump as a .gz file and uploads it.


BUG=363459

Review URL: https://codereview.chromium.org/264793017

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273745 0039d316-1c4b-4281-b951-d872f2087c98

Bitmaps allocated in the browser process may have to be shared between compositors (for surfaces) but not between processes, so they should be backed by new[] and use the SharedBitmap mechanism.

BUG=

Review URL: https://codereview.chromium.org/307973004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273743 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273742 0039d316-1c4b-4281-b951-d872f2087c98

This avoids a DCHECK in ProfileDestroyer::DestroyProfileWhenAppropriate.

I also added a Chrome-side browsertest for the Service Worker system, where we
can check other things that only show up with this embedder.

I tried to move the DCHECK into ~BrowserContext so that we wouldn't need to
expose the Terminate() function across the Content API, but:
* BrowserContext doesn't know whether it's a "testing" profile.
* There is no BrowserContext::GetOffTheRecordProfile with which to check that
  transitive processes have been shut down.
* (Minor) GetHostsForProfile() works by iterating over all hosts, and moving the
  DCHECK to ~BrowserContext would have doubled the number of iterations needed.

BUG=368570

Review URL: https://codereview.chromium.org/305723003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273741 0039d316-1c4b-4281-b951-d872f2087c98

Introduce MicButtonTargeter, a derived class of MaskedViewTargeter,
to define the circular hit-test region of MicButton.

We will need to be able to access the hit test mask corresponding
to a view by calling into its installed MaskedViewTargeter, so
GetHitTestMask() should be made public. Furthermore, add a version
of GetEventTargeter() to EventTarget that returns a const pointer.

BUG=377537, 378485
TEST=SpeechViewTest.ClickMicButton

Review URL: https://codereview.chromium.org/302653002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273740 0039d316-1c4b-4281-b951-d872f2087c98

This is based on a patch sent by Jiang Kelvin, with some modifications.

BUG=369613

Review URL: https://codereview.chromium.org/296983014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273739 0039d316-1c4b-4281-b951-d872f2087c98

This was missing from: https://codereview.chromium.org/281513003/

BUG=303491
R=mpearson@chromium.org

Review URL: https://codereview.chromium.org/307453009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273738 0039d316-1c4b-4281-b951-d872f2087c98

BUG=378612
R=mattm@chromium.org
TBR=asvitkine@chromium.org

Review URL: https://codereview.chromium.org/304873002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273737 0039d316-1c4b-4281-b951-d872f2087c98

Fixing crash in PictureLayerImpl::MarkVisibleResourcesAsRequired when low-res tiles are disabled and adjusting unit tests for it.

Disabling low resolution tiling was added here:
https://codereview.chromium.org/196473007/
However, if we just add switch --disable-low-res-tiling,
right now it will cause crash (null pointer) in
PictureLayerImpl::MarkVisibleResourcesAsRequired. Also
adding some unittests for disabled low res tiles, including
the ones, that would fail without fix in
PictureLayerImpl::MarkVisibleResourcesAsRequired:
NoLowResPictureLayerImplTest.NothingRequiredIfAllHighResTilesShared
and NoLowResPictureLayerImplTest.NothingRequiredIfActiveMissingTiles.

BUG=none

Review URL: https://codereview.chromium.org/260963008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273736 0039d316-1c4b-4281-b951-d872f2087c98

> Wire input events through the ViewManagerClient interface.
> 
> R=sky@chromium.org
> http://crbug.com/365012
> 
> Review URL: https://codereview.chromium.org/300863003

TBR=ben@chromium.org

Review URL: https://codereview.chromium.org/304323005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273735 0039d316-1c4b-4281-b951-d872f2087c98

> Fix bustage, missed files
> 
> BUG=none
> 
> Review URL: https://codereview.chromium.org/304253006

TBR=ben@chromium.org

Review URL: https://codereview.chromium.org/306013003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273734 0039d316-1c4b-4281-b951-d872f2087c98

Not all guests are created equal. <webview> guests are isolated from other processes.
In other words, the content of a <webview> cannot live in the same process as any other
non-<webview> content.

This is not true for <appview>. <appview> guests are NOT isolated. In the future,
BrowserPluginGuest::IsGuest and GuestViewBase::IsGuest may return true even if
RenderProcessHost::IsIsolatedGuest returns false.

BUG=364141, 330264

Review URL: https://codereview.chromium.org/293093010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273733 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none

Review URL: https://codereview.chromium.org/304253006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273732 0039d316-1c4b-4281-b951-d872f2087c98

When we receive a nack for a frame, we can cancel re-sending of any packet that is NOT missing.
This is a fairly minor optimzation, but helps when we're close to the available bandwidth.
(According to simulation.)

Review URL: https://codereview.chromium.org/302913004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273729 0039d316-1c4b-4281-b951-d872f2087c98

> Merge HarfBuzz kerning support from Blink
>
> Original CL: https://codereview.chromium.org/69513002
>
> BUG=321868
> R=msw@chromium.org
>
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=273181

BUG=321868
R=msw@chromium.org

Review URL: https://codereview.chromium.org/308433003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273727 0039d316-1c4b-4281-b951-d872f2087c98

Add a mechansim to send messages to running micro benchmarks through the
GPU benchmarking extension. This will be used for example to shut down a
running micro benchmark from telemetry.

R=vmpstr@chromium.org,jamesr@chromium.org
BUG=376379

Review URL: https://codereview.chromium.org/300963004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273726 0039d316-1c4b-4281-b951-d872f2087c98

    
Looking through crash output, we should do the right thing when chrome is run like this: "DISPLAY= chrome --display=:0.0"

BUG=376971

Review URL: https://codereview.chromium.org/297983005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273725 0039d316-1c4b-4281-b951-d872f2087c98

This broke browser_test on Linux ChromiumOS Tests (dbg). I need to ivestigate it.

> Change directory if the active list item on navigation list is changed.
>
> We want to change current directory if the active item on navigation list is changed.
> But we don't want to change current directory when listed items are spliced or permuted.
> In both cases, 'change' event is dispatched from selection model and we can't distinguish these cases from event itself.
> currentActiveItem_ is introduced to detect change of actual active item.
>
> BUG=375663
> TEST=go through steps in Issue 375663 and 361047
>
> Review URL: https://codereview.chromium.org/303503004

TBR=yoshiki@chromium.org

Review URL: https://codereview.chromium.org/306053004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273724 0039d316-1c4b-4281-b951-d872f2087c98

R=sky@chromium.org
http://crbug.com/365012

Review URL: https://codereview.chromium.org/300863003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273723 0039d316-1c4b-4281-b951-d872f2087c98

BufferedSpdyFramerVisitorInterface::OnPushPromise

BUG=377538

R=jgraettinger@chromium.org

TEST=BufferedSpdyFramerTest.ReadPushPromiseHeaderBlock

Review URL: https://codereview.chromium.org/300553013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273722 0039d316-1c4b-4281-b951-d872f2087c98

Passes the "v" level to DomDistiller for debug output. Now starting
chrome with something like:
out/Debug/chrome --enable-dom-distiller --vmodule=dom_distiller*=1
will log the boilerpipe debug phases to the console.

For content extractor (which doesn't use DomDistillerServiceFactory),
adds a --debug-level flag.

BUG=375443

Review URL: https://codereview.chromium.org/302793002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273721 0039d316-1c4b-4281-b951-d872f2087c98

TBR=sky@chromium.org
BUG=vtl's bitching

Review URL: https://codereview.chromium.org/303873003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273720 0039d316-1c4b-4281-b951-d872f2087c98

This just takes the root surface and displays it using a delegated renderer layer.

BUG=334090

Review URL: https://codereview.chromium.org/302903003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273719 0039d316-1c4b-4281-b951-d872f2087c98

athena_ui_main is a small executable that creates an aura host window
and builds the athena UI on it. This should make it easier to play with
the UI (a la ash_shell etc.)

BUG=362288
R=oshima@chromium.org
TBR=ben@chromium.org

Review URL: https://codereview.chromium.org/302473005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273718 0039d316-1c4b-4281-b951-d872f2087c98

BUG=374439

Review URL: https://codereview.chromium.org/272323003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273717 0039d316-1c4b-4281-b951-d872f2087c98

The test currently depends on timers in a few places. Replace those with more
deterministic behaviour by using the appropriate observers to make sure that
the message-box dialog shows up, the dialog receives the tab key event, and
the dialog is terminated correctly.

BUG=none
R=sky@chromium.org

Review URL: https://codereview.chromium.org/305093002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273716 0039d316-1c4b-4281-b951-d872f2087c98

The SendMessageToExtension action was not working because <webview> originally cast it as a WebRequestEvent which it is not. This CL fires a filtered webview.onMessage event if the action is being operated upon a webview guest.

BUG=

Review URL: https://codereview.chromium.org/307543005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273715 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=true

Review URL: https://codereview.chromium.org/306923004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273714 0039d316-1c4b-4281-b951-d872f2087c98

This change introduces IPC::Channel::Create*() API to turn
IPC::Channel into a heap allocated object. This will allow us to
make Channel a polymorphic class.

This change also tries to hide Channel::Mode from public API
so that we can simplify channel creation code paths cleaner in
following changes. ChannelProxy has to follow same pattern to
finish this cleanup. Such changes will follow.

TEST=none
BUG=377980
R=darin@chromium.org,cpu@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=273575

Review URL: https://codereview.chromium.org/307653003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273713 0039d316-1c4b-4281-b951-d872f2087c98

This way failures in one benchmark won't stop the rest of the tests, and it'll print separate GTest steps for each benchmark.
The test names follow the format BenchmarkSmokeTest.benchmark.name so you can still find BenchmarkSmokeTest with a search.

Also rename it to the more apt BenchmarkSmokeTest.

BUG=368401
TEST=tools/perf/run_tests benchmark

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=270766

Review URL: https://codereview.chromium.org/287773002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273712 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273705 0039d316-1c4b-4281-b951-d872f2087c98

BUG=375809

Review URL: https://codereview.chromium.org/301293003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273702 0039d316-1c4b-4281-b951-d872f2087c98

I see "Could not get the download directory" in a lot of test output.
This is not relevant to any of these tests, so this log should just
become a vlog that can be turned on via command line by developers who
care about it.

R=vrk@chromium.org
BUG=none

Review URL: https://codereview.chromium.org/308673003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273701 0039d316-1c4b-4281-b951-d872f2087c98

This is the last prerequisite before merging AudioReceiver with
VideoReceiver.  As it turns out, certain configuration details that
seemed to be audio-specific or video-specific are in-fact applicable to
both.  Examples: 1) max_frame_rate for audio should be 100 (implying
10ms frames).  2) frequency (i.e., RTP timebase) for video should be 90
kHz, by convention.

BUG=378568

Review URL: https://codereview.chromium.org/306783002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273700 0039d316-1c4b-4281-b951-d872f2087c98

BUG=374833
TEST=visual & unit test

Review URL: https://codereview.chromium.org/302653008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273699 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273698 0039d316-1c4b-4281-b951-d872f2087c98

This patch bumps the version of the external libaddressinput to r258.
The external version is at src/third_party/libaddressinput/src and is
not yet used in Chrome. Chrome uses the forked version at
src/third_party/libaddressinput/chromium/. The r258 version of the
external libaddressinput is ready to begin integrating into Chrome.

The next step is to setup Chrome's translations infrastructure to
retrieve the existing translations for the strings in the external
libaddressinput version. No translation effort is necessary.

BUG=327046

Review URL: https://codereview.chromium.org/302003003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@273697 0039d316-1c4b-4281-b951-d872f2087c98